Commit 85be0b8c6589f58d5667d0a8a8f94524d5de5ba6 | zig

author	Andrew Kelley <andrew@ziglang.org> 2023-01-26 13:15:35-07:00
committer	Andrew Kelley <andrew@ziglang.org> 2023-01-26 16:36:13-07:00
log	85be0b8c6589f58d5667d0a8a8f94524d5de5ba6
tree	30aaa511ff01f908a867a2e43c1f6d1df7110a9c
parent	b260bf42d3cadc12e814220b80fe9b41eeccb345

update C headers to LLVM 16

upstream commit 0604154e006e88e9e7f82d8ee5fd076bda206613

59 files changed, 37368 insertions(+), 35053 deletions(-)

lib
- include

lib/include/__clang_cuda_texture_intrinsics.h+2

...	...	@@ -666,6 +666,7 @@ __device__ static void __tex_fetch(__T *__ptr, cudaTextureObject_t __handle,
666	666	__tex_fetch_v4<__op>::template __run<__FetchT>(__handle, __args...));
667	667	}
668	668
	669	#if CUDA_VERSION < 12000
669	670	// texture<> objects get magically converted into a texture reference. However,
670	671	// there's no way to convert them to cudaTextureObject_t on C++ level. So, we
671	672	// cheat a bit and use inline assembly to do it. It costs us an extra register
...	...	@@ -713,6 +714,7 @@ __tex_fetch(__DataT , __RetT __ptr,
713	714	__tex_fetch_v4<__op>::template __run<__FetchT>(
714	715	__tex_handle_to_obj(__handle), __args...));
715	716	}
	717	#endif // CUDA_VERSION
716	718	} // namespace __cuda_tex
717	719	} // namespace
718	720	#pragma pop_macro("__ASM_OUT")

lib/include/__clang_hip_libdevice_declares.h+5

...	...	@@ -288,12 +288,17 @@ __llvm_amdgcn_rsq_f64(double __x) {
288	288
289	289	__device__ __attribute__((const)) _Float16 __ocml_ceil_f16(_Float16);
290	290	__device__ _Float16 __ocml_cos_f16(_Float16);
	291	__device__ __attribute__((const)) _Float16 __ocml_cvtrtn_f16_f32(float);
	292	__device__ __attribute__((const)) _Float16 __ocml_cvtrtp_f16_f32(float);
	293	__device__ __attribute__((const)) _Float16 __ocml_cvtrtz_f16_f32(float);
291	294	__device__ __attribute__((pure)) _Float16 __ocml_exp_f16(_Float16);
292	295	__device__ __attribute__((pure)) _Float16 __ocml_exp10_f16(_Float16);
293	296	__device__ __attribute__((pure)) _Float16 __ocml_exp2_f16(_Float16);
294	297	__device__ __attribute__((const)) _Float16 __ocml_floor_f16(_Float16);
295	298	__device__ __attribute__((const)) _Float16 __ocml_fma_f16(_Float16, _Float16,
296	299	_Float16);
	300	__device__ __attribute__((const)) _Float16 __ocml_fmax_f16(_Float16, _Float16);
	301	__device__ __attribute__((const)) _Float16 __ocml_fmin_f16(_Float16, _Float16);
297	302	__device__ __attribute__((const)) _Float16 __ocml_fabs_f16(_Float16);
298	303	__device__ __attribute__((const)) int __ocml_isinf_f16(_Float16);
299	304	__device__ __attribute__((const)) int __ocml_isnan_f16(_Float16);

lib/include/__clang_hip_math.h+10-13

...	...	@@ -70,9 +70,9 @@ __DEVICE__ void __static_assert_equal_size() {
70	70	#endif
71	71
72	72	__DEVICE__
73		uint64_t __make_mantissa_base8(const char *__tagp) {
	73	uint64_t __make_mantissa_base8(const char *__tagp __attribute__((nonnull))) {
74	74	uint64_t __r = 0;
75		while (__tagp) {
	75	while (*__tagp != '\0') {
76	76	char __tmp = *__tagp;
77	77
78	78	if (__tmp >= '0' && __tmp <= '7')
...	...	@@ -87,9 +87,9 @@ uint64_t __make_mantissa_base8(const char *__tagp) {
87	87	}
88	88
89	89	__DEVICE__
90		uint64_t __make_mantissa_base10(const char *__tagp) {
	90	uint64_t __make_mantissa_base10(const char *__tagp __attribute__((nonnull))) {
91	91	uint64_t __r = 0;
92		while (__tagp) {
	92	while (*__tagp != '\0') {
93	93	char __tmp = *__tagp;
94	94
95	95	if (__tmp >= '0' && __tmp <= '9')
...	...	@@ -104,9 +104,9 @@ uint64_t __make_mantissa_base10(const char *__tagp) {
104	104	}
105	105
106	106	__DEVICE__
107		uint64_t __make_mantissa_base16(const char *__tagp) {
	107	uint64_t __make_mantissa_base16(const char *__tagp __attribute__((nonnull))) {
108	108	uint64_t __r = 0;
109		while (__tagp) {
	109	while (*__tagp != '\0') {
110	110	char __tmp = *__tagp;
111	111
112	112	if (__tmp >= '0' && __tmp <= '9')
...	...	@@ -125,10 +125,7 @@ uint64_t __make_mantissa_base16(const char *__tagp) {
125	125	}
126	126
127	127	__DEVICE__
128		uint64_t __make_mantissa(const char *__tagp) {
129		if (!__tagp)
130		return 0u;
131
	128	uint64_t __make_mantissa(const char *__tagp __attribute__((nonnull))) {
132	129	if (*__tagp == '0') {
133	130	++__tagp;
134	131
...	...	@@ -233,7 +230,7 @@ __DEVICE__
233	230	float expm1f(float __x) { return __ocml_expm1_f32(__x); }
234	231
235	232	__DEVICE__
236		float fabsf(float __x) { return __ocml_fabs_f32(__x); }
	233	float fabsf(float __x) { return __builtin_fabsf(__x); }
237	234
238	235	__DEVICE__
239	236	float fdimf(float __x, float __y) { return __ocml_fdim_f32(__x, __y); }
...	...	@@ -359,7 +356,7 @@ float modff(float __x, float *__iptr) {
359	356	}
360	357
361	358	__DEVICE__
362		float nanf(const char *__tagp) {
	359	float nanf(const char *__tagp __attribute__((nonnull))) {
363	360	union {
364	361	float val;
365	362	struct ieee_float {
...	...	@@ -792,7 +789,7 @@ __DEVICE__
792	789	double expm1(double __x) { return __ocml_expm1_f64(__x); }
793	790
794	791	__DEVICE__
795		double fabs(double __x) { return __ocml_fabs_f64(__x); }
	792	double fabs(double __x) { return __builtin_fabs(__x); }
796	793
797	794	__DEVICE__
798	795	double fdim(double __x, double __y) { return __ocml_fdim_f64(__x, __y); }

lib/include/__clang_hip_runtime_wrapper.h+1

...	...	@@ -113,6 +113,7 @@ __attribute__((weak)) inline __device__ void free(void *__ptr) {
113	113
114	114	#include <__clang_hip_libdevice_declares.h>
115	115	#include <__clang_hip_math.h>
	116	#include <__clang_hip_stdlib.h>
116	117
117	118	#if defined(__HIPCC_RTC__)
118	119	#include <__clang_hip_cmath.h>

lib/include/__clang_hip_stdlib.h created+43

...	...	@@ -0,0 +1,43 @@
	1	/*===---- __clang_hip_stdlib.h - Device-side HIP math support --------------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9	#ifndef __CLANG_HIP_STDLIB_H__
	10
	11	#if !defined(__HIP__) && !defined(__OPENMP_AMDGCN__)
	12	#error "This file is for HIP and OpenMP AMDGCN device compilation only."
	13	#endif
	14
	15	#if !defined(__cplusplus)
	16
	17	#include <limits.h>
	18
	19	#ifdef __OPENMP_AMDGCN__
	20	#define __DEVICE__ static inline __attribute__((always_inline, nothrow))
	21	#else
	22	#define __DEVICE__ static __device__ inline __attribute__((always_inline))
	23	#endif
	24
	25	__DEVICE__
	26	int abs(int __x) {
	27	int __sgn = __x >> (sizeof(int) * CHAR_BIT - 1);
	28	return (__x ^ __sgn) - __sgn;
	29	}
	30	__DEVICE__
	31	long labs(long __x) {
	32	long __sgn = __x >> (sizeof(long) * CHAR_BIT - 1);
	33	return (__x ^ __sgn) - __sgn;
	34	}
	35	__DEVICE__
	36	long long llabs(long long __x) {
	37	long long __sgn = __x >> (sizeof(long long) * CHAR_BIT - 1);
	38	return (__x ^ __sgn) - __sgn;
	39	}
	40
	41	#endif // !defined(__cplusplus)
	42
	43	#endif // #define __CLANG_HIP_STDLIB_H__

lib/include/altivec.h+14-14

...	...	@@ -17323,32 +17323,32 @@ provided.
17323	17323	#define vec_ncipherlast_be __builtin_altivec_crypto_vncipherlast
17324	17324
17325	17325	#ifdef __VSX__
17326		static __inline__ vector unsigned long long __attribute__((__always_inline__))
17327		__builtin_crypto_vsbox(vector unsigned long long __a) {
	17326	static __inline__ vector unsigned char __attribute__((__always_inline__))
	17327	__builtin_crypto_vsbox(vector unsigned char __a) {
17328	17328	return __builtin_altivec_crypto_vsbox(__a);
17329	17329	}
17330	17330
17331		static __inline__ vector unsigned long long __attribute__((__always_inline__))
17332		__builtin_crypto_vcipher(vector unsigned long long __a,
17333		vector unsigned long long __b) {
	17331	static __inline__ vector unsigned char __attribute__((__always_inline__))
	17332	__builtin_crypto_vcipher(vector unsigned char __a,
	17333	vector unsigned char __b) {
17334	17334	return __builtin_altivec_crypto_vcipher(__a, __b);
17335	17335	}
17336	17336
17337		static __inline__ vector unsigned long long __attribute__((__always_inline__))
17338		__builtin_crypto_vcipherlast(vector unsigned long long __a,
17339		vector unsigned long long __b) {
	17337	static __inline__ vector unsigned char __attribute__((__always_inline__))
	17338	__builtin_crypto_vcipherlast(vector unsigned char __a,
	17339	vector unsigned char __b) {
17340	17340	return __builtin_altivec_crypto_vcipherlast(__a, __b);
17341	17341	}
17342	17342
17343		static __inline__ vector unsigned long long __attribute__((__always_inline__))
17344		__builtin_crypto_vncipher(vector unsigned long long __a,
17345		vector unsigned long long __b) {
	17343	static __inline__ vector unsigned char __attribute__((__always_inline__))
	17344	__builtin_crypto_vncipher(vector unsigned char __a,
	17345	vector unsigned char __b) {
17346	17346	return __builtin_altivec_crypto_vncipher(__a, __b);
17347	17347	}
17348	17348
17349		static __inline__ vector unsigned long long __attribute__((__always_inline__))
17350		__builtin_crypto_vncipherlast(vector unsigned long long __a,
17351		vector unsigned long long __b) {
	17349	static __inline__ vector unsigned char __attribute__((__always_inline__))
	17350	__builtin_crypto_vncipherlast(vector unsigned char __a,
	17351	vector unsigned char __b) {
17352	17352	return __builtin_altivec_crypto_vncipherlast(__a, __b);
17353	17353	}
17354	17354	#endif /* __VSX__ */

lib/include/amxfp16intrin.h created+58

...	...	@@ -0,0 +1,58 @@
	1	/===------------- amxfp16intrin.h - AMX_FP16 intrinsics -- C++ -*---------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===------------------------------------------------------------------------===
	8	*/
	9
	10	#ifndef __IMMINTRIN_H
	11	#error "Never use <amxfp16intrin.h> directly; use <immintrin.h> instead."
	12	#endif /* __IMMINTRIN_H */
	13
	14	#ifndef __AMX_FP16INTRIN_H
	15	#define __AMX_FP16INTRIN_H
	16	#ifdef __x86_64__
	17
	18	/// Compute dot-product of FP16 (16-bit) floating-point pairs in tiles \a a
	19	/// and \a b, accumulating the intermediate single-precision (32-bit)
	20	/// floating-point elements with elements in \a dst, and store the 32-bit
	21	/// result back to tile \a dst.
	22	///
	23	/// \headerfile <immintrin.h>
	24	///
	25	/// \code
	26	/// void _tile_dpfp16ps (__tile dst, __tile a, __tile b)
	27	/// \endcode
	28	///
	29	/// \code{.operation}
	30	/// FOR m := 0 TO dst.rows - 1
	31	/// tmp := dst.row[m]
	32	/// FOR k := 0 TO (a.colsb / 4) - 1
	33	/// FOR n := 0 TO (dst.colsb / 4) - 1
	34	/// tmp.fp32[n] += FP32(a.row[m].fp16[2k+0])
	35	/// FP32(b.row[k].fp16[2*n+0])
	36	/// tmp.fp32[n] += FP32(a.row[m].fp16[2k+1])
	37	/// FP32(b.row[k].fp16[2*n+1])
	38	/// ENDFOR
	39	/// ENDFOR
	40	/// write_row_and_zero(dst, m, tmp, dst.colsb)
	41	/// ENDFOR
	42	/// zero_upper_rows(dst, dst.rows)
	43	/// zero_tileconfig_start()
	44	/// \endcode
	45	///
	46	/// This intrinsic corresponds to the \c TDPFP16PS instruction.
	47	///
	48	/// \param dst
	49	/// The destination tile. Max size is 1024 Bytes.
	50	/// \param a
	51	/// The 1st source tile. Max size is 1024 Bytes.
	52	/// \param b
	53	/// The 2nd source tile. Max size is 1024 Bytes.
	54	#define _tile_dpfp16ps(dst, a, b) \
	55	__builtin_ia32_tdpfp16ps(dst, a, b)
	56
	57	#endif /* __x86_64__ */
	58	#endif /* __AMX_FP16INTRIN_H */

lib/include/amxintrin.h+32

...	...	@@ -22,6 +22,8 @@
22	22	__attribute__((__always_inline__, __nodebug__, __target__("amx-int8")))
23	23	#define __DEFAULT_FN_ATTRS_BF16 \
24	24	__attribute__((__always_inline__, __nodebug__, __target__("amx-bf16")))
	25	#define __DEFAULT_FN_ATTRS_FP16 \
	26	__attribute__((__always_inline__, __nodebug__, __target__("amx-fp16")))
25	27
26	28	/// Load tile configuration from a 64-byte memory location specified by
27	29	/// "mem_addr". The tile configuration includes the tile type palette, the
...	...	@@ -290,6 +292,13 @@ _tile_dpbf16ps_internal(unsigned short m, unsigned short n, unsigned short k,
290	292	return __builtin_ia32_tdpbf16ps_internal(m, n, k, dst, src1, src2);
291	293	}
292	294
	295	/// This is internal intrinsic. C/C++ user should avoid calling it directly.
	296	static __inline__ _tile1024i __DEFAULT_FN_ATTRS_FP16
	297	_tile_dpfp16ps_internal(unsigned short m, unsigned short n, unsigned short k,
	298	_tile1024i dst, _tile1024i src1, _tile1024i src2) {
	299	return __builtin_ia32_tdpfp16ps_internal(m, n, k, dst, src1, src2);
	300	}
	301
293	302	/// This struct pack the shape and tile data together for user. We suggest
294	303	/// initializing the struct as early as possible, because compiler depends
295	304	/// on the shape information to do configure. The constant value is preferred
...	...	@@ -484,9 +493,32 @@ static __inline__ void __tile_dpbf16ps(__tile1024i *dst, __tile1024i src0,
484	493	src0.tile, src1.tile);
485	494	}
486	495
	496	/// Compute dot-product of FP16 (16-bit) floating-point pairs in tiles src0 and
	497	/// src1, accumulating the intermediate single-precision (32-bit) floating-point
	498	/// elements with elements in "dst", and store the 32-bit result back to tile
	499	/// "dst".
	500	///
	501	/// \headerfile <immintrin.h>
	502	///
	503	/// This intrinsic corresponds to the <c> TDPFP16PS </c> instruction.
	504	///
	505	/// \param dst
	506	/// The destination tile. Max size is 1024 Bytes.
	507	/// \param src0
	508	/// The 1st source tile. Max size is 1024 Bytes.
	509	/// \param src1
	510	/// The 2nd source tile. Max size is 1024 Bytes.
	511	__DEFAULT_FN_ATTRS_FP16
	512	static __inline__ void __tile_dpfp16ps(__tile1024i *dst, __tile1024i src0,
	513	__tile1024i src1) {
	514	dst->tile = _tile_dpfp16ps_internal(src0.row, src1.col, src0.col, dst->tile,
	515	src0.tile, src1.tile);
	516	}
	517
487	518	#undef __DEFAULT_FN_ATTRS_TILE
488	519	#undef __DEFAULT_FN_ATTRS_INT8
489	520	#undef __DEFAULT_FN_ATTRS_BF16
	521	#undef __DEFAULT_FN_ATTRS_FP16
490	522
491	523	#endif /* __x86_64__ */
492	524	#endif /* __AMXINTRIN_H */

lib/include/arm_acle.h+76-75

...	...	@@ -64,7 +64,7 @@ static __inline__ void __attribute__((__always_inline__, __nodebug__)) __yield(v
64	64	}
65	65	#endif
66	66
67		#if __ARM_32BIT_STATE
	67	#if defined(__ARM_32BIT_STATE) && __ARM_32BIT_STATE
68	68	#define __dbg(t) __builtin_arm_dbg(t)
69	69	#endif
70	70
...	...	@@ -82,7 +82,7 @@ __swp(uint32_t __x, volatile uint32_t *__p) {
82	82	/* 8.6.1 Data prefetch */
83	83	#define __pld(addr) __pldx(0, 0, 0, addr)
84	84
85		#if __ARM_32BIT_STATE
	85	#if defined(__ARM_32BIT_STATE) && __ARM_32BIT_STATE
86	86	#define __pldx(access_kind, cache_level, retention_policy, addr) \
87	87	__builtin_arm_prefetch(addr, access_kind, 1)
88	88	#else
...	...	@@ -93,7 +93,7 @@ __swp(uint32_t __x, volatile uint32_t *__p) {
93	93	/* 8.6.2 Instruction prefetch */
94	94	#define __pli(addr) __plix(0, 0, addr)
95	95
96		#if __ARM_32BIT_STATE
	96	#if defined(__ARM_32BIT_STATE) && __ARM_32BIT_STATE
97	97	#define __plix(cache_level, retention_policy, addr) \
98	98	__builtin_arm_prefetch(addr, 0, 0)
99	99	#else
...	...	@@ -140,17 +140,17 @@ __rorl(unsigned long __x, uint32_t __y) {
140	140	/* CLZ */
141	141	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
142	142	__clz(uint32_t __t) {
143		return __builtin_clz(__t);
	143	return (uint32_t)__builtin_clz(__t);
144	144	}
145	145
146	146	static __inline__ unsigned long __attribute__((__always_inline__, __nodebug__))
147	147	__clzl(unsigned long __t) {
148		return __builtin_clzl(__t);
	148	return (unsigned long)__builtin_clzl(__t);
149	149	}
150	150
151	151	static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__))
152	152	__clzll(uint64_t __t) {
153		return __builtin_clzll(__t);
	153	return (uint64_t)__builtin_clzll(__t);
154	154	}
155	155
156	156	/* CLS */
...	...	@@ -201,7 +201,7 @@ __rev16(uint32_t __t) {
201	201
202	202	static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__))
203	203	__rev16ll(uint64_t __t) {
204		return (((uint64_t)__rev16(__t >> 32)) << 32) \| __rev16(__t);
	204	return (((uint64_t)__rev16(__t >> 32)) << 32) \| (uint64_t)__rev16((uint32_t)__t);
205	205	}
206	206
207	207	static __inline__ unsigned long __attribute__((__always_inline__, __nodebug__))
...	...	@@ -216,7 +216,7 @@ __rev16l(unsigned long __t) {
216	216	/* REVSH */
217	217	static __inline__ int16_t __attribute__((__always_inline__, __nodebug__))
218	218	__revsh(int16_t __t) {
219		return __builtin_bswap16(__t);
	219	return (int16_t)__builtin_bswap16((uint16_t)__t);
220	220	}
221	221
222	222	/* RBIT */
...	...	@@ -227,7 +227,7 @@ __rbit(uint32_t __t) {
227	227
228	228	static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__))
229	229	__rbitll(uint64_t __t) {
230		#if __ARM_32BIT_STATE
	230	#if defined(__ARM_32BIT_STATE) && __ARM_32BIT_STATE
231	231	return (((uint64_t)__builtin_arm_rbit(__t)) << 32) \|
232	232	__builtin_arm_rbit(__t >> 32);
233	233	#else
...	...	@@ -247,7 +247,7 @@ __rbitl(unsigned long __t) {
247	247	/*
248	248	* 9.3 16-bit multiplications
249	249	*/
250		#if __ARM_FEATURE_DSP
	250	#if defined(__ARM_FEATURE_DSP) && __ARM_FEATURE_DSP
251	251	static __inline__ int32_t __attribute__((__always_inline__,__nodebug__))
252	252	__smulbb(int32_t __a, int32_t __b) {
253	253	return __builtin_arm_smulbb(__a, __b);
...	...	@@ -277,17 +277,17 @@ __smulwt(int32_t __a, int32_t __b) {
277	277	/*
278	278	* 9.4 Saturating intrinsics
279	279	*
280		* FIXME: Change guard to their corrosponding __ARM_FEATURE flag when Q flag
	280	* FIXME: Change guard to their corresponding __ARM_FEATURE flag when Q flag
281	281	* intrinsics are implemented and the flag is enabled.
282	282	*/
283	283	/* 9.4.1 Width-specified saturation intrinsics */
284		#if __ARM_FEATURE_SAT
	284	#if defined(__ARM_FEATURE_SAT) && __ARM_FEATURE_SAT
285	285	#define __ssat(x, y) __builtin_arm_ssat(x, y)
286	286	#define __usat(x, y) __builtin_arm_usat(x, y)
287	287	#endif
288	288
289	289	/* 9.4.2 Saturating addition and subtraction intrinsics */
290		#if __ARM_FEATURE_DSP
	290	#if defined(__ARM_FEATURE_DSP) && __ARM_FEATURE_DSP
291	291	static __inline__ int32_t __attribute__((__always_inline__, __nodebug__))
292	292	__qadd(int32_t __t, int32_t __v) {
293	293	return __builtin_arm_qadd(__t, __v);
...	...	@@ -305,7 +305,7 @@ __qdbl(int32_t __t) {
305	305	#endif
306	306
307	307	/* 9.4.3 Accumultating multiplications */
308		#if __ARM_FEATURE_DSP
	308	#if defined(__ARM_FEATURE_DSP) && __ARM_FEATURE_DSP
309	309	static __inline__ int32_t __attribute__((__always_inline__, __nodebug__))
310	310	__smlabb(int32_t __a, int32_t __b, int32_t __c) {
311	311	return __builtin_arm_smlabb(__a, __b, __c);
...	...	@@ -334,13 +334,13 @@ __smlawt(int32_t __a, int32_t __b, int32_t __c) {
334	334
335	335
336	336	/* 9.5.4 Parallel 16-bit saturation */
337		#if __ARM_FEATURE_SIMD32
	337	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
338	338	#define __ssat16(x, y) __builtin_arm_ssat16(x, y)
339	339	#define __usat16(x, y) __builtin_arm_usat16(x, y)
340	340	#endif
341	341
342	342	/* 9.5.5 Packing and unpacking */
343		#if __ARM_FEATURE_SIMD32
	343	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
344	344	typedef int32_t int8x4_t;
345	345	typedef int32_t int16x2_t;
346	346	typedef uint32_t uint8x4_t;
...	...	@@ -365,7 +365,7 @@ __uxtb16(int8x4_t __a) {
365	365	#endif
366	366
367	367	/* 9.5.6 Parallel selection */
368		#if __ARM_FEATURE_SIMD32
	368	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
369	369	static __inline__ uint8x4_t __attribute__((__always_inline__, __nodebug__))
370	370	__sel(uint8x4_t __a, uint8x4_t __b) {
371	371	return __builtin_arm_sel(__a, __b);
...	...	@@ -373,7 +373,7 @@ __sel(uint8x4_t __a, uint8x4_t __b) {
373	373	#endif
374	374
375	375	/* 9.5.7 Parallel 8-bit addition and subtraction */
376		#if __ARM_FEATURE_SIMD32
	376	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
377	377	static __inline__ int8x4_t __attribute__((__always_inline__, __nodebug__))
378	378	__qadd8(int8x4_t __a, int8x4_t __b) {
379	379	return __builtin_arm_qadd8(__a, __b);
...	...	@@ -425,7 +425,7 @@ __usub8(uint8x4_t __a, uint8x4_t __b) {
425	425	#endif
426	426
427	427	/* 9.5.8 Sum of 8-bit absolute differences */
428		#if __ARM_FEATURE_SIMD32
	428	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
429	429	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
430	430	__usad8(uint8x4_t __a, uint8x4_t __b) {
431	431	return __builtin_arm_usad8(__a, __b);
...	...	@@ -437,7 +437,7 @@ __usada8(uint8x4_t __a, uint8x4_t __b, uint32_t __c) {
437	437	#endif
438	438
439	439	/* 9.5.9 Parallel 16-bit addition and subtraction */
440		#if __ARM_FEATURE_SIMD32
	440	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
441	441	static __inline__ int16x2_t __attribute__((__always_inline__, __nodebug__))
442	442	__qadd16(int16x2_t __a, int16x2_t __b) {
443	443	return __builtin_arm_qadd16(__a, __b);
...	...	@@ -537,7 +537,7 @@ __usub16(uint16x2_t __a, uint16x2_t __b) {
537	537	#endif
538	538
539	539	/* 9.5.10 Parallel 16-bit multiplications */
540		#if __ARM_FEATURE_SIMD32
	540	#if defined(__ARM_FEATURE_SIMD32) && __ARM_FEATURE_SIMD32
541	541	static __inline__ int32_t __attribute__((__always_inline__, __nodebug__))
542	542	__smlad(int16x2_t __a, int16x2_t __b, int32_t __c) {
543	543	return __builtin_arm_smlad(__a, __b, __c);
...	...	@@ -589,155 +589,156 @@ __smusdx(int16x2_t __a, int16x2_t __b) {
589	589	#endif
590	590
591	591	/* 9.7 CRC32 intrinsics */
592		#if __ARM_FEATURE_CRC32
593		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	592	#if (defined(__ARM_FEATURE_CRC32) && __ARM_FEATURE_CRC32) \|\| \
	593	(defined(__ARM_64BIT_STATE) && __ARM_64BIT_STATE)
	594	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
594	595	__crc32b(uint32_t __a, uint8_t __b) {
595	596	return __builtin_arm_crc32b(__a, __b);
596	597	}
597	598
598		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	599	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
599	600	__crc32h(uint32_t __a, uint16_t __b) {
600	601	return __builtin_arm_crc32h(__a, __b);
601	602	}
602	603
603		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	604	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
604	605	__crc32w(uint32_t __a, uint32_t __b) {
605	606	return __builtin_arm_crc32w(__a, __b);
606	607	}
607	608
608		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	609	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
609	610	__crc32d(uint32_t __a, uint64_t __b) {
610	611	return __builtin_arm_crc32d(__a, __b);
611	612	}
612	613
613		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	614	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
614	615	__crc32cb(uint32_t __a, uint8_t __b) {
615	616	return __builtin_arm_crc32cb(__a, __b);
616	617	}
617	618
618		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	619	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
619	620	__crc32ch(uint32_t __a, uint16_t __b) {
620	621	return __builtin_arm_crc32ch(__a, __b);
621	622	}
622	623
623		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	624	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
624	625	__crc32cw(uint32_t __a, uint32_t __b) {
625	626	return __builtin_arm_crc32cw(__a, __b);
626	627	}
627	628
628		static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__))
	629	static __inline__ uint32_t __attribute__((__always_inline__, __nodebug__, target("crc")))
629	630	__crc32cd(uint32_t __a, uint64_t __b) {
630	631	return __builtin_arm_crc32cd(__a, __b);
631	632	}
632	633	#endif
633	634
634	635	/* Armv8.3-A Javascript conversion intrinsic */
635		#if __ARM_64BIT_STATE && defined(__ARM_FEATURE_JCVT)
636		static __inline__ int32_t __attribute__((__always_inline__, __nodebug__))
	636	#if defined(__ARM_64BIT_STATE) && __ARM_64BIT_STATE
	637	static __inline__ int32_t __attribute__((__always_inline__, __nodebug__, target("v8.3a")))
637	638	__jcvt(double __a) {
638	639	return __builtin_arm_jcvt(__a);
639	640	}
640	641	#endif
641	642
642	643	/* Armv8.5-A FP rounding intrinsics */
643		#if __ARM_64BIT_STATE && defined(__ARM_FEATURE_FRINT)
644		static __inline__ float __attribute__((__always_inline__, __nodebug__))
645		__frint32zf(float __a) {
646		return __builtin_arm_frint32zf(__a);
	644	#if defined(__ARM_64BIT_STATE) && __ARM_64BIT_STATE
	645	static __inline__ float __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	646	__rint32zf(float __a) {
	647	return __builtin_arm_rint32zf(__a);
647	648	}
648	649
649		static __inline__ double __attribute__((__always_inline__, __nodebug__))
650		__frint32z(double __a) {
651		return __builtin_arm_frint32z(__a);
	650	static __inline__ double __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	651	__rint32z(double __a) {
	652	return __builtin_arm_rint32z(__a);
652	653	}
653	654
654		static __inline__ float __attribute__((__always_inline__, __nodebug__))
655		__frint64zf(float __a) {
656		return __builtin_arm_frint64zf(__a);
	655	static __inline__ float __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	656	__rint64zf(float __a) {
	657	return __builtin_arm_rint64zf(__a);
657	658	}
658	659
659		static __inline__ double __attribute__((__always_inline__, __nodebug__))
660		__frint64z(double __a) {
661		return __builtin_arm_frint64z(__a);
	660	static __inline__ double __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	661	__rint64z(double __a) {
	662	return __builtin_arm_rint64z(__a);
662	663	}
663	664
664		static __inline__ float __attribute__((__always_inline__, __nodebug__))
665		__frint32xf(float __a) {
666		return __builtin_arm_frint32xf(__a);
	665	static __inline__ float __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	666	__rint32xf(float __a) {
	667	return __builtin_arm_rint32xf(__a);
667	668	}
668	669
669		static __inline__ double __attribute__((__always_inline__, __nodebug__))
670		__frint32x(double __a) {
671		return __builtin_arm_frint32x(__a);
	670	static __inline__ double __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	671	__rint32x(double __a) {
	672	return __builtin_arm_rint32x(__a);
672	673	}
673	674
674		static __inline__ float __attribute__((__always_inline__, __nodebug__))
675		__frint64xf(float __a) {
676		return __builtin_arm_frint64xf(__a);
	675	static __inline__ float __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	676	__rint64xf(float __a) {
	677	return __builtin_arm_rint64xf(__a);
677	678	}
678	679
679		static __inline__ double __attribute__((__always_inline__, __nodebug__))
680		__frint64x(double __a) {
681		return __builtin_arm_frint64x(__a);
	680	static __inline__ double __attribute__((__always_inline__, __nodebug__, target("v8.5a")))
	681	__rint64x(double __a) {
	682	return __builtin_arm_rint64x(__a);
682	683	}
683	684	#endif
684	685
685	686	/* Armv8.7-A load/store 64-byte intrinsics */
686		#if __ARM_64BIT_STATE && defined(__ARM_FEATURE_LS64)
	687	#if defined(__ARM_64BIT_STATE) && __ARM_64BIT_STATE
687	688	typedef struct {
688	689	uint64_t val[8];
689	690	} data512_t;
690	691
691		static __inline__ data512_t __attribute__((__always_inline__, __nodebug__))
	692	static __inline__ data512_t __attribute__((__always_inline__, __nodebug__, target("ls64")))
692	693	__arm_ld64b(const void *__addr) {
693		data512_t __value;
694		__builtin_arm_ld64b(__addr, __value.val);
695		return __value;
	694	data512_t __value;
	695	__builtin_arm_ld64b(__addr, __value.val);
	696	return __value;
696	697	}
697		static __inline__ void __attribute__((__always_inline__, __nodebug__))
	698	static __inline__ void __attribute__((__always_inline__, __nodebug__, target("ls64")))
698	699	__arm_st64b(void *__addr, data512_t __value) {
699		__builtin_arm_st64b(__addr, __value.val);
	700	__builtin_arm_st64b(__addr, __value.val);
700	701	}
701		static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__))
	702	static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__, target("ls64")))
702	703	__arm_st64bv(void *__addr, data512_t __value) {
703		return __builtin_arm_st64bv(__addr, __value.val);
	704	return __builtin_arm_st64bv(__addr, __value.val);
704	705	}
705		static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__))
	706	static __inline__ uint64_t __attribute__((__always_inline__, __nodebug__, target("ls64")))
706	707	__arm_st64bv0(void *__addr, data512_t __value) {
707		return __builtin_arm_st64bv0(__addr, __value.val);
	708	return __builtin_arm_st64bv0(__addr, __value.val);
708	709	}
709	710	#endif
710	711
711	712	/* 10.1 Special register intrinsics */
712	713	#define __arm_rsr(sysreg) __builtin_arm_rsr(sysreg)
713	714	#define __arm_rsr64(sysreg) __builtin_arm_rsr64(sysreg)
	715	#define __arm_rsr128(sysreg) __builtin_arm_rsr128(sysreg)
714	716	#define __arm_rsrp(sysreg) __builtin_arm_rsrp(sysreg)
715	717	#define __arm_rsrf(sysreg) __builtin_bit_cast(float, __arm_rsr(sysreg))
716	718	#define __arm_rsrf64(sysreg) __builtin_bit_cast(double, __arm_rsr64(sysreg))
717	719	#define __arm_wsr(sysreg, v) __builtin_arm_wsr(sysreg, v)
718	720	#define __arm_wsr64(sysreg, v) __builtin_arm_wsr64(sysreg, v)
	721	#define __arm_wsr128(sysreg, v) __builtin_arm_wsr128(sysreg, v)
719	722	#define __arm_wsrp(sysreg, v) __builtin_arm_wsrp(sysreg, v)
720	723	#define __arm_wsrf(sysreg, v) __arm_wsr(sysreg, __builtin_bit_cast(uint32_t, v))
721	724	#define __arm_wsrf64(sysreg, v) __arm_wsr64(sysreg, __builtin_bit_cast(uint64_t, v))
722	725
723	726	/* Memory Tagging Extensions (MTE) Intrinsics */
724		#if __ARM_FEATURE_MEMORY_TAGGING
	727	#if defined(__ARM_64BIT_STATE) && __ARM_64BIT_STATE
725	728	#define __arm_mte_create_random_tag(__ptr, __mask) __builtin_arm_irg(__ptr, __mask)
726	729	#define __arm_mte_increment_tag(__ptr, __tag_offset) __builtin_arm_addg(__ptr, __tag_offset)
727	730	#define __arm_mte_exclude_tag(__ptr, __excluded) __builtin_arm_gmi(__ptr, __excluded)
728	731	#define __arm_mte_get_tag(__ptr) __builtin_arm_ldg(__ptr)
729	732	#define __arm_mte_set_tag(__ptr) __builtin_arm_stg(__ptr)
730	733	#define __arm_mte_ptrdiff(__ptra, __ptrb) __builtin_arm_subp(__ptra, __ptrb)
731		#endif
732	734
733	735	/* Memory Operations Intrinsics */
734		#if __ARM_FEATURE_MOPS && __ARM_FEATURE_MEMORY_TAGGING
735	736	#define __arm_mops_memset_tag(__tagged_address, __value, __size) \
736	737	__builtin_arm_mops_memset_tag(__tagged_address, __value, __size)
737	738	#endif
738	739
739	740	/* Transactional Memory Extension (TME) Intrinsics */
740		#if __ARM_FEATURE_TME
	741	#if defined(__ARM_FEATURE_TME) && __ARM_FEATURE_TME
741	742
742	743	#define _TMFAILURE_REASON 0x00007fffu
743	744	#define _TMFAILURE_RTRY 0x00008000u
...	...	@@ -759,12 +760,12 @@ __arm_st64bv0(void *__addr, data512_t __value) {
759	760	#endif /* __ARM_FEATURE_TME */
760	761
761	762	/* Armv8.5-A Random number generation intrinsics */
762		#if __ARM_64BIT_STATE && defined(__ARM_FEATURE_RNG)
763		static __inline__ int __attribute__((__always_inline__, __nodebug__))
	763	#if defined(__ARM_64BIT_STATE) && __ARM_64BIT_STATE
	764	static __inline__ int __attribute__((__always_inline__, __nodebug__, target("rand")))
764	765	__rndr(uint64_t *__p) {
765	766	return __builtin_arm_rndr(__p);
766	767	}
767		static __inline__ int __attribute__((__always_inline__, __nodebug__))
	768	static __inline__ int __attribute__((__always_inline__, __nodebug__, target("rand")))
768	769	__rndrrs(uint64_t *__p) {
769	770	return __builtin_arm_rndrrs(__p);
770	771	}

lib/include/arm_fp16.h+1-1

...	...	@@ -29,7 +29,7 @@
29	29	typedef __fp16 float16_t;
30	30	#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
31	31
32		#if defined(__ARM_FEATURE_FP16_SCALAR_ARITHMETIC) && defined(__aarch64__)
	32	#if defined(__aarch64__)
33	33	#define vabdh_f16(__p0, __p1) __extension__ ({ \
34	34	float16_t __ret; \
35	35	float16_t __s0 = __p0; \

lib/include/arm_neon.h+26428-26598

...	...	@@ -34,11 +34,8 @@
34	34
35	35	#include <stdint.h>
36	36
37		#ifdef __ARM_FEATURE_BF16
38	37	#include <arm_bf16.h>
39	38	typedef __bf16 bfloat16_t;
40		#endif
41
42	39	typedef float float32_t;
43	40	typedef __fp16 float16_t;
44	41	#ifdef __aarch64__
...	...	@@ -428,7 +425,6 @@ typedef struct poly64x2x4_t {
428	425	poly64x2_t val[4];
429	426	} poly64x2x4_t;
430	427
431		#ifdef __ARM_FEATURE_BF16
432	428	typedef __attribute__((neon_vector_type(4))) bfloat16_t bfloat16x4_t;
433	429	typedef __attribute__((neon_vector_type(8))) bfloat16_t bfloat16x8_t;
434	430
...	...	@@ -456,8 +452,6 @@ typedef struct bfloat16x8x4_t {
456	452	bfloat16x8_t val[4];
457	453	} bfloat16x8x4_t;
458	454
459		#endif
460
461	455	#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
462	456
463	457	#ifdef __LITTLE_ENDIAN__
...	...	@@ -6124,194 +6118,230 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
6124	6118	#endif
6125	6119
6126	6120	#ifdef __LITTLE_ENDIAN__
6127		#define vdupq_lane_s32(__p0_20, __p1_20) __extension__ ({ \
6128		int32x4_t __ret_20; \
6129		int32x2_t __s0_20 = __p0_20; \
6130		__ret_20 = splatq_lane_s32(__s0_20, __p1_20); \
	6121	#define vdupq_lane_f16(__p0_20, __p1_20) __extension__ ({ \
	6122	float16x8_t __ret_20; \
	6123	float16x4_t __s0_20 = __p0_20; \
	6124	__ret_20 = splatq_lane_f16(__s0_20, __p1_20); \
6131	6125	__ret_20; \
6132	6126	})
6133	6127	#else
6134		#define vdupq_lane_s32(__p0_21, __p1_21) __extension__ ({ \
6135		int32x4_t __ret_21; \
6136		int32x2_t __s0_21 = __p0_21; \
6137		int32x2_t __rev0_21; __rev0_21 = __builtin_shufflevector(__s0_21, __s0_21, 1, 0); \
6138		__ret_21 = __noswap_splatq_lane_s32(__rev0_21, __p1_21); \
6139		__ret_21 = __builtin_shufflevector(__ret_21, __ret_21, 3, 2, 1, 0); \
	6128	#define vdupq_lane_f16(__p0_21, __p1_21) __extension__ ({ \
	6129	float16x8_t __ret_21; \
	6130	float16x4_t __s0_21 = __p0_21; \
	6131	float16x4_t __rev0_21; __rev0_21 = __builtin_shufflevector(__s0_21, __s0_21, 3, 2, 1, 0); \
	6132	__ret_21 = __noswap_splatq_lane_f16(__rev0_21, __p1_21); \
	6133	__ret_21 = __builtin_shufflevector(__ret_21, __ret_21, 7, 6, 5, 4, 3, 2, 1, 0); \
6140	6134	__ret_21; \
6141	6135	})
6142	6136	#endif
6143	6137
6144	6138	#ifdef __LITTLE_ENDIAN__
6145		#define vdupq_lane_s64(__p0_22, __p1_22) __extension__ ({ \
6146		int64x2_t __ret_22; \
6147		int64x1_t __s0_22 = __p0_22; \
6148		__ret_22 = splatq_lane_s64(__s0_22, __p1_22); \
	6139	#define vdupq_lane_s32(__p0_22, __p1_22) __extension__ ({ \
	6140	int32x4_t __ret_22; \
	6141	int32x2_t __s0_22 = __p0_22; \
	6142	__ret_22 = splatq_lane_s32(__s0_22, __p1_22); \
6149	6143	__ret_22; \
6150	6144	})
6151	6145	#else
6152		#define vdupq_lane_s64(__p0_23, __p1_23) __extension__ ({ \
6153		int64x2_t __ret_23; \
6154		int64x1_t __s0_23 = __p0_23; \
6155		__ret_23 = __noswap_splatq_lane_s64(__s0_23, __p1_23); \
6156		__ret_23 = __builtin_shufflevector(__ret_23, __ret_23, 1, 0); \
	6146	#define vdupq_lane_s32(__p0_23, __p1_23) __extension__ ({ \
	6147	int32x4_t __ret_23; \
	6148	int32x2_t __s0_23 = __p0_23; \
	6149	int32x2_t __rev0_23; __rev0_23 = __builtin_shufflevector(__s0_23, __s0_23, 1, 0); \
	6150	__ret_23 = __noswap_splatq_lane_s32(__rev0_23, __p1_23); \
	6151	__ret_23 = __builtin_shufflevector(__ret_23, __ret_23, 3, 2, 1, 0); \
6157	6152	__ret_23; \
6158	6153	})
6159	6154	#endif
6160	6155
6161	6156	#ifdef __LITTLE_ENDIAN__
6162		#define vdupq_lane_s16(__p0_24, __p1_24) __extension__ ({ \
6163		int16x8_t __ret_24; \
6164		int16x4_t __s0_24 = __p0_24; \
6165		__ret_24 = splatq_lane_s16(__s0_24, __p1_24); \
	6157	#define vdupq_lane_s64(__p0_24, __p1_24) __extension__ ({ \
	6158	int64x2_t __ret_24; \
	6159	int64x1_t __s0_24 = __p0_24; \
	6160	__ret_24 = splatq_lane_s64(__s0_24, __p1_24); \
6166	6161	__ret_24; \
6167	6162	})
6168	6163	#else
6169		#define vdupq_lane_s16(__p0_25, __p1_25) __extension__ ({ \
6170		int16x8_t __ret_25; \
6171		int16x4_t __s0_25 = __p0_25; \
6172		int16x4_t __rev0_25; __rev0_25 = __builtin_shufflevector(__s0_25, __s0_25, 3, 2, 1, 0); \
6173		__ret_25 = __noswap_splatq_lane_s16(__rev0_25, __p1_25); \
6174		__ret_25 = __builtin_shufflevector(__ret_25, __ret_25, 7, 6, 5, 4, 3, 2, 1, 0); \
	6164	#define vdupq_lane_s64(__p0_25, __p1_25) __extension__ ({ \
	6165	int64x2_t __ret_25; \
	6166	int64x1_t __s0_25 = __p0_25; \
	6167	__ret_25 = __noswap_splatq_lane_s64(__s0_25, __p1_25); \
	6168	__ret_25 = __builtin_shufflevector(__ret_25, __ret_25, 1, 0); \
6175	6169	__ret_25; \
6176	6170	})
6177	6171	#endif
6178	6172
6179	6173	#ifdef __LITTLE_ENDIAN__
6180		#define vdup_lane_u8(__p0_26, __p1_26) __extension__ ({ \
6181		uint8x8_t __ret_26; \
6182		uint8x8_t __s0_26 = __p0_26; \
6183		__ret_26 = splat_lane_u8(__s0_26, __p1_26); \
	6174	#define vdupq_lane_s16(__p0_26, __p1_26) __extension__ ({ \
	6175	int16x8_t __ret_26; \
	6176	int16x4_t __s0_26 = __p0_26; \
	6177	__ret_26 = splatq_lane_s16(__s0_26, __p1_26); \
6184	6178	__ret_26; \
6185	6179	})
6186	6180	#else
6187		#define vdup_lane_u8(__p0_27, __p1_27) __extension__ ({ \
6188		uint8x8_t __ret_27; \
6189		uint8x8_t __s0_27 = __p0_27; \
6190		uint8x8_t __rev0_27; __rev0_27 = __builtin_shufflevector(__s0_27, __s0_27, 7, 6, 5, 4, 3, 2, 1, 0); \
6191		__ret_27 = __noswap_splat_lane_u8(__rev0_27, __p1_27); \
	6181	#define vdupq_lane_s16(__p0_27, __p1_27) __extension__ ({ \
	6182	int16x8_t __ret_27; \
	6183	int16x4_t __s0_27 = __p0_27; \
	6184	int16x4_t __rev0_27; __rev0_27 = __builtin_shufflevector(__s0_27, __s0_27, 3, 2, 1, 0); \
	6185	__ret_27 = __noswap_splatq_lane_s16(__rev0_27, __p1_27); \
6192	6186	__ret_27 = __builtin_shufflevector(__ret_27, __ret_27, 7, 6, 5, 4, 3, 2, 1, 0); \
6193	6187	__ret_27; \
6194	6188	})
6195	6189	#endif
6196	6190
6197	6191	#ifdef __LITTLE_ENDIAN__
6198		#define vdup_lane_u32(__p0_28, __p1_28) __extension__ ({ \
6199		uint32x2_t __ret_28; \
6200		uint32x2_t __s0_28 = __p0_28; \
6201		__ret_28 = splat_lane_u32(__s0_28, __p1_28); \
	6192	#define vdup_lane_u8(__p0_28, __p1_28) __extension__ ({ \
	6193	uint8x8_t __ret_28; \
	6194	uint8x8_t __s0_28 = __p0_28; \
	6195	__ret_28 = splat_lane_u8(__s0_28, __p1_28); \
6202	6196	__ret_28; \
6203	6197	})
6204	6198	#else
6205		#define vdup_lane_u32(__p0_29, __p1_29) __extension__ ({ \
6206		uint32x2_t __ret_29; \
6207		uint32x2_t __s0_29 = __p0_29; \
6208		uint32x2_t __rev0_29; __rev0_29 = __builtin_shufflevector(__s0_29, __s0_29, 1, 0); \
6209		__ret_29 = __noswap_splat_lane_u32(__rev0_29, __p1_29); \
6210		__ret_29 = __builtin_shufflevector(__ret_29, __ret_29, 1, 0); \
	6199	#define vdup_lane_u8(__p0_29, __p1_29) __extension__ ({ \
	6200	uint8x8_t __ret_29; \
	6201	uint8x8_t __s0_29 = __p0_29; \
	6202	uint8x8_t __rev0_29; __rev0_29 = __builtin_shufflevector(__s0_29, __s0_29, 7, 6, 5, 4, 3, 2, 1, 0); \
	6203	__ret_29 = __noswap_splat_lane_u8(__rev0_29, __p1_29); \
	6204	__ret_29 = __builtin_shufflevector(__ret_29, __ret_29, 7, 6, 5, 4, 3, 2, 1, 0); \
6211	6205	__ret_29; \
6212	6206	})
6213	6207	#endif
6214	6208
6215		#define vdup_lane_u64(__p0_30, __p1_30) __extension__ ({ \
6216		uint64x1_t __ret_30; \
6217		uint64x1_t __s0_30 = __p0_30; \
6218		__ret_30 = splat_lane_u64(__s0_30, __p1_30); \
6219		__ret_30; \
6220		})
6221	6209	#ifdef __LITTLE_ENDIAN__
6222		#define vdup_lane_u16(__p0_31, __p1_31) __extension__ ({ \
6223		uint16x4_t __ret_31; \
6224		uint16x4_t __s0_31 = __p0_31; \
6225		__ret_31 = splat_lane_u16(__s0_31, __p1_31); \
6226		__ret_31; \
	6210	#define vdup_lane_u32(__p0_30, __p1_30) __extension__ ({ \
	6211	uint32x2_t __ret_30; \
	6212	uint32x2_t __s0_30 = __p0_30; \
	6213	__ret_30 = splat_lane_u32(__s0_30, __p1_30); \
	6214	__ret_30; \
6227	6215	})
6228	6216	#else
6229		#define vdup_lane_u16(__p0_32, __p1_32) __extension__ ({ \
6230		uint16x4_t __ret_32; \
6231		uint16x4_t __s0_32 = __p0_32; \
6232		uint16x4_t __rev0_32; __rev0_32 = __builtin_shufflevector(__s0_32, __s0_32, 3, 2, 1, 0); \
6233		__ret_32 = __noswap_splat_lane_u16(__rev0_32, __p1_32); \
6234		__ret_32 = __builtin_shufflevector(__ret_32, __ret_32, 3, 2, 1, 0); \
6235		__ret_32; \
	6217	#define vdup_lane_u32(__p0_31, __p1_31) __extension__ ({ \
	6218	uint32x2_t __ret_31; \
	6219	uint32x2_t __s0_31 = __p0_31; \
	6220	uint32x2_t __rev0_31; __rev0_31 = __builtin_shufflevector(__s0_31, __s0_31, 1, 0); \
	6221	__ret_31 = __noswap_splat_lane_u32(__rev0_31, __p1_31); \
	6222	__ret_31 = __builtin_shufflevector(__ret_31, __ret_31, 1, 0); \
	6223	__ret_31; \
6236	6224	})
6237	6225	#endif
6238	6226
	6227	#define vdup_lane_u64(__p0_32, __p1_32) __extension__ ({ \
	6228	uint64x1_t __ret_32; \
	6229	uint64x1_t __s0_32 = __p0_32; \
	6230	__ret_32 = splat_lane_u64(__s0_32, __p1_32); \
	6231	__ret_32; \
	6232	})
6239	6233	#ifdef __LITTLE_ENDIAN__
6240		#define vdup_lane_s8(__p0_33, __p1_33) __extension__ ({ \
6241		int8x8_t __ret_33; \
6242		int8x8_t __s0_33 = __p0_33; \
6243		__ret_33 = splat_lane_s8(__s0_33, __p1_33); \
	6234	#define vdup_lane_u16(__p0_33, __p1_33) __extension__ ({ \
	6235	uint16x4_t __ret_33; \
	6236	uint16x4_t __s0_33 = __p0_33; \
	6237	__ret_33 = splat_lane_u16(__s0_33, __p1_33); \
6244	6238	__ret_33; \
6245	6239	})
6246	6240	#else
6247		#define vdup_lane_s8(__p0_34, __p1_34) __extension__ ({ \
6248		int8x8_t __ret_34; \
6249		int8x8_t __s0_34 = __p0_34; \
6250		int8x8_t __rev0_34; __rev0_34 = __builtin_shufflevector(__s0_34, __s0_34, 7, 6, 5, 4, 3, 2, 1, 0); \
6251		__ret_34 = __noswap_splat_lane_s8(__rev0_34, __p1_34); \
6252		__ret_34 = __builtin_shufflevector(__ret_34, __ret_34, 7, 6, 5, 4, 3, 2, 1, 0); \
	6241	#define vdup_lane_u16(__p0_34, __p1_34) __extension__ ({ \
	6242	uint16x4_t __ret_34; \
	6243	uint16x4_t __s0_34 = __p0_34; \
	6244	uint16x4_t __rev0_34; __rev0_34 = __builtin_shufflevector(__s0_34, __s0_34, 3, 2, 1, 0); \
	6245	__ret_34 = __noswap_splat_lane_u16(__rev0_34, __p1_34); \
	6246	__ret_34 = __builtin_shufflevector(__ret_34, __ret_34, 3, 2, 1, 0); \
6253	6247	__ret_34; \
6254	6248	})
6255	6249	#endif
6256	6250
6257	6251	#ifdef __LITTLE_ENDIAN__
6258		#define vdup_lane_f32(__p0_35, __p1_35) __extension__ ({ \
6259		float32x2_t __ret_35; \
6260		float32x2_t __s0_35 = __p0_35; \
6261		__ret_35 = splat_lane_f32(__s0_35, __p1_35); \
	6252	#define vdup_lane_s8(__p0_35, __p1_35) __extension__ ({ \
	6253	int8x8_t __ret_35; \
	6254	int8x8_t __s0_35 = __p0_35; \
	6255	__ret_35 = splat_lane_s8(__s0_35, __p1_35); \
6262	6256	__ret_35; \
6263	6257	})
6264	6258	#else
6265		#define vdup_lane_f32(__p0_36, __p1_36) __extension__ ({ \
6266		float32x2_t __ret_36; \
6267		float32x2_t __s0_36 = __p0_36; \
6268		float32x2_t __rev0_36; __rev0_36 = __builtin_shufflevector(__s0_36, __s0_36, 1, 0); \
6269		__ret_36 = __noswap_splat_lane_f32(__rev0_36, __p1_36); \
6270		__ret_36 = __builtin_shufflevector(__ret_36, __ret_36, 1, 0); \
	6259	#define vdup_lane_s8(__p0_36, __p1_36) __extension__ ({ \
	6260	int8x8_t __ret_36; \
	6261	int8x8_t __s0_36 = __p0_36; \
	6262	int8x8_t __rev0_36; __rev0_36 = __builtin_shufflevector(__s0_36, __s0_36, 7, 6, 5, 4, 3, 2, 1, 0); \
	6263	__ret_36 = __noswap_splat_lane_s8(__rev0_36, __p1_36); \
	6264	__ret_36 = __builtin_shufflevector(__ret_36, __ret_36, 7, 6, 5, 4, 3, 2, 1, 0); \
6271	6265	__ret_36; \
6272	6266	})
6273	6267	#endif
6274	6268
6275	6269	#ifdef __LITTLE_ENDIAN__
6276		#define vdup_lane_s32(__p0_37, __p1_37) __extension__ ({ \
6277		int32x2_t __ret_37; \
6278		int32x2_t __s0_37 = __p0_37; \
6279		__ret_37 = splat_lane_s32(__s0_37, __p1_37); \
	6270	#define vdup_lane_f32(__p0_37, __p1_37) __extension__ ({ \
	6271	float32x2_t __ret_37; \
	6272	float32x2_t __s0_37 = __p0_37; \
	6273	__ret_37 = splat_lane_f32(__s0_37, __p1_37); \
6280	6274	__ret_37; \
6281	6275	})
6282	6276	#else
6283		#define vdup_lane_s32(__p0_38, __p1_38) __extension__ ({ \
6284		int32x2_t __ret_38; \
6285		int32x2_t __s0_38 = __p0_38; \
6286		int32x2_t __rev0_38; __rev0_38 = __builtin_shufflevector(__s0_38, __s0_38, 1, 0); \
6287		__ret_38 = __noswap_splat_lane_s32(__rev0_38, __p1_38); \
	6277	#define vdup_lane_f32(__p0_38, __p1_38) __extension__ ({ \
	6278	float32x2_t __ret_38; \
	6279	float32x2_t __s0_38 = __p0_38; \
	6280	float32x2_t __rev0_38; __rev0_38 = __builtin_shufflevector(__s0_38, __s0_38, 1, 0); \
	6281	__ret_38 = __noswap_splat_lane_f32(__rev0_38, __p1_38); \
6288	6282	__ret_38 = __builtin_shufflevector(__ret_38, __ret_38, 1, 0); \
6289	6283	__ret_38; \
6290	6284	})
6291	6285	#endif
6292	6286
6293		#define vdup_lane_s64(__p0_39, __p1_39) __extension__ ({ \
6294		int64x1_t __ret_39; \
6295		int64x1_t __s0_39 = __p0_39; \
6296		__ret_39 = splat_lane_s64(__s0_39, __p1_39); \
	6287	#ifdef __LITTLE_ENDIAN__
	6288	#define vdup_lane_f16(__p0_39, __p1_39) __extension__ ({ \
	6289	float16x4_t __ret_39; \
	6290	float16x4_t __s0_39 = __p0_39; \
	6291	__ret_39 = splat_lane_f16(__s0_39, __p1_39); \
6297	6292	__ret_39; \
6298	6293	})
6299		#ifdef __LITTLE_ENDIAN__
6300		#define vdup_lane_s16(__p0_40, __p1_40) __extension__ ({ \
6301		int16x4_t __ret_40; \
6302		int16x4_t __s0_40 = __p0_40; \
6303		__ret_40 = splat_lane_s16(__s0_40, __p1_40); \
	6294	#else
	6295	#define vdup_lane_f16(__p0_40, __p1_40) __extension__ ({ \
	6296	float16x4_t __ret_40; \
	6297	float16x4_t __s0_40 = __p0_40; \
	6298	float16x4_t __rev0_40; __rev0_40 = __builtin_shufflevector(__s0_40, __s0_40, 3, 2, 1, 0); \
	6299	__ret_40 = __noswap_splat_lane_f16(__rev0_40, __p1_40); \
	6300	__ret_40 = __builtin_shufflevector(__ret_40, __ret_40, 3, 2, 1, 0); \
6304	6301	__ret_40; \
6305	6302	})
6306		#else
6307		#define vdup_lane_s16(__p0_41, __p1_41) __extension__ ({ \
6308		int16x4_t __ret_41; \
6309		int16x4_t __s0_41 = __p0_41; \
6310		int16x4_t __rev0_41; __rev0_41 = __builtin_shufflevector(__s0_41, __s0_41, 3, 2, 1, 0); \
6311		__ret_41 = __noswap_splat_lane_s16(__rev0_41, __p1_41); \
6312		__ret_41 = __builtin_shufflevector(__ret_41, __ret_41, 3, 2, 1, 0); \
	6303	#endif
	6304
	6305	#ifdef __LITTLE_ENDIAN__
	6306	#define vdup_lane_s32(__p0_41, __p1_41) __extension__ ({ \
	6307	int32x2_t __ret_41; \
	6308	int32x2_t __s0_41 = __p0_41; \
	6309	__ret_41 = splat_lane_s32(__s0_41, __p1_41); \
6313	6310	__ret_41; \
6314	6311	})
	6312	#else
	6313	#define vdup_lane_s32(__p0_42, __p1_42) __extension__ ({ \
	6314	int32x2_t __ret_42; \
	6315	int32x2_t __s0_42 = __p0_42; \
	6316	int32x2_t __rev0_42; __rev0_42 = __builtin_shufflevector(__s0_42, __s0_42, 1, 0); \
	6317	__ret_42 = __noswap_splat_lane_s32(__rev0_42, __p1_42); \
	6318	__ret_42 = __builtin_shufflevector(__ret_42, __ret_42, 1, 0); \
	6319	__ret_42; \
	6320	})
	6321	#endif
	6322
	6323	#define vdup_lane_s64(__p0_43, __p1_43) __extension__ ({ \
	6324	int64x1_t __ret_43; \
	6325	int64x1_t __s0_43 = __p0_43; \
	6326	__ret_43 = splat_lane_s64(__s0_43, __p1_43); \
	6327	__ret_43; \
	6328	})
	6329	#ifdef __LITTLE_ENDIAN__
	6330	#define vdup_lane_s16(__p0_44, __p1_44) __extension__ ({ \
	6331	int16x4_t __ret_44; \
	6332	int16x4_t __s0_44 = __p0_44; \
	6333	__ret_44 = splat_lane_s16(__s0_44, __p1_44); \
	6334	__ret_44; \
	6335	})
	6336	#else
	6337	#define vdup_lane_s16(__p0_45, __p1_45) __extension__ ({ \
	6338	int16x4_t __ret_45; \
	6339	int16x4_t __s0_45 = __p0_45; \
	6340	int16x4_t __rev0_45; __rev0_45 = __builtin_shufflevector(__s0_45, __s0_45, 3, 2, 1, 0); \
	6341	__ret_45 = __noswap_splat_lane_s16(__rev0_45, __p1_45); \
	6342	__ret_45 = __builtin_shufflevector(__ret_45, __ret_45, 3, 2, 1, 0); \
	6343	__ret_45; \
	6344	})
6315	6345	#endif
6316	6346
6317	6347	#ifdef __LITTLE_ENDIAN__
...	...	@@ -14668,245 +14698,245 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
14668	14698	#endif
14669	14699
14670	14700	#ifdef __LITTLE_ENDIAN__
14671		#define vmlaq_lane_u32(__p0_42, __p1_42, __p2_42, __p3_42) __extension__ ({ \
14672		uint32x4_t __ret_42; \
14673		uint32x4_t __s0_42 = __p0_42; \
14674		uint32x4_t __s1_42 = __p1_42; \
14675		uint32x2_t __s2_42 = __p2_42; \
14676		__ret_42 = __s0_42 + __s1_42 * splatq_lane_u32(__s2_42, __p3_42); \
14677		__ret_42; \
14678		})
14679		#else
14680		#define vmlaq_lane_u32(__p0_43, __p1_43, __p2_43, __p3_43) __extension__ ({ \
14681		uint32x4_t __ret_43; \
14682		uint32x4_t __s0_43 = __p0_43; \
14683		uint32x4_t __s1_43 = __p1_43; \
14684		uint32x2_t __s2_43 = __p2_43; \
14685		uint32x4_t __rev0_43; __rev0_43 = __builtin_shufflevector(__s0_43, __s0_43, 3, 2, 1, 0); \
14686		uint32x4_t __rev1_43; __rev1_43 = __builtin_shufflevector(__s1_43, __s1_43, 3, 2, 1, 0); \
14687		uint32x2_t __rev2_43; __rev2_43 = __builtin_shufflevector(__s2_43, __s2_43, 1, 0); \
14688		__ret_43 = __rev0_43 + __rev1_43 * __noswap_splatq_lane_u32(__rev2_43, __p3_43); \
14689		__ret_43 = __builtin_shufflevector(__ret_43, __ret_43, 3, 2, 1, 0); \
14690		__ret_43; \
14691		})
14692		#endif
14693
14694		#ifdef __LITTLE_ENDIAN__
14695		#define vmlaq_lane_u16(__p0_44, __p1_44, __p2_44, __p3_44) __extension__ ({ \
14696		uint16x8_t __ret_44; \
14697		uint16x8_t __s0_44 = __p0_44; \
14698		uint16x8_t __s1_44 = __p1_44; \
14699		uint16x4_t __s2_44 = __p2_44; \
14700		__ret_44 = __s0_44 + __s1_44 * splatq_lane_u16(__s2_44, __p3_44); \
14701		__ret_44; \
14702		})
14703		#else
14704		#define vmlaq_lane_u16(__p0_45, __p1_45, __p2_45, __p3_45) __extension__ ({ \
14705		uint16x8_t __ret_45; \
14706		uint16x8_t __s0_45 = __p0_45; \
14707		uint16x8_t __s1_45 = __p1_45; \
14708		uint16x4_t __s2_45 = __p2_45; \
14709		uint16x8_t __rev0_45; __rev0_45 = __builtin_shufflevector(__s0_45, __s0_45, 7, 6, 5, 4, 3, 2, 1, 0); \
14710		uint16x8_t __rev1_45; __rev1_45 = __builtin_shufflevector(__s1_45, __s1_45, 7, 6, 5, 4, 3, 2, 1, 0); \
14711		uint16x4_t __rev2_45; __rev2_45 = __builtin_shufflevector(__s2_45, __s2_45, 3, 2, 1, 0); \
14712		__ret_45 = __rev0_45 + __rev1_45 * __noswap_splatq_lane_u16(__rev2_45, __p3_45); \
14713		__ret_45 = __builtin_shufflevector(__ret_45, __ret_45, 7, 6, 5, 4, 3, 2, 1, 0); \
14714		__ret_45; \
14715		})
14716		#endif
14717
14718		#ifdef __LITTLE_ENDIAN__
14719		#define vmlaq_lane_f32(__p0_46, __p1_46, __p2_46, __p3_46) __extension__ ({ \
14720		float32x4_t __ret_46; \
14721		float32x4_t __s0_46 = __p0_46; \
14722		float32x4_t __s1_46 = __p1_46; \
14723		float32x2_t __s2_46 = __p2_46; \
14724		__ret_46 = __s0_46 + __s1_46 * splatq_lane_f32(__s2_46, __p3_46); \
	14701	#define vmlaq_lane_u32(__p0_46, __p1_46, __p2_46, __p3_46) __extension__ ({ \
	14702	uint32x4_t __ret_46; \
	14703	uint32x4_t __s0_46 = __p0_46; \
	14704	uint32x4_t __s1_46 = __p1_46; \
	14705	uint32x2_t __s2_46 = __p2_46; \
	14706	__ret_46 = __s0_46 + __s1_46 * splatq_lane_u32(__s2_46, __p3_46); \
14725	14707	__ret_46; \
14726	14708	})
14727	14709	#else
14728		#define vmlaq_lane_f32(__p0_47, __p1_47, __p2_47, __p3_47) __extension__ ({ \
14729		float32x4_t __ret_47; \
14730		float32x4_t __s0_47 = __p0_47; \
14731		float32x4_t __s1_47 = __p1_47; \
14732		float32x2_t __s2_47 = __p2_47; \
14733		float32x4_t __rev0_47; __rev0_47 = __builtin_shufflevector(__s0_47, __s0_47, 3, 2, 1, 0); \
14734		float32x4_t __rev1_47; __rev1_47 = __builtin_shufflevector(__s1_47, __s1_47, 3, 2, 1, 0); \
14735		float32x2_t __rev2_47; __rev2_47 = __builtin_shufflevector(__s2_47, __s2_47, 1, 0); \
14736		__ret_47 = __rev0_47 + __rev1_47 * __noswap_splatq_lane_f32(__rev2_47, __p3_47); \
	14710	#define vmlaq_lane_u32(__p0_47, __p1_47, __p2_47, __p3_47) __extension__ ({ \
	14711	uint32x4_t __ret_47; \
	14712	uint32x4_t __s0_47 = __p0_47; \
	14713	uint32x4_t __s1_47 = __p1_47; \
	14714	uint32x2_t __s2_47 = __p2_47; \
	14715	uint32x4_t __rev0_47; __rev0_47 = __builtin_shufflevector(__s0_47, __s0_47, 3, 2, 1, 0); \
	14716	uint32x4_t __rev1_47; __rev1_47 = __builtin_shufflevector(__s1_47, __s1_47, 3, 2, 1, 0); \
	14717	uint32x2_t __rev2_47; __rev2_47 = __builtin_shufflevector(__s2_47, __s2_47, 1, 0); \
	14718	__ret_47 = __rev0_47 + __rev1_47 * __noswap_splatq_lane_u32(__rev2_47, __p3_47); \
14737	14719	__ret_47 = __builtin_shufflevector(__ret_47, __ret_47, 3, 2, 1, 0); \
14738	14720	__ret_47; \
14739	14721	})
14740	14722	#endif
14741	14723
14742	14724	#ifdef __LITTLE_ENDIAN__
14743		#define vmlaq_lane_s32(__p0_48, __p1_48, __p2_48, __p3_48) __extension__ ({ \
14744		int32x4_t __ret_48; \
14745		int32x4_t __s0_48 = __p0_48; \
14746		int32x4_t __s1_48 = __p1_48; \
14747		int32x2_t __s2_48 = __p2_48; \
14748		__ret_48 = __s0_48 + __s1_48 * splatq_lane_s32(__s2_48, __p3_48); \
	14725	#define vmlaq_lane_u16(__p0_48, __p1_48, __p2_48, __p3_48) __extension__ ({ \
	14726	uint16x8_t __ret_48; \
	14727	uint16x8_t __s0_48 = __p0_48; \
	14728	uint16x8_t __s1_48 = __p1_48; \
	14729	uint16x4_t __s2_48 = __p2_48; \
	14730	__ret_48 = __s0_48 + __s1_48 * splatq_lane_u16(__s2_48, __p3_48); \
14749	14731	__ret_48; \
14750	14732	})
14751	14733	#else
14752		#define vmlaq_lane_s32(__p0_49, __p1_49, __p2_49, __p3_49) __extension__ ({ \
14753		int32x4_t __ret_49; \
14754		int32x4_t __s0_49 = __p0_49; \
14755		int32x4_t __s1_49 = __p1_49; \
14756		int32x2_t __s2_49 = __p2_49; \
14757		int32x4_t __rev0_49; __rev0_49 = __builtin_shufflevector(__s0_49, __s0_49, 3, 2, 1, 0); \
14758		int32x4_t __rev1_49; __rev1_49 = __builtin_shufflevector(__s1_49, __s1_49, 3, 2, 1, 0); \
14759		int32x2_t __rev2_49; __rev2_49 = __builtin_shufflevector(__s2_49, __s2_49, 1, 0); \
14760		__ret_49 = __rev0_49 + __rev1_49 * __noswap_splatq_lane_s32(__rev2_49, __p3_49); \
14761		__ret_49 = __builtin_shufflevector(__ret_49, __ret_49, 3, 2, 1, 0); \
	14734	#define vmlaq_lane_u16(__p0_49, __p1_49, __p2_49, __p3_49) __extension__ ({ \
	14735	uint16x8_t __ret_49; \
	14736	uint16x8_t __s0_49 = __p0_49; \
	14737	uint16x8_t __s1_49 = __p1_49; \
	14738	uint16x4_t __s2_49 = __p2_49; \
	14739	uint16x8_t __rev0_49; __rev0_49 = __builtin_shufflevector(__s0_49, __s0_49, 7, 6, 5, 4, 3, 2, 1, 0); \
	14740	uint16x8_t __rev1_49; __rev1_49 = __builtin_shufflevector(__s1_49, __s1_49, 7, 6, 5, 4, 3, 2, 1, 0); \
	14741	uint16x4_t __rev2_49; __rev2_49 = __builtin_shufflevector(__s2_49, __s2_49, 3, 2, 1, 0); \
	14742	__ret_49 = __rev0_49 + __rev1_49 * __noswap_splatq_lane_u16(__rev2_49, __p3_49); \
	14743	__ret_49 = __builtin_shufflevector(__ret_49, __ret_49, 7, 6, 5, 4, 3, 2, 1, 0); \
14762	14744	__ret_49; \
14763	14745	})
14764	14746	#endif
14765	14747
14766	14748	#ifdef __LITTLE_ENDIAN__
14767		#define vmlaq_lane_s16(__p0_50, __p1_50, __p2_50, __p3_50) __extension__ ({ \
14768		int16x8_t __ret_50; \
14769		int16x8_t __s0_50 = __p0_50; \
14770		int16x8_t __s1_50 = __p1_50; \
14771		int16x4_t __s2_50 = __p2_50; \
14772		__ret_50 = __s0_50 + __s1_50 * splatq_lane_s16(__s2_50, __p3_50); \
	14749	#define vmlaq_lane_f32(__p0_50, __p1_50, __p2_50, __p3_50) __extension__ ({ \
	14750	float32x4_t __ret_50; \
	14751	float32x4_t __s0_50 = __p0_50; \
	14752	float32x4_t __s1_50 = __p1_50; \
	14753	float32x2_t __s2_50 = __p2_50; \
	14754	__ret_50 = __s0_50 + __s1_50 * splatq_lane_f32(__s2_50, __p3_50); \
14773	14755	__ret_50; \
14774	14756	})
14775	14757	#else
14776		#define vmlaq_lane_s16(__p0_51, __p1_51, __p2_51, __p3_51) __extension__ ({ \
14777		int16x8_t __ret_51; \
14778		int16x8_t __s0_51 = __p0_51; \
14779		int16x8_t __s1_51 = __p1_51; \
14780		int16x4_t __s2_51 = __p2_51; \
14781		int16x8_t __rev0_51; __rev0_51 = __builtin_shufflevector(__s0_51, __s0_51, 7, 6, 5, 4, 3, 2, 1, 0); \
14782		int16x8_t __rev1_51; __rev1_51 = __builtin_shufflevector(__s1_51, __s1_51, 7, 6, 5, 4, 3, 2, 1, 0); \
14783		int16x4_t __rev2_51; __rev2_51 = __builtin_shufflevector(__s2_51, __s2_51, 3, 2, 1, 0); \
14784		__ret_51 = __rev0_51 + __rev1_51 * __noswap_splatq_lane_s16(__rev2_51, __p3_51); \
14785		__ret_51 = __builtin_shufflevector(__ret_51, __ret_51, 7, 6, 5, 4, 3, 2, 1, 0); \
	14758	#define vmlaq_lane_f32(__p0_51, __p1_51, __p2_51, __p3_51) __extension__ ({ \
	14759	float32x4_t __ret_51; \
	14760	float32x4_t __s0_51 = __p0_51; \
	14761	float32x4_t __s1_51 = __p1_51; \
	14762	float32x2_t __s2_51 = __p2_51; \
	14763	float32x4_t __rev0_51; __rev0_51 = __builtin_shufflevector(__s0_51, __s0_51, 3, 2, 1, 0); \
	14764	float32x4_t __rev1_51; __rev1_51 = __builtin_shufflevector(__s1_51, __s1_51, 3, 2, 1, 0); \
	14765	float32x2_t __rev2_51; __rev2_51 = __builtin_shufflevector(__s2_51, __s2_51, 1, 0); \
	14766	__ret_51 = __rev0_51 + __rev1_51 * __noswap_splatq_lane_f32(__rev2_51, __p3_51); \
	14767	__ret_51 = __builtin_shufflevector(__ret_51, __ret_51, 3, 2, 1, 0); \
14786	14768	__ret_51; \
14787	14769	})
14788	14770	#endif
14789	14771
14790	14772	#ifdef __LITTLE_ENDIAN__
14791		#define vmla_lane_u32(__p0_52, __p1_52, __p2_52, __p3_52) __extension__ ({ \
14792		uint32x2_t __ret_52; \
14793		uint32x2_t __s0_52 = __p0_52; \
14794		uint32x2_t __s1_52 = __p1_52; \
14795		uint32x2_t __s2_52 = __p2_52; \
14796		__ret_52 = __s0_52 + __s1_52 * splat_lane_u32(__s2_52, __p3_52); \
	14773	#define vmlaq_lane_s32(__p0_52, __p1_52, __p2_52, __p3_52) __extension__ ({ \
	14774	int32x4_t __ret_52; \
	14775	int32x4_t __s0_52 = __p0_52; \
	14776	int32x4_t __s1_52 = __p1_52; \
	14777	int32x2_t __s2_52 = __p2_52; \
	14778	__ret_52 = __s0_52 + __s1_52 * splatq_lane_s32(__s2_52, __p3_52); \
14797	14779	__ret_52; \
14798	14780	})
14799	14781	#else
14800		#define vmla_lane_u32(__p0_53, __p1_53, __p2_53, __p3_53) __extension__ ({ \
14801		uint32x2_t __ret_53; \
14802		uint32x2_t __s0_53 = __p0_53; \
14803		uint32x2_t __s1_53 = __p1_53; \
14804		uint32x2_t __s2_53 = __p2_53; \
14805		uint32x2_t __rev0_53; __rev0_53 = __builtin_shufflevector(__s0_53, __s0_53, 1, 0); \
14806		uint32x2_t __rev1_53; __rev1_53 = __builtin_shufflevector(__s1_53, __s1_53, 1, 0); \
14807		uint32x2_t __rev2_53; __rev2_53 = __builtin_shufflevector(__s2_53, __s2_53, 1, 0); \
14808		__ret_53 = __rev0_53 + __rev1_53 * __noswap_splat_lane_u32(__rev2_53, __p3_53); \
14809		__ret_53 = __builtin_shufflevector(__ret_53, __ret_53, 1, 0); \
	14782	#define vmlaq_lane_s32(__p0_53, __p1_53, __p2_53, __p3_53) __extension__ ({ \
	14783	int32x4_t __ret_53; \
	14784	int32x4_t __s0_53 = __p0_53; \
	14785	int32x4_t __s1_53 = __p1_53; \
	14786	int32x2_t __s2_53 = __p2_53; \
	14787	int32x4_t __rev0_53; __rev0_53 = __builtin_shufflevector(__s0_53, __s0_53, 3, 2, 1, 0); \
	14788	int32x4_t __rev1_53; __rev1_53 = __builtin_shufflevector(__s1_53, __s1_53, 3, 2, 1, 0); \
	14789	int32x2_t __rev2_53; __rev2_53 = __builtin_shufflevector(__s2_53, __s2_53, 1, 0); \
	14790	__ret_53 = __rev0_53 + __rev1_53 * __noswap_splatq_lane_s32(__rev2_53, __p3_53); \
	14791	__ret_53 = __builtin_shufflevector(__ret_53, __ret_53, 3, 2, 1, 0); \
14810	14792	__ret_53; \
14811	14793	})
14812	14794	#endif
14813	14795
14814	14796	#ifdef __LITTLE_ENDIAN__
14815		#define vmla_lane_u16(__p0_54, __p1_54, __p2_54, __p3_54) __extension__ ({ \
14816		uint16x4_t __ret_54; \
14817		uint16x4_t __s0_54 = __p0_54; \
14818		uint16x4_t __s1_54 = __p1_54; \
14819		uint16x4_t __s2_54 = __p2_54; \
14820		__ret_54 = __s0_54 + __s1_54 * splat_lane_u16(__s2_54, __p3_54); \
	14797	#define vmlaq_lane_s16(__p0_54, __p1_54, __p2_54, __p3_54) __extension__ ({ \
	14798	int16x8_t __ret_54; \
	14799	int16x8_t __s0_54 = __p0_54; \
	14800	int16x8_t __s1_54 = __p1_54; \
	14801	int16x4_t __s2_54 = __p2_54; \
	14802	__ret_54 = __s0_54 + __s1_54 * splatq_lane_s16(__s2_54, __p3_54); \
14821	14803	__ret_54; \
14822	14804	})
14823	14805	#else
14824		#define vmla_lane_u16(__p0_55, __p1_55, __p2_55, __p3_55) __extension__ ({ \
14825		uint16x4_t __ret_55; \
14826		uint16x4_t __s0_55 = __p0_55; \
14827		uint16x4_t __s1_55 = __p1_55; \
14828		uint16x4_t __s2_55 = __p2_55; \
14829		uint16x4_t __rev0_55; __rev0_55 = __builtin_shufflevector(__s0_55, __s0_55, 3, 2, 1, 0); \
14830		uint16x4_t __rev1_55; __rev1_55 = __builtin_shufflevector(__s1_55, __s1_55, 3, 2, 1, 0); \
14831		uint16x4_t __rev2_55; __rev2_55 = __builtin_shufflevector(__s2_55, __s2_55, 3, 2, 1, 0); \
14832		__ret_55 = __rev0_55 + __rev1_55 * __noswap_splat_lane_u16(__rev2_55, __p3_55); \
14833		__ret_55 = __builtin_shufflevector(__ret_55, __ret_55, 3, 2, 1, 0); \
	14806	#define vmlaq_lane_s16(__p0_55, __p1_55, __p2_55, __p3_55) __extension__ ({ \
	14807	int16x8_t __ret_55; \
	14808	int16x8_t __s0_55 = __p0_55; \
	14809	int16x8_t __s1_55 = __p1_55; \
	14810	int16x4_t __s2_55 = __p2_55; \
	14811	int16x8_t __rev0_55; __rev0_55 = __builtin_shufflevector(__s0_55, __s0_55, 7, 6, 5, 4, 3, 2, 1, 0); \
	14812	int16x8_t __rev1_55; __rev1_55 = __builtin_shufflevector(__s1_55, __s1_55, 7, 6, 5, 4, 3, 2, 1, 0); \
	14813	int16x4_t __rev2_55; __rev2_55 = __builtin_shufflevector(__s2_55, __s2_55, 3, 2, 1, 0); \
	14814	__ret_55 = __rev0_55 + __rev1_55 * __noswap_splatq_lane_s16(__rev2_55, __p3_55); \
	14815	__ret_55 = __builtin_shufflevector(__ret_55, __ret_55, 7, 6, 5, 4, 3, 2, 1, 0); \
14834	14816	__ret_55; \
14835	14817	})
14836	14818	#endif
14837	14819
14838	14820	#ifdef __LITTLE_ENDIAN__
14839		#define vmla_lane_f32(__p0_56, __p1_56, __p2_56, __p3_56) __extension__ ({ \
14840		float32x2_t __ret_56; \
14841		float32x2_t __s0_56 = __p0_56; \
14842		float32x2_t __s1_56 = __p1_56; \
14843		float32x2_t __s2_56 = __p2_56; \
14844		__ret_56 = __s0_56 + __s1_56 * splat_lane_f32(__s2_56, __p3_56); \
	14821	#define vmla_lane_u32(__p0_56, __p1_56, __p2_56, __p3_56) __extension__ ({ \
	14822	uint32x2_t __ret_56; \
	14823	uint32x2_t __s0_56 = __p0_56; \
	14824	uint32x2_t __s1_56 = __p1_56; \
	14825	uint32x2_t __s2_56 = __p2_56; \
	14826	__ret_56 = __s0_56 + __s1_56 * splat_lane_u32(__s2_56, __p3_56); \
14845	14827	__ret_56; \
14846	14828	})
14847	14829	#else
14848		#define vmla_lane_f32(__p0_57, __p1_57, __p2_57, __p3_57) __extension__ ({ \
14849		float32x2_t __ret_57; \
14850		float32x2_t __s0_57 = __p0_57; \
14851		float32x2_t __s1_57 = __p1_57; \
14852		float32x2_t __s2_57 = __p2_57; \
14853		float32x2_t __rev0_57; __rev0_57 = __builtin_shufflevector(__s0_57, __s0_57, 1, 0); \
14854		float32x2_t __rev1_57; __rev1_57 = __builtin_shufflevector(__s1_57, __s1_57, 1, 0); \
14855		float32x2_t __rev2_57; __rev2_57 = __builtin_shufflevector(__s2_57, __s2_57, 1, 0); \
14856		__ret_57 = __rev0_57 + __rev1_57 * __noswap_splat_lane_f32(__rev2_57, __p3_57); \
	14830	#define vmla_lane_u32(__p0_57, __p1_57, __p2_57, __p3_57) __extension__ ({ \
	14831	uint32x2_t __ret_57; \
	14832	uint32x2_t __s0_57 = __p0_57; \
	14833	uint32x2_t __s1_57 = __p1_57; \
	14834	uint32x2_t __s2_57 = __p2_57; \
	14835	uint32x2_t __rev0_57; __rev0_57 = __builtin_shufflevector(__s0_57, __s0_57, 1, 0); \
	14836	uint32x2_t __rev1_57; __rev1_57 = __builtin_shufflevector(__s1_57, __s1_57, 1, 0); \
	14837	uint32x2_t __rev2_57; __rev2_57 = __builtin_shufflevector(__s2_57, __s2_57, 1, 0); \
	14838	__ret_57 = __rev0_57 + __rev1_57 * __noswap_splat_lane_u32(__rev2_57, __p3_57); \
14857	14839	__ret_57 = __builtin_shufflevector(__ret_57, __ret_57, 1, 0); \
14858	14840	__ret_57; \
14859	14841	})
14860	14842	#endif
14861	14843
14862	14844	#ifdef __LITTLE_ENDIAN__
14863		#define vmla_lane_s32(__p0_58, __p1_58, __p2_58, __p3_58) __extension__ ({ \
14864		int32x2_t __ret_58; \
14865		int32x2_t __s0_58 = __p0_58; \
14866		int32x2_t __s1_58 = __p1_58; \
14867		int32x2_t __s2_58 = __p2_58; \
14868		__ret_58 = __s0_58 + __s1_58 * splat_lane_s32(__s2_58, __p3_58); \
	14845	#define vmla_lane_u16(__p0_58, __p1_58, __p2_58, __p3_58) __extension__ ({ \
	14846	uint16x4_t __ret_58; \
	14847	uint16x4_t __s0_58 = __p0_58; \
	14848	uint16x4_t __s1_58 = __p1_58; \
	14849	uint16x4_t __s2_58 = __p2_58; \
	14850	__ret_58 = __s0_58 + __s1_58 * splat_lane_u16(__s2_58, __p3_58); \
14869	14851	__ret_58; \
14870	14852	})
14871	14853	#else
14872		#define vmla_lane_s32(__p0_59, __p1_59, __p2_59, __p3_59) __extension__ ({ \
14873		int32x2_t __ret_59; \
14874		int32x2_t __s0_59 = __p0_59; \
14875		int32x2_t __s1_59 = __p1_59; \
14876		int32x2_t __s2_59 = __p2_59; \
14877		int32x2_t __rev0_59; __rev0_59 = __builtin_shufflevector(__s0_59, __s0_59, 1, 0); \
14878		int32x2_t __rev1_59; __rev1_59 = __builtin_shufflevector(__s1_59, __s1_59, 1, 0); \
14879		int32x2_t __rev2_59; __rev2_59 = __builtin_shufflevector(__s2_59, __s2_59, 1, 0); \
14880		__ret_59 = __rev0_59 + __rev1_59 * __noswap_splat_lane_s32(__rev2_59, __p3_59); \
14881		__ret_59 = __builtin_shufflevector(__ret_59, __ret_59, 1, 0); \
	14854	#define vmla_lane_u16(__p0_59, __p1_59, __p2_59, __p3_59) __extension__ ({ \
	14855	uint16x4_t __ret_59; \
	14856	uint16x4_t __s0_59 = __p0_59; \
	14857	uint16x4_t __s1_59 = __p1_59; \
	14858	uint16x4_t __s2_59 = __p2_59; \
	14859	uint16x4_t __rev0_59; __rev0_59 = __builtin_shufflevector(__s0_59, __s0_59, 3, 2, 1, 0); \
	14860	uint16x4_t __rev1_59; __rev1_59 = __builtin_shufflevector(__s1_59, __s1_59, 3, 2, 1, 0); \
	14861	uint16x4_t __rev2_59; __rev2_59 = __builtin_shufflevector(__s2_59, __s2_59, 3, 2, 1, 0); \
	14862	__ret_59 = __rev0_59 + __rev1_59 * __noswap_splat_lane_u16(__rev2_59, __p3_59); \
	14863	__ret_59 = __builtin_shufflevector(__ret_59, __ret_59, 3, 2, 1, 0); \
14882	14864	__ret_59; \
14883	14865	})
14884	14866	#endif
14885	14867
14886	14868	#ifdef __LITTLE_ENDIAN__
14887		#define vmla_lane_s16(__p0_60, __p1_60, __p2_60, __p3_60) __extension__ ({ \
14888		int16x4_t __ret_60; \
14889		int16x4_t __s0_60 = __p0_60; \
14890		int16x4_t __s1_60 = __p1_60; \
14891		int16x4_t __s2_60 = __p2_60; \
14892		__ret_60 = __s0_60 + __s1_60 * splat_lane_s16(__s2_60, __p3_60); \
	14869	#define vmla_lane_f32(__p0_60, __p1_60, __p2_60, __p3_60) __extension__ ({ \
	14870	float32x2_t __ret_60; \
	14871	float32x2_t __s0_60 = __p0_60; \
	14872	float32x2_t __s1_60 = __p1_60; \
	14873	float32x2_t __s2_60 = __p2_60; \
	14874	__ret_60 = __s0_60 + __s1_60 * splat_lane_f32(__s2_60, __p3_60); \
14893	14875	__ret_60; \
14894	14876	})
14895	14877	#else
14896		#define vmla_lane_s16(__p0_61, __p1_61, __p2_61, __p3_61) __extension__ ({ \
14897		int16x4_t __ret_61; \
14898		int16x4_t __s0_61 = __p0_61; \
14899		int16x4_t __s1_61 = __p1_61; \
14900		int16x4_t __s2_61 = __p2_61; \
14901		int16x4_t __rev0_61; __rev0_61 = __builtin_shufflevector(__s0_61, __s0_61, 3, 2, 1, 0); \
14902		int16x4_t __rev1_61; __rev1_61 = __builtin_shufflevector(__s1_61, __s1_61, 3, 2, 1, 0); \
14903		int16x4_t __rev2_61; __rev2_61 = __builtin_shufflevector(__s2_61, __s2_61, 3, 2, 1, 0); \
14904		__ret_61 = __rev0_61 + __rev1_61 * __noswap_splat_lane_s16(__rev2_61, __p3_61); \
14905		__ret_61 = __builtin_shufflevector(__ret_61, __ret_61, 3, 2, 1, 0); \
	14878	#define vmla_lane_f32(__p0_61, __p1_61, __p2_61, __p3_61) __extension__ ({ \
	14879	float32x2_t __ret_61; \
	14880	float32x2_t __s0_61 = __p0_61; \
	14881	float32x2_t __s1_61 = __p1_61; \
	14882	float32x2_t __s2_61 = __p2_61; \
	14883	float32x2_t __rev0_61; __rev0_61 = __builtin_shufflevector(__s0_61, __s0_61, 1, 0); \
	14884	float32x2_t __rev1_61; __rev1_61 = __builtin_shufflevector(__s1_61, __s1_61, 1, 0); \
	14885	float32x2_t __rev2_61; __rev2_61 = __builtin_shufflevector(__s2_61, __s2_61, 1, 0); \
	14886	__ret_61 = __rev0_61 + __rev1_61 * __noswap_splat_lane_f32(__rev2_61, __p3_61); \
	14887	__ret_61 = __builtin_shufflevector(__ret_61, __ret_61, 1, 0); \
14906	14888	__ret_61; \
14907	14889	})
14908	14890	#endif
14909	14891
	14892	#ifdef __LITTLE_ENDIAN__
	14893	#define vmla_lane_s32(__p0_62, __p1_62, __p2_62, __p3_62) __extension__ ({ \
	14894	int32x2_t __ret_62; \
	14895	int32x2_t __s0_62 = __p0_62; \
	14896	int32x2_t __s1_62 = __p1_62; \
	14897	int32x2_t __s2_62 = __p2_62; \
	14898	__ret_62 = __s0_62 + __s1_62 * splat_lane_s32(__s2_62, __p3_62); \
	14899	__ret_62; \
	14900	})
	14901	#else
	14902	#define vmla_lane_s32(__p0_63, __p1_63, __p2_63, __p3_63) __extension__ ({ \
	14903	int32x2_t __ret_63; \
	14904	int32x2_t __s0_63 = __p0_63; \
	14905	int32x2_t __s1_63 = __p1_63; \
	14906	int32x2_t __s2_63 = __p2_63; \
	14907	int32x2_t __rev0_63; __rev0_63 = __builtin_shufflevector(__s0_63, __s0_63, 1, 0); \
	14908	int32x2_t __rev1_63; __rev1_63 = __builtin_shufflevector(__s1_63, __s1_63, 1, 0); \
	14909	int32x2_t __rev2_63; __rev2_63 = __builtin_shufflevector(__s2_63, __s2_63, 1, 0); \
	14910	__ret_63 = __rev0_63 + __rev1_63 * __noswap_splat_lane_s32(__rev2_63, __p3_63); \
	14911	__ret_63 = __builtin_shufflevector(__ret_63, __ret_63, 1, 0); \
	14912	__ret_63; \
	14913	})
	14914	#endif
	14915
	14916	#ifdef __LITTLE_ENDIAN__
	14917	#define vmla_lane_s16(__p0_64, __p1_64, __p2_64, __p3_64) __extension__ ({ \
	14918	int16x4_t __ret_64; \
	14919	int16x4_t __s0_64 = __p0_64; \
	14920	int16x4_t __s1_64 = __p1_64; \
	14921	int16x4_t __s2_64 = __p2_64; \
	14922	__ret_64 = __s0_64 + __s1_64 * splat_lane_s16(__s2_64, __p3_64); \
	14923	__ret_64; \
	14924	})
	14925	#else
	14926	#define vmla_lane_s16(__p0_65, __p1_65, __p2_65, __p3_65) __extension__ ({ \
	14927	int16x4_t __ret_65; \
	14928	int16x4_t __s0_65 = __p0_65; \
	14929	int16x4_t __s1_65 = __p1_65; \
	14930	int16x4_t __s2_65 = __p2_65; \
	14931	int16x4_t __rev0_65; __rev0_65 = __builtin_shufflevector(__s0_65, __s0_65, 3, 2, 1, 0); \
	14932	int16x4_t __rev1_65; __rev1_65 = __builtin_shufflevector(__s1_65, __s1_65, 3, 2, 1, 0); \
	14933	int16x4_t __rev2_65; __rev2_65 = __builtin_shufflevector(__s2_65, __s2_65, 3, 2, 1, 0); \
	14934	__ret_65 = __rev0_65 + __rev1_65 * __noswap_splat_lane_s16(__rev2_65, __p3_65); \
	14935	__ret_65 = __builtin_shufflevector(__ret_65, __ret_65, 3, 2, 1, 0); \
	14936	__ret_65; \
	14937	})
	14938	#endif
	14939
14910	14940	#ifdef __LITTLE_ENDIAN__
14911	14941	__ai uint32x4_t vmlaq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
14912	14942	uint32x4_t __ret;
...	...	@@ -15330,245 +15360,245 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
15330	15360	#endif
15331	15361
15332	15362	#ifdef __LITTLE_ENDIAN__
15333		#define vmlsq_lane_u32(__p0_62, __p1_62, __p2_62, __p3_62) __extension__ ({ \
15334		uint32x4_t __ret_62; \
15335		uint32x4_t __s0_62 = __p0_62; \
15336		uint32x4_t __s1_62 = __p1_62; \
15337		uint32x2_t __s2_62 = __p2_62; \
15338		__ret_62 = __s0_62 - __s1_62 * splatq_lane_u32(__s2_62, __p3_62); \
15339		__ret_62; \
15340		})
15341		#else
15342		#define vmlsq_lane_u32(__p0_63, __p1_63, __p2_63, __p3_63) __extension__ ({ \
15343		uint32x4_t __ret_63; \
15344		uint32x4_t __s0_63 = __p0_63; \
15345		uint32x4_t __s1_63 = __p1_63; \
15346		uint32x2_t __s2_63 = __p2_63; \
15347		uint32x4_t __rev0_63; __rev0_63 = __builtin_shufflevector(__s0_63, __s0_63, 3, 2, 1, 0); \
15348		uint32x4_t __rev1_63; __rev1_63 = __builtin_shufflevector(__s1_63, __s1_63, 3, 2, 1, 0); \
15349		uint32x2_t __rev2_63; __rev2_63 = __builtin_shufflevector(__s2_63, __s2_63, 1, 0); \
15350		__ret_63 = __rev0_63 - __rev1_63 * __noswap_splatq_lane_u32(__rev2_63, __p3_63); \
15351		__ret_63 = __builtin_shufflevector(__ret_63, __ret_63, 3, 2, 1, 0); \
15352		__ret_63; \
15353		})
15354		#endif
15355
15356		#ifdef __LITTLE_ENDIAN__
15357		#define vmlsq_lane_u16(__p0_64, __p1_64, __p2_64, __p3_64) __extension__ ({ \
15358		uint16x8_t __ret_64; \
15359		uint16x8_t __s0_64 = __p0_64; \
15360		uint16x8_t __s1_64 = __p1_64; \
15361		uint16x4_t __s2_64 = __p2_64; \
15362		__ret_64 = __s0_64 - __s1_64 * splatq_lane_u16(__s2_64, __p3_64); \
15363		__ret_64; \
15364		})
15365		#else
15366		#define vmlsq_lane_u16(__p0_65, __p1_65, __p2_65, __p3_65) __extension__ ({ \
15367		uint16x8_t __ret_65; \
15368		uint16x8_t __s0_65 = __p0_65; \
15369		uint16x8_t __s1_65 = __p1_65; \
15370		uint16x4_t __s2_65 = __p2_65; \
15371		uint16x8_t __rev0_65; __rev0_65 = __builtin_shufflevector(__s0_65, __s0_65, 7, 6, 5, 4, 3, 2, 1, 0); \
15372		uint16x8_t __rev1_65; __rev1_65 = __builtin_shufflevector(__s1_65, __s1_65, 7, 6, 5, 4, 3, 2, 1, 0); \
15373		uint16x4_t __rev2_65; __rev2_65 = __builtin_shufflevector(__s2_65, __s2_65, 3, 2, 1, 0); \
15374		__ret_65 = __rev0_65 - __rev1_65 * __noswap_splatq_lane_u16(__rev2_65, __p3_65); \
15375		__ret_65 = __builtin_shufflevector(__ret_65, __ret_65, 7, 6, 5, 4, 3, 2, 1, 0); \
15376		__ret_65; \
15377		})
15378		#endif
15379
15380		#ifdef __LITTLE_ENDIAN__
15381		#define vmlsq_lane_f32(__p0_66, __p1_66, __p2_66, __p3_66) __extension__ ({ \
15382		float32x4_t __ret_66; \
15383		float32x4_t __s0_66 = __p0_66; \
15384		float32x4_t __s1_66 = __p1_66; \
15385		float32x2_t __s2_66 = __p2_66; \
15386		__ret_66 = __s0_66 - __s1_66 * splatq_lane_f32(__s2_66, __p3_66); \
	15363	#define vmlsq_lane_u32(__p0_66, __p1_66, __p2_66, __p3_66) __extension__ ({ \
	15364	uint32x4_t __ret_66; \
	15365	uint32x4_t __s0_66 = __p0_66; \
	15366	uint32x4_t __s1_66 = __p1_66; \
	15367	uint32x2_t __s2_66 = __p2_66; \
	15368	__ret_66 = __s0_66 - __s1_66 * splatq_lane_u32(__s2_66, __p3_66); \
15387	15369	__ret_66; \
15388	15370	})
15389	15371	#else
15390		#define vmlsq_lane_f32(__p0_67, __p1_67, __p2_67, __p3_67) __extension__ ({ \
15391		float32x4_t __ret_67; \
15392		float32x4_t __s0_67 = __p0_67; \
15393		float32x4_t __s1_67 = __p1_67; \
15394		float32x2_t __s2_67 = __p2_67; \
15395		float32x4_t __rev0_67; __rev0_67 = __builtin_shufflevector(__s0_67, __s0_67, 3, 2, 1, 0); \
15396		float32x4_t __rev1_67; __rev1_67 = __builtin_shufflevector(__s1_67, __s1_67, 3, 2, 1, 0); \
15397		float32x2_t __rev2_67; __rev2_67 = __builtin_shufflevector(__s2_67, __s2_67, 1, 0); \
15398		__ret_67 = __rev0_67 - __rev1_67 * __noswap_splatq_lane_f32(__rev2_67, __p3_67); \
	15372	#define vmlsq_lane_u32(__p0_67, __p1_67, __p2_67, __p3_67) __extension__ ({ \
	15373	uint32x4_t __ret_67; \
	15374	uint32x4_t __s0_67 = __p0_67; \
	15375	uint32x4_t __s1_67 = __p1_67; \
	15376	uint32x2_t __s2_67 = __p2_67; \
	15377	uint32x4_t __rev0_67; __rev0_67 = __builtin_shufflevector(__s0_67, __s0_67, 3, 2, 1, 0); \
	15378	uint32x4_t __rev1_67; __rev1_67 = __builtin_shufflevector(__s1_67, __s1_67, 3, 2, 1, 0); \
	15379	uint32x2_t __rev2_67; __rev2_67 = __builtin_shufflevector(__s2_67, __s2_67, 1, 0); \
	15380	__ret_67 = __rev0_67 - __rev1_67 * __noswap_splatq_lane_u32(__rev2_67, __p3_67); \
15399	15381	__ret_67 = __builtin_shufflevector(__ret_67, __ret_67, 3, 2, 1, 0); \
15400	15382	__ret_67; \
15401	15383	})
15402	15384	#endif
15403	15385
15404	15386	#ifdef __LITTLE_ENDIAN__
15405		#define vmlsq_lane_s32(__p0_68, __p1_68, __p2_68, __p3_68) __extension__ ({ \
15406		int32x4_t __ret_68; \
15407		int32x4_t __s0_68 = __p0_68; \
15408		int32x4_t __s1_68 = __p1_68; \
15409		int32x2_t __s2_68 = __p2_68; \
15410		__ret_68 = __s0_68 - __s1_68 * splatq_lane_s32(__s2_68, __p3_68); \
	15387	#define vmlsq_lane_u16(__p0_68, __p1_68, __p2_68, __p3_68) __extension__ ({ \
	15388	uint16x8_t __ret_68; \
	15389	uint16x8_t __s0_68 = __p0_68; \
	15390	uint16x8_t __s1_68 = __p1_68; \
	15391	uint16x4_t __s2_68 = __p2_68; \
	15392	__ret_68 = __s0_68 - __s1_68 * splatq_lane_u16(__s2_68, __p3_68); \
15411	15393	__ret_68; \
15412	15394	})
15413	15395	#else
15414		#define vmlsq_lane_s32(__p0_69, __p1_69, __p2_69, __p3_69) __extension__ ({ \
15415		int32x4_t __ret_69; \
15416		int32x4_t __s0_69 = __p0_69; \
15417		int32x4_t __s1_69 = __p1_69; \
15418		int32x2_t __s2_69 = __p2_69; \
15419		int32x4_t __rev0_69; __rev0_69 = __builtin_shufflevector(__s0_69, __s0_69, 3, 2, 1, 0); \
15420		int32x4_t __rev1_69; __rev1_69 = __builtin_shufflevector(__s1_69, __s1_69, 3, 2, 1, 0); \
15421		int32x2_t __rev2_69; __rev2_69 = __builtin_shufflevector(__s2_69, __s2_69, 1, 0); \
15422		__ret_69 = __rev0_69 - __rev1_69 * __noswap_splatq_lane_s32(__rev2_69, __p3_69); \
15423		__ret_69 = __builtin_shufflevector(__ret_69, __ret_69, 3, 2, 1, 0); \
	15396	#define vmlsq_lane_u16(__p0_69, __p1_69, __p2_69, __p3_69) __extension__ ({ \
	15397	uint16x8_t __ret_69; \
	15398	uint16x8_t __s0_69 = __p0_69; \
	15399	uint16x8_t __s1_69 = __p1_69; \
	15400	uint16x4_t __s2_69 = __p2_69; \
	15401	uint16x8_t __rev0_69; __rev0_69 = __builtin_shufflevector(__s0_69, __s0_69, 7, 6, 5, 4, 3, 2, 1, 0); \
	15402	uint16x8_t __rev1_69; __rev1_69 = __builtin_shufflevector(__s1_69, __s1_69, 7, 6, 5, 4, 3, 2, 1, 0); \
	15403	uint16x4_t __rev2_69; __rev2_69 = __builtin_shufflevector(__s2_69, __s2_69, 3, 2, 1, 0); \
	15404	__ret_69 = __rev0_69 - __rev1_69 * __noswap_splatq_lane_u16(__rev2_69, __p3_69); \
	15405	__ret_69 = __builtin_shufflevector(__ret_69, __ret_69, 7, 6, 5, 4, 3, 2, 1, 0); \
15424	15406	__ret_69; \
15425	15407	})
15426	15408	#endif
15427	15409
15428	15410	#ifdef __LITTLE_ENDIAN__
15429		#define vmlsq_lane_s16(__p0_70, __p1_70, __p2_70, __p3_70) __extension__ ({ \
15430		int16x8_t __ret_70; \
15431		int16x8_t __s0_70 = __p0_70; \
15432		int16x8_t __s1_70 = __p1_70; \
15433		int16x4_t __s2_70 = __p2_70; \
15434		__ret_70 = __s0_70 - __s1_70 * splatq_lane_s16(__s2_70, __p3_70); \
	15411	#define vmlsq_lane_f32(__p0_70, __p1_70, __p2_70, __p3_70) __extension__ ({ \
	15412	float32x4_t __ret_70; \
	15413	float32x4_t __s0_70 = __p0_70; \
	15414	float32x4_t __s1_70 = __p1_70; \
	15415	float32x2_t __s2_70 = __p2_70; \
	15416	__ret_70 = __s0_70 - __s1_70 * splatq_lane_f32(__s2_70, __p3_70); \
15435	15417	__ret_70; \
15436	15418	})
15437	15419	#else
15438		#define vmlsq_lane_s16(__p0_71, __p1_71, __p2_71, __p3_71) __extension__ ({ \
15439		int16x8_t __ret_71; \
15440		int16x8_t __s0_71 = __p0_71; \
15441		int16x8_t __s1_71 = __p1_71; \
15442		int16x4_t __s2_71 = __p2_71; \
15443		int16x8_t __rev0_71; __rev0_71 = __builtin_shufflevector(__s0_71, __s0_71, 7, 6, 5, 4, 3, 2, 1, 0); \
15444		int16x8_t __rev1_71; __rev1_71 = __builtin_shufflevector(__s1_71, __s1_71, 7, 6, 5, 4, 3, 2, 1, 0); \
15445		int16x4_t __rev2_71; __rev2_71 = __builtin_shufflevector(__s2_71, __s2_71, 3, 2, 1, 0); \
15446		__ret_71 = __rev0_71 - __rev1_71 * __noswap_splatq_lane_s16(__rev2_71, __p3_71); \
15447		__ret_71 = __builtin_shufflevector(__ret_71, __ret_71, 7, 6, 5, 4, 3, 2, 1, 0); \
	15420	#define vmlsq_lane_f32(__p0_71, __p1_71, __p2_71, __p3_71) __extension__ ({ \
	15421	float32x4_t __ret_71; \
	15422	float32x4_t __s0_71 = __p0_71; \
	15423	float32x4_t __s1_71 = __p1_71; \
	15424	float32x2_t __s2_71 = __p2_71; \
	15425	float32x4_t __rev0_71; __rev0_71 = __builtin_shufflevector(__s0_71, __s0_71, 3, 2, 1, 0); \
	15426	float32x4_t __rev1_71; __rev1_71 = __builtin_shufflevector(__s1_71, __s1_71, 3, 2, 1, 0); \
	15427	float32x2_t __rev2_71; __rev2_71 = __builtin_shufflevector(__s2_71, __s2_71, 1, 0); \
	15428	__ret_71 = __rev0_71 - __rev1_71 * __noswap_splatq_lane_f32(__rev2_71, __p3_71); \
	15429	__ret_71 = __builtin_shufflevector(__ret_71, __ret_71, 3, 2, 1, 0); \
15448	15430	__ret_71; \
15449	15431	})
15450	15432	#endif
15451	15433
15452	15434	#ifdef __LITTLE_ENDIAN__
15453		#define vmls_lane_u32(__p0_72, __p1_72, __p2_72, __p3_72) __extension__ ({ \
15454		uint32x2_t __ret_72; \
15455		uint32x2_t __s0_72 = __p0_72; \
15456		uint32x2_t __s1_72 = __p1_72; \
15457		uint32x2_t __s2_72 = __p2_72; \
15458		__ret_72 = __s0_72 - __s1_72 * splat_lane_u32(__s2_72, __p3_72); \
	15435	#define vmlsq_lane_s32(__p0_72, __p1_72, __p2_72, __p3_72) __extension__ ({ \
	15436	int32x4_t __ret_72; \
	15437	int32x4_t __s0_72 = __p0_72; \
	15438	int32x4_t __s1_72 = __p1_72; \
	15439	int32x2_t __s2_72 = __p2_72; \
	15440	__ret_72 = __s0_72 - __s1_72 * splatq_lane_s32(__s2_72, __p3_72); \
15459	15441	__ret_72; \
15460	15442	})
15461	15443	#else
15462		#define vmls_lane_u32(__p0_73, __p1_73, __p2_73, __p3_73) __extension__ ({ \
15463		uint32x2_t __ret_73; \
15464		uint32x2_t __s0_73 = __p0_73; \
15465		uint32x2_t __s1_73 = __p1_73; \
15466		uint32x2_t __s2_73 = __p2_73; \
15467		uint32x2_t __rev0_73; __rev0_73 = __builtin_shufflevector(__s0_73, __s0_73, 1, 0); \
15468		uint32x2_t __rev1_73; __rev1_73 = __builtin_shufflevector(__s1_73, __s1_73, 1, 0); \
15469		uint32x2_t __rev2_73; __rev2_73 = __builtin_shufflevector(__s2_73, __s2_73, 1, 0); \
15470		__ret_73 = __rev0_73 - __rev1_73 * __noswap_splat_lane_u32(__rev2_73, __p3_73); \
15471		__ret_73 = __builtin_shufflevector(__ret_73, __ret_73, 1, 0); \
	15444	#define vmlsq_lane_s32(__p0_73, __p1_73, __p2_73, __p3_73) __extension__ ({ \
	15445	int32x4_t __ret_73; \
	15446	int32x4_t __s0_73 = __p0_73; \
	15447	int32x4_t __s1_73 = __p1_73; \
	15448	int32x2_t __s2_73 = __p2_73; \
	15449	int32x4_t __rev0_73; __rev0_73 = __builtin_shufflevector(__s0_73, __s0_73, 3, 2, 1, 0); \
	15450	int32x4_t __rev1_73; __rev1_73 = __builtin_shufflevector(__s1_73, __s1_73, 3, 2, 1, 0); \
	15451	int32x2_t __rev2_73; __rev2_73 = __builtin_shufflevector(__s2_73, __s2_73, 1, 0); \
	15452	__ret_73 = __rev0_73 - __rev1_73 * __noswap_splatq_lane_s32(__rev2_73, __p3_73); \
	15453	__ret_73 = __builtin_shufflevector(__ret_73, __ret_73, 3, 2, 1, 0); \
15472	15454	__ret_73; \
15473	15455	})
15474	15456	#endif
15475	15457
15476	15458	#ifdef __LITTLE_ENDIAN__
15477		#define vmls_lane_u16(__p0_74, __p1_74, __p2_74, __p3_74) __extension__ ({ \
15478		uint16x4_t __ret_74; \
15479		uint16x4_t __s0_74 = __p0_74; \
15480		uint16x4_t __s1_74 = __p1_74; \
15481		uint16x4_t __s2_74 = __p2_74; \
15482		__ret_74 = __s0_74 - __s1_74 * splat_lane_u16(__s2_74, __p3_74); \
	15459	#define vmlsq_lane_s16(__p0_74, __p1_74, __p2_74, __p3_74) __extension__ ({ \
	15460	int16x8_t __ret_74; \
	15461	int16x8_t __s0_74 = __p0_74; \
	15462	int16x8_t __s1_74 = __p1_74; \
	15463	int16x4_t __s2_74 = __p2_74; \
	15464	__ret_74 = __s0_74 - __s1_74 * splatq_lane_s16(__s2_74, __p3_74); \
15483	15465	__ret_74; \
15484	15466	})
15485	15467	#else
15486		#define vmls_lane_u16(__p0_75, __p1_75, __p2_75, __p3_75) __extension__ ({ \
15487		uint16x4_t __ret_75; \
15488		uint16x4_t __s0_75 = __p0_75; \
15489		uint16x4_t __s1_75 = __p1_75; \
15490		uint16x4_t __s2_75 = __p2_75; \
15491		uint16x4_t __rev0_75; __rev0_75 = __builtin_shufflevector(__s0_75, __s0_75, 3, 2, 1, 0); \
15492		uint16x4_t __rev1_75; __rev1_75 = __builtin_shufflevector(__s1_75, __s1_75, 3, 2, 1, 0); \
15493		uint16x4_t __rev2_75; __rev2_75 = __builtin_shufflevector(__s2_75, __s2_75, 3, 2, 1, 0); \
15494		__ret_75 = __rev0_75 - __rev1_75 * __noswap_splat_lane_u16(__rev2_75, __p3_75); \
15495		__ret_75 = __builtin_shufflevector(__ret_75, __ret_75, 3, 2, 1, 0); \
	15468	#define vmlsq_lane_s16(__p0_75, __p1_75, __p2_75, __p3_75) __extension__ ({ \
	15469	int16x8_t __ret_75; \
	15470	int16x8_t __s0_75 = __p0_75; \
	15471	int16x8_t __s1_75 = __p1_75; \
	15472	int16x4_t __s2_75 = __p2_75; \
	15473	int16x8_t __rev0_75; __rev0_75 = __builtin_shufflevector(__s0_75, __s0_75, 7, 6, 5, 4, 3, 2, 1, 0); \
	15474	int16x8_t __rev1_75; __rev1_75 = __builtin_shufflevector(__s1_75, __s1_75, 7, 6, 5, 4, 3, 2, 1, 0); \
	15475	int16x4_t __rev2_75; __rev2_75 = __builtin_shufflevector(__s2_75, __s2_75, 3, 2, 1, 0); \
	15476	__ret_75 = __rev0_75 - __rev1_75 * __noswap_splatq_lane_s16(__rev2_75, __p3_75); \
	15477	__ret_75 = __builtin_shufflevector(__ret_75, __ret_75, 7, 6, 5, 4, 3, 2, 1, 0); \
15496	15478	__ret_75; \
15497	15479	})
15498	15480	#endif
15499	15481
15500	15482	#ifdef __LITTLE_ENDIAN__
15501		#define vmls_lane_f32(__p0_76, __p1_76, __p2_76, __p3_76) __extension__ ({ \
15502		float32x2_t __ret_76; \
15503		float32x2_t __s0_76 = __p0_76; \
15504		float32x2_t __s1_76 = __p1_76; \
15505		float32x2_t __s2_76 = __p2_76; \
15506		__ret_76 = __s0_76 - __s1_76 * splat_lane_f32(__s2_76, __p3_76); \
	15483	#define vmls_lane_u32(__p0_76, __p1_76, __p2_76, __p3_76) __extension__ ({ \
	15484	uint32x2_t __ret_76; \
	15485	uint32x2_t __s0_76 = __p0_76; \
	15486	uint32x2_t __s1_76 = __p1_76; \
	15487	uint32x2_t __s2_76 = __p2_76; \
	15488	__ret_76 = __s0_76 - __s1_76 * splat_lane_u32(__s2_76, __p3_76); \
15507	15489	__ret_76; \
15508	15490	})
15509	15491	#else
15510		#define vmls_lane_f32(__p0_77, __p1_77, __p2_77, __p3_77) __extension__ ({ \
15511		float32x2_t __ret_77; \
15512		float32x2_t __s0_77 = __p0_77; \
15513		float32x2_t __s1_77 = __p1_77; \
15514		float32x2_t __s2_77 = __p2_77; \
15515		float32x2_t __rev0_77; __rev0_77 = __builtin_shufflevector(__s0_77, __s0_77, 1, 0); \
15516		float32x2_t __rev1_77; __rev1_77 = __builtin_shufflevector(__s1_77, __s1_77, 1, 0); \
15517		float32x2_t __rev2_77; __rev2_77 = __builtin_shufflevector(__s2_77, __s2_77, 1, 0); \
15518		__ret_77 = __rev0_77 - __rev1_77 * __noswap_splat_lane_f32(__rev2_77, __p3_77); \
	15492	#define vmls_lane_u32(__p0_77, __p1_77, __p2_77, __p3_77) __extension__ ({ \
	15493	uint32x2_t __ret_77; \
	15494	uint32x2_t __s0_77 = __p0_77; \
	15495	uint32x2_t __s1_77 = __p1_77; \
	15496	uint32x2_t __s2_77 = __p2_77; \
	15497	uint32x2_t __rev0_77; __rev0_77 = __builtin_shufflevector(__s0_77, __s0_77, 1, 0); \
	15498	uint32x2_t __rev1_77; __rev1_77 = __builtin_shufflevector(__s1_77, __s1_77, 1, 0); \
	15499	uint32x2_t __rev2_77; __rev2_77 = __builtin_shufflevector(__s2_77, __s2_77, 1, 0); \
	15500	__ret_77 = __rev0_77 - __rev1_77 * __noswap_splat_lane_u32(__rev2_77, __p3_77); \
15519	15501	__ret_77 = __builtin_shufflevector(__ret_77, __ret_77, 1, 0); \
15520	15502	__ret_77; \
15521	15503	})
15522	15504	#endif
15523	15505
15524	15506	#ifdef __LITTLE_ENDIAN__
15525		#define vmls_lane_s32(__p0_78, __p1_78, __p2_78, __p3_78) __extension__ ({ \
15526		int32x2_t __ret_78; \
15527		int32x2_t __s0_78 = __p0_78; \
15528		int32x2_t __s1_78 = __p1_78; \
15529		int32x2_t __s2_78 = __p2_78; \
15530		__ret_78 = __s0_78 - __s1_78 * splat_lane_s32(__s2_78, __p3_78); \
	15507	#define vmls_lane_u16(__p0_78, __p1_78, __p2_78, __p3_78) __extension__ ({ \
	15508	uint16x4_t __ret_78; \
	15509	uint16x4_t __s0_78 = __p0_78; \
	15510	uint16x4_t __s1_78 = __p1_78; \
	15511	uint16x4_t __s2_78 = __p2_78; \
	15512	__ret_78 = __s0_78 - __s1_78 * splat_lane_u16(__s2_78, __p3_78); \
15531	15513	__ret_78; \
15532	15514	})
15533	15515	#else
15534		#define vmls_lane_s32(__p0_79, __p1_79, __p2_79, __p3_79) __extension__ ({ \
15535		int32x2_t __ret_79; \
15536		int32x2_t __s0_79 = __p0_79; \
15537		int32x2_t __s1_79 = __p1_79; \
15538		int32x2_t __s2_79 = __p2_79; \
15539		int32x2_t __rev0_79; __rev0_79 = __builtin_shufflevector(__s0_79, __s0_79, 1, 0); \
15540		int32x2_t __rev1_79; __rev1_79 = __builtin_shufflevector(__s1_79, __s1_79, 1, 0); \
15541		int32x2_t __rev2_79; __rev2_79 = __builtin_shufflevector(__s2_79, __s2_79, 1, 0); \
15542		__ret_79 = __rev0_79 - __rev1_79 * __noswap_splat_lane_s32(__rev2_79, __p3_79); \
15543		__ret_79 = __builtin_shufflevector(__ret_79, __ret_79, 1, 0); \
	15516	#define vmls_lane_u16(__p0_79, __p1_79, __p2_79, __p3_79) __extension__ ({ \
	15517	uint16x4_t __ret_79; \
	15518	uint16x4_t __s0_79 = __p0_79; \
	15519	uint16x4_t __s1_79 = __p1_79; \
	15520	uint16x4_t __s2_79 = __p2_79; \
	15521	uint16x4_t __rev0_79; __rev0_79 = __builtin_shufflevector(__s0_79, __s0_79, 3, 2, 1, 0); \
	15522	uint16x4_t __rev1_79; __rev1_79 = __builtin_shufflevector(__s1_79, __s1_79, 3, 2, 1, 0); \
	15523	uint16x4_t __rev2_79; __rev2_79 = __builtin_shufflevector(__s2_79, __s2_79, 3, 2, 1, 0); \
	15524	__ret_79 = __rev0_79 - __rev1_79 * __noswap_splat_lane_u16(__rev2_79, __p3_79); \
	15525	__ret_79 = __builtin_shufflevector(__ret_79, __ret_79, 3, 2, 1, 0); \
15544	15526	__ret_79; \
15545	15527	})
15546	15528	#endif
15547	15529
15548	15530	#ifdef __LITTLE_ENDIAN__
15549		#define vmls_lane_s16(__p0_80, __p1_80, __p2_80, __p3_80) __extension__ ({ \
15550		int16x4_t __ret_80; \
15551		int16x4_t __s0_80 = __p0_80; \
15552		int16x4_t __s1_80 = __p1_80; \
15553		int16x4_t __s2_80 = __p2_80; \
15554		__ret_80 = __s0_80 - __s1_80 * splat_lane_s16(__s2_80, __p3_80); \
	15531	#define vmls_lane_f32(__p0_80, __p1_80, __p2_80, __p3_80) __extension__ ({ \
	15532	float32x2_t __ret_80; \
	15533	float32x2_t __s0_80 = __p0_80; \
	15534	float32x2_t __s1_80 = __p1_80; \
	15535	float32x2_t __s2_80 = __p2_80; \
	15536	__ret_80 = __s0_80 - __s1_80 * splat_lane_f32(__s2_80, __p3_80); \
15555	15537	__ret_80; \
15556	15538	})
15557	15539	#else
15558		#define vmls_lane_s16(__p0_81, __p1_81, __p2_81, __p3_81) __extension__ ({ \
15559		int16x4_t __ret_81; \
15560		int16x4_t __s0_81 = __p0_81; \
15561		int16x4_t __s1_81 = __p1_81; \
15562		int16x4_t __s2_81 = __p2_81; \
15563		int16x4_t __rev0_81; __rev0_81 = __builtin_shufflevector(__s0_81, __s0_81, 3, 2, 1, 0); \
15564		int16x4_t __rev1_81; __rev1_81 = __builtin_shufflevector(__s1_81, __s1_81, 3, 2, 1, 0); \
15565		int16x4_t __rev2_81; __rev2_81 = __builtin_shufflevector(__s2_81, __s2_81, 3, 2, 1, 0); \
15566		__ret_81 = __rev0_81 - __rev1_81 * __noswap_splat_lane_s16(__rev2_81, __p3_81); \
15567		__ret_81 = __builtin_shufflevector(__ret_81, __ret_81, 3, 2, 1, 0); \
	15540	#define vmls_lane_f32(__p0_81, __p1_81, __p2_81, __p3_81) __extension__ ({ \
	15541	float32x2_t __ret_81; \
	15542	float32x2_t __s0_81 = __p0_81; \
	15543	float32x2_t __s1_81 = __p1_81; \
	15544	float32x2_t __s2_81 = __p2_81; \
	15545	float32x2_t __rev0_81; __rev0_81 = __builtin_shufflevector(__s0_81, __s0_81, 1, 0); \
	15546	float32x2_t __rev1_81; __rev1_81 = __builtin_shufflevector(__s1_81, __s1_81, 1, 0); \
	15547	float32x2_t __rev2_81; __rev2_81 = __builtin_shufflevector(__s2_81, __s2_81, 1, 0); \
	15548	__ret_81 = __rev0_81 - __rev1_81 * __noswap_splat_lane_f32(__rev2_81, __p3_81); \
	15549	__ret_81 = __builtin_shufflevector(__ret_81, __ret_81, 1, 0); \
15568	15550	__ret_81; \
15569	15551	})
15570	15552	#endif
15571	15553
	15554	#ifdef __LITTLE_ENDIAN__
	15555	#define vmls_lane_s32(__p0_82, __p1_82, __p2_82, __p3_82) __extension__ ({ \
	15556	int32x2_t __ret_82; \
	15557	int32x2_t __s0_82 = __p0_82; \
	15558	int32x2_t __s1_82 = __p1_82; \
	15559	int32x2_t __s2_82 = __p2_82; \
	15560	__ret_82 = __s0_82 - __s1_82 * splat_lane_s32(__s2_82, __p3_82); \
	15561	__ret_82; \
	15562	})
	15563	#else
	15564	#define vmls_lane_s32(__p0_83, __p1_83, __p2_83, __p3_83) __extension__ ({ \
	15565	int32x2_t __ret_83; \
	15566	int32x2_t __s0_83 = __p0_83; \
	15567	int32x2_t __s1_83 = __p1_83; \
	15568	int32x2_t __s2_83 = __p2_83; \
	15569	int32x2_t __rev0_83; __rev0_83 = __builtin_shufflevector(__s0_83, __s0_83, 1, 0); \
	15570	int32x2_t __rev1_83; __rev1_83 = __builtin_shufflevector(__s1_83, __s1_83, 1, 0); \
	15571	int32x2_t __rev2_83; __rev2_83 = __builtin_shufflevector(__s2_83, __s2_83, 1, 0); \
	15572	__ret_83 = __rev0_83 - __rev1_83 * __noswap_splat_lane_s32(__rev2_83, __p3_83); \
	15573	__ret_83 = __builtin_shufflevector(__ret_83, __ret_83, 1, 0); \
	15574	__ret_83; \
	15575	})
	15576	#endif
	15577
	15578	#ifdef __LITTLE_ENDIAN__
	15579	#define vmls_lane_s16(__p0_84, __p1_84, __p2_84, __p3_84) __extension__ ({ \
	15580	int16x4_t __ret_84; \
	15581	int16x4_t __s0_84 = __p0_84; \
	15582	int16x4_t __s1_84 = __p1_84; \
	15583	int16x4_t __s2_84 = __p2_84; \
	15584	__ret_84 = __s0_84 - __s1_84 * splat_lane_s16(__s2_84, __p3_84); \
	15585	__ret_84; \
	15586	})
	15587	#else
	15588	#define vmls_lane_s16(__p0_85, __p1_85, __p2_85, __p3_85) __extension__ ({ \
	15589	int16x4_t __ret_85; \
	15590	int16x4_t __s0_85 = __p0_85; \
	15591	int16x4_t __s1_85 = __p1_85; \
	15592	int16x4_t __s2_85 = __p2_85; \
	15593	int16x4_t __rev0_85; __rev0_85 = __builtin_shufflevector(__s0_85, __s0_85, 3, 2, 1, 0); \
	15594	int16x4_t __rev1_85; __rev1_85 = __builtin_shufflevector(__s1_85, __s1_85, 3, 2, 1, 0); \
	15595	int16x4_t __rev2_85; __rev2_85 = __builtin_shufflevector(__s2_85, __s2_85, 3, 2, 1, 0); \
	15596	__ret_85 = __rev0_85 - __rev1_85 * __noswap_splat_lane_s16(__rev2_85, __p3_85); \
	15597	__ret_85 = __builtin_shufflevector(__ret_85, __ret_85, 3, 2, 1, 0); \
	15598	__ret_85; \
	15599	})
	15600	#endif
	15601
15572	15602	#ifdef __LITTLE_ENDIAN__
15573	15603	__ai uint32x4_t vmlsq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
15574	15604	uint32x4_t __ret;
...	...	@@ -16608,215 +16638,215 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
16608	16638	#endif
16609	16639
16610	16640	#ifdef __LITTLE_ENDIAN__
16611		#define vmulq_lane_u32(__p0_82, __p1_82, __p2_82) __extension__ ({ \
16612		uint32x4_t __ret_82; \
16613		uint32x4_t __s0_82 = __p0_82; \
16614		uint32x2_t __s1_82 = __p1_82; \
16615		__ret_82 = __s0_82 * splatq_lane_u32(__s1_82, __p2_82); \
16616		__ret_82; \
16617		})
16618		#else
16619		#define vmulq_lane_u32(__p0_83, __p1_83, __p2_83) __extension__ ({ \
16620		uint32x4_t __ret_83; \
16621		uint32x4_t __s0_83 = __p0_83; \
16622		uint32x2_t __s1_83 = __p1_83; \
16623		uint32x4_t __rev0_83; __rev0_83 = __builtin_shufflevector(__s0_83, __s0_83, 3, 2, 1, 0); \
16624		uint32x2_t __rev1_83; __rev1_83 = __builtin_shufflevector(__s1_83, __s1_83, 1, 0); \
16625		__ret_83 = __rev0_83 * __noswap_splatq_lane_u32(__rev1_83, __p2_83); \
16626		__ret_83 = __builtin_shufflevector(__ret_83, __ret_83, 3, 2, 1, 0); \
16627		__ret_83; \
16628		})
16629		#endif
16630
16631		#ifdef __LITTLE_ENDIAN__
16632		#define vmulq_lane_u16(__p0_84, __p1_84, __p2_84) __extension__ ({ \
16633		uint16x8_t __ret_84; \
16634		uint16x8_t __s0_84 = __p0_84; \
16635		uint16x4_t __s1_84 = __p1_84; \
16636		__ret_84 = __s0_84 * splatq_lane_u16(__s1_84, __p2_84); \
16637		__ret_84; \
16638		})
16639		#else
16640		#define vmulq_lane_u16(__p0_85, __p1_85, __p2_85) __extension__ ({ \
16641		uint16x8_t __ret_85; \
16642		uint16x8_t __s0_85 = __p0_85; \
16643		uint16x4_t __s1_85 = __p1_85; \
16644		uint16x8_t __rev0_85; __rev0_85 = __builtin_shufflevector(__s0_85, __s0_85, 7, 6, 5, 4, 3, 2, 1, 0); \
16645		uint16x4_t __rev1_85; __rev1_85 = __builtin_shufflevector(__s1_85, __s1_85, 3, 2, 1, 0); \
16646		__ret_85 = __rev0_85 * __noswap_splatq_lane_u16(__rev1_85, __p2_85); \
16647		__ret_85 = __builtin_shufflevector(__ret_85, __ret_85, 7, 6, 5, 4, 3, 2, 1, 0); \
16648		__ret_85; \
16649		})
16650		#endif
16651
16652		#ifdef __LITTLE_ENDIAN__
16653		#define vmulq_lane_f32(__p0_86, __p1_86, __p2_86) __extension__ ({ \
16654		float32x4_t __ret_86; \
16655		float32x4_t __s0_86 = __p0_86; \
16656		float32x2_t __s1_86 = __p1_86; \
16657		__ret_86 = __s0_86 * splatq_lane_f32(__s1_86, __p2_86); \
	16641	#define vmulq_lane_u32(__p0_86, __p1_86, __p2_86) __extension__ ({ \
	16642	uint32x4_t __ret_86; \
	16643	uint32x4_t __s0_86 = __p0_86; \
	16644	uint32x2_t __s1_86 = __p1_86; \
	16645	__ret_86 = __s0_86 * splatq_lane_u32(__s1_86, __p2_86); \
16658	16646	__ret_86; \
16659	16647	})
16660	16648	#else
16661		#define vmulq_lane_f32(__p0_87, __p1_87, __p2_87) __extension__ ({ \
16662		float32x4_t __ret_87; \
16663		float32x4_t __s0_87 = __p0_87; \
16664		float32x2_t __s1_87 = __p1_87; \
16665		float32x4_t __rev0_87; __rev0_87 = __builtin_shufflevector(__s0_87, __s0_87, 3, 2, 1, 0); \
16666		float32x2_t __rev1_87; __rev1_87 = __builtin_shufflevector(__s1_87, __s1_87, 1, 0); \
16667		__ret_87 = __rev0_87 * __noswap_splatq_lane_f32(__rev1_87, __p2_87); \
	16649	#define vmulq_lane_u32(__p0_87, __p1_87, __p2_87) __extension__ ({ \
	16650	uint32x4_t __ret_87; \
	16651	uint32x4_t __s0_87 = __p0_87; \
	16652	uint32x2_t __s1_87 = __p1_87; \
	16653	uint32x4_t __rev0_87; __rev0_87 = __builtin_shufflevector(__s0_87, __s0_87, 3, 2, 1, 0); \
	16654	uint32x2_t __rev1_87; __rev1_87 = __builtin_shufflevector(__s1_87, __s1_87, 1, 0); \
	16655	__ret_87 = __rev0_87 * __noswap_splatq_lane_u32(__rev1_87, __p2_87); \
16668	16656	__ret_87 = __builtin_shufflevector(__ret_87, __ret_87, 3, 2, 1, 0); \
16669	16657	__ret_87; \
16670	16658	})
16671	16659	#endif
16672	16660
16673	16661	#ifdef __LITTLE_ENDIAN__
16674		#define vmulq_lane_s32(__p0_88, __p1_88, __p2_88) __extension__ ({ \
16675		int32x4_t __ret_88; \
16676		int32x4_t __s0_88 = __p0_88; \
16677		int32x2_t __s1_88 = __p1_88; \
16678		__ret_88 = __s0_88 * splatq_lane_s32(__s1_88, __p2_88); \
	16662	#define vmulq_lane_u16(__p0_88, __p1_88, __p2_88) __extension__ ({ \
	16663	uint16x8_t __ret_88; \
	16664	uint16x8_t __s0_88 = __p0_88; \
	16665	uint16x4_t __s1_88 = __p1_88; \
	16666	__ret_88 = __s0_88 * splatq_lane_u16(__s1_88, __p2_88); \
16679	16667	__ret_88; \
16680	16668	})
16681	16669	#else
16682		#define vmulq_lane_s32(__p0_89, __p1_89, __p2_89) __extension__ ({ \
16683		int32x4_t __ret_89; \
16684		int32x4_t __s0_89 = __p0_89; \
16685		int32x2_t __s1_89 = __p1_89; \
16686		int32x4_t __rev0_89; __rev0_89 = __builtin_shufflevector(__s0_89, __s0_89, 3, 2, 1, 0); \
16687		int32x2_t __rev1_89; __rev1_89 = __builtin_shufflevector(__s1_89, __s1_89, 1, 0); \
16688		__ret_89 = __rev0_89 * __noswap_splatq_lane_s32(__rev1_89, __p2_89); \
16689		__ret_89 = __builtin_shufflevector(__ret_89, __ret_89, 3, 2, 1, 0); \
	16670	#define vmulq_lane_u16(__p0_89, __p1_89, __p2_89) __extension__ ({ \
	16671	uint16x8_t __ret_89; \
	16672	uint16x8_t __s0_89 = __p0_89; \
	16673	uint16x4_t __s1_89 = __p1_89; \
	16674	uint16x8_t __rev0_89; __rev0_89 = __builtin_shufflevector(__s0_89, __s0_89, 7, 6, 5, 4, 3, 2, 1, 0); \
	16675	uint16x4_t __rev1_89; __rev1_89 = __builtin_shufflevector(__s1_89, __s1_89, 3, 2, 1, 0); \
	16676	__ret_89 = __rev0_89 * __noswap_splatq_lane_u16(__rev1_89, __p2_89); \
	16677	__ret_89 = __builtin_shufflevector(__ret_89, __ret_89, 7, 6, 5, 4, 3, 2, 1, 0); \
16690	16678	__ret_89; \
16691	16679	})
16692	16680	#endif
16693	16681
16694	16682	#ifdef __LITTLE_ENDIAN__
16695		#define vmulq_lane_s16(__p0_90, __p1_90, __p2_90) __extension__ ({ \
16696		int16x8_t __ret_90; \
16697		int16x8_t __s0_90 = __p0_90; \
16698		int16x4_t __s1_90 = __p1_90; \
16699		__ret_90 = __s0_90 * splatq_lane_s16(__s1_90, __p2_90); \
	16683	#define vmulq_lane_f32(__p0_90, __p1_90, __p2_90) __extension__ ({ \
	16684	float32x4_t __ret_90; \
	16685	float32x4_t __s0_90 = __p0_90; \
	16686	float32x2_t __s1_90 = __p1_90; \
	16687	__ret_90 = __s0_90 * splatq_lane_f32(__s1_90, __p2_90); \
16700	16688	__ret_90; \
16701	16689	})
16702	16690	#else
16703		#define vmulq_lane_s16(__p0_91, __p1_91, __p2_91) __extension__ ({ \
16704		int16x8_t __ret_91; \
16705		int16x8_t __s0_91 = __p0_91; \
16706		int16x4_t __s1_91 = __p1_91; \
16707		int16x8_t __rev0_91; __rev0_91 = __builtin_shufflevector(__s0_91, __s0_91, 7, 6, 5, 4, 3, 2, 1, 0); \
16708		int16x4_t __rev1_91; __rev1_91 = __builtin_shufflevector(__s1_91, __s1_91, 3, 2, 1, 0); \
16709		__ret_91 = __rev0_91 * __noswap_splatq_lane_s16(__rev1_91, __p2_91); \
16710		__ret_91 = __builtin_shufflevector(__ret_91, __ret_91, 7, 6, 5, 4, 3, 2, 1, 0); \
	16691	#define vmulq_lane_f32(__p0_91, __p1_91, __p2_91) __extension__ ({ \
	16692	float32x4_t __ret_91; \
	16693	float32x4_t __s0_91 = __p0_91; \
	16694	float32x2_t __s1_91 = __p1_91; \
	16695	float32x4_t __rev0_91; __rev0_91 = __builtin_shufflevector(__s0_91, __s0_91, 3, 2, 1, 0); \
	16696	float32x2_t __rev1_91; __rev1_91 = __builtin_shufflevector(__s1_91, __s1_91, 1, 0); \
	16697	__ret_91 = __rev0_91 * __noswap_splatq_lane_f32(__rev1_91, __p2_91); \
	16698	__ret_91 = __builtin_shufflevector(__ret_91, __ret_91, 3, 2, 1, 0); \
16711	16699	__ret_91; \
16712	16700	})
16713	16701	#endif
16714	16702
16715	16703	#ifdef __LITTLE_ENDIAN__
16716		#define vmul_lane_u32(__p0_92, __p1_92, __p2_92) __extension__ ({ \
16717		uint32x2_t __ret_92; \
16718		uint32x2_t __s0_92 = __p0_92; \
16719		uint32x2_t __s1_92 = __p1_92; \
16720		__ret_92 = __s0_92 * splat_lane_u32(__s1_92, __p2_92); \
	16704	#define vmulq_lane_s32(__p0_92, __p1_92, __p2_92) __extension__ ({ \
	16705	int32x4_t __ret_92; \
	16706	int32x4_t __s0_92 = __p0_92; \
	16707	int32x2_t __s1_92 = __p1_92; \
	16708	__ret_92 = __s0_92 * splatq_lane_s32(__s1_92, __p2_92); \
16721	16709	__ret_92; \
16722	16710	})
16723	16711	#else
16724		#define vmul_lane_u32(__p0_93, __p1_93, __p2_93) __extension__ ({ \
16725		uint32x2_t __ret_93; \
16726		uint32x2_t __s0_93 = __p0_93; \
16727		uint32x2_t __s1_93 = __p1_93; \
16728		uint32x2_t __rev0_93; __rev0_93 = __builtin_shufflevector(__s0_93, __s0_93, 1, 0); \
16729		uint32x2_t __rev1_93; __rev1_93 = __builtin_shufflevector(__s1_93, __s1_93, 1, 0); \
16730		__ret_93 = __rev0_93 * __noswap_splat_lane_u32(__rev1_93, __p2_93); \
16731		__ret_93 = __builtin_shufflevector(__ret_93, __ret_93, 1, 0); \
	16712	#define vmulq_lane_s32(__p0_93, __p1_93, __p2_93) __extension__ ({ \
	16713	int32x4_t __ret_93; \
	16714	int32x4_t __s0_93 = __p0_93; \
	16715	int32x2_t __s1_93 = __p1_93; \
	16716	int32x4_t __rev0_93; __rev0_93 = __builtin_shufflevector(__s0_93, __s0_93, 3, 2, 1, 0); \
	16717	int32x2_t __rev1_93; __rev1_93 = __builtin_shufflevector(__s1_93, __s1_93, 1, 0); \
	16718	__ret_93 = __rev0_93 * __noswap_splatq_lane_s32(__rev1_93, __p2_93); \
	16719	__ret_93 = __builtin_shufflevector(__ret_93, __ret_93, 3, 2, 1, 0); \
16732	16720	__ret_93; \
16733	16721	})
16734	16722	#endif
16735	16723
16736	16724	#ifdef __LITTLE_ENDIAN__
16737		#define vmul_lane_u16(__p0_94, __p1_94, __p2_94) __extension__ ({ \
16738		uint16x4_t __ret_94; \
16739		uint16x4_t __s0_94 = __p0_94; \
16740		uint16x4_t __s1_94 = __p1_94; \
16741		__ret_94 = __s0_94 * splat_lane_u16(__s1_94, __p2_94); \
	16725	#define vmulq_lane_s16(__p0_94, __p1_94, __p2_94) __extension__ ({ \
	16726	int16x8_t __ret_94; \
	16727	int16x8_t __s0_94 = __p0_94; \
	16728	int16x4_t __s1_94 = __p1_94; \
	16729	__ret_94 = __s0_94 * splatq_lane_s16(__s1_94, __p2_94); \
16742	16730	__ret_94; \
16743	16731	})
16744	16732	#else
16745		#define vmul_lane_u16(__p0_95, __p1_95, __p2_95) __extension__ ({ \
16746		uint16x4_t __ret_95; \
16747		uint16x4_t __s0_95 = __p0_95; \
16748		uint16x4_t __s1_95 = __p1_95; \
16749		uint16x4_t __rev0_95; __rev0_95 = __builtin_shufflevector(__s0_95, __s0_95, 3, 2, 1, 0); \
16750		uint16x4_t __rev1_95; __rev1_95 = __builtin_shufflevector(__s1_95, __s1_95, 3, 2, 1, 0); \
16751		__ret_95 = __rev0_95 * __noswap_splat_lane_u16(__rev1_95, __p2_95); \
16752		__ret_95 = __builtin_shufflevector(__ret_95, __ret_95, 3, 2, 1, 0); \
	16733	#define vmulq_lane_s16(__p0_95, __p1_95, __p2_95) __extension__ ({ \
	16734	int16x8_t __ret_95; \
	16735	int16x8_t __s0_95 = __p0_95; \
	16736	int16x4_t __s1_95 = __p1_95; \
	16737	int16x8_t __rev0_95; __rev0_95 = __builtin_shufflevector(__s0_95, __s0_95, 7, 6, 5, 4, 3, 2, 1, 0); \
	16738	int16x4_t __rev1_95; __rev1_95 = __builtin_shufflevector(__s1_95, __s1_95, 3, 2, 1, 0); \
	16739	__ret_95 = __rev0_95 * __noswap_splatq_lane_s16(__rev1_95, __p2_95); \
	16740	__ret_95 = __builtin_shufflevector(__ret_95, __ret_95, 7, 6, 5, 4, 3, 2, 1, 0); \
16753	16741	__ret_95; \
16754	16742	})
16755	16743	#endif
16756	16744
16757	16745	#ifdef __LITTLE_ENDIAN__
16758		#define vmul_lane_f32(__p0_96, __p1_96, __p2_96) __extension__ ({ \
16759		float32x2_t __ret_96; \
16760		float32x2_t __s0_96 = __p0_96; \
16761		float32x2_t __s1_96 = __p1_96; \
16762		__ret_96 = __s0_96 * splat_lane_f32(__s1_96, __p2_96); \
	16746	#define vmul_lane_u32(__p0_96, __p1_96, __p2_96) __extension__ ({ \
	16747	uint32x2_t __ret_96; \
	16748	uint32x2_t __s0_96 = __p0_96; \
	16749	uint32x2_t __s1_96 = __p1_96; \
	16750	__ret_96 = __s0_96 * splat_lane_u32(__s1_96, __p2_96); \
16763	16751	__ret_96; \
16764	16752	})
16765	16753	#else
16766		#define vmul_lane_f32(__p0_97, __p1_97, __p2_97) __extension__ ({ \
16767		float32x2_t __ret_97; \
16768		float32x2_t __s0_97 = __p0_97; \
16769		float32x2_t __s1_97 = __p1_97; \
16770		float32x2_t __rev0_97; __rev0_97 = __builtin_shufflevector(__s0_97, __s0_97, 1, 0); \
16771		float32x2_t __rev1_97; __rev1_97 = __builtin_shufflevector(__s1_97, __s1_97, 1, 0); \
16772		__ret_97 = __rev0_97 * __noswap_splat_lane_f32(__rev1_97, __p2_97); \
	16754	#define vmul_lane_u32(__p0_97, __p1_97, __p2_97) __extension__ ({ \
	16755	uint32x2_t __ret_97; \
	16756	uint32x2_t __s0_97 = __p0_97; \
	16757	uint32x2_t __s1_97 = __p1_97; \
	16758	uint32x2_t __rev0_97; __rev0_97 = __builtin_shufflevector(__s0_97, __s0_97, 1, 0); \
	16759	uint32x2_t __rev1_97; __rev1_97 = __builtin_shufflevector(__s1_97, __s1_97, 1, 0); \
	16760	__ret_97 = __rev0_97 * __noswap_splat_lane_u32(__rev1_97, __p2_97); \
16773	16761	__ret_97 = __builtin_shufflevector(__ret_97, __ret_97, 1, 0); \
16774	16762	__ret_97; \
16775	16763	})
16776	16764	#endif
16777	16765
16778	16766	#ifdef __LITTLE_ENDIAN__
16779		#define vmul_lane_s32(__p0_98, __p1_98, __p2_98) __extension__ ({ \
16780		int32x2_t __ret_98; \
16781		int32x2_t __s0_98 = __p0_98; \
16782		int32x2_t __s1_98 = __p1_98; \
16783		__ret_98 = __s0_98 * splat_lane_s32(__s1_98, __p2_98); \
	16767	#define vmul_lane_u16(__p0_98, __p1_98, __p2_98) __extension__ ({ \
	16768	uint16x4_t __ret_98; \
	16769	uint16x4_t __s0_98 = __p0_98; \
	16770	uint16x4_t __s1_98 = __p1_98; \
	16771	__ret_98 = __s0_98 * splat_lane_u16(__s1_98, __p2_98); \
16784	16772	__ret_98; \
16785	16773	})
16786	16774	#else
16787		#define vmul_lane_s32(__p0_99, __p1_99, __p2_99) __extension__ ({ \
16788		int32x2_t __ret_99; \
16789		int32x2_t __s0_99 = __p0_99; \
16790		int32x2_t __s1_99 = __p1_99; \
16791		int32x2_t __rev0_99; __rev0_99 = __builtin_shufflevector(__s0_99, __s0_99, 1, 0); \
16792		int32x2_t __rev1_99; __rev1_99 = __builtin_shufflevector(__s1_99, __s1_99, 1, 0); \
16793		__ret_99 = __rev0_99 * __noswap_splat_lane_s32(__rev1_99, __p2_99); \
16794		__ret_99 = __builtin_shufflevector(__ret_99, __ret_99, 1, 0); \
	16775	#define vmul_lane_u16(__p0_99, __p1_99, __p2_99) __extension__ ({ \
	16776	uint16x4_t __ret_99; \
	16777	uint16x4_t __s0_99 = __p0_99; \
	16778	uint16x4_t __s1_99 = __p1_99; \
	16779	uint16x4_t __rev0_99; __rev0_99 = __builtin_shufflevector(__s0_99, __s0_99, 3, 2, 1, 0); \
	16780	uint16x4_t __rev1_99; __rev1_99 = __builtin_shufflevector(__s1_99, __s1_99, 3, 2, 1, 0); \
	16781	__ret_99 = __rev0_99 * __noswap_splat_lane_u16(__rev1_99, __p2_99); \
	16782	__ret_99 = __builtin_shufflevector(__ret_99, __ret_99, 3, 2, 1, 0); \
16795	16783	__ret_99; \
16796	16784	})
16797	16785	#endif
16798	16786
16799	16787	#ifdef __LITTLE_ENDIAN__
16800		#define vmul_lane_s16(__p0_100, __p1_100, __p2_100) __extension__ ({ \
16801		int16x4_t __ret_100; \
16802		int16x4_t __s0_100 = __p0_100; \
16803		int16x4_t __s1_100 = __p1_100; \
16804		__ret_100 = __s0_100 * splat_lane_s16(__s1_100, __p2_100); \
	16788	#define vmul_lane_f32(__p0_100, __p1_100, __p2_100) __extension__ ({ \
	16789	float32x2_t __ret_100; \
	16790	float32x2_t __s0_100 = __p0_100; \
	16791	float32x2_t __s1_100 = __p1_100; \
	16792	__ret_100 = __s0_100 * splat_lane_f32(__s1_100, __p2_100); \
16805	16793	__ret_100; \
16806	16794	})
16807	16795	#else
16808		#define vmul_lane_s16(__p0_101, __p1_101, __p2_101) __extension__ ({ \
16809		int16x4_t __ret_101; \
16810		int16x4_t __s0_101 = __p0_101; \
16811		int16x4_t __s1_101 = __p1_101; \
16812		int16x4_t __rev0_101; __rev0_101 = __builtin_shufflevector(__s0_101, __s0_101, 3, 2, 1, 0); \
16813		int16x4_t __rev1_101; __rev1_101 = __builtin_shufflevector(__s1_101, __s1_101, 3, 2, 1, 0); \
16814		__ret_101 = __rev0_101 * __noswap_splat_lane_s16(__rev1_101, __p2_101); \
16815		__ret_101 = __builtin_shufflevector(__ret_101, __ret_101, 3, 2, 1, 0); \
	16796	#define vmul_lane_f32(__p0_101, __p1_101, __p2_101) __extension__ ({ \
	16797	float32x2_t __ret_101; \
	16798	float32x2_t __s0_101 = __p0_101; \
	16799	float32x2_t __s1_101 = __p1_101; \
	16800	float32x2_t __rev0_101; __rev0_101 = __builtin_shufflevector(__s0_101, __s0_101, 1, 0); \
	16801	float32x2_t __rev1_101; __rev1_101 = __builtin_shufflevector(__s1_101, __s1_101, 1, 0); \
	16802	__ret_101 = __rev0_101 * __noswap_splat_lane_f32(__rev1_101, __p2_101); \
	16803	__ret_101 = __builtin_shufflevector(__ret_101, __ret_101, 1, 0); \
16816	16804	__ret_101; \
16817	16805	})
16818	16806	#endif
16819	16807
	16808	#ifdef __LITTLE_ENDIAN__
	16809	#define vmul_lane_s32(__p0_102, __p1_102, __p2_102) __extension__ ({ \
	16810	int32x2_t __ret_102; \
	16811	int32x2_t __s0_102 = __p0_102; \
	16812	int32x2_t __s1_102 = __p1_102; \
	16813	__ret_102 = __s0_102 * splat_lane_s32(__s1_102, __p2_102); \
	16814	__ret_102; \
	16815	})
	16816	#else
	16817	#define vmul_lane_s32(__p0_103, __p1_103, __p2_103) __extension__ ({ \
	16818	int32x2_t __ret_103; \
	16819	int32x2_t __s0_103 = __p0_103; \
	16820	int32x2_t __s1_103 = __p1_103; \
	16821	int32x2_t __rev0_103; __rev0_103 = __builtin_shufflevector(__s0_103, __s0_103, 1, 0); \
	16822	int32x2_t __rev1_103; __rev1_103 = __builtin_shufflevector(__s1_103, __s1_103, 1, 0); \
	16823	__ret_103 = __rev0_103 * __noswap_splat_lane_s32(__rev1_103, __p2_103); \
	16824	__ret_103 = __builtin_shufflevector(__ret_103, __ret_103, 1, 0); \
	16825	__ret_103; \
	16826	})
	16827	#endif
	16828
	16829	#ifdef __LITTLE_ENDIAN__
	16830	#define vmul_lane_s16(__p0_104, __p1_104, __p2_104) __extension__ ({ \
	16831	int16x4_t __ret_104; \
	16832	int16x4_t __s0_104 = __p0_104; \
	16833	int16x4_t __s1_104 = __p1_104; \
	16834	__ret_104 = __s0_104 * splat_lane_s16(__s1_104, __p2_104); \
	16835	__ret_104; \
	16836	})
	16837	#else
	16838	#define vmul_lane_s16(__p0_105, __p1_105, __p2_105) __extension__ ({ \
	16839	int16x4_t __ret_105; \
	16840	int16x4_t __s0_105 = __p0_105; \
	16841	int16x4_t __s1_105 = __p1_105; \
	16842	int16x4_t __rev0_105; __rev0_105 = __builtin_shufflevector(__s0_105, __s0_105, 3, 2, 1, 0); \
	16843	int16x4_t __rev1_105; __rev1_105 = __builtin_shufflevector(__s1_105, __s1_105, 3, 2, 1, 0); \
	16844	__ret_105 = __rev0_105 * __noswap_splat_lane_s16(__rev1_105, __p2_105); \
	16845	__ret_105 = __builtin_shufflevector(__ret_105, __ret_105, 3, 2, 1, 0); \
	16846	__ret_105; \
	16847	})
	16848	#endif
	16849
16820	16850	#ifdef __LITTLE_ENDIAN__
16821	16851	__ai uint32x4_t vmulq_n_u32(uint32x4_t __p0, uint32_t __p1) {
16822	16852	uint32x4_t __ret;
...	...	@@ -17132,86 +17162,86 @@ __ai int32x4_t __noswap_vmull_s16(int16x4_t __p0, int16x4_t __p1) {
17132	17162	#endif
17133	17163
17134	17164	#ifdef __LITTLE_ENDIAN__
17135		#define vmull_lane_u32(__p0_102, __p1_102, __p2_102) __extension__ ({ \
17136		uint64x2_t __ret_102; \
17137		uint32x2_t __s0_102 = __p0_102; \
17138		uint32x2_t __s1_102 = __p1_102; \
17139		__ret_102 = vmull_u32(__s0_102, splat_lane_u32(__s1_102, __p2_102)); \
17140		__ret_102; \
	17165	#define vmull_lane_u32(__p0_106, __p1_106, __p2_106) __extension__ ({ \
	17166	uint64x2_t __ret_106; \
	17167	uint32x2_t __s0_106 = __p0_106; \
	17168	uint32x2_t __s1_106 = __p1_106; \
	17169	__ret_106 = vmull_u32(__s0_106, splat_lane_u32(__s1_106, __p2_106)); \
	17170	__ret_106; \
17141	17171	})
17142	17172	#else
17143		#define vmull_lane_u32(__p0_103, __p1_103, __p2_103) __extension__ ({ \
17144		uint64x2_t __ret_103; \
17145		uint32x2_t __s0_103 = __p0_103; \
17146		uint32x2_t __s1_103 = __p1_103; \
17147		uint32x2_t __rev0_103; __rev0_103 = __builtin_shufflevector(__s0_103, __s0_103, 1, 0); \
17148		uint32x2_t __rev1_103; __rev1_103 = __builtin_shufflevector(__s1_103, __s1_103, 1, 0); \
17149		__ret_103 = __noswap_vmull_u32(__rev0_103, __noswap_splat_lane_u32(__rev1_103, __p2_103)); \
17150		__ret_103 = __builtin_shufflevector(__ret_103, __ret_103, 1, 0); \
17151		__ret_103; \
	17173	#define vmull_lane_u32(__p0_107, __p1_107, __p2_107) __extension__ ({ \
	17174	uint64x2_t __ret_107; \
	17175	uint32x2_t __s0_107 = __p0_107; \
	17176	uint32x2_t __s1_107 = __p1_107; \
	17177	uint32x2_t __rev0_107; __rev0_107 = __builtin_shufflevector(__s0_107, __s0_107, 1, 0); \
	17178	uint32x2_t __rev1_107; __rev1_107 = __builtin_shufflevector(__s1_107, __s1_107, 1, 0); \
	17179	__ret_107 = __noswap_vmull_u32(__rev0_107, __noswap_splat_lane_u32(__rev1_107, __p2_107)); \
	17180	__ret_107 = __builtin_shufflevector(__ret_107, __ret_107, 1, 0); \
	17181	__ret_107; \
17152	17182	})
17153	17183	#endif
17154	17184
17155	17185	#ifdef __LITTLE_ENDIAN__
17156		#define vmull_lane_u16(__p0_104, __p1_104, __p2_104) __extension__ ({ \
17157		uint32x4_t __ret_104; \
17158		uint16x4_t __s0_104 = __p0_104; \
17159		uint16x4_t __s1_104 = __p1_104; \
17160		__ret_104 = vmull_u16(__s0_104, splat_lane_u16(__s1_104, __p2_104)); \
17161		__ret_104; \
	17186	#define vmull_lane_u16(__p0_108, __p1_108, __p2_108) __extension__ ({ \
	17187	uint32x4_t __ret_108; \
	17188	uint16x4_t __s0_108 = __p0_108; \
	17189	uint16x4_t __s1_108 = __p1_108; \
	17190	__ret_108 = vmull_u16(__s0_108, splat_lane_u16(__s1_108, __p2_108)); \
	17191	__ret_108; \
17162	17192	})
17163	17193	#else
17164		#define vmull_lane_u16(__p0_105, __p1_105, __p2_105) __extension__ ({ \
17165		uint32x4_t __ret_105; \
17166		uint16x4_t __s0_105 = __p0_105; \
17167		uint16x4_t __s1_105 = __p1_105; \
17168		uint16x4_t __rev0_105; __rev0_105 = __builtin_shufflevector(__s0_105, __s0_105, 3, 2, 1, 0); \
17169		uint16x4_t __rev1_105; __rev1_105 = __builtin_shufflevector(__s1_105, __s1_105, 3, 2, 1, 0); \
17170		__ret_105 = __noswap_vmull_u16(__rev0_105, __noswap_splat_lane_u16(__rev1_105, __p2_105)); \
17171		__ret_105 = __builtin_shufflevector(__ret_105, __ret_105, 3, 2, 1, 0); \
17172		__ret_105; \
	17194	#define vmull_lane_u16(__p0_109, __p1_109, __p2_109) __extension__ ({ \
	17195	uint32x4_t __ret_109; \
	17196	uint16x4_t __s0_109 = __p0_109; \
	17197	uint16x4_t __s1_109 = __p1_109; \
	17198	uint16x4_t __rev0_109; __rev0_109 = __builtin_shufflevector(__s0_109, __s0_109, 3, 2, 1, 0); \
	17199	uint16x4_t __rev1_109; __rev1_109 = __builtin_shufflevector(__s1_109, __s1_109, 3, 2, 1, 0); \
	17200	__ret_109 = __noswap_vmull_u16(__rev0_109, __noswap_splat_lane_u16(__rev1_109, __p2_109)); \
	17201	__ret_109 = __builtin_shufflevector(__ret_109, __ret_109, 3, 2, 1, 0); \
	17202	__ret_109; \
17173	17203	})
17174	17204	#endif
17175	17205
17176	17206	#ifdef __LITTLE_ENDIAN__
17177		#define vmull_lane_s32(__p0_106, __p1_106, __p2_106) __extension__ ({ \
17178		int64x2_t __ret_106; \
17179		int32x2_t __s0_106 = __p0_106; \
17180		int32x2_t __s1_106 = __p1_106; \
17181		__ret_106 = vmull_s32(__s0_106, splat_lane_s32(__s1_106, __p2_106)); \
17182		__ret_106; \
	17207	#define vmull_lane_s32(__p0_110, __p1_110, __p2_110) __extension__ ({ \
	17208	int64x2_t __ret_110; \
	17209	int32x2_t __s0_110 = __p0_110; \
	17210	int32x2_t __s1_110 = __p1_110; \
	17211	__ret_110 = vmull_s32(__s0_110, splat_lane_s32(__s1_110, __p2_110)); \
	17212	__ret_110; \
17183	17213	})
17184	17214	#else
17185		#define vmull_lane_s32(__p0_107, __p1_107, __p2_107) __extension__ ({ \
17186		int64x2_t __ret_107; \
17187		int32x2_t __s0_107 = __p0_107; \
17188		int32x2_t __s1_107 = __p1_107; \
17189		int32x2_t __rev0_107; __rev0_107 = __builtin_shufflevector(__s0_107, __s0_107, 1, 0); \
17190		int32x2_t __rev1_107; __rev1_107 = __builtin_shufflevector(__s1_107, __s1_107, 1, 0); \
17191		__ret_107 = __noswap_vmull_s32(__rev0_107, __noswap_splat_lane_s32(__rev1_107, __p2_107)); \
17192		__ret_107 = __builtin_shufflevector(__ret_107, __ret_107, 1, 0); \
17193		__ret_107; \
	17215	#define vmull_lane_s32(__p0_111, __p1_111, __p2_111) __extension__ ({ \
	17216	int64x2_t __ret_111; \
	17217	int32x2_t __s0_111 = __p0_111; \
	17218	int32x2_t __s1_111 = __p1_111; \
	17219	int32x2_t __rev0_111; __rev0_111 = __builtin_shufflevector(__s0_111, __s0_111, 1, 0); \
	17220	int32x2_t __rev1_111; __rev1_111 = __builtin_shufflevector(__s1_111, __s1_111, 1, 0); \
	17221	__ret_111 = __noswap_vmull_s32(__rev0_111, __noswap_splat_lane_s32(__rev1_111, __p2_111)); \
	17222	__ret_111 = __builtin_shufflevector(__ret_111, __ret_111, 1, 0); \
	17223	__ret_111; \
17194	17224	})
17195	17225	#endif
17196	17226
17197	17227	#ifdef __LITTLE_ENDIAN__
17198		#define vmull_lane_s16(__p0_108, __p1_108, __p2_108) __extension__ ({ \
17199		int32x4_t __ret_108; \
17200		int16x4_t __s0_108 = __p0_108; \
17201		int16x4_t __s1_108 = __p1_108; \
17202		__ret_108 = vmull_s16(__s0_108, splat_lane_s16(__s1_108, __p2_108)); \
17203		__ret_108; \
	17228	#define vmull_lane_s16(__p0_112, __p1_112, __p2_112) __extension__ ({ \
	17229	int32x4_t __ret_112; \
	17230	int16x4_t __s0_112 = __p0_112; \
	17231	int16x4_t __s1_112 = __p1_112; \
	17232	__ret_112 = vmull_s16(__s0_112, splat_lane_s16(__s1_112, __p2_112)); \
	17233	__ret_112; \
17204	17234	})
17205	17235	#else
17206		#define vmull_lane_s16(__p0_109, __p1_109, __p2_109) __extension__ ({ \
17207		int32x4_t __ret_109; \
17208		int16x4_t __s0_109 = __p0_109; \
17209		int16x4_t __s1_109 = __p1_109; \
17210		int16x4_t __rev0_109; __rev0_109 = __builtin_shufflevector(__s0_109, __s0_109, 3, 2, 1, 0); \
17211		int16x4_t __rev1_109; __rev1_109 = __builtin_shufflevector(__s1_109, __s1_109, 3, 2, 1, 0); \
17212		__ret_109 = __noswap_vmull_s16(__rev0_109, __noswap_splat_lane_s16(__rev1_109, __p2_109)); \
17213		__ret_109 = __builtin_shufflevector(__ret_109, __ret_109, 3, 2, 1, 0); \
17214		__ret_109; \
	17236	#define vmull_lane_s16(__p0_113, __p1_113, __p2_113) __extension__ ({ \
	17237	int32x4_t __ret_113; \
	17238	int16x4_t __s0_113 = __p0_113; \
	17239	int16x4_t __s1_113 = __p1_113; \
	17240	int16x4_t __rev0_113; __rev0_113 = __builtin_shufflevector(__s0_113, __s0_113, 3, 2, 1, 0); \
	17241	int16x4_t __rev1_113; __rev1_113 = __builtin_shufflevector(__s1_113, __s1_113, 3, 2, 1, 0); \
	17242	__ret_113 = __noswap_vmull_s16(__rev0_113, __noswap_splat_lane_s16(__rev1_113, __p2_113)); \
	17243	__ret_113 = __builtin_shufflevector(__ret_113, __ret_113, 3, 2, 1, 0); \
	17244	__ret_113; \
17215	17245	})
17216	17246	#endif
17217	17247
...	...	@@ -19285,50 +19315,50 @@ __ai int32x4_t __noswap_vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __
19285	19315	#endif
19286	19316
19287	19317	#ifdef __LITTLE_ENDIAN__
19288		#define vqdmlal_lane_s32(__p0_110, __p1_110, __p2_110, __p3_110) __extension__ ({ \
19289		int64x2_t __ret_110; \
19290		int64x2_t __s0_110 = __p0_110; \
19291		int32x2_t __s1_110 = __p1_110; \
19292		int32x2_t __s2_110 = __p2_110; \
19293		__ret_110 = vqdmlal_s32(__s0_110, __s1_110, splat_lane_s32(__s2_110, __p3_110)); \
19294		__ret_110; \
	19318	#define vqdmlal_lane_s32(__p0_114, __p1_114, __p2_114, __p3_114) __extension__ ({ \
	19319	int64x2_t __ret_114; \
	19320	int64x2_t __s0_114 = __p0_114; \
	19321	int32x2_t __s1_114 = __p1_114; \
	19322	int32x2_t __s2_114 = __p2_114; \
	19323	__ret_114 = vqdmlal_s32(__s0_114, __s1_114, splat_lane_s32(__s2_114, __p3_114)); \
	19324	__ret_114; \
19295	19325	})
19296	19326	#else
19297		#define vqdmlal_lane_s32(__p0_111, __p1_111, __p2_111, __p3_111) __extension__ ({ \
19298		int64x2_t __ret_111; \
19299		int64x2_t __s0_111 = __p0_111; \
19300		int32x2_t __s1_111 = __p1_111; \
19301		int32x2_t __s2_111 = __p2_111; \
19302		int64x2_t __rev0_111; __rev0_111 = __builtin_shufflevector(__s0_111, __s0_111, 1, 0); \
19303		int32x2_t __rev1_111; __rev1_111 = __builtin_shufflevector(__s1_111, __s1_111, 1, 0); \
19304		int32x2_t __rev2_111; __rev2_111 = __builtin_shufflevector(__s2_111, __s2_111, 1, 0); \
19305		__ret_111 = __noswap_vqdmlal_s32(__rev0_111, __rev1_111, __noswap_splat_lane_s32(__rev2_111, __p3_111)); \
19306		__ret_111 = __builtin_shufflevector(__ret_111, __ret_111, 1, 0); \
19307		__ret_111; \
	19327	#define vqdmlal_lane_s32(__p0_115, __p1_115, __p2_115, __p3_115) __extension__ ({ \
	19328	int64x2_t __ret_115; \
	19329	int64x2_t __s0_115 = __p0_115; \
	19330	int32x2_t __s1_115 = __p1_115; \
	19331	int32x2_t __s2_115 = __p2_115; \
	19332	int64x2_t __rev0_115; __rev0_115 = __builtin_shufflevector(__s0_115, __s0_115, 1, 0); \
	19333	int32x2_t __rev1_115; __rev1_115 = __builtin_shufflevector(__s1_115, __s1_115, 1, 0); \
	19334	int32x2_t __rev2_115; __rev2_115 = __builtin_shufflevector(__s2_115, __s2_115, 1, 0); \
	19335	__ret_115 = __noswap_vqdmlal_s32(__rev0_115, __rev1_115, __noswap_splat_lane_s32(__rev2_115, __p3_115)); \
	19336	__ret_115 = __builtin_shufflevector(__ret_115, __ret_115, 1, 0); \
	19337	__ret_115; \
19308	19338	})
19309	19339	#endif
19310	19340
19311	19341	#ifdef __LITTLE_ENDIAN__
19312		#define vqdmlal_lane_s16(__p0_112, __p1_112, __p2_112, __p3_112) __extension__ ({ \
19313		int32x4_t __ret_112; \
19314		int32x4_t __s0_112 = __p0_112; \
19315		int16x4_t __s1_112 = __p1_112; \
19316		int16x4_t __s2_112 = __p2_112; \
19317		__ret_112 = vqdmlal_s16(__s0_112, __s1_112, splat_lane_s16(__s2_112, __p3_112)); \
19318		__ret_112; \
	19342	#define vqdmlal_lane_s16(__p0_116, __p1_116, __p2_116, __p3_116) __extension__ ({ \
	19343	int32x4_t __ret_116; \
	19344	int32x4_t __s0_116 = __p0_116; \
	19345	int16x4_t __s1_116 = __p1_116; \
	19346	int16x4_t __s2_116 = __p2_116; \
	19347	__ret_116 = vqdmlal_s16(__s0_116, __s1_116, splat_lane_s16(__s2_116, __p3_116)); \
	19348	__ret_116; \
19319	19349	})
19320	19350	#else
19321		#define vqdmlal_lane_s16(__p0_113, __p1_113, __p2_113, __p3_113) __extension__ ({ \
19322		int32x4_t __ret_113; \
19323		int32x4_t __s0_113 = __p0_113; \
19324		int16x4_t __s1_113 = __p1_113; \
19325		int16x4_t __s2_113 = __p2_113; \
19326		int32x4_t __rev0_113; __rev0_113 = __builtin_shufflevector(__s0_113, __s0_113, 3, 2, 1, 0); \
19327		int16x4_t __rev1_113; __rev1_113 = __builtin_shufflevector(__s1_113, __s1_113, 3, 2, 1, 0); \
19328		int16x4_t __rev2_113; __rev2_113 = __builtin_shufflevector(__s2_113, __s2_113, 3, 2, 1, 0); \
19329		__ret_113 = __noswap_vqdmlal_s16(__rev0_113, __rev1_113, __noswap_splat_lane_s16(__rev2_113, __p3_113)); \
19330		__ret_113 = __builtin_shufflevector(__ret_113, __ret_113, 3, 2, 1, 0); \
19331		__ret_113; \
	19351	#define vqdmlal_lane_s16(__p0_117, __p1_117, __p2_117, __p3_117) __extension__ ({ \
	19352	int32x4_t __ret_117; \
	19353	int32x4_t __s0_117 = __p0_117; \
	19354	int16x4_t __s1_117 = __p1_117; \
	19355	int16x4_t __s2_117 = __p2_117; \
	19356	int32x4_t __rev0_117; __rev0_117 = __builtin_shufflevector(__s0_117, __s0_117, 3, 2, 1, 0); \
	19357	int16x4_t __rev1_117; __rev1_117 = __builtin_shufflevector(__s1_117, __s1_117, 3, 2, 1, 0); \
	19358	int16x4_t __rev2_117; __rev2_117 = __builtin_shufflevector(__s2_117, __s2_117, 3, 2, 1, 0); \
	19359	__ret_117 = __noswap_vqdmlal_s16(__rev0_117, __rev1_117, __noswap_splat_lane_s16(__rev2_117, __p3_117)); \
	19360	__ret_117 = __builtin_shufflevector(__ret_117, __ret_117, 3, 2, 1, 0); \
	19361	__ret_117; \
19332	19362	})
19333	19363	#endif
19334	19364
...	...	@@ -19423,50 +19453,50 @@ __ai int32x4_t __noswap_vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __
19423	19453	#endif
19424	19454
19425	19455	#ifdef __LITTLE_ENDIAN__
19426		#define vqdmlsl_lane_s32(__p0_114, __p1_114, __p2_114, __p3_114) __extension__ ({ \
19427		int64x2_t __ret_114; \
19428		int64x2_t __s0_114 = __p0_114; \
19429		int32x2_t __s1_114 = __p1_114; \
19430		int32x2_t __s2_114 = __p2_114; \
19431		__ret_114 = vqdmlsl_s32(__s0_114, __s1_114, splat_lane_s32(__s2_114, __p3_114)); \
19432		__ret_114; \
	19456	#define vqdmlsl_lane_s32(__p0_118, __p1_118, __p2_118, __p3_118) __extension__ ({ \
	19457	int64x2_t __ret_118; \
	19458	int64x2_t __s0_118 = __p0_118; \
	19459	int32x2_t __s1_118 = __p1_118; \
	19460	int32x2_t __s2_118 = __p2_118; \
	19461	__ret_118 = vqdmlsl_s32(__s0_118, __s1_118, splat_lane_s32(__s2_118, __p3_118)); \
	19462	__ret_118; \
19433	19463	})
19434	19464	#else
19435		#define vqdmlsl_lane_s32(__p0_115, __p1_115, __p2_115, __p3_115) __extension__ ({ \
19436		int64x2_t __ret_115; \
19437		int64x2_t __s0_115 = __p0_115; \
19438		int32x2_t __s1_115 = __p1_115; \
19439		int32x2_t __s2_115 = __p2_115; \
19440		int64x2_t __rev0_115; __rev0_115 = __builtin_shufflevector(__s0_115, __s0_115, 1, 0); \
19441		int32x2_t __rev1_115; __rev1_115 = __builtin_shufflevector(__s1_115, __s1_115, 1, 0); \
19442		int32x2_t __rev2_115; __rev2_115 = __builtin_shufflevector(__s2_115, __s2_115, 1, 0); \
19443		__ret_115 = __noswap_vqdmlsl_s32(__rev0_115, __rev1_115, __noswap_splat_lane_s32(__rev2_115, __p3_115)); \
19444		__ret_115 = __builtin_shufflevector(__ret_115, __ret_115, 1, 0); \
19445		__ret_115; \
	19465	#define vqdmlsl_lane_s32(__p0_119, __p1_119, __p2_119, __p3_119) __extension__ ({ \
	19466	int64x2_t __ret_119; \
	19467	int64x2_t __s0_119 = __p0_119; \
	19468	int32x2_t __s1_119 = __p1_119; \
	19469	int32x2_t __s2_119 = __p2_119; \
	19470	int64x2_t __rev0_119; __rev0_119 = __builtin_shufflevector(__s0_119, __s0_119, 1, 0); \
	19471	int32x2_t __rev1_119; __rev1_119 = __builtin_shufflevector(__s1_119, __s1_119, 1, 0); \
	19472	int32x2_t __rev2_119; __rev2_119 = __builtin_shufflevector(__s2_119, __s2_119, 1, 0); \
	19473	__ret_119 = __noswap_vqdmlsl_s32(__rev0_119, __rev1_119, __noswap_splat_lane_s32(__rev2_119, __p3_119)); \
	19474	__ret_119 = __builtin_shufflevector(__ret_119, __ret_119, 1, 0); \
	19475	__ret_119; \
19446	19476	})
19447	19477	#endif
19448	19478
19449	19479	#ifdef __LITTLE_ENDIAN__
19450		#define vqdmlsl_lane_s16(__p0_116, __p1_116, __p2_116, __p3_116) __extension__ ({ \
19451		int32x4_t __ret_116; \
19452		int32x4_t __s0_116 = __p0_116; \
19453		int16x4_t __s1_116 = __p1_116; \
19454		int16x4_t __s2_116 = __p2_116; \
19455		__ret_116 = vqdmlsl_s16(__s0_116, __s1_116, splat_lane_s16(__s2_116, __p3_116)); \
19456		__ret_116; \
	19480	#define vqdmlsl_lane_s16(__p0_120, __p1_120, __p2_120, __p3_120) __extension__ ({ \
	19481	int32x4_t __ret_120; \
	19482	int32x4_t __s0_120 = __p0_120; \
	19483	int16x4_t __s1_120 = __p1_120; \
	19484	int16x4_t __s2_120 = __p2_120; \
	19485	__ret_120 = vqdmlsl_s16(__s0_120, __s1_120, splat_lane_s16(__s2_120, __p3_120)); \
	19486	__ret_120; \
19457	19487	})
19458	19488	#else
19459		#define vqdmlsl_lane_s16(__p0_117, __p1_117, __p2_117, __p3_117) __extension__ ({ \
19460		int32x4_t __ret_117; \
19461		int32x4_t __s0_117 = __p0_117; \
19462		int16x4_t __s1_117 = __p1_117; \
19463		int16x4_t __s2_117 = __p2_117; \
19464		int32x4_t __rev0_117; __rev0_117 = __builtin_shufflevector(__s0_117, __s0_117, 3, 2, 1, 0); \
19465		int16x4_t __rev1_117; __rev1_117 = __builtin_shufflevector(__s1_117, __s1_117, 3, 2, 1, 0); \
19466		int16x4_t __rev2_117; __rev2_117 = __builtin_shufflevector(__s2_117, __s2_117, 3, 2, 1, 0); \
19467		__ret_117 = __noswap_vqdmlsl_s16(__rev0_117, __rev1_117, __noswap_splat_lane_s16(__rev2_117, __p3_117)); \
19468		__ret_117 = __builtin_shufflevector(__ret_117, __ret_117, 3, 2, 1, 0); \
19469		__ret_117; \
	19489	#define vqdmlsl_lane_s16(__p0_121, __p1_121, __p2_121, __p3_121) __extension__ ({ \
	19490	int32x4_t __ret_121; \
	19491	int32x4_t __s0_121 = __p0_121; \
	19492	int16x4_t __s1_121 = __p1_121; \
	19493	int16x4_t __s2_121 = __p2_121; \
	19494	int32x4_t __rev0_121; __rev0_121 = __builtin_shufflevector(__s0_121, __s0_121, 3, 2, 1, 0); \
	19495	int16x4_t __rev1_121; __rev1_121 = __builtin_shufflevector(__s1_121, __s1_121, 3, 2, 1, 0); \
	19496	int16x4_t __rev2_121; __rev2_121 = __builtin_shufflevector(__s2_121, __s2_121, 3, 2, 1, 0); \
	19497	__ret_121 = __noswap_vqdmlsl_s16(__rev0_121, __rev1_121, __noswap_splat_lane_s16(__rev2_121, __p3_121)); \
	19498	__ret_121 = __builtin_shufflevector(__ret_121, __ret_121, 3, 2, 1, 0); \
	19499	__ret_121; \
19470	19500	})
19471	19501	#endif
19472	19502
...	...	@@ -19711,44 +19741,44 @@ __ai int32x4_t __noswap_vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
19711	19741	#endif
19712	19742
19713	19743	#ifdef __LITTLE_ENDIAN__
19714		#define vqdmull_lane_s32(__p0_118, __p1_118, __p2_118) __extension__ ({ \
19715		int64x2_t __ret_118; \
19716		int32x2_t __s0_118 = __p0_118; \
19717		int32x2_t __s1_118 = __p1_118; \
19718		__ret_118 = vqdmull_s32(__s0_118, splat_lane_s32(__s1_118, __p2_118)); \
19719		__ret_118; \
	19744	#define vqdmull_lane_s32(__p0_122, __p1_122, __p2_122) __extension__ ({ \
	19745	int64x2_t __ret_122; \
	19746	int32x2_t __s0_122 = __p0_122; \
	19747	int32x2_t __s1_122 = __p1_122; \
	19748	__ret_122 = vqdmull_s32(__s0_122, splat_lane_s32(__s1_122, __p2_122)); \
	19749	__ret_122; \
19720	19750	})
19721	19751	#else
19722		#define vqdmull_lane_s32(__p0_119, __p1_119, __p2_119) __extension__ ({ \
19723		int64x2_t __ret_119; \
19724		int32x2_t __s0_119 = __p0_119; \
19725		int32x2_t __s1_119 = __p1_119; \
19726		int32x2_t __rev0_119; __rev0_119 = __builtin_shufflevector(__s0_119, __s0_119, 1, 0); \
19727		int32x2_t __rev1_119; __rev1_119 = __builtin_shufflevector(__s1_119, __s1_119, 1, 0); \
19728		__ret_119 = __noswap_vqdmull_s32(__rev0_119, __noswap_splat_lane_s32(__rev1_119, __p2_119)); \
19729		__ret_119 = __builtin_shufflevector(__ret_119, __ret_119, 1, 0); \
19730		__ret_119; \
	19752	#define vqdmull_lane_s32(__p0_123, __p1_123, __p2_123) __extension__ ({ \
	19753	int64x2_t __ret_123; \
	19754	int32x2_t __s0_123 = __p0_123; \
	19755	int32x2_t __s1_123 = __p1_123; \
	19756	int32x2_t __rev0_123; __rev0_123 = __builtin_shufflevector(__s0_123, __s0_123, 1, 0); \
	19757	int32x2_t __rev1_123; __rev1_123 = __builtin_shufflevector(__s1_123, __s1_123, 1, 0); \
	19758	__ret_123 = __noswap_vqdmull_s32(__rev0_123, __noswap_splat_lane_s32(__rev1_123, __p2_123)); \
	19759	__ret_123 = __builtin_shufflevector(__ret_123, __ret_123, 1, 0); \
	19760	__ret_123; \
19731	19761	})
19732	19762	#endif
19733	19763
19734	19764	#ifdef __LITTLE_ENDIAN__
19735		#define vqdmull_lane_s16(__p0_120, __p1_120, __p2_120) __extension__ ({ \
19736		int32x4_t __ret_120; \
19737		int16x4_t __s0_120 = __p0_120; \
19738		int16x4_t __s1_120 = __p1_120; \
19739		__ret_120 = vqdmull_s16(__s0_120, splat_lane_s16(__s1_120, __p2_120)); \
19740		__ret_120; \
	19765	#define vqdmull_lane_s16(__p0_124, __p1_124, __p2_124) __extension__ ({ \
	19766	int32x4_t __ret_124; \
	19767	int16x4_t __s0_124 = __p0_124; \
	19768	int16x4_t __s1_124 = __p1_124; \
	19769	__ret_124 = vqdmull_s16(__s0_124, splat_lane_s16(__s1_124, __p2_124)); \
	19770	__ret_124; \
19741	19771	})
19742	19772	#else
19743		#define vqdmull_lane_s16(__p0_121, __p1_121, __p2_121) __extension__ ({ \
19744		int32x4_t __ret_121; \
19745		int16x4_t __s0_121 = __p0_121; \
19746		int16x4_t __s1_121 = __p1_121; \
19747		int16x4_t __rev0_121; __rev0_121 = __builtin_shufflevector(__s0_121, __s0_121, 3, 2, 1, 0); \
19748		int16x4_t __rev1_121; __rev1_121 = __builtin_shufflevector(__s1_121, __s1_121, 3, 2, 1, 0); \
19749		__ret_121 = __noswap_vqdmull_s16(__rev0_121, __noswap_splat_lane_s16(__rev1_121, __p2_121)); \
19750		__ret_121 = __builtin_shufflevector(__ret_121, __ret_121, 3, 2, 1, 0); \
19751		__ret_121; \
	19773	#define vqdmull_lane_s16(__p0_125, __p1_125, __p2_125) __extension__ ({ \
	19774	int32x4_t __ret_125; \
	19775	int16x4_t __s0_125 = __p0_125; \
	19776	int16x4_t __s1_125 = __p1_125; \
	19777	int16x4_t __rev0_125; __rev0_125 = __builtin_shufflevector(__s0_125, __s0_125, 3, 2, 1, 0); \
	19778	int16x4_t __rev1_125; __rev1_125 = __builtin_shufflevector(__s1_125, __s1_125, 3, 2, 1, 0); \
	19779	__ret_125 = __noswap_vqdmull_s16(__rev0_125, __noswap_splat_lane_s16(__rev1_125, __p2_125)); \
	19780	__ret_125 = __builtin_shufflevector(__ret_125, __ret_125, 3, 2, 1, 0); \
	19781	__ret_125; \
19752	19782	})
19753	19783	#endif
19754	19784
...	...	@@ -32235,4366 +32265,5707 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
32235	32265	}
32236	32266	#endif
32237	32267
32238		#if !defined(__aarch64__)
32239	32268	#ifdef __LITTLE_ENDIAN__
32240		#define vdupq_lane_f16(__p0_122, __p1_122) __extension__ ({ \
32241		float16x8_t __ret_122; \
32242		float16x4_t __s0_122 = __p0_122; \
32243		__ret_122 = splatq_lane_f16(__s0_122, __p1_122); \
32244		__ret_122; \
	32269	#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
	32270	bfloat16x8_t __ret; \
	32271	bfloat16x4_t __s0 = __p0; \
	32272	__ret = (bfloat16x8_t) __builtin_neon_splatq_lane_bf16((int8x8_t)__s0, __p1, 11); \
	32273	__ret; \
32245	32274	})
32246	32275	#else
32247		#define vdupq_lane_f16(__p0_123, __p1_123) __extension__ ({ \
32248		float16x8_t __ret_123; \
32249		float16x4_t __s0_123 = __p0_123; \
32250		float16x4_t __rev0_123; __rev0_123 = __builtin_shufflevector(__s0_123, __s0_123, 3, 2, 1, 0); \
32251		__ret_123 = __noswap_splatq_lane_f16(__rev0_123, __p1_123); \
32252		__ret_123 = __builtin_shufflevector(__ret_123, __ret_123, 7, 6, 5, 4, 3, 2, 1, 0); \
32253		__ret_123; \
	32276	#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
	32277	bfloat16x8_t __ret; \
	32278	bfloat16x4_t __s0 = __p0; \
	32279	bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	32280	__ret = (bfloat16x8_t) __builtin_neon_splatq_lane_bf16((int8x8_t)__rev0, __p1, 11); \
	32281	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	32282	__ret; \
	32283	})
	32284	#define __noswap_splatq_lane_bf16(__p0, __p1) __extension__ ({ \
	32285	bfloat16x8_t __ret; \
	32286	bfloat16x4_t __s0 = __p0; \
	32287	__ret = (bfloat16x8_t) __builtin_neon_splatq_lane_bf16((int8x8_t)__s0, __p1, 11); \
	32288	__ret; \
32254	32289	})
32255	32290	#endif
32256	32291
32257	32292	#ifdef __LITTLE_ENDIAN__
32258		#define vdup_lane_f16(__p0_124, __p1_124) __extension__ ({ \
32259		float16x4_t __ret_124; \
32260		float16x4_t __s0_124 = __p0_124; \
32261		__ret_124 = splat_lane_f16(__s0_124, __p1_124); \
32262		__ret_124; \
	32293	#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
	32294	bfloat16x4_t __ret; \
	32295	bfloat16x4_t __s0 = __p0; \
	32296	__ret = (bfloat16x4_t) __builtin_neon_splat_lane_bf16((int8x8_t)__s0, __p1, 11); \
	32297	__ret; \
32263	32298	})
32264	32299	#else
32265		#define vdup_lane_f16(__p0_125, __p1_125) __extension__ ({ \
32266		float16x4_t __ret_125; \
32267		float16x4_t __s0_125 = __p0_125; \
32268		float16x4_t __rev0_125; __rev0_125 = __builtin_shufflevector(__s0_125, __s0_125, 3, 2, 1, 0); \
32269		__ret_125 = __noswap_splat_lane_f16(__rev0_125, __p1_125); \
32270		__ret_125 = __builtin_shufflevector(__ret_125, __ret_125, 3, 2, 1, 0); \
32271		__ret_125; \
	32300	#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
	32301	bfloat16x4_t __ret; \
	32302	bfloat16x4_t __s0 = __p0; \
	32303	bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	32304	__ret = (bfloat16x4_t) __builtin_neon_splat_lane_bf16((int8x8_t)__rev0, __p1, 11); \
	32305	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	32306	__ret; \
	32307	})
	32308	#define __noswap_splat_lane_bf16(__p0, __p1) __extension__ ({ \
	32309	bfloat16x4_t __ret; \
	32310	bfloat16x4_t __s0 = __p0; \
	32311	__ret = (bfloat16x4_t) __builtin_neon_splat_lane_bf16((int8x8_t)__s0, __p1, 11); \
	32312	__ret; \
32272	32313	})
32273	32314	#endif
32274	32315
32275	32316	#ifdef __LITTLE_ENDIAN__
32276		#define vdupq_n_f16(__p0) __extension__ ({ \
32277		float16x8_t __ret; \
32278		float16_t __s0 = __p0; \
32279		__ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
	32317	#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
	32318	bfloat16x8_t __ret; \
	32319	bfloat16x8_t __s0 = __p0; \
	32320	__ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_bf16((int8x16_t)__s0, __p1, 43); \
32280	32321	__ret; \
32281	32322	})
32282	32323	#else
32283		#define vdupq_n_f16(__p0) __extension__ ({ \
32284		float16x8_t __ret; \
32285		float16_t __s0 = __p0; \
32286		__ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
	32324	#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
	32325	bfloat16x8_t __ret; \
	32326	bfloat16x8_t __s0 = __p0; \
	32327	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	32328	__ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_bf16((int8x16_t)__rev0, __p1, 43); \
32287	32329	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
32288	32330	__ret; \
32289	32331	})
	32332	#define __noswap_splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
	32333	bfloat16x8_t __ret; \
	32334	bfloat16x8_t __s0 = __p0; \
	32335	__ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_bf16((int8x16_t)__s0, __p1, 43); \
	32336	__ret; \
	32337	})
32290	32338	#endif
32291	32339
32292	32340	#ifdef __LITTLE_ENDIAN__
32293		#define vdup_n_f16(__p0) __extension__ ({ \
32294		float16x4_t __ret; \
32295		float16_t __s0 = __p0; \
32296		__ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
	32341	#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
	32342	bfloat16x4_t __ret; \
	32343	bfloat16x8_t __s0 = __p0; \
	32344	__ret = (bfloat16x4_t) __builtin_neon_splat_laneq_bf16((int8x16_t)__s0, __p1, 43); \
32297	32345	__ret; \
32298	32346	})
32299	32347	#else
32300		#define vdup_n_f16(__p0) __extension__ ({ \
32301		float16x4_t __ret; \
32302		float16_t __s0 = __p0; \
32303		__ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
	32348	#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
	32349	bfloat16x4_t __ret; \
	32350	bfloat16x8_t __s0 = __p0; \
	32351	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	32352	__ret = (bfloat16x4_t) __builtin_neon_splat_laneq_bf16((int8x16_t)__rev0, __p1, 43); \
32304	32353	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
32305	32354	__ret; \
32306	32355	})
	32356	#define __noswap_splat_laneq_bf16(__p0, __p1) __extension__ ({ \
	32357	bfloat16x4_t __ret; \
	32358	bfloat16x8_t __s0 = __p0; \
	32359	__ret = (bfloat16x4_t) __builtin_neon_splat_laneq_bf16((int8x16_t)__s0, __p1, 43); \
	32360	__ret; \
	32361	})
32307	32362	#endif
32308	32363
32309	32364	#ifdef __LITTLE_ENDIAN__
32310		#define vmovq_n_f16(__p0) __extension__ ({ \
32311		float16x8_t __ret; \
32312		float16_t __s0 = __p0; \
32313		__ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
32314		__ret; \
32315		})
	32365	__ai __attribute__((target("bf16"))) float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32366	float32x4_t __ret;
	32367	__ret = (float32x4_t) __builtin_neon_vbfdotq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	32368	return __ret;
	32369	}
32316	32370	#else
32317		#define vmovq_n_f16(__p0) __extension__ ({ \
32318		float16x8_t __ret; \
32319		float16_t __s0 = __p0; \
32320		__ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
32321		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
32322		__ret; \
32323		})
	32371	__ai __attribute__((target("bf16"))) float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32372	float32x4_t __ret;
	32373	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	32374	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	32375	bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	32376	__ret = (float32x4_t) __builtin_neon_vbfdotq_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	32377	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	32378	return __ret;
	32379	}
	32380	__ai __attribute__((target("bf16"))) float32x4_t __noswap_vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32381	float32x4_t __ret;
	32382	__ret = (float32x4_t) __builtin_neon_vbfdotq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	32383	return __ret;
	32384	}
32324	32385	#endif
32325	32386
32326	32387	#ifdef __LITTLE_ENDIAN__
32327		#define vmov_n_f16(__p0) __extension__ ({ \
32328		float16x4_t __ret; \
32329		float16_t __s0 = __p0; \
32330		__ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
32331		__ret; \
32332		})
	32388	__ai __attribute__((target("bf16"))) float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
	32389	float32x2_t __ret;
	32390	__ret = (float32x2_t) __builtin_neon_vbfdot_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	32391	return __ret;
	32392	}
32333	32393	#else
32334		#define vmov_n_f16(__p0) __extension__ ({ \
32335		float16x4_t __ret; \
32336		float16_t __s0 = __p0; \
32337		__ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
32338		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
32339		__ret; \
32340		})
	32394	__ai __attribute__((target("bf16"))) float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
	32395	float32x2_t __ret;
	32396	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	32397	bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	32398	bfloat16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	32399	__ret = (float32x2_t) __builtin_neon_vbfdot_f32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	32400	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	32401	return __ret;
	32402	}
	32403	__ai __attribute__((target("bf16"))) float32x2_t __noswap_vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
	32404	float32x2_t __ret;
	32405	__ret = (float32x2_t) __builtin_neon_vbfdot_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	32406	return __ret;
	32407	}
32341	32408	#endif
32342	32409
32343	32410	#ifdef __LITTLE_ENDIAN__
32344		#define vqdmulhq_lane_s32(__p0_126, __p1_126, __p2_126) __extension__ ({ \
32345		int32x4_t __ret_126; \
32346		int32x4_t __s0_126 = __p0_126; \
32347		int32x2_t __s1_126 = __p1_126; \
32348		__ret_126 = vqdmulhq_s32(__s0_126, splatq_lane_s32(__s1_126, __p2_126)); \
	32411	#define vbfdotq_lane_f32(__p0_126, __p1_126, __p2_126, __p3_126) __extension__ ({ \
	32412	float32x4_t __ret_126; \
	32413	float32x4_t __s0_126 = __p0_126; \
	32414	bfloat16x8_t __s1_126 = __p1_126; \
	32415	bfloat16x4_t __s2_126 = __p2_126; \
	32416	bfloat16x4_t __reint_126 = __s2_126; \
	32417	float32x4_t __reint1_126 = splatq_lane_f32((float32x2_t ) &__reint_126, __p3_126); \
	32418	__ret_126 = vbfdotq_f32(__s0_126, __s1_126, (bfloat16x8_t ) &__reint1_126); \
32349	32419	__ret_126; \
32350	32420	})
32351	32421	#else
32352		#define vqdmulhq_lane_s32(__p0_127, __p1_127, __p2_127) __extension__ ({ \
32353		int32x4_t __ret_127; \
32354		int32x4_t __s0_127 = __p0_127; \
32355		int32x2_t __s1_127 = __p1_127; \
32356		int32x4_t __rev0_127; __rev0_127 = __builtin_shufflevector(__s0_127, __s0_127, 3, 2, 1, 0); \
32357		int32x2_t __rev1_127; __rev1_127 = __builtin_shufflevector(__s1_127, __s1_127, 1, 0); \
32358		__ret_127 = __noswap_vqdmulhq_s32(__rev0_127, __noswap_splatq_lane_s32(__rev1_127, __p2_127)); \
	32422	#define vbfdotq_lane_f32(__p0_127, __p1_127, __p2_127, __p3_127) __extension__ ({ \
	32423	float32x4_t __ret_127; \
	32424	float32x4_t __s0_127 = __p0_127; \
	32425	bfloat16x8_t __s1_127 = __p1_127; \
	32426	bfloat16x4_t __s2_127 = __p2_127; \
	32427	float32x4_t __rev0_127; __rev0_127 = __builtin_shufflevector(__s0_127, __s0_127, 3, 2, 1, 0); \
	32428	bfloat16x8_t __rev1_127; __rev1_127 = __builtin_shufflevector(__s1_127, __s1_127, 7, 6, 5, 4, 3, 2, 1, 0); \
	32429	bfloat16x4_t __rev2_127; __rev2_127 = __builtin_shufflevector(__s2_127, __s2_127, 3, 2, 1, 0); \
	32430	bfloat16x4_t __reint_127 = __rev2_127; \
	32431	float32x4_t __reint1_127 = __noswap_splatq_lane_f32((float32x2_t ) &__reint_127, __p3_127); \
	32432	__ret_127 = __noswap_vbfdotq_f32(__rev0_127, __rev1_127, (bfloat16x8_t ) &__reint1_127); \
32359	32433	__ret_127 = __builtin_shufflevector(__ret_127, __ret_127, 3, 2, 1, 0); \
32360	32434	__ret_127; \
32361	32435	})
32362	32436	#endif
32363	32437
32364	32438	#ifdef __LITTLE_ENDIAN__
32365		#define vqdmulhq_lane_s16(__p0_128, __p1_128, __p2_128) __extension__ ({ \
32366		int16x8_t __ret_128; \
32367		int16x8_t __s0_128 = __p0_128; \
32368		int16x4_t __s1_128 = __p1_128; \
32369		__ret_128 = vqdmulhq_s16(__s0_128, splatq_lane_s16(__s1_128, __p2_128)); \
	32439	#define vbfdot_lane_f32(__p0_128, __p1_128, __p2_128, __p3_128) __extension__ ({ \
	32440	float32x2_t __ret_128; \
	32441	float32x2_t __s0_128 = __p0_128; \
	32442	bfloat16x4_t __s1_128 = __p1_128; \
	32443	bfloat16x4_t __s2_128 = __p2_128; \
	32444	bfloat16x4_t __reint_128 = __s2_128; \
	32445	float32x2_t __reint1_128 = splat_lane_f32((float32x2_t ) &__reint_128, __p3_128); \
	32446	__ret_128 = vbfdot_f32(__s0_128, __s1_128, (bfloat16x4_t ) &__reint1_128); \
32370	32447	__ret_128; \
32371	32448	})
32372	32449	#else
32373		#define vqdmulhq_lane_s16(__p0_129, __p1_129, __p2_129) __extension__ ({ \
32374		int16x8_t __ret_129; \
32375		int16x8_t __s0_129 = __p0_129; \
32376		int16x4_t __s1_129 = __p1_129; \
32377		int16x8_t __rev0_129; __rev0_129 = __builtin_shufflevector(__s0_129, __s0_129, 7, 6, 5, 4, 3, 2, 1, 0); \
32378		int16x4_t __rev1_129; __rev1_129 = __builtin_shufflevector(__s1_129, __s1_129, 3, 2, 1, 0); \
32379		__ret_129 = __noswap_vqdmulhq_s16(__rev0_129, __noswap_splatq_lane_s16(__rev1_129, __p2_129)); \
32380		__ret_129 = __builtin_shufflevector(__ret_129, __ret_129, 7, 6, 5, 4, 3, 2, 1, 0); \
	32450	#define vbfdot_lane_f32(__p0_129, __p1_129, __p2_129, __p3_129) __extension__ ({ \
	32451	float32x2_t __ret_129; \
	32452	float32x2_t __s0_129 = __p0_129; \
	32453	bfloat16x4_t __s1_129 = __p1_129; \
	32454	bfloat16x4_t __s2_129 = __p2_129; \
	32455	float32x2_t __rev0_129; __rev0_129 = __builtin_shufflevector(__s0_129, __s0_129, 1, 0); \
	32456	bfloat16x4_t __rev1_129; __rev1_129 = __builtin_shufflevector(__s1_129, __s1_129, 3, 2, 1, 0); \
	32457	bfloat16x4_t __rev2_129; __rev2_129 = __builtin_shufflevector(__s2_129, __s2_129, 3, 2, 1, 0); \
	32458	bfloat16x4_t __reint_129 = __rev2_129; \
	32459	float32x2_t __reint1_129 = __noswap_splat_lane_f32((float32x2_t ) &__reint_129, __p3_129); \
	32460	__ret_129 = __noswap_vbfdot_f32(__rev0_129, __rev1_129, (bfloat16x4_t ) &__reint1_129); \
	32461	__ret_129 = __builtin_shufflevector(__ret_129, __ret_129, 1, 0); \
32381	32462	__ret_129; \
32382	32463	})
32383	32464	#endif
32384	32465
32385	32466	#ifdef __LITTLE_ENDIAN__
32386		#define vqdmulh_lane_s32(__p0_130, __p1_130, __p2_130) __extension__ ({ \
32387		int32x2_t __ret_130; \
32388		int32x2_t __s0_130 = __p0_130; \
32389		int32x2_t __s1_130 = __p1_130; \
32390		__ret_130 = vqdmulh_s32(__s0_130, splat_lane_s32(__s1_130, __p2_130)); \
	32467	#define vbfdotq_laneq_f32(__p0_130, __p1_130, __p2_130, __p3_130) __extension__ ({ \
	32468	float32x4_t __ret_130; \
	32469	float32x4_t __s0_130 = __p0_130; \
	32470	bfloat16x8_t __s1_130 = __p1_130; \
	32471	bfloat16x8_t __s2_130 = __p2_130; \
	32472	bfloat16x8_t __reint_130 = __s2_130; \
	32473	float32x4_t __reint1_130 = splatq_laneq_f32((float32x4_t ) &__reint_130, __p3_130); \
	32474	__ret_130 = vbfdotq_f32(__s0_130, __s1_130, (bfloat16x8_t ) &__reint1_130); \
32391	32475	__ret_130; \
32392	32476	})
32393	32477	#else
32394		#define vqdmulh_lane_s32(__p0_131, __p1_131, __p2_131) __extension__ ({ \
32395		int32x2_t __ret_131; \
32396		int32x2_t __s0_131 = __p0_131; \
32397		int32x2_t __s1_131 = __p1_131; \
32398		int32x2_t __rev0_131; __rev0_131 = __builtin_shufflevector(__s0_131, __s0_131, 1, 0); \
32399		int32x2_t __rev1_131; __rev1_131 = __builtin_shufflevector(__s1_131, __s1_131, 1, 0); \
32400		__ret_131 = __noswap_vqdmulh_s32(__rev0_131, __noswap_splat_lane_s32(__rev1_131, __p2_131)); \
32401		__ret_131 = __builtin_shufflevector(__ret_131, __ret_131, 1, 0); \
	32478	#define vbfdotq_laneq_f32(__p0_131, __p1_131, __p2_131, __p3_131) __extension__ ({ \
	32479	float32x4_t __ret_131; \
	32480	float32x4_t __s0_131 = __p0_131; \
	32481	bfloat16x8_t __s1_131 = __p1_131; \
	32482	bfloat16x8_t __s2_131 = __p2_131; \
	32483	float32x4_t __rev0_131; __rev0_131 = __builtin_shufflevector(__s0_131, __s0_131, 3, 2, 1, 0); \
	32484	bfloat16x8_t __rev1_131; __rev1_131 = __builtin_shufflevector(__s1_131, __s1_131, 7, 6, 5, 4, 3, 2, 1, 0); \
	32485	bfloat16x8_t __rev2_131; __rev2_131 = __builtin_shufflevector(__s2_131, __s2_131, 7, 6, 5, 4, 3, 2, 1, 0); \
	32486	bfloat16x8_t __reint_131 = __rev2_131; \
	32487	float32x4_t __reint1_131 = __noswap_splatq_laneq_f32((float32x4_t ) &__reint_131, __p3_131); \
	32488	__ret_131 = __noswap_vbfdotq_f32(__rev0_131, __rev1_131, (bfloat16x8_t ) &__reint1_131); \
	32489	__ret_131 = __builtin_shufflevector(__ret_131, __ret_131, 3, 2, 1, 0); \
32402	32490	__ret_131; \
32403	32491	})
32404	32492	#endif
32405	32493
32406	32494	#ifdef __LITTLE_ENDIAN__
32407		#define vqdmulh_lane_s16(__p0_132, __p1_132, __p2_132) __extension__ ({ \
32408		int16x4_t __ret_132; \
32409		int16x4_t __s0_132 = __p0_132; \
32410		int16x4_t __s1_132 = __p1_132; \
32411		__ret_132 = vqdmulh_s16(__s0_132, splat_lane_s16(__s1_132, __p2_132)); \
	32495	#define vbfdot_laneq_f32(__p0_132, __p1_132, __p2_132, __p3_132) __extension__ ({ \
	32496	float32x2_t __ret_132; \
	32497	float32x2_t __s0_132 = __p0_132; \
	32498	bfloat16x4_t __s1_132 = __p1_132; \
	32499	bfloat16x8_t __s2_132 = __p2_132; \
	32500	bfloat16x8_t __reint_132 = __s2_132; \
	32501	float32x2_t __reint1_132 = splat_laneq_f32((float32x4_t ) &__reint_132, __p3_132); \
	32502	__ret_132 = vbfdot_f32(__s0_132, __s1_132, (bfloat16x4_t ) &__reint1_132); \
32412	32503	__ret_132; \
32413	32504	})
32414	32505	#else
32415		#define vqdmulh_lane_s16(__p0_133, __p1_133, __p2_133) __extension__ ({ \
32416		int16x4_t __ret_133; \
32417		int16x4_t __s0_133 = __p0_133; \
32418		int16x4_t __s1_133 = __p1_133; \
32419		int16x4_t __rev0_133; __rev0_133 = __builtin_shufflevector(__s0_133, __s0_133, 3, 2, 1, 0); \
32420		int16x4_t __rev1_133; __rev1_133 = __builtin_shufflevector(__s1_133, __s1_133, 3, 2, 1, 0); \
32421		__ret_133 = __noswap_vqdmulh_s16(__rev0_133, __noswap_splat_lane_s16(__rev1_133, __p2_133)); \
32422		__ret_133 = __builtin_shufflevector(__ret_133, __ret_133, 3, 2, 1, 0); \
	32506	#define vbfdot_laneq_f32(__p0_133, __p1_133, __p2_133, __p3_133) __extension__ ({ \
	32507	float32x2_t __ret_133; \
	32508	float32x2_t __s0_133 = __p0_133; \
	32509	bfloat16x4_t __s1_133 = __p1_133; \
	32510	bfloat16x8_t __s2_133 = __p2_133; \
	32511	float32x2_t __rev0_133; __rev0_133 = __builtin_shufflevector(__s0_133, __s0_133, 1, 0); \
	32512	bfloat16x4_t __rev1_133; __rev1_133 = __builtin_shufflevector(__s1_133, __s1_133, 3, 2, 1, 0); \
	32513	bfloat16x8_t __rev2_133; __rev2_133 = __builtin_shufflevector(__s2_133, __s2_133, 7, 6, 5, 4, 3, 2, 1, 0); \
	32514	bfloat16x8_t __reint_133 = __rev2_133; \
	32515	float32x2_t __reint1_133 = __noswap_splat_laneq_f32((float32x4_t ) &__reint_133, __p3_133); \
	32516	__ret_133 = __noswap_vbfdot_f32(__rev0_133, __rev1_133, (bfloat16x4_t ) &__reint1_133); \
	32517	__ret_133 = __builtin_shufflevector(__ret_133, __ret_133, 1, 0); \
32423	32518	__ret_133; \
32424	32519	})
32425	32520	#endif
32426	32521
32427	32522	#ifdef __LITTLE_ENDIAN__
32428		#define vqrdmulhq_lane_s32(__p0_134, __p1_134, __p2_134) __extension__ ({ \
32429		int32x4_t __ret_134; \
32430		int32x4_t __s0_134 = __p0_134; \
32431		int32x2_t __s1_134 = __p1_134; \
32432		__ret_134 = vqrdmulhq_s32(__s0_134, splatq_lane_s32(__s1_134, __p2_134)); \
32433		__ret_134; \
32434		})
32435		#else
32436		#define vqrdmulhq_lane_s32(__p0_135, __p1_135, __p2_135) __extension__ ({ \
32437		int32x4_t __ret_135; \
32438		int32x4_t __s0_135 = __p0_135; \
32439		int32x2_t __s1_135 = __p1_135; \
32440		int32x4_t __rev0_135; __rev0_135 = __builtin_shufflevector(__s0_135, __s0_135, 3, 2, 1, 0); \
32441		int32x2_t __rev1_135; __rev1_135 = __builtin_shufflevector(__s1_135, __s1_135, 1, 0); \
32442		__ret_135 = __noswap_vqrdmulhq_s32(__rev0_135, __noswap_splatq_lane_s32(__rev1_135, __p2_135)); \
32443		__ret_135 = __builtin_shufflevector(__ret_135, __ret_135, 3, 2, 1, 0); \
32444		__ret_135; \
32445		})
32446		#endif
32447
32448		#ifdef __LITTLE_ENDIAN__
32449		#define vqrdmulhq_lane_s16(__p0_136, __p1_136, __p2_136) __extension__ ({ \
32450		int16x8_t __ret_136; \
32451		int16x8_t __s0_136 = __p0_136; \
32452		int16x4_t __s1_136 = __p1_136; \
32453		__ret_136 = vqrdmulhq_s16(__s0_136, splatq_lane_s16(__s1_136, __p2_136)); \
32454		__ret_136; \
32455		})
32456		#else
32457		#define vqrdmulhq_lane_s16(__p0_137, __p1_137, __p2_137) __extension__ ({ \
32458		int16x8_t __ret_137; \
32459		int16x8_t __s0_137 = __p0_137; \
32460		int16x4_t __s1_137 = __p1_137; \
32461		int16x8_t __rev0_137; __rev0_137 = __builtin_shufflevector(__s0_137, __s0_137, 7, 6, 5, 4, 3, 2, 1, 0); \
32462		int16x4_t __rev1_137; __rev1_137 = __builtin_shufflevector(__s1_137, __s1_137, 3, 2, 1, 0); \
32463		__ret_137 = __noswap_vqrdmulhq_s16(__rev0_137, __noswap_splatq_lane_s16(__rev1_137, __p2_137)); \
32464		__ret_137 = __builtin_shufflevector(__ret_137, __ret_137, 7, 6, 5, 4, 3, 2, 1, 0); \
32465		__ret_137; \
32466		})
32467		#endif
32468
32469		#ifdef __LITTLE_ENDIAN__
32470		#define vqrdmulh_lane_s32(__p0_138, __p1_138, __p2_138) __extension__ ({ \
32471		int32x2_t __ret_138; \
32472		int32x2_t __s0_138 = __p0_138; \
32473		int32x2_t __s1_138 = __p1_138; \
32474		__ret_138 = vqrdmulh_s32(__s0_138, splat_lane_s32(__s1_138, __p2_138)); \
32475		__ret_138; \
32476		})
32477		#else
32478		#define vqrdmulh_lane_s32(__p0_139, __p1_139, __p2_139) __extension__ ({ \
32479		int32x2_t __ret_139; \
32480		int32x2_t __s0_139 = __p0_139; \
32481		int32x2_t __s1_139 = __p1_139; \
32482		int32x2_t __rev0_139; __rev0_139 = __builtin_shufflevector(__s0_139, __s0_139, 1, 0); \
32483		int32x2_t __rev1_139; __rev1_139 = __builtin_shufflevector(__s1_139, __s1_139, 1, 0); \
32484		__ret_139 = __noswap_vqrdmulh_s32(__rev0_139, __noswap_splat_lane_s32(__rev1_139, __p2_139)); \
32485		__ret_139 = __builtin_shufflevector(__ret_139, __ret_139, 1, 0); \
32486		__ret_139; \
32487		})
32488		#endif
32489
32490		#ifdef __LITTLE_ENDIAN__
32491		#define vqrdmulh_lane_s16(__p0_140, __p1_140, __p2_140) __extension__ ({ \
32492		int16x4_t __ret_140; \
32493		int16x4_t __s0_140 = __p0_140; \
32494		int16x4_t __s1_140 = __p1_140; \
32495		__ret_140 = vqrdmulh_s16(__s0_140, splat_lane_s16(__s1_140, __p2_140)); \
32496		__ret_140; \
32497		})
32498		#else
32499		#define vqrdmulh_lane_s16(__p0_141, __p1_141, __p2_141) __extension__ ({ \
32500		int16x4_t __ret_141; \
32501		int16x4_t __s0_141 = __p0_141; \
32502		int16x4_t __s1_141 = __p1_141; \
32503		int16x4_t __rev0_141; __rev0_141 = __builtin_shufflevector(__s0_141, __s0_141, 3, 2, 1, 0); \
32504		int16x4_t __rev1_141; __rev1_141 = __builtin_shufflevector(__s1_141, __s1_141, 3, 2, 1, 0); \
32505		__ret_141 = __noswap_vqrdmulh_s16(__rev0_141, __noswap_splat_lane_s16(__rev1_141, __p2_141)); \
32506		__ret_141 = __builtin_shufflevector(__ret_141, __ret_141, 3, 2, 1, 0); \
32507		__ret_141; \
32508		})
32509		#endif
32510
32511		__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
32512		poly8x8_t __ret;
32513		__ret = (poly8x8_t)(__p0);
32514		return __ret;
32515		}
32516		__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
32517		poly8x8_t __ret;
32518		__ret = (poly8x8_t)(__p0);
32519		return __ret;
32520		}
32521		__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
32522		poly8x8_t __ret;
32523		__ret = (poly8x8_t)(__p0);
32524		return __ret;
32525		}
32526		__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
32527		poly8x8_t __ret;
32528		__ret = (poly8x8_t)(__p0);
32529		return __ret;
32530		}
32531		__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
32532		poly8x8_t __ret;
32533		__ret = (poly8x8_t)(__p0);
32534		return __ret;
32535		}
32536		__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
32537		poly8x8_t __ret;
32538		__ret = (poly8x8_t)(__p0);
32539		return __ret;
32540		}
32541		__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
32542		poly8x8_t __ret;
32543		__ret = (poly8x8_t)(__p0);
32544		return __ret;
32545		}
32546		__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
32547		poly8x8_t __ret;
32548		__ret = (poly8x8_t)(__p0);
32549		return __ret;
32550		}
32551		__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
32552		poly8x8_t __ret;
32553		__ret = (poly8x8_t)(__p0);
32554		return __ret;
32555		}
32556		__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
32557		poly8x8_t __ret;
32558		__ret = (poly8x8_t)(__p0);
32559		return __ret;
32560		}
32561		__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
32562		poly8x8_t __ret;
32563		__ret = (poly8x8_t)(__p0);
32564		return __ret;
32565		}
32566		__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
32567		poly16x4_t __ret;
32568		__ret = (poly16x4_t)(__p0);
32569		return __ret;
32570		}
32571		__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
32572		poly16x4_t __ret;
32573		__ret = (poly16x4_t)(__p0);
	32523	__ai __attribute__((target("bf16"))) float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32524	float32x4_t __ret;
	32525	__ret = (float32x4_t) __builtin_neon_vbfmlalbq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
32574	32526	return __ret;
32575	32527	}
32576		__ai poly16x4_t vreinterpret_p16_u32(uint32x2_t __p0) {
32577		poly16x4_t __ret;
32578		__ret = (poly16x4_t)(__p0);
	32528	#else
	32529	__ai __attribute__((target("bf16"))) float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32530	float32x4_t __ret;
	32531	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	32532	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	32533	bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	32534	__ret = (float32x4_t) __builtin_neon_vbfmlalbq_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	32535	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32579	32536	return __ret;
32580	32537	}
32581		__ai poly16x4_t vreinterpret_p16_u64(uint64x1_t __p0) {
32582		poly16x4_t __ret;
32583		__ret = (poly16x4_t)(__p0);
	32538	__ai __attribute__((target("bf16"))) float32x4_t __noswap_vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32539	float32x4_t __ret;
	32540	__ret = (float32x4_t) __builtin_neon_vbfmlalbq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
32584	32541	return __ret;
32585	32542	}
32586		__ai poly16x4_t vreinterpret_p16_u16(uint16x4_t __p0) {
32587		poly16x4_t __ret;
32588		__ret = (poly16x4_t)(__p0);
	32543	#endif
	32544
	32545	#ifdef __LITTLE_ENDIAN__
	32546	__ai __attribute__((target("bf16"))) float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32547	float32x4_t __ret;
	32548	__ret = (float32x4_t) __builtin_neon_vbfmlaltq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
32589	32549	return __ret;
32590	32550	}
32591		__ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
32592		poly16x4_t __ret;
32593		__ret = (poly16x4_t)(__p0);
	32551	#else
	32552	__ai __attribute__((target("bf16"))) float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32553	float32x4_t __ret;
	32554	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	32555	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	32556	bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	32557	__ret = (float32x4_t) __builtin_neon_vbfmlaltq_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	32558	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32594	32559	return __ret;
32595	32560	}
32596		__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
32597		poly16x4_t __ret;
32598		__ret = (poly16x4_t)(__p0);
	32561	__ai __attribute__((target("bf16"))) float32x4_t __noswap_vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32562	float32x4_t __ret;
	32563	__ret = (float32x4_t) __builtin_neon_vbfmlaltq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
32599	32564	return __ret;
32600	32565	}
32601		__ai poly16x4_t vreinterpret_p16_f16(float16x4_t __p0) {
32602		poly16x4_t __ret;
32603		__ret = (poly16x4_t)(__p0);
	32566	#endif
	32567
	32568	#ifdef __LITTLE_ENDIAN__
	32569	__ai __attribute__((target("bf16"))) float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32570	float32x4_t __ret;
	32571	__ret = (float32x4_t) __builtin_neon_vbfmmlaq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
32604	32572	return __ret;
32605	32573	}
32606		__ai poly16x4_t vreinterpret_p16_s32(int32x2_t __p0) {
32607		poly16x4_t __ret;
32608		__ret = (poly16x4_t)(__p0);
	32574	#else
	32575	__ai __attribute__((target("bf16"))) float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	32576	float32x4_t __ret;
	32577	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	32578	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	32579	bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	32580	__ret = (float32x4_t) __builtin_neon_vbfmmlaq_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	32581	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32609	32582	return __ret;
32610	32583	}
32611		__ai poly16x4_t vreinterpret_p16_s64(int64x1_t __p0) {
32612		poly16x4_t __ret;
32613		__ret = (poly16x4_t)(__p0);
	32584	#endif
	32585
	32586	#ifdef __LITTLE_ENDIAN__
	32587	__ai __attribute__((target("bf16"))) bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
	32588	bfloat16x8_t __ret;
	32589	__ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
32614	32590	return __ret;
32615	32591	}
32616		__ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
32617		poly16x4_t __ret;
32618		__ret = (poly16x4_t)(__p0);
	32592	#else
	32593	__ai __attribute__((target("bf16"))) bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
	32594	bfloat16x8_t __ret;
	32595	bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	32596	bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	32597	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
	32598	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
32619	32599	return __ret;
32620	32600	}
32621		__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
32622		poly8x16_t __ret;
32623		__ret = (poly8x16_t)(__p0);
	32601	__ai __attribute__((target("bf16"))) bfloat16x8_t __noswap_vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
	32602	bfloat16x8_t __ret;
	32603	__ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
32624	32604	return __ret;
32625	32605	}
32626		__ai poly8x16_t vreinterpretq_p8_u8(uint8x16_t __p0) {
32627		poly8x16_t __ret;
32628		__ret = (poly8x16_t)(__p0);
32629		return __ret;
	32606	#endif
	32607
	32608	#define vcreate_bf16(__p0) __extension__ ({ \
	32609	bfloat16x4_t __ret; \
	32610	uint64_t __promote = __p0; \
	32611	__ret = (bfloat16x4_t)(__promote); \
	32612	__ret; \
	32613	})
	32614	#ifdef __LITTLE_ENDIAN__
	32615	__ai __attribute__((target("bf16"))) float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_134) {
	32616	float32x4_t __ret_134;
	32617	bfloat16x4_t __reint_134 = __p0_134;
	32618	int32x4_t __reint1_134 = vshll_n_s16((int16x4_t ) &__reint_134, 16);
	32619	__ret_134 = (float32x4_t ) &__reint1_134;
	32620	return __ret_134;
32630	32621	}
32631		__ai poly8x16_t vreinterpretq_p8_u32(uint32x4_t __p0) {
32632		poly8x16_t __ret;
32633		__ret = (poly8x16_t)(__p0);
32634		return __ret;
	32622	#else
	32623	__ai __attribute__((target("bf16"))) float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_135) {
	32624	float32x4_t __ret_135;
	32625	bfloat16x4_t __rev0_135; __rev0_135 = __builtin_shufflevector(__p0_135, __p0_135, 3, 2, 1, 0);
	32626	bfloat16x4_t __reint_135 = __rev0_135;
	32627	int32x4_t __reint1_135 = __noswap_vshll_n_s16((int16x4_t ) &__reint_135, 16);
	32628	__ret_135 = (float32x4_t ) &__reint1_135;
	32629	__ret_135 = __builtin_shufflevector(__ret_135, __ret_135, 3, 2, 1, 0);
	32630	return __ret_135;
32635	32631	}
32636		__ai poly8x16_t vreinterpretq_p8_u64(uint64x2_t __p0) {
32637		poly8x16_t __ret;
32638		__ret = (poly8x16_t)(__p0);
32639		return __ret;
	32632	__ai __attribute__((target("bf16"))) float32x4_t __noswap_vcvt_f32_bf16(bfloat16x4_t __p0_136) {
	32633	float32x4_t __ret_136;
	32634	bfloat16x4_t __reint_136 = __p0_136;
	32635	int32x4_t __reint1_136 = __noswap_vshll_n_s16((int16x4_t ) &__reint_136, 16);
	32636	__ret_136 = (float32x4_t ) &__reint1_136;
	32637	return __ret_136;
32640	32638	}
32641		__ai poly8x16_t vreinterpretq_p8_u16(uint16x8_t __p0) {
32642		poly8x16_t __ret;
32643		__ret = (poly8x16_t)(__p0);
	32639	#endif
	32640
	32641	__ai __attribute__((target("bf16"))) float32_t vcvtah_f32_bf16(bfloat16_t __p0) {
	32642	float32_t __ret;
	32643	bfloat16_t __reint = __p0;
	32644	int32_t __reint1 = (int32_t ) &__reint << 16;
	32645	__ret = (float32_t ) &__reint1;
32644	32646	return __ret;
32645	32647	}
32646		__ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
32647		poly8x16_t __ret;
32648		__ret = (poly8x16_t)(__p0);
	32648	__ai __attribute__((target("bf16"))) bfloat16_t vcvth_bf16_f32(float32_t __p0) {
	32649	bfloat16_t __ret;
	32650	__ret = (bfloat16_t) __builtin_neon_vcvth_bf16_f32(__p0);
32649	32651	return __ret;
32650	32652	}
32651		__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
32652		poly8x16_t __ret;
32653		__ret = (poly8x16_t)(__p0);
	32653	#ifdef __LITTLE_ENDIAN__
	32654	#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
	32655	bfloat16_t __ret; \
	32656	bfloat16x4_t __s0 = __p0; \
	32657	__ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__s0, __p1); \
	32658	__ret; \
	32659	})
	32660	#else
	32661	#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
	32662	bfloat16_t __ret; \
	32663	bfloat16x4_t __s0 = __p0; \
	32664	bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	32665	__ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__rev0, __p1); \
	32666	__ret; \
	32667	})
	32668	#endif
	32669
	32670	#ifdef __LITTLE_ENDIAN__
	32671	#define vdupq_lane_bf16(__p0_137, __p1_137) __extension__ ({ \
	32672	bfloat16x8_t __ret_137; \
	32673	bfloat16x4_t __s0_137 = __p0_137; \
	32674	__ret_137 = splatq_lane_bf16(__s0_137, __p1_137); \
	32675	__ret_137; \
	32676	})
	32677	#else
	32678	#define vdupq_lane_bf16(__p0_138, __p1_138) __extension__ ({ \
	32679	bfloat16x8_t __ret_138; \
	32680	bfloat16x4_t __s0_138 = __p0_138; \
	32681	bfloat16x4_t __rev0_138; __rev0_138 = __builtin_shufflevector(__s0_138, __s0_138, 3, 2, 1, 0); \
	32682	__ret_138 = __noswap_splatq_lane_bf16(__rev0_138, __p1_138); \
	32683	__ret_138 = __builtin_shufflevector(__ret_138, __ret_138, 7, 6, 5, 4, 3, 2, 1, 0); \
	32684	__ret_138; \
	32685	})
	32686	#endif
	32687
	32688	#ifdef __LITTLE_ENDIAN__
	32689	#define vdup_lane_bf16(__p0_139, __p1_139) __extension__ ({ \
	32690	bfloat16x4_t __ret_139; \
	32691	bfloat16x4_t __s0_139 = __p0_139; \
	32692	__ret_139 = splat_lane_bf16(__s0_139, __p1_139); \
	32693	__ret_139; \
	32694	})
	32695	#else
	32696	#define vdup_lane_bf16(__p0_140, __p1_140) __extension__ ({ \
	32697	bfloat16x4_t __ret_140; \
	32698	bfloat16x4_t __s0_140 = __p0_140; \
	32699	bfloat16x4_t __rev0_140; __rev0_140 = __builtin_shufflevector(__s0_140, __s0_140, 3, 2, 1, 0); \
	32700	__ret_140 = __noswap_splat_lane_bf16(__rev0_140, __p1_140); \
	32701	__ret_140 = __builtin_shufflevector(__ret_140, __ret_140, 3, 2, 1, 0); \
	32702	__ret_140; \
	32703	})
	32704	#endif
	32705
	32706	#ifdef __LITTLE_ENDIAN__
	32707	#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
	32708	bfloat16_t __ret; \
	32709	bfloat16x8_t __s0 = __p0; \
	32710	__ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__s0, __p1); \
	32711	__ret; \
	32712	})
	32713	#else
	32714	#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
	32715	bfloat16_t __ret; \
	32716	bfloat16x8_t __s0 = __p0; \
	32717	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	32718	__ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__rev0, __p1); \
	32719	__ret; \
	32720	})
	32721	#endif
	32722
	32723	#ifdef __LITTLE_ENDIAN__
	32724	#define vdupq_laneq_bf16(__p0_141, __p1_141) __extension__ ({ \
	32725	bfloat16x8_t __ret_141; \
	32726	bfloat16x8_t __s0_141 = __p0_141; \
	32727	__ret_141 = splatq_laneq_bf16(__s0_141, __p1_141); \
	32728	__ret_141; \
	32729	})
	32730	#else
	32731	#define vdupq_laneq_bf16(__p0_142, __p1_142) __extension__ ({ \
	32732	bfloat16x8_t __ret_142; \
	32733	bfloat16x8_t __s0_142 = __p0_142; \
	32734	bfloat16x8_t __rev0_142; __rev0_142 = __builtin_shufflevector(__s0_142, __s0_142, 7, 6, 5, 4, 3, 2, 1, 0); \
	32735	__ret_142 = __noswap_splatq_laneq_bf16(__rev0_142, __p1_142); \
	32736	__ret_142 = __builtin_shufflevector(__ret_142, __ret_142, 7, 6, 5, 4, 3, 2, 1, 0); \
	32737	__ret_142; \
	32738	})
	32739	#endif
	32740
	32741	#ifdef __LITTLE_ENDIAN__
	32742	#define vdup_laneq_bf16(__p0_143, __p1_143) __extension__ ({ \
	32743	bfloat16x4_t __ret_143; \
	32744	bfloat16x8_t __s0_143 = __p0_143; \
	32745	__ret_143 = splat_laneq_bf16(__s0_143, __p1_143); \
	32746	__ret_143; \
	32747	})
	32748	#else
	32749	#define vdup_laneq_bf16(__p0_144, __p1_144) __extension__ ({ \
	32750	bfloat16x4_t __ret_144; \
	32751	bfloat16x8_t __s0_144 = __p0_144; \
	32752	bfloat16x8_t __rev0_144; __rev0_144 = __builtin_shufflevector(__s0_144, __s0_144, 7, 6, 5, 4, 3, 2, 1, 0); \
	32753	__ret_144 = __noswap_splat_laneq_bf16(__rev0_144, __p1_144); \
	32754	__ret_144 = __builtin_shufflevector(__ret_144, __ret_144, 3, 2, 1, 0); \
	32755	__ret_144; \
	32756	})
	32757	#endif
	32758
	32759	#ifdef __LITTLE_ENDIAN__
	32760	__ai __attribute__((target("bf16"))) bfloat16x8_t vdupq_n_bf16(bfloat16_t __p0) {
	32761	bfloat16x8_t __ret;
	32762	__ret = (bfloat16x8_t) {__p0, __p0, __p0, __p0, __p0, __p0, __p0, __p0};
32654	32763	return __ret;
32655	32764	}
32656		__ai poly8x16_t vreinterpretq_p8_f16(float16x8_t __p0) {
32657		poly8x16_t __ret;
32658		__ret = (poly8x16_t)(__p0);
	32765	#else
	32766	__ai __attribute__((target("bf16"))) bfloat16x8_t vdupq_n_bf16(bfloat16_t __p0) {
	32767	bfloat16x8_t __ret;
	32768	__ret = (bfloat16x8_t) {__p0, __p0, __p0, __p0, __p0, __p0, __p0, __p0};
	32769	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
32659	32770	return __ret;
32660	32771	}
32661		__ai poly8x16_t vreinterpretq_p8_s32(int32x4_t __p0) {
32662		poly8x16_t __ret;
32663		__ret = (poly8x16_t)(__p0);
	32772	#endif
	32773
	32774	#ifdef __LITTLE_ENDIAN__
	32775	__ai __attribute__((target("bf16"))) bfloat16x4_t vdup_n_bf16(bfloat16_t __p0) {
	32776	bfloat16x4_t __ret;
	32777	__ret = (bfloat16x4_t) {__p0, __p0, __p0, __p0};
32664	32778	return __ret;
32665	32779	}
32666		__ai poly8x16_t vreinterpretq_p8_s64(int64x2_t __p0) {
32667		poly8x16_t __ret;
32668		__ret = (poly8x16_t)(__p0);
	32780	#else
	32781	__ai __attribute__((target("bf16"))) bfloat16x4_t vdup_n_bf16(bfloat16_t __p0) {
	32782	bfloat16x4_t __ret;
	32783	__ret = (bfloat16x4_t) {__p0, __p0, __p0, __p0};
	32784	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32669	32785	return __ret;
32670	32786	}
32671		__ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
32672		poly8x16_t __ret;
32673		__ret = (poly8x16_t)(__p0);
	32787	#endif
	32788
	32789	#ifdef __LITTLE_ENDIAN__
	32790	__ai __attribute__((target("bf16"))) bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
	32791	bfloat16x4_t __ret;
	32792	__ret = __builtin_shufflevector(__p0, __p0, 4, 5, 6, 7);
32674	32793	return __ret;
32675	32794	}
32676		__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
32677		poly16x8_t __ret;
32678		__ret = (poly16x8_t)(__p0);
	32795	#else
	32796	__ai __attribute__((target("bf16"))) bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
	32797	bfloat16x4_t __ret;
	32798	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	32799	__ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
	32800	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32679	32801	return __ret;
32680	32802	}
32681		__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
32682		poly16x8_t __ret;
32683		__ret = (poly16x8_t)(__p0);
	32803	__ai __attribute__((target("bf16"))) bfloat16x4_t __noswap_vget_high_bf16(bfloat16x8_t __p0) {
	32804	bfloat16x4_t __ret;
	32805	__ret = __builtin_shufflevector(__p0, __p0, 4, 5, 6, 7);
32684	32806	return __ret;
32685	32807	}
32686		__ai poly16x8_t vreinterpretq_p16_u32(uint32x4_t __p0) {
32687		poly16x8_t __ret;
32688		__ret = (poly16x8_t)(__p0);
	32808	#endif
	32809
	32810	#ifdef __LITTLE_ENDIAN__
	32811	#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
	32812	bfloat16_t __ret; \
	32813	bfloat16x8_t __s0 = __p0; \
	32814	__ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
	32815	__ret; \
	32816	})
	32817	#else
	32818	#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
	32819	bfloat16_t __ret; \
	32820	bfloat16x8_t __s0 = __p0; \
	32821	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	32822	__ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__rev0, __p1); \
	32823	__ret; \
	32824	})
	32825	#define __noswap_vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
	32826	bfloat16_t __ret; \
	32827	bfloat16x8_t __s0 = __p0; \
	32828	__ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
	32829	__ret; \
	32830	})
	32831	#endif
	32832
	32833	#ifdef __LITTLE_ENDIAN__
	32834	#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
	32835	bfloat16_t __ret; \
	32836	bfloat16x4_t __s0 = __p0; \
	32837	__ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
	32838	__ret; \
	32839	})
	32840	#else
	32841	#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
	32842	bfloat16_t __ret; \
	32843	bfloat16x4_t __s0 = __p0; \
	32844	bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	32845	__ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__rev0, __p1); \
	32846	__ret; \
	32847	})
	32848	#define __noswap_vget_lane_bf16(__p0, __p1) __extension__ ({ \
	32849	bfloat16_t __ret; \
	32850	bfloat16x4_t __s0 = __p0; \
	32851	__ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
	32852	__ret; \
	32853	})
	32854	#endif
	32855
	32856	#ifdef __LITTLE_ENDIAN__
	32857	__ai __attribute__((target("bf16"))) bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
	32858	bfloat16x4_t __ret;
	32859	__ret = __builtin_shufflevector(__p0, __p0, 0, 1, 2, 3);
32689	32860	return __ret;
32690	32861	}
32691		__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
32692		poly16x8_t __ret;
32693		__ret = (poly16x8_t)(__p0);
	32862	#else
	32863	__ai __attribute__((target("bf16"))) bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
	32864	bfloat16x4_t __ret;
	32865	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	32866	__ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
	32867	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32694	32868	return __ret;
32695	32869	}
32696		__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
32697		poly16x8_t __ret;
32698		__ret = (poly16x8_t)(__p0);
	32870	__ai __attribute__((target("bf16"))) bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
	32871	bfloat16x4_t __ret;
	32872	__ret = __builtin_shufflevector(__p0, __p0, 0, 1, 2, 3);
32699	32873	return __ret;
32700	32874	}
32701		__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
32702		poly16x8_t __ret;
32703		__ret = (poly16x8_t)(__p0);
32704		return __ret;
32705		}
32706		__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
32707		poly16x8_t __ret;
32708		__ret = (poly16x8_t)(__p0);
32709		return __ret;
32710		}
32711		__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
32712		poly16x8_t __ret;
32713		__ret = (poly16x8_t)(__p0);
32714		return __ret;
32715		}
32716		__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
32717		poly16x8_t __ret;
32718		__ret = (poly16x8_t)(__p0);
32719		return __ret;
32720		}
32721		__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
32722		poly16x8_t __ret;
32723		__ret = (poly16x8_t)(__p0);
32724		return __ret;
32725		}
32726		__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
32727		poly16x8_t __ret;
32728		__ret = (poly16x8_t)(__p0);
32729		return __ret;
32730		}
32731		__ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
32732		uint8x16_t __ret;
32733		__ret = (uint8x16_t)(__p0);
32734		return __ret;
32735		}
32736		__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
32737		uint8x16_t __ret;
32738		__ret = (uint8x16_t)(__p0);
32739		return __ret;
32740		}
32741		__ai uint8x16_t vreinterpretq_u8_u32(uint32x4_t __p0) {
32742		uint8x16_t __ret;
32743		__ret = (uint8x16_t)(__p0);
32744		return __ret;
32745		}
32746		__ai uint8x16_t vreinterpretq_u8_u64(uint64x2_t __p0) {
32747		uint8x16_t __ret;
32748		__ret = (uint8x16_t)(__p0);
32749		return __ret;
32750		}
32751		__ai uint8x16_t vreinterpretq_u8_u16(uint16x8_t __p0) {
32752		uint8x16_t __ret;
32753		__ret = (uint8x16_t)(__p0);
32754		return __ret;
32755		}
32756		__ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
32757		uint8x16_t __ret;
32758		__ret = (uint8x16_t)(__p0);
32759		return __ret;
32760		}
32761		__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
32762		uint8x16_t __ret;
32763		__ret = (uint8x16_t)(__p0);
32764		return __ret;
32765		}
32766		__ai uint8x16_t vreinterpretq_u8_f16(float16x8_t __p0) {
32767		uint8x16_t __ret;
32768		__ret = (uint8x16_t)(__p0);
32769		return __ret;
32770		}
32771		__ai uint8x16_t vreinterpretq_u8_s32(int32x4_t __p0) {
32772		uint8x16_t __ret;
32773		__ret = (uint8x16_t)(__p0);
32774		return __ret;
32775		}
32776		__ai uint8x16_t vreinterpretq_u8_s64(int64x2_t __p0) {
32777		uint8x16_t __ret;
32778		__ret = (uint8x16_t)(__p0);
32779		return __ret;
32780		}
32781		__ai uint8x16_t vreinterpretq_u8_s16(int16x8_t __p0) {
32782		uint8x16_t __ret;
32783		__ret = (uint8x16_t)(__p0);
32784		return __ret;
32785		}
32786		__ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
32787		uint32x4_t __ret;
32788		__ret = (uint32x4_t)(__p0);
32789		return __ret;
32790		}
32791		__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
32792		uint32x4_t __ret;
32793		__ret = (uint32x4_t)(__p0);
32794		return __ret;
32795		}
32796		__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
32797		uint32x4_t __ret;
32798		__ret = (uint32x4_t)(__p0);
32799		return __ret;
32800		}
32801		__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
32802		uint32x4_t __ret;
32803		__ret = (uint32x4_t)(__p0);
32804		return __ret;
32805		}
32806		__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
32807		uint32x4_t __ret;
32808		__ret = (uint32x4_t)(__p0);
32809		return __ret;
32810		}
32811		__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
32812		uint32x4_t __ret;
32813		__ret = (uint32x4_t)(__p0);
32814		return __ret;
32815		}
32816		__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
32817		uint32x4_t __ret;
32818		__ret = (uint32x4_t)(__p0);
32819		return __ret;
32820		}
32821		__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
32822		uint32x4_t __ret;
32823		__ret = (uint32x4_t)(__p0);
32824		return __ret;
32825		}
32826		__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
32827		uint32x4_t __ret;
32828		__ret = (uint32x4_t)(__p0);
32829		return __ret;
32830		}
32831		__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
32832		uint32x4_t __ret;
32833		__ret = (uint32x4_t)(__p0);
32834		return __ret;
32835		}
32836		__ai uint32x4_t vreinterpretq_u32_s16(int16x8_t __p0) {
32837		uint32x4_t __ret;
32838		__ret = (uint32x4_t)(__p0);
32839		return __ret;
32840		}
32841		__ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
32842		uint64x2_t __ret;
32843		__ret = (uint64x2_t)(__p0);
32844		return __ret;
32845		}
32846		__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
32847		uint64x2_t __ret;
32848		__ret = (uint64x2_t)(__p0);
32849		return __ret;
32850		}
32851		__ai uint64x2_t vreinterpretq_u64_u8(uint8x16_t __p0) {
32852		uint64x2_t __ret;
32853		__ret = (uint64x2_t)(__p0);
32854		return __ret;
32855		}
32856		__ai uint64x2_t vreinterpretq_u64_u32(uint32x4_t __p0) {
32857		uint64x2_t __ret;
32858		__ret = (uint64x2_t)(__p0);
32859		return __ret;
32860		}
32861		__ai uint64x2_t vreinterpretq_u64_u16(uint16x8_t __p0) {
32862		uint64x2_t __ret;
32863		__ret = (uint64x2_t)(__p0);
32864		return __ret;
32865		}
32866		__ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
32867		uint64x2_t __ret;
32868		__ret = (uint64x2_t)(__p0);
32869		return __ret;
32870		}
32871		__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
32872		uint64x2_t __ret;
32873		__ret = (uint64x2_t)(__p0);
32874		return __ret;
32875		}
32876		__ai uint64x2_t vreinterpretq_u64_f16(float16x8_t __p0) {
32877		uint64x2_t __ret;
32878		__ret = (uint64x2_t)(__p0);
32879		return __ret;
32880		}
32881		__ai uint64x2_t vreinterpretq_u64_s32(int32x4_t __p0) {
32882		uint64x2_t __ret;
32883		__ret = (uint64x2_t)(__p0);
32884		return __ret;
32885		}
32886		__ai uint64x2_t vreinterpretq_u64_s64(int64x2_t __p0) {
32887		uint64x2_t __ret;
32888		__ret = (uint64x2_t)(__p0);
32889		return __ret;
32890		}
32891		__ai uint64x2_t vreinterpretq_u64_s16(int16x8_t __p0) {
32892		uint64x2_t __ret;
32893		__ret = (uint64x2_t)(__p0);
32894		return __ret;
32895		}
32896		__ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
32897		uint16x8_t __ret;
32898		__ret = (uint16x8_t)(__p0);
32899		return __ret;
32900		}
32901		__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
32902		uint16x8_t __ret;
32903		__ret = (uint16x8_t)(__p0);
32904		return __ret;
32905		}
32906		__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
32907		uint16x8_t __ret;
32908		__ret = (uint16x8_t)(__p0);
32909		return __ret;
32910		}
32911		__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
32912		uint16x8_t __ret;
32913		__ret = (uint16x8_t)(__p0);
32914		return __ret;
32915		}
32916		__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
32917		uint16x8_t __ret;
32918		__ret = (uint16x8_t)(__p0);
32919		return __ret;
32920		}
32921		__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
32922		uint16x8_t __ret;
32923		__ret = (uint16x8_t)(__p0);
32924		return __ret;
32925		}
32926		__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
32927		uint16x8_t __ret;
32928		__ret = (uint16x8_t)(__p0);
32929		return __ret;
32930		}
32931		__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
32932		uint16x8_t __ret;
32933		__ret = (uint16x8_t)(__p0);
32934		return __ret;
32935		}
32936		__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
32937		uint16x8_t __ret;
32938		__ret = (uint16x8_t)(__p0);
32939		return __ret;
32940		}
32941		__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
32942		uint16x8_t __ret;
32943		__ret = (uint16x8_t)(__p0);
32944		return __ret;
32945		}
32946		__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
32947		uint16x8_t __ret;
32948		__ret = (uint16x8_t)(__p0);
32949		return __ret;
32950		}
32951		__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
32952		int8x16_t __ret;
32953		__ret = (int8x16_t)(__p0);
32954		return __ret;
32955		}
32956		__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
32957		int8x16_t __ret;
32958		__ret = (int8x16_t)(__p0);
32959		return __ret;
32960		}
32961		__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
32962		int8x16_t __ret;
32963		__ret = (int8x16_t)(__p0);
32964		return __ret;
32965		}
32966		__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
32967		int8x16_t __ret;
32968		__ret = (int8x16_t)(__p0);
32969		return __ret;
32970		}
32971		__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
32972		int8x16_t __ret;
32973		__ret = (int8x16_t)(__p0);
32974		return __ret;
32975		}
32976		__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
32977		int8x16_t __ret;
32978		__ret = (int8x16_t)(__p0);
32979		return __ret;
32980		}
32981		__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
32982		int8x16_t __ret;
32983		__ret = (int8x16_t)(__p0);
32984		return __ret;
32985		}
32986		__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
32987		int8x16_t __ret;
32988		__ret = (int8x16_t)(__p0);
32989		return __ret;
32990		}
32991		__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
32992		int8x16_t __ret;
32993		__ret = (int8x16_t)(__p0);
32994		return __ret;
32995		}
32996		__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
32997		int8x16_t __ret;
32998		__ret = (int8x16_t)(__p0);
32999		return __ret;
33000		}
33001		__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
33002		int8x16_t __ret;
33003		__ret = (int8x16_t)(__p0);
33004		return __ret;
33005		}
33006		__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
33007		float32x4_t __ret;
33008		__ret = (float32x4_t)(__p0);
33009		return __ret;
33010		}
33011		__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
33012		float32x4_t __ret;
33013		__ret = (float32x4_t)(__p0);
33014		return __ret;
33015		}
33016		__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
33017		float32x4_t __ret;
33018		__ret = (float32x4_t)(__p0);
33019		return __ret;
33020		}
33021		__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
33022		float32x4_t __ret;
33023		__ret = (float32x4_t)(__p0);
33024		return __ret;
33025		}
33026		__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
33027		float32x4_t __ret;
33028		__ret = (float32x4_t)(__p0);
33029		return __ret;
33030		}
33031		__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
33032		float32x4_t __ret;
33033		__ret = (float32x4_t)(__p0);
33034		return __ret;
33035		}
33036		__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
33037		float32x4_t __ret;
33038		__ret = (float32x4_t)(__p0);
33039		return __ret;
33040		}
33041		__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
33042		float32x4_t __ret;
33043		__ret = (float32x4_t)(__p0);
33044		return __ret;
33045		}
33046		__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
33047		float32x4_t __ret;
33048		__ret = (float32x4_t)(__p0);
33049		return __ret;
33050		}
33051		__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
33052		float32x4_t __ret;
33053		__ret = (float32x4_t)(__p0);
33054		return __ret;
33055		}
33056		__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
33057		float32x4_t __ret;
33058		__ret = (float32x4_t)(__p0);
33059		return __ret;
33060		}
33061		__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
33062		float16x8_t __ret;
33063		__ret = (float16x8_t)(__p0);
33064		return __ret;
33065		}
33066		__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
33067		float16x8_t __ret;
33068		__ret = (float16x8_t)(__p0);
33069		return __ret;
33070		}
33071		__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
33072		float16x8_t __ret;
33073		__ret = (float16x8_t)(__p0);
33074		return __ret;
33075		}
33076		__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
33077		float16x8_t __ret;
33078		__ret = (float16x8_t)(__p0);
33079		return __ret;
33080		}
33081		__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
33082		float16x8_t __ret;
33083		__ret = (float16x8_t)(__p0);
33084		return __ret;
33085		}
33086		__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
33087		float16x8_t __ret;
33088		__ret = (float16x8_t)(__p0);
33089		return __ret;
33090		}
33091		__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
33092		float16x8_t __ret;
33093		__ret = (float16x8_t)(__p0);
33094		return __ret;
33095		}
33096		__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
33097		float16x8_t __ret;
33098		__ret = (float16x8_t)(__p0);
33099		return __ret;
33100		}
33101		__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
33102		float16x8_t __ret;
33103		__ret = (float16x8_t)(__p0);
33104		return __ret;
33105		}
33106		__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
33107		float16x8_t __ret;
33108		__ret = (float16x8_t)(__p0);
33109		return __ret;
33110		}
33111		__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
33112		float16x8_t __ret;
33113		__ret = (float16x8_t)(__p0);
33114		return __ret;
33115		}
33116		__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
33117		int32x4_t __ret;
33118		__ret = (int32x4_t)(__p0);
33119		return __ret;
33120		}
33121		__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
33122		int32x4_t __ret;
33123		__ret = (int32x4_t)(__p0);
33124		return __ret;
33125		}
33126		__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
33127		int32x4_t __ret;
33128		__ret = (int32x4_t)(__p0);
33129		return __ret;
33130		}
33131		__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
33132		int32x4_t __ret;
33133		__ret = (int32x4_t)(__p0);
33134		return __ret;
33135		}
33136		__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
33137		int32x4_t __ret;
33138		__ret = (int32x4_t)(__p0);
33139		return __ret;
33140		}
33141		__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
33142		int32x4_t __ret;
33143		__ret = (int32x4_t)(__p0);
	32875	#endif
	32876
	32877	#ifdef __LITTLE_ENDIAN__
	32878	#define vld1q_bf16(__p0) __extension__ ({ \
	32879	bfloat16x8_t __ret; \
	32880	__ret = (bfloat16x8_t) __builtin_neon_vld1q_bf16(__p0, 43); \
	32881	__ret; \
	32882	})
	32883	#else
	32884	#define vld1q_bf16(__p0) __extension__ ({ \
	32885	bfloat16x8_t __ret; \
	32886	__ret = (bfloat16x8_t) __builtin_neon_vld1q_bf16(__p0, 43); \
	32887	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	32888	__ret; \
	32889	})
	32890	#endif
	32891
	32892	#ifdef __LITTLE_ENDIAN__
	32893	#define vld1_bf16(__p0) __extension__ ({ \
	32894	bfloat16x4_t __ret; \
	32895	__ret = (bfloat16x4_t) __builtin_neon_vld1_bf16(__p0, 11); \
	32896	__ret; \
	32897	})
	32898	#else
	32899	#define vld1_bf16(__p0) __extension__ ({ \
	32900	bfloat16x4_t __ret; \
	32901	__ret = (bfloat16x4_t) __builtin_neon_vld1_bf16(__p0, 11); \
	32902	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	32903	__ret; \
	32904	})
	32905	#endif
	32906
	32907	#ifdef __LITTLE_ENDIAN__
	32908	#define vld1q_dup_bf16(__p0) __extension__ ({ \
	32909	bfloat16x8_t __ret; \
	32910	__ret = (bfloat16x8_t) __builtin_neon_vld1q_dup_bf16(__p0, 43); \
	32911	__ret; \
	32912	})
	32913	#else
	32914	#define vld1q_dup_bf16(__p0) __extension__ ({ \
	32915	bfloat16x8_t __ret; \
	32916	__ret = (bfloat16x8_t) __builtin_neon_vld1q_dup_bf16(__p0, 43); \
	32917	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	32918	__ret; \
	32919	})
	32920	#endif
	32921
	32922	#ifdef __LITTLE_ENDIAN__
	32923	#define vld1_dup_bf16(__p0) __extension__ ({ \
	32924	bfloat16x4_t __ret; \
	32925	__ret = (bfloat16x4_t) __builtin_neon_vld1_dup_bf16(__p0, 11); \
	32926	__ret; \
	32927	})
	32928	#else
	32929	#define vld1_dup_bf16(__p0) __extension__ ({ \
	32930	bfloat16x4_t __ret; \
	32931	__ret = (bfloat16x4_t) __builtin_neon_vld1_dup_bf16(__p0, 11); \
	32932	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	32933	__ret; \
	32934	})
	32935	#endif
	32936
	32937	#ifdef __LITTLE_ENDIAN__
	32938	#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	32939	bfloat16x8_t __ret; \
	32940	bfloat16x8_t __s1 = __p1; \
	32941	__ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_bf16(__p0, (int8x16_t)__s1, __p2, 43); \
	32942	__ret; \
	32943	})
	32944	#else
	32945	#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	32946	bfloat16x8_t __ret; \
	32947	bfloat16x8_t __s1 = __p1; \
	32948	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	32949	__ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_bf16(__p0, (int8x16_t)__rev1, __p2, 43); \
	32950	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	32951	__ret; \
	32952	})
	32953	#endif
	32954
	32955	#ifdef __LITTLE_ENDIAN__
	32956	#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	32957	bfloat16x4_t __ret; \
	32958	bfloat16x4_t __s1 = __p1; \
	32959	__ret = (bfloat16x4_t) __builtin_neon_vld1_lane_bf16(__p0, (int8x8_t)__s1, __p2, 11); \
	32960	__ret; \
	32961	})
	32962	#else
	32963	#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	32964	bfloat16x4_t __ret; \
	32965	bfloat16x4_t __s1 = __p1; \
	32966	bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	32967	__ret = (bfloat16x4_t) __builtin_neon_vld1_lane_bf16(__p0, (int8x8_t)__rev1, __p2, 11); \
	32968	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	32969	__ret; \
	32970	})
	32971	#endif
	32972
	32973	#ifdef __LITTLE_ENDIAN__
	32974	#define vld1q_bf16_x2(__p0) __extension__ ({ \
	32975	bfloat16x8x2_t __ret; \
	32976	__builtin_neon_vld1q_bf16_x2(&__ret, __p0, 43); \
	32977	__ret; \
	32978	})
	32979	#else
	32980	#define vld1q_bf16_x2(__p0) __extension__ ({ \
	32981	bfloat16x8x2_t __ret; \
	32982	__builtin_neon_vld1q_bf16_x2(&__ret, __p0, 43); \
	32983	\
	32984	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	32985	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	32986	__ret; \
	32987	})
	32988	#endif
	32989
	32990	#ifdef __LITTLE_ENDIAN__
	32991	#define vld1_bf16_x2(__p0) __extension__ ({ \
	32992	bfloat16x4x2_t __ret; \
	32993	__builtin_neon_vld1_bf16_x2(&__ret, __p0, 11); \
	32994	__ret; \
	32995	})
	32996	#else
	32997	#define vld1_bf16_x2(__p0) __extension__ ({ \
	32998	bfloat16x4x2_t __ret; \
	32999	__builtin_neon_vld1_bf16_x2(&__ret, __p0, 11); \
	33000	\
	33001	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33002	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33003	__ret; \
	33004	})
	33005	#endif
	33006
	33007	#ifdef __LITTLE_ENDIAN__
	33008	#define vld1q_bf16_x3(__p0) __extension__ ({ \
	33009	bfloat16x8x3_t __ret; \
	33010	__builtin_neon_vld1q_bf16_x3(&__ret, __p0, 43); \
	33011	__ret; \
	33012	})
	33013	#else
	33014	#define vld1q_bf16_x3(__p0) __extension__ ({ \
	33015	bfloat16x8x3_t __ret; \
	33016	__builtin_neon_vld1q_bf16_x3(&__ret, __p0, 43); \
	33017	\
	33018	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33019	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33020	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33021	__ret; \
	33022	})
	33023	#endif
	33024
	33025	#ifdef __LITTLE_ENDIAN__
	33026	#define vld1_bf16_x3(__p0) __extension__ ({ \
	33027	bfloat16x4x3_t __ret; \
	33028	__builtin_neon_vld1_bf16_x3(&__ret, __p0, 11); \
	33029	__ret; \
	33030	})
	33031	#else
	33032	#define vld1_bf16_x3(__p0) __extension__ ({ \
	33033	bfloat16x4x3_t __ret; \
	33034	__builtin_neon_vld1_bf16_x3(&__ret, __p0, 11); \
	33035	\
	33036	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33037	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33038	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33039	__ret; \
	33040	})
	33041	#endif
	33042
	33043	#ifdef __LITTLE_ENDIAN__
	33044	#define vld1q_bf16_x4(__p0) __extension__ ({ \
	33045	bfloat16x8x4_t __ret; \
	33046	__builtin_neon_vld1q_bf16_x4(&__ret, __p0, 43); \
	33047	__ret; \
	33048	})
	33049	#else
	33050	#define vld1q_bf16_x4(__p0) __extension__ ({ \
	33051	bfloat16x8x4_t __ret; \
	33052	__builtin_neon_vld1q_bf16_x4(&__ret, __p0, 43); \
	33053	\
	33054	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33055	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33056	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33057	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33058	__ret; \
	33059	})
	33060	#endif
	33061
	33062	#ifdef __LITTLE_ENDIAN__
	33063	#define vld1_bf16_x4(__p0) __extension__ ({ \
	33064	bfloat16x4x4_t __ret; \
	33065	__builtin_neon_vld1_bf16_x4(&__ret, __p0, 11); \
	33066	__ret; \
	33067	})
	33068	#else
	33069	#define vld1_bf16_x4(__p0) __extension__ ({ \
	33070	bfloat16x4x4_t __ret; \
	33071	__builtin_neon_vld1_bf16_x4(&__ret, __p0, 11); \
	33072	\
	33073	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33074	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33075	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33076	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	33077	__ret; \
	33078	})
	33079	#endif
	33080
	33081	#ifdef __LITTLE_ENDIAN__
	33082	#define vld2q_bf16(__p0) __extension__ ({ \
	33083	bfloat16x8x2_t __ret; \
	33084	__builtin_neon_vld2q_bf16(&__ret, __p0, 43); \
	33085	__ret; \
	33086	})
	33087	#else
	33088	#define vld2q_bf16(__p0) __extension__ ({ \
	33089	bfloat16x8x2_t __ret; \
	33090	__builtin_neon_vld2q_bf16(&__ret, __p0, 43); \
	33091	\
	33092	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33093	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33094	__ret; \
	33095	})
	33096	#endif
	33097
	33098	#ifdef __LITTLE_ENDIAN__
	33099	#define vld2_bf16(__p0) __extension__ ({ \
	33100	bfloat16x4x2_t __ret; \
	33101	__builtin_neon_vld2_bf16(&__ret, __p0, 11); \
	33102	__ret; \
	33103	})
	33104	#else
	33105	#define vld2_bf16(__p0) __extension__ ({ \
	33106	bfloat16x4x2_t __ret; \
	33107	__builtin_neon_vld2_bf16(&__ret, __p0, 11); \
	33108	\
	33109	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33110	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33111	__ret; \
	33112	})
	33113	#endif
	33114
	33115	#ifdef __LITTLE_ENDIAN__
	33116	#define vld2q_dup_bf16(__p0) __extension__ ({ \
	33117	bfloat16x8x2_t __ret; \
	33118	__builtin_neon_vld2q_dup_bf16(&__ret, __p0, 43); \
	33119	__ret; \
	33120	})
	33121	#else
	33122	#define vld2q_dup_bf16(__p0) __extension__ ({ \
	33123	bfloat16x8x2_t __ret; \
	33124	__builtin_neon_vld2q_dup_bf16(&__ret, __p0, 43); \
	33125	\
	33126	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33127	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33128	__ret; \
	33129	})
	33130	#endif
	33131
	33132	#ifdef __LITTLE_ENDIAN__
	33133	#define vld2_dup_bf16(__p0) __extension__ ({ \
	33134	bfloat16x4x2_t __ret; \
	33135	__builtin_neon_vld2_dup_bf16(&__ret, __p0, 11); \
	33136	__ret; \
	33137	})
	33138	#else
	33139	#define vld2_dup_bf16(__p0) __extension__ ({ \
	33140	bfloat16x4x2_t __ret; \
	33141	__builtin_neon_vld2_dup_bf16(&__ret, __p0, 11); \
	33142	\
	33143	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33144	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33145	__ret; \
	33146	})
	33147	#endif
	33148
	33149	#ifdef __LITTLE_ENDIAN__
	33150	#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33151	bfloat16x8x2_t __ret; \
	33152	bfloat16x8x2_t __s1 = __p1; \
	33153	__builtin_neon_vld2q_lane_bf16(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
	33154	__ret; \
	33155	})
	33156	#else
	33157	#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33158	bfloat16x8x2_t __ret; \
	33159	bfloat16x8x2_t __s1 = __p1; \
	33160	bfloat16x8x2_t __rev1; \
	33161	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33162	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33163	__builtin_neon_vld2q_lane_bf16(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
	33164	\
	33165	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33166	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33167	__ret; \
	33168	})
	33169	#endif
	33170
	33171	#ifdef __LITTLE_ENDIAN__
	33172	#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33173	bfloat16x4x2_t __ret; \
	33174	bfloat16x4x2_t __s1 = __p1; \
	33175	__builtin_neon_vld2_lane_bf16(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
	33176	__ret; \
	33177	})
	33178	#else
	33179	#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33180	bfloat16x4x2_t __ret; \
	33181	bfloat16x4x2_t __s1 = __p1; \
	33182	bfloat16x4x2_t __rev1; \
	33183	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33184	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33185	__builtin_neon_vld2_lane_bf16(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
	33186	\
	33187	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33188	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33189	__ret; \
	33190	})
	33191	#endif
	33192
	33193	#ifdef __LITTLE_ENDIAN__
	33194	#define vld3q_bf16(__p0) __extension__ ({ \
	33195	bfloat16x8x3_t __ret; \
	33196	__builtin_neon_vld3q_bf16(&__ret, __p0, 43); \
	33197	__ret; \
	33198	})
	33199	#else
	33200	#define vld3q_bf16(__p0) __extension__ ({ \
	33201	bfloat16x8x3_t __ret; \
	33202	__builtin_neon_vld3q_bf16(&__ret, __p0, 43); \
	33203	\
	33204	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33205	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33206	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33207	__ret; \
	33208	})
	33209	#endif
	33210
	33211	#ifdef __LITTLE_ENDIAN__
	33212	#define vld3_bf16(__p0) __extension__ ({ \
	33213	bfloat16x4x3_t __ret; \
	33214	__builtin_neon_vld3_bf16(&__ret, __p0, 11); \
	33215	__ret; \
	33216	})
	33217	#else
	33218	#define vld3_bf16(__p0) __extension__ ({ \
	33219	bfloat16x4x3_t __ret; \
	33220	__builtin_neon_vld3_bf16(&__ret, __p0, 11); \
	33221	\
	33222	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33223	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33224	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33225	__ret; \
	33226	})
	33227	#endif
	33228
	33229	#ifdef __LITTLE_ENDIAN__
	33230	#define vld3q_dup_bf16(__p0) __extension__ ({ \
	33231	bfloat16x8x3_t __ret; \
	33232	__builtin_neon_vld3q_dup_bf16(&__ret, __p0, 43); \
	33233	__ret; \
	33234	})
	33235	#else
	33236	#define vld3q_dup_bf16(__p0) __extension__ ({ \
	33237	bfloat16x8x3_t __ret; \
	33238	__builtin_neon_vld3q_dup_bf16(&__ret, __p0, 43); \
	33239	\
	33240	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33241	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33242	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33243	__ret; \
	33244	})
	33245	#endif
	33246
	33247	#ifdef __LITTLE_ENDIAN__
	33248	#define vld3_dup_bf16(__p0) __extension__ ({ \
	33249	bfloat16x4x3_t __ret; \
	33250	__builtin_neon_vld3_dup_bf16(&__ret, __p0, 11); \
	33251	__ret; \
	33252	})
	33253	#else
	33254	#define vld3_dup_bf16(__p0) __extension__ ({ \
	33255	bfloat16x4x3_t __ret; \
	33256	__builtin_neon_vld3_dup_bf16(&__ret, __p0, 11); \
	33257	\
	33258	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33259	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33260	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33261	__ret; \
	33262	})
	33263	#endif
	33264
	33265	#ifdef __LITTLE_ENDIAN__
	33266	#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33267	bfloat16x8x3_t __ret; \
	33268	bfloat16x8x3_t __s1 = __p1; \
	33269	__builtin_neon_vld3q_lane_bf16(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
	33270	__ret; \
	33271	})
	33272	#else
	33273	#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33274	bfloat16x8x3_t __ret; \
	33275	bfloat16x8x3_t __s1 = __p1; \
	33276	bfloat16x8x3_t __rev1; \
	33277	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33278	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33279	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33280	__builtin_neon_vld3q_lane_bf16(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
	33281	\
	33282	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33283	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33284	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33285	__ret; \
	33286	})
	33287	#endif
	33288
	33289	#ifdef __LITTLE_ENDIAN__
	33290	#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33291	bfloat16x4x3_t __ret; \
	33292	bfloat16x4x3_t __s1 = __p1; \
	33293	__builtin_neon_vld3_lane_bf16(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
	33294	__ret; \
	33295	})
	33296	#else
	33297	#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33298	bfloat16x4x3_t __ret; \
	33299	bfloat16x4x3_t __s1 = __p1; \
	33300	bfloat16x4x3_t __rev1; \
	33301	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33302	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33303	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33304	__builtin_neon_vld3_lane_bf16(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
	33305	\
	33306	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33307	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33308	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33309	__ret; \
	33310	})
	33311	#endif
	33312
	33313	#ifdef __LITTLE_ENDIAN__
	33314	#define vld4q_bf16(__p0) __extension__ ({ \
	33315	bfloat16x8x4_t __ret; \
	33316	__builtin_neon_vld4q_bf16(&__ret, __p0, 43); \
	33317	__ret; \
	33318	})
	33319	#else
	33320	#define vld4q_bf16(__p0) __extension__ ({ \
	33321	bfloat16x8x4_t __ret; \
	33322	__builtin_neon_vld4q_bf16(&__ret, __p0, 43); \
	33323	\
	33324	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33325	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33326	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33327	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33328	__ret; \
	33329	})
	33330	#endif
	33331
	33332	#ifdef __LITTLE_ENDIAN__
	33333	#define vld4_bf16(__p0) __extension__ ({ \
	33334	bfloat16x4x4_t __ret; \
	33335	__builtin_neon_vld4_bf16(&__ret, __p0, 11); \
	33336	__ret; \
	33337	})
	33338	#else
	33339	#define vld4_bf16(__p0) __extension__ ({ \
	33340	bfloat16x4x4_t __ret; \
	33341	__builtin_neon_vld4_bf16(&__ret, __p0, 11); \
	33342	\
	33343	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33344	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33345	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33346	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	33347	__ret; \
	33348	})
	33349	#endif
	33350
	33351	#ifdef __LITTLE_ENDIAN__
	33352	#define vld4q_dup_bf16(__p0) __extension__ ({ \
	33353	bfloat16x8x4_t __ret; \
	33354	__builtin_neon_vld4q_dup_bf16(&__ret, __p0, 43); \
	33355	__ret; \
	33356	})
	33357	#else
	33358	#define vld4q_dup_bf16(__p0) __extension__ ({ \
	33359	bfloat16x8x4_t __ret; \
	33360	__builtin_neon_vld4q_dup_bf16(&__ret, __p0, 43); \
	33361	\
	33362	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33363	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33364	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33365	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33366	__ret; \
	33367	})
	33368	#endif
	33369
	33370	#ifdef __LITTLE_ENDIAN__
	33371	#define vld4_dup_bf16(__p0) __extension__ ({ \
	33372	bfloat16x4x4_t __ret; \
	33373	__builtin_neon_vld4_dup_bf16(&__ret, __p0, 11); \
	33374	__ret; \
	33375	})
	33376	#else
	33377	#define vld4_dup_bf16(__p0) __extension__ ({ \
	33378	bfloat16x4x4_t __ret; \
	33379	__builtin_neon_vld4_dup_bf16(&__ret, __p0, 11); \
	33380	\
	33381	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33382	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33383	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33384	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	33385	__ret; \
	33386	})
	33387	#endif
	33388
	33389	#ifdef __LITTLE_ENDIAN__
	33390	#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33391	bfloat16x8x4_t __ret; \
	33392	bfloat16x8x4_t __s1 = __p1; \
	33393	__builtin_neon_vld4q_lane_bf16(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 43); \
	33394	__ret; \
	33395	})
	33396	#else
	33397	#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33398	bfloat16x8x4_t __ret; \
	33399	bfloat16x8x4_t __s1 = __p1; \
	33400	bfloat16x8x4_t __rev1; \
	33401	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33402	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33403	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33404	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33405	__builtin_neon_vld4q_lane_bf16(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 43); \
	33406	\
	33407	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33408	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33409	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33410	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33411	__ret; \
	33412	})
	33413	#endif
	33414
	33415	#ifdef __LITTLE_ENDIAN__
	33416	#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33417	bfloat16x4x4_t __ret; \
	33418	bfloat16x4x4_t __s1 = __p1; \
	33419	__builtin_neon_vld4_lane_bf16(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 11); \
	33420	__ret; \
	33421	})
	33422	#else
	33423	#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33424	bfloat16x4x4_t __ret; \
	33425	bfloat16x4x4_t __s1 = __p1; \
	33426	bfloat16x4x4_t __rev1; \
	33427	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33428	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33429	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33430	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	33431	__builtin_neon_vld4_lane_bf16(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 11); \
	33432	\
	33433	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	33434	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	33435	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	33436	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	33437	__ret; \
	33438	})
	33439	#endif
	33440
	33441	#ifdef __LITTLE_ENDIAN__
	33442	#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33443	bfloat16x8_t __ret; \
	33444	bfloat16_t __s0 = __p0; \
	33445	bfloat16x8_t __s1 = __p1; \
	33446	__ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
	33447	__ret; \
	33448	})
	33449	#else
	33450	#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33451	bfloat16x8_t __ret; \
	33452	bfloat16_t __s0 = __p0; \
	33453	bfloat16x8_t __s1 = __p1; \
	33454	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	33455	__ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__rev1, __p2); \
	33456	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	33457	__ret; \
	33458	})
	33459	#define __noswap_vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33460	bfloat16x8_t __ret; \
	33461	bfloat16_t __s0 = __p0; \
	33462	bfloat16x8_t __s1 = __p1; \
	33463	__ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
	33464	__ret; \
	33465	})
	33466	#endif
	33467
	33468	#ifdef __LITTLE_ENDIAN__
	33469	#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33470	bfloat16x4_t __ret; \
	33471	bfloat16_t __s0 = __p0; \
	33472	bfloat16x4_t __s1 = __p1; \
	33473	__ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
	33474	__ret; \
	33475	})
	33476	#else
	33477	#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33478	bfloat16x4_t __ret; \
	33479	bfloat16_t __s0 = __p0; \
	33480	bfloat16x4_t __s1 = __p1; \
	33481	bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	33482	__ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__rev1, __p2); \
	33483	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	33484	__ret; \
	33485	})
	33486	#define __noswap_vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33487	bfloat16x4_t __ret; \
	33488	bfloat16_t __s0 = __p0; \
	33489	bfloat16x4_t __s1 = __p1; \
	33490	__ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
	33491	__ret; \
	33492	})
	33493	#endif
	33494
	33495	#ifdef __LITTLE_ENDIAN__
	33496	#define vst1q_bf16(__p0, __p1) __extension__ ({ \
	33497	bfloat16x8_t __s1 = __p1; \
	33498	__builtin_neon_vst1q_bf16(__p0, (int8x16_t)__s1, 43); \
	33499	})
	33500	#else
	33501	#define vst1q_bf16(__p0, __p1) __extension__ ({ \
	33502	bfloat16x8_t __s1 = __p1; \
	33503	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	33504	__builtin_neon_vst1q_bf16(__p0, (int8x16_t)__rev1, 43); \
	33505	})
	33506	#endif
	33507
	33508	#ifdef __LITTLE_ENDIAN__
	33509	#define vst1_bf16(__p0, __p1) __extension__ ({ \
	33510	bfloat16x4_t __s1 = __p1; \
	33511	__builtin_neon_vst1_bf16(__p0, (int8x8_t)__s1, 11); \
	33512	})
	33513	#else
	33514	#define vst1_bf16(__p0, __p1) __extension__ ({ \
	33515	bfloat16x4_t __s1 = __p1; \
	33516	bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	33517	__builtin_neon_vst1_bf16(__p0, (int8x8_t)__rev1, 11); \
	33518	})
	33519	#endif
	33520
	33521	#ifdef __LITTLE_ENDIAN__
	33522	#define vst1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33523	bfloat16x8_t __s1 = __p1; \
	33524	__builtin_neon_vst1q_lane_bf16(__p0, (int8x16_t)__s1, __p2, 43); \
	33525	})
	33526	#else
	33527	#define vst1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33528	bfloat16x8_t __s1 = __p1; \
	33529	bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	33530	__builtin_neon_vst1q_lane_bf16(__p0, (int8x16_t)__rev1, __p2, 43); \
	33531	})
	33532	#endif
	33533
	33534	#ifdef __LITTLE_ENDIAN__
	33535	#define vst1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33536	bfloat16x4_t __s1 = __p1; \
	33537	__builtin_neon_vst1_lane_bf16(__p0, (int8x8_t)__s1, __p2, 11); \
	33538	})
	33539	#else
	33540	#define vst1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33541	bfloat16x4_t __s1 = __p1; \
	33542	bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	33543	__builtin_neon_vst1_lane_bf16(__p0, (int8x8_t)__rev1, __p2, 11); \
	33544	})
	33545	#endif
	33546
	33547	#ifdef __LITTLE_ENDIAN__
	33548	#define vst1q_bf16_x2(__p0, __p1) __extension__ ({ \
	33549	bfloat16x8x2_t __s1 = __p1; \
	33550	__builtin_neon_vst1q_bf16_x2(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 43); \
	33551	})
	33552	#else
	33553	#define vst1q_bf16_x2(__p0, __p1) __extension__ ({ \
	33554	bfloat16x8x2_t __s1 = __p1; \
	33555	bfloat16x8x2_t __rev1; \
	33556	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33557	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33558	__builtin_neon_vst1q_bf16_x2(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 43); \
	33559	})
	33560	#endif
	33561
	33562	#ifdef __LITTLE_ENDIAN__
	33563	#define vst1_bf16_x2(__p0, __p1) __extension__ ({ \
	33564	bfloat16x4x2_t __s1 = __p1; \
	33565	__builtin_neon_vst1_bf16_x2(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 11); \
	33566	})
	33567	#else
	33568	#define vst1_bf16_x2(__p0, __p1) __extension__ ({ \
	33569	bfloat16x4x2_t __s1 = __p1; \
	33570	bfloat16x4x2_t __rev1; \
	33571	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33572	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33573	__builtin_neon_vst1_bf16_x2(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 11); \
	33574	})
	33575	#endif
	33576
	33577	#ifdef __LITTLE_ENDIAN__
	33578	#define vst1q_bf16_x3(__p0, __p1) __extension__ ({ \
	33579	bfloat16x8x3_t __s1 = __p1; \
	33580	__builtin_neon_vst1q_bf16_x3(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 43); \
	33581	})
	33582	#else
	33583	#define vst1q_bf16_x3(__p0, __p1) __extension__ ({ \
	33584	bfloat16x8x3_t __s1 = __p1; \
	33585	bfloat16x8x3_t __rev1; \
	33586	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33587	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33588	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33589	__builtin_neon_vst1q_bf16_x3(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 43); \
	33590	})
	33591	#endif
	33592
	33593	#ifdef __LITTLE_ENDIAN__
	33594	#define vst1_bf16_x3(__p0, __p1) __extension__ ({ \
	33595	bfloat16x4x3_t __s1 = __p1; \
	33596	__builtin_neon_vst1_bf16_x3(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 11); \
	33597	})
	33598	#else
	33599	#define vst1_bf16_x3(__p0, __p1) __extension__ ({ \
	33600	bfloat16x4x3_t __s1 = __p1; \
	33601	bfloat16x4x3_t __rev1; \
	33602	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33603	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33604	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33605	__builtin_neon_vst1_bf16_x3(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 11); \
	33606	})
	33607	#endif
	33608
	33609	#ifdef __LITTLE_ENDIAN__
	33610	#define vst1q_bf16_x4(__p0, __p1) __extension__ ({ \
	33611	bfloat16x8x4_t __s1 = __p1; \
	33612	__builtin_neon_vst1q_bf16_x4(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 43); \
	33613	})
	33614	#else
	33615	#define vst1q_bf16_x4(__p0, __p1) __extension__ ({ \
	33616	bfloat16x8x4_t __s1 = __p1; \
	33617	bfloat16x8x4_t __rev1; \
	33618	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33619	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33620	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33621	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33622	__builtin_neon_vst1q_bf16_x4(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 43); \
	33623	})
	33624	#endif
	33625
	33626	#ifdef __LITTLE_ENDIAN__
	33627	#define vst1_bf16_x4(__p0, __p1) __extension__ ({ \
	33628	bfloat16x4x4_t __s1 = __p1; \
	33629	__builtin_neon_vst1_bf16_x4(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 11); \
	33630	})
	33631	#else
	33632	#define vst1_bf16_x4(__p0, __p1) __extension__ ({ \
	33633	bfloat16x4x4_t __s1 = __p1; \
	33634	bfloat16x4x4_t __rev1; \
	33635	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33636	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33637	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33638	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	33639	__builtin_neon_vst1_bf16_x4(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 11); \
	33640	})
	33641	#endif
	33642
	33643	#ifdef __LITTLE_ENDIAN__
	33644	#define vst2q_bf16(__p0, __p1) __extension__ ({ \
	33645	bfloat16x8x2_t __s1 = __p1; \
	33646	__builtin_neon_vst2q_bf16(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 43); \
	33647	})
	33648	#else
	33649	#define vst2q_bf16(__p0, __p1) __extension__ ({ \
	33650	bfloat16x8x2_t __s1 = __p1; \
	33651	bfloat16x8x2_t __rev1; \
	33652	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33653	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33654	__builtin_neon_vst2q_bf16(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 43); \
	33655	})
	33656	#endif
	33657
	33658	#ifdef __LITTLE_ENDIAN__
	33659	#define vst2_bf16(__p0, __p1) __extension__ ({ \
	33660	bfloat16x4x2_t __s1 = __p1; \
	33661	__builtin_neon_vst2_bf16(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 11); \
	33662	})
	33663	#else
	33664	#define vst2_bf16(__p0, __p1) __extension__ ({ \
	33665	bfloat16x4x2_t __s1 = __p1; \
	33666	bfloat16x4x2_t __rev1; \
	33667	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33668	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33669	__builtin_neon_vst2_bf16(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 11); \
	33670	})
	33671	#endif
	33672
	33673	#ifdef __LITTLE_ENDIAN__
	33674	#define vst2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33675	bfloat16x8x2_t __s1 = __p1; \
	33676	__builtin_neon_vst2q_lane_bf16(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
	33677	})
	33678	#else
	33679	#define vst2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33680	bfloat16x8x2_t __s1 = __p1; \
	33681	bfloat16x8x2_t __rev1; \
	33682	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33683	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33684	__builtin_neon_vst2q_lane_bf16(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
	33685	})
	33686	#endif
	33687
	33688	#ifdef __LITTLE_ENDIAN__
	33689	#define vst2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33690	bfloat16x4x2_t __s1 = __p1; \
	33691	__builtin_neon_vst2_lane_bf16(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
	33692	})
	33693	#else
	33694	#define vst2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33695	bfloat16x4x2_t __s1 = __p1; \
	33696	bfloat16x4x2_t __rev1; \
	33697	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33698	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33699	__builtin_neon_vst2_lane_bf16(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
	33700	})
	33701	#endif
	33702
	33703	#ifdef __LITTLE_ENDIAN__
	33704	#define vst3q_bf16(__p0, __p1) __extension__ ({ \
	33705	bfloat16x8x3_t __s1 = __p1; \
	33706	__builtin_neon_vst3q_bf16(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 43); \
	33707	})
	33708	#else
	33709	#define vst3q_bf16(__p0, __p1) __extension__ ({ \
	33710	bfloat16x8x3_t __s1 = __p1; \
	33711	bfloat16x8x3_t __rev1; \
	33712	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33713	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33714	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33715	__builtin_neon_vst3q_bf16(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 43); \
	33716	})
	33717	#endif
	33718
	33719	#ifdef __LITTLE_ENDIAN__
	33720	#define vst3_bf16(__p0, __p1) __extension__ ({ \
	33721	bfloat16x4x3_t __s1 = __p1; \
	33722	__builtin_neon_vst3_bf16(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 11); \
	33723	})
	33724	#else
	33725	#define vst3_bf16(__p0, __p1) __extension__ ({ \
	33726	bfloat16x4x3_t __s1 = __p1; \
	33727	bfloat16x4x3_t __rev1; \
	33728	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33729	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33730	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33731	__builtin_neon_vst3_bf16(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 11); \
	33732	})
	33733	#endif
	33734
	33735	#ifdef __LITTLE_ENDIAN__
	33736	#define vst3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33737	bfloat16x8x3_t __s1 = __p1; \
	33738	__builtin_neon_vst3q_lane_bf16(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
	33739	})
	33740	#else
	33741	#define vst3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33742	bfloat16x8x3_t __s1 = __p1; \
	33743	bfloat16x8x3_t __rev1; \
	33744	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33745	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33746	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33747	__builtin_neon_vst3q_lane_bf16(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
	33748	})
	33749	#endif
	33750
	33751	#ifdef __LITTLE_ENDIAN__
	33752	#define vst3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33753	bfloat16x4x3_t __s1 = __p1; \
	33754	__builtin_neon_vst3_lane_bf16(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
	33755	})
	33756	#else
	33757	#define vst3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33758	bfloat16x4x3_t __s1 = __p1; \
	33759	bfloat16x4x3_t __rev1; \
	33760	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33761	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33762	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33763	__builtin_neon_vst3_lane_bf16(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
	33764	})
	33765	#endif
	33766
	33767	#ifdef __LITTLE_ENDIAN__
	33768	#define vst4q_bf16(__p0, __p1) __extension__ ({ \
	33769	bfloat16x8x4_t __s1 = __p1; \
	33770	__builtin_neon_vst4q_bf16(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 43); \
	33771	})
	33772	#else
	33773	#define vst4q_bf16(__p0, __p1) __extension__ ({ \
	33774	bfloat16x8x4_t __s1 = __p1; \
	33775	bfloat16x8x4_t __rev1; \
	33776	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33777	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33778	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33779	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33780	__builtin_neon_vst4q_bf16(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 43); \
	33781	})
	33782	#endif
	33783
	33784	#ifdef __LITTLE_ENDIAN__
	33785	#define vst4_bf16(__p0, __p1) __extension__ ({ \
	33786	bfloat16x4x4_t __s1 = __p1; \
	33787	__builtin_neon_vst4_bf16(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 11); \
	33788	})
	33789	#else
	33790	#define vst4_bf16(__p0, __p1) __extension__ ({ \
	33791	bfloat16x4x4_t __s1 = __p1; \
	33792	bfloat16x4x4_t __rev1; \
	33793	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33794	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33795	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33796	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	33797	__builtin_neon_vst4_bf16(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 11); \
	33798	})
	33799	#endif
	33800
	33801	#ifdef __LITTLE_ENDIAN__
	33802	#define vst4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33803	bfloat16x8x4_t __s1 = __p1; \
	33804	__builtin_neon_vst4q_lane_bf16(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 43); \
	33805	})
	33806	#else
	33807	#define vst4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33808	bfloat16x8x4_t __s1 = __p1; \
	33809	bfloat16x8x4_t __rev1; \
	33810	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	33811	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	33812	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	33813	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	33814	__builtin_neon_vst4q_lane_bf16(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 43); \
	33815	})
	33816	#endif
	33817
	33818	#ifdef __LITTLE_ENDIAN__
	33819	#define vst4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33820	bfloat16x4x4_t __s1 = __p1; \
	33821	__builtin_neon_vst4_lane_bf16(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 11); \
	33822	})
	33823	#else
	33824	#define vst4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
	33825	bfloat16x4x4_t __s1 = __p1; \
	33826	bfloat16x4x4_t __rev1; \
	33827	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	33828	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	33829	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	33830	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	33831	__builtin_neon_vst4_lane_bf16(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 11); \
	33832	})
	33833	#endif
	33834
	33835	#ifdef __LITTLE_ENDIAN__
	33836	__ai __attribute__((target("dotprod"))) uint32x4_t vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	33837	uint32x4_t __ret;
	33838	__ret = (uint32x4_t) __builtin_neon_vdotq_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
33144	33839	return __ret;
33145	33840	}
33146		__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
33147		int32x4_t __ret;
33148		__ret = (int32x4_t)(__p0);
	33841	#else
	33842	__ai __attribute__((target("dotprod"))) uint32x4_t vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	33843	uint32x4_t __ret;
	33844	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	33845	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	33846	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	33847	__ret = (uint32x4_t) __builtin_neon_vdotq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	33848	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33149	33849	return __ret;
33150	33850	}
33151		__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
33152		int32x4_t __ret;
33153		__ret = (int32x4_t)(__p0);
	33851	__ai __attribute__((target("dotprod"))) uint32x4_t __noswap_vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	33852	uint32x4_t __ret;
	33853	__ret = (uint32x4_t) __builtin_neon_vdotq_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
33154	33854	return __ret;
33155	33855	}
33156		__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
	33856	#endif
	33857
	33858	#ifdef __LITTLE_ENDIAN__
	33859	__ai __attribute__((target("dotprod"))) int32x4_t vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
33157	33860	int32x4_t __ret;
33158		__ret = (int32x4_t)(__p0);
	33861	__ret = (int32x4_t) __builtin_neon_vdotq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
33159	33862	return __ret;
33160	33863	}
33161		__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
	33864	#else
	33865	__ai __attribute__((target("dotprod"))) int32x4_t vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
33162	33866	int32x4_t __ret;
33163		__ret = (int32x4_t)(__p0);
	33867	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	33868	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	33869	int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	33870	__ret = (int32x4_t) __builtin_neon_vdotq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
	33871	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33164	33872	return __ret;
33165	33873	}
33166		__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
	33874	__ai __attribute__((target("dotprod"))) int32x4_t __noswap_vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
33167	33875	int32x4_t __ret;
33168		__ret = (int32x4_t)(__p0);
33169		return __ret;
33170		}
33171		__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
33172		int64x2_t __ret;
33173		__ret = (int64x2_t)(__p0);
33174		return __ret;
33175		}
33176		__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
33177		int64x2_t __ret;
33178		__ret = (int64x2_t)(__p0);
33179		return __ret;
33180		}
33181		__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
33182		int64x2_t __ret;
33183		__ret = (int64x2_t)(__p0);
33184		return __ret;
33185		}
33186		__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
33187		int64x2_t __ret;
33188		__ret = (int64x2_t)(__p0);
33189		return __ret;
33190		}
33191		__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
33192		int64x2_t __ret;
33193		__ret = (int64x2_t)(__p0);
33194		return __ret;
33195		}
33196		__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
33197		int64x2_t __ret;
33198		__ret = (int64x2_t)(__p0);
33199		return __ret;
33200		}
33201		__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
33202		int64x2_t __ret;
33203		__ret = (int64x2_t)(__p0);
33204		return __ret;
33205		}
33206		__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
33207		int64x2_t __ret;
33208		__ret = (int64x2_t)(__p0);
33209		return __ret;
33210		}
33211		__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
33212		int64x2_t __ret;
33213		__ret = (int64x2_t)(__p0);
33214		return __ret;
33215		}
33216		__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
33217		int64x2_t __ret;
33218		__ret = (int64x2_t)(__p0);
33219		return __ret;
33220		}
33221		__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
33222		int64x2_t __ret;
33223		__ret = (int64x2_t)(__p0);
33224		return __ret;
33225		}
33226		__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
33227		int16x8_t __ret;
33228		__ret = (int16x8_t)(__p0);
33229		return __ret;
33230		}
33231		__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
33232		int16x8_t __ret;
33233		__ret = (int16x8_t)(__p0);
33234		return __ret;
33235		}
33236		__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
33237		int16x8_t __ret;
33238		__ret = (int16x8_t)(__p0);
33239		return __ret;
33240		}
33241		__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
33242		int16x8_t __ret;
33243		__ret = (int16x8_t)(__p0);
33244		return __ret;
33245		}
33246		__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
33247		int16x8_t __ret;
33248		__ret = (int16x8_t)(__p0);
33249		return __ret;
33250		}
33251		__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
33252		int16x8_t __ret;
33253		__ret = (int16x8_t)(__p0);
33254		return __ret;
33255		}
33256		__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
33257		int16x8_t __ret;
33258		__ret = (int16x8_t)(__p0);
33259		return __ret;
33260		}
33261		__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
33262		int16x8_t __ret;
33263		__ret = (int16x8_t)(__p0);
33264		return __ret;
33265		}
33266		__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
33267		int16x8_t __ret;
33268		__ret = (int16x8_t)(__p0);
33269		return __ret;
33270		}
33271		__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
33272		int16x8_t __ret;
33273		__ret = (int16x8_t)(__p0);
33274		return __ret;
33275		}
33276		__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
33277		int16x8_t __ret;
33278		__ret = (int16x8_t)(__p0);
33279		return __ret;
33280		}
33281		__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
33282		uint8x8_t __ret;
33283		__ret = (uint8x8_t)(__p0);
33284		return __ret;
33285		}
33286		__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
33287		uint8x8_t __ret;
33288		__ret = (uint8x8_t)(__p0);
33289		return __ret;
33290		}
33291		__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
33292		uint8x8_t __ret;
33293		__ret = (uint8x8_t)(__p0);
33294		return __ret;
33295		}
33296		__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
33297		uint8x8_t __ret;
33298		__ret = (uint8x8_t)(__p0);
33299		return __ret;
33300		}
33301		__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
33302		uint8x8_t __ret;
33303		__ret = (uint8x8_t)(__p0);
33304		return __ret;
33305		}
33306		__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
33307		uint8x8_t __ret;
33308		__ret = (uint8x8_t)(__p0);
33309		return __ret;
33310		}
33311		__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
33312		uint8x8_t __ret;
33313		__ret = (uint8x8_t)(__p0);
33314		return __ret;
33315		}
33316		__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
33317		uint8x8_t __ret;
33318		__ret = (uint8x8_t)(__p0);
33319		return __ret;
33320		}
33321		__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
33322		uint8x8_t __ret;
33323		__ret = (uint8x8_t)(__p0);
33324		return __ret;
33325		}
33326		__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
33327		uint8x8_t __ret;
33328		__ret = (uint8x8_t)(__p0);
33329		return __ret;
33330		}
33331		__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
33332		uint8x8_t __ret;
33333		__ret = (uint8x8_t)(__p0);
33334		return __ret;
33335		}
33336		__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
33337		uint32x2_t __ret;
33338		__ret = (uint32x2_t)(__p0);
33339		return __ret;
33340		}
33341		__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
33342		uint32x2_t __ret;
33343		__ret = (uint32x2_t)(__p0);
33344		return __ret;
33345		}
33346		__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
33347		uint32x2_t __ret;
33348		__ret = (uint32x2_t)(__p0);
33349		return __ret;
33350		}
33351		__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
33352		uint32x2_t __ret;
33353		__ret = (uint32x2_t)(__p0);
33354		return __ret;
33355		}
33356		__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
33357		uint32x2_t __ret;
33358		__ret = (uint32x2_t)(__p0);
33359		return __ret;
33360		}
33361		__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
33362		uint32x2_t __ret;
33363		__ret = (uint32x2_t)(__p0);
33364		return __ret;
33365		}
33366		__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
33367		uint32x2_t __ret;
33368		__ret = (uint32x2_t)(__p0);
33369		return __ret;
33370		}
33371		__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
33372		uint32x2_t __ret;
33373		__ret = (uint32x2_t)(__p0);
	33876	__ret = (int32x4_t) __builtin_neon_vdotq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
33374	33877	return __ret;
33375	33878	}
33376		__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
	33879	#endif
	33880
	33881	#ifdef __LITTLE_ENDIAN__
	33882	__ai __attribute__((target("dotprod"))) uint32x2_t vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
33377	33883	uint32x2_t __ret;
33378		__ret = (uint32x2_t)(__p0);
	33884	__ret = (uint32x2_t) __builtin_neon_vdot_u32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 18);
33379	33885	return __ret;
33380	33886	}
33381		__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
	33887	#else
	33888	__ai __attribute__((target("dotprod"))) uint32x2_t vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
33382	33889	uint32x2_t __ret;
33383		__ret = (uint32x2_t)(__p0);
	33890	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	33891	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	33892	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	33893	__ret = (uint32x2_t) __builtin_neon_vdot_u32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 18);
	33894	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
33384	33895	return __ret;
33385	33896	}
33386		__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
	33897	__ai __attribute__((target("dotprod"))) uint32x2_t __noswap_vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
33387	33898	uint32x2_t __ret;
33388		__ret = (uint32x2_t)(__p0);
33389		return __ret;
33390		}
33391		__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
33392		uint64x1_t __ret;
33393		__ret = (uint64x1_t)(__p0);
33394		return __ret;
33395		}
33396		__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
33397		uint64x1_t __ret;
33398		__ret = (uint64x1_t)(__p0);
33399		return __ret;
33400		}
33401		__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
33402		uint64x1_t __ret;
33403		__ret = (uint64x1_t)(__p0);
33404		return __ret;
33405		}
33406		__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
33407		uint64x1_t __ret;
33408		__ret = (uint64x1_t)(__p0);
33409		return __ret;
33410		}
33411		__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
33412		uint64x1_t __ret;
33413		__ret = (uint64x1_t)(__p0);
33414		return __ret;
33415		}
33416		__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
33417		uint64x1_t __ret;
33418		__ret = (uint64x1_t)(__p0);
33419		return __ret;
33420		}
33421		__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
33422		uint64x1_t __ret;
33423		__ret = (uint64x1_t)(__p0);
33424		return __ret;
33425		}
33426		__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
33427		uint64x1_t __ret;
33428		__ret = (uint64x1_t)(__p0);
33429		return __ret;
33430		}
33431		__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
33432		uint64x1_t __ret;
33433		__ret = (uint64x1_t)(__p0);
33434		return __ret;
33435		}
33436		__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
33437		uint64x1_t __ret;
33438		__ret = (uint64x1_t)(__p0);
	33899	__ret = (uint32x2_t) __builtin_neon_vdot_u32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 18);
33439	33900	return __ret;
33440	33901	}
33441		__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
33442		uint64x1_t __ret;
33443		__ret = (uint64x1_t)(__p0);
	33902	#endif
	33903
	33904	#ifdef __LITTLE_ENDIAN__
	33905	__ai __attribute__((target("dotprod"))) int32x2_t vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
	33906	int32x2_t __ret;
	33907	__ret = (int32x2_t) __builtin_neon_vdot_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
33444	33908	return __ret;
33445	33909	}
33446		__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
33447		uint16x4_t __ret;
33448		__ret = (uint16x4_t)(__p0);
	33910	#else
	33911	__ai __attribute__((target("dotprod"))) int32x2_t vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
	33912	int32x2_t __ret;
	33913	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	33914	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	33915	int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	33916	__ret = (int32x2_t) __builtin_neon_vdot_s32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
	33917	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
33449	33918	return __ret;
33450	33919	}
33451		__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
33452		uint16x4_t __ret;
33453		__ret = (uint16x4_t)(__p0);
	33920	__ai __attribute__((target("dotprod"))) int32x2_t __noswap_vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
	33921	int32x2_t __ret;
	33922	__ret = (int32x2_t) __builtin_neon_vdot_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
33454	33923	return __ret;
33455	33924	}
33456		__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
33457		uint16x4_t __ret;
33458		__ret = (uint16x4_t)(__p0);
	33925	#endif
	33926
	33927	#ifdef __LITTLE_ENDIAN__
	33928	#define vdotq_lane_u32(__p0_145, __p1_145, __p2_145, __p3_145) __extension__ ({ \
	33929	uint32x4_t __ret_145; \
	33930	uint32x4_t __s0_145 = __p0_145; \
	33931	uint8x16_t __s1_145 = __p1_145; \
	33932	uint8x8_t __s2_145 = __p2_145; \
	33933	uint8x8_t __reint_145 = __s2_145; \
	33934	uint32x4_t __reint1_145 = splatq_lane_u32((uint32x2_t ) &__reint_145, __p3_145); \
	33935	__ret_145 = vdotq_u32(__s0_145, __s1_145, (uint8x16_t ) &__reint1_145); \
	33936	__ret_145; \
	33937	})
	33938	#else
	33939	#define vdotq_lane_u32(__p0_146, __p1_146, __p2_146, __p3_146) __extension__ ({ \
	33940	uint32x4_t __ret_146; \
	33941	uint32x4_t __s0_146 = __p0_146; \
	33942	uint8x16_t __s1_146 = __p1_146; \
	33943	uint8x8_t __s2_146 = __p2_146; \
	33944	uint32x4_t __rev0_146; __rev0_146 = __builtin_shufflevector(__s0_146, __s0_146, 3, 2, 1, 0); \
	33945	uint8x16_t __rev1_146; __rev1_146 = __builtin_shufflevector(__s1_146, __s1_146, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	33946	uint8x8_t __rev2_146; __rev2_146 = __builtin_shufflevector(__s2_146, __s2_146, 7, 6, 5, 4, 3, 2, 1, 0); \
	33947	uint8x8_t __reint_146 = __rev2_146; \
	33948	uint32x4_t __reint1_146 = __noswap_splatq_lane_u32((uint32x2_t ) &__reint_146, __p3_146); \
	33949	__ret_146 = __noswap_vdotq_u32(__rev0_146, __rev1_146, (uint8x16_t ) &__reint1_146); \
	33950	__ret_146 = __builtin_shufflevector(__ret_146, __ret_146, 3, 2, 1, 0); \
	33951	__ret_146; \
	33952	})
	33953	#endif
	33954
	33955	#ifdef __LITTLE_ENDIAN__
	33956	#define vdotq_lane_s32(__p0_147, __p1_147, __p2_147, __p3_147) __extension__ ({ \
	33957	int32x4_t __ret_147; \
	33958	int32x4_t __s0_147 = __p0_147; \
	33959	int8x16_t __s1_147 = __p1_147; \
	33960	int8x8_t __s2_147 = __p2_147; \
	33961	int8x8_t __reint_147 = __s2_147; \
	33962	int32x4_t __reint1_147 = splatq_lane_s32((int32x2_t ) &__reint_147, __p3_147); \
	33963	__ret_147 = vdotq_s32(__s0_147, __s1_147, (int8x16_t ) &__reint1_147); \
	33964	__ret_147; \
	33965	})
	33966	#else
	33967	#define vdotq_lane_s32(__p0_148, __p1_148, __p2_148, __p3_148) __extension__ ({ \
	33968	int32x4_t __ret_148; \
	33969	int32x4_t __s0_148 = __p0_148; \
	33970	int8x16_t __s1_148 = __p1_148; \
	33971	int8x8_t __s2_148 = __p2_148; \
	33972	int32x4_t __rev0_148; __rev0_148 = __builtin_shufflevector(__s0_148, __s0_148, 3, 2, 1, 0); \
	33973	int8x16_t __rev1_148; __rev1_148 = __builtin_shufflevector(__s1_148, __s1_148, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	33974	int8x8_t __rev2_148; __rev2_148 = __builtin_shufflevector(__s2_148, __s2_148, 7, 6, 5, 4, 3, 2, 1, 0); \
	33975	int8x8_t __reint_148 = __rev2_148; \
	33976	int32x4_t __reint1_148 = __noswap_splatq_lane_s32((int32x2_t ) &__reint_148, __p3_148); \
	33977	__ret_148 = __noswap_vdotq_s32(__rev0_148, __rev1_148, (int8x16_t ) &__reint1_148); \
	33978	__ret_148 = __builtin_shufflevector(__ret_148, __ret_148, 3, 2, 1, 0); \
	33979	__ret_148; \
	33980	})
	33981	#endif
	33982
	33983	#ifdef __LITTLE_ENDIAN__
	33984	#define vdot_lane_u32(__p0_149, __p1_149, __p2_149, __p3_149) __extension__ ({ \
	33985	uint32x2_t __ret_149; \
	33986	uint32x2_t __s0_149 = __p0_149; \
	33987	uint8x8_t __s1_149 = __p1_149; \
	33988	uint8x8_t __s2_149 = __p2_149; \
	33989	uint8x8_t __reint_149 = __s2_149; \
	33990	uint32x2_t __reint1_149 = splat_lane_u32((uint32x2_t ) &__reint_149, __p3_149); \
	33991	__ret_149 = vdot_u32(__s0_149, __s1_149, (uint8x8_t ) &__reint1_149); \
	33992	__ret_149; \
	33993	})
	33994	#else
	33995	#define vdot_lane_u32(__p0_150, __p1_150, __p2_150, __p3_150) __extension__ ({ \
	33996	uint32x2_t __ret_150; \
	33997	uint32x2_t __s0_150 = __p0_150; \
	33998	uint8x8_t __s1_150 = __p1_150; \
	33999	uint8x8_t __s2_150 = __p2_150; \
	34000	uint32x2_t __rev0_150; __rev0_150 = __builtin_shufflevector(__s0_150, __s0_150, 1, 0); \
	34001	uint8x8_t __rev1_150; __rev1_150 = __builtin_shufflevector(__s1_150, __s1_150, 7, 6, 5, 4, 3, 2, 1, 0); \
	34002	uint8x8_t __rev2_150; __rev2_150 = __builtin_shufflevector(__s2_150, __s2_150, 7, 6, 5, 4, 3, 2, 1, 0); \
	34003	uint8x8_t __reint_150 = __rev2_150; \
	34004	uint32x2_t __reint1_150 = __noswap_splat_lane_u32((uint32x2_t ) &__reint_150, __p3_150); \
	34005	__ret_150 = __noswap_vdot_u32(__rev0_150, __rev1_150, (uint8x8_t ) &__reint1_150); \
	34006	__ret_150 = __builtin_shufflevector(__ret_150, __ret_150, 1, 0); \
	34007	__ret_150; \
	34008	})
	34009	#endif
	34010
	34011	#ifdef __LITTLE_ENDIAN__
	34012	#define vdot_lane_s32(__p0_151, __p1_151, __p2_151, __p3_151) __extension__ ({ \
	34013	int32x2_t __ret_151; \
	34014	int32x2_t __s0_151 = __p0_151; \
	34015	int8x8_t __s1_151 = __p1_151; \
	34016	int8x8_t __s2_151 = __p2_151; \
	34017	int8x8_t __reint_151 = __s2_151; \
	34018	int32x2_t __reint1_151 = splat_lane_s32((int32x2_t ) &__reint_151, __p3_151); \
	34019	__ret_151 = vdot_s32(__s0_151, __s1_151, (int8x8_t ) &__reint1_151); \
	34020	__ret_151; \
	34021	})
	34022	#else
	34023	#define vdot_lane_s32(__p0_152, __p1_152, __p2_152, __p3_152) __extension__ ({ \
	34024	int32x2_t __ret_152; \
	34025	int32x2_t __s0_152 = __p0_152; \
	34026	int8x8_t __s1_152 = __p1_152; \
	34027	int8x8_t __s2_152 = __p2_152; \
	34028	int32x2_t __rev0_152; __rev0_152 = __builtin_shufflevector(__s0_152, __s0_152, 1, 0); \
	34029	int8x8_t __rev1_152; __rev1_152 = __builtin_shufflevector(__s1_152, __s1_152, 7, 6, 5, 4, 3, 2, 1, 0); \
	34030	int8x8_t __rev2_152; __rev2_152 = __builtin_shufflevector(__s2_152, __s2_152, 7, 6, 5, 4, 3, 2, 1, 0); \
	34031	int8x8_t __reint_152 = __rev2_152; \
	34032	int32x2_t __reint1_152 = __noswap_splat_lane_s32((int32x2_t ) &__reint_152, __p3_152); \
	34033	__ret_152 = __noswap_vdot_s32(__rev0_152, __rev1_152, (int8x8_t ) &__reint1_152); \
	34034	__ret_152 = __builtin_shufflevector(__ret_152, __ret_152, 1, 0); \
	34035	__ret_152; \
	34036	})
	34037	#endif
	34038
	34039	#ifdef __LITTLE_ENDIAN__
	34040	__ai __attribute__((target("fullfp16"))) float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
	34041	float16x8_t __ret;
	34042	__ret = (float16x8_t) __builtin_neon_vabdq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
33459	34043	return __ret;
33460	34044	}
33461		__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
33462		uint16x4_t __ret;
33463		__ret = (uint16x4_t)(__p0);
	34045	#else
	34046	__ai __attribute__((target("fullfp16"))) float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
	34047	float16x8_t __ret;
	34048	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34049	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34050	__ret = (float16x8_t) __builtin_neon_vabdq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	34051	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33464	34052	return __ret;
33465	34053	}
33466		__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
33467		uint16x4_t __ret;
33468		__ret = (uint16x4_t)(__p0);
	34054	#endif
	34055
	34056	#ifdef __LITTLE_ENDIAN__
	34057	__ai __attribute__((target("fullfp16"))) float16x4_t vabd_f16(float16x4_t __p0, float16x4_t __p1) {
	34058	float16x4_t __ret;
	34059	__ret = (float16x4_t) __builtin_neon_vabd_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
33469	34060	return __ret;
33470	34061	}
33471		__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
33472		uint16x4_t __ret;
33473		__ret = (uint16x4_t)(__p0);
	34062	#else
	34063	__ai __attribute__((target("fullfp16"))) float16x4_t vabd_f16(float16x4_t __p0, float16x4_t __p1) {
	34064	float16x4_t __ret;
	34065	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34066	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34067	__ret = (float16x4_t) __builtin_neon_vabd_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	34068	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33474	34069	return __ret;
33475	34070	}
33476		__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
33477		uint16x4_t __ret;
33478		__ret = (uint16x4_t)(__p0);
	34071	#endif
	34072
	34073	#ifdef __LITTLE_ENDIAN__
	34074	__ai __attribute__((target("fullfp16"))) float16x8_t vabsq_f16(float16x8_t __p0) {
	34075	float16x8_t __ret;
	34076	__ret = (float16x8_t) __builtin_neon_vabsq_f16((int8x16_t)__p0, 40);
33479	34077	return __ret;
33480	34078	}
33481		__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
33482		uint16x4_t __ret;
33483		__ret = (uint16x4_t)(__p0);
	34079	#else
	34080	__ai __attribute__((target("fullfp16"))) float16x8_t vabsq_f16(float16x8_t __p0) {
	34081	float16x8_t __ret;
	34082	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34083	__ret = (float16x8_t) __builtin_neon_vabsq_f16((int8x16_t)__rev0, 40);
	34084	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33484	34085	return __ret;
33485	34086	}
33486		__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
33487		uint16x4_t __ret;
33488		__ret = (uint16x4_t)(__p0);
	34087	#endif
	34088
	34089	#ifdef __LITTLE_ENDIAN__
	34090	__ai __attribute__((target("fullfp16"))) float16x4_t vabs_f16(float16x4_t __p0) {
	34091	float16x4_t __ret;
	34092	__ret = (float16x4_t) __builtin_neon_vabs_f16((int8x8_t)__p0, 8);
33489	34093	return __ret;
33490	34094	}
33491		__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
33492		uint16x4_t __ret;
33493		__ret = (uint16x4_t)(__p0);
	34095	#else
	34096	__ai __attribute__((target("fullfp16"))) float16x4_t vabs_f16(float16x4_t __p0) {
	34097	float16x4_t __ret;
	34098	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34099	__ret = (float16x4_t) __builtin_neon_vabs_f16((int8x8_t)__rev0, 8);
	34100	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33494	34101	return __ret;
33495	34102	}
33496		__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
33497		uint16x4_t __ret;
33498		__ret = (uint16x4_t)(__p0);
	34103	#endif
	34104
	34105	#ifdef __LITTLE_ENDIAN__
	34106	__ai __attribute__((target("fullfp16"))) float16x8_t vaddq_f16(float16x8_t __p0, float16x8_t __p1) {
	34107	float16x8_t __ret;
	34108	__ret = __p0 + __p1;
33499	34109	return __ret;
33500	34110	}
33501		__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
33502		int8x8_t __ret;
33503		__ret = (int8x8_t)(__p0);
	34111	#else
	34112	__ai __attribute__((target("fullfp16"))) float16x8_t vaddq_f16(float16x8_t __p0, float16x8_t __p1) {
	34113	float16x8_t __ret;
	34114	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34115	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34116	__ret = __rev0 + __rev1;
	34117	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33504	34118	return __ret;
33505	34119	}
33506		__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
33507		int8x8_t __ret;
33508		__ret = (int8x8_t)(__p0);
	34120	#endif
	34121
	34122	#ifdef __LITTLE_ENDIAN__
	34123	__ai __attribute__((target("fullfp16"))) float16x4_t vadd_f16(float16x4_t __p0, float16x4_t __p1) {
	34124	float16x4_t __ret;
	34125	__ret = __p0 + __p1;
33509	34126	return __ret;
33510	34127	}
33511		__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
33512		int8x8_t __ret;
33513		__ret = (int8x8_t)(__p0);
	34128	#else
	34129	__ai __attribute__((target("fullfp16"))) float16x4_t vadd_f16(float16x4_t __p0, float16x4_t __p1) {
	34130	float16x4_t __ret;
	34131	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34132	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34133	__ret = __rev0 + __rev1;
	34134	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33514	34135	return __ret;
33515	34136	}
33516		__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
33517		int8x8_t __ret;
33518		__ret = (int8x8_t)(__p0);
	34137	#endif
	34138
	34139	#ifdef __LITTLE_ENDIAN__
	34140	__ai __attribute__((target("fullfp16"))) float16x8_t vbslq_f16(uint16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	34141	float16x8_t __ret;
	34142	__ret = (float16x8_t) __builtin_neon_vbslq_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
33519	34143	return __ret;
33520	34144	}
33521		__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
33522		int8x8_t __ret;
33523		__ret = (int8x8_t)(__p0);
	34145	#else
	34146	__ai __attribute__((target("fullfp16"))) float16x8_t vbslq_f16(uint16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	34147	float16x8_t __ret;
	34148	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34149	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34150	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	34151	__ret = (float16x8_t) __builtin_neon_vbslq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
	34152	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33524	34153	return __ret;
33525	34154	}
33526		__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
33527		int8x8_t __ret;
33528		__ret = (int8x8_t)(__p0);
	34155	#endif
	34156
	34157	#ifdef __LITTLE_ENDIAN__
	34158	__ai __attribute__((target("fullfp16"))) float16x4_t vbsl_f16(uint16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	34159	float16x4_t __ret;
	34160	__ret = (float16x4_t) __builtin_neon_vbsl_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
33529	34161	return __ret;
33530	34162	}
33531		__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
33532		int8x8_t __ret;
33533		__ret = (int8x8_t)(__p0);
	34163	#else
	34164	__ai __attribute__((target("fullfp16"))) float16x4_t vbsl_f16(uint16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	34165	float16x4_t __ret;
	34166	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34167	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34168	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	34169	__ret = (float16x4_t) __builtin_neon_vbsl_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
	34170	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33534	34171	return __ret;
33535	34172	}
33536		__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
33537		int8x8_t __ret;
33538		__ret = (int8x8_t)(__p0);
	34173	#endif
	34174
	34175	#ifdef __LITTLE_ENDIAN__
	34176	__ai __attribute__((target("fullfp16"))) uint16x8_t vcageq_f16(float16x8_t __p0, float16x8_t __p1) {
	34177	uint16x8_t __ret;
	34178	__ret = (uint16x8_t) __builtin_neon_vcageq_f16((int8x16_t)__p0, (int8x16_t)__p1, 49);
33539	34179	return __ret;
33540	34180	}
33541		__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
33542		int8x8_t __ret;
33543		__ret = (int8x8_t)(__p0);
	34181	#else
	34182	__ai __attribute__((target("fullfp16"))) uint16x8_t vcageq_f16(float16x8_t __p0, float16x8_t __p1) {
	34183	uint16x8_t __ret;
	34184	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34185	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34186	__ret = (uint16x8_t) __builtin_neon_vcageq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	34187	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33544	34188	return __ret;
33545	34189	}
33546		__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
33547		int8x8_t __ret;
33548		__ret = (int8x8_t)(__p0);
	34190	#endif
	34191
	34192	#ifdef __LITTLE_ENDIAN__
	34193	__ai __attribute__((target("fullfp16"))) uint16x4_t vcage_f16(float16x4_t __p0, float16x4_t __p1) {
	34194	uint16x4_t __ret;
	34195	__ret = (uint16x4_t) __builtin_neon_vcage_f16((int8x8_t)__p0, (int8x8_t)__p1, 17);
33549	34196	return __ret;
33550	34197	}
33551		__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
33552		int8x8_t __ret;
33553		__ret = (int8x8_t)(__p0);
	34198	#else
	34199	__ai __attribute__((target("fullfp16"))) uint16x4_t vcage_f16(float16x4_t __p0, float16x4_t __p1) {
	34200	uint16x4_t __ret;
	34201	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34202	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34203	__ret = (uint16x4_t) __builtin_neon_vcage_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
	34204	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33554	34205	return __ret;
33555	34206	}
33556		__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
33557		float32x2_t __ret;
33558		__ret = (float32x2_t)(__p0);
	34207	#endif
	34208
	34209	#ifdef __LITTLE_ENDIAN__
	34210	__ai __attribute__((target("fullfp16"))) uint16x8_t vcagtq_f16(float16x8_t __p0, float16x8_t __p1) {
	34211	uint16x8_t __ret;
	34212	__ret = (uint16x8_t) __builtin_neon_vcagtq_f16((int8x16_t)__p0, (int8x16_t)__p1, 49);
33559	34213	return __ret;
33560	34214	}
33561		__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
33562		float32x2_t __ret;
33563		__ret = (float32x2_t)(__p0);
	34215	#else
	34216	__ai __attribute__((target("fullfp16"))) uint16x8_t vcagtq_f16(float16x8_t __p0, float16x8_t __p1) {
	34217	uint16x8_t __ret;
	34218	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34219	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34220	__ret = (uint16x8_t) __builtin_neon_vcagtq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	34221	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33564	34222	return __ret;
33565	34223	}
33566		__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
33567		float32x2_t __ret;
33568		__ret = (float32x2_t)(__p0);
	34224	#endif
	34225
	34226	#ifdef __LITTLE_ENDIAN__
	34227	__ai __attribute__((target("fullfp16"))) uint16x4_t vcagt_f16(float16x4_t __p0, float16x4_t __p1) {
	34228	uint16x4_t __ret;
	34229	__ret = (uint16x4_t) __builtin_neon_vcagt_f16((int8x8_t)__p0, (int8x8_t)__p1, 17);
33569	34230	return __ret;
33570	34231	}
33571		__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
33572		float32x2_t __ret;
33573		__ret = (float32x2_t)(__p0);
	34232	#else
	34233	__ai __attribute__((target("fullfp16"))) uint16x4_t vcagt_f16(float16x4_t __p0, float16x4_t __p1) {
	34234	uint16x4_t __ret;
	34235	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34236	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34237	__ret = (uint16x4_t) __builtin_neon_vcagt_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
	34238	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33574	34239	return __ret;
33575	34240	}
33576		__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
33577		float32x2_t __ret;
33578		__ret = (float32x2_t)(__p0);
	34241	#endif
	34242
	34243	#ifdef __LITTLE_ENDIAN__
	34244	__ai __attribute__((target("fullfp16"))) uint16x8_t vcaleq_f16(float16x8_t __p0, float16x8_t __p1) {
	34245	uint16x8_t __ret;
	34246	__ret = (uint16x8_t) __builtin_neon_vcaleq_f16((int8x16_t)__p0, (int8x16_t)__p1, 49);
33579	34247	return __ret;
33580	34248	}
33581		__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
33582		float32x2_t __ret;
33583		__ret = (float32x2_t)(__p0);
	34249	#else
	34250	__ai __attribute__((target("fullfp16"))) uint16x8_t vcaleq_f16(float16x8_t __p0, float16x8_t __p1) {
	34251	uint16x8_t __ret;
	34252	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34253	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34254	__ret = (uint16x8_t) __builtin_neon_vcaleq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	34255	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33584	34256	return __ret;
33585	34257	}
33586		__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
33587		float32x2_t __ret;
33588		__ret = (float32x2_t)(__p0);
	34258	#endif
	34259
	34260	#ifdef __LITTLE_ENDIAN__
	34261	__ai __attribute__((target("fullfp16"))) uint16x4_t vcale_f16(float16x4_t __p0, float16x4_t __p1) {
	34262	uint16x4_t __ret;
	34263	__ret = (uint16x4_t) __builtin_neon_vcale_f16((int8x8_t)__p0, (int8x8_t)__p1, 17);
33589	34264	return __ret;
33590	34265	}
33591		__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
33592		float32x2_t __ret;
33593		__ret = (float32x2_t)(__p0);
	34266	#else
	34267	__ai __attribute__((target("fullfp16"))) uint16x4_t vcale_f16(float16x4_t __p0, float16x4_t __p1) {
	34268	uint16x4_t __ret;
	34269	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34270	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34271	__ret = (uint16x4_t) __builtin_neon_vcale_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
	34272	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33594	34273	return __ret;
33595	34274	}
33596		__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
33597		float32x2_t __ret;
33598		__ret = (float32x2_t)(__p0);
	34275	#endif
	34276
	34277	#ifdef __LITTLE_ENDIAN__
	34278	__ai __attribute__((target("fullfp16"))) uint16x8_t vcaltq_f16(float16x8_t __p0, float16x8_t __p1) {
	34279	uint16x8_t __ret;
	34280	__ret = (uint16x8_t) __builtin_neon_vcaltq_f16((int8x16_t)__p0, (int8x16_t)__p1, 49);
33599	34281	return __ret;
33600	34282	}
33601		__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
33602		float32x2_t __ret;
33603		__ret = (float32x2_t)(__p0);
	34283	#else
	34284	__ai __attribute__((target("fullfp16"))) uint16x8_t vcaltq_f16(float16x8_t __p0, float16x8_t __p1) {
	34285	uint16x8_t __ret;
	34286	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34287	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34288	__ret = (uint16x8_t) __builtin_neon_vcaltq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	34289	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33604	34290	return __ret;
33605	34291	}
33606		__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
33607		float32x2_t __ret;
33608		__ret = (float32x2_t)(__p0);
	34292	#endif
	34293
	34294	#ifdef __LITTLE_ENDIAN__
	34295	__ai __attribute__((target("fullfp16"))) uint16x4_t vcalt_f16(float16x4_t __p0, float16x4_t __p1) {
	34296	uint16x4_t __ret;
	34297	__ret = (uint16x4_t) __builtin_neon_vcalt_f16((int8x8_t)__p0, (int8x8_t)__p1, 17);
33609	34298	return __ret;
33610	34299	}
33611		__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
33612		float16x4_t __ret;
33613		__ret = (float16x4_t)(__p0);
	34300	#else
	34301	__ai __attribute__((target("fullfp16"))) uint16x4_t vcalt_f16(float16x4_t __p0, float16x4_t __p1) {
	34302	uint16x4_t __ret;
	34303	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34304	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34305	__ret = (uint16x4_t) __builtin_neon_vcalt_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
	34306	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33614	34307	return __ret;
33615	34308	}
33616		__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
33617		float16x4_t __ret;
33618		__ret = (float16x4_t)(__p0);
	34309	#endif
	34310
	34311	#ifdef __LITTLE_ENDIAN__
	34312	__ai __attribute__((target("fullfp16"))) uint16x8_t vceqq_f16(float16x8_t __p0, float16x8_t __p1) {
	34313	uint16x8_t __ret;
	34314	__ret = (uint16x8_t)(__p0 == __p1);
33619	34315	return __ret;
33620	34316	}
33621		__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
33622		float16x4_t __ret;
33623		__ret = (float16x4_t)(__p0);
	34317	#else
	34318	__ai __attribute__((target("fullfp16"))) uint16x8_t vceqq_f16(float16x8_t __p0, float16x8_t __p1) {
	34319	uint16x8_t __ret;
	34320	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34321	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34322	__ret = (uint16x8_t)(__rev0 == __rev1);
	34323	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33624	34324	return __ret;
33625	34325	}
33626		__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
33627		float16x4_t __ret;
33628		__ret = (float16x4_t)(__p0);
	34326	#endif
	34327
	34328	#ifdef __LITTLE_ENDIAN__
	34329	__ai __attribute__((target("fullfp16"))) uint16x4_t vceq_f16(float16x4_t __p0, float16x4_t __p1) {
	34330	uint16x4_t __ret;
	34331	__ret = (uint16x4_t)(__p0 == __p1);
33629	34332	return __ret;
33630	34333	}
33631		__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
33632		float16x4_t __ret;
33633		__ret = (float16x4_t)(__p0);
	34334	#else
	34335	__ai __attribute__((target("fullfp16"))) uint16x4_t vceq_f16(float16x4_t __p0, float16x4_t __p1) {
	34336	uint16x4_t __ret;
	34337	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34338	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34339	__ret = (uint16x4_t)(__rev0 == __rev1);
	34340	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33634	34341	return __ret;
33635	34342	}
33636		__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
33637		float16x4_t __ret;
33638		__ret = (float16x4_t)(__p0);
	34343	#endif
	34344
	34345	#ifdef __LITTLE_ENDIAN__
	34346	__ai __attribute__((target("fullfp16"))) uint16x8_t vceqzq_f16(float16x8_t __p0) {
	34347	uint16x8_t __ret;
	34348	__ret = (uint16x8_t) __builtin_neon_vceqzq_f16((int8x16_t)__p0, 49);
33639	34349	return __ret;
33640	34350	}
33641		__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
33642		float16x4_t __ret;
33643		__ret = (float16x4_t)(__p0);
	34351	#else
	34352	__ai __attribute__((target("fullfp16"))) uint16x8_t vceqzq_f16(float16x8_t __p0) {
	34353	uint16x8_t __ret;
	34354	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34355	__ret = (uint16x8_t) __builtin_neon_vceqzq_f16((int8x16_t)__rev0, 49);
	34356	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33644	34357	return __ret;
33645	34358	}
33646		__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
33647		float16x4_t __ret;
33648		__ret = (float16x4_t)(__p0);
	34359	#endif
	34360
	34361	#ifdef __LITTLE_ENDIAN__
	34362	__ai __attribute__((target("fullfp16"))) uint16x4_t vceqz_f16(float16x4_t __p0) {
	34363	uint16x4_t __ret;
	34364	__ret = (uint16x4_t) __builtin_neon_vceqz_f16((int8x8_t)__p0, 17);
33649	34365	return __ret;
33650	34366	}
33651		__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
33652		float16x4_t __ret;
33653		__ret = (float16x4_t)(__p0);
	34367	#else
	34368	__ai __attribute__((target("fullfp16"))) uint16x4_t vceqz_f16(float16x4_t __p0) {
	34369	uint16x4_t __ret;
	34370	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34371	__ret = (uint16x4_t) __builtin_neon_vceqz_f16((int8x8_t)__rev0, 17);
	34372	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33654	34373	return __ret;
33655	34374	}
33656		__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
33657		float16x4_t __ret;
33658		__ret = (float16x4_t)(__p0);
	34375	#endif
	34376
	34377	#ifdef __LITTLE_ENDIAN__
	34378	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgeq_f16(float16x8_t __p0, float16x8_t __p1) {
	34379	uint16x8_t __ret;
	34380	__ret = (uint16x8_t)(__p0 >= __p1);
33659	34381	return __ret;
33660	34382	}
33661		__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
33662		float16x4_t __ret;
33663		__ret = (float16x4_t)(__p0);
	34383	#else
	34384	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgeq_f16(float16x8_t __p0, float16x8_t __p1) {
	34385	uint16x8_t __ret;
	34386	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34387	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34388	__ret = (uint16x8_t)(__rev0 >= __rev1);
	34389	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33664	34390	return __ret;
33665	34391	}
33666		__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
33667		int32x2_t __ret;
33668		__ret = (int32x2_t)(__p0);
	34392	#endif
	34393
	34394	#ifdef __LITTLE_ENDIAN__
	34395	__ai __attribute__((target("fullfp16"))) uint16x4_t vcge_f16(float16x4_t __p0, float16x4_t __p1) {
	34396	uint16x4_t __ret;
	34397	__ret = (uint16x4_t)(__p0 >= __p1);
33669	34398	return __ret;
33670	34399	}
33671		__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
33672		int32x2_t __ret;
33673		__ret = (int32x2_t)(__p0);
	34400	#else
	34401	__ai __attribute__((target("fullfp16"))) uint16x4_t vcge_f16(float16x4_t __p0, float16x4_t __p1) {
	34402	uint16x4_t __ret;
	34403	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34404	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34405	__ret = (uint16x4_t)(__rev0 >= __rev1);
	34406	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33674	34407	return __ret;
33675	34408	}
33676		__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
33677		int32x2_t __ret;
33678		__ret = (int32x2_t)(__p0);
	34409	#endif
	34410
	34411	#ifdef __LITTLE_ENDIAN__
	34412	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgezq_f16(float16x8_t __p0) {
	34413	uint16x8_t __ret;
	34414	__ret = (uint16x8_t) __builtin_neon_vcgezq_f16((int8x16_t)__p0, 49);
33679	34415	return __ret;
33680	34416	}
33681		__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
33682		int32x2_t __ret;
33683		__ret = (int32x2_t)(__p0);
	34417	#else
	34418	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgezq_f16(float16x8_t __p0) {
	34419	uint16x8_t __ret;
	34420	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34421	__ret = (uint16x8_t) __builtin_neon_vcgezq_f16((int8x16_t)__rev0, 49);
	34422	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33684	34423	return __ret;
33685	34424	}
33686		__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
33687		int32x2_t __ret;
33688		__ret = (int32x2_t)(__p0);
	34425	#endif
	34426
	34427	#ifdef __LITTLE_ENDIAN__
	34428	__ai __attribute__((target("fullfp16"))) uint16x4_t vcgez_f16(float16x4_t __p0) {
	34429	uint16x4_t __ret;
	34430	__ret = (uint16x4_t) __builtin_neon_vcgez_f16((int8x8_t)__p0, 17);
33689	34431	return __ret;
33690	34432	}
33691		__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
33692		int32x2_t __ret;
33693		__ret = (int32x2_t)(__p0);
	34433	#else
	34434	__ai __attribute__((target("fullfp16"))) uint16x4_t vcgez_f16(float16x4_t __p0) {
	34435	uint16x4_t __ret;
	34436	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34437	__ret = (uint16x4_t) __builtin_neon_vcgez_f16((int8x8_t)__rev0, 17);
	34438	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33694	34439	return __ret;
33695	34440	}
33696		__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
33697		int32x2_t __ret;
33698		__ret = (int32x2_t)(__p0);
	34441	#endif
	34442
	34443	#ifdef __LITTLE_ENDIAN__
	34444	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgtq_f16(float16x8_t __p0, float16x8_t __p1) {
	34445	uint16x8_t __ret;
	34446	__ret = (uint16x8_t)(__p0 > __p1);
33699	34447	return __ret;
33700	34448	}
33701		__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
33702		int32x2_t __ret;
33703		__ret = (int32x2_t)(__p0);
	34449	#else
	34450	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgtq_f16(float16x8_t __p0, float16x8_t __p1) {
	34451	uint16x8_t __ret;
	34452	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34453	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34454	__ret = (uint16x8_t)(__rev0 > __rev1);
	34455	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33704	34456	return __ret;
33705	34457	}
33706		__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
33707		int32x2_t __ret;
33708		__ret = (int32x2_t)(__p0);
	34458	#endif
	34459
	34460	#ifdef __LITTLE_ENDIAN__
	34461	__ai __attribute__((target("fullfp16"))) uint16x4_t vcgt_f16(float16x4_t __p0, float16x4_t __p1) {
	34462	uint16x4_t __ret;
	34463	__ret = (uint16x4_t)(__p0 > __p1);
33709	34464	return __ret;
33710	34465	}
33711		__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
33712		int32x2_t __ret;
33713		__ret = (int32x2_t)(__p0);
	34466	#else
	34467	__ai __attribute__((target("fullfp16"))) uint16x4_t vcgt_f16(float16x4_t __p0, float16x4_t __p1) {
	34468	uint16x4_t __ret;
	34469	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34470	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34471	__ret = (uint16x4_t)(__rev0 > __rev1);
	34472	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33714	34473	return __ret;
33715	34474	}
33716		__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
33717		int32x2_t __ret;
33718		__ret = (int32x2_t)(__p0);
	34475	#endif
	34476
	34477	#ifdef __LITTLE_ENDIAN__
	34478	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgtzq_f16(float16x8_t __p0) {
	34479	uint16x8_t __ret;
	34480	__ret = (uint16x8_t) __builtin_neon_vcgtzq_f16((int8x16_t)__p0, 49);
33719	34481	return __ret;
33720	34482	}
33721		__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
33722		int64x1_t __ret;
33723		__ret = (int64x1_t)(__p0);
	34483	#else
	34484	__ai __attribute__((target("fullfp16"))) uint16x8_t vcgtzq_f16(float16x8_t __p0) {
	34485	uint16x8_t __ret;
	34486	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34487	__ret = (uint16x8_t) __builtin_neon_vcgtzq_f16((int8x16_t)__rev0, 49);
	34488	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33724	34489	return __ret;
33725	34490	}
33726		__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
33727		int64x1_t __ret;
33728		__ret = (int64x1_t)(__p0);
	34491	#endif
	34492
	34493	#ifdef __LITTLE_ENDIAN__
	34494	__ai __attribute__((target("fullfp16"))) uint16x4_t vcgtz_f16(float16x4_t __p0) {
	34495	uint16x4_t __ret;
	34496	__ret = (uint16x4_t) __builtin_neon_vcgtz_f16((int8x8_t)__p0, 17);
33729	34497	return __ret;
33730	34498	}
33731		__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
33732		int64x1_t __ret;
33733		__ret = (int64x1_t)(__p0);
	34499	#else
	34500	__ai __attribute__((target("fullfp16"))) uint16x4_t vcgtz_f16(float16x4_t __p0) {
	34501	uint16x4_t __ret;
	34502	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34503	__ret = (uint16x4_t) __builtin_neon_vcgtz_f16((int8x8_t)__rev0, 17);
	34504	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33734	34505	return __ret;
33735	34506	}
33736		__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
33737		int64x1_t __ret;
33738		__ret = (int64x1_t)(__p0);
	34507	#endif
	34508
	34509	#ifdef __LITTLE_ENDIAN__
	34510	__ai __attribute__((target("fullfp16"))) uint16x8_t vcleq_f16(float16x8_t __p0, float16x8_t __p1) {
	34511	uint16x8_t __ret;
	34512	__ret = (uint16x8_t)(__p0 <= __p1);
33739	34513	return __ret;
33740	34514	}
33741		__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
33742		int64x1_t __ret;
33743		__ret = (int64x1_t)(__p0);
	34515	#else
	34516	__ai __attribute__((target("fullfp16"))) uint16x8_t vcleq_f16(float16x8_t __p0, float16x8_t __p1) {
	34517	uint16x8_t __ret;
	34518	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34519	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34520	__ret = (uint16x8_t)(__rev0 <= __rev1);
	34521	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33744	34522	return __ret;
33745	34523	}
33746		__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
33747		int64x1_t __ret;
33748		__ret = (int64x1_t)(__p0);
	34524	#endif
	34525
	34526	#ifdef __LITTLE_ENDIAN__
	34527	__ai __attribute__((target("fullfp16"))) uint16x4_t vcle_f16(float16x4_t __p0, float16x4_t __p1) {
	34528	uint16x4_t __ret;
	34529	__ret = (uint16x4_t)(__p0 <= __p1);
	34530	return __ret;
	34531	}
	34532	#else
	34533	__ai __attribute__((target("fullfp16"))) uint16x4_t vcle_f16(float16x4_t __p0, float16x4_t __p1) {
	34534	uint16x4_t __ret;
	34535	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34536	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34537	__ret = (uint16x4_t)(__rev0 <= __rev1);
	34538	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33749	34539	return __ret;
33750	34540	}
33751		__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
33752		int64x1_t __ret;
33753		__ret = (int64x1_t)(__p0);
	34541	#endif
	34542
	34543	#ifdef __LITTLE_ENDIAN__
	34544	__ai __attribute__((target("fullfp16"))) uint16x8_t vclezq_f16(float16x8_t __p0) {
	34545	uint16x8_t __ret;
	34546	__ret = (uint16x8_t) __builtin_neon_vclezq_f16((int8x16_t)__p0, 49);
33754	34547	return __ret;
33755	34548	}
33756		__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
33757		int64x1_t __ret;
33758		__ret = (int64x1_t)(__p0);
	34549	#else
	34550	__ai __attribute__((target("fullfp16"))) uint16x8_t vclezq_f16(float16x8_t __p0) {
	34551	uint16x8_t __ret;
	34552	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34553	__ret = (uint16x8_t) __builtin_neon_vclezq_f16((int8x16_t)__rev0, 49);
	34554	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33759	34555	return __ret;
33760	34556	}
33761		__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
33762		int64x1_t __ret;
33763		__ret = (int64x1_t)(__p0);
	34557	#endif
	34558
	34559	#ifdef __LITTLE_ENDIAN__
	34560	__ai __attribute__((target("fullfp16"))) uint16x4_t vclez_f16(float16x4_t __p0) {
	34561	uint16x4_t __ret;
	34562	__ret = (uint16x4_t) __builtin_neon_vclez_f16((int8x8_t)__p0, 17);
33764	34563	return __ret;
33765	34564	}
33766		__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
33767		int64x1_t __ret;
33768		__ret = (int64x1_t)(__p0);
	34565	#else
	34566	__ai __attribute__((target("fullfp16"))) uint16x4_t vclez_f16(float16x4_t __p0) {
	34567	uint16x4_t __ret;
	34568	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34569	__ret = (uint16x4_t) __builtin_neon_vclez_f16((int8x8_t)__rev0, 17);
	34570	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33769	34571	return __ret;
33770	34572	}
33771		__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
33772		int64x1_t __ret;
33773		__ret = (int64x1_t)(__p0);
	34573	#endif
	34574
	34575	#ifdef __LITTLE_ENDIAN__
	34576	__ai __attribute__((target("fullfp16"))) uint16x8_t vcltq_f16(float16x8_t __p0, float16x8_t __p1) {
	34577	uint16x8_t __ret;
	34578	__ret = (uint16x8_t)(__p0 < __p1);
33774	34579	return __ret;
33775	34580	}
33776		__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
33777		int16x4_t __ret;
33778		__ret = (int16x4_t)(__p0);
	34581	#else
	34582	__ai __attribute__((target("fullfp16"))) uint16x8_t vcltq_f16(float16x8_t __p0, float16x8_t __p1) {
	34583	uint16x8_t __ret;
	34584	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34585	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	34586	__ret = (uint16x8_t)(__rev0 < __rev1);
	34587	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33779	34588	return __ret;
33780	34589	}
33781		__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
33782		int16x4_t __ret;
33783		__ret = (int16x4_t)(__p0);
	34590	#endif
	34591
	34592	#ifdef __LITTLE_ENDIAN__
	34593	__ai __attribute__((target("fullfp16"))) uint16x4_t vclt_f16(float16x4_t __p0, float16x4_t __p1) {
	34594	uint16x4_t __ret;
	34595	__ret = (uint16x4_t)(__p0 < __p1);
33784	34596	return __ret;
33785	34597	}
33786		__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
33787		int16x4_t __ret;
33788		__ret = (int16x4_t)(__p0);
	34598	#else
	34599	__ai __attribute__((target("fullfp16"))) uint16x4_t vclt_f16(float16x4_t __p0, float16x4_t __p1) {
	34600	uint16x4_t __ret;
	34601	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34602	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	34603	__ret = (uint16x4_t)(__rev0 < __rev1);
	34604	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33789	34605	return __ret;
33790	34606	}
33791		__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
33792		int16x4_t __ret;
33793		__ret = (int16x4_t)(__p0);
	34607	#endif
	34608
	34609	#ifdef __LITTLE_ENDIAN__
	34610	__ai __attribute__((target("fullfp16"))) uint16x8_t vcltzq_f16(float16x8_t __p0) {
	34611	uint16x8_t __ret;
	34612	__ret = (uint16x8_t) __builtin_neon_vcltzq_f16((int8x16_t)__p0, 49);
33794	34613	return __ret;
33795	34614	}
33796		__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
33797		int16x4_t __ret;
33798		__ret = (int16x4_t)(__p0);
	34615	#else
	34616	__ai __attribute__((target("fullfp16"))) uint16x8_t vcltzq_f16(float16x8_t __p0) {
	34617	uint16x8_t __ret;
	34618	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34619	__ret = (uint16x8_t) __builtin_neon_vcltzq_f16((int8x16_t)__rev0, 49);
	34620	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33799	34621	return __ret;
33800	34622	}
33801		__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
33802		int16x4_t __ret;
33803		__ret = (int16x4_t)(__p0);
	34623	#endif
	34624
	34625	#ifdef __LITTLE_ENDIAN__
	34626	__ai __attribute__((target("fullfp16"))) uint16x4_t vcltz_f16(float16x4_t __p0) {
	34627	uint16x4_t __ret;
	34628	__ret = (uint16x4_t) __builtin_neon_vcltz_f16((int8x8_t)__p0, 17);
33804	34629	return __ret;
33805	34630	}
33806		__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
33807		int16x4_t __ret;
33808		__ret = (int16x4_t)(__p0);
	34631	#else
	34632	__ai __attribute__((target("fullfp16"))) uint16x4_t vcltz_f16(float16x4_t __p0) {
	34633	uint16x4_t __ret;
	34634	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34635	__ret = (uint16x4_t) __builtin_neon_vcltz_f16((int8x8_t)__rev0, 17);
	34636	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33809	34637	return __ret;
33810	34638	}
33811		__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
33812		int16x4_t __ret;
33813		__ret = (int16x4_t)(__p0);
	34639	#endif
	34640
	34641	#ifdef __LITTLE_ENDIAN__
	34642	__ai __attribute__((target("fullfp16"))) float16x8_t vcvtq_f16_u16(uint16x8_t __p0) {
	34643	float16x8_t __ret;
	34644	__ret = (float16x8_t) __builtin_neon_vcvtq_f16_u16((int8x16_t)__p0, 49);
33814	34645	return __ret;
33815	34646	}
33816		__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
33817		int16x4_t __ret;
33818		__ret = (int16x4_t)(__p0);
	34647	#else
	34648	__ai __attribute__((target("fullfp16"))) float16x8_t vcvtq_f16_u16(uint16x8_t __p0) {
	34649	float16x8_t __ret;
	34650	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34651	__ret = (float16x8_t) __builtin_neon_vcvtq_f16_u16((int8x16_t)__rev0, 49);
	34652	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33819	34653	return __ret;
33820	34654	}
33821		__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
33822		int16x4_t __ret;
33823		__ret = (int16x4_t)(__p0);
	34655	#endif
	34656
	34657	#ifdef __LITTLE_ENDIAN__
	34658	__ai __attribute__((target("fullfp16"))) float16x8_t vcvtq_f16_s16(int16x8_t __p0) {
	34659	float16x8_t __ret;
	34660	__ret = (float16x8_t) __builtin_neon_vcvtq_f16_s16((int8x16_t)__p0, 33);
33824	34661	return __ret;
33825	34662	}
33826		__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
33827		int16x4_t __ret;
33828		__ret = (int16x4_t)(__p0);
	34663	#else
	34664	__ai __attribute__((target("fullfp16"))) float16x8_t vcvtq_f16_s16(int16x8_t __p0) {
	34665	float16x8_t __ret;
	34666	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34667	__ret = (float16x8_t) __builtin_neon_vcvtq_f16_s16((int8x16_t)__rev0, 33);
	34668	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33829	34669	return __ret;
33830	34670	}
33831	34671	#endif
33832		#if (__ARM_FP & 2)
	34672
33833	34673	#ifdef __LITTLE_ENDIAN__
33834		__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
	34674	__ai __attribute__((target("fullfp16"))) float16x4_t vcvt_f16_u16(uint16x4_t __p0) {
33835	34675	float16x4_t __ret;
33836		__ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 41);
	34676	__ret = (float16x4_t) __builtin_neon_vcvt_f16_u16((int8x8_t)__p0, 17);
33837	34677	return __ret;
33838	34678	}
33839	34679	#else
33840		__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
	34680	__ai __attribute__((target("fullfp16"))) float16x4_t vcvt_f16_u16(uint16x4_t __p0) {
33841	34681	float16x4_t __ret;
33842		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
33843		__ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__rev0, 41);
	34682	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34683	__ret = (float16x4_t) __builtin_neon_vcvt_f16_u16((int8x8_t)__rev0, 17);
33844	34684	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33845	34685	return __ret;
33846	34686	}
33847		__ai float16x4_t __noswap_vcvt_f16_f32(float32x4_t __p0) {
33848		float16x4_t __ret;
33849		__ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 41);
33850		return __ret;
33851		}
33852	34687	#endif
33853	34688
33854	34689	#ifdef __LITTLE_ENDIAN__
33855		__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
33856		float32x4_t __ret;
33857		__ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 8);
	34690	__ai __attribute__((target("fullfp16"))) float16x4_t vcvt_f16_s16(int16x4_t __p0) {
	34691	float16x4_t __ret;
	34692	__ret = (float16x4_t) __builtin_neon_vcvt_f16_s16((int8x8_t)__p0, 1);
33858	34693	return __ret;
33859	34694	}
33860	34695	#else
33861		__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
33862		float32x4_t __ret;
33863		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
33864		__ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__rev0, 8);
	34696	__ai __attribute__((target("fullfp16"))) float16x4_t vcvt_f16_s16(int16x4_t __p0) {
	34697	float16x4_t __ret;
	34698	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34699	__ret = (float16x4_t) __builtin_neon_vcvt_f16_s16((int8x8_t)__rev0, 1);
33865	34700	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33866	34701	return __ret;
33867	34702	}
33868		__ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
33869		float32x4_t __ret;
33870		__ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 8);
33871		return __ret;
33872		}
33873	34703	#endif
33874	34704
33875	34705	#ifdef __LITTLE_ENDIAN__
33876		#define vld1q_f16(__p0) __extension__ ({ \
	34706	#define vcvtq_n_f16_u16(__p0, __p1) __extension__ ({ \
33877	34707	float16x8_t __ret; \
33878		__ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
	34708	uint16x8_t __s0 = __p0; \
	34709	__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_u16((int8x16_t)__s0, __p1, 49); \
33879	34710	__ret; \
33880	34711	})
33881	34712	#else
33882		#define vld1q_f16(__p0) __extension__ ({ \
	34713	#define vcvtq_n_f16_u16(__p0, __p1) __extension__ ({ \
33883	34714	float16x8_t __ret; \
33884		__ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
	34715	uint16x8_t __s0 = __p0; \
	34716	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	34717	__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_u16((int8x16_t)__rev0, __p1, 49); \
33885	34718	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
33886	34719	__ret; \
33887	34720	})
33888	34721	#endif
33889	34722
33890	34723	#ifdef __LITTLE_ENDIAN__
33891		#define vld1_f16(__p0) __extension__ ({ \
33892		float16x4_t __ret; \
33893		__ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
	34724	#define vcvtq_n_f16_s16(__p0, __p1) __extension__ ({ \
	34725	float16x8_t __ret; \
	34726	int16x8_t __s0 = __p0; \
	34727	__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_s16((int8x16_t)__s0, __p1, 33); \
33894	34728	__ret; \
33895	34729	})
33896	34730	#else
33897		#define vld1_f16(__p0) __extension__ ({ \
33898		float16x4_t __ret; \
33899		__ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
33900		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	34731	#define vcvtq_n_f16_s16(__p0, __p1) __extension__ ({ \
	34732	float16x8_t __ret; \
	34733	int16x8_t __s0 = __p0; \
	34734	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	34735	__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_s16((int8x16_t)__rev0, __p1, 33); \
	34736	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
33901	34737	__ret; \
33902	34738	})
33903	34739	#endif
33904	34740
33905	34741	#ifdef __LITTLE_ENDIAN__
33906		#define vld1q_dup_f16(__p0) __extension__ ({ \
33907		float16x8_t __ret; \
33908		__ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
	34742	#define vcvt_n_f16_u16(__p0, __p1) __extension__ ({ \
	34743	float16x4_t __ret; \
	34744	uint16x4_t __s0 = __p0; \
	34745	__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_u16((int8x8_t)__s0, __p1, 17); \
33909	34746	__ret; \
33910	34747	})
33911	34748	#else
33912		#define vld1q_dup_f16(__p0) __extension__ ({ \
33913		float16x8_t __ret; \
33914		__ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
33915		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	34749	#define vcvt_n_f16_u16(__p0, __p1) __extension__ ({ \
	34750	float16x4_t __ret; \
	34751	uint16x4_t __s0 = __p0; \
	34752	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	34753	__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_u16((int8x8_t)__rev0, __p1, 17); \
	34754	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
33916	34755	__ret; \
33917	34756	})
33918	34757	#endif
33919	34758
33920	34759	#ifdef __LITTLE_ENDIAN__
33921		#define vld1_dup_f16(__p0) __extension__ ({ \
	34760	#define vcvt_n_f16_s16(__p0, __p1) __extension__ ({ \
33922	34761	float16x4_t __ret; \
33923		__ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
	34762	int16x4_t __s0 = __p0; \
	34763	__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_s16((int8x8_t)__s0, __p1, 1); \
33924	34764	__ret; \
33925	34765	})
33926	34766	#else
33927		#define vld1_dup_f16(__p0) __extension__ ({ \
	34767	#define vcvt_n_f16_s16(__p0, __p1) __extension__ ({ \
33928	34768	float16x4_t __ret; \
33929		__ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
	34769	int16x4_t __s0 = __p0; \
	34770	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	34771	__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_s16((int8x8_t)__rev0, __p1, 1); \
33930	34772	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
33931	34773	__ret; \
33932	34774	})
33933	34775	#endif
33934	34776
33935	34777	#ifdef __LITTLE_ENDIAN__
33936		#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33937		float16x8_t __ret; \
33938		float16x8_t __s1 = __p1; \
33939		__ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
	34778	#define vcvtq_n_s16_f16(__p0, __p1) __extension__ ({ \
	34779	int16x8_t __ret; \
	34780	float16x8_t __s0 = __p0; \
	34781	__ret = (int16x8_t) __builtin_neon_vcvtq_n_s16_f16((int8x16_t)__s0, __p1, 33); \
33940	34782	__ret; \
33941	34783	})
33942	34784	#else
33943		#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33944		float16x8_t __ret; \
33945		float16x8_t __s1 = __p1; \
33946		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
33947		__ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
	34785	#define vcvtq_n_s16_f16(__p0, __p1) __extension__ ({ \
	34786	int16x8_t __ret; \
	34787	float16x8_t __s0 = __p0; \
	34788	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	34789	__ret = (int16x8_t) __builtin_neon_vcvtq_n_s16_f16((int8x16_t)__rev0, __p1, 33); \
33948	34790	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
33949	34791	__ret; \
33950	34792	})
33951	34793	#endif
33952	34794
33953	34795	#ifdef __LITTLE_ENDIAN__
33954		#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33955		float16x4_t __ret; \
33956		float16x4_t __s1 = __p1; \
33957		__ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
	34796	#define vcvt_n_s16_f16(__p0, __p1) __extension__ ({ \
	34797	int16x4_t __ret; \
	34798	float16x4_t __s0 = __p0; \
	34799	__ret = (int16x4_t) __builtin_neon_vcvt_n_s16_f16((int8x8_t)__s0, __p1, 1); \
33958	34800	__ret; \
33959	34801	})
33960	34802	#else
33961		#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33962		float16x4_t __ret; \
33963		float16x4_t __s1 = __p1; \
33964		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
33965		__ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
	34803	#define vcvt_n_s16_f16(__p0, __p1) __extension__ ({ \
	34804	int16x4_t __ret; \
	34805	float16x4_t __s0 = __p0; \
	34806	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	34807	__ret = (int16x4_t) __builtin_neon_vcvt_n_s16_f16((int8x8_t)__rev0, __p1, 1); \
33966	34808	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
33967	34809	__ret; \
33968	34810	})
33969	34811	#endif
33970	34812
33971	34813	#ifdef __LITTLE_ENDIAN__
33972		#define vld1q_f16_x2(__p0) __extension__ ({ \
33973		float16x8x2_t __ret; \
33974		__builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
	34814	#define vcvtq_n_u16_f16(__p0, __p1) __extension__ ({ \
	34815	uint16x8_t __ret; \
	34816	float16x8_t __s0 = __p0; \
	34817	__ret = (uint16x8_t) __builtin_neon_vcvtq_n_u16_f16((int8x16_t)__s0, __p1, 49); \
33975	34818	__ret; \
33976	34819	})
33977	34820	#else
33978		#define vld1q_f16_x2(__p0) __extension__ ({ \
33979		float16x8x2_t __ret; \
33980		__builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
33981		\
33982		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
33983		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	34821	#define vcvtq_n_u16_f16(__p0, __p1) __extension__ ({ \
	34822	uint16x8_t __ret; \
	34823	float16x8_t __s0 = __p0; \
	34824	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	34825	__ret = (uint16x8_t) __builtin_neon_vcvtq_n_u16_f16((int8x16_t)__rev0, __p1, 49); \
	34826	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
33984	34827	__ret; \
33985	34828	})
33986	34829	#endif
33987	34830
33988	34831	#ifdef __LITTLE_ENDIAN__
33989		#define vld1_f16_x2(__p0) __extension__ ({ \
33990		float16x4x2_t __ret; \
33991		__builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
	34832	#define vcvt_n_u16_f16(__p0, __p1) __extension__ ({ \
	34833	uint16x4_t __ret; \
	34834	float16x4_t __s0 = __p0; \
	34835	__ret = (uint16x4_t) __builtin_neon_vcvt_n_u16_f16((int8x8_t)__s0, __p1, 17); \
33992	34836	__ret; \
33993	34837	})
33994	34838	#else
33995		#define vld1_f16_x2(__p0) __extension__ ({ \
33996		float16x4x2_t __ret; \
33997		__builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
33998		\
33999		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34000		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	34839	#define vcvt_n_u16_f16(__p0, __p1) __extension__ ({ \
	34840	uint16x4_t __ret; \
	34841	float16x4_t __s0 = __p0; \
	34842	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	34843	__ret = (uint16x4_t) __builtin_neon_vcvt_n_u16_f16((int8x8_t)__rev0, __p1, 17); \
	34844	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
34001	34845	__ret; \
34002	34846	})
34003	34847	#endif
34004	34848
34005	34849	#ifdef __LITTLE_ENDIAN__
34006		#define vld1q_f16_x3(__p0) __extension__ ({ \
34007		float16x8x3_t __ret; \
34008		__builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
34009		__ret; \
34010		})
	34850	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtq_s16_f16(float16x8_t __p0) {
	34851	int16x8_t __ret;
	34852	__ret = (int16x8_t) __builtin_neon_vcvtq_s16_f16((int8x16_t)__p0, 33);
	34853	return __ret;
	34854	}
34011	34855	#else
34012		#define vld1q_f16_x3(__p0) __extension__ ({ \
34013		float16x8x3_t __ret; \
34014		__builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
34015		\
34016		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34017		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34018		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34019		__ret; \
34020		})
	34856	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtq_s16_f16(float16x8_t __p0) {
	34857	int16x8_t __ret;
	34858	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34859	__ret = (int16x8_t) __builtin_neon_vcvtq_s16_f16((int8x16_t)__rev0, 33);
	34860	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	34861	return __ret;
	34862	}
34021	34863	#endif
34022	34864
34023	34865	#ifdef __LITTLE_ENDIAN__
34024		#define vld1_f16_x3(__p0) __extension__ ({ \
34025		float16x4x3_t __ret; \
34026		__builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
34027		__ret; \
34028		})
	34866	__ai __attribute__((target("fullfp16"))) int16x4_t vcvt_s16_f16(float16x4_t __p0) {
	34867	int16x4_t __ret;
	34868	__ret = (int16x4_t) __builtin_neon_vcvt_s16_f16((int8x8_t)__p0, 1);
	34869	return __ret;
	34870	}
34029	34871	#else
34030		#define vld1_f16_x3(__p0) __extension__ ({ \
34031		float16x4x3_t __ret; \
34032		__builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
34033		\
34034		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34035		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34036		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34037		__ret; \
34038		})
	34872	__ai __attribute__((target("fullfp16"))) int16x4_t vcvt_s16_f16(float16x4_t __p0) {
	34873	int16x4_t __ret;
	34874	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34875	__ret = (int16x4_t) __builtin_neon_vcvt_s16_f16((int8x8_t)__rev0, 1);
	34876	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	34877	return __ret;
	34878	}
34039	34879	#endif
34040	34880
34041	34881	#ifdef __LITTLE_ENDIAN__
34042		#define vld1q_f16_x4(__p0) __extension__ ({ \
34043		float16x8x4_t __ret; \
34044		__builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
34045		__ret; \
34046		})
	34882	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtq_u16_f16(float16x8_t __p0) {
	34883	uint16x8_t __ret;
	34884	__ret = (uint16x8_t) __builtin_neon_vcvtq_u16_f16((int8x16_t)__p0, 49);
	34885	return __ret;
	34886	}
34047	34887	#else
34048		#define vld1q_f16_x4(__p0) __extension__ ({ \
34049		float16x8x4_t __ret; \
34050		__builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
34051		\
34052		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34053		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34054		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34055		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34056		__ret; \
34057		})
	34888	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtq_u16_f16(float16x8_t __p0) {
	34889	uint16x8_t __ret;
	34890	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34891	__ret = (uint16x8_t) __builtin_neon_vcvtq_u16_f16((int8x16_t)__rev0, 49);
	34892	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	34893	return __ret;
	34894	}
34058	34895	#endif
34059	34896
34060	34897	#ifdef __LITTLE_ENDIAN__
34061		#define vld1_f16_x4(__p0) __extension__ ({ \
34062		float16x4x4_t __ret; \
34063		__builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
34064		__ret; \
34065		})
	34898	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvt_u16_f16(float16x4_t __p0) {
	34899	uint16x4_t __ret;
	34900	__ret = (uint16x4_t) __builtin_neon_vcvt_u16_f16((int8x8_t)__p0, 17);
	34901	return __ret;
	34902	}
34066	34903	#else
34067		#define vld1_f16_x4(__p0) __extension__ ({ \
34068		float16x4x4_t __ret; \
34069		__builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
34070		\
34071		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34072		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34073		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34074		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
34075		__ret; \
34076		})
	34904	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvt_u16_f16(float16x4_t __p0) {
	34905	uint16x4_t __ret;
	34906	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34907	__ret = (uint16x4_t) __builtin_neon_vcvt_u16_f16((int8x8_t)__rev0, 17);
	34908	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	34909	return __ret;
	34910	}
34077	34911	#endif
34078	34912
34079	34913	#ifdef __LITTLE_ENDIAN__
34080		#define vld2q_f16(__p0) __extension__ ({ \
34081		float16x8x2_t __ret; \
34082		__builtin_neon_vld2q_v(&__ret, __p0, 40); \
34083		__ret; \
34084		})
	34914	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtaq_s16_f16(float16x8_t __p0) {
	34915	int16x8_t __ret;
	34916	__ret = (int16x8_t) __builtin_neon_vcvtaq_s16_f16((int8x16_t)__p0, 33);
	34917	return __ret;
	34918	}
34085	34919	#else
34086		#define vld2q_f16(__p0) __extension__ ({ \
34087		float16x8x2_t __ret; \
34088		__builtin_neon_vld2q_v(&__ret, __p0, 40); \
34089		\
34090		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34091		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34092		__ret; \
34093		})
	34920	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtaq_s16_f16(float16x8_t __p0) {
	34921	int16x8_t __ret;
	34922	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34923	__ret = (int16x8_t) __builtin_neon_vcvtaq_s16_f16((int8x16_t)__rev0, 33);
	34924	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	34925	return __ret;
	34926	}
34094	34927	#endif
34095	34928
34096	34929	#ifdef __LITTLE_ENDIAN__
34097		#define vld2_f16(__p0) __extension__ ({ \
34098		float16x4x2_t __ret; \
34099		__builtin_neon_vld2_v(&__ret, __p0, 8); \
34100		__ret; \
34101		})
	34930	__ai __attribute__((target("fullfp16"))) int16x4_t vcvta_s16_f16(float16x4_t __p0) {
	34931	int16x4_t __ret;
	34932	__ret = (int16x4_t) __builtin_neon_vcvta_s16_f16((int8x8_t)__p0, 1);
	34933	return __ret;
	34934	}
34102	34935	#else
34103		#define vld2_f16(__p0) __extension__ ({ \
34104		float16x4x2_t __ret; \
34105		__builtin_neon_vld2_v(&__ret, __p0, 8); \
34106		\
34107		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34108		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34109		__ret; \
34110		})
	34936	__ai __attribute__((target("fullfp16"))) int16x4_t vcvta_s16_f16(float16x4_t __p0) {
	34937	int16x4_t __ret;
	34938	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34939	__ret = (int16x4_t) __builtin_neon_vcvta_s16_f16((int8x8_t)__rev0, 1);
	34940	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	34941	return __ret;
	34942	}
34111	34943	#endif
34112	34944
34113	34945	#ifdef __LITTLE_ENDIAN__
34114		#define vld2q_dup_f16(__p0) __extension__ ({ \
34115		float16x8x2_t __ret; \
34116		__builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
34117		__ret; \
34118		})
	34946	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtaq_u16_f16(float16x8_t __p0) {
	34947	uint16x8_t __ret;
	34948	__ret = (uint16x8_t) __builtin_neon_vcvtaq_u16_f16((int8x16_t)__p0, 49);
	34949	return __ret;
	34950	}
34119	34951	#else
34120		#define vld2q_dup_f16(__p0) __extension__ ({ \
34121		float16x8x2_t __ret; \
34122		__builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
34123		\
34124		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34125		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34126		__ret; \
34127		})
	34952	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtaq_u16_f16(float16x8_t __p0) {
	34953	uint16x8_t __ret;
	34954	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34955	__ret = (uint16x8_t) __builtin_neon_vcvtaq_u16_f16((int8x16_t)__rev0, 49);
	34956	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	34957	return __ret;
	34958	}
34128	34959	#endif
34129	34960
34130	34961	#ifdef __LITTLE_ENDIAN__
34131		#define vld2_dup_f16(__p0) __extension__ ({ \
34132		float16x4x2_t __ret; \
34133		__builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
34134		__ret; \
34135		})
	34962	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvta_u16_f16(float16x4_t __p0) {
	34963	uint16x4_t __ret;
	34964	__ret = (uint16x4_t) __builtin_neon_vcvta_u16_f16((int8x8_t)__p0, 17);
	34965	return __ret;
	34966	}
	34967	#else
	34968	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvta_u16_f16(float16x4_t __p0) {
	34969	uint16x4_t __ret;
	34970	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	34971	__ret = (uint16x4_t) __builtin_neon_vcvta_u16_f16((int8x8_t)__rev0, 17);
	34972	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	34973	return __ret;
	34974	}
	34975	#endif
	34976
	34977	#ifdef __LITTLE_ENDIAN__
	34978	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtmq_s16_f16(float16x8_t __p0) {
	34979	int16x8_t __ret;
	34980	__ret = (int16x8_t) __builtin_neon_vcvtmq_s16_f16((int8x16_t)__p0, 33);
	34981	return __ret;
	34982	}
	34983	#else
	34984	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtmq_s16_f16(float16x8_t __p0) {
	34985	int16x8_t __ret;
	34986	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	34987	__ret = (int16x8_t) __builtin_neon_vcvtmq_s16_f16((int8x16_t)__rev0, 33);
	34988	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	34989	return __ret;
	34990	}
	34991	#endif
	34992
	34993	#ifdef __LITTLE_ENDIAN__
	34994	__ai __attribute__((target("fullfp16"))) int16x4_t vcvtm_s16_f16(float16x4_t __p0) {
	34995	int16x4_t __ret;
	34996	__ret = (int16x4_t) __builtin_neon_vcvtm_s16_f16((int8x8_t)__p0, 1);
	34997	return __ret;
	34998	}
	34999	#else
	35000	__ai __attribute__((target("fullfp16"))) int16x4_t vcvtm_s16_f16(float16x4_t __p0) {
	35001	int16x4_t __ret;
	35002	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35003	__ret = (int16x4_t) __builtin_neon_vcvtm_s16_f16((int8x8_t)__rev0, 1);
	35004	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35005	return __ret;
	35006	}
	35007	#endif
	35008
	35009	#ifdef __LITTLE_ENDIAN__
	35010	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtmq_u16_f16(float16x8_t __p0) {
	35011	uint16x8_t __ret;
	35012	__ret = (uint16x8_t) __builtin_neon_vcvtmq_u16_f16((int8x16_t)__p0, 49);
	35013	return __ret;
	35014	}
	35015	#else
	35016	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtmq_u16_f16(float16x8_t __p0) {
	35017	uint16x8_t __ret;
	35018	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35019	__ret = (uint16x8_t) __builtin_neon_vcvtmq_u16_f16((int8x16_t)__rev0, 49);
	35020	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35021	return __ret;
	35022	}
	35023	#endif
	35024
	35025	#ifdef __LITTLE_ENDIAN__
	35026	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvtm_u16_f16(float16x4_t __p0) {
	35027	uint16x4_t __ret;
	35028	__ret = (uint16x4_t) __builtin_neon_vcvtm_u16_f16((int8x8_t)__p0, 17);
	35029	return __ret;
	35030	}
	35031	#else
	35032	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvtm_u16_f16(float16x4_t __p0) {
	35033	uint16x4_t __ret;
	35034	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35035	__ret = (uint16x4_t) __builtin_neon_vcvtm_u16_f16((int8x8_t)__rev0, 17);
	35036	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35037	return __ret;
	35038	}
	35039	#endif
	35040
	35041	#ifdef __LITTLE_ENDIAN__
	35042	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtnq_s16_f16(float16x8_t __p0) {
	35043	int16x8_t __ret;
	35044	__ret = (int16x8_t) __builtin_neon_vcvtnq_s16_f16((int8x16_t)__p0, 33);
	35045	return __ret;
	35046	}
	35047	#else
	35048	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtnq_s16_f16(float16x8_t __p0) {
	35049	int16x8_t __ret;
	35050	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35051	__ret = (int16x8_t) __builtin_neon_vcvtnq_s16_f16((int8x16_t)__rev0, 33);
	35052	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35053	return __ret;
	35054	}
	35055	#endif
	35056
	35057	#ifdef __LITTLE_ENDIAN__
	35058	__ai __attribute__((target("fullfp16"))) int16x4_t vcvtn_s16_f16(float16x4_t __p0) {
	35059	int16x4_t __ret;
	35060	__ret = (int16x4_t) __builtin_neon_vcvtn_s16_f16((int8x8_t)__p0, 1);
	35061	return __ret;
	35062	}
	35063	#else
	35064	__ai __attribute__((target("fullfp16"))) int16x4_t vcvtn_s16_f16(float16x4_t __p0) {
	35065	int16x4_t __ret;
	35066	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35067	__ret = (int16x4_t) __builtin_neon_vcvtn_s16_f16((int8x8_t)__rev0, 1);
	35068	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35069	return __ret;
	35070	}
	35071	#endif
	35072
	35073	#ifdef __LITTLE_ENDIAN__
	35074	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtnq_u16_f16(float16x8_t __p0) {
	35075	uint16x8_t __ret;
	35076	__ret = (uint16x8_t) __builtin_neon_vcvtnq_u16_f16((int8x16_t)__p0, 49);
	35077	return __ret;
	35078	}
	35079	#else
	35080	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtnq_u16_f16(float16x8_t __p0) {
	35081	uint16x8_t __ret;
	35082	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35083	__ret = (uint16x8_t) __builtin_neon_vcvtnq_u16_f16((int8x16_t)__rev0, 49);
	35084	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35085	return __ret;
	35086	}
	35087	#endif
	35088
	35089	#ifdef __LITTLE_ENDIAN__
	35090	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvtn_u16_f16(float16x4_t __p0) {
	35091	uint16x4_t __ret;
	35092	__ret = (uint16x4_t) __builtin_neon_vcvtn_u16_f16((int8x8_t)__p0, 17);
	35093	return __ret;
	35094	}
34136	35095	#else
34137		#define vld2_dup_f16(__p0) __extension__ ({ \
34138		float16x4x2_t __ret; \
34139		__builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
34140		\
34141		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34142		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34143		__ret; \
34144		})
	35096	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvtn_u16_f16(float16x4_t __p0) {
	35097	uint16x4_t __ret;
	35098	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35099	__ret = (uint16x4_t) __builtin_neon_vcvtn_u16_f16((int8x8_t)__rev0, 17);
	35100	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35101	return __ret;
	35102	}
34145	35103	#endif
34146	35104
34147	35105	#ifdef __LITTLE_ENDIAN__
34148		#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34149		float16x8x2_t __ret; \
34150		float16x8x2_t __s1 = __p1; \
34151		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
34152		__ret; \
34153		})
	35106	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtpq_s16_f16(float16x8_t __p0) {
	35107	int16x8_t __ret;
	35108	__ret = (int16x8_t) __builtin_neon_vcvtpq_s16_f16((int8x16_t)__p0, 33);
	35109	return __ret;
	35110	}
34154	35111	#else
34155		#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34156		float16x8x2_t __ret; \
34157		float16x8x2_t __s1 = __p1; \
34158		float16x8x2_t __rev1; \
34159		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34160		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34161		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
34162		\
34163		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34164		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34165		__ret; \
34166		})
	35112	__ai __attribute__((target("fullfp16"))) int16x8_t vcvtpq_s16_f16(float16x8_t __p0) {
	35113	int16x8_t __ret;
	35114	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35115	__ret = (int16x8_t) __builtin_neon_vcvtpq_s16_f16((int8x16_t)__rev0, 33);
	35116	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35117	return __ret;
	35118	}
34167	35119	#endif
34168	35120
34169	35121	#ifdef __LITTLE_ENDIAN__
34170		#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34171		float16x4x2_t __ret; \
34172		float16x4x2_t __s1 = __p1; \
34173		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
34174		__ret; \
34175		})
	35122	__ai __attribute__((target("fullfp16"))) int16x4_t vcvtp_s16_f16(float16x4_t __p0) {
	35123	int16x4_t __ret;
	35124	__ret = (int16x4_t) __builtin_neon_vcvtp_s16_f16((int8x8_t)__p0, 1);
	35125	return __ret;
	35126	}
34176	35127	#else
34177		#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34178		float16x4x2_t __ret; \
34179		float16x4x2_t __s1 = __p1; \
34180		float16x4x2_t __rev1; \
34181		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34182		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34183		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
34184		\
34185		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34186		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34187		__ret; \
34188		})
	35128	__ai __attribute__((target("fullfp16"))) int16x4_t vcvtp_s16_f16(float16x4_t __p0) {
	35129	int16x4_t __ret;
	35130	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35131	__ret = (int16x4_t) __builtin_neon_vcvtp_s16_f16((int8x8_t)__rev0, 1);
	35132	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35133	return __ret;
	35134	}
34189	35135	#endif
34190	35136
34191	35137	#ifdef __LITTLE_ENDIAN__
34192		#define vld3q_f16(__p0) __extension__ ({ \
34193		float16x8x3_t __ret; \
34194		__builtin_neon_vld3q_v(&__ret, __p0, 40); \
34195		__ret; \
34196		})
	35138	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtpq_u16_f16(float16x8_t __p0) {
	35139	uint16x8_t __ret;
	35140	__ret = (uint16x8_t) __builtin_neon_vcvtpq_u16_f16((int8x16_t)__p0, 49);
	35141	return __ret;
	35142	}
34197	35143	#else
34198		#define vld3q_f16(__p0) __extension__ ({ \
34199		float16x8x3_t __ret; \
34200		__builtin_neon_vld3q_v(&__ret, __p0, 40); \
34201		\
34202		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34203		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34204		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34205		__ret; \
34206		})
	35144	__ai __attribute__((target("fullfp16"))) uint16x8_t vcvtpq_u16_f16(float16x8_t __p0) {
	35145	uint16x8_t __ret;
	35146	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35147	__ret = (uint16x8_t) __builtin_neon_vcvtpq_u16_f16((int8x16_t)__rev0, 49);
	35148	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35149	return __ret;
	35150	}
34207	35151	#endif
34208	35152
34209	35153	#ifdef __LITTLE_ENDIAN__
34210		#define vld3_f16(__p0) __extension__ ({ \
34211		float16x4x3_t __ret; \
34212		__builtin_neon_vld3_v(&__ret, __p0, 8); \
34213		__ret; \
34214		})
	35154	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
	35155	uint16x4_t __ret;
	35156	__ret = (uint16x4_t) __builtin_neon_vcvtp_u16_f16((int8x8_t)__p0, 17);
	35157	return __ret;
	35158	}
34215	35159	#else
34216		#define vld3_f16(__p0) __extension__ ({ \
34217		float16x4x3_t __ret; \
34218		__builtin_neon_vld3_v(&__ret, __p0, 8); \
34219		\
34220		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34221		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34222		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34223		__ret; \
34224		})
	35160	__ai __attribute__((target("fullfp16"))) uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
	35161	uint16x4_t __ret;
	35162	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35163	__ret = (uint16x4_t) __builtin_neon_vcvtp_u16_f16((int8x8_t)__rev0, 17);
	35164	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35165	return __ret;
	35166	}
34225	35167	#endif
34226	35168
34227	35169	#ifdef __LITTLE_ENDIAN__
34228		#define vld3q_dup_f16(__p0) __extension__ ({ \
34229		float16x8x3_t __ret; \
34230		__builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
	35170	#define vextq_f16(__p0, __p1, __p2) __extension__ ({ \
	35171	float16x8_t __ret; \
	35172	float16x8_t __s0 = __p0; \
	35173	float16x8_t __s1 = __p1; \
	35174	__ret = (float16x8_t) __builtin_neon_vextq_f16((int8x16_t)__s0, (int8x16_t)__s1, __p2, 40); \
34231	35175	__ret; \
34232	35176	})
34233	35177	#else
34234		#define vld3q_dup_f16(__p0) __extension__ ({ \
34235		float16x8x3_t __ret; \
34236		__builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
34237		\
34238		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34239		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34240		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	35178	#define vextq_f16(__p0, __p1, __p2) __extension__ ({ \
	35179	float16x8_t __ret; \
	35180	float16x8_t __s0 = __p0; \
	35181	float16x8_t __s1 = __p1; \
	35182	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	35183	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	35184	__ret = (float16x8_t) __builtin_neon_vextq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 40); \
	35185	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
34241	35186	__ret; \
34242	35187	})
34243	35188	#endif
34244	35189
34245	35190	#ifdef __LITTLE_ENDIAN__
34246		#define vld3_dup_f16(__p0) __extension__ ({ \
34247		float16x4x3_t __ret; \
34248		__builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
	35191	#define vext_f16(__p0, __p1, __p2) __extension__ ({ \
	35192	float16x4_t __ret; \
	35193	float16x4_t __s0 = __p0; \
	35194	float16x4_t __s1 = __p1; \
	35195	__ret = (float16x4_t) __builtin_neon_vext_f16((int8x8_t)__s0, (int8x8_t)__s1, __p2, 8); \
34249	35196	__ret; \
34250	35197	})
34251	35198	#else
34252		#define vld3_dup_f16(__p0) __extension__ ({ \
34253		float16x4x3_t __ret; \
34254		__builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
34255		\
34256		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34257		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34258		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	35199	#define vext_f16(__p0, __p1, __p2) __extension__ ({ \
	35200	float16x4_t __ret; \
	35201	float16x4_t __s0 = __p0; \
	35202	float16x4_t __s1 = __p1; \
	35203	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	35204	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	35205	__ret = (float16x4_t) __builtin_neon_vext_f16((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 8); \
	35206	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
34259	35207	__ret; \
34260	35208	})
34261	35209	#endif
34262	35210
34263	35211	#ifdef __LITTLE_ENDIAN__
34264		#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34265		float16x8x3_t __ret; \
34266		float16x8x3_t __s1 = __p1; \
34267		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
34268		__ret; \
34269		})
	35212	__ai __attribute__((target("fullfp16"))) float16x8_t vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	35213	float16x8_t __ret;
	35214	__ret = (float16x8_t) __builtin_neon_vfmaq_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	35215	return __ret;
	35216	}
34270	35217	#else
34271		#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34272		float16x8x3_t __ret; \
34273		float16x8x3_t __s1 = __p1; \
34274		float16x8x3_t __rev1; \
34275		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34276		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34277		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34278		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
34279		\
34280		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34281		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34282		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34283		__ret; \
34284		})
	35218	__ai __attribute__((target("fullfp16"))) float16x8_t vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	35219	float16x8_t __ret;
	35220	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35221	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35222	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	35223	__ret = (float16x8_t) __builtin_neon_vfmaq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
	35224	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35225	return __ret;
	35226	}
	35227	__ai __attribute__((target("fullfp16"))) float16x8_t __noswap_vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	35228	float16x8_t __ret;
	35229	__ret = (float16x8_t) __builtin_neon_vfmaq_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	35230	return __ret;
	35231	}
34285	35232	#endif
34286	35233
34287	35234	#ifdef __LITTLE_ENDIAN__
34288		#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34289		float16x4x3_t __ret; \
34290		float16x4x3_t __s1 = __p1; \
34291		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
34292		__ret; \
34293		})
	35235	__ai __attribute__((target("fullfp16"))) float16x4_t vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	35236	float16x4_t __ret;
	35237	__ret = (float16x4_t) __builtin_neon_vfma_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	35238	return __ret;
	35239	}
34294	35240	#else
34295		#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34296		float16x4x3_t __ret; \
34297		float16x4x3_t __s1 = __p1; \
34298		float16x4x3_t __rev1; \
34299		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34300		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34301		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34302		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
34303		\
34304		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34305		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34306		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34307		__ret; \
34308		})
	35241	__ai __attribute__((target("fullfp16"))) float16x4_t vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	35242	float16x4_t __ret;
	35243	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35244	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35245	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	35246	__ret = (float16x4_t) __builtin_neon_vfma_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
	35247	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35248	return __ret;
	35249	}
	35250	__ai __attribute__((target("fullfp16"))) float16x4_t __noswap_vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	35251	float16x4_t __ret;
	35252	__ret = (float16x4_t) __builtin_neon_vfma_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	35253	return __ret;
	35254	}
34309	35255	#endif
34310	35256
34311	35257	#ifdef __LITTLE_ENDIAN__
34312		#define vld4q_f16(__p0) __extension__ ({ \
34313		float16x8x4_t __ret; \
34314		__builtin_neon_vld4q_v(&__ret, __p0, 40); \
34315		__ret; \
34316		})
	35258	__ai __attribute__((target("fullfp16"))) float16x8_t vfmsq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	35259	float16x8_t __ret;
	35260	__ret = vfmaq_f16(__p0, -__p1, __p2);
	35261	return __ret;
	35262	}
34317	35263	#else
34318		#define vld4q_f16(__p0) __extension__ ({ \
34319		float16x8x4_t __ret; \
34320		__builtin_neon_vld4q_v(&__ret, __p0, 40); \
34321		\
34322		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34323		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34324		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34325		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34326		__ret; \
34327		})
	35264	__ai __attribute__((target("fullfp16"))) float16x8_t vfmsq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	35265	float16x8_t __ret;
	35266	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35267	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35268	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	35269	__ret = __noswap_vfmaq_f16(__rev0, -__rev1, __rev2);
	35270	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35271	return __ret;
	35272	}
34328	35273	#endif
34329	35274
34330	35275	#ifdef __LITTLE_ENDIAN__
34331		#define vld4_f16(__p0) __extension__ ({ \
34332		float16x4x4_t __ret; \
34333		__builtin_neon_vld4_v(&__ret, __p0, 8); \
34334		__ret; \
34335		})
	35276	__ai __attribute__((target("fullfp16"))) float16x4_t vfms_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	35277	float16x4_t __ret;
	35278	__ret = vfma_f16(__p0, -__p1, __p2);
	35279	return __ret;
	35280	}
34336	35281	#else
34337		#define vld4_f16(__p0) __extension__ ({ \
34338		float16x4x4_t __ret; \
34339		__builtin_neon_vld4_v(&__ret, __p0, 8); \
34340		\
34341		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34342		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34343		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34344		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
34345		__ret; \
34346		})
	35282	__ai __attribute__((target("fullfp16"))) float16x4_t vfms_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	35283	float16x4_t __ret;
	35284	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35285	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35286	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	35287	__ret = __noswap_vfma_f16(__rev0, -__rev1, __rev2);
	35288	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35289	return __ret;
	35290	}
34347	35291	#endif
34348	35292
34349	35293	#ifdef __LITTLE_ENDIAN__
34350		#define vld4q_dup_f16(__p0) __extension__ ({ \
34351		float16x8x4_t __ret; \
34352		__builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
34353		__ret; \
34354		})
	35294	__ai __attribute__((target("fullfp16"))) float16x8_t vmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
	35295	float16x8_t __ret;
	35296	__ret = (float16x8_t) __builtin_neon_vmaxq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
	35297	return __ret;
	35298	}
34355	35299	#else
34356		#define vld4q_dup_f16(__p0) __extension__ ({ \
34357		float16x8x4_t __ret; \
34358		__builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
34359		\
34360		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34361		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34362		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34363		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34364		__ret; \
34365		})
	35300	__ai __attribute__((target("fullfp16"))) float16x8_t vmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
	35301	float16x8_t __ret;
	35302	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35303	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35304	__ret = (float16x8_t) __builtin_neon_vmaxq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35305	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35306	return __ret;
	35307	}
34366	35308	#endif
34367	35309
34368	35310	#ifdef __LITTLE_ENDIAN__
34369		#define vld4_dup_f16(__p0) __extension__ ({ \
34370		float16x4x4_t __ret; \
34371		__builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
34372		__ret; \
34373		})
	35311	__ai __attribute__((target("fullfp16"))) float16x4_t vmax_f16(float16x4_t __p0, float16x4_t __p1) {
	35312	float16x4_t __ret;
	35313	__ret = (float16x4_t) __builtin_neon_vmax_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35314	return __ret;
	35315	}
34374	35316	#else
34375		#define vld4_dup_f16(__p0) __extension__ ({ \
34376		float16x4x4_t __ret; \
34377		__builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
34378		\
34379		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34380		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34381		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34382		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
34383		__ret; \
34384		})
	35317	__ai __attribute__((target("fullfp16"))) float16x4_t vmax_f16(float16x4_t __p0, float16x4_t __p1) {
	35318	float16x4_t __ret;
	35319	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35320	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35321	__ret = (float16x4_t) __builtin_neon_vmax_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35322	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35323	return __ret;
	35324	}
34385	35325	#endif
34386	35326
34387	35327	#ifdef __LITTLE_ENDIAN__
34388		#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34389		float16x8x4_t __ret; \
34390		float16x8x4_t __s1 = __p1; \
34391		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 40); \
34392		__ret; \
34393		})
	35328	__ai __attribute__((target("fullfp16"))) float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
	35329	float16x8_t __ret;
	35330	__ret = (float16x8_t) __builtin_neon_vminq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
	35331	return __ret;
	35332	}
34394	35333	#else
34395		#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34396		float16x8x4_t __ret; \
34397		float16x8x4_t __s1 = __p1; \
34398		float16x8x4_t __rev1; \
34399		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34400		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34401		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34402		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34403		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 40); \
34404		\
34405		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34406		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34407		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34408		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34409		__ret; \
34410		})
	35334	__ai __attribute__((target("fullfp16"))) float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
	35335	float16x8_t __ret;
	35336	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35337	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35338	__ret = (float16x8_t) __builtin_neon_vminq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35339	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35340	return __ret;
	35341	}
34411	35342	#endif
34412	35343
34413	35344	#ifdef __LITTLE_ENDIAN__
34414		#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34415		float16x4x4_t __ret; \
34416		float16x4x4_t __s1 = __p1; \
34417		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 8); \
34418		__ret; \
34419		})
	35345	__ai __attribute__((target("fullfp16"))) float16x4_t vmin_f16(float16x4_t __p0, float16x4_t __p1) {
	35346	float16x4_t __ret;
	35347	__ret = (float16x4_t) __builtin_neon_vmin_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35348	return __ret;
	35349	}
34420	35350	#else
34421		#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34422		float16x4x4_t __ret; \
34423		float16x4x4_t __s1 = __p1; \
34424		float16x4x4_t __rev1; \
34425		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34426		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34427		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34428		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
34429		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 8); \
34430		\
34431		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
34432		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
34433		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
34434		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
34435		__ret; \
34436		})
	35351	__ai __attribute__((target("fullfp16"))) float16x4_t vmin_f16(float16x4_t __p0, float16x4_t __p1) {
	35352	float16x4_t __ret;
	35353	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35354	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35355	__ret = (float16x4_t) __builtin_neon_vmin_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35356	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35357	return __ret;
	35358	}
34437	35359	#endif
34438	35360
34439	35361	#ifdef __LITTLE_ENDIAN__
34440		#define vst1q_f16(__p0, __p1) __extension__ ({ \
34441		float16x8_t __s1 = __p1; \
34442		__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 40); \
34443		})
	35362	__ai __attribute__((target("fullfp16"))) float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
	35363	float16x8_t __ret;
	35364	__ret = __p0 * __p1;
	35365	return __ret;
	35366	}
34444	35367	#else
34445		#define vst1q_f16(__p0, __p1) __extension__ ({ \
34446		float16x8_t __s1 = __p1; \
34447		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
34448		__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 40); \
34449		})
	35368	__ai __attribute__((target("fullfp16"))) float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
	35369	float16x8_t __ret;
	35370	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35371	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35372	__ret = __rev0 * __rev1;
	35373	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35374	return __ret;
	35375	}
34450	35376	#endif
34451	35377
34452	35378	#ifdef __LITTLE_ENDIAN__
34453		#define vst1_f16(__p0, __p1) __extension__ ({ \
34454		float16x4_t __s1 = __p1; \
34455		__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 8); \
34456		})
	35379	__ai __attribute__((target("fullfp16"))) float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
	35380	float16x4_t __ret;
	35381	__ret = __p0 * __p1;
	35382	return __ret;
	35383	}
34457	35384	#else
34458		#define vst1_f16(__p0, __p1) __extension__ ({ \
34459		float16x4_t __s1 = __p1; \
34460		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
34461		__builtin_neon_vst1_v(__p0, (int8x8_t)__rev1, 8); \
34462		})
	35385	__ai __attribute__((target("fullfp16"))) float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
	35386	float16x4_t __ret;
	35387	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35388	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35389	__ret = __rev0 * __rev1;
	35390	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35391	return __ret;
	35392	}
34463	35393	#endif
34464	35394
34465	35395	#ifdef __LITTLE_ENDIAN__
34466		#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34467		float16x8_t __s1 = __p1; \
34468		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
	35396	#define vmulq_lane_f16(__p0_153, __p1_153, __p2_153) __extension__ ({ \
	35397	float16x8_t __ret_153; \
	35398	float16x8_t __s0_153 = __p0_153; \
	35399	float16x4_t __s1_153 = __p1_153; \
	35400	__ret_153 = __s0_153 * splatq_lane_f16(__s1_153, __p2_153); \
	35401	__ret_153; \
34469	35402	})
34470	35403	#else
34471		#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34472		float16x8_t __s1 = __p1; \
34473		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
34474		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
	35404	#define vmulq_lane_f16(__p0_154, __p1_154, __p2_154) __extension__ ({ \
	35405	float16x8_t __ret_154; \
	35406	float16x8_t __s0_154 = __p0_154; \
	35407	float16x4_t __s1_154 = __p1_154; \
	35408	float16x8_t __rev0_154; __rev0_154 = __builtin_shufflevector(__s0_154, __s0_154, 7, 6, 5, 4, 3, 2, 1, 0); \
	35409	float16x4_t __rev1_154; __rev1_154 = __builtin_shufflevector(__s1_154, __s1_154, 3, 2, 1, 0); \
	35410	__ret_154 = __rev0_154 * __noswap_splatq_lane_f16(__rev1_154, __p2_154); \
	35411	__ret_154 = __builtin_shufflevector(__ret_154, __ret_154, 7, 6, 5, 4, 3, 2, 1, 0); \
	35412	__ret_154; \
34475	35413	})
34476	35414	#endif
34477	35415
34478	35416	#ifdef __LITTLE_ENDIAN__
34479		#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34480		float16x4_t __s1 = __p1; \
34481		__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
	35417	#define vmul_lane_f16(__p0_155, __p1_155, __p2_155) __extension__ ({ \
	35418	float16x4_t __ret_155; \
	35419	float16x4_t __s0_155 = __p0_155; \
	35420	float16x4_t __s1_155 = __p1_155; \
	35421	__ret_155 = __s0_155 * splat_lane_f16(__s1_155, __p2_155); \
	35422	__ret_155; \
34482	35423	})
34483	35424	#else
34484		#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34485		float16x4_t __s1 = __p1; \
34486		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
34487		__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
	35425	#define vmul_lane_f16(__p0_156, __p1_156, __p2_156) __extension__ ({ \
	35426	float16x4_t __ret_156; \
	35427	float16x4_t __s0_156 = __p0_156; \
	35428	float16x4_t __s1_156 = __p1_156; \
	35429	float16x4_t __rev0_156; __rev0_156 = __builtin_shufflevector(__s0_156, __s0_156, 3, 2, 1, 0); \
	35430	float16x4_t __rev1_156; __rev1_156 = __builtin_shufflevector(__s1_156, __s1_156, 3, 2, 1, 0); \
	35431	__ret_156 = __rev0_156 * __noswap_splat_lane_f16(__rev1_156, __p2_156); \
	35432	__ret_156 = __builtin_shufflevector(__ret_156, __ret_156, 3, 2, 1, 0); \
	35433	__ret_156; \
34488	35434	})
34489	35435	#endif
34490	35436
34491	35437	#ifdef __LITTLE_ENDIAN__
34492		#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
34493		float16x8x2_t __s1 = __p1; \
34494		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 40); \
	35438	#define vmulq_n_f16(__p0, __p1) __extension__ ({ \
	35439	float16x8_t __ret; \
	35440	float16x8_t __s0 = __p0; \
	35441	float16_t __s1 = __p1; \
	35442	__ret = __s0 * (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}; \
	35443	__ret; \
34495	35444	})
34496	35445	#else
34497		#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
34498		float16x8x2_t __s1 = __p1; \
34499		float16x8x2_t __rev1; \
34500		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34501		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34502		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 40); \
	35446	#define vmulq_n_f16(__p0, __p1) __extension__ ({ \
	35447	float16x8_t __ret; \
	35448	float16x8_t __s0 = __p0; \
	35449	float16_t __s1 = __p1; \
	35450	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	35451	__ret = __rev0 * (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}; \
	35452	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	35453	__ret; \
34503	35454	})
34504	35455	#endif
34505	35456
34506	35457	#ifdef __LITTLE_ENDIAN__
34507		#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
34508		float16x4x2_t __s1 = __p1; \
34509		__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 8); \
	35458	#define vmul_n_f16(__p0, __p1) __extension__ ({ \
	35459	float16x4_t __ret; \
	35460	float16x4_t __s0 = __p0; \
	35461	float16_t __s1 = __p1; \
	35462	__ret = __s0 * (float16x4_t) {__s1, __s1, __s1, __s1}; \
	35463	__ret; \
34510	35464	})
34511	35465	#else
34512		#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
34513		float16x4x2_t __s1 = __p1; \
34514		float16x4x2_t __rev1; \
34515		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34516		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34517		__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 8); \
	35466	#define vmul_n_f16(__p0, __p1) __extension__ ({ \
	35467	float16x4_t __ret; \
	35468	float16x4_t __s0 = __p0; \
	35469	float16_t __s1 = __p1; \
	35470	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	35471	__ret = __rev0 * (float16x4_t) {__s1, __s1, __s1, __s1}; \
	35472	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	35473	__ret; \
34518	35474	})
34519	35475	#endif
34520	35476
34521	35477	#ifdef __LITTLE_ENDIAN__
34522		#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
34523		float16x8x3_t __s1 = __p1; \
34524		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 40); \
34525		})
	35478	__ai __attribute__((target("fullfp16"))) float16x8_t vnegq_f16(float16x8_t __p0) {
	35479	float16x8_t __ret;
	35480	__ret = -__p0;
	35481	return __ret;
	35482	}
34526	35483	#else
34527		#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
34528		float16x8x3_t __s1 = __p1; \
34529		float16x8x3_t __rev1; \
34530		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34531		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34532		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34533		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 40); \
34534		})
	35484	__ai __attribute__((target("fullfp16"))) float16x8_t vnegq_f16(float16x8_t __p0) {
	35485	float16x8_t __ret;
	35486	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35487	__ret = -__rev0;
	35488	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35489	return __ret;
	35490	}
34535	35491	#endif
34536	35492
34537	35493	#ifdef __LITTLE_ENDIAN__
34538		#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
34539		float16x4x3_t __s1 = __p1; \
34540		__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 8); \
34541		})
	35494	__ai __attribute__((target("fullfp16"))) float16x4_t vneg_f16(float16x4_t __p0) {
	35495	float16x4_t __ret;
	35496	__ret = -__p0;
	35497	return __ret;
	35498	}
34542	35499	#else
34543		#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
34544		float16x4x3_t __s1 = __p1; \
34545		float16x4x3_t __rev1; \
34546		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34547		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34548		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34549		__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 8); \
34550		})
	35500	__ai __attribute__((target("fullfp16"))) float16x4_t vneg_f16(float16x4_t __p0) {
	35501	float16x4_t __ret;
	35502	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35503	__ret = -__rev0;
	35504	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35505	return __ret;
	35506	}
34551	35507	#endif
34552	35508
34553	35509	#ifdef __LITTLE_ENDIAN__
34554		#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
34555		float16x8x4_t __s1 = __p1; \
34556		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 40); \
34557		})
	35510	__ai __attribute__((target("fullfp16"))) float16x4_t vpadd_f16(float16x4_t __p0, float16x4_t __p1) {
	35511	float16x4_t __ret;
	35512	__ret = (float16x4_t) __builtin_neon_vpadd_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35513	return __ret;
	35514	}
34558	35515	#else
34559		#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
34560		float16x8x4_t __s1 = __p1; \
34561		float16x8x4_t __rev1; \
34562		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34563		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34564		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34565		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34566		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 40); \
34567		})
	35516	__ai __attribute__((target("fullfp16"))) float16x4_t vpadd_f16(float16x4_t __p0, float16x4_t __p1) {
	35517	float16x4_t __ret;
	35518	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35519	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35520	__ret = (float16x4_t) __builtin_neon_vpadd_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35521	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35522	return __ret;
	35523	}
34568	35524	#endif
34569	35525
34570	35526	#ifdef __LITTLE_ENDIAN__
34571		#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
34572		float16x4x4_t __s1 = __p1; \
34573		__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 8); \
34574		})
	35527	__ai __attribute__((target("fullfp16"))) float16x4_t vpmax_f16(float16x4_t __p0, float16x4_t __p1) {
	35528	float16x4_t __ret;
	35529	__ret = (float16x4_t) __builtin_neon_vpmax_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35530	return __ret;
	35531	}
34575	35532	#else
34576		#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
34577		float16x4x4_t __s1 = __p1; \
34578		float16x4x4_t __rev1; \
34579		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34580		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34581		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34582		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
34583		__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 8); \
34584		})
	35533	__ai __attribute__((target("fullfp16"))) float16x4_t vpmax_f16(float16x4_t __p0, float16x4_t __p1) {
	35534	float16x4_t __ret;
	35535	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35536	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35537	__ret = (float16x4_t) __builtin_neon_vpmax_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35538	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35539	return __ret;
	35540	}
34585	35541	#endif
34586	35542
34587	35543	#ifdef __LITTLE_ENDIAN__
34588		#define vst2q_f16(__p0, __p1) __extension__ ({ \
34589		float16x8x2_t __s1 = __p1; \
34590		__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 40); \
34591		})
	35544	__ai __attribute__((target("fullfp16"))) float16x4_t vpmin_f16(float16x4_t __p0, float16x4_t __p1) {
	35545	float16x4_t __ret;
	35546	__ret = (float16x4_t) __builtin_neon_vpmin_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35547	return __ret;
	35548	}
34592	35549	#else
34593		#define vst2q_f16(__p0, __p1) __extension__ ({ \
34594		float16x8x2_t __s1 = __p1; \
34595		float16x8x2_t __rev1; \
34596		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34597		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34598		__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 40); \
34599		})
	35550	__ai __attribute__((target("fullfp16"))) float16x4_t vpmin_f16(float16x4_t __p0, float16x4_t __p1) {
	35551	float16x4_t __ret;
	35552	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35553	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35554	__ret = (float16x4_t) __builtin_neon_vpmin_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35555	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35556	return __ret;
	35557	}
34600	35558	#endif
34601	35559
34602	35560	#ifdef __LITTLE_ENDIAN__
34603		#define vst2_f16(__p0, __p1) __extension__ ({ \
34604		float16x4x2_t __s1 = __p1; \
34605		__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 8); \
34606		})
	35561	__ai __attribute__((target("fullfp16"))) float16x8_t vrecpeq_f16(float16x8_t __p0) {
	35562	float16x8_t __ret;
	35563	__ret = (float16x8_t) __builtin_neon_vrecpeq_f16((int8x16_t)__p0, 40);
	35564	return __ret;
	35565	}
34607	35566	#else
34608		#define vst2_f16(__p0, __p1) __extension__ ({ \
34609		float16x4x2_t __s1 = __p1; \
34610		float16x4x2_t __rev1; \
34611		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34612		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34613		__builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 8); \
34614		})
	35567	__ai __attribute__((target("fullfp16"))) float16x8_t vrecpeq_f16(float16x8_t __p0) {
	35568	float16x8_t __ret;
	35569	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35570	__ret = (float16x8_t) __builtin_neon_vrecpeq_f16((int8x16_t)__rev0, 40);
	35571	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35572	return __ret;
	35573	}
34615	35574	#endif
34616	35575
34617	35576	#ifdef __LITTLE_ENDIAN__
34618		#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34619		float16x8x2_t __s1 = __p1; \
34620		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
34621		})
	35577	__ai __attribute__((target("fullfp16"))) float16x4_t vrecpe_f16(float16x4_t __p0) {
	35578	float16x4_t __ret;
	35579	__ret = (float16x4_t) __builtin_neon_vrecpe_f16((int8x8_t)__p0, 8);
	35580	return __ret;
	35581	}
34622	35582	#else
34623		#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34624		float16x8x2_t __s1 = __p1; \
34625		float16x8x2_t __rev1; \
34626		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34627		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34628		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
34629		})
	35583	__ai __attribute__((target("fullfp16"))) float16x4_t vrecpe_f16(float16x4_t __p0) {
	35584	float16x4_t __ret;
	35585	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35586	__ret = (float16x4_t) __builtin_neon_vrecpe_f16((int8x8_t)__rev0, 8);
	35587	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35588	return __ret;
	35589	}
34630	35590	#endif
34631	35591
34632	35592	#ifdef __LITTLE_ENDIAN__
34633		#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34634		float16x4x2_t __s1 = __p1; \
34635		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
34636		})
	35593	__ai __attribute__((target("fullfp16"))) float16x8_t vrecpsq_f16(float16x8_t __p0, float16x8_t __p1) {
	35594	float16x8_t __ret;
	35595	__ret = (float16x8_t) __builtin_neon_vrecpsq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
	35596	return __ret;
	35597	}
34637	35598	#else
34638		#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34639		float16x4x2_t __s1 = __p1; \
34640		float16x4x2_t __rev1; \
34641		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34642		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34643		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
34644		})
	35599	__ai __attribute__((target("fullfp16"))) float16x8_t vrecpsq_f16(float16x8_t __p0, float16x8_t __p1) {
	35600	float16x8_t __ret;
	35601	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35602	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35603	__ret = (float16x8_t) __builtin_neon_vrecpsq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35604	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35605	return __ret;
	35606	}
34645	35607	#endif
34646	35608
34647	35609	#ifdef __LITTLE_ENDIAN__
34648		#define vst3q_f16(__p0, __p1) __extension__ ({ \
34649		float16x8x3_t __s1 = __p1; \
34650		__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 40); \
34651		})
	35610	__ai __attribute__((target("fullfp16"))) float16x4_t vrecps_f16(float16x4_t __p0, float16x4_t __p1) {
	35611	float16x4_t __ret;
	35612	__ret = (float16x4_t) __builtin_neon_vrecps_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35613	return __ret;
	35614	}
34652	35615	#else
34653		#define vst3q_f16(__p0, __p1) __extension__ ({ \
34654		float16x8x3_t __s1 = __p1; \
34655		float16x8x3_t __rev1; \
34656		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34657		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34658		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34659		__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 40); \
34660		})
	35616	__ai __attribute__((target("fullfp16"))) float16x4_t vrecps_f16(float16x4_t __p0, float16x4_t __p1) {
	35617	float16x4_t __ret;
	35618	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35619	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35620	__ret = (float16x4_t) __builtin_neon_vrecps_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35621	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35622	return __ret;
	35623	}
34661	35624	#endif
34662	35625
34663	35626	#ifdef __LITTLE_ENDIAN__
34664		#define vst3_f16(__p0, __p1) __extension__ ({ \
34665		float16x4x3_t __s1 = __p1; \
34666		__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 8); \
34667		})
	35627	__ai __attribute__((target("fullfp16"))) float16x8_t vrev64q_f16(float16x8_t __p0) {
	35628	float16x8_t __ret;
	35629	__ret = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0, 7, 6, 5, 4);
	35630	return __ret;
	35631	}
34668	35632	#else
34669		#define vst3_f16(__p0, __p1) __extension__ ({ \
34670		float16x4x3_t __s1 = __p1; \
34671		float16x4x3_t __rev1; \
34672		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34673		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34674		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34675		__builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 8); \
34676		})
	35633	__ai __attribute__((target("fullfp16"))) float16x8_t vrev64q_f16(float16x8_t __p0) {
	35634	float16x8_t __ret;
	35635	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35636	__ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
	35637	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35638	return __ret;
	35639	}
34677	35640	#endif
34678	35641
34679	35642	#ifdef __LITTLE_ENDIAN__
34680		#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34681		float16x8x3_t __s1 = __p1; \
34682		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
34683		})
	35643	__ai __attribute__((target("fullfp16"))) float16x4_t vrev64_f16(float16x4_t __p0) {
	35644	float16x4_t __ret;
	35645	__ret = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35646	return __ret;
	35647	}
34684	35648	#else
34685		#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34686		float16x8x3_t __s1 = __p1; \
34687		float16x8x3_t __rev1; \
34688		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34689		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34690		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34691		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
34692		})
	35649	__ai __attribute__((target("fullfp16"))) float16x4_t vrev64_f16(float16x4_t __p0) {
	35650	float16x4_t __ret;
	35651	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35652	__ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0);
	35653	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35654	return __ret;
	35655	}
34693	35656	#endif
34694	35657
34695	35658	#ifdef __LITTLE_ENDIAN__
34696		#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34697		float16x4x3_t __s1 = __p1; \
34698		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
34699		})
	35659	__ai __attribute__((target("fullfp16"))) float16x8_t vrsqrteq_f16(float16x8_t __p0) {
	35660	float16x8_t __ret;
	35661	__ret = (float16x8_t) __builtin_neon_vrsqrteq_f16((int8x16_t)__p0, 40);
	35662	return __ret;
	35663	}
34700	35664	#else
34701		#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34702		float16x4x3_t __s1 = __p1; \
34703		float16x4x3_t __rev1; \
34704		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34705		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34706		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34707		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
34708		})
	35665	__ai __attribute__((target("fullfp16"))) float16x8_t vrsqrteq_f16(float16x8_t __p0) {
	35666	float16x8_t __ret;
	35667	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35668	__ret = (float16x8_t) __builtin_neon_vrsqrteq_f16((int8x16_t)__rev0, 40);
	35669	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35670	return __ret;
	35671	}
34709	35672	#endif
34710	35673
34711	35674	#ifdef __LITTLE_ENDIAN__
34712		#define vst4q_f16(__p0, __p1) __extension__ ({ \
34713		float16x8x4_t __s1 = __p1; \
34714		__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 40); \
34715		})
	35675	__ai __attribute__((target("fullfp16"))) float16x4_t vrsqrte_f16(float16x4_t __p0) {
	35676	float16x4_t __ret;
	35677	__ret = (float16x4_t) __builtin_neon_vrsqrte_f16((int8x8_t)__p0, 8);
	35678	return __ret;
	35679	}
34716	35680	#else
34717		#define vst4q_f16(__p0, __p1) __extension__ ({ \
34718		float16x8x4_t __s1 = __p1; \
34719		float16x8x4_t __rev1; \
34720		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34721		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34722		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34723		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34724		__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 40); \
34725		})
	35681	__ai __attribute__((target("fullfp16"))) float16x4_t vrsqrte_f16(float16x4_t __p0) {
	35682	float16x4_t __ret;
	35683	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35684	__ret = (float16x4_t) __builtin_neon_vrsqrte_f16((int8x8_t)__rev0, 8);
	35685	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35686	return __ret;
	35687	}
34726	35688	#endif
34727	35689
34728	35690	#ifdef __LITTLE_ENDIAN__
34729		#define vst4_f16(__p0, __p1) __extension__ ({ \
34730		float16x4x4_t __s1 = __p1; \
34731		__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 8); \
34732		})
	35691	__ai __attribute__((target("fullfp16"))) float16x8_t vrsqrtsq_f16(float16x8_t __p0, float16x8_t __p1) {
	35692	float16x8_t __ret;
	35693	__ret = (float16x8_t) __builtin_neon_vrsqrtsq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
	35694	return __ret;
	35695	}
34733	35696	#else
34734		#define vst4_f16(__p0, __p1) __extension__ ({ \
34735		float16x4x4_t __s1 = __p1; \
34736		float16x4x4_t __rev1; \
34737		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34738		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34739		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34740		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
34741		__builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 8); \
34742		})
	35697	__ai __attribute__((target("fullfp16"))) float16x8_t vrsqrtsq_f16(float16x8_t __p0, float16x8_t __p1) {
	35698	float16x8_t __ret;
	35699	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35700	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35701	__ret = (float16x8_t) __builtin_neon_vrsqrtsq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35702	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35703	return __ret;
	35704	}
34743	35705	#endif
34744	35706
34745	35707	#ifdef __LITTLE_ENDIAN__
34746		#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34747		float16x8x4_t __s1 = __p1; \
34748		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 40); \
34749		})
	35708	__ai __attribute__((target("fullfp16"))) float16x4_t vrsqrts_f16(float16x4_t __p0, float16x4_t __p1) {
	35709	float16x4_t __ret;
	35710	__ret = (float16x4_t) __builtin_neon_vrsqrts_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	35711	return __ret;
	35712	}
34750	35713	#else
34751		#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34752		float16x8x4_t __s1 = __p1; \
34753		float16x8x4_t __rev1; \
34754		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
34755		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34756		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34757		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34758		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 40); \
34759		})
	35714	__ai __attribute__((target("fullfp16"))) float16x4_t vrsqrts_f16(float16x4_t __p0, float16x4_t __p1) {
	35715	float16x4_t __ret;
	35716	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35717	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35718	__ret = (float16x4_t) __builtin_neon_vrsqrts_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35719	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35720	return __ret;
	35721	}
34760	35722	#endif
34761	35723
34762	35724	#ifdef __LITTLE_ENDIAN__
34763		#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34764		float16x4x4_t __s1 = __p1; \
34765		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 8); \
34766		})
	35725	__ai __attribute__((target("fullfp16"))) float16x8_t vsubq_f16(float16x8_t __p0, float16x8_t __p1) {
	35726	float16x8_t __ret;
	35727	__ret = __p0 - __p1;
	35728	return __ret;
	35729	}
34767	35730	#else
34768		#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34769		float16x4x4_t __s1 = __p1; \
34770		float16x4x4_t __rev1; \
34771		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
34772		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34773		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34774		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
34775		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 8); \
34776		})
	35731	__ai __attribute__((target("fullfp16"))) float16x8_t vsubq_f16(float16x8_t __p0, float16x8_t __p1) {
	35732	float16x8_t __ret;
	35733	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35734	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35735	__ret = __rev0 - __rev1;
	35736	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	35737	return __ret;
	35738	}
34777	35739	#endif
34778	35740
34779		#endif
34780		#if __ARM_ARCH >= 8
34781	35741	#ifdef __LITTLE_ENDIAN__
34782		__ai int32x4_t vcvtaq_s32_f32(float32x4_t __p0) {
34783		int32x4_t __ret;
34784		__ret = (int32x4_t) __builtin_neon_vcvtaq_s32_v((int8x16_t)__p0, 34);
	35742	__ai __attribute__((target("fullfp16"))) float16x4_t vsub_f16(float16x4_t __p0, float16x4_t __p1) {
	35743	float16x4_t __ret;
	35744	__ret = __p0 - __p1;
34785	35745	return __ret;
34786	35746	}
34787	35747	#else
34788		__ai int32x4_t vcvtaq_s32_f32(float32x4_t __p0) {
34789		int32x4_t __ret;
34790		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34791		__ret = (int32x4_t) __builtin_neon_vcvtaq_s32_v((int8x16_t)__rev0, 34);
	35748	__ai __attribute__((target("fullfp16"))) float16x4_t vsub_f16(float16x4_t __p0, float16x4_t __p1) {
	35749	float16x4_t __ret;
	35750	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35751	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35752	__ret = __rev0 - __rev1;
34792	35753	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
34793	35754	return __ret;
34794	35755	}
34795	35756	#endif
34796	35757
34797	35758	#ifdef __LITTLE_ENDIAN__
34798		__ai int32x2_t vcvta_s32_f32(float32x2_t __p0) {
34799		int32x2_t __ret;
34800		__ret = (int32x2_t) __builtin_neon_vcvta_s32_v((int8x8_t)__p0, 2);
	35759	__ai __attribute__((target("fullfp16"))) float16x8x2_t vtrnq_f16(float16x8_t __p0, float16x8_t __p1) {
	35760	float16x8x2_t __ret;
	35761	__builtin_neon_vtrnq_f16(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 40);
34801	35762	return __ret;
34802	35763	}
34803	35764	#else
34804		__ai int32x2_t vcvta_s32_f32(float32x2_t __p0) {
34805		int32x2_t __ret;
34806		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34807		__ret = (int32x2_t) __builtin_neon_vcvta_s32_v((int8x8_t)__rev0, 2);
34808		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	35765	__ai __attribute__((target("fullfp16"))) float16x8x2_t vtrnq_f16(float16x8_t __p0, float16x8_t __p1) {
	35766	float16x8x2_t __ret;
	35767	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35768	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35769	__builtin_neon_vtrnq_f16(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35770
	35771	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
	35772	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
34809	35773	return __ret;
34810	35774	}
34811	35775	#endif
34812	35776
34813	35777	#ifdef __LITTLE_ENDIAN__
34814		__ai uint32x4_t vcvtaq_u32_f32(float32x4_t __p0) {
34815		uint32x4_t __ret;
34816		__ret = (uint32x4_t) __builtin_neon_vcvtaq_u32_v((int8x16_t)__p0, 50);
	35778	__ai __attribute__((target("fullfp16"))) float16x4x2_t vtrn_f16(float16x4_t __p0, float16x4_t __p1) {
	35779	float16x4x2_t __ret;
	35780	__builtin_neon_vtrn_f16(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 8);
34817	35781	return __ret;
34818	35782	}
34819	35783	#else
34820		__ai uint32x4_t vcvtaq_u32_f32(float32x4_t __p0) {
34821		uint32x4_t __ret;
34822		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34823		__ret = (uint32x4_t) __builtin_neon_vcvtaq_u32_v((int8x16_t)__rev0, 50);
34824		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35784	__ai __attribute__((target("fullfp16"))) float16x4x2_t vtrn_f16(float16x4_t __p0, float16x4_t __p1) {
	35785	float16x4x2_t __ret;
	35786	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35787	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35788	__builtin_neon_vtrn_f16(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35789
	35790	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
	35791	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0);
34825	35792	return __ret;
34826	35793	}
34827	35794	#endif
34828	35795
34829	35796	#ifdef __LITTLE_ENDIAN__
34830		__ai uint32x2_t vcvta_u32_f32(float32x2_t __p0) {
34831		uint32x2_t __ret;
34832		__ret = (uint32x2_t) __builtin_neon_vcvta_u32_v((int8x8_t)__p0, 18);
	35797	__ai __attribute__((target("fullfp16"))) float16x8x2_t vuzpq_f16(float16x8_t __p0, float16x8_t __p1) {
	35798	float16x8x2_t __ret;
	35799	__builtin_neon_vuzpq_f16(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 40);
34833	35800	return __ret;
34834	35801	}
34835	35802	#else
34836		__ai uint32x2_t vcvta_u32_f32(float32x2_t __p0) {
34837		uint32x2_t __ret;
34838		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34839		__ret = (uint32x2_t) __builtin_neon_vcvta_u32_v((int8x8_t)__rev0, 18);
34840		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	35803	__ai __attribute__((target("fullfp16"))) float16x8x2_t vuzpq_f16(float16x8_t __p0, float16x8_t __p1) {
	35804	float16x8x2_t __ret;
	35805	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35806	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35807	__builtin_neon_vuzpq_f16(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35808
	35809	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
	35810	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
34841	35811	return __ret;
34842	35812	}
34843	35813	#endif
34844	35814
34845	35815	#ifdef __LITTLE_ENDIAN__
34846		__ai int32x4_t vcvtmq_s32_f32(float32x4_t __p0) {
34847		int32x4_t __ret;
34848		__ret = (int32x4_t) __builtin_neon_vcvtmq_s32_v((int8x16_t)__p0, 34);
	35816	__ai __attribute__((target("fullfp16"))) float16x4x2_t vuzp_f16(float16x4_t __p0, float16x4_t __p1) {
	35817	float16x4x2_t __ret;
	35818	__builtin_neon_vuzp_f16(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 8);
34849	35819	return __ret;
34850	35820	}
34851	35821	#else
34852		__ai int32x4_t vcvtmq_s32_f32(float32x4_t __p0) {
34853		int32x4_t __ret;
34854		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34855		__ret = (int32x4_t) __builtin_neon_vcvtmq_s32_v((int8x16_t)__rev0, 34);
34856		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	35822	__ai __attribute__((target("fullfp16"))) float16x4x2_t vuzp_f16(float16x4_t __p0, float16x4_t __p1) {
	35823	float16x4x2_t __ret;
	35824	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35825	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35826	__builtin_neon_vuzp_f16(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35827
	35828	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
	35829	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0);
34857	35830	return __ret;
34858	35831	}
34859	35832	#endif
34860	35833
34861	35834	#ifdef __LITTLE_ENDIAN__
34862		__ai int32x2_t vcvtm_s32_f32(float32x2_t __p0) {
34863		int32x2_t __ret;
34864		__ret = (int32x2_t) __builtin_neon_vcvtm_s32_v((int8x8_t)__p0, 2);
	35835	__ai __attribute__((target("fullfp16"))) float16x8x2_t vzipq_f16(float16x8_t __p0, float16x8_t __p1) {
	35836	float16x8x2_t __ret;
	35837	__builtin_neon_vzipq_f16(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 40);
34865	35838	return __ret;
34866	35839	}
34867	35840	#else
34868		__ai int32x2_t vcvtm_s32_f32(float32x2_t __p0) {
34869		int32x2_t __ret;
34870		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34871		__ret = (int32x2_t) __builtin_neon_vcvtm_s32_v((int8x8_t)__rev0, 2);
34872		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	35841	__ai __attribute__((target("fullfp16"))) float16x8x2_t vzipq_f16(float16x8_t __p0, float16x8_t __p1) {
	35842	float16x8x2_t __ret;
	35843	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	35844	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35845	__builtin_neon_vzipq_f16(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	35846
	35847	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
	35848	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
34873	35849	return __ret;
34874	35850	}
34875	35851	#endif
34876	35852
34877	35853	#ifdef __LITTLE_ENDIAN__
34878		__ai uint32x4_t vcvtmq_u32_f32(float32x4_t __p0) {
	35854	__ai __attribute__((target("fullfp16"))) float16x4x2_t vzip_f16(float16x4_t __p0, float16x4_t __p1) {
	35855	float16x4x2_t __ret;
	35856	__builtin_neon_vzip_f16(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 8);
	35857	return __ret;
	35858	}
	35859	#else
	35860	__ai __attribute__((target("fullfp16"))) float16x4x2_t vzip_f16(float16x4_t __p0, float16x4_t __p1) {
	35861	float16x4x2_t __ret;
	35862	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35863	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	35864	__builtin_neon_vzip_f16(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	35865
	35866	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
	35867	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0);
	35868	return __ret;
	35869	}
	35870	#endif
	35871
	35872	#ifdef __LITTLE_ENDIAN__
	35873	__ai __attribute__((target("i8mm"))) uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
34879	35874	uint32x4_t __ret;
34880		__ret = (uint32x4_t) __builtin_neon_vcvtmq_u32_v((int8x16_t)__p0, 50);
	35875	__ret = (uint32x4_t) __builtin_neon_vmmlaq_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
34881	35876	return __ret;
34882	35877	}
34883	35878	#else
34884		__ai uint32x4_t vcvtmq_u32_f32(float32x4_t __p0) {
	35879	__ai __attribute__((target("i8mm"))) uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
34885	35880	uint32x4_t __ret;
34886		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34887		__ret = (uint32x4_t) __builtin_neon_vcvtmq_u32_v((int8x16_t)__rev0, 50);
	35881	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35882	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	35883	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	35884	__ret = (uint32x4_t) __builtin_neon_vmmlaq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
34888	35885	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
34889	35886	return __ret;
34890	35887	}
34891	35888	#endif
34892	35889
34893	35890	#ifdef __LITTLE_ENDIAN__
34894		__ai uint32x2_t vcvtm_u32_f32(float32x2_t __p0) {
34895		uint32x2_t __ret;
34896		__ret = (uint32x2_t) __builtin_neon_vcvtm_u32_v((int8x8_t)__p0, 18);
	35891	__ai __attribute__((target("i8mm"))) int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
	35892	int32x4_t __ret;
	35893	__ret = (int32x4_t) __builtin_neon_vmmlaq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
34897	35894	return __ret;
34898	35895	}
34899	35896	#else
34900		__ai uint32x2_t vcvtm_u32_f32(float32x2_t __p0) {
34901		uint32x2_t __ret;
34902		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34903		__ret = (uint32x2_t) __builtin_neon_vcvtm_u32_v((int8x8_t)__rev0, 18);
34904		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	35897	__ai __attribute__((target("i8mm"))) int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
	35898	int32x4_t __ret;
	35899	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35900	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	35901	int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	35902	__ret = (int32x4_t) __builtin_neon_vmmlaq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
	35903	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
34905	35904	return __ret;
34906	35905	}
34907	35906	#endif
34908	35907
34909	35908	#ifdef __LITTLE_ENDIAN__
34910		__ai int32x4_t vcvtnq_s32_f32(float32x4_t __p0) {
	35909	__ai __attribute__((target("i8mm"))) int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
34911	35910	int32x4_t __ret;
34912		__ret = (int32x4_t) __builtin_neon_vcvtnq_s32_v((int8x16_t)__p0, 34);
	35911	__ret = (int32x4_t) __builtin_neon_vusdotq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
34913	35912	return __ret;
34914	35913	}
34915	35914	#else
34916		__ai int32x4_t vcvtnq_s32_f32(float32x4_t __p0) {
	35915	__ai __attribute__((target("i8mm"))) int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
34917	35916	int32x4_t __ret;
34918		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34919		__ret = (int32x4_t) __builtin_neon_vcvtnq_s32_v((int8x16_t)__rev0, 34);
	35917	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	35918	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	35919	int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	35920	__ret = (int32x4_t) __builtin_neon_vusdotq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
34920	35921	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
34921	35922	return __ret;
34922	35923	}
	35924	__ai __attribute__((target("i8mm"))) int32x4_t __noswap_vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
	35925	int32x4_t __ret;
	35926	__ret = (int32x4_t) __builtin_neon_vusdotq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
	35927	return __ret;
	35928	}
34923	35929	#endif
34924	35930
34925	35931	#ifdef __LITTLE_ENDIAN__
34926		__ai int32x2_t vcvtn_s32_f32(float32x2_t __p0) {
	35932	__ai __attribute__((target("i8mm"))) int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
34927	35933	int32x2_t __ret;
34928		__ret = (int32x2_t) __builtin_neon_vcvtn_s32_v((int8x8_t)__p0, 2);
	35934	__ret = (int32x2_t) __builtin_neon_vusdot_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
34929	35935	return __ret;
34930	35936	}
34931	35937	#else
34932		__ai int32x2_t vcvtn_s32_f32(float32x2_t __p0) {
	35938	__ai __attribute__((target("i8mm"))) int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
34933	35939	int32x2_t __ret;
34934		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34935		__ret = (int32x2_t) __builtin_neon_vcvtn_s32_v((int8x8_t)__rev0, 2);
	35940	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	35941	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	35942	int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	35943	__ret = (int32x2_t) __builtin_neon_vusdot_s32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
34936	35944	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
34937	35945	return __ret;
34938	35946	}
	35947	__ai __attribute__((target("i8mm"))) int32x2_t __noswap_vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
	35948	int32x2_t __ret;
	35949	__ret = (int32x2_t) __builtin_neon_vusdot_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
	35950	return __ret;
	35951	}
34939	35952	#endif
34940	35953
34941	35954	#ifdef __LITTLE_ENDIAN__
34942		__ai uint32x4_t vcvtnq_u32_f32(float32x4_t __p0) {
34943		uint32x4_t __ret;
34944		__ret = (uint32x4_t) __builtin_neon_vcvtnq_u32_v((int8x16_t)__p0, 50);
	35955	#define vusdotq_lane_s32(__p0_157, __p1_157, __p2_157, __p3_157) __extension__ ({ \
	35956	int32x4_t __ret_157; \
	35957	int32x4_t __s0_157 = __p0_157; \
	35958	uint8x16_t __s1_157 = __p1_157; \
	35959	int8x8_t __s2_157 = __p2_157; \
	35960	int8x8_t __reint_157 = __s2_157; \
	35961	__ret_157 = vusdotq_s32(__s0_157, __s1_157, (int8x16_t)(splatq_lane_s32((int32x2_t ) &__reint_157, __p3_157))); \
	35962	__ret_157; \
	35963	})
	35964	#else
	35965	#define vusdotq_lane_s32(__p0_158, __p1_158, __p2_158, __p3_158) __extension__ ({ \
	35966	int32x4_t __ret_158; \
	35967	int32x4_t __s0_158 = __p0_158; \
	35968	uint8x16_t __s1_158 = __p1_158; \
	35969	int8x8_t __s2_158 = __p2_158; \
	35970	int32x4_t __rev0_158; __rev0_158 = __builtin_shufflevector(__s0_158, __s0_158, 3, 2, 1, 0); \
	35971	uint8x16_t __rev1_158; __rev1_158 = __builtin_shufflevector(__s1_158, __s1_158, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	35972	int8x8_t __rev2_158; __rev2_158 = __builtin_shufflevector(__s2_158, __s2_158, 7, 6, 5, 4, 3, 2, 1, 0); \
	35973	int8x8_t __reint_158 = __rev2_158; \
	35974	__ret_158 = __noswap_vusdotq_s32(__rev0_158, __rev1_158, (int8x16_t)(__noswap_splatq_lane_s32((int32x2_t ) &__reint_158, __p3_158))); \
	35975	__ret_158 = __builtin_shufflevector(__ret_158, __ret_158, 3, 2, 1, 0); \
	35976	__ret_158; \
	35977	})
	35978	#endif
	35979
	35980	#ifdef __LITTLE_ENDIAN__
	35981	#define vusdot_lane_s32(__p0_159, __p1_159, __p2_159, __p3_159) __extension__ ({ \
	35982	int32x2_t __ret_159; \
	35983	int32x2_t __s0_159 = __p0_159; \
	35984	uint8x8_t __s1_159 = __p1_159; \
	35985	int8x8_t __s2_159 = __p2_159; \
	35986	int8x8_t __reint_159 = __s2_159; \
	35987	__ret_159 = vusdot_s32(__s0_159, __s1_159, (int8x8_t)(splat_lane_s32((int32x2_t ) &__reint_159, __p3_159))); \
	35988	__ret_159; \
	35989	})
	35990	#else
	35991	#define vusdot_lane_s32(__p0_160, __p1_160, __p2_160, __p3_160) __extension__ ({ \
	35992	int32x2_t __ret_160; \
	35993	int32x2_t __s0_160 = __p0_160; \
	35994	uint8x8_t __s1_160 = __p1_160; \
	35995	int8x8_t __s2_160 = __p2_160; \
	35996	int32x2_t __rev0_160; __rev0_160 = __builtin_shufflevector(__s0_160, __s0_160, 1, 0); \
	35997	uint8x8_t __rev1_160; __rev1_160 = __builtin_shufflevector(__s1_160, __s1_160, 7, 6, 5, 4, 3, 2, 1, 0); \
	35998	int8x8_t __rev2_160; __rev2_160 = __builtin_shufflevector(__s2_160, __s2_160, 7, 6, 5, 4, 3, 2, 1, 0); \
	35999	int8x8_t __reint_160 = __rev2_160; \
	36000	__ret_160 = __noswap_vusdot_s32(__rev0_160, __rev1_160, (int8x8_t)(__noswap_splat_lane_s32((int32x2_t ) &__reint_160, __p3_160))); \
	36001	__ret_160 = __builtin_shufflevector(__ret_160, __ret_160, 1, 0); \
	36002	__ret_160; \
	36003	})
	36004	#endif
	36005
	36006	#ifdef __LITTLE_ENDIAN__
	36007	__ai __attribute__((target("i8mm"))) int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
	36008	int32x4_t __ret;
	36009	__ret = (int32x4_t) __builtin_neon_vusmmlaq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
34945	36010	return __ret;
34946	36011	}
34947	36012	#else
34948		__ai uint32x4_t vcvtnq_u32_f32(float32x4_t __p0) {
34949		uint32x4_t __ret;
34950		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34951		__ret = (uint32x4_t) __builtin_neon_vcvtnq_u32_v((int8x16_t)__rev0, 50);
	36013	__ai __attribute__((target("i8mm"))) int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
	36014	int32x4_t __ret;
	36015	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36016	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	36017	int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	36018	__ret = (int32x4_t) __builtin_neon_vusmmlaq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
34952	36019	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
34953	36020	return __ret;
34954	36021	}
34955	36022	#endif
34956	36023
34957	36024	#ifdef __LITTLE_ENDIAN__
34958		__ai uint32x2_t vcvtn_u32_f32(float32x2_t __p0) {
34959		uint32x2_t __ret;
34960		__ret = (uint32x2_t) __builtin_neon_vcvtn_u32_v((int8x8_t)__p0, 18);
	36025	__ai __attribute__((target("v8.1a"))) int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	36026	int32x4_t __ret;
	36027	__ret = (int32x4_t) __builtin_neon_vqrdmlahq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
34961	36028	return __ret;
34962	36029	}
34963	36030	#else
34964		__ai uint32x2_t vcvtn_u32_f32(float32x2_t __p0) {
34965		uint32x2_t __ret;
34966		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34967		__ret = (uint32x2_t) __builtin_neon_vcvtn_u32_v((int8x8_t)__rev0, 18);
34968		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	36031	__ai __attribute__((target("v8.1a"))) int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	36032	int32x4_t __ret;
	36033	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36034	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36035	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36036	__ret = (int32x4_t) __builtin_neon_vqrdmlahq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
	36037	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36038	return __ret;
	36039	}
	36040	__ai __attribute__((target("v8.1a"))) int32x4_t __noswap_vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	36041	int32x4_t __ret;
	36042	__ret = (int32x4_t) __builtin_neon_vqrdmlahq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
34969	36043	return __ret;
34970	36044	}
34971	36045	#endif
34972	36046
34973	36047	#ifdef __LITTLE_ENDIAN__
34974		__ai int32x4_t vcvtpq_s32_f32(float32x4_t __p0) {
34975		int32x4_t __ret;
34976		__ret = (int32x4_t) __builtin_neon_vcvtpq_s32_v((int8x16_t)__p0, 34);
	36048	__ai __attribute__((target("v8.1a"))) int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	36049	int16x8_t __ret;
	36050	__ret = (int16x8_t) __builtin_neon_vqrdmlahq_s16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
34977	36051	return __ret;
34978	36052	}
34979	36053	#else
34980		__ai int32x4_t vcvtpq_s32_f32(float32x4_t __p0) {
34981		int32x4_t __ret;
34982		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34983		__ret = (int32x4_t) __builtin_neon_vcvtpq_s32_v((int8x16_t)__rev0, 34);
34984		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36054	__ai __attribute__((target("v8.1a"))) int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	36055	int16x8_t __ret;
	36056	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	36057	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	36058	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	36059	__ret = (int16x8_t) __builtin_neon_vqrdmlahq_s16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
	36060	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	36061	return __ret;
	36062	}
	36063	__ai __attribute__((target("v8.1a"))) int16x8_t __noswap_vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	36064	int16x8_t __ret;
	36065	__ret = (int16x8_t) __builtin_neon_vqrdmlahq_s16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
34985	36066	return __ret;
34986	36067	}
34987	36068	#endif
34988	36069
34989	36070	#ifdef __LITTLE_ENDIAN__
34990		__ai int32x2_t vcvtp_s32_f32(float32x2_t __p0) {
	36071	__ai __attribute__((target("v8.1a"))) int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
34991	36072	int32x2_t __ret;
34992		__ret = (int32x2_t) __builtin_neon_vcvtp_s32_v((int8x8_t)__p0, 2);
	36073	__ret = (int32x2_t) __builtin_neon_vqrdmlah_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
34993	36074	return __ret;
34994	36075	}
34995	36076	#else
34996		__ai int32x2_t vcvtp_s32_f32(float32x2_t __p0) {
	36077	__ai __attribute__((target("v8.1a"))) int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
34997	36078	int32x2_t __ret;
34998		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34999		__ret = (int32x2_t) __builtin_neon_vcvtp_s32_v((int8x8_t)__rev0, 2);
	36079	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	36080	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36081	int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	36082	__ret = (int32x2_t) __builtin_neon_vqrdmlah_s32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
35000	36083	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35001	36084	return __ret;
35002	36085	}
	36086	__ai __attribute__((target("v8.1a"))) int32x2_t __noswap_vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
	36087	int32x2_t __ret;
	36088	__ret = (int32x2_t) __builtin_neon_vqrdmlah_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
	36089	return __ret;
	36090	}
35003	36091	#endif
35004	36092
35005	36093	#ifdef __LITTLE_ENDIAN__
35006		__ai uint32x4_t vcvtpq_u32_f32(float32x4_t __p0) {
35007		uint32x4_t __ret;
35008		__ret = (uint32x4_t) __builtin_neon_vcvtpq_u32_v((int8x16_t)__p0, 50);
	36094	__ai __attribute__((target("v8.1a"))) int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
	36095	int16x4_t __ret;
	36096	__ret = (int16x4_t) __builtin_neon_vqrdmlah_s16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
35009	36097	return __ret;
35010	36098	}
35011	36099	#else
35012		__ai uint32x4_t vcvtpq_u32_f32(float32x4_t __p0) {
35013		uint32x4_t __ret;
35014		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35015		__ret = (uint32x4_t) __builtin_neon_vcvtpq_u32_v((int8x16_t)__rev0, 50);
	36100	__ai __attribute__((target("v8.1a"))) int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
	36101	int16x4_t __ret;
	36102	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36103	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36104	int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36105	__ret = (int16x4_t) __builtin_neon_vqrdmlah_s16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
35016	36106	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35017	36107	return __ret;
35018	36108	}
	36109	__ai __attribute__((target("v8.1a"))) int16x4_t __noswap_vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
	36110	int16x4_t __ret;
	36111	__ret = (int16x4_t) __builtin_neon_vqrdmlah_s16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
	36112	return __ret;
	36113	}
35019	36114	#endif
35020	36115
35021	36116	#ifdef __LITTLE_ENDIAN__
35022		__ai uint32x2_t vcvtp_u32_f32(float32x2_t __p0) {
35023		uint32x2_t __ret;
35024		__ret = (uint32x2_t) __builtin_neon_vcvtp_u32_v((int8x8_t)__p0, 18);
35025		return __ret;
35026		}
	36117	#define vqrdmlahq_lane_s32(__p0_161, __p1_161, __p2_161, __p3_161) __extension__ ({ \
	36118	int32x4_t __ret_161; \
	36119	int32x4_t __s0_161 = __p0_161; \
	36120	int32x4_t __s1_161 = __p1_161; \
	36121	int32x2_t __s2_161 = __p2_161; \
	36122	__ret_161 = vqrdmlahq_s32(__s0_161, __s1_161, splatq_lane_s32(__s2_161, __p3_161)); \
	36123	__ret_161; \
	36124	})
35027	36125	#else
35028		__ai uint32x2_t vcvtp_u32_f32(float32x2_t __p0) {
35029		uint32x2_t __ret;
35030		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35031		__ret = (uint32x2_t) __builtin_neon_vcvtp_u32_v((int8x8_t)__rev0, 18);
35032		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35033		return __ret;
35034		}
	36126	#define vqrdmlahq_lane_s32(__p0_162, __p1_162, __p2_162, __p3_162) __extension__ ({ \
	36127	int32x4_t __ret_162; \
	36128	int32x4_t __s0_162 = __p0_162; \
	36129	int32x4_t __s1_162 = __p1_162; \
	36130	int32x2_t __s2_162 = __p2_162; \
	36131	int32x4_t __rev0_162; __rev0_162 = __builtin_shufflevector(__s0_162, __s0_162, 3, 2, 1, 0); \
	36132	int32x4_t __rev1_162; __rev1_162 = __builtin_shufflevector(__s1_162, __s1_162, 3, 2, 1, 0); \
	36133	int32x2_t __rev2_162; __rev2_162 = __builtin_shufflevector(__s2_162, __s2_162, 1, 0); \
	36134	__ret_162 = __noswap_vqrdmlahq_s32(__rev0_162, __rev1_162, __noswap_splatq_lane_s32(__rev2_162, __p3_162)); \
	36135	__ret_162 = __builtin_shufflevector(__ret_162, __ret_162, 3, 2, 1, 0); \
	36136	__ret_162; \
	36137	})
35035	36138	#endif
35036	36139
35037		#endif
35038		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_AES)
35039	36140	#ifdef __LITTLE_ENDIAN__
35040		__ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
35041		uint8x16_t __ret;
35042		__ret = (uint8x16_t) __builtin_neon_vaesdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
35043		return __ret;
35044		}
	36141	#define vqrdmlahq_lane_s16(__p0_163, __p1_163, __p2_163, __p3_163) __extension__ ({ \
	36142	int16x8_t __ret_163; \
	36143	int16x8_t __s0_163 = __p0_163; \
	36144	int16x8_t __s1_163 = __p1_163; \
	36145	int16x4_t __s2_163 = __p2_163; \
	36146	__ret_163 = vqrdmlahq_s16(__s0_163, __s1_163, splatq_lane_s16(__s2_163, __p3_163)); \
	36147	__ret_163; \
	36148	})
35045	36149	#else
35046		__ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
35047		uint8x16_t __ret;
35048		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35049		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35050		__ret = (uint8x16_t) __builtin_neon_vaesdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
35051		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35052		return __ret;
35053		}
	36150	#define vqrdmlahq_lane_s16(__p0_164, __p1_164, __p2_164, __p3_164) __extension__ ({ \
	36151	int16x8_t __ret_164; \
	36152	int16x8_t __s0_164 = __p0_164; \
	36153	int16x8_t __s1_164 = __p1_164; \
	36154	int16x4_t __s2_164 = __p2_164; \
	36155	int16x8_t __rev0_164; __rev0_164 = __builtin_shufflevector(__s0_164, __s0_164, 7, 6, 5, 4, 3, 2, 1, 0); \
	36156	int16x8_t __rev1_164; __rev1_164 = __builtin_shufflevector(__s1_164, __s1_164, 7, 6, 5, 4, 3, 2, 1, 0); \
	36157	int16x4_t __rev2_164; __rev2_164 = __builtin_shufflevector(__s2_164, __s2_164, 3, 2, 1, 0); \
	36158	__ret_164 = __noswap_vqrdmlahq_s16(__rev0_164, __rev1_164, __noswap_splatq_lane_s16(__rev2_164, __p3_164)); \
	36159	__ret_164 = __builtin_shufflevector(__ret_164, __ret_164, 7, 6, 5, 4, 3, 2, 1, 0); \
	36160	__ret_164; \
	36161	})
35054	36162	#endif
35055	36163
35056	36164	#ifdef __LITTLE_ENDIAN__
35057		__ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
35058		uint8x16_t __ret;
35059		__ret = (uint8x16_t) __builtin_neon_vaeseq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
35060		return __ret;
35061		}
	36165	#define vqrdmlah_lane_s32(__p0_165, __p1_165, __p2_165, __p3_165) __extension__ ({ \
	36166	int32x2_t __ret_165; \
	36167	int32x2_t __s0_165 = __p0_165; \
	36168	int32x2_t __s1_165 = __p1_165; \
	36169	int32x2_t __s2_165 = __p2_165; \
	36170	__ret_165 = vqrdmlah_s32(__s0_165, __s1_165, splat_lane_s32(__s2_165, __p3_165)); \
	36171	__ret_165; \
	36172	})
35062	36173	#else
35063		__ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
35064		uint8x16_t __ret;
35065		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35066		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35067		__ret = (uint8x16_t) __builtin_neon_vaeseq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
35068		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35069		return __ret;
35070		}
	36174	#define vqrdmlah_lane_s32(__p0_166, __p1_166, __p2_166, __p3_166) __extension__ ({ \
	36175	int32x2_t __ret_166; \
	36176	int32x2_t __s0_166 = __p0_166; \
	36177	int32x2_t __s1_166 = __p1_166; \
	36178	int32x2_t __s2_166 = __p2_166; \
	36179	int32x2_t __rev0_166; __rev0_166 = __builtin_shufflevector(__s0_166, __s0_166, 1, 0); \
	36180	int32x2_t __rev1_166; __rev1_166 = __builtin_shufflevector(__s1_166, __s1_166, 1, 0); \
	36181	int32x2_t __rev2_166; __rev2_166 = __builtin_shufflevector(__s2_166, __s2_166, 1, 0); \
	36182	__ret_166 = __noswap_vqrdmlah_s32(__rev0_166, __rev1_166, __noswap_splat_lane_s32(__rev2_166, __p3_166)); \
	36183	__ret_166 = __builtin_shufflevector(__ret_166, __ret_166, 1, 0); \
	36184	__ret_166; \
	36185	})
35071	36186	#endif
35072	36187
35073	36188	#ifdef __LITTLE_ENDIAN__
35074		__ai uint8x16_t vaesimcq_u8(uint8x16_t __p0) {
35075		uint8x16_t __ret;
35076		__ret = (uint8x16_t) __builtin_neon_vaesimcq_v((int8x16_t)__p0, 48);
35077		return __ret;
35078		}
	36189	#define vqrdmlah_lane_s16(__p0_167, __p1_167, __p2_167, __p3_167) __extension__ ({ \
	36190	int16x4_t __ret_167; \
	36191	int16x4_t __s0_167 = __p0_167; \
	36192	int16x4_t __s1_167 = __p1_167; \
	36193	int16x4_t __s2_167 = __p2_167; \
	36194	__ret_167 = vqrdmlah_s16(__s0_167, __s1_167, splat_lane_s16(__s2_167, __p3_167)); \
	36195	__ret_167; \
	36196	})
35079	36197	#else
35080		__ai uint8x16_t vaesimcq_u8(uint8x16_t __p0) {
35081		uint8x16_t __ret;
35082		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35083		__ret = (uint8x16_t) __builtin_neon_vaesimcq_v((int8x16_t)__rev0, 48);
35084		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35085		return __ret;
35086		}
	36198	#define vqrdmlah_lane_s16(__p0_168, __p1_168, __p2_168, __p3_168) __extension__ ({ \
	36199	int16x4_t __ret_168; \
	36200	int16x4_t __s0_168 = __p0_168; \
	36201	int16x4_t __s1_168 = __p1_168; \
	36202	int16x4_t __s2_168 = __p2_168; \
	36203	int16x4_t __rev0_168; __rev0_168 = __builtin_shufflevector(__s0_168, __s0_168, 3, 2, 1, 0); \
	36204	int16x4_t __rev1_168; __rev1_168 = __builtin_shufflevector(__s1_168, __s1_168, 3, 2, 1, 0); \
	36205	int16x4_t __rev2_168; __rev2_168 = __builtin_shufflevector(__s2_168, __s2_168, 3, 2, 1, 0); \
	36206	__ret_168 = __noswap_vqrdmlah_s16(__rev0_168, __rev1_168, __noswap_splat_lane_s16(__rev2_168, __p3_168)); \
	36207	__ret_168 = __builtin_shufflevector(__ret_168, __ret_168, 3, 2, 1, 0); \
	36208	__ret_168; \
	36209	})
35087	36210	#endif
35088	36211
35089	36212	#ifdef __LITTLE_ENDIAN__
35090		__ai uint8x16_t vaesmcq_u8(uint8x16_t __p0) {
35091		uint8x16_t __ret;
35092		__ret = (uint8x16_t) __builtin_neon_vaesmcq_v((int8x16_t)__p0, 48);
	36213	__ai __attribute__((target("v8.1a"))) int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	36214	int32x4_t __ret;
	36215	__ret = (int32x4_t) __builtin_neon_vqrdmlshq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
35093	36216	return __ret;
35094	36217	}
35095	36218	#else
35096		__ai uint8x16_t vaesmcq_u8(uint8x16_t __p0) {
35097		uint8x16_t __ret;
35098		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35099		__ret = (uint8x16_t) __builtin_neon_vaesmcq_v((int8x16_t)__rev0, 48);
35100		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	36219	__ai __attribute__((target("v8.1a"))) int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	36220	int32x4_t __ret;
	36221	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36222	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36223	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36224	__ret = (int32x4_t) __builtin_neon_vqrdmlshq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
	36225	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36226	return __ret;
	36227	}
	36228	__ai __attribute__((target("v8.1a"))) int32x4_t __noswap_vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	36229	int32x4_t __ret;
	36230	__ret = (int32x4_t) __builtin_neon_vqrdmlshq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
35101	36231	return __ret;
35102	36232	}
35103	36233	#endif
35104	36234
35105		#endif
35106		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_DIRECTED_ROUNDING)
35107	36235	#ifdef __LITTLE_ENDIAN__
35108		__ai float32x4_t vrndq_f32(float32x4_t __p0) {
35109		float32x4_t __ret;
35110		__ret = (float32x4_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 41);
	36236	__ai __attribute__((target("v8.1a"))) int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	36237	int16x8_t __ret;
	36238	__ret = (int16x8_t) __builtin_neon_vqrdmlshq_s16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
35111	36239	return __ret;
35112	36240	}
35113	36241	#else
35114		__ai float32x4_t vrndq_f32(float32x4_t __p0) {
35115		float32x4_t __ret;
35116		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35117		__ret = (float32x4_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 41);
35118		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36242	__ai __attribute__((target("v8.1a"))) int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	36243	int16x8_t __ret;
	36244	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	36245	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	36246	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	36247	__ret = (int16x8_t) __builtin_neon_vqrdmlshq_s16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
	36248	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	36249	return __ret;
	36250	}
	36251	__ai __attribute__((target("v8.1a"))) int16x8_t __noswap_vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	36252	int16x8_t __ret;
	36253	__ret = (int16x8_t) __builtin_neon_vqrdmlshq_s16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
35119	36254	return __ret;
35120	36255	}
35121	36256	#endif
35122	36257
35123	36258	#ifdef __LITTLE_ENDIAN__
35124		__ai float32x2_t vrnd_f32(float32x2_t __p0) {
35125		float32x2_t __ret;
35126		__ret = (float32x2_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 9);
	36259	__ai __attribute__((target("v8.1a"))) int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
	36260	int32x2_t __ret;
	36261	__ret = (int32x2_t) __builtin_neon_vqrdmlsh_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
35127	36262	return __ret;
35128	36263	}
35129	36264	#else
35130		__ai float32x2_t vrnd_f32(float32x2_t __p0) {
35131		float32x2_t __ret;
35132		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35133		__ret = (float32x2_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 9);
	36265	__ai __attribute__((target("v8.1a"))) int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
	36266	int32x2_t __ret;
	36267	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	36268	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36269	int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	36270	__ret = (int32x2_t) __builtin_neon_vqrdmlsh_s32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
35134	36271	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35135	36272	return __ret;
35136	36273	}
	36274	__ai __attribute__((target("v8.1a"))) int32x2_t __noswap_vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
	36275	int32x2_t __ret;
	36276	__ret = (int32x2_t) __builtin_neon_vqrdmlsh_s32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
	36277	return __ret;
	36278	}
35137	36279	#endif
35138	36280
35139	36281	#ifdef __LITTLE_ENDIAN__
35140		__ai float32x4_t vrndaq_f32(float32x4_t __p0) {
35141		float32x4_t __ret;
35142		__ret = (float32x4_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 41);
	36282	__ai __attribute__((target("v8.1a"))) int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
	36283	int16x4_t __ret;
	36284	__ret = (int16x4_t) __builtin_neon_vqrdmlsh_s16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
35143	36285	return __ret;
35144	36286	}
35145	36287	#else
35146		__ai float32x4_t vrndaq_f32(float32x4_t __p0) {
35147		float32x4_t __ret;
35148		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35149		__ret = (float32x4_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 41);
	36288	__ai __attribute__((target("v8.1a"))) int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
	36289	int16x4_t __ret;
	36290	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36291	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36292	int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36293	__ret = (int16x4_t) __builtin_neon_vqrdmlsh_s16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
35150	36294	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35151	36295	return __ret;
35152	36296	}
	36297	__ai __attribute__((target("v8.1a"))) int16x4_t __noswap_vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
	36298	int16x4_t __ret;
	36299	__ret = (int16x4_t) __builtin_neon_vqrdmlsh_s16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
	36300	return __ret;
	36301	}
35153	36302	#endif
35154	36303
35155	36304	#ifdef __LITTLE_ENDIAN__
35156		__ai float32x2_t vrnda_f32(float32x2_t __p0) {
35157		float32x2_t __ret;
35158		__ret = (float32x2_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 9);
35159		return __ret;
35160		}
	36305	#define vqrdmlshq_lane_s32(__p0_169, __p1_169, __p2_169, __p3_169) __extension__ ({ \
	36306	int32x4_t __ret_169; \
	36307	int32x4_t __s0_169 = __p0_169; \
	36308	int32x4_t __s1_169 = __p1_169; \
	36309	int32x2_t __s2_169 = __p2_169; \
	36310	__ret_169 = vqrdmlshq_s32(__s0_169, __s1_169, splatq_lane_s32(__s2_169, __p3_169)); \
	36311	__ret_169; \
	36312	})
35161	36313	#else
35162		__ai float32x2_t vrnda_f32(float32x2_t __p0) {
35163		float32x2_t __ret;
35164		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35165		__ret = (float32x2_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 9);
35166		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35167		return __ret;
35168		}
	36314	#define vqrdmlshq_lane_s32(__p0_170, __p1_170, __p2_170, __p3_170) __extension__ ({ \
	36315	int32x4_t __ret_170; \
	36316	int32x4_t __s0_170 = __p0_170; \
	36317	int32x4_t __s1_170 = __p1_170; \
	36318	int32x2_t __s2_170 = __p2_170; \
	36319	int32x4_t __rev0_170; __rev0_170 = __builtin_shufflevector(__s0_170, __s0_170, 3, 2, 1, 0); \
	36320	int32x4_t __rev1_170; __rev1_170 = __builtin_shufflevector(__s1_170, __s1_170, 3, 2, 1, 0); \
	36321	int32x2_t __rev2_170; __rev2_170 = __builtin_shufflevector(__s2_170, __s2_170, 1, 0); \
	36322	__ret_170 = __noswap_vqrdmlshq_s32(__rev0_170, __rev1_170, __noswap_splatq_lane_s32(__rev2_170, __p3_170)); \
	36323	__ret_170 = __builtin_shufflevector(__ret_170, __ret_170, 3, 2, 1, 0); \
	36324	__ret_170; \
	36325	})
35169	36326	#endif
35170	36327
35171	36328	#ifdef __LITTLE_ENDIAN__
35172		__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
35173		float32x4_t __ret;
35174		__ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 41);
35175		return __ret;
35176		}
	36329	#define vqrdmlshq_lane_s16(__p0_171, __p1_171, __p2_171, __p3_171) __extension__ ({ \
	36330	int16x8_t __ret_171; \
	36331	int16x8_t __s0_171 = __p0_171; \
	36332	int16x8_t __s1_171 = __p1_171; \
	36333	int16x4_t __s2_171 = __p2_171; \
	36334	__ret_171 = vqrdmlshq_s16(__s0_171, __s1_171, splatq_lane_s16(__s2_171, __p3_171)); \
	36335	__ret_171; \
	36336	})
35177	36337	#else
35178		__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
35179		float32x4_t __ret;
35180		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35181		__ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 41);
35182		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35183		return __ret;
35184		}
	36338	#define vqrdmlshq_lane_s16(__p0_172, __p1_172, __p2_172, __p3_172) __extension__ ({ \
	36339	int16x8_t __ret_172; \
	36340	int16x8_t __s0_172 = __p0_172; \
	36341	int16x8_t __s1_172 = __p1_172; \
	36342	int16x4_t __s2_172 = __p2_172; \
	36343	int16x8_t __rev0_172; __rev0_172 = __builtin_shufflevector(__s0_172, __s0_172, 7, 6, 5, 4, 3, 2, 1, 0); \
	36344	int16x8_t __rev1_172; __rev1_172 = __builtin_shufflevector(__s1_172, __s1_172, 7, 6, 5, 4, 3, 2, 1, 0); \
	36345	int16x4_t __rev2_172; __rev2_172 = __builtin_shufflevector(__s2_172, __s2_172, 3, 2, 1, 0); \
	36346	__ret_172 = __noswap_vqrdmlshq_s16(__rev0_172, __rev1_172, __noswap_splatq_lane_s16(__rev2_172, __p3_172)); \
	36347	__ret_172 = __builtin_shufflevector(__ret_172, __ret_172, 7, 6, 5, 4, 3, 2, 1, 0); \
	36348	__ret_172; \
	36349	})
35185	36350	#endif
35186	36351
35187	36352	#ifdef __LITTLE_ENDIAN__
35188		__ai float32x2_t vrndi_f32(float32x2_t __p0) {
35189		float32x2_t __ret;
35190		__ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 9);
35191		return __ret;
35192		}
	36353	#define vqrdmlsh_lane_s32(__p0_173, __p1_173, __p2_173, __p3_173) __extension__ ({ \
	36354	int32x2_t __ret_173; \
	36355	int32x2_t __s0_173 = __p0_173; \
	36356	int32x2_t __s1_173 = __p1_173; \
	36357	int32x2_t __s2_173 = __p2_173; \
	36358	__ret_173 = vqrdmlsh_s32(__s0_173, __s1_173, splat_lane_s32(__s2_173, __p3_173)); \
	36359	__ret_173; \
	36360	})
35193	36361	#else
35194		__ai float32x2_t vrndi_f32(float32x2_t __p0) {
35195		float32x2_t __ret;
35196		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35197		__ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 9);
35198		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35199		return __ret;
35200		}
	36362	#define vqrdmlsh_lane_s32(__p0_174, __p1_174, __p2_174, __p3_174) __extension__ ({ \
	36363	int32x2_t __ret_174; \
	36364	int32x2_t __s0_174 = __p0_174; \
	36365	int32x2_t __s1_174 = __p1_174; \
	36366	int32x2_t __s2_174 = __p2_174; \
	36367	int32x2_t __rev0_174; __rev0_174 = __builtin_shufflevector(__s0_174, __s0_174, 1, 0); \
	36368	int32x2_t __rev1_174; __rev1_174 = __builtin_shufflevector(__s1_174, __s1_174, 1, 0); \
	36369	int32x2_t __rev2_174; __rev2_174 = __builtin_shufflevector(__s2_174, __s2_174, 1, 0); \
	36370	__ret_174 = __noswap_vqrdmlsh_s32(__rev0_174, __rev1_174, __noswap_splat_lane_s32(__rev2_174, __p3_174)); \
	36371	__ret_174 = __builtin_shufflevector(__ret_174, __ret_174, 1, 0); \
	36372	__ret_174; \
	36373	})
35201	36374	#endif
35202	36375
35203	36376	#ifdef __LITTLE_ENDIAN__
35204		__ai float32x4_t vrndmq_f32(float32x4_t __p0) {
35205		float32x4_t __ret;
35206		__ret = (float32x4_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 41);
35207		return __ret;
35208		}
	36377	#define vqrdmlsh_lane_s16(__p0_175, __p1_175, __p2_175, __p3_175) __extension__ ({ \
	36378	int16x4_t __ret_175; \
	36379	int16x4_t __s0_175 = __p0_175; \
	36380	int16x4_t __s1_175 = __p1_175; \
	36381	int16x4_t __s2_175 = __p2_175; \
	36382	__ret_175 = vqrdmlsh_s16(__s0_175, __s1_175, splat_lane_s16(__s2_175, __p3_175)); \
	36383	__ret_175; \
	36384	})
35209	36385	#else
35210		__ai float32x4_t vrndmq_f32(float32x4_t __p0) {
35211		float32x4_t __ret;
35212		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35213		__ret = (float32x4_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 41);
35214		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35215		return __ret;
35216		}
	36386	#define vqrdmlsh_lane_s16(__p0_176, __p1_176, __p2_176, __p3_176) __extension__ ({ \
	36387	int16x4_t __ret_176; \
	36388	int16x4_t __s0_176 = __p0_176; \
	36389	int16x4_t __s1_176 = __p1_176; \
	36390	int16x4_t __s2_176 = __p2_176; \
	36391	int16x4_t __rev0_176; __rev0_176 = __builtin_shufflevector(__s0_176, __s0_176, 3, 2, 1, 0); \
	36392	int16x4_t __rev1_176; __rev1_176 = __builtin_shufflevector(__s1_176, __s1_176, 3, 2, 1, 0); \
	36393	int16x4_t __rev2_176; __rev2_176 = __builtin_shufflevector(__s2_176, __s2_176, 3, 2, 1, 0); \
	36394	__ret_176 = __noswap_vqrdmlsh_s16(__rev0_176, __rev1_176, __noswap_splat_lane_s16(__rev2_176, __p3_176)); \
	36395	__ret_176 = __builtin_shufflevector(__ret_176, __ret_176, 3, 2, 1, 0); \
	36396	__ret_176; \
	36397	})
35217	36398	#endif
35218	36399
35219	36400	#ifdef __LITTLE_ENDIAN__
35220		__ai float32x2_t vrndm_f32(float32x2_t __p0) {
	36401	__ai __attribute__((target("v8.3a"))) float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
35221	36402	float32x2_t __ret;
35222		__ret = (float32x2_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 9);
	36403	__ret = (float32x2_t) __builtin_neon_vcadd_rot270_f32((int8x8_t)__p0, (int8x8_t)__p1, 9);
35223	36404	return __ret;
35224	36405	}
35225	36406	#else
35226		__ai float32x2_t vrndm_f32(float32x2_t __p0) {
	36407	__ai __attribute__((target("v8.3a"))) float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
35227	36408	float32x2_t __ret;
35228	36409	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35229		__ret = (float32x2_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 9);
	36410	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36411	__ret = (float32x2_t) __builtin_neon_vcadd_rot270_f32((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
35230	36412	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35231	36413	return __ret;
35232	36414	}
35233	36415	#endif
35234	36416
35235	36417	#ifdef __LITTLE_ENDIAN__
35236		__ai float32x4_t vrndnq_f32(float32x4_t __p0) {
35237		float32x4_t __ret;
35238		__ret = (float32x4_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 41);
35239		return __ret;
35240		}
35241		#else
35242		__ai float32x4_t vrndnq_f32(float32x4_t __p0) {
35243		float32x4_t __ret;
35244		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35245		__ret = (float32x4_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 41);
35246		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35247		return __ret;
35248		}
35249		#endif
35250
35251		#ifdef __LITTLE_ENDIAN__
35252		__ai float32x2_t vrndn_f32(float32x2_t __p0) {
	36418	__ai __attribute__((target("v8.3a"))) float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
35253	36419	float32x2_t __ret;
35254		__ret = (float32x2_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 9);
	36420	__ret = (float32x2_t) __builtin_neon_vcadd_rot90_f32((int8x8_t)__p0, (int8x8_t)__p1, 9);
35255	36421	return __ret;
35256	36422	}
35257	36423	#else
35258		__ai float32x2_t vrndn_f32(float32x2_t __p0) {
	36424	__ai __attribute__((target("v8.3a"))) float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
35259	36425	float32x2_t __ret;
35260	36426	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35261		__ret = (float32x2_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 9);
	36427	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36428	__ret = (float32x2_t) __builtin_neon_vcadd_rot90_f32((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
35262	36429	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35263	36430	return __ret;
35264	36431	}
35265	36432	#endif
35266	36433
35267		__ai float32_t vrndns_f32(float32_t __p0) {
35268		float32_t __ret;
35269		__ret = (float32_t) __builtin_neon_vrndns_f32(__p0);
35270		return __ret;
35271		}
35272	36434	#ifdef __LITTLE_ENDIAN__
35273		__ai float32x4_t vrndpq_f32(float32x4_t __p0) {
	36435	__ai __attribute__((target("v8.3a"))) float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
35274	36436	float32x4_t __ret;
35275		__ret = (float32x4_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 41);
	36437	__ret = (float32x4_t) __builtin_neon_vcaddq_rot270_f32((int8x16_t)__p0, (int8x16_t)__p1, 41);
35276	36438	return __ret;
35277	36439	}
35278	36440	#else
35279		__ai float32x4_t vrndpq_f32(float32x4_t __p0) {
	36441	__ai __attribute__((target("v8.3a"))) float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
35280	36442	float32x4_t __ret;
35281	36443	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35282		__ret = (float32x4_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 41);
	36444	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36445	__ret = (float32x4_t) __builtin_neon_vcaddq_rot270_f32((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
35283	36446	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35284	36447	return __ret;
35285	36448	}
35286	36449	#endif
35287	36450
35288	36451	#ifdef __LITTLE_ENDIAN__
35289		__ai float32x2_t vrndp_f32(float32x2_t __p0) {
35290		float32x2_t __ret;
35291		__ret = (float32x2_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 9);
	36452	__ai __attribute__((target("v8.3a"))) float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
	36453	float32x4_t __ret;
	36454	__ret = (float32x4_t) __builtin_neon_vcaddq_rot90_f32((int8x16_t)__p0, (int8x16_t)__p1, 41);
35292	36455	return __ret;
35293	36456	}
35294	36457	#else
35295		__ai float32x2_t vrndp_f32(float32x2_t __p0) {
35296		float32x2_t __ret;
35297		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35298		__ret = (float32x2_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 9);
35299		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	36458	__ai __attribute__((target("v8.3a"))) float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
	36459	float32x4_t __ret;
	36460	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36461	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36462	__ret = (float32x4_t) __builtin_neon_vcaddq_rot90_f32((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	36463	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35300	36464	return __ret;
35301	36465	}
35302	36466	#endif
35303	36467
35304	36468	#ifdef __LITTLE_ENDIAN__
35305		__ai float32x4_t vrndxq_f32(float32x4_t __p0) {
	36469	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
35306	36470	float32x4_t __ret;
35307		__ret = (float32x4_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 41);
	36471	__ret = (float32x4_t) __builtin_neon_vcmlaq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
35308	36472	return __ret;
35309	36473	}
35310	36474	#else
35311		__ai float32x4_t vrndxq_f32(float32x4_t __p0) {
	36475	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
35312	36476	float32x4_t __ret;
35313	36477	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35314		__ret = (float32x4_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 41);
	36478	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36479	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36480	__ret = (float32x4_t) __builtin_neon_vcmlaq_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
35315	36481	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35316	36482	return __ret;
35317	36483	}
	36484	__ai __attribute__((target("v8.3a"))) float32x4_t __noswap_vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36485	float32x4_t __ret;
	36486	__ret = (float32x4_t) __builtin_neon_vcmlaq_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	36487	return __ret;
	36488	}
35318	36489	#endif
35319	36490
35320	36491	#ifdef __LITTLE_ENDIAN__
35321		__ai float32x2_t vrndx_f32(float32x2_t __p0) {
	36492	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
35322	36493	float32x2_t __ret;
35323		__ret = (float32x2_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 9);
	36494	__ret = (float32x2_t) __builtin_neon_vcmla_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35324	36495	return __ret;
35325	36496	}
35326	36497	#else
35327		__ai float32x2_t vrndx_f32(float32x2_t __p0) {
	36498	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
35328	36499	float32x2_t __ret;
35329	36500	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35330		__ret = (float32x2_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 9);
	36501	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36502	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	36503	__ret = (float32x2_t) __builtin_neon_vcmla_f32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
35331	36504	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35332	36505	return __ret;
35333	36506	}
35334		#endif
35335
35336		#endif
35337		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_DIRECTED_ROUNDING) && defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
35338		#ifdef __LITTLE_ENDIAN__
35339		__ai float16x8_t vrndq_f16(float16x8_t __p0) {
35340		float16x8_t __ret;
35341		__ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 40);
35342		return __ret;
35343		}
35344		#else
35345		__ai float16x8_t vrndq_f16(float16x8_t __p0) {
35346		float16x8_t __ret;
35347		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35348		__ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 40);
35349		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	36507	__ai __attribute__((target("v8.3a"))) float32x2_t __noswap_vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36508	float32x2_t __ret;
	36509	__ret = (float32x2_t) __builtin_neon_vcmla_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35350	36510	return __ret;
35351	36511	}
35352	36512	#endif
35353	36513
35354	36514	#ifdef __LITTLE_ENDIAN__
35355		__ai float16x4_t vrnd_f16(float16x4_t __p0) {
35356		float16x4_t __ret;
35357		__ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 8);
35358		return __ret;
35359		}
	36515	#define vcmla_lane_f32(__p0_177, __p1_177, __p2_177, __p3_177) __extension__ ({ \
	36516	float32x2_t __ret_177; \
	36517	float32x2_t __s0_177 = __p0_177; \
	36518	float32x2_t __s1_177 = __p1_177; \
	36519	float32x2_t __s2_177 = __p2_177; \
	36520	float32x2_t __reint_177 = __s2_177; \
	36521	uint64x1_t __reint1_177 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_177, __p3_177)}; \
	36522	__ret_177 = vcmla_f32(__s0_177, __s1_177, (float32x2_t ) &__reint1_177); \
	36523	__ret_177; \
	36524	})
35360	36525	#else
35361		__ai float16x4_t vrnd_f16(float16x4_t __p0) {
35362		float16x4_t __ret;
35363		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35364		__ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 8);
35365		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35366		return __ret;
35367		}
	36526	#define vcmla_lane_f32(__p0_178, __p1_178, __p2_178, __p3_178) __extension__ ({ \
	36527	float32x2_t __ret_178; \
	36528	float32x2_t __s0_178 = __p0_178; \
	36529	float32x2_t __s1_178 = __p1_178; \
	36530	float32x2_t __s2_178 = __p2_178; \
	36531	float32x2_t __rev0_178; __rev0_178 = __builtin_shufflevector(__s0_178, __s0_178, 1, 0); \
	36532	float32x2_t __rev1_178; __rev1_178 = __builtin_shufflevector(__s1_178, __s1_178, 1, 0); \
	36533	float32x2_t __rev2_178; __rev2_178 = __builtin_shufflevector(__s2_178, __s2_178, 1, 0); \
	36534	float32x2_t __reint_178 = __rev2_178; \
	36535	uint64x1_t __reint1_178 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_178, __p3_178)}; \
	36536	__ret_178 = __noswap_vcmla_f32(__rev0_178, __rev1_178, (float32x2_t ) &__reint1_178); \
	36537	__ret_178 = __builtin_shufflevector(__ret_178, __ret_178, 1, 0); \
	36538	__ret_178; \
	36539	})
35368	36540	#endif
35369	36541
35370	36542	#ifdef __LITTLE_ENDIAN__
35371		__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
35372		float16x8_t __ret;
35373		__ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 40);
35374		return __ret;
35375		}
	36543	#define vcmlaq_lane_f32(__p0_179, __p1_179, __p2_179, __p3_179) __extension__ ({ \
	36544	float32x4_t __ret_179; \
	36545	float32x4_t __s0_179 = __p0_179; \
	36546	float32x4_t __s1_179 = __p1_179; \
	36547	float32x2_t __s2_179 = __p2_179; \
	36548	float32x2_t __reint_179 = __s2_179; \
	36549	uint64x2_t __reint1_179 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_179, __p3_179), vget_lane_u64((uint64x1_t ) &__reint_179, __p3_179)}; \
	36550	__ret_179 = vcmlaq_f32(__s0_179, __s1_179, (float32x4_t ) &__reint1_179); \
	36551	__ret_179; \
	36552	})
35376	36553	#else
35377		__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
35378		float16x8_t __ret;
35379		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35380		__ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 40);
35381		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35382		return __ret;
35383		}
	36554	#define vcmlaq_lane_f32(__p0_180, __p1_180, __p2_180, __p3_180) __extension__ ({ \
	36555	float32x4_t __ret_180; \
	36556	float32x4_t __s0_180 = __p0_180; \
	36557	float32x4_t __s1_180 = __p1_180; \
	36558	float32x2_t __s2_180 = __p2_180; \
	36559	float32x4_t __rev0_180; __rev0_180 = __builtin_shufflevector(__s0_180, __s0_180, 3, 2, 1, 0); \
	36560	float32x4_t __rev1_180; __rev1_180 = __builtin_shufflevector(__s1_180, __s1_180, 3, 2, 1, 0); \
	36561	float32x2_t __rev2_180; __rev2_180 = __builtin_shufflevector(__s2_180, __s2_180, 1, 0); \
	36562	float32x2_t __reint_180 = __rev2_180; \
	36563	uint64x2_t __reint1_180 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_180, __p3_180), vget_lane_u64((uint64x1_t ) &__reint_180, __p3_180)}; \
	36564	__ret_180 = __noswap_vcmlaq_f32(__rev0_180, __rev1_180, (float32x4_t ) &__reint1_180); \
	36565	__ret_180 = __builtin_shufflevector(__ret_180, __ret_180, 3, 2, 1, 0); \
	36566	__ret_180; \
	36567	})
35384	36568	#endif
35385	36569
35386	36570	#ifdef __LITTLE_ENDIAN__
35387		__ai float16x4_t vrnda_f16(float16x4_t __p0) {
35388		float16x4_t __ret;
35389		__ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 8);
35390		return __ret;
35391		}
	36571	#define vcmla_laneq_f32(__p0_181, __p1_181, __p2_181, __p3_181) __extension__ ({ \
	36572	float32x2_t __ret_181; \
	36573	float32x2_t __s0_181 = __p0_181; \
	36574	float32x2_t __s1_181 = __p1_181; \
	36575	float32x4_t __s2_181 = __p2_181; \
	36576	float32x4_t __reint_181 = __s2_181; \
	36577	uint64x1_t __reint1_181 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_181, __p3_181)}; \
	36578	__ret_181 = vcmla_f32(__s0_181, __s1_181, (float32x2_t ) &__reint1_181); \
	36579	__ret_181; \
	36580	})
35392	36581	#else
35393		__ai float16x4_t vrnda_f16(float16x4_t __p0) {
35394		float16x4_t __ret;
35395		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35396		__ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 8);
35397		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35398		return __ret;
35399		}
	36582	#define vcmla_laneq_f32(__p0_182, __p1_182, __p2_182, __p3_182) __extension__ ({ \
	36583	float32x2_t __ret_182; \
	36584	float32x2_t __s0_182 = __p0_182; \
	36585	float32x2_t __s1_182 = __p1_182; \
	36586	float32x4_t __s2_182 = __p2_182; \
	36587	float32x2_t __rev0_182; __rev0_182 = __builtin_shufflevector(__s0_182, __s0_182, 1, 0); \
	36588	float32x2_t __rev1_182; __rev1_182 = __builtin_shufflevector(__s1_182, __s1_182, 1, 0); \
	36589	float32x4_t __rev2_182; __rev2_182 = __builtin_shufflevector(__s2_182, __s2_182, 3, 2, 1, 0); \
	36590	float32x4_t __reint_182 = __rev2_182; \
	36591	uint64x1_t __reint1_182 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_182, __p3_182)}; \
	36592	__ret_182 = __noswap_vcmla_f32(__rev0_182, __rev1_182, (float32x2_t ) &__reint1_182); \
	36593	__ret_182 = __builtin_shufflevector(__ret_182, __ret_182, 1, 0); \
	36594	__ret_182; \
	36595	})
35400	36596	#endif
35401	36597
35402	36598	#ifdef __LITTLE_ENDIAN__
35403		__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
35404		float16x8_t __ret;
35405		__ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 40);
35406		return __ret;
35407		}
	36599	#define vcmlaq_laneq_f32(__p0_183, __p1_183, __p2_183, __p3_183) __extension__ ({ \
	36600	float32x4_t __ret_183; \
	36601	float32x4_t __s0_183 = __p0_183; \
	36602	float32x4_t __s1_183 = __p1_183; \
	36603	float32x4_t __s2_183 = __p2_183; \
	36604	float32x4_t __reint_183 = __s2_183; \
	36605	uint64x2_t __reint1_183 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_183, __p3_183), vgetq_lane_u64((uint64x2_t ) &__reint_183, __p3_183)}; \
	36606	__ret_183 = vcmlaq_f32(__s0_183, __s1_183, (float32x4_t ) &__reint1_183); \
	36607	__ret_183; \
	36608	})
35408	36609	#else
35409		__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
35410		float16x8_t __ret;
35411		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35412		__ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 40);
35413		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35414		return __ret;
35415		}
	36610	#define vcmlaq_laneq_f32(__p0_184, __p1_184, __p2_184, __p3_184) __extension__ ({ \
	36611	float32x4_t __ret_184; \
	36612	float32x4_t __s0_184 = __p0_184; \
	36613	float32x4_t __s1_184 = __p1_184; \
	36614	float32x4_t __s2_184 = __p2_184; \
	36615	float32x4_t __rev0_184; __rev0_184 = __builtin_shufflevector(__s0_184, __s0_184, 3, 2, 1, 0); \
	36616	float32x4_t __rev1_184; __rev1_184 = __builtin_shufflevector(__s1_184, __s1_184, 3, 2, 1, 0); \
	36617	float32x4_t __rev2_184; __rev2_184 = __builtin_shufflevector(__s2_184, __s2_184, 3, 2, 1, 0); \
	36618	float32x4_t __reint_184 = __rev2_184; \
	36619	uint64x2_t __reint1_184 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_184, __p3_184), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_184, __p3_184)}; \
	36620	__ret_184 = __noswap_vcmlaq_f32(__rev0_184, __rev1_184, (float32x4_t ) &__reint1_184); \
	36621	__ret_184 = __builtin_shufflevector(__ret_184, __ret_184, 3, 2, 1, 0); \
	36622	__ret_184; \
	36623	})
35416	36624	#endif
35417	36625
35418	36626	#ifdef __LITTLE_ENDIAN__
35419		__ai float16x4_t vrndm_f16(float16x4_t __p0) {
35420		float16x4_t __ret;
35421		__ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 8);
	36627	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36628	float32x4_t __ret;
	36629	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
35422	36630	return __ret;
35423	36631	}
35424	36632	#else
35425		__ai float16x4_t vrndm_f16(float16x4_t __p0) {
35426		float16x4_t __ret;
35427		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35428		__ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 8);
	36633	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36634	float32x4_t __ret;
	36635	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36636	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36637	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36638	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
35429	36639	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35430	36640	return __ret;
35431	36641	}
35432		#endif
35433
35434		#ifdef __LITTLE_ENDIAN__
35435		__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
35436		float16x8_t __ret;
35437		__ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 40);
35438		return __ret;
35439		}
35440		#else
35441		__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
35442		float16x8_t __ret;
35443		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35444		__ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 40);
35445		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	36642	__ai __attribute__((target("v8.3a"))) float32x4_t __noswap_vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36643	float32x4_t __ret;
	36644	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
35446	36645	return __ret;
35447	36646	}
35448	36647	#endif
35449	36648
35450	36649	#ifdef __LITTLE_ENDIAN__
35451		__ai float16x4_t vrndn_f16(float16x4_t __p0) {
35452		float16x4_t __ret;
35453		__ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 8);
	36650	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36651	float32x2_t __ret;
	36652	__ret = (float32x2_t) __builtin_neon_vcmla_rot180_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35454	36653	return __ret;
35455	36654	}
35456	36655	#else
35457		__ai float16x4_t vrndn_f16(float16x4_t __p0) {
35458		float16x4_t __ret;
35459		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35460		__ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 8);
35461		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36656	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36657	float32x2_t __ret;
	36658	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	36659	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36660	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	36661	__ret = (float32x2_t) __builtin_neon_vcmla_rot180_f32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	36662	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	36663	return __ret;
	36664	}
	36665	__ai __attribute__((target("v8.3a"))) float32x2_t __noswap_vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36666	float32x2_t __ret;
	36667	__ret = (float32x2_t) __builtin_neon_vcmla_rot180_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35462	36668	return __ret;
35463	36669	}
35464	36670	#endif
35465	36671
35466	36672	#ifdef __LITTLE_ENDIAN__
35467		__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
35468		float16x8_t __ret;
35469		__ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 40);
35470		return __ret;
35471		}
	36673	#define vcmla_rot180_lane_f32(__p0_185, __p1_185, __p2_185, __p3_185) __extension__ ({ \
	36674	float32x2_t __ret_185; \
	36675	float32x2_t __s0_185 = __p0_185; \
	36676	float32x2_t __s1_185 = __p1_185; \
	36677	float32x2_t __s2_185 = __p2_185; \
	36678	float32x2_t __reint_185 = __s2_185; \
	36679	uint64x1_t __reint1_185 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_185, __p3_185)}; \
	36680	__ret_185 = vcmla_rot180_f32(__s0_185, __s1_185, (float32x2_t ) &__reint1_185); \
	36681	__ret_185; \
	36682	})
35472	36683	#else
35473		__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
35474		float16x8_t __ret;
35475		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35476		__ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 40);
35477		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35478		return __ret;
35479		}
	36684	#define vcmla_rot180_lane_f32(__p0_186, __p1_186, __p2_186, __p3_186) __extension__ ({ \
	36685	float32x2_t __ret_186; \
	36686	float32x2_t __s0_186 = __p0_186; \
	36687	float32x2_t __s1_186 = __p1_186; \
	36688	float32x2_t __s2_186 = __p2_186; \
	36689	float32x2_t __rev0_186; __rev0_186 = __builtin_shufflevector(__s0_186, __s0_186, 1, 0); \
	36690	float32x2_t __rev1_186; __rev1_186 = __builtin_shufflevector(__s1_186, __s1_186, 1, 0); \
	36691	float32x2_t __rev2_186; __rev2_186 = __builtin_shufflevector(__s2_186, __s2_186, 1, 0); \
	36692	float32x2_t __reint_186 = __rev2_186; \
	36693	uint64x1_t __reint1_186 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_186, __p3_186)}; \
	36694	__ret_186 = __noswap_vcmla_rot180_f32(__rev0_186, __rev1_186, (float32x2_t ) &__reint1_186); \
	36695	__ret_186 = __builtin_shufflevector(__ret_186, __ret_186, 1, 0); \
	36696	__ret_186; \
	36697	})
35480	36698	#endif
35481	36699
35482	36700	#ifdef __LITTLE_ENDIAN__
35483		__ai float16x4_t vrndp_f16(float16x4_t __p0) {
35484		float16x4_t __ret;
35485		__ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 8);
35486		return __ret;
35487		}
	36701	#define vcmlaq_rot180_lane_f32(__p0_187, __p1_187, __p2_187, __p3_187) __extension__ ({ \
	36702	float32x4_t __ret_187; \
	36703	float32x4_t __s0_187 = __p0_187; \
	36704	float32x4_t __s1_187 = __p1_187; \
	36705	float32x2_t __s2_187 = __p2_187; \
	36706	float32x2_t __reint_187 = __s2_187; \
	36707	uint64x2_t __reint1_187 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_187, __p3_187), vget_lane_u64((uint64x1_t ) &__reint_187, __p3_187)}; \
	36708	__ret_187 = vcmlaq_rot180_f32(__s0_187, __s1_187, (float32x4_t ) &__reint1_187); \
	36709	__ret_187; \
	36710	})
35488	36711	#else
35489		__ai float16x4_t vrndp_f16(float16x4_t __p0) {
35490		float16x4_t __ret;
35491		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35492		__ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 8);
35493		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35494		return __ret;
35495		}
	36712	#define vcmlaq_rot180_lane_f32(__p0_188, __p1_188, __p2_188, __p3_188) __extension__ ({ \
	36713	float32x4_t __ret_188; \
	36714	float32x4_t __s0_188 = __p0_188; \
	36715	float32x4_t __s1_188 = __p1_188; \
	36716	float32x2_t __s2_188 = __p2_188; \
	36717	float32x4_t __rev0_188; __rev0_188 = __builtin_shufflevector(__s0_188, __s0_188, 3, 2, 1, 0); \
	36718	float32x4_t __rev1_188; __rev1_188 = __builtin_shufflevector(__s1_188, __s1_188, 3, 2, 1, 0); \
	36719	float32x2_t __rev2_188; __rev2_188 = __builtin_shufflevector(__s2_188, __s2_188, 1, 0); \
	36720	float32x2_t __reint_188 = __rev2_188; \
	36721	uint64x2_t __reint1_188 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_188, __p3_188), vget_lane_u64((uint64x1_t ) &__reint_188, __p3_188)}; \
	36722	__ret_188 = __noswap_vcmlaq_rot180_f32(__rev0_188, __rev1_188, (float32x4_t ) &__reint1_188); \
	36723	__ret_188 = __builtin_shufflevector(__ret_188, __ret_188, 3, 2, 1, 0); \
	36724	__ret_188; \
	36725	})
35496	36726	#endif
35497	36727
35498	36728	#ifdef __LITTLE_ENDIAN__
35499		__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
35500		float16x8_t __ret;
35501		__ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 40);
35502		return __ret;
35503		}
	36729	#define vcmla_rot180_laneq_f32(__p0_189, __p1_189, __p2_189, __p3_189) __extension__ ({ \
	36730	float32x2_t __ret_189; \
	36731	float32x2_t __s0_189 = __p0_189; \
	36732	float32x2_t __s1_189 = __p1_189; \
	36733	float32x4_t __s2_189 = __p2_189; \
	36734	float32x4_t __reint_189 = __s2_189; \
	36735	uint64x1_t __reint1_189 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_189, __p3_189)}; \
	36736	__ret_189 = vcmla_rot180_f32(__s0_189, __s1_189, (float32x2_t ) &__reint1_189); \
	36737	__ret_189; \
	36738	})
35504	36739	#else
35505		__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
35506		float16x8_t __ret;
35507		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35508		__ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 40);
35509		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35510		return __ret;
35511		}
	36740	#define vcmla_rot180_laneq_f32(__p0_190, __p1_190, __p2_190, __p3_190) __extension__ ({ \
	36741	float32x2_t __ret_190; \
	36742	float32x2_t __s0_190 = __p0_190; \
	36743	float32x2_t __s1_190 = __p1_190; \
	36744	float32x4_t __s2_190 = __p2_190; \
	36745	float32x2_t __rev0_190; __rev0_190 = __builtin_shufflevector(__s0_190, __s0_190, 1, 0); \
	36746	float32x2_t __rev1_190; __rev1_190 = __builtin_shufflevector(__s1_190, __s1_190, 1, 0); \
	36747	float32x4_t __rev2_190; __rev2_190 = __builtin_shufflevector(__s2_190, __s2_190, 3, 2, 1, 0); \
	36748	float32x4_t __reint_190 = __rev2_190; \
	36749	uint64x1_t __reint1_190 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_190, __p3_190)}; \
	36750	__ret_190 = __noswap_vcmla_rot180_f32(__rev0_190, __rev1_190, (float32x2_t ) &__reint1_190); \
	36751	__ret_190 = __builtin_shufflevector(__ret_190, __ret_190, 1, 0); \
	36752	__ret_190; \
	36753	})
35512	36754	#endif
35513	36755
35514	36756	#ifdef __LITTLE_ENDIAN__
35515		__ai float16x4_t vrndx_f16(float16x4_t __p0) {
35516		float16x4_t __ret;
35517		__ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 8);
35518		return __ret;
35519		}
	36757	#define vcmlaq_rot180_laneq_f32(__p0_191, __p1_191, __p2_191, __p3_191) __extension__ ({ \
	36758	float32x4_t __ret_191; \
	36759	float32x4_t __s0_191 = __p0_191; \
	36760	float32x4_t __s1_191 = __p1_191; \
	36761	float32x4_t __s2_191 = __p2_191; \
	36762	float32x4_t __reint_191 = __s2_191; \
	36763	uint64x2_t __reint1_191 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_191, __p3_191), vgetq_lane_u64((uint64x2_t ) &__reint_191, __p3_191)}; \
	36764	__ret_191 = vcmlaq_rot180_f32(__s0_191, __s1_191, (float32x4_t ) &__reint1_191); \
	36765	__ret_191; \
	36766	})
35520	36767	#else
35521		__ai float16x4_t vrndx_f16(float16x4_t __p0) {
35522		float16x4_t __ret;
35523		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35524		__ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 8);
35525		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35526		return __ret;
35527		}
	36768	#define vcmlaq_rot180_laneq_f32(__p0_192, __p1_192, __p2_192, __p3_192) __extension__ ({ \
	36769	float32x4_t __ret_192; \
	36770	float32x4_t __s0_192 = __p0_192; \
	36771	float32x4_t __s1_192 = __p1_192; \
	36772	float32x4_t __s2_192 = __p2_192; \
	36773	float32x4_t __rev0_192; __rev0_192 = __builtin_shufflevector(__s0_192, __s0_192, 3, 2, 1, 0); \
	36774	float32x4_t __rev1_192; __rev1_192 = __builtin_shufflevector(__s1_192, __s1_192, 3, 2, 1, 0); \
	36775	float32x4_t __rev2_192; __rev2_192 = __builtin_shufflevector(__s2_192, __s2_192, 3, 2, 1, 0); \
	36776	float32x4_t __reint_192 = __rev2_192; \
	36777	uint64x2_t __reint1_192 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_192, __p3_192), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_192, __p3_192)}; \
	36778	__ret_192 = __noswap_vcmlaq_rot180_f32(__rev0_192, __rev1_192, (float32x4_t ) &__reint1_192); \
	36779	__ret_192 = __builtin_shufflevector(__ret_192, __ret_192, 3, 2, 1, 0); \
	36780	__ret_192; \
	36781	})
35528	36782	#endif
35529	36783
35530		#endif
35531		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
35532	36784	#ifdef __LITTLE_ENDIAN__
35533		__ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	36785	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
35534	36786	float32x4_t __ret;
35535		__ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	36787	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
35536	36788	return __ret;
35537	36789	}
35538	36790	#else
35539		__ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	36791	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
35540	36792	float32x4_t __ret;
35541	36793	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35542	36794	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35543		__ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	36795	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36796	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
35544	36797	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35545	36798	return __ret;
35546	36799	}
	36800	__ai __attribute__((target("v8.3a"))) float32x4_t __noswap_vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36801	float32x4_t __ret;
	36802	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	36803	return __ret;
	36804	}
35547	36805	#endif
35548	36806
35549	36807	#ifdef __LITTLE_ENDIAN__
35550		__ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
	36808	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
35551	36809	float32x2_t __ret;
35552		__ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	36810	__ret = (float32x2_t) __builtin_neon_vcmla_rot270_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35553	36811	return __ret;
35554	36812	}
35555	36813	#else
35556		__ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
	36814	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
35557	36815	float32x2_t __ret;
35558	36816	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35559	36817	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
35560		__ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
	36818	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	36819	__ret = (float32x2_t) __builtin_neon_vcmla_rot270_f32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
35561	36820	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35562	36821	return __ret;
35563	36822	}
35564		#endif
35565
35566		#ifdef __LITTLE_ENDIAN__
35567		__ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
35568		float32x4_t __ret;
35569		__ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
35570		return __ret;
35571		}
35572		#else
35573		__ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
35574		float32x4_t __ret;
35575		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35576		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35577		__ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
35578		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36823	__ai __attribute__((target("v8.3a"))) float32x2_t __noswap_vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36824	float32x2_t __ret;
	36825	__ret = (float32x2_t) __builtin_neon_vcmla_rot270_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35579	36826	return __ret;
35580	36827	}
35581	36828	#endif
35582	36829
35583	36830	#ifdef __LITTLE_ENDIAN__
35584		__ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
35585		float32x2_t __ret;
35586		__ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
35587		return __ret;
35588		}
	36831	#define vcmla_rot270_lane_f32(__p0_193, __p1_193, __p2_193, __p3_193) __extension__ ({ \
	36832	float32x2_t __ret_193; \
	36833	float32x2_t __s0_193 = __p0_193; \
	36834	float32x2_t __s1_193 = __p1_193; \
	36835	float32x2_t __s2_193 = __p2_193; \
	36836	float32x2_t __reint_193 = __s2_193; \
	36837	uint64x1_t __reint1_193 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_193, __p3_193)}; \
	36838	__ret_193 = vcmla_rot270_f32(__s0_193, __s1_193, (float32x2_t ) &__reint1_193); \
	36839	__ret_193; \
	36840	})
35589	36841	#else
35590		__ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
35591		float32x2_t __ret;
35592		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35593		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
35594		__ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
35595		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35596		return __ret;
35597		}
	36842	#define vcmla_rot270_lane_f32(__p0_194, __p1_194, __p2_194, __p3_194) __extension__ ({ \
	36843	float32x2_t __ret_194; \
	36844	float32x2_t __s0_194 = __p0_194; \
	36845	float32x2_t __s1_194 = __p1_194; \
	36846	float32x2_t __s2_194 = __p2_194; \
	36847	float32x2_t __rev0_194; __rev0_194 = __builtin_shufflevector(__s0_194, __s0_194, 1, 0); \
	36848	float32x2_t __rev1_194; __rev1_194 = __builtin_shufflevector(__s1_194, __s1_194, 1, 0); \
	36849	float32x2_t __rev2_194; __rev2_194 = __builtin_shufflevector(__s2_194, __s2_194, 1, 0); \
	36850	float32x2_t __reint_194 = __rev2_194; \
	36851	uint64x1_t __reint1_194 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_194, __p3_194)}; \
	36852	__ret_194 = __noswap_vcmla_rot270_f32(__rev0_194, __rev1_194, (float32x2_t ) &__reint1_194); \
	36853	__ret_194 = __builtin_shufflevector(__ret_194, __ret_194, 1, 0); \
	36854	__ret_194; \
	36855	})
35598	36856	#endif
35599	36857
35600		#endif
35601		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_NUMERIC_MAXMIN) && defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
35602	36858	#ifdef __LITTLE_ENDIAN__
35603		__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
35604		float16x8_t __ret;
35605		__ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
35606		return __ret;
35607		}
	36859	#define vcmlaq_rot270_lane_f32(__p0_195, __p1_195, __p2_195, __p3_195) __extension__ ({ \
	36860	float32x4_t __ret_195; \
	36861	float32x4_t __s0_195 = __p0_195; \
	36862	float32x4_t __s1_195 = __p1_195; \
	36863	float32x2_t __s2_195 = __p2_195; \
	36864	float32x2_t __reint_195 = __s2_195; \
	36865	uint64x2_t __reint1_195 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_195, __p3_195), vget_lane_u64((uint64x1_t ) &__reint_195, __p3_195)}; \
	36866	__ret_195 = vcmlaq_rot270_f32(__s0_195, __s1_195, (float32x4_t ) &__reint1_195); \
	36867	__ret_195; \
	36868	})
35608	36869	#else
35609		__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
35610		float16x8_t __ret;
35611		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35612		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
35613		__ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
35614		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35615		return __ret;
35616		}
	36870	#define vcmlaq_rot270_lane_f32(__p0_196, __p1_196, __p2_196, __p3_196) __extension__ ({ \
	36871	float32x4_t __ret_196; \
	36872	float32x4_t __s0_196 = __p0_196; \
	36873	float32x4_t __s1_196 = __p1_196; \
	36874	float32x2_t __s2_196 = __p2_196; \
	36875	float32x4_t __rev0_196; __rev0_196 = __builtin_shufflevector(__s0_196, __s0_196, 3, 2, 1, 0); \
	36876	float32x4_t __rev1_196; __rev1_196 = __builtin_shufflevector(__s1_196, __s1_196, 3, 2, 1, 0); \
	36877	float32x2_t __rev2_196; __rev2_196 = __builtin_shufflevector(__s2_196, __s2_196, 1, 0); \
	36878	float32x2_t __reint_196 = __rev2_196; \
	36879	uint64x2_t __reint1_196 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_196, __p3_196), vget_lane_u64((uint64x1_t ) &__reint_196, __p3_196)}; \
	36880	__ret_196 = __noswap_vcmlaq_rot270_f32(__rev0_196, __rev1_196, (float32x4_t ) &__reint1_196); \
	36881	__ret_196 = __builtin_shufflevector(__ret_196, __ret_196, 3, 2, 1, 0); \
	36882	__ret_196; \
	36883	})
35617	36884	#endif
35618	36885
35619	36886	#ifdef __LITTLE_ENDIAN__
35620		__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
35621		float16x4_t __ret;
35622		__ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
35623		return __ret;
35624		}
	36887	#define vcmla_rot270_laneq_f32(__p0_197, __p1_197, __p2_197, __p3_197) __extension__ ({ \
	36888	float32x2_t __ret_197; \
	36889	float32x2_t __s0_197 = __p0_197; \
	36890	float32x2_t __s1_197 = __p1_197; \
	36891	float32x4_t __s2_197 = __p2_197; \
	36892	float32x4_t __reint_197 = __s2_197; \
	36893	uint64x1_t __reint1_197 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_197, __p3_197)}; \
	36894	__ret_197 = vcmla_rot270_f32(__s0_197, __s1_197, (float32x2_t ) &__reint1_197); \
	36895	__ret_197; \
	36896	})
35625	36897	#else
35626		__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
35627		float16x4_t __ret;
35628		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35629		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35630		__ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
35631		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35632		return __ret;
35633		}
	36898	#define vcmla_rot270_laneq_f32(__p0_198, __p1_198, __p2_198, __p3_198) __extension__ ({ \
	36899	float32x2_t __ret_198; \
	36900	float32x2_t __s0_198 = __p0_198; \
	36901	float32x2_t __s1_198 = __p1_198; \
	36902	float32x4_t __s2_198 = __p2_198; \
	36903	float32x2_t __rev0_198; __rev0_198 = __builtin_shufflevector(__s0_198, __s0_198, 1, 0); \
	36904	float32x2_t __rev1_198; __rev1_198 = __builtin_shufflevector(__s1_198, __s1_198, 1, 0); \
	36905	float32x4_t __rev2_198; __rev2_198 = __builtin_shufflevector(__s2_198, __s2_198, 3, 2, 1, 0); \
	36906	float32x4_t __reint_198 = __rev2_198; \
	36907	uint64x1_t __reint1_198 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_198, __p3_198)}; \
	36908	__ret_198 = __noswap_vcmla_rot270_f32(__rev0_198, __rev1_198, (float32x2_t ) &__reint1_198); \
	36909	__ret_198 = __builtin_shufflevector(__ret_198, __ret_198, 1, 0); \
	36910	__ret_198; \
	36911	})
35634	36912	#endif
35635	36913
35636	36914	#ifdef __LITTLE_ENDIAN__
35637		__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
35638		float16x8_t __ret;
35639		__ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
35640		return __ret;
35641		}
	36915	#define vcmlaq_rot270_laneq_f32(__p0_199, __p1_199, __p2_199, __p3_199) __extension__ ({ \
	36916	float32x4_t __ret_199; \
	36917	float32x4_t __s0_199 = __p0_199; \
	36918	float32x4_t __s1_199 = __p1_199; \
	36919	float32x4_t __s2_199 = __p2_199; \
	36920	float32x4_t __reint_199 = __s2_199; \
	36921	uint64x2_t __reint1_199 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_199, __p3_199), vgetq_lane_u64((uint64x2_t ) &__reint_199, __p3_199)}; \
	36922	__ret_199 = vcmlaq_rot270_f32(__s0_199, __s1_199, (float32x4_t ) &__reint1_199); \
	36923	__ret_199; \
	36924	})
35642	36925	#else
35643		__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
35644		float16x8_t __ret;
35645		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35646		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
35647		__ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
35648		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35649		return __ret;
35650		}
	36926	#define vcmlaq_rot270_laneq_f32(__p0_200, __p1_200, __p2_200, __p3_200) __extension__ ({ \
	36927	float32x4_t __ret_200; \
	36928	float32x4_t __s0_200 = __p0_200; \
	36929	float32x4_t __s1_200 = __p1_200; \
	36930	float32x4_t __s2_200 = __p2_200; \
	36931	float32x4_t __rev0_200; __rev0_200 = __builtin_shufflevector(__s0_200, __s0_200, 3, 2, 1, 0); \
	36932	float32x4_t __rev1_200; __rev1_200 = __builtin_shufflevector(__s1_200, __s1_200, 3, 2, 1, 0); \
	36933	float32x4_t __rev2_200; __rev2_200 = __builtin_shufflevector(__s2_200, __s2_200, 3, 2, 1, 0); \
	36934	float32x4_t __reint_200 = __rev2_200; \
	36935	uint64x2_t __reint1_200 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_200, __p3_200), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_200, __p3_200)}; \
	36936	__ret_200 = __noswap_vcmlaq_rot270_f32(__rev0_200, __rev1_200, (float32x4_t ) &__reint1_200); \
	36937	__ret_200 = __builtin_shufflevector(__ret_200, __ret_200, 3, 2, 1, 0); \
	36938	__ret_200; \
	36939	})
35651	36940	#endif
35652	36941
35653	36942	#ifdef __LITTLE_ENDIAN__
35654		__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
35655		float16x4_t __ret;
35656		__ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	36943	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36944	float32x4_t __ret;
	36945	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
35657	36946	return __ret;
35658	36947	}
35659	36948	#else
35660		__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
35661		float16x4_t __ret;
35662		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35663		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35664		__ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	36949	__ai __attribute__((target("v8.3a"))) float32x4_t vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36950	float32x4_t __ret;
	36951	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	36952	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	36953	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	36954	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_f32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
35665	36955	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35666	36956	return __ret;
35667	36957	}
	36958	__ai __attribute__((target("v8.3a"))) float32x4_t __noswap_vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	36959	float32x4_t __ret;
	36960	__ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_f32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	36961	return __ret;
	36962	}
35668	36963	#endif
35669	36964
35670		#endif
35671		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_SHA2)
35672	36965	#ifdef __LITTLE_ENDIAN__
35673		__ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35674		uint32x4_t __ret;
35675		__ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__p0, __p1, __p2);
	36966	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36967	float32x2_t __ret;
	36968	__ret = (float32x2_t) __builtin_neon_vcmla_rot90_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35676	36969	return __ret;
35677	36970	}
35678	36971	#else
35679		__ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35680		uint32x4_t __ret;
35681		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35682		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35683		__ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__rev0, __p1, __rev2);
35684		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	36972	__ai __attribute__((target("v8.3a"))) float32x2_t vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36973	float32x2_t __ret;
	36974	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	36975	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	36976	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	36977	__ret = (float32x2_t) __builtin_neon_vcmla_rot90_f32((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	36978	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35685	36979	return __ret;
35686	36980	}
35687		#endif
35688
35689		__ai uint32_t vsha1h_u32(uint32_t __p0) {
35690		uint32_t __ret;
35691		__ret = (uint32_t) __builtin_neon_vsha1h_u32(__p0);
	36981	__ai __attribute__((target("v8.3a"))) float32x2_t __noswap_vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	36982	float32x2_t __ret;
	36983	__ret = (float32x2_t) __builtin_neon_vcmla_rot90_f32((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
35692	36984	return __ret;
35693	36985	}
	36986	#endif
	36987
35694	36988	#ifdef __LITTLE_ENDIAN__
35695		__ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35696		uint32x4_t __ret;
35697		__ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__p0, __p1, __p2);
35698		return __ret;
35699		}
	36989	#define vcmla_rot90_lane_f32(__p0_201, __p1_201, __p2_201, __p3_201) __extension__ ({ \
	36990	float32x2_t __ret_201; \
	36991	float32x2_t __s0_201 = __p0_201; \
	36992	float32x2_t __s1_201 = __p1_201; \
	36993	float32x2_t __s2_201 = __p2_201; \
	36994	float32x2_t __reint_201 = __s2_201; \
	36995	uint64x1_t __reint1_201 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_201, __p3_201)}; \
	36996	__ret_201 = vcmla_rot90_f32(__s0_201, __s1_201, (float32x2_t ) &__reint1_201); \
	36997	__ret_201; \
	36998	})
35700	36999	#else
35701		__ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35702		uint32x4_t __ret;
35703		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35704		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35705		__ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__rev0, __p1, __rev2);
35706		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35707		return __ret;
35708		}
	37000	#define vcmla_rot90_lane_f32(__p0_202, __p1_202, __p2_202, __p3_202) __extension__ ({ \
	37001	float32x2_t __ret_202; \
	37002	float32x2_t __s0_202 = __p0_202; \
	37003	float32x2_t __s1_202 = __p1_202; \
	37004	float32x2_t __s2_202 = __p2_202; \
	37005	float32x2_t __rev0_202; __rev0_202 = __builtin_shufflevector(__s0_202, __s0_202, 1, 0); \
	37006	float32x2_t __rev1_202; __rev1_202 = __builtin_shufflevector(__s1_202, __s1_202, 1, 0); \
	37007	float32x2_t __rev2_202; __rev2_202 = __builtin_shufflevector(__s2_202, __s2_202, 1, 0); \
	37008	float32x2_t __reint_202 = __rev2_202; \
	37009	uint64x1_t __reint1_202 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_202, __p3_202)}; \
	37010	__ret_202 = __noswap_vcmla_rot90_f32(__rev0_202, __rev1_202, (float32x2_t ) &__reint1_202); \
	37011	__ret_202 = __builtin_shufflevector(__ret_202, __ret_202, 1, 0); \
	37012	__ret_202; \
	37013	})
35709	37014	#endif
35710	37015
35711	37016	#ifdef __LITTLE_ENDIAN__
35712		__ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35713		uint32x4_t __ret;
35714		__ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__p0, __p1, __p2);
35715		return __ret;
35716		}
	37017	#define vcmlaq_rot90_lane_f32(__p0_203, __p1_203, __p2_203, __p3_203) __extension__ ({ \
	37018	float32x4_t __ret_203; \
	37019	float32x4_t __s0_203 = __p0_203; \
	37020	float32x4_t __s1_203 = __p1_203; \
	37021	float32x2_t __s2_203 = __p2_203; \
	37022	float32x2_t __reint_203 = __s2_203; \
	37023	uint64x2_t __reint1_203 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_203, __p3_203), vget_lane_u64((uint64x1_t ) &__reint_203, __p3_203)}; \
	37024	__ret_203 = vcmlaq_rot90_f32(__s0_203, __s1_203, (float32x4_t ) &__reint1_203); \
	37025	__ret_203; \
	37026	})
35717	37027	#else
35718		__ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35719		uint32x4_t __ret;
35720		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35721		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35722		__ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__rev0, __p1, __rev2);
35723		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35724		return __ret;
35725		}
	37028	#define vcmlaq_rot90_lane_f32(__p0_204, __p1_204, __p2_204, __p3_204) __extension__ ({ \
	37029	float32x4_t __ret_204; \
	37030	float32x4_t __s0_204 = __p0_204; \
	37031	float32x4_t __s1_204 = __p1_204; \
	37032	float32x2_t __s2_204 = __p2_204; \
	37033	float32x4_t __rev0_204; __rev0_204 = __builtin_shufflevector(__s0_204, __s0_204, 3, 2, 1, 0); \
	37034	float32x4_t __rev1_204; __rev1_204 = __builtin_shufflevector(__s1_204, __s1_204, 3, 2, 1, 0); \
	37035	float32x2_t __rev2_204; __rev2_204 = __builtin_shufflevector(__s2_204, __s2_204, 1, 0); \
	37036	float32x2_t __reint_204 = __rev2_204; \
	37037	uint64x2_t __reint1_204 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_204, __p3_204), vget_lane_u64((uint64x1_t ) &__reint_204, __p3_204)}; \
	37038	__ret_204 = __noswap_vcmlaq_rot90_f32(__rev0_204, __rev1_204, (float32x4_t ) &__reint1_204); \
	37039	__ret_204 = __builtin_shufflevector(__ret_204, __ret_204, 3, 2, 1, 0); \
	37040	__ret_204; \
	37041	})
35726	37042	#endif
35727	37043
35728	37044	#ifdef __LITTLE_ENDIAN__
35729		__ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35730		uint32x4_t __ret;
35731		__ret = (uint32x4_t) __builtin_neon_vsha1su0q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
35732		return __ret;
35733		}
	37045	#define vcmla_rot90_laneq_f32(__p0_205, __p1_205, __p2_205, __p3_205) __extension__ ({ \
	37046	float32x2_t __ret_205; \
	37047	float32x2_t __s0_205 = __p0_205; \
	37048	float32x2_t __s1_205 = __p1_205; \
	37049	float32x4_t __s2_205 = __p2_205; \
	37050	float32x4_t __reint_205 = __s2_205; \
	37051	uint64x1_t __reint1_205 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_205, __p3_205)}; \
	37052	__ret_205 = vcmla_rot90_f32(__s0_205, __s1_205, (float32x2_t ) &__reint1_205); \
	37053	__ret_205; \
	37054	})
35734	37055	#else
35735		__ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35736		uint32x4_t __ret;
35737		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35738		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35739		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35740		__ret = (uint32x4_t) __builtin_neon_vsha1su0q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
35741		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35742		return __ret;
35743		}
	37056	#define vcmla_rot90_laneq_f32(__p0_206, __p1_206, __p2_206, __p3_206) __extension__ ({ \
	37057	float32x2_t __ret_206; \
	37058	float32x2_t __s0_206 = __p0_206; \
	37059	float32x2_t __s1_206 = __p1_206; \
	37060	float32x4_t __s2_206 = __p2_206; \
	37061	float32x2_t __rev0_206; __rev0_206 = __builtin_shufflevector(__s0_206, __s0_206, 1, 0); \
	37062	float32x2_t __rev1_206; __rev1_206 = __builtin_shufflevector(__s1_206, __s1_206, 1, 0); \
	37063	float32x4_t __rev2_206; __rev2_206 = __builtin_shufflevector(__s2_206, __s2_206, 3, 2, 1, 0); \
	37064	float32x4_t __reint_206 = __rev2_206; \
	37065	uint64x1_t __reint1_206 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_206, __p3_206)}; \
	37066	__ret_206 = __noswap_vcmla_rot90_f32(__rev0_206, __rev1_206, (float32x2_t ) &__reint1_206); \
	37067	__ret_206 = __builtin_shufflevector(__ret_206, __ret_206, 1, 0); \
	37068	__ret_206; \
	37069	})
35744	37070	#endif
35745	37071
35746	37072	#ifdef __LITTLE_ENDIAN__
35747		__ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
35748		uint32x4_t __ret;
35749		__ret = (uint32x4_t) __builtin_neon_vsha1su1q_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
35750		return __ret;
35751		}
	37073	#define vcmlaq_rot90_laneq_f32(__p0_207, __p1_207, __p2_207, __p3_207) __extension__ ({ \
	37074	float32x4_t __ret_207; \
	37075	float32x4_t __s0_207 = __p0_207; \
	37076	float32x4_t __s1_207 = __p1_207; \
	37077	float32x4_t __s2_207 = __p2_207; \
	37078	float32x4_t __reint_207 = __s2_207; \
	37079	uint64x2_t __reint1_207 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_207, __p3_207), vgetq_lane_u64((uint64x2_t ) &__reint_207, __p3_207)}; \
	37080	__ret_207 = vcmlaq_rot90_f32(__s0_207, __s1_207, (float32x4_t ) &__reint1_207); \
	37081	__ret_207; \
	37082	})
35752	37083	#else
35753		__ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
35754		uint32x4_t __ret;
35755		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35756		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35757		__ret = (uint32x4_t) __builtin_neon_vsha1su1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
35758		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35759		return __ret;
35760		}
	37084	#define vcmlaq_rot90_laneq_f32(__p0_208, __p1_208, __p2_208, __p3_208) __extension__ ({ \
	37085	float32x4_t __ret_208; \
	37086	float32x4_t __s0_208 = __p0_208; \
	37087	float32x4_t __s1_208 = __p1_208; \
	37088	float32x4_t __s2_208 = __p2_208; \
	37089	float32x4_t __rev0_208; __rev0_208 = __builtin_shufflevector(__s0_208, __s0_208, 3, 2, 1, 0); \
	37090	float32x4_t __rev1_208; __rev1_208 = __builtin_shufflevector(__s1_208, __s1_208, 3, 2, 1, 0); \
	37091	float32x4_t __rev2_208; __rev2_208 = __builtin_shufflevector(__s2_208, __s2_208, 3, 2, 1, 0); \
	37092	float32x4_t __reint_208 = __rev2_208; \
	37093	uint64x2_t __reint1_208 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_208, __p3_208), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_208, __p3_208)}; \
	37094	__ret_208 = __noswap_vcmlaq_rot90_f32(__rev0_208, __rev1_208, (float32x4_t ) &__reint1_208); \
	37095	__ret_208 = __builtin_shufflevector(__ret_208, __ret_208, 3, 2, 1, 0); \
	37096	__ret_208; \
	37097	})
35761	37098	#endif
35762	37099
35763	37100	#ifdef __LITTLE_ENDIAN__
35764		__ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35765		uint32x4_t __ret;
35766		__ret = (uint32x4_t) __builtin_neon_vsha256hq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37101	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
	37102	float16x4_t __ret;
	37103	__ret = (float16x4_t) __builtin_neon_vcadd_rot270_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
35767	37104	return __ret;
35768	37105	}
35769	37106	#else
35770		__ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35771		uint32x4_t __ret;
35772		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35773		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35774		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35775		__ret = (uint32x4_t) __builtin_neon_vsha256hq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	37107	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
	37108	float16x4_t __ret;
	37109	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	37110	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	37111	__ret = (float16x4_t) __builtin_neon_vcadd_rot270_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
35776	37112	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35777	37113	return __ret;
35778	37114	}
35779	37115	#endif
35780	37116
35781	37117	#ifdef __LITTLE_ENDIAN__
35782		__ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35783		uint32x4_t __ret;
35784		__ret = (uint32x4_t) __builtin_neon_vsha256h2q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37118	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
	37119	float16x4_t __ret;
	37120	__ret = (float16x4_t) __builtin_neon_vcadd_rot90_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
35785	37121	return __ret;
35786	37122	}
35787	37123	#else
35788		__ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35789		uint32x4_t __ret;
35790		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35791		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35792		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35793		__ret = (uint32x4_t) __builtin_neon_vsha256h2q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	37124	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
	37125	float16x4_t __ret;
	37126	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	37127	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	37128	__ret = (float16x4_t) __builtin_neon_vcadd_rot90_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
35794	37129	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35795	37130	return __ret;
35796	37131	}
35797	37132	#endif
35798	37133
35799	37134	#ifdef __LITTLE_ENDIAN__
35800		__ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
35801		uint32x4_t __ret;
35802		__ret = (uint32x4_t) __builtin_neon_vsha256su0q_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	37135	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
	37136	float16x8_t __ret;
	37137	__ret = (float16x8_t) __builtin_neon_vcaddq_rot270_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
35803	37138	return __ret;
35804	37139	}
35805	37140	#else
35806		__ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
35807		uint32x4_t __ret;
35808		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35809		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35810		__ret = (uint32x4_t) __builtin_neon_vsha256su0q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
35811		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	37141	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
	37142	float16x8_t __ret;
	37143	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	37144	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	37145	__ret = (float16x8_t) __builtin_neon_vcaddq_rot270_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	37146	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35812	37147	return __ret;
35813	37148	}
35814	37149	#endif
35815	37150
35816	37151	#ifdef __LITTLE_ENDIAN__
35817		__ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35818		uint32x4_t __ret;
35819		__ret = (uint32x4_t) __builtin_neon_vsha256su1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37152	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
	37153	float16x8_t __ret;
	37154	__ret = (float16x8_t) __builtin_neon_vcaddq_rot90_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
35820	37155	return __ret;
35821	37156	}
35822	37157	#else
35823		__ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35824		uint32x4_t __ret;
35825		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35826		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35827		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35828		__ret = (uint32x4_t) __builtin_neon_vsha256su1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
35829		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	37158	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
	37159	float16x8_t __ret;
	37160	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	37161	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	37162	__ret = (float16x8_t) __builtin_neon_vcaddq_rot90_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	37163	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35830	37164	return __ret;
35831	37165	}
35832	37166	#endif
35833	37167
35834		#endif
35835		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_SHA3) && defined(__aarch64__)
35836	37168	#ifdef __LITTLE_ENDIAN__
35837		__ai uint8x16_t vbcaxq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
35838		uint8x16_t __ret;
35839		__ret = (uint8x16_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
	37169	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37170	float16x8_t __ret;
	37171	__ret = (float16x8_t) __builtin_neon_vcmlaq_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
35840	37172	return __ret;
35841	37173	}
35842	37174	#else
35843		__ai uint8x16_t vbcaxq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
35844		uint8x16_t __ret;
35845		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35846		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35847		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35848		__ret = (uint8x16_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
35849		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	37175	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37176	float16x8_t __ret;
	37177	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	37178	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	37179	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	37180	__ret = (float16x8_t) __builtin_neon_vcmlaq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
	37181	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	37182	return __ret;
	37183	}
	37184	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t __noswap_vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37185	float16x8_t __ret;
	37186	__ret = (float16x8_t) __builtin_neon_vcmlaq_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
35850	37187	return __ret;
35851	37188	}
35852	37189	#endif
35853	37190
35854	37191	#ifdef __LITTLE_ENDIAN__
35855		__ai uint32x4_t vbcaxq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35856		uint32x4_t __ret;
35857		__ret = (uint32x4_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37192	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37193	float16x4_t __ret;
	37194	__ret = (float16x4_t) __builtin_neon_vcmla_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
35858	37195	return __ret;
35859	37196	}
35860	37197	#else
35861		__ai uint32x4_t vbcaxq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35862		uint32x4_t __ret;
35863		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35864		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35865		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35866		__ret = (uint32x4_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	37198	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37199	float16x4_t __ret;
	37200	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	37201	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	37202	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	37203	__ret = (float16x4_t) __builtin_neon_vcmla_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
35867	37204	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35868	37205	return __ret;
35869	37206	}
35870		#endif
35871
35872		#ifdef __LITTLE_ENDIAN__
35873		__ai uint64x2_t vbcaxq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
35874		uint64x2_t __ret;
35875		__ret = (uint64x2_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
35876		return __ret;
35877		}
35878		#else
35879		__ai uint64x2_t vbcaxq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
35880		uint64x2_t __ret;
35881		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35882		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
35883		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
35884		__ret = (uint64x2_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
35885		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	37207	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t __noswap_vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37208	float16x4_t __ret;
	37209	__ret = (float16x4_t) __builtin_neon_vcmla_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
35886	37210	return __ret;
35887	37211	}
35888	37212	#endif
35889	37213
35890	37214	#ifdef __LITTLE_ENDIAN__
35891		__ai uint16x8_t vbcaxq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
35892		uint16x8_t __ret;
35893		__ret = (uint16x8_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 49);
35894		return __ret;
35895		}
	37215	#define vcmla_lane_f16(__p0_209, __p1_209, __p2_209, __p3_209) __extension__ ({ \
	37216	float16x4_t __ret_209; \
	37217	float16x4_t __s0_209 = __p0_209; \
	37218	float16x4_t __s1_209 = __p1_209; \
	37219	float16x4_t __s2_209 = __p2_209; \
	37220	float16x4_t __reint_209 = __s2_209; \
	37221	uint32x2_t __reint1_209 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_209, __p3_209), vget_lane_u32((uint32x2_t ) &__reint_209, __p3_209)}; \
	37222	__ret_209 = vcmla_f16(__s0_209, __s1_209, (float16x4_t ) &__reint1_209); \
	37223	__ret_209; \
	37224	})
35896	37225	#else
35897		__ai uint16x8_t vbcaxq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
35898		uint16x8_t __ret;
35899		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35900		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
35901		uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
35902		__ret = (uint16x8_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
35903		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35904		return __ret;
35905		}
	37226	#define vcmla_lane_f16(__p0_210, __p1_210, __p2_210, __p3_210) __extension__ ({ \
	37227	float16x4_t __ret_210; \
	37228	float16x4_t __s0_210 = __p0_210; \
	37229	float16x4_t __s1_210 = __p1_210; \
	37230	float16x4_t __s2_210 = __p2_210; \
	37231	float16x4_t __rev0_210; __rev0_210 = __builtin_shufflevector(__s0_210, __s0_210, 3, 2, 1, 0); \
	37232	float16x4_t __rev1_210; __rev1_210 = __builtin_shufflevector(__s1_210, __s1_210, 3, 2, 1, 0); \
	37233	float16x4_t __rev2_210; __rev2_210 = __builtin_shufflevector(__s2_210, __s2_210, 3, 2, 1, 0); \
	37234	float16x4_t __reint_210 = __rev2_210; \
	37235	uint32x2_t __reint1_210 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_210, __p3_210), __noswap_vget_lane_u32((uint32x2_t ) &__reint_210, __p3_210)}; \
	37236	__ret_210 = __noswap_vcmla_f16(__rev0_210, __rev1_210, (float16x4_t ) &__reint1_210); \
	37237	__ret_210 = __builtin_shufflevector(__ret_210, __ret_210, 3, 2, 1, 0); \
	37238	__ret_210; \
	37239	})
35906	37240	#endif
35907	37241
35908	37242	#ifdef __LITTLE_ENDIAN__
35909		__ai int8x16_t vbcaxq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
35910		int8x16_t __ret;
35911		__ret = (int8x16_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
35912		return __ret;
35913		}
	37243	#define vcmlaq_lane_f16(__p0_211, __p1_211, __p2_211, __p3_211) __extension__ ({ \
	37244	float16x8_t __ret_211; \
	37245	float16x8_t __s0_211 = __p0_211; \
	37246	float16x8_t __s1_211 = __p1_211; \
	37247	float16x4_t __s2_211 = __p2_211; \
	37248	float16x4_t __reint_211 = __s2_211; \
	37249	uint32x4_t __reint1_211 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211), vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211), vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211), vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211)}; \
	37250	__ret_211 = vcmlaq_f16(__s0_211, __s1_211, (float16x8_t ) &__reint1_211); \
	37251	__ret_211; \
	37252	})
35914	37253	#else
35915		__ai int8x16_t vbcaxq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
35916		int8x16_t __ret;
35917		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35918		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35919		int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35920		__ret = (int8x16_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
35921		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35922		return __ret;
35923		}
	37254	#define vcmlaq_lane_f16(__p0_212, __p1_212, __p2_212, __p3_212) __extension__ ({ \
	37255	float16x8_t __ret_212; \
	37256	float16x8_t __s0_212 = __p0_212; \
	37257	float16x8_t __s1_212 = __p1_212; \
	37258	float16x4_t __s2_212 = __p2_212; \
	37259	float16x8_t __rev0_212; __rev0_212 = __builtin_shufflevector(__s0_212, __s0_212, 7, 6, 5, 4, 3, 2, 1, 0); \
	37260	float16x8_t __rev1_212; __rev1_212 = __builtin_shufflevector(__s1_212, __s1_212, 7, 6, 5, 4, 3, 2, 1, 0); \
	37261	float16x4_t __rev2_212; __rev2_212 = __builtin_shufflevector(__s2_212, __s2_212, 3, 2, 1, 0); \
	37262	float16x4_t __reint_212 = __rev2_212; \
	37263	uint32x4_t __reint1_212 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212), __noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212), __noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212), __noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212)}; \
	37264	__ret_212 = __noswap_vcmlaq_f16(__rev0_212, __rev1_212, (float16x8_t ) &__reint1_212); \
	37265	__ret_212 = __builtin_shufflevector(__ret_212, __ret_212, 7, 6, 5, 4, 3, 2, 1, 0); \
	37266	__ret_212; \
	37267	})
35924	37268	#endif
35925	37269
35926	37270	#ifdef __LITTLE_ENDIAN__
35927		__ai int32x4_t vbcaxq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
35928		int32x4_t __ret;
35929		__ret = (int32x4_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
35930		return __ret;
35931		}
	37271	#define vcmla_laneq_f16(__p0_213, __p1_213, __p2_213, __p3_213) __extension__ ({ \
	37272	float16x4_t __ret_213; \
	37273	float16x4_t __s0_213 = __p0_213; \
	37274	float16x4_t __s1_213 = __p1_213; \
	37275	float16x8_t __s2_213 = __p2_213; \
	37276	float16x8_t __reint_213 = __s2_213; \
	37277	uint32x2_t __reint1_213 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_213, __p3_213), vgetq_lane_u32((uint32x4_t ) &__reint_213, __p3_213)}; \
	37278	__ret_213 = vcmla_f16(__s0_213, __s1_213, (float16x4_t ) &__reint1_213); \
	37279	__ret_213; \
	37280	})
35932	37281	#else
35933		__ai int32x4_t vbcaxq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
35934		int32x4_t __ret;
35935		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35936		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35937		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35938		__ret = (int32x4_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
35939		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35940		return __ret;
35941		}
	37282	#define vcmla_laneq_f16(__p0_214, __p1_214, __p2_214, __p3_214) __extension__ ({ \
	37283	float16x4_t __ret_214; \
	37284	float16x4_t __s0_214 = __p0_214; \
	37285	float16x4_t __s1_214 = __p1_214; \
	37286	float16x8_t __s2_214 = __p2_214; \
	37287	float16x4_t __rev0_214; __rev0_214 = __builtin_shufflevector(__s0_214, __s0_214, 3, 2, 1, 0); \
	37288	float16x4_t __rev1_214; __rev1_214 = __builtin_shufflevector(__s1_214, __s1_214, 3, 2, 1, 0); \
	37289	float16x8_t __rev2_214; __rev2_214 = __builtin_shufflevector(__s2_214, __s2_214, 7, 6, 5, 4, 3, 2, 1, 0); \
	37290	float16x8_t __reint_214 = __rev2_214; \
	37291	uint32x2_t __reint1_214 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_214, __p3_214), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_214, __p3_214)}; \
	37292	__ret_214 = __noswap_vcmla_f16(__rev0_214, __rev1_214, (float16x4_t ) &__reint1_214); \
	37293	__ret_214 = __builtin_shufflevector(__ret_214, __ret_214, 3, 2, 1, 0); \
	37294	__ret_214; \
	37295	})
35942	37296	#endif
35943	37297
35944	37298	#ifdef __LITTLE_ENDIAN__
35945		__ai int64x2_t vbcaxq_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
35946		int64x2_t __ret;
35947		__ret = (int64x2_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 35);
35948		return __ret;
35949		}
	37299	#define vcmlaq_laneq_f16(__p0_215, __p1_215, __p2_215, __p3_215) __extension__ ({ \
	37300	float16x8_t __ret_215; \
	37301	float16x8_t __s0_215 = __p0_215; \
	37302	float16x8_t __s1_215 = __p1_215; \
	37303	float16x8_t __s2_215 = __p2_215; \
	37304	float16x8_t __reint_215 = __s2_215; \
	37305	uint32x4_t __reint1_215 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215), vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215), vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215), vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215)}; \
	37306	__ret_215 = vcmlaq_f16(__s0_215, __s1_215, (float16x8_t ) &__reint1_215); \
	37307	__ret_215; \
	37308	})
35950	37309	#else
35951		__ai int64x2_t vbcaxq_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
35952		int64x2_t __ret;
35953		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35954		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
35955		int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
35956		__ret = (int64x2_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
35957		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35958		return __ret;
35959		}
	37310	#define vcmlaq_laneq_f16(__p0_216, __p1_216, __p2_216, __p3_216) __extension__ ({ \
	37311	float16x8_t __ret_216; \
	37312	float16x8_t __s0_216 = __p0_216; \
	37313	float16x8_t __s1_216 = __p1_216; \
	37314	float16x8_t __s2_216 = __p2_216; \
	37315	float16x8_t __rev0_216; __rev0_216 = __builtin_shufflevector(__s0_216, __s0_216, 7, 6, 5, 4, 3, 2, 1, 0); \
	37316	float16x8_t __rev1_216; __rev1_216 = __builtin_shufflevector(__s1_216, __s1_216, 7, 6, 5, 4, 3, 2, 1, 0); \
	37317	float16x8_t __rev2_216; __rev2_216 = __builtin_shufflevector(__s2_216, __s2_216, 7, 6, 5, 4, 3, 2, 1, 0); \
	37318	float16x8_t __reint_216 = __rev2_216; \
	37319	uint32x4_t __reint1_216 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216)}; \
	37320	__ret_216 = __noswap_vcmlaq_f16(__rev0_216, __rev1_216, (float16x8_t ) &__reint1_216); \
	37321	__ret_216 = __builtin_shufflevector(__ret_216, __ret_216, 7, 6, 5, 4, 3, 2, 1, 0); \
	37322	__ret_216; \
	37323	})
35960	37324	#endif
35961	37325
35962	37326	#ifdef __LITTLE_ENDIAN__
35963		__ai int16x8_t vbcaxq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
35964		int16x8_t __ret;
35965		__ret = (int16x8_t) __builtin_neon_vbcaxq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
	37327	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37328	float16x8_t __ret;
	37329	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
35966	37330	return __ret;
35967	37331	}
35968	37332	#else
35969		__ai int16x8_t vbcaxq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
35970		int16x8_t __ret;
35971		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
35972		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
35973		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
35974		__ret = (int16x8_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
	37333	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37334	float16x8_t __ret;
	37335	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	37336	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	37337	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	37338	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
35975	37339	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
35976	37340	return __ret;
35977	37341	}
35978		#endif
35979
35980		#ifdef __LITTLE_ENDIAN__
35981		__ai uint8x16_t veor3q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
35982		uint8x16_t __ret;
35983		__ret = (uint8x16_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
35984		return __ret;
35985		}
35986		#else
35987		__ai uint8x16_t veor3q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
35988		uint8x16_t __ret;
35989		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35990		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35991		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35992		__ret = (uint8x16_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
35993		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	37342	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t __noswap_vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37343	float16x8_t __ret;
	37344	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
35994	37345	return __ret;
35995	37346	}
35996	37347	#endif
35997	37348
35998	37349	#ifdef __LITTLE_ENDIAN__
35999		__ai uint32x4_t veor3q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36000		uint32x4_t __ret;
36001		__ret = (uint32x4_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37350	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37351	float16x4_t __ret;
	37352	__ret = (float16x4_t) __builtin_neon_vcmla_rot180_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
36002	37353	return __ret;
36003	37354	}
36004	37355	#else
36005		__ai uint32x4_t veor3q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36006		uint32x4_t __ret;
36007		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36008		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36009		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36010		__ret = (uint32x4_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	37356	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37357	float16x4_t __ret;
	37358	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	37359	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	37360	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	37361	__ret = (float16x4_t) __builtin_neon_vcmla_rot180_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
36011	37362	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36012	37363	return __ret;
36013	37364	}
36014		#endif
36015
36016		#ifdef __LITTLE_ENDIAN__
36017		__ai uint64x2_t veor3q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36018		uint64x2_t __ret;
36019		__ret = (uint64x2_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
36020		return __ret;
36021		}
36022		#else
36023		__ai uint64x2_t veor3q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36024		uint64x2_t __ret;
36025		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36026		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36027		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36028		__ret = (uint64x2_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
36029		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	37365	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t __noswap_vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37366	float16x4_t __ret;
	37367	__ret = (float16x4_t) __builtin_neon_vcmla_rot180_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
36030	37368	return __ret;
36031	37369	}
36032	37370	#endif
36033	37371
36034	37372	#ifdef __LITTLE_ENDIAN__
36035		__ai uint16x8_t veor3q_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
36036		uint16x8_t __ret;
36037		__ret = (uint16x8_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 49);
36038		return __ret;
36039		}
	37373	#define vcmla_rot180_lane_f16(__p0_217, __p1_217, __p2_217, __p3_217) __extension__ ({ \
	37374	float16x4_t __ret_217; \
	37375	float16x4_t __s0_217 = __p0_217; \
	37376	float16x4_t __s1_217 = __p1_217; \
	37377	float16x4_t __s2_217 = __p2_217; \
	37378	float16x4_t __reint_217 = __s2_217; \
	37379	uint32x2_t __reint1_217 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_217, __p3_217), vget_lane_u32((uint32x2_t ) &__reint_217, __p3_217)}; \
	37380	__ret_217 = vcmla_rot180_f16(__s0_217, __s1_217, (float16x4_t ) &__reint1_217); \
	37381	__ret_217; \
	37382	})
36040	37383	#else
36041		__ai uint16x8_t veor3q_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
36042		uint16x8_t __ret;
36043		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
36044		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
36045		uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
36046		__ret = (uint16x8_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
36047		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
36048		return __ret;
36049		}
	37384	#define vcmla_rot180_lane_f16(__p0_218, __p1_218, __p2_218, __p3_218) __extension__ ({ \
	37385	float16x4_t __ret_218; \
	37386	float16x4_t __s0_218 = __p0_218; \
	37387	float16x4_t __s1_218 = __p1_218; \
	37388	float16x4_t __s2_218 = __p2_218; \
	37389	float16x4_t __rev0_218; __rev0_218 = __builtin_shufflevector(__s0_218, __s0_218, 3, 2, 1, 0); \
	37390	float16x4_t __rev1_218; __rev1_218 = __builtin_shufflevector(__s1_218, __s1_218, 3, 2, 1, 0); \
	37391	float16x4_t __rev2_218; __rev2_218 = __builtin_shufflevector(__s2_218, __s2_218, 3, 2, 1, 0); \
	37392	float16x4_t __reint_218 = __rev2_218; \
	37393	uint32x2_t __reint1_218 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_218, __p3_218), __noswap_vget_lane_u32((uint32x2_t ) &__reint_218, __p3_218)}; \
	37394	__ret_218 = __noswap_vcmla_rot180_f16(__rev0_218, __rev1_218, (float16x4_t ) &__reint1_218); \
	37395	__ret_218 = __builtin_shufflevector(__ret_218, __ret_218, 3, 2, 1, 0); \
	37396	__ret_218; \
	37397	})
36050	37398	#endif
36051	37399
36052	37400	#ifdef __LITTLE_ENDIAN__
36053		__ai int8x16_t veor3q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
36054		int8x16_t __ret;
36055		__ret = (int8x16_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
36056		return __ret;
36057		}
	37401	#define vcmlaq_rot180_lane_f16(__p0_219, __p1_219, __p2_219, __p3_219) __extension__ ({ \
	37402	float16x8_t __ret_219; \
	37403	float16x8_t __s0_219 = __p0_219; \
	37404	float16x8_t __s1_219 = __p1_219; \
	37405	float16x4_t __s2_219 = __p2_219; \
	37406	float16x4_t __reint_219 = __s2_219; \
	37407	uint32x4_t __reint1_219 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219), vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219), vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219), vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219)}; \
	37408	__ret_219 = vcmlaq_rot180_f16(__s0_219, __s1_219, (float16x8_t ) &__reint1_219); \
	37409	__ret_219; \
	37410	})
36058	37411	#else
36059		__ai int8x16_t veor3q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
36060		int8x16_t __ret;
36061		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36062		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36063		int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36064		__ret = (int8x16_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
36065		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36066		return __ret;
36067		}
	37412	#define vcmlaq_rot180_lane_f16(__p0_220, __p1_220, __p2_220, __p3_220) __extension__ ({ \
	37413	float16x8_t __ret_220; \
	37414	float16x8_t __s0_220 = __p0_220; \
	37415	float16x8_t __s1_220 = __p1_220; \
	37416	float16x4_t __s2_220 = __p2_220; \
	37417	float16x8_t __rev0_220; __rev0_220 = __builtin_shufflevector(__s0_220, __s0_220, 7, 6, 5, 4, 3, 2, 1, 0); \
	37418	float16x8_t __rev1_220; __rev1_220 = __builtin_shufflevector(__s1_220, __s1_220, 7, 6, 5, 4, 3, 2, 1, 0); \
	37419	float16x4_t __rev2_220; __rev2_220 = __builtin_shufflevector(__s2_220, __s2_220, 3, 2, 1, 0); \
	37420	float16x4_t __reint_220 = __rev2_220; \
	37421	uint32x4_t __reint1_220 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220), __noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220), __noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220), __noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220)}; \
	37422	__ret_220 = __noswap_vcmlaq_rot180_f16(__rev0_220, __rev1_220, (float16x8_t ) &__reint1_220); \
	37423	__ret_220 = __builtin_shufflevector(__ret_220, __ret_220, 7, 6, 5, 4, 3, 2, 1, 0); \
	37424	__ret_220; \
	37425	})
36068	37426	#endif
36069	37427
36070	37428	#ifdef __LITTLE_ENDIAN__
36071		__ai int32x4_t veor3q_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
36072		int32x4_t __ret;
36073		__ret = (int32x4_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
36074		return __ret;
36075		}
	37429	#define vcmla_rot180_laneq_f16(__p0_221, __p1_221, __p2_221, __p3_221) __extension__ ({ \
	37430	float16x4_t __ret_221; \
	37431	float16x4_t __s0_221 = __p0_221; \
	37432	float16x4_t __s1_221 = __p1_221; \
	37433	float16x8_t __s2_221 = __p2_221; \
	37434	float16x8_t __reint_221 = __s2_221; \
	37435	uint32x2_t __reint1_221 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_221, __p3_221), vgetq_lane_u32((uint32x4_t ) &__reint_221, __p3_221)}; \
	37436	__ret_221 = vcmla_rot180_f16(__s0_221, __s1_221, (float16x4_t ) &__reint1_221); \
	37437	__ret_221; \
	37438	})
36076	37439	#else
36077		__ai int32x4_t veor3q_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
36078		int32x4_t __ret;
36079		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36080		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36081		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36082		__ret = (int32x4_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
36083		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36084		return __ret;
36085		}
	37440	#define vcmla_rot180_laneq_f16(__p0_222, __p1_222, __p2_222, __p3_222) __extension__ ({ \
	37441	float16x4_t __ret_222; \
	37442	float16x4_t __s0_222 = __p0_222; \
	37443	float16x4_t __s1_222 = __p1_222; \
	37444	float16x8_t __s2_222 = __p2_222; \
	37445	float16x4_t __rev0_222; __rev0_222 = __builtin_shufflevector(__s0_222, __s0_222, 3, 2, 1, 0); \
	37446	float16x4_t __rev1_222; __rev1_222 = __builtin_shufflevector(__s1_222, __s1_222, 3, 2, 1, 0); \
	37447	float16x8_t __rev2_222; __rev2_222 = __builtin_shufflevector(__s2_222, __s2_222, 7, 6, 5, 4, 3, 2, 1, 0); \
	37448	float16x8_t __reint_222 = __rev2_222; \
	37449	uint32x2_t __reint1_222 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_222, __p3_222), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_222, __p3_222)}; \
	37450	__ret_222 = __noswap_vcmla_rot180_f16(__rev0_222, __rev1_222, (float16x4_t ) &__reint1_222); \
	37451	__ret_222 = __builtin_shufflevector(__ret_222, __ret_222, 3, 2, 1, 0); \
	37452	__ret_222; \
	37453	})
36086	37454	#endif
36087	37455
36088	37456	#ifdef __LITTLE_ENDIAN__
36089		__ai int64x2_t veor3q_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
36090		int64x2_t __ret;
36091		__ret = (int64x2_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 35);
36092		return __ret;
36093		}
	37457	#define vcmlaq_rot180_laneq_f16(__p0_223, __p1_223, __p2_223, __p3_223) __extension__ ({ \
	37458	float16x8_t __ret_223; \
	37459	float16x8_t __s0_223 = __p0_223; \
	37460	float16x8_t __s1_223 = __p1_223; \
	37461	float16x8_t __s2_223 = __p2_223; \
	37462	float16x8_t __reint_223 = __s2_223; \
	37463	uint32x4_t __reint1_223 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223), vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223), vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223), vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223)}; \
	37464	__ret_223 = vcmlaq_rot180_f16(__s0_223, __s1_223, (float16x8_t ) &__reint1_223); \
	37465	__ret_223; \
	37466	})
36094	37467	#else
36095		__ai int64x2_t veor3q_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
36096		int64x2_t __ret;
36097		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36098		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36099		int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36100		__ret = (int64x2_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
36101		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36102		return __ret;
36103		}
	37468	#define vcmlaq_rot180_laneq_f16(__p0_224, __p1_224, __p2_224, __p3_224) __extension__ ({ \
	37469	float16x8_t __ret_224; \
	37470	float16x8_t __s0_224 = __p0_224; \
	37471	float16x8_t __s1_224 = __p1_224; \
	37472	float16x8_t __s2_224 = __p2_224; \
	37473	float16x8_t __rev0_224; __rev0_224 = __builtin_shufflevector(__s0_224, __s0_224, 7, 6, 5, 4, 3, 2, 1, 0); \
	37474	float16x8_t __rev1_224; __rev1_224 = __builtin_shufflevector(__s1_224, __s1_224, 7, 6, 5, 4, 3, 2, 1, 0); \
	37475	float16x8_t __rev2_224; __rev2_224 = __builtin_shufflevector(__s2_224, __s2_224, 7, 6, 5, 4, 3, 2, 1, 0); \
	37476	float16x8_t __reint_224 = __rev2_224; \
	37477	uint32x4_t __reint1_224 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224)}; \
	37478	__ret_224 = __noswap_vcmlaq_rot180_f16(__rev0_224, __rev1_224, (float16x8_t ) &__reint1_224); \
	37479	__ret_224 = __builtin_shufflevector(__ret_224, __ret_224, 7, 6, 5, 4, 3, 2, 1, 0); \
	37480	__ret_224; \
	37481	})
36104	37482	#endif
36105	37483
36106	37484	#ifdef __LITTLE_ENDIAN__
36107		__ai int16x8_t veor3q_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
36108		int16x8_t __ret;
36109		__ret = (int16x8_t) __builtin_neon_veor3q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
	37485	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37486	float16x8_t __ret;
	37487	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
36110	37488	return __ret;
36111	37489	}
36112	37490	#else
36113		__ai int16x8_t veor3q_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
36114		int16x8_t __ret;
36115		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
36116		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
36117		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
36118		__ret = (int16x8_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
	37491	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37492	float16x8_t __ret;
	37493	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	37494	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	37495	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	37496	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
36119	37497	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
36120	37498	return __ret;
36121	37499	}
	37500	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t __noswap_vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37501	float16x8_t __ret;
	37502	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	37503	return __ret;
	37504	}
36122	37505	#endif
36123	37506
36124	37507	#ifdef __LITTLE_ENDIAN__
36125		__ai uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
36126		uint64x2_t __ret;
36127		__ret = (uint64x2_t) __builtin_neon_vrax1q_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	37508	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37509	float16x4_t __ret;
	37510	__ret = (float16x4_t) __builtin_neon_vcmla_rot270_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
36128	37511	return __ret;
36129	37512	}
36130	37513	#else
36131		__ai uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
36132		uint64x2_t __ret;
36133		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36134		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36135		__ret = (uint64x2_t) __builtin_neon_vrax1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
36136		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	37514	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37515	float16x4_t __ret;
	37516	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	37517	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	37518	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	37519	__ret = (float16x4_t) __builtin_neon_vcmla_rot270_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
	37520	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	37521	return __ret;
	37522	}
	37523	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t __noswap_vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37524	float16x4_t __ret;
	37525	__ret = (float16x4_t) __builtin_neon_vcmla_rot270_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
36137	37526	return __ret;
36138	37527	}
36139	37528	#endif
36140	37529
36141	37530	#ifdef __LITTLE_ENDIAN__
36142		#define vxarq_u64(__p0, __p1, __p2) __extension__ ({ \
36143		uint64x2_t __ret; \
36144		uint64x2_t __s0 = __p0; \
36145		uint64x2_t __s1 = __p1; \
36146		__ret = (uint64x2_t) __builtin_neon_vxarq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
36147		__ret; \
	37531	#define vcmla_rot270_lane_f16(__p0_225, __p1_225, __p2_225, __p3_225) __extension__ ({ \
	37532	float16x4_t __ret_225; \
	37533	float16x4_t __s0_225 = __p0_225; \
	37534	float16x4_t __s1_225 = __p1_225; \
	37535	float16x4_t __s2_225 = __p2_225; \
	37536	float16x4_t __reint_225 = __s2_225; \
	37537	uint32x2_t __reint1_225 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_225, __p3_225), vget_lane_u32((uint32x2_t ) &__reint_225, __p3_225)}; \
	37538	__ret_225 = vcmla_rot270_f16(__s0_225, __s1_225, (float16x4_t ) &__reint1_225); \
	37539	__ret_225; \
36148	37540	})
36149	37541	#else
36150		#define vxarq_u64(__p0, __p1, __p2) __extension__ ({ \
36151		uint64x2_t __ret; \
36152		uint64x2_t __s0 = __p0; \
36153		uint64x2_t __s1 = __p1; \
36154		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
36155		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
36156		__ret = (uint64x2_t) __builtin_neon_vxarq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
36157		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
36158		__ret; \
	37542	#define vcmla_rot270_lane_f16(__p0_226, __p1_226, __p2_226, __p3_226) __extension__ ({ \
	37543	float16x4_t __ret_226; \
	37544	float16x4_t __s0_226 = __p0_226; \
	37545	float16x4_t __s1_226 = __p1_226; \
	37546	float16x4_t __s2_226 = __p2_226; \
	37547	float16x4_t __rev0_226; __rev0_226 = __builtin_shufflevector(__s0_226, __s0_226, 3, 2, 1, 0); \
	37548	float16x4_t __rev1_226; __rev1_226 = __builtin_shufflevector(__s1_226, __s1_226, 3, 2, 1, 0); \
	37549	float16x4_t __rev2_226; __rev2_226 = __builtin_shufflevector(__s2_226, __s2_226, 3, 2, 1, 0); \
	37550	float16x4_t __reint_226 = __rev2_226; \
	37551	uint32x2_t __reint1_226 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_226, __p3_226), __noswap_vget_lane_u32((uint32x2_t ) &__reint_226, __p3_226)}; \
	37552	__ret_226 = __noswap_vcmla_rot270_f16(__rev0_226, __rev1_226, (float16x4_t ) &__reint1_226); \
	37553	__ret_226 = __builtin_shufflevector(__ret_226, __ret_226, 3, 2, 1, 0); \
	37554	__ret_226; \
36159	37555	})
36160	37556	#endif
36161	37557
36162		#endif
36163		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_SHA512) && defined(__aarch64__)
36164	37558	#ifdef __LITTLE_ENDIAN__
36165		__ai uint64x2_t vsha512hq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36166		uint64x2_t __ret;
36167		__ret = (uint64x2_t) __builtin_neon_vsha512hq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
36168		return __ret;
36169		}
	37559	#define vcmlaq_rot270_lane_f16(__p0_227, __p1_227, __p2_227, __p3_227) __extension__ ({ \
	37560	float16x8_t __ret_227; \
	37561	float16x8_t __s0_227 = __p0_227; \
	37562	float16x8_t __s1_227 = __p1_227; \
	37563	float16x4_t __s2_227 = __p2_227; \
	37564	float16x4_t __reint_227 = __s2_227; \
	37565	uint32x4_t __reint1_227 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227), vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227), vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227), vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227)}; \
	37566	__ret_227 = vcmlaq_rot270_f16(__s0_227, __s1_227, (float16x8_t ) &__reint1_227); \
	37567	__ret_227; \
	37568	})
36170	37569	#else
36171		__ai uint64x2_t vsha512hq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36172		uint64x2_t __ret;
36173		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36174		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36175		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36176		__ret = (uint64x2_t) __builtin_neon_vsha512hq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
36177		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36178		return __ret;
36179		}
	37570	#define vcmlaq_rot270_lane_f16(__p0_228, __p1_228, __p2_228, __p3_228) __extension__ ({ \
	37571	float16x8_t __ret_228; \
	37572	float16x8_t __s0_228 = __p0_228; \
	37573	float16x8_t __s1_228 = __p1_228; \
	37574	float16x4_t __s2_228 = __p2_228; \
	37575	float16x8_t __rev0_228; __rev0_228 = __builtin_shufflevector(__s0_228, __s0_228, 7, 6, 5, 4, 3, 2, 1, 0); \
	37576	float16x8_t __rev1_228; __rev1_228 = __builtin_shufflevector(__s1_228, __s1_228, 7, 6, 5, 4, 3, 2, 1, 0); \
	37577	float16x4_t __rev2_228; __rev2_228 = __builtin_shufflevector(__s2_228, __s2_228, 3, 2, 1, 0); \
	37578	float16x4_t __reint_228 = __rev2_228; \
	37579	uint32x4_t __reint1_228 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228), __noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228), __noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228), __noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228)}; \
	37580	__ret_228 = __noswap_vcmlaq_rot270_f16(__rev0_228, __rev1_228, (float16x8_t ) &__reint1_228); \
	37581	__ret_228 = __builtin_shufflevector(__ret_228, __ret_228, 7, 6, 5, 4, 3, 2, 1, 0); \
	37582	__ret_228; \
	37583	})
36180	37584	#endif
36181	37585
36182	37586	#ifdef __LITTLE_ENDIAN__
36183		__ai uint64x2_t vsha512h2q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36184		uint64x2_t __ret;
36185		__ret = (uint64x2_t) __builtin_neon_vsha512h2q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
36186		return __ret;
36187		}
	37587	#define vcmla_rot270_laneq_f16(__p0_229, __p1_229, __p2_229, __p3_229) __extension__ ({ \
	37588	float16x4_t __ret_229; \
	37589	float16x4_t __s0_229 = __p0_229; \
	37590	float16x4_t __s1_229 = __p1_229; \
	37591	float16x8_t __s2_229 = __p2_229; \
	37592	float16x8_t __reint_229 = __s2_229; \
	37593	uint32x2_t __reint1_229 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_229, __p3_229), vgetq_lane_u32((uint32x4_t ) &__reint_229, __p3_229)}; \
	37594	__ret_229 = vcmla_rot270_f16(__s0_229, __s1_229, (float16x4_t ) &__reint1_229); \
	37595	__ret_229; \
	37596	})
36188	37597	#else
36189		__ai uint64x2_t vsha512h2q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36190		uint64x2_t __ret;
36191		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36192		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36193		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36194		__ret = (uint64x2_t) __builtin_neon_vsha512h2q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
36195		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36196		return __ret;
36197		}
	37598	#define vcmla_rot270_laneq_f16(__p0_230, __p1_230, __p2_230, __p3_230) __extension__ ({ \
	37599	float16x4_t __ret_230; \
	37600	float16x4_t __s0_230 = __p0_230; \
	37601	float16x4_t __s1_230 = __p1_230; \
	37602	float16x8_t __s2_230 = __p2_230; \
	37603	float16x4_t __rev0_230; __rev0_230 = __builtin_shufflevector(__s0_230, __s0_230, 3, 2, 1, 0); \
	37604	float16x4_t __rev1_230; __rev1_230 = __builtin_shufflevector(__s1_230, __s1_230, 3, 2, 1, 0); \
	37605	float16x8_t __rev2_230; __rev2_230 = __builtin_shufflevector(__s2_230, __s2_230, 7, 6, 5, 4, 3, 2, 1, 0); \
	37606	float16x8_t __reint_230 = __rev2_230; \
	37607	uint32x2_t __reint1_230 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_230, __p3_230), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_230, __p3_230)}; \
	37608	__ret_230 = __noswap_vcmla_rot270_f16(__rev0_230, __rev1_230, (float16x4_t ) &__reint1_230); \
	37609	__ret_230 = __builtin_shufflevector(__ret_230, __ret_230, 3, 2, 1, 0); \
	37610	__ret_230; \
	37611	})
36198	37612	#endif
36199	37613
36200	37614	#ifdef __LITTLE_ENDIAN__
36201		__ai uint64x2_t vsha512su0q_u64(uint64x2_t __p0, uint64x2_t __p1) {
36202		uint64x2_t __ret;
36203		__ret = (uint64x2_t) __builtin_neon_vsha512su0q_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
36204		return __ret;
36205		}
	37615	#define vcmlaq_rot270_laneq_f16(__p0_231, __p1_231, __p2_231, __p3_231) __extension__ ({ \
	37616	float16x8_t __ret_231; \
	37617	float16x8_t __s0_231 = __p0_231; \
	37618	float16x8_t __s1_231 = __p1_231; \
	37619	float16x8_t __s2_231 = __p2_231; \
	37620	float16x8_t __reint_231 = __s2_231; \
	37621	uint32x4_t __reint1_231 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231), vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231), vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231), vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231)}; \
	37622	__ret_231 = vcmlaq_rot270_f16(__s0_231, __s1_231, (float16x8_t ) &__reint1_231); \
	37623	__ret_231; \
	37624	})
36206	37625	#else
36207		__ai uint64x2_t vsha512su0q_u64(uint64x2_t __p0, uint64x2_t __p1) {
36208		uint64x2_t __ret;
36209		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36210		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36211		__ret = (uint64x2_t) __builtin_neon_vsha512su0q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
36212		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36213		return __ret;
36214		}
	37626	#define vcmlaq_rot270_laneq_f16(__p0_232, __p1_232, __p2_232, __p3_232) __extension__ ({ \
	37627	float16x8_t __ret_232; \
	37628	float16x8_t __s0_232 = __p0_232; \
	37629	float16x8_t __s1_232 = __p1_232; \
	37630	float16x8_t __s2_232 = __p2_232; \
	37631	float16x8_t __rev0_232; __rev0_232 = __builtin_shufflevector(__s0_232, __s0_232, 7, 6, 5, 4, 3, 2, 1, 0); \
	37632	float16x8_t __rev1_232; __rev1_232 = __builtin_shufflevector(__s1_232, __s1_232, 7, 6, 5, 4, 3, 2, 1, 0); \
	37633	float16x8_t __rev2_232; __rev2_232 = __builtin_shufflevector(__s2_232, __s2_232, 7, 6, 5, 4, 3, 2, 1, 0); \
	37634	float16x8_t __reint_232 = __rev2_232; \
	37635	uint32x4_t __reint1_232 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232)}; \
	37636	__ret_232 = __noswap_vcmlaq_rot270_f16(__rev0_232, __rev1_232, (float16x8_t ) &__reint1_232); \
	37637	__ret_232 = __builtin_shufflevector(__ret_232, __ret_232, 7, 6, 5, 4, 3, 2, 1, 0); \
	37638	__ret_232; \
	37639	})
36215	37640	#endif
36216	37641
36217	37642	#ifdef __LITTLE_ENDIAN__
36218		__ai uint64x2_t vsha512su1q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36219		uint64x2_t __ret;
36220		__ret = (uint64x2_t) __builtin_neon_vsha512su1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
	37643	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37644	float16x8_t __ret;
	37645	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
36221	37646	return __ret;
36222	37647	}
36223	37648	#else
36224		__ai uint64x2_t vsha512su1q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36225		uint64x2_t __ret;
36226		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36227		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36228		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36229		__ret = (uint64x2_t) __builtin_neon_vsha512su1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
36230		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36231		return __ret;
36232		}
36233		#endif
36234
36235		#endif
36236		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_SM3) && defined(__aarch64__)
36237		#ifdef __LITTLE_ENDIAN__
36238		__ai uint32x4_t vsm3partw1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36239		uint32x4_t __ret;
36240		__ret = (uint32x4_t) __builtin_neon_vsm3partw1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37649	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37650	float16x8_t __ret;
	37651	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	37652	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	37653	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	37654	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
	37655	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
36241	37656	return __ret;
36242	37657	}
36243		#else
36244		__ai uint32x4_t vsm3partw1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36245		uint32x4_t __ret;
36246		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36247		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36248		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36249		__ret = (uint32x4_t) __builtin_neon_vsm3partw1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
36250		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	37658	__ai __attribute__((target("v8.3a,fullfp16"))) float16x8_t __noswap_vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
	37659	float16x8_t __ret;
	37660	__ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
36251	37661	return __ret;
36252	37662	}
36253	37663	#endif
36254	37664
36255	37665	#ifdef __LITTLE_ENDIAN__
36256		__ai uint32x4_t vsm3partw2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36257		uint32x4_t __ret;
36258		__ret = (uint32x4_t) __builtin_neon_vsm3partw2q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	37666	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37667	float16x4_t __ret;
	37668	__ret = (float16x4_t) __builtin_neon_vcmla_rot90_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
36259	37669	return __ret;
36260	37670	}
36261	37671	#else
36262		__ai uint32x4_t vsm3partw2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36263		uint32x4_t __ret;
36264		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36265		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36266		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36267		__ret = (uint32x4_t) __builtin_neon_vsm3partw2q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	37672	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37673	float16x4_t __ret;
	37674	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	37675	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	37676	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	37677	__ret = (float16x4_t) __builtin_neon_vcmla_rot90_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
36268	37678	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36269	37679	return __ret;
36270	37680	}
36271		#endif
36272
36273		#ifdef __LITTLE_ENDIAN__
36274		__ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36275		uint32x4_t __ret;
36276		__ret = (uint32x4_t) __builtin_neon_vsm3ss1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
36277		return __ret;
36278		}
36279		#else
36280		__ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36281		uint32x4_t __ret;
36282		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36283		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36284		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36285		__ret = (uint32x4_t) __builtin_neon_vsm3ss1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
36286		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	37681	__ai __attribute__((target("v8.3a,fullfp16"))) float16x4_t __noswap_vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
	37682	float16x4_t __ret;
	37683	__ret = (float16x4_t) __builtin_neon_vcmla_rot90_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
36287	37684	return __ret;
36288	37685	}
36289	37686	#endif
36290	37687
36291	37688	#ifdef __LITTLE_ENDIAN__
36292		#define vsm3tt1aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36293		uint32x4_t __ret; \
36294		uint32x4_t __s0 = __p0; \
36295		uint32x4_t __s1 = __p1; \
36296		uint32x4_t __s2 = __p2; \
36297		__ret = (uint32x4_t) __builtin_neon_vsm3tt1aq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
36298		__ret; \
	37689	#define vcmla_rot90_lane_f16(__p0_233, __p1_233, __p2_233, __p3_233) __extension__ ({ \
	37690	float16x4_t __ret_233; \
	37691	float16x4_t __s0_233 = __p0_233; \
	37692	float16x4_t __s1_233 = __p1_233; \
	37693	float16x4_t __s2_233 = __p2_233; \
	37694	float16x4_t __reint_233 = __s2_233; \
	37695	uint32x2_t __reint1_233 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_233, __p3_233), vget_lane_u32((uint32x2_t ) &__reint_233, __p3_233)}; \
	37696	__ret_233 = vcmla_rot90_f16(__s0_233, __s1_233, (float16x4_t ) &__reint1_233); \
	37697	__ret_233; \
36299	37698	})
36300	37699	#else
36301		#define vsm3tt1aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36302		uint32x4_t __ret; \
36303		uint32x4_t __s0 = __p0; \
36304		uint32x4_t __s1 = __p1; \
36305		uint32x4_t __s2 = __p2; \
36306		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
36307		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
36308		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36309		__ret = (uint32x4_t) __builtin_neon_vsm3tt1aq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
36310		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
36311		__ret; \
	37700	#define vcmla_rot90_lane_f16(__p0_234, __p1_234, __p2_234, __p3_234) __extension__ ({ \
	37701	float16x4_t __ret_234; \
	37702	float16x4_t __s0_234 = __p0_234; \
	37703	float16x4_t __s1_234 = __p1_234; \
	37704	float16x4_t __s2_234 = __p2_234; \
	37705	float16x4_t __rev0_234; __rev0_234 = __builtin_shufflevector(__s0_234, __s0_234, 3, 2, 1, 0); \
	37706	float16x4_t __rev1_234; __rev1_234 = __builtin_shufflevector(__s1_234, __s1_234, 3, 2, 1, 0); \
	37707	float16x4_t __rev2_234; __rev2_234 = __builtin_shufflevector(__s2_234, __s2_234, 3, 2, 1, 0); \
	37708	float16x4_t __reint_234 = __rev2_234; \
	37709	uint32x2_t __reint1_234 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_234, __p3_234), __noswap_vget_lane_u32((uint32x2_t ) &__reint_234, __p3_234)}; \
	37710	__ret_234 = __noswap_vcmla_rot90_f16(__rev0_234, __rev1_234, (float16x4_t ) &__reint1_234); \
	37711	__ret_234 = __builtin_shufflevector(__ret_234, __ret_234, 3, 2, 1, 0); \
	37712	__ret_234; \
36312	37713	})
36313	37714	#endif
36314	37715
36315	37716	#ifdef __LITTLE_ENDIAN__
36316		#define vsm3tt1bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36317		uint32x4_t __ret; \
36318		uint32x4_t __s0 = __p0; \
36319		uint32x4_t __s1 = __p1; \
36320		uint32x4_t __s2 = __p2; \
36321		__ret = (uint32x4_t) __builtin_neon_vsm3tt1bq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
36322		__ret; \
	37717	#define vcmlaq_rot90_lane_f16(__p0_235, __p1_235, __p2_235, __p3_235) __extension__ ({ \
	37718	float16x8_t __ret_235; \
	37719	float16x8_t __s0_235 = __p0_235; \
	37720	float16x8_t __s1_235 = __p1_235; \
	37721	float16x4_t __s2_235 = __p2_235; \
	37722	float16x4_t __reint_235 = __s2_235; \
	37723	uint32x4_t __reint1_235 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_235, __p3_235), vget_lane_u32((uint32x2_t ) &__reint_235, __p3_235), vget_lane_u32((uint32x2_t ) &__reint_235, __p3_235), vget_lane_u32((uint32x2_t ) &__reint_235, __p3_235)}; \
	37724	__ret_235 = vcmlaq_rot90_f16(__s0_235, __s1_235, (float16x8_t ) &__reint1_235); \
	37725	__ret_235; \
36323	37726	})
36324	37727	#else
36325		#define vsm3tt1bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36326		uint32x4_t __ret; \
36327		uint32x4_t __s0 = __p0; \
36328		uint32x4_t __s1 = __p1; \
36329		uint32x4_t __s2 = __p2; \
36330		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
36331		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
36332		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36333		__ret = (uint32x4_t) __builtin_neon_vsm3tt1bq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
36334		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
36335		__ret; \
	37728	#define vcmlaq_rot90_lane_f16(__p0_236, __p1_236, __p2_236, __p3_236) __extension__ ({ \
	37729	float16x8_t __ret_236; \
	37730	float16x8_t __s0_236 = __p0_236; \
	37731	float16x8_t __s1_236 = __p1_236; \
	37732	float16x4_t __s2_236 = __p2_236; \
	37733	float16x8_t __rev0_236; __rev0_236 = __builtin_shufflevector(__s0_236, __s0_236, 7, 6, 5, 4, 3, 2, 1, 0); \
	37734	float16x8_t __rev1_236; __rev1_236 = __builtin_shufflevector(__s1_236, __s1_236, 7, 6, 5, 4, 3, 2, 1, 0); \
	37735	float16x4_t __rev2_236; __rev2_236 = __builtin_shufflevector(__s2_236, __s2_236, 3, 2, 1, 0); \
	37736	float16x4_t __reint_236 = __rev2_236; \
	37737	uint32x4_t __reint1_236 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_236, __p3_236), __noswap_vget_lane_u32((uint32x2_t ) &__reint_236, __p3_236), __noswap_vget_lane_u32((uint32x2_t ) &__reint_236, __p3_236), __noswap_vget_lane_u32((uint32x2_t ) &__reint_236, __p3_236)}; \
	37738	__ret_236 = __noswap_vcmlaq_rot90_f16(__rev0_236, __rev1_236, (float16x8_t ) &__reint1_236); \
	37739	__ret_236 = __builtin_shufflevector(__ret_236, __ret_236, 7, 6, 5, 4, 3, 2, 1, 0); \
	37740	__ret_236; \
36336	37741	})
36337	37742	#endif
36338	37743
36339	37744	#ifdef __LITTLE_ENDIAN__
36340		#define vsm3tt2aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36341		uint32x4_t __ret; \
36342		uint32x4_t __s0 = __p0; \
36343		uint32x4_t __s1 = __p1; \
36344		uint32x4_t __s2 = __p2; \
36345		__ret = (uint32x4_t) __builtin_neon_vsm3tt2aq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
36346		__ret; \
	37745	#define vcmla_rot90_laneq_f16(__p0_237, __p1_237, __p2_237, __p3_237) __extension__ ({ \
	37746	float16x4_t __ret_237; \
	37747	float16x4_t __s0_237 = __p0_237; \
	37748	float16x4_t __s1_237 = __p1_237; \
	37749	float16x8_t __s2_237 = __p2_237; \
	37750	float16x8_t __reint_237 = __s2_237; \
	37751	uint32x2_t __reint1_237 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_237, __p3_237), vgetq_lane_u32((uint32x4_t ) &__reint_237, __p3_237)}; \
	37752	__ret_237 = vcmla_rot90_f16(__s0_237, __s1_237, (float16x4_t ) &__reint1_237); \
	37753	__ret_237; \
36347	37754	})
36348	37755	#else
36349		#define vsm3tt2aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36350		uint32x4_t __ret; \
36351		uint32x4_t __s0 = __p0; \
36352		uint32x4_t __s1 = __p1; \
36353		uint32x4_t __s2 = __p2; \
36354		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
36355		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
36356		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36357		__ret = (uint32x4_t) __builtin_neon_vsm3tt2aq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
36358		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
36359		__ret; \
	37756	#define vcmla_rot90_laneq_f16(__p0_238, __p1_238, __p2_238, __p3_238) __extension__ ({ \
	37757	float16x4_t __ret_238; \
	37758	float16x4_t __s0_238 = __p0_238; \
	37759	float16x4_t __s1_238 = __p1_238; \
	37760	float16x8_t __s2_238 = __p2_238; \
	37761	float16x4_t __rev0_238; __rev0_238 = __builtin_shufflevector(__s0_238, __s0_238, 3, 2, 1, 0); \
	37762	float16x4_t __rev1_238; __rev1_238 = __builtin_shufflevector(__s1_238, __s1_238, 3, 2, 1, 0); \
	37763	float16x8_t __rev2_238; __rev2_238 = __builtin_shufflevector(__s2_238, __s2_238, 7, 6, 5, 4, 3, 2, 1, 0); \
	37764	float16x8_t __reint_238 = __rev2_238; \
	37765	uint32x2_t __reint1_238 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_238, __p3_238), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_238, __p3_238)}; \
	37766	__ret_238 = __noswap_vcmla_rot90_f16(__rev0_238, __rev1_238, (float16x4_t ) &__reint1_238); \
	37767	__ret_238 = __builtin_shufflevector(__ret_238, __ret_238, 3, 2, 1, 0); \
	37768	__ret_238; \
36360	37769	})
36361	37770	#endif
36362	37771
36363	37772	#ifdef __LITTLE_ENDIAN__
36364		#define vsm3tt2bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36365		uint32x4_t __ret; \
36366		uint32x4_t __s0 = __p0; \
36367		uint32x4_t __s1 = __p1; \
36368		uint32x4_t __s2 = __p2; \
36369		__ret = (uint32x4_t) __builtin_neon_vsm3tt2bq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
36370		__ret; \
	37773	#define vcmlaq_rot90_laneq_f16(__p0_239, __p1_239, __p2_239, __p3_239) __extension__ ({ \
	37774	float16x8_t __ret_239; \
	37775	float16x8_t __s0_239 = __p0_239; \
	37776	float16x8_t __s1_239 = __p1_239; \
	37777	float16x8_t __s2_239 = __p2_239; \
	37778	float16x8_t __reint_239 = __s2_239; \
	37779	uint32x4_t __reint1_239 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_239, __p3_239), vgetq_lane_u32((uint32x4_t ) &__reint_239, __p3_239), vgetq_lane_u32((uint32x4_t ) &__reint_239, __p3_239), vgetq_lane_u32((uint32x4_t ) &__reint_239, __p3_239)}; \
	37780	__ret_239 = vcmlaq_rot90_f16(__s0_239, __s1_239, (float16x8_t ) &__reint1_239); \
	37781	__ret_239; \
36371	37782	})
36372	37783	#else
36373		#define vsm3tt2bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36374		uint32x4_t __ret; \
36375		uint32x4_t __s0 = __p0; \
36376		uint32x4_t __s1 = __p1; \
36377		uint32x4_t __s2 = __p2; \
36378		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
36379		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
36380		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36381		__ret = (uint32x4_t) __builtin_neon_vsm3tt2bq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
36382		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
36383		__ret; \
	37784	#define vcmlaq_rot90_laneq_f16(__p0_240, __p1_240, __p2_240, __p3_240) __extension__ ({ \
	37785	float16x8_t __ret_240; \
	37786	float16x8_t __s0_240 = __p0_240; \
	37787	float16x8_t __s1_240 = __p1_240; \
	37788	float16x8_t __s2_240 = __p2_240; \
	37789	float16x8_t __rev0_240; __rev0_240 = __builtin_shufflevector(__s0_240, __s0_240, 7, 6, 5, 4, 3, 2, 1, 0); \
	37790	float16x8_t __rev1_240; __rev1_240 = __builtin_shufflevector(__s1_240, __s1_240, 7, 6, 5, 4, 3, 2, 1, 0); \
	37791	float16x8_t __rev2_240; __rev2_240 = __builtin_shufflevector(__s2_240, __s2_240, 7, 6, 5, 4, 3, 2, 1, 0); \
	37792	float16x8_t __reint_240 = __rev2_240; \
	37793	uint32x4_t __reint1_240 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_240, __p3_240), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_240, __p3_240), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_240, __p3_240), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_240, __p3_240)}; \
	37794	__ret_240 = __noswap_vcmlaq_rot90_f16(__rev0_240, __rev1_240, (float16x8_t ) &__reint1_240); \
	37795	__ret_240 = __builtin_shufflevector(__ret_240, __ret_240, 7, 6, 5, 4, 3, 2, 1, 0); \
	37796	__ret_240; \
36384	37797	})
36385	37798	#endif
36386	37799
36387		#endif
36388		#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_SM4) && defined(__aarch64__)
36389		#ifdef __LITTLE_ENDIAN__
36390		__ai uint32x4_t vsm4eq_u32(uint32x4_t __p0, uint32x4_t __p1) {
36391		uint32x4_t __ret;
36392		__ret = (uint32x4_t) __builtin_neon_vsm4eq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
36393		return __ret;
36394		}
36395		#else
36396		__ai uint32x4_t vsm4eq_u32(uint32x4_t __p0, uint32x4_t __p1) {
36397		uint32x4_t __ret;
36398		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36399		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36400		__ret = (uint32x4_t) __builtin_neon_vsm4eq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
36401		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36402		return __ret;
36403		}
36404		#endif
36405
36406		#ifdef __LITTLE_ENDIAN__
36407		__ai uint32x4_t vsm4ekeyq_u32(uint32x4_t __p0, uint32x4_t __p1) {
36408		uint32x4_t __ret;
36409		__ret = (uint32x4_t) __builtin_neon_vsm4ekeyq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
36410		return __ret;
36411		}
36412		#else
36413		__ai uint32x4_t vsm4ekeyq_u32(uint32x4_t __p0, uint32x4_t __p1) {
36414		uint32x4_t __ret;
36415		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36416		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36417		__ret = (uint32x4_t) __builtin_neon_vsm4ekeyq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
36418		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36419		return __ret;
36420		}
36421		#endif
36422
36423		#endif
36424		#if __ARM_ARCH >= 8 && defined(__aarch64__)
	37800	#if !defined(__aarch64__)
36425	37801	#ifdef __LITTLE_ENDIAN__
36426		__ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
36427		int64x2_t __ret;
36428		__ret = (int64x2_t) __builtin_neon_vcvtaq_s64_v((int8x16_t)__p0, 35);
36429		return __ret;
36430		}
	37802	#define vqdmulhq_lane_s32(__p0_241, __p1_241, __p2_241) __extension__ ({ \
	37803	int32x4_t __ret_241; \
	37804	int32x4_t __s0_241 = __p0_241; \
	37805	int32x2_t __s1_241 = __p1_241; \
	37806	__ret_241 = vqdmulhq_s32(__s0_241, splatq_lane_s32(__s1_241, __p2_241)); \
	37807	__ret_241; \
	37808	})
36431	37809	#else
36432		__ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
36433		int64x2_t __ret;
36434		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36435		__ret = (int64x2_t) __builtin_neon_vcvtaq_s64_v((int8x16_t)__rev0, 35);
36436		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36437		return __ret;
36438		}
	37810	#define vqdmulhq_lane_s32(__p0_242, __p1_242, __p2_242) __extension__ ({ \
	37811	int32x4_t __ret_242; \
	37812	int32x4_t __s0_242 = __p0_242; \
	37813	int32x2_t __s1_242 = __p1_242; \
	37814	int32x4_t __rev0_242; __rev0_242 = __builtin_shufflevector(__s0_242, __s0_242, 3, 2, 1, 0); \
	37815	int32x2_t __rev1_242; __rev1_242 = __builtin_shufflevector(__s1_242, __s1_242, 1, 0); \
	37816	__ret_242 = __noswap_vqdmulhq_s32(__rev0_242, __noswap_splatq_lane_s32(__rev1_242, __p2_242)); \
	37817	__ret_242 = __builtin_shufflevector(__ret_242, __ret_242, 3, 2, 1, 0); \
	37818	__ret_242; \
	37819	})
36439	37820	#endif
36440	37821
36441		__ai int64x1_t vcvta_s64_f64(float64x1_t __p0) {
36442		int64x1_t __ret;
36443		__ret = (int64x1_t) __builtin_neon_vcvta_s64_v((int8x8_t)__p0, 3);
36444		return __ret;
36445		}
36446	37822	#ifdef __LITTLE_ENDIAN__
36447		__ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
36448		uint64x2_t __ret;
36449		__ret = (uint64x2_t) __builtin_neon_vcvtaq_u64_v((int8x16_t)__p0, 51);
36450		return __ret;
36451		}
	37823	#define vqdmulhq_lane_s16(__p0_243, __p1_243, __p2_243) __extension__ ({ \
	37824	int16x8_t __ret_243; \
	37825	int16x8_t __s0_243 = __p0_243; \
	37826	int16x4_t __s1_243 = __p1_243; \
	37827	__ret_243 = vqdmulhq_s16(__s0_243, splatq_lane_s16(__s1_243, __p2_243)); \
	37828	__ret_243; \
	37829	})
36452	37830	#else
36453		__ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
36454		uint64x2_t __ret;
36455		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36456		__ret = (uint64x2_t) __builtin_neon_vcvtaq_u64_v((int8x16_t)__rev0, 51);
36457		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36458		return __ret;
36459		}
	37831	#define vqdmulhq_lane_s16(__p0_244, __p1_244, __p2_244) __extension__ ({ \
	37832	int16x8_t __ret_244; \
	37833	int16x8_t __s0_244 = __p0_244; \
	37834	int16x4_t __s1_244 = __p1_244; \
	37835	int16x8_t __rev0_244; __rev0_244 = __builtin_shufflevector(__s0_244, __s0_244, 7, 6, 5, 4, 3, 2, 1, 0); \
	37836	int16x4_t __rev1_244; __rev1_244 = __builtin_shufflevector(__s1_244, __s1_244, 3, 2, 1, 0); \
	37837	__ret_244 = __noswap_vqdmulhq_s16(__rev0_244, __noswap_splatq_lane_s16(__rev1_244, __p2_244)); \
	37838	__ret_244 = __builtin_shufflevector(__ret_244, __ret_244, 7, 6, 5, 4, 3, 2, 1, 0); \
	37839	__ret_244; \
	37840	})
36460	37841	#endif
36461	37842
36462		__ai uint64x1_t vcvta_u64_f64(float64x1_t __p0) {
36463		uint64x1_t __ret;
36464		__ret = (uint64x1_t) __builtin_neon_vcvta_u64_v((int8x8_t)__p0, 19);
36465		return __ret;
36466		}
36467	37843	#ifdef __LITTLE_ENDIAN__
36468		__ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
36469		int64x2_t __ret;
36470		__ret = (int64x2_t) __builtin_neon_vcvtmq_s64_v((int8x16_t)__p0, 35);
36471		return __ret;
36472		}
	37844	#define vqdmulh_lane_s32(__p0_245, __p1_245, __p2_245) __extension__ ({ \
	37845	int32x2_t __ret_245; \
	37846	int32x2_t __s0_245 = __p0_245; \
	37847	int32x2_t __s1_245 = __p1_245; \
	37848	__ret_245 = vqdmulh_s32(__s0_245, splat_lane_s32(__s1_245, __p2_245)); \
	37849	__ret_245; \
	37850	})
36473	37851	#else
36474		__ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
36475		int64x2_t __ret;
36476		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36477		__ret = (int64x2_t) __builtin_neon_vcvtmq_s64_v((int8x16_t)__rev0, 35);
36478		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36479		return __ret;
36480		}
	37852	#define vqdmulh_lane_s32(__p0_246, __p1_246, __p2_246) __extension__ ({ \
	37853	int32x2_t __ret_246; \
	37854	int32x2_t __s0_246 = __p0_246; \
	37855	int32x2_t __s1_246 = __p1_246; \
	37856	int32x2_t __rev0_246; __rev0_246 = __builtin_shufflevector(__s0_246, __s0_246, 1, 0); \
	37857	int32x2_t __rev1_246; __rev1_246 = __builtin_shufflevector(__s1_246, __s1_246, 1, 0); \
	37858	__ret_246 = __noswap_vqdmulh_s32(__rev0_246, __noswap_splat_lane_s32(__rev1_246, __p2_246)); \
	37859	__ret_246 = __builtin_shufflevector(__ret_246, __ret_246, 1, 0); \
	37860	__ret_246; \
	37861	})
36481	37862	#endif
36482	37863
36483		__ai int64x1_t vcvtm_s64_f64(float64x1_t __p0) {
36484		int64x1_t __ret;
36485		__ret = (int64x1_t) __builtin_neon_vcvtm_s64_v((int8x8_t)__p0, 3);
36486		return __ret;
36487		}
36488	37864	#ifdef __LITTLE_ENDIAN__
36489		__ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
36490		uint64x2_t __ret;
36491		__ret = (uint64x2_t) __builtin_neon_vcvtmq_u64_v((int8x16_t)__p0, 51);
36492		return __ret;
36493		}
	37865	#define vqdmulh_lane_s16(__p0_247, __p1_247, __p2_247) __extension__ ({ \
	37866	int16x4_t __ret_247; \
	37867	int16x4_t __s0_247 = __p0_247; \
	37868	int16x4_t __s1_247 = __p1_247; \
	37869	__ret_247 = vqdmulh_s16(__s0_247, splat_lane_s16(__s1_247, __p2_247)); \
	37870	__ret_247; \
	37871	})
36494	37872	#else
36495		__ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
36496		uint64x2_t __ret;
36497		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36498		__ret = (uint64x2_t) __builtin_neon_vcvtmq_u64_v((int8x16_t)__rev0, 51);
36499		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36500		return __ret;
36501		}
	37873	#define vqdmulh_lane_s16(__p0_248, __p1_248, __p2_248) __extension__ ({ \
	37874	int16x4_t __ret_248; \
	37875	int16x4_t __s0_248 = __p0_248; \
	37876	int16x4_t __s1_248 = __p1_248; \
	37877	int16x4_t __rev0_248; __rev0_248 = __builtin_shufflevector(__s0_248, __s0_248, 3, 2, 1, 0); \
	37878	int16x4_t __rev1_248; __rev1_248 = __builtin_shufflevector(__s1_248, __s1_248, 3, 2, 1, 0); \
	37879	__ret_248 = __noswap_vqdmulh_s16(__rev0_248, __noswap_splat_lane_s16(__rev1_248, __p2_248)); \
	37880	__ret_248 = __builtin_shufflevector(__ret_248, __ret_248, 3, 2, 1, 0); \
	37881	__ret_248; \
	37882	})
36502	37883	#endif
36503	37884
36504		__ai uint64x1_t vcvtm_u64_f64(float64x1_t __p0) {
36505		uint64x1_t __ret;
36506		__ret = (uint64x1_t) __builtin_neon_vcvtm_u64_v((int8x8_t)__p0, 19);
36507		return __ret;
36508		}
36509	37885	#ifdef __LITTLE_ENDIAN__
36510		__ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
36511		int64x2_t __ret;
36512		__ret = (int64x2_t) __builtin_neon_vcvtnq_s64_v((int8x16_t)__p0, 35);
36513		return __ret;
36514		}
	37886	#define vqrdmulhq_lane_s32(__p0_249, __p1_249, __p2_249) __extension__ ({ \
	37887	int32x4_t __ret_249; \
	37888	int32x4_t __s0_249 = __p0_249; \
	37889	int32x2_t __s1_249 = __p1_249; \
	37890	__ret_249 = vqrdmulhq_s32(__s0_249, splatq_lane_s32(__s1_249, __p2_249)); \
	37891	__ret_249; \
	37892	})
36515	37893	#else
36516		__ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
36517		int64x2_t __ret;
36518		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36519		__ret = (int64x2_t) __builtin_neon_vcvtnq_s64_v((int8x16_t)__rev0, 35);
36520		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36521		return __ret;
36522		}
	37894	#define vqrdmulhq_lane_s32(__p0_250, __p1_250, __p2_250) __extension__ ({ \
	37895	int32x4_t __ret_250; \
	37896	int32x4_t __s0_250 = __p0_250; \
	37897	int32x2_t __s1_250 = __p1_250; \
	37898	int32x4_t __rev0_250; __rev0_250 = __builtin_shufflevector(__s0_250, __s0_250, 3, 2, 1, 0); \
	37899	int32x2_t __rev1_250; __rev1_250 = __builtin_shufflevector(__s1_250, __s1_250, 1, 0); \
	37900	__ret_250 = __noswap_vqrdmulhq_s32(__rev0_250, __noswap_splatq_lane_s32(__rev1_250, __p2_250)); \
	37901	__ret_250 = __builtin_shufflevector(__ret_250, __ret_250, 3, 2, 1, 0); \
	37902	__ret_250; \
	37903	})
36523	37904	#endif
36524	37905
36525		__ai int64x1_t vcvtn_s64_f64(float64x1_t __p0) {
36526		int64x1_t __ret;
36527		__ret = (int64x1_t) __builtin_neon_vcvtn_s64_v((int8x8_t)__p0, 3);
36528		return __ret;
36529		}
36530	37906	#ifdef __LITTLE_ENDIAN__
36531		__ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
36532		uint64x2_t __ret;
36533		__ret = (uint64x2_t) __builtin_neon_vcvtnq_u64_v((int8x16_t)__p0, 51);
36534		return __ret;
36535		}
	37907	#define vqrdmulhq_lane_s16(__p0_251, __p1_251, __p2_251) __extension__ ({ \
	37908	int16x8_t __ret_251; \
	37909	int16x8_t __s0_251 = __p0_251; \
	37910	int16x4_t __s1_251 = __p1_251; \
	37911	__ret_251 = vqrdmulhq_s16(__s0_251, splatq_lane_s16(__s1_251, __p2_251)); \
	37912	__ret_251; \
	37913	})
36536	37914	#else
36537		__ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
36538		uint64x2_t __ret;
36539		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36540		__ret = (uint64x2_t) __builtin_neon_vcvtnq_u64_v((int8x16_t)__rev0, 51);
36541		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36542		return __ret;
36543		}
	37915	#define vqrdmulhq_lane_s16(__p0_252, __p1_252, __p2_252) __extension__ ({ \
	37916	int16x8_t __ret_252; \
	37917	int16x8_t __s0_252 = __p0_252; \
	37918	int16x4_t __s1_252 = __p1_252; \
	37919	int16x8_t __rev0_252; __rev0_252 = __builtin_shufflevector(__s0_252, __s0_252, 7, 6, 5, 4, 3, 2, 1, 0); \
	37920	int16x4_t __rev1_252; __rev1_252 = __builtin_shufflevector(__s1_252, __s1_252, 3, 2, 1, 0); \
	37921	__ret_252 = __noswap_vqrdmulhq_s16(__rev0_252, __noswap_splatq_lane_s16(__rev1_252, __p2_252)); \
	37922	__ret_252 = __builtin_shufflevector(__ret_252, __ret_252, 7, 6, 5, 4, 3, 2, 1, 0); \
	37923	__ret_252; \
	37924	})
36544	37925	#endif
36545	37926
36546		__ai uint64x1_t vcvtn_u64_f64(float64x1_t __p0) {
36547		uint64x1_t __ret;
36548		__ret = (uint64x1_t) __builtin_neon_vcvtn_u64_v((int8x8_t)__p0, 19);
36549		return __ret;
36550		}
36551	37927	#ifdef __LITTLE_ENDIAN__
36552		__ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
36553		int64x2_t __ret;
36554		__ret = (int64x2_t) __builtin_neon_vcvtpq_s64_v((int8x16_t)__p0, 35);
36555		return __ret;
36556		}
	37928	#define vqrdmulh_lane_s32(__p0_253, __p1_253, __p2_253) __extension__ ({ \
	37929	int32x2_t __ret_253; \
	37930	int32x2_t __s0_253 = __p0_253; \
	37931	int32x2_t __s1_253 = __p1_253; \
	37932	__ret_253 = vqrdmulh_s32(__s0_253, splat_lane_s32(__s1_253, __p2_253)); \
	37933	__ret_253; \
	37934	})
36557	37935	#else
36558		__ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
36559		int64x2_t __ret;
36560		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36561		__ret = (int64x2_t) __builtin_neon_vcvtpq_s64_v((int8x16_t)__rev0, 35);
36562		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36563		return __ret;
36564		}
	37936	#define vqrdmulh_lane_s32(__p0_254, __p1_254, __p2_254) __extension__ ({ \
	37937	int32x2_t __ret_254; \
	37938	int32x2_t __s0_254 = __p0_254; \
	37939	int32x2_t __s1_254 = __p1_254; \
	37940	int32x2_t __rev0_254; __rev0_254 = __builtin_shufflevector(__s0_254, __s0_254, 1, 0); \
	37941	int32x2_t __rev1_254; __rev1_254 = __builtin_shufflevector(__s1_254, __s1_254, 1, 0); \
	37942	__ret_254 = __noswap_vqrdmulh_s32(__rev0_254, __noswap_splat_lane_s32(__rev1_254, __p2_254)); \
	37943	__ret_254 = __builtin_shufflevector(__ret_254, __ret_254, 1, 0); \
	37944	__ret_254; \
	37945	})
36565	37946	#endif
36566	37947
36567		__ai int64x1_t vcvtp_s64_f64(float64x1_t __p0) {
36568		int64x1_t __ret;
36569		__ret = (int64x1_t) __builtin_neon_vcvtp_s64_v((int8x8_t)__p0, 3);
36570		return __ret;
36571		}
36572	37948	#ifdef __LITTLE_ENDIAN__
36573		__ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
36574		uint64x2_t __ret;
36575		__ret = (uint64x2_t) __builtin_neon_vcvtpq_u64_v((int8x16_t)__p0, 51);
36576		return __ret;
36577		}
	37949	#define vqrdmulh_lane_s16(__p0_255, __p1_255, __p2_255) __extension__ ({ \
	37950	int16x4_t __ret_255; \
	37951	int16x4_t __s0_255 = __p0_255; \
	37952	int16x4_t __s1_255 = __p1_255; \
	37953	__ret_255 = vqrdmulh_s16(__s0_255, splat_lane_s16(__s1_255, __p2_255)); \
	37954	__ret_255; \
	37955	})
36578	37956	#else
36579		__ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
36580		uint64x2_t __ret;
36581		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36582		__ret = (uint64x2_t) __builtin_neon_vcvtpq_u64_v((int8x16_t)__rev0, 51);
36583		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36584		return __ret;
36585		}
	37957	#define vqrdmulh_lane_s16(__p0_256, __p1_256, __p2_256) __extension__ ({ \
	37958	int16x4_t __ret_256; \
	37959	int16x4_t __s0_256 = __p0_256; \
	37960	int16x4_t __s1_256 = __p1_256; \
	37961	int16x4_t __rev0_256; __rev0_256 = __builtin_shufflevector(__s0_256, __s0_256, 3, 2, 1, 0); \
	37962	int16x4_t __rev1_256; __rev1_256 = __builtin_shufflevector(__s1_256, __s1_256, 3, 2, 1, 0); \
	37963	__ret_256 = __noswap_vqrdmulh_s16(__rev0_256, __noswap_splat_lane_s16(__rev1_256, __p2_256)); \
	37964	__ret_256 = __builtin_shufflevector(__ret_256, __ret_256, 3, 2, 1, 0); \
	37965	__ret_256; \
	37966	})
36586	37967	#endif
36587	37968
36588		__ai uint64x1_t vcvtp_u64_f64(float64x1_t __p0) {
36589		uint64x1_t __ret;
36590		__ret = (uint64x1_t) __builtin_neon_vcvtp_u64_v((int8x8_t)__p0, 19);
36591		return __ret;
36592		}
36593		__ai poly8x8_t vreinterpret_p8_p64(poly64x1_t __p0) {
36594		poly8x8_t __ret;
36595		__ret = (poly8x8_t)(__p0);
36596		return __ret;
36597		}
36598	37969	__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
36599	37970	poly8x8_t __ret;
36600	37971	__ret = (poly8x8_t)(__p0);
...	...	@@ -36625,11 +37996,6 @@ __ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
36625	37996	__ret = (poly8x8_t)(__p0);
36626	37997	return __ret;
36627	37998	}
36628		__ai poly8x8_t vreinterpret_p8_f64(float64x1_t __p0) {
36629		poly8x8_t __ret;
36630		__ret = (poly8x8_t)(__p0);
36631		return __ret;
36632		}
36633	37999	__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
36634	38000	poly8x8_t __ret;
36635	38001	__ret = (poly8x8_t)(__p0);
...	...	@@ -36655,81 +38021,11 @@ __ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
36655	38021	__ret = (poly8x8_t)(__p0);
36656	38022	return __ret;
36657	38023	}
36658		__ai poly64x1_t vreinterpret_p64_p8(poly8x8_t __p0) {
36659		poly64x1_t __ret;
36660		__ret = (poly64x1_t)(__p0);
36661		return __ret;
36662		}
36663		__ai poly64x1_t vreinterpret_p64_p16(poly16x4_t __p0) {
36664		poly64x1_t __ret;
36665		__ret = (poly64x1_t)(__p0);
36666		return __ret;
36667		}
36668		__ai poly64x1_t vreinterpret_p64_u8(uint8x8_t __p0) {
36669		poly64x1_t __ret;
36670		__ret = (poly64x1_t)(__p0);
36671		return __ret;
36672		}
36673		__ai poly64x1_t vreinterpret_p64_u32(uint32x2_t __p0) {
36674		poly64x1_t __ret;
36675		__ret = (poly64x1_t)(__p0);
36676		return __ret;
36677		}
36678		__ai poly64x1_t vreinterpret_p64_u64(uint64x1_t __p0) {
36679		poly64x1_t __ret;
36680		__ret = (poly64x1_t)(__p0);
36681		return __ret;
36682		}
36683		__ai poly64x1_t vreinterpret_p64_u16(uint16x4_t __p0) {
36684		poly64x1_t __ret;
36685		__ret = (poly64x1_t)(__p0);
36686		return __ret;
36687		}
36688		__ai poly64x1_t vreinterpret_p64_s8(int8x8_t __p0) {
36689		poly64x1_t __ret;
36690		__ret = (poly64x1_t)(__p0);
36691		return __ret;
36692		}
36693		__ai poly64x1_t vreinterpret_p64_f64(float64x1_t __p0) {
36694		poly64x1_t __ret;
36695		__ret = (poly64x1_t)(__p0);
36696		return __ret;
36697		}
36698		__ai poly64x1_t vreinterpret_p64_f32(float32x2_t __p0) {
36699		poly64x1_t __ret;
36700		__ret = (poly64x1_t)(__p0);
36701		return __ret;
36702		}
36703		__ai poly64x1_t vreinterpret_p64_f16(float16x4_t __p0) {
36704		poly64x1_t __ret;
36705		__ret = (poly64x1_t)(__p0);
36706		return __ret;
36707		}
36708		__ai poly64x1_t vreinterpret_p64_s32(int32x2_t __p0) {
36709		poly64x1_t __ret;
36710		__ret = (poly64x1_t)(__p0);
36711		return __ret;
36712		}
36713		__ai poly64x1_t vreinterpret_p64_s64(int64x1_t __p0) {
36714		poly64x1_t __ret;
36715		__ret = (poly64x1_t)(__p0);
36716		return __ret;
36717		}
36718		__ai poly64x1_t vreinterpret_p64_s16(int16x4_t __p0) {
36719		poly64x1_t __ret;
36720		__ret = (poly64x1_t)(__p0);
36721		return __ret;
36722		}
36723	38024	__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
36724	38025	poly16x4_t __ret;
36725	38026	__ret = (poly16x4_t)(__p0);
36726	38027	return __ret;
36727	38028	}
36728		__ai poly16x4_t vreinterpret_p16_p64(poly64x1_t __p0) {
36729		poly16x4_t __ret;
36730		__ret = (poly16x4_t)(__p0);
36731		return __ret;
36732		}
36733	38029	__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
36734	38030	poly16x4_t __ret;
36735	38031	__ret = (poly16x4_t)(__p0);
...	...	@@ -36755,11 +38051,6 @@ __ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
36755	38051	__ret = (poly16x4_t)(__p0);
36756	38052	return __ret;
36757	38053	}
36758		__ai poly16x4_t vreinterpret_p16_f64(float64x1_t __p0) {
36759		poly16x4_t __ret;
36760		__ret = (poly16x4_t)(__p0);
36761		return __ret;
36762		}
36763	38054	__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
36764	38055	poly16x4_t __ret;
36765	38056	__ret = (poly16x4_t)(__p0);
...	...	@@ -36785,16 +38076,6 @@ __ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
36785	38076	__ret = (poly16x4_t)(__p0);
36786	38077	return __ret;
36787	38078	}
36788		__ai poly8x16_t vreinterpretq_p8_p128(poly128_t __p0) {
36789		poly8x16_t __ret;
36790		__ret = (poly8x16_t)(__p0);
36791		return __ret;
36792		}
36793		__ai poly8x16_t vreinterpretq_p8_p64(poly64x2_t __p0) {
36794		poly8x16_t __ret;
36795		__ret = (poly8x16_t)(__p0);
36796		return __ret;
36797		}
36798	38079	__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
36799	38080	poly8x16_t __ret;
36800	38081	__ret = (poly8x16_t)(__p0);
...	...	@@ -36825,11 +38106,6 @@ __ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
36825	38106	__ret = (poly8x16_t)(__p0);
36826	38107	return __ret;
36827	38108	}
36828		__ai poly8x16_t vreinterpretq_p8_f64(float64x2_t __p0) {
36829		poly8x16_t __ret;
36830		__ret = (poly8x16_t)(__p0);
36831		return __ret;
36832		}
36833	38109	__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
36834	38110	poly8x16_t __ret;
36835	38111	__ret = (poly8x16_t)(__p0);
...	...	@@ -36855,161 +38131,11 @@ __ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
36855	38131	__ret = (poly8x16_t)(__p0);
36856	38132	return __ret;
36857	38133	}
36858		__ai poly128_t vreinterpretq_p128_p8(poly8x16_t __p0) {
36859		poly128_t __ret;
36860		__ret = (poly128_t)(__p0);
36861		return __ret;
36862		}
36863		__ai poly128_t vreinterpretq_p128_p64(poly64x2_t __p0) {
36864		poly128_t __ret;
36865		__ret = (poly128_t)(__p0);
36866		return __ret;
36867		}
36868		__ai poly128_t vreinterpretq_p128_p16(poly16x8_t __p0) {
36869		poly128_t __ret;
36870		__ret = (poly128_t)(__p0);
36871		return __ret;
36872		}
36873		__ai poly128_t vreinterpretq_p128_u8(uint8x16_t __p0) {
36874		poly128_t __ret;
36875		__ret = (poly128_t)(__p0);
36876		return __ret;
36877		}
36878		__ai poly128_t vreinterpretq_p128_u32(uint32x4_t __p0) {
36879		poly128_t __ret;
36880		__ret = (poly128_t)(__p0);
36881		return __ret;
36882		}
36883		__ai poly128_t vreinterpretq_p128_u64(uint64x2_t __p0) {
36884		poly128_t __ret;
36885		__ret = (poly128_t)(__p0);
36886		return __ret;
36887		}
36888		__ai poly128_t vreinterpretq_p128_u16(uint16x8_t __p0) {
36889		poly128_t __ret;
36890		__ret = (poly128_t)(__p0);
36891		return __ret;
36892		}
36893		__ai poly128_t vreinterpretq_p128_s8(int8x16_t __p0) {
36894		poly128_t __ret;
36895		__ret = (poly128_t)(__p0);
36896		return __ret;
36897		}
36898		__ai poly128_t vreinterpretq_p128_f64(float64x2_t __p0) {
36899		poly128_t __ret;
36900		__ret = (poly128_t)(__p0);
36901		return __ret;
36902		}
36903		__ai poly128_t vreinterpretq_p128_f32(float32x4_t __p0) {
36904		poly128_t __ret;
36905		__ret = (poly128_t)(__p0);
36906		return __ret;
36907		}
36908		__ai poly128_t vreinterpretq_p128_f16(float16x8_t __p0) {
36909		poly128_t __ret;
36910		__ret = (poly128_t)(__p0);
36911		return __ret;
36912		}
36913		__ai poly128_t vreinterpretq_p128_s32(int32x4_t __p0) {
36914		poly128_t __ret;
36915		__ret = (poly128_t)(__p0);
36916		return __ret;
36917		}
36918		__ai poly128_t vreinterpretq_p128_s64(int64x2_t __p0) {
36919		poly128_t __ret;
36920		__ret = (poly128_t)(__p0);
36921		return __ret;
36922		}
36923		__ai poly128_t vreinterpretq_p128_s16(int16x8_t __p0) {
36924		poly128_t __ret;
36925		__ret = (poly128_t)(__p0);
36926		return __ret;
36927		}
36928		__ai poly64x2_t vreinterpretq_p64_p8(poly8x16_t __p0) {
36929		poly64x2_t __ret;
36930		__ret = (poly64x2_t)(__p0);
36931		return __ret;
36932		}
36933		__ai poly64x2_t vreinterpretq_p64_p128(poly128_t __p0) {
36934		poly64x2_t __ret;
36935		__ret = (poly64x2_t)(__p0);
36936		return __ret;
36937		}
36938		__ai poly64x2_t vreinterpretq_p64_p16(poly16x8_t __p0) {
36939		poly64x2_t __ret;
36940		__ret = (poly64x2_t)(__p0);
36941		return __ret;
36942		}
36943		__ai poly64x2_t vreinterpretq_p64_u8(uint8x16_t __p0) {
36944		poly64x2_t __ret;
36945		__ret = (poly64x2_t)(__p0);
36946		return __ret;
36947		}
36948		__ai poly64x2_t vreinterpretq_p64_u32(uint32x4_t __p0) {
36949		poly64x2_t __ret;
36950		__ret = (poly64x2_t)(__p0);
36951		return __ret;
36952		}
36953		__ai poly64x2_t vreinterpretq_p64_u64(uint64x2_t __p0) {
36954		poly64x2_t __ret;
36955		__ret = (poly64x2_t)(__p0);
36956		return __ret;
36957		}
36958		__ai poly64x2_t vreinterpretq_p64_u16(uint16x8_t __p0) {
36959		poly64x2_t __ret;
36960		__ret = (poly64x2_t)(__p0);
36961		return __ret;
36962		}
36963		__ai poly64x2_t vreinterpretq_p64_s8(int8x16_t __p0) {
36964		poly64x2_t __ret;
36965		__ret = (poly64x2_t)(__p0);
36966		return __ret;
36967		}
36968		__ai poly64x2_t vreinterpretq_p64_f64(float64x2_t __p0) {
36969		poly64x2_t __ret;
36970		__ret = (poly64x2_t)(__p0);
36971		return __ret;
36972		}
36973		__ai poly64x2_t vreinterpretq_p64_f32(float32x4_t __p0) {
36974		poly64x2_t __ret;
36975		__ret = (poly64x2_t)(__p0);
36976		return __ret;
36977		}
36978		__ai poly64x2_t vreinterpretq_p64_f16(float16x8_t __p0) {
36979		poly64x2_t __ret;
36980		__ret = (poly64x2_t)(__p0);
36981		return __ret;
36982		}
36983		__ai poly64x2_t vreinterpretq_p64_s32(int32x4_t __p0) {
36984		poly64x2_t __ret;
36985		__ret = (poly64x2_t)(__p0);
36986		return __ret;
36987		}
36988		__ai poly64x2_t vreinterpretq_p64_s64(int64x2_t __p0) {
36989		poly64x2_t __ret;
36990		__ret = (poly64x2_t)(__p0);
36991		return __ret;
36992		}
36993		__ai poly64x2_t vreinterpretq_p64_s16(int16x8_t __p0) {
36994		poly64x2_t __ret;
36995		__ret = (poly64x2_t)(__p0);
36996		return __ret;
36997		}
36998	38134	__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
36999	38135	poly16x8_t __ret;
37000	38136	__ret = (poly16x8_t)(__p0);
37001	38137	return __ret;
37002	38138	}
37003		__ai poly16x8_t vreinterpretq_p16_p128(poly128_t __p0) {
37004		poly16x8_t __ret;
37005		__ret = (poly16x8_t)(__p0);
37006		return __ret;
37007		}
37008		__ai poly16x8_t vreinterpretq_p16_p64(poly64x2_t __p0) {
37009		poly16x8_t __ret;
37010		__ret = (poly16x8_t)(__p0);
37011		return __ret;
37012		}
37013	38139	__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
37014	38140	poly16x8_t __ret;
37015	38141	__ret = (poly16x8_t)(__p0);
...	...	@@ -37035,11 +38161,6 @@ __ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
37035	38161	__ret = (poly16x8_t)(__p0);
37036	38162	return __ret;
37037	38163	}
37038		__ai poly16x8_t vreinterpretq_p16_f64(float64x2_t __p0) {
37039		poly16x8_t __ret;
37040		__ret = (poly16x8_t)(__p0);
37041		return __ret;
37042		}
37043	38164	__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
37044	38165	poly16x8_t __ret;
37045	38166	__ret = (poly16x8_t)(__p0);
...	...	@@ -37070,16 +38191,6 @@ __ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
37070	38191	__ret = (uint8x16_t)(__p0);
37071	38192	return __ret;
37072	38193	}
37073		__ai uint8x16_t vreinterpretq_u8_p128(poly128_t __p0) {
37074		uint8x16_t __ret;
37075		__ret = (uint8x16_t)(__p0);
37076		return __ret;
37077		}
37078		__ai uint8x16_t vreinterpretq_u8_p64(poly64x2_t __p0) {
37079		uint8x16_t __ret;
37080		__ret = (uint8x16_t)(__p0);
37081		return __ret;
37082		}
37083	38194	__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
37084	38195	uint8x16_t __ret;
37085	38196	__ret = (uint8x16_t)(__p0);
...	...	@@ -37105,11 +38216,6 @@ __ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
37105	38216	__ret = (uint8x16_t)(__p0);
37106	38217	return __ret;
37107	38218	}
37108		__ai uint8x16_t vreinterpretq_u8_f64(float64x2_t __p0) {
37109		uint8x16_t __ret;
37110		__ret = (uint8x16_t)(__p0);
37111		return __ret;
37112		}
37113	38219	__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
37114	38220	uint8x16_t __ret;
37115	38221	__ret = (uint8x16_t)(__p0);
...	...	@@ -37140,16 +38246,6 @@ __ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
37140	38246	__ret = (uint32x4_t)(__p0);
37141	38247	return __ret;
37142	38248	}
37143		__ai uint32x4_t vreinterpretq_u32_p128(poly128_t __p0) {
37144		uint32x4_t __ret;
37145		__ret = (uint32x4_t)(__p0);
37146		return __ret;
37147		}
37148		__ai uint32x4_t vreinterpretq_u32_p64(poly64x2_t __p0) {
37149		uint32x4_t __ret;
37150		__ret = (uint32x4_t)(__p0);
37151		return __ret;
37152		}
37153	38249	__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
37154	38250	uint32x4_t __ret;
37155	38251	__ret = (uint32x4_t)(__p0);
...	...	@@ -37175,11 +38271,6 @@ __ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
37175	38271	__ret = (uint32x4_t)(__p0);
37176	38272	return __ret;
37177	38273	}
37178		__ai uint32x4_t vreinterpretq_u32_f64(float64x2_t __p0) {
37179		uint32x4_t __ret;
37180		__ret = (uint32x4_t)(__p0);
37181		return __ret;
37182		}
37183	38274	__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
37184	38275	uint32x4_t __ret;
37185	38276	__ret = (uint32x4_t)(__p0);
...	...	@@ -37210,16 +38301,6 @@ __ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
37210	38301	__ret = (uint64x2_t)(__p0);
37211	38302	return __ret;
37212	38303	}
37213		__ai uint64x2_t vreinterpretq_u64_p128(poly128_t __p0) {
37214		uint64x2_t __ret;
37215		__ret = (uint64x2_t)(__p0);
37216		return __ret;
37217		}
37218		__ai uint64x2_t vreinterpretq_u64_p64(poly64x2_t __p0) {
37219		uint64x2_t __ret;
37220		__ret = (uint64x2_t)(__p0);
37221		return __ret;
37222		}
37223	38304	__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
37224	38305	uint64x2_t __ret;
37225	38306	__ret = (uint64x2_t)(__p0);
...	...	@@ -37245,11 +38326,6 @@ __ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
37245	38326	__ret = (uint64x2_t)(__p0);
37246	38327	return __ret;
37247	38328	}
37248		__ai uint64x2_t vreinterpretq_u64_f64(float64x2_t __p0) {
37249		uint64x2_t __ret;
37250		__ret = (uint64x2_t)(__p0);
37251		return __ret;
37252		}
37253	38329	__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
37254	38330	uint64x2_t __ret;
37255	38331	__ret = (uint64x2_t)(__p0);
...	...	@@ -37280,16 +38356,6 @@ __ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
37280	38356	__ret = (uint16x8_t)(__p0);
37281	38357	return __ret;
37282	38358	}
37283		__ai uint16x8_t vreinterpretq_u16_p128(poly128_t __p0) {
37284		uint16x8_t __ret;
37285		__ret = (uint16x8_t)(__p0);
37286		return __ret;
37287		}
37288		__ai uint16x8_t vreinterpretq_u16_p64(poly64x2_t __p0) {
37289		uint16x8_t __ret;
37290		__ret = (uint16x8_t)(__p0);
37291		return __ret;
37292		}
37293	38359	__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
37294	38360	uint16x8_t __ret;
37295	38361	__ret = (uint16x8_t)(__p0);
...	...	@@ -37315,11 +38381,6 @@ __ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
37315	38381	__ret = (uint16x8_t)(__p0);
37316	38382	return __ret;
37317	38383	}
37318		__ai uint16x8_t vreinterpretq_u16_f64(float64x2_t __p0) {
37319		uint16x8_t __ret;
37320		__ret = (uint16x8_t)(__p0);
37321		return __ret;
37322		}
37323	38384	__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
37324	38385	uint16x8_t __ret;
37325	38386	__ret = (uint16x8_t)(__p0);
...	...	@@ -37350,16 +38411,6 @@ __ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
37350	38411	__ret = (int8x16_t)(__p0);
37351	38412	return __ret;
37352	38413	}
37353		__ai int8x16_t vreinterpretq_s8_p128(poly128_t __p0) {
37354		int8x16_t __ret;
37355		__ret = (int8x16_t)(__p0);
37356		return __ret;
37357		}
37358		__ai int8x16_t vreinterpretq_s8_p64(poly64x2_t __p0) {
37359		int8x16_t __ret;
37360		__ret = (int8x16_t)(__p0);
37361		return __ret;
37362		}
37363	38414	__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
37364	38415	int8x16_t __ret;
37365	38416	__ret = (int8x16_t)(__p0);
...	...	@@ -37385,11 +38436,6 @@ __ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
37385	38436	__ret = (int8x16_t)(__p0);
37386	38437	return __ret;
37387	38438	}
37388		__ai int8x16_t vreinterpretq_s8_f64(float64x2_t __p0) {
37389		int8x16_t __ret;
37390		__ret = (int8x16_t)(__p0);
37391		return __ret;
37392		}
37393	38439	__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
37394	38440	int8x16_t __ret;
37395	38441	__ret = (int8x16_t)(__p0);
...	...	@@ -37399,103 +38445,23 @@ __ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
37399	38445	int8x16_t __ret;
37400	38446	__ret = (int8x16_t)(__p0);
37401	38447	return __ret;
37402		}
37403		__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
37404		int8x16_t __ret;
37405		__ret = (int8x16_t)(__p0);
37406		return __ret;
37407		}
37408		__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
37409		int8x16_t __ret;
37410		__ret = (int8x16_t)(__p0);
37411		return __ret;
37412		}
37413		__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
37414		int8x16_t __ret;
37415		__ret = (int8x16_t)(__p0);
37416		return __ret;
37417		}
37418		__ai float64x2_t vreinterpretq_f64_p8(poly8x16_t __p0) {
37419		float64x2_t __ret;
37420		__ret = (float64x2_t)(__p0);
37421		return __ret;
37422		}
37423		__ai float64x2_t vreinterpretq_f64_p128(poly128_t __p0) {
37424		float64x2_t __ret;
37425		__ret = (float64x2_t)(__p0);
37426		return __ret;
37427		}
37428		__ai float64x2_t vreinterpretq_f64_p64(poly64x2_t __p0) {
37429		float64x2_t __ret;
37430		__ret = (float64x2_t)(__p0);
37431		return __ret;
37432		}
37433		__ai float64x2_t vreinterpretq_f64_p16(poly16x8_t __p0) {
37434		float64x2_t __ret;
37435		__ret = (float64x2_t)(__p0);
37436		return __ret;
37437		}
37438		__ai float64x2_t vreinterpretq_f64_u8(uint8x16_t __p0) {
37439		float64x2_t __ret;
37440		__ret = (float64x2_t)(__p0);
37441		return __ret;
37442		}
37443		__ai float64x2_t vreinterpretq_f64_u32(uint32x4_t __p0) {
37444		float64x2_t __ret;
37445		__ret = (float64x2_t)(__p0);
37446		return __ret;
37447		}
37448		__ai float64x2_t vreinterpretq_f64_u64(uint64x2_t __p0) {
37449		float64x2_t __ret;
37450		__ret = (float64x2_t)(__p0);
37451		return __ret;
37452		}
37453		__ai float64x2_t vreinterpretq_f64_u16(uint16x8_t __p0) {
37454		float64x2_t __ret;
37455		__ret = (float64x2_t)(__p0);
37456		return __ret;
37457		}
37458		__ai float64x2_t vreinterpretq_f64_s8(int8x16_t __p0) {
37459		float64x2_t __ret;
37460		__ret = (float64x2_t)(__p0);
37461		return __ret;
37462		}
37463		__ai float64x2_t vreinterpretq_f64_f32(float32x4_t __p0) {
37464		float64x2_t __ret;
37465		__ret = (float64x2_t)(__p0);
37466		return __ret;
37467		}
37468		__ai float64x2_t vreinterpretq_f64_f16(float16x8_t __p0) {
37469		float64x2_t __ret;
37470		__ret = (float64x2_t)(__p0);
37471		return __ret;
37472		}
37473		__ai float64x2_t vreinterpretq_f64_s32(int32x4_t __p0) {
37474		float64x2_t __ret;
37475		__ret = (float64x2_t)(__p0);
37476		return __ret;
37477		}
37478		__ai float64x2_t vreinterpretq_f64_s64(int64x2_t __p0) {
37479		float64x2_t __ret;
37480		__ret = (float64x2_t)(__p0);
37481		return __ret;
37482		}
37483		__ai float64x2_t vreinterpretq_f64_s16(int16x8_t __p0) {
37484		float64x2_t __ret;
37485		__ret = (float64x2_t)(__p0);
	38448	}
	38449	__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
	38450	int8x16_t __ret;
	38451	__ret = (int8x16_t)(__p0);
37486	38452	return __ret;
37487	38453	}
37488		__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
37489		float32x4_t __ret;
37490		__ret = (float32x4_t)(__p0);
	38454	__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
	38455	int8x16_t __ret;
	38456	__ret = (int8x16_t)(__p0);
37491	38457	return __ret;
37492	38458	}
37493		__ai float32x4_t vreinterpretq_f32_p128(poly128_t __p0) {
37494		float32x4_t __ret;
37495		__ret = (float32x4_t)(__p0);
	38459	__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
	38460	int8x16_t __ret;
	38461	__ret = (int8x16_t)(__p0);
37496	38462	return __ret;
37497	38463	}
37498		__ai float32x4_t vreinterpretq_f32_p64(poly64x2_t __p0) {
	38464	__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
37499	38465	float32x4_t __ret;
37500	38466	__ret = (float32x4_t)(__p0);
37501	38467	return __ret;
...	...	@@ -37530,11 +38496,6 @@ __ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
37530	38496	__ret = (float32x4_t)(__p0);
37531	38497	return __ret;
37532	38498	}
37533		__ai float32x4_t vreinterpretq_f32_f64(float64x2_t __p0) {
37534		float32x4_t __ret;
37535		__ret = (float32x4_t)(__p0);
37536		return __ret;
37537		}
37538	38499	__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
37539	38500	float32x4_t __ret;
37540	38501	__ret = (float32x4_t)(__p0);
...	...	@@ -37560,16 +38521,6 @@ __ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
37560	38521	__ret = (float16x8_t)(__p0);
37561	38522	return __ret;
37562	38523	}
37563		__ai float16x8_t vreinterpretq_f16_p128(poly128_t __p0) {
37564		float16x8_t __ret;
37565		__ret = (float16x8_t)(__p0);
37566		return __ret;
37567		}
37568		__ai float16x8_t vreinterpretq_f16_p64(poly64x2_t __p0) {
37569		float16x8_t __ret;
37570		__ret = (float16x8_t)(__p0);
37571		return __ret;
37572		}
37573	38524	__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
37574	38525	float16x8_t __ret;
37575	38526	__ret = (float16x8_t)(__p0);
...	...	@@ -37600,11 +38551,6 @@ __ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
37600	38551	__ret = (float16x8_t)(__p0);
37601	38552	return __ret;
37602	38553	}
37603		__ai float16x8_t vreinterpretq_f16_f64(float64x2_t __p0) {
37604		float16x8_t __ret;
37605		__ret = (float16x8_t)(__p0);
37606		return __ret;
37607		}
37608	38554	__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
37609	38555	float16x8_t __ret;
37610	38556	__ret = (float16x8_t)(__p0);
...	...	@@ -37630,16 +38576,6 @@ __ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
37630	38576	__ret = (int32x4_t)(__p0);
37631	38577	return __ret;
37632	38578	}
37633		__ai int32x4_t vreinterpretq_s32_p128(poly128_t __p0) {
37634		int32x4_t __ret;
37635		__ret = (int32x4_t)(__p0);
37636		return __ret;
37637		}
37638		__ai int32x4_t vreinterpretq_s32_p64(poly64x2_t __p0) {
37639		int32x4_t __ret;
37640		__ret = (int32x4_t)(__p0);
37641		return __ret;
37642		}
37643	38579	__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
37644	38580	int32x4_t __ret;
37645	38581	__ret = (int32x4_t)(__p0);
...	...	@@ -37670,11 +38606,6 @@ __ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
37670	38606	__ret = (int32x4_t)(__p0);
37671	38607	return __ret;
37672	38608	}
37673		__ai int32x4_t vreinterpretq_s32_f64(float64x2_t __p0) {
37674		int32x4_t __ret;
37675		__ret = (int32x4_t)(__p0);
37676		return __ret;
37677		}
37678	38609	__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
37679	38610	int32x4_t __ret;
37680	38611	__ret = (int32x4_t)(__p0);
...	...	@@ -37700,16 +38631,6 @@ __ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
37700	38631	__ret = (int64x2_t)(__p0);
37701	38632	return __ret;
37702	38633	}
37703		__ai int64x2_t vreinterpretq_s64_p128(poly128_t __p0) {
37704		int64x2_t __ret;
37705		__ret = (int64x2_t)(__p0);
37706		return __ret;
37707		}
37708		__ai int64x2_t vreinterpretq_s64_p64(poly64x2_t __p0) {
37709		int64x2_t __ret;
37710		__ret = (int64x2_t)(__p0);
37711		return __ret;
37712		}
37713	38634	__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
37714	38635	int64x2_t __ret;
37715	38636	__ret = (int64x2_t)(__p0);
...	...	@@ -37740,11 +38661,6 @@ __ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
37740	38661	__ret = (int64x2_t)(__p0);
37741	38662	return __ret;
37742	38663	}
37743		__ai int64x2_t vreinterpretq_s64_f64(float64x2_t __p0) {
37744		int64x2_t __ret;
37745		__ret = (int64x2_t)(__p0);
37746		return __ret;
37747		}
37748	38664	__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
37749	38665	int64x2_t __ret;
37750	38666	__ret = (int64x2_t)(__p0);
...	...	@@ -37770,16 +38686,6 @@ __ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
37770	38686	__ret = (int16x8_t)(__p0);
37771	38687	return __ret;
37772	38688	}
37773		__ai int16x8_t vreinterpretq_s16_p128(poly128_t __p0) {
37774		int16x8_t __ret;
37775		__ret = (int16x8_t)(__p0);
37776		return __ret;
37777		}
37778		__ai int16x8_t vreinterpretq_s16_p64(poly64x2_t __p0) {
37779		int16x8_t __ret;
37780		__ret = (int16x8_t)(__p0);
37781		return __ret;
37782		}
37783	38689	__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
37784	38690	int16x8_t __ret;
37785	38691	__ret = (int16x8_t)(__p0);
...	...	@@ -37810,11 +38716,6 @@ __ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
37810	38716	__ret = (int16x8_t)(__p0);
37811	38717	return __ret;
37812	38718	}
37813		__ai int16x8_t vreinterpretq_s16_f64(float64x2_t __p0) {
37814		int16x8_t __ret;
37815		__ret = (int16x8_t)(__p0);
37816		return __ret;
37817		}
37818	38719	__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
37819	38720	int16x8_t __ret;
37820	38721	__ret = (int16x8_t)(__p0);
...	...	@@ -37840,11 +38741,6 @@ __ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
37840	38741	__ret = (uint8x8_t)(__p0);
37841	38742	return __ret;
37842	38743	}
37843		__ai uint8x8_t vreinterpret_u8_p64(poly64x1_t __p0) {
37844		uint8x8_t __ret;
37845		__ret = (uint8x8_t)(__p0);
37846		return __ret;
37847		}
37848	38744	__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
37849	38745	uint8x8_t __ret;
37850	38746	__ret = (uint8x8_t)(__p0);
...	...	@@ -37870,11 +38766,6 @@ __ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
37870	38766	__ret = (uint8x8_t)(__p0);
37871	38767	return __ret;
37872	38768	}
37873		__ai uint8x8_t vreinterpret_u8_f64(float64x1_t __p0) {
37874		uint8x8_t __ret;
37875		__ret = (uint8x8_t)(__p0);
37876		return __ret;
37877		}
37878	38769	__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
37879	38770	uint8x8_t __ret;
37880	38771	__ret = (uint8x8_t)(__p0);
...	...	@@ -37905,11 +38796,6 @@ __ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
37905	38796	__ret = (uint32x2_t)(__p0);
37906	38797	return __ret;
37907	38798	}
37908		__ai uint32x2_t vreinterpret_u32_p64(poly64x1_t __p0) {
37909		uint32x2_t __ret;
37910		__ret = (uint32x2_t)(__p0);
37911		return __ret;
37912		}
37913	38799	__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
37914	38800	uint32x2_t __ret;
37915	38801	__ret = (uint32x2_t)(__p0);
...	...	@@ -37935,11 +38821,6 @@ __ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
37935	38821	__ret = (uint32x2_t)(__p0);
37936	38822	return __ret;
37937	38823	}
37938		__ai uint32x2_t vreinterpret_u32_f64(float64x1_t __p0) {
37939		uint32x2_t __ret;
37940		__ret = (uint32x2_t)(__p0);
37941		return __ret;
37942		}
37943	38824	__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
37944	38825	uint32x2_t __ret;
37945	38826	__ret = (uint32x2_t)(__p0);
...	...	@@ -37970,11 +38851,6 @@ __ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
37970	38851	__ret = (uint64x1_t)(__p0);
37971	38852	return __ret;
37972	38853	}
37973		__ai uint64x1_t vreinterpret_u64_p64(poly64x1_t __p0) {
37974		uint64x1_t __ret;
37975		__ret = (uint64x1_t)(__p0);
37976		return __ret;
37977		}
37978	38854	__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
37979	38855	uint64x1_t __ret;
37980	38856	__ret = (uint64x1_t)(__p0);
...	...	@@ -38000,11 +38876,6 @@ __ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
38000	38876	__ret = (uint64x1_t)(__p0);
38001	38877	return __ret;
38002	38878	}
38003		__ai uint64x1_t vreinterpret_u64_f64(float64x1_t __p0) {
38004		uint64x1_t __ret;
38005		__ret = (uint64x1_t)(__p0);
38006		return __ret;
38007		}
38008	38879	__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
38009	38880	uint64x1_t __ret;
38010	38881	__ret = (uint64x1_t)(__p0);
...	...	@@ -38035,11 +38906,6 @@ __ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
38035	38906	__ret = (uint16x4_t)(__p0);
38036	38907	return __ret;
38037	38908	}
38038		__ai uint16x4_t vreinterpret_u16_p64(poly64x1_t __p0) {
38039		uint16x4_t __ret;
38040		__ret = (uint16x4_t)(__p0);
38041		return __ret;
38042		}
38043	38909	__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
38044	38910	uint16x4_t __ret;
38045	38911	__ret = (uint16x4_t)(__p0);
...	...	@@ -38065,11 +38931,6 @@ __ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
38065	38931	__ret = (uint16x4_t)(__p0);
38066	38932	return __ret;
38067	38933	}
38068		__ai uint16x4_t vreinterpret_u16_f64(float64x1_t __p0) {
38069		uint16x4_t __ret;
38070		__ret = (uint16x4_t)(__p0);
38071		return __ret;
38072		}
38073	38934	__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
38074	38935	uint16x4_t __ret;
38075	38936	__ret = (uint16x4_t)(__p0);
...	...	@@ -38100,11 +38961,6 @@ __ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
38100	38961	__ret = (int8x8_t)(__p0);
38101	38962	return __ret;
38102	38963	}
38103		__ai int8x8_t vreinterpret_s8_p64(poly64x1_t __p0) {
38104		int8x8_t __ret;
38105		__ret = (int8x8_t)(__p0);
38106		return __ret;
38107		}
38108	38964	__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
38109	38965	int8x8_t __ret;
38110	38966	__ret = (int8x8_t)(__p0);
...	...	@@ -38130,11 +38986,6 @@ __ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
38130	38986	__ret = (int8x8_t)(__p0);
38131	38987	return __ret;
38132	38988	}
38133		__ai int8x8_t vreinterpret_s8_f64(float64x1_t __p0) {
38134		int8x8_t __ret;
38135		__ret = (int8x8_t)(__p0);
38136		return __ret;
38137		}
38138	38989	__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
38139	38990	int8x8_t __ret;
38140	38991	__ret = (int8x8_t)(__p0);
...	...	@@ -38160,81 +39011,11 @@ __ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
38160	39011	__ret = (int8x8_t)(__p0);
38161	39012	return __ret;
38162	39013	}
38163		__ai float64x1_t vreinterpret_f64_p8(poly8x8_t __p0) {
38164		float64x1_t __ret;
38165		__ret = (float64x1_t)(__p0);
38166		return __ret;
38167		}
38168		__ai float64x1_t vreinterpret_f64_p64(poly64x1_t __p0) {
38169		float64x1_t __ret;
38170		__ret = (float64x1_t)(__p0);
38171		return __ret;
38172		}
38173		__ai float64x1_t vreinterpret_f64_p16(poly16x4_t __p0) {
38174		float64x1_t __ret;
38175		__ret = (float64x1_t)(__p0);
38176		return __ret;
38177		}
38178		__ai float64x1_t vreinterpret_f64_u8(uint8x8_t __p0) {
38179		float64x1_t __ret;
38180		__ret = (float64x1_t)(__p0);
38181		return __ret;
38182		}
38183		__ai float64x1_t vreinterpret_f64_u32(uint32x2_t __p0) {
38184		float64x1_t __ret;
38185		__ret = (float64x1_t)(__p0);
38186		return __ret;
38187		}
38188		__ai float64x1_t vreinterpret_f64_u64(uint64x1_t __p0) {
38189		float64x1_t __ret;
38190		__ret = (float64x1_t)(__p0);
38191		return __ret;
38192		}
38193		__ai float64x1_t vreinterpret_f64_u16(uint16x4_t __p0) {
38194		float64x1_t __ret;
38195		__ret = (float64x1_t)(__p0);
38196		return __ret;
38197		}
38198		__ai float64x1_t vreinterpret_f64_s8(int8x8_t __p0) {
38199		float64x1_t __ret;
38200		__ret = (float64x1_t)(__p0);
38201		return __ret;
38202		}
38203		__ai float64x1_t vreinterpret_f64_f32(float32x2_t __p0) {
38204		float64x1_t __ret;
38205		__ret = (float64x1_t)(__p0);
38206		return __ret;
38207		}
38208		__ai float64x1_t vreinterpret_f64_f16(float16x4_t __p0) {
38209		float64x1_t __ret;
38210		__ret = (float64x1_t)(__p0);
38211		return __ret;
38212		}
38213		__ai float64x1_t vreinterpret_f64_s32(int32x2_t __p0) {
38214		float64x1_t __ret;
38215		__ret = (float64x1_t)(__p0);
38216		return __ret;
38217		}
38218		__ai float64x1_t vreinterpret_f64_s64(int64x1_t __p0) {
38219		float64x1_t __ret;
38220		__ret = (float64x1_t)(__p0);
38221		return __ret;
38222		}
38223		__ai float64x1_t vreinterpret_f64_s16(int16x4_t __p0) {
38224		float64x1_t __ret;
38225		__ret = (float64x1_t)(__p0);
38226		return __ret;
38227		}
38228	39014	__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
38229	39015	float32x2_t __ret;
38230	39016	__ret = (float32x2_t)(__p0);
38231	39017	return __ret;
38232	39018	}
38233		__ai float32x2_t vreinterpret_f32_p64(poly64x1_t __p0) {
38234		float32x2_t __ret;
38235		__ret = (float32x2_t)(__p0);
38236		return __ret;
38237		}
38238	39019	__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
38239	39020	float32x2_t __ret;
38240	39021	__ret = (float32x2_t)(__p0);
...	...	@@ -38265,11 +39046,6 @@ __ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
38265	39046	__ret = (float32x2_t)(__p0);
38266	39047	return __ret;
38267	39048	}
38268		__ai float32x2_t vreinterpret_f32_f64(float64x1_t __p0) {
38269		float32x2_t __ret;
38270		__ret = (float32x2_t)(__p0);
38271		return __ret;
38272		}
38273	39049	__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
38274	39050	float32x2_t __ret;
38275	39051	__ret = (float32x2_t)(__p0);
...	...	@@ -38295,11 +39071,6 @@ __ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
38295	39071	__ret = (float16x4_t)(__p0);
38296	39072	return __ret;
38297	39073	}
38298		__ai float16x4_t vreinterpret_f16_p64(poly64x1_t __p0) {
38299		float16x4_t __ret;
38300		__ret = (float16x4_t)(__p0);
38301		return __ret;
38302		}
38303	39074	__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
38304	39075	float16x4_t __ret;
38305	39076	__ret = (float16x4_t)(__p0);
...	...	@@ -38325,28571 +39096,27638 @@ __ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
38325	39096	__ret = (float16x4_t)(__p0);
38326	39097	return __ret;
38327	39098	}
38328		__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
38329		float16x4_t __ret;
38330		__ret = (float16x4_t)(__p0);
	39099	__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
	39100	float16x4_t __ret;
	39101	__ret = (float16x4_t)(__p0);
	39102	return __ret;
	39103	}
	39104	__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
	39105	float16x4_t __ret;
	39106	__ret = (float16x4_t)(__p0);
	39107	return __ret;
	39108	}
	39109	__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
	39110	float16x4_t __ret;
	39111	__ret = (float16x4_t)(__p0);
	39112	return __ret;
	39113	}
	39114	__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
	39115	float16x4_t __ret;
	39116	__ret = (float16x4_t)(__p0);
	39117	return __ret;
	39118	}
	39119	__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
	39120	float16x4_t __ret;
	39121	__ret = (float16x4_t)(__p0);
	39122	return __ret;
	39123	}
	39124	__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
	39125	int32x2_t __ret;
	39126	__ret = (int32x2_t)(__p0);
	39127	return __ret;
	39128	}
	39129	__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
	39130	int32x2_t __ret;
	39131	__ret = (int32x2_t)(__p0);
	39132	return __ret;
	39133	}
	39134	__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
	39135	int32x2_t __ret;
	39136	__ret = (int32x2_t)(__p0);
	39137	return __ret;
	39138	}
	39139	__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
	39140	int32x2_t __ret;
	39141	__ret = (int32x2_t)(__p0);
	39142	return __ret;
	39143	}
	39144	__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
	39145	int32x2_t __ret;
	39146	__ret = (int32x2_t)(__p0);
	39147	return __ret;
	39148	}
	39149	__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
	39150	int32x2_t __ret;
	39151	__ret = (int32x2_t)(__p0);
	39152	return __ret;
	39153	}
	39154	__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
	39155	int32x2_t __ret;
	39156	__ret = (int32x2_t)(__p0);
	39157	return __ret;
	39158	}
	39159	__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
	39160	int32x2_t __ret;
	39161	__ret = (int32x2_t)(__p0);
	39162	return __ret;
	39163	}
	39164	__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
	39165	int32x2_t __ret;
	39166	__ret = (int32x2_t)(__p0);
	39167	return __ret;
	39168	}
	39169	__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
	39170	int32x2_t __ret;
	39171	__ret = (int32x2_t)(__p0);
	39172	return __ret;
	39173	}
	39174	__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
	39175	int32x2_t __ret;
	39176	__ret = (int32x2_t)(__p0);
	39177	return __ret;
	39178	}
	39179	__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
	39180	int64x1_t __ret;
	39181	__ret = (int64x1_t)(__p0);
	39182	return __ret;
	39183	}
	39184	__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
	39185	int64x1_t __ret;
	39186	__ret = (int64x1_t)(__p0);
	39187	return __ret;
	39188	}
	39189	__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
	39190	int64x1_t __ret;
	39191	__ret = (int64x1_t)(__p0);
	39192	return __ret;
	39193	}
	39194	__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
	39195	int64x1_t __ret;
	39196	__ret = (int64x1_t)(__p0);
	39197	return __ret;
	39198	}
	39199	__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
	39200	int64x1_t __ret;
	39201	__ret = (int64x1_t)(__p0);
	39202	return __ret;
	39203	}
	39204	__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
	39205	int64x1_t __ret;
	39206	__ret = (int64x1_t)(__p0);
	39207	return __ret;
	39208	}
	39209	__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
	39210	int64x1_t __ret;
	39211	__ret = (int64x1_t)(__p0);
	39212	return __ret;
	39213	}
	39214	__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
	39215	int64x1_t __ret;
	39216	__ret = (int64x1_t)(__p0);
	39217	return __ret;
	39218	}
	39219	__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
	39220	int64x1_t __ret;
	39221	__ret = (int64x1_t)(__p0);
	39222	return __ret;
	39223	}
	39224	__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
	39225	int64x1_t __ret;
	39226	__ret = (int64x1_t)(__p0);
	39227	return __ret;
	39228	}
	39229	__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
	39230	int64x1_t __ret;
	39231	__ret = (int64x1_t)(__p0);
	39232	return __ret;
	39233	}
	39234	__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
	39235	int16x4_t __ret;
	39236	__ret = (int16x4_t)(__p0);
	39237	return __ret;
	39238	}
	39239	__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
	39240	int16x4_t __ret;
	39241	__ret = (int16x4_t)(__p0);
	39242	return __ret;
	39243	}
	39244	__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
	39245	int16x4_t __ret;
	39246	__ret = (int16x4_t)(__p0);
	39247	return __ret;
	39248	}
	39249	__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
	39250	int16x4_t __ret;
	39251	__ret = (int16x4_t)(__p0);
	39252	return __ret;
	39253	}
	39254	__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
	39255	int16x4_t __ret;
	39256	__ret = (int16x4_t)(__p0);
	39257	return __ret;
	39258	}
	39259	__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
	39260	int16x4_t __ret;
	39261	__ret = (int16x4_t)(__p0);
	39262	return __ret;
	39263	}
	39264	__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
	39265	int16x4_t __ret;
	39266	__ret = (int16x4_t)(__p0);
	39267	return __ret;
	39268	}
	39269	__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
	39270	int16x4_t __ret;
	39271	__ret = (int16x4_t)(__p0);
	39272	return __ret;
	39273	}
	39274	__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
	39275	int16x4_t __ret;
	39276	__ret = (int16x4_t)(__p0);
	39277	return __ret;
	39278	}
	39279	__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
	39280	int16x4_t __ret;
	39281	__ret = (int16x4_t)(__p0);
	39282	return __ret;
	39283	}
	39284	__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
	39285	int16x4_t __ret;
	39286	__ret = (int16x4_t)(__p0);
	39287	return __ret;
	39288	}
	39289	#ifdef __LITTLE_ENDIAN__
	39290	__ai __attribute__((target("bf16"))) bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
	39291	bfloat16x4_t __ret;
	39292	__ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_f32((int8x16_t)__p0, 11);
	39293	return __ret;
	39294	}
	39295	#else
	39296	__ai __attribute__((target("bf16"))) bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
	39297	bfloat16x4_t __ret;
	39298	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	39299	__ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_f32((int8x16_t)__rev0, 11);
	39300	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	39301	return __ret;
	39302	}
	39303	__ai __attribute__((target("bf16"))) bfloat16x4_t __noswap___a32_vcvt_bf16_f32(float32x4_t __p0) {
	39304	bfloat16x4_t __ret;
	39305	__ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_f32((int8x16_t)__p0, 11);
	39306	return __ret;
	39307	}
	39308	#endif
	39309
	39310	#ifdef __LITTLE_ENDIAN__
	39311	__ai __attribute__((target("bf16"))) bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
	39312	bfloat16x4_t __ret;
	39313	__ret = __a32_vcvt_bf16_f32(__p0);
	39314	return __ret;
	39315	}
	39316	#else
	39317	__ai __attribute__((target("bf16"))) bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
	39318	bfloat16x4_t __ret;
	39319	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	39320	__ret = __noswap___a32_vcvt_bf16_f32(__rev0);
	39321	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	39322	return __ret;
	39323	}
	39324	#endif
	39325
	39326	#ifdef __LITTLE_ENDIAN__
	39327	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
	39328	bfloat16x8_t __ret;
	39329	__ret = vcombine_bf16(__a32_vcvt_bf16_f32(__p1), vget_low_bf16(__p0));
	39330	return __ret;
	39331	}
	39332	#else
	39333	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
	39334	bfloat16x8_t __ret;
	39335	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	39336	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	39337	__ret = __noswap_vcombine_bf16(__noswap___a32_vcvt_bf16_f32(__rev1), __noswap_vget_low_bf16(__rev0));
	39338	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	39339	return __ret;
	39340	}
	39341	#endif
	39342
	39343	#ifdef __LITTLE_ENDIAN__
	39344	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
	39345	bfloat16x8_t __ret;
	39346	__ret = vcombine_bf16((bfloat16x4_t)(0ULL), __a32_vcvt_bf16_f32(__p0));
	39347	return __ret;
	39348	}
	39349	#else
	39350	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
	39351	bfloat16x8_t __ret;
	39352	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	39353	__ret = __noswap_vcombine_bf16((bfloat16x4_t)(0ULL), __noswap___a32_vcvt_bf16_f32(__rev0));
	39354	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	39355	return __ret;
	39356	}
	39357	#endif
	39358
	39359	__ai __attribute__((target("bf16"))) poly8x8_t vreinterpret_p8_bf16(bfloat16x4_t __p0) {
	39360	poly8x8_t __ret;
	39361	__ret = (poly8x8_t)(__p0);
	39362	return __ret;
	39363	}
	39364	__ai __attribute__((target("bf16"))) poly64x1_t vreinterpret_p64_bf16(bfloat16x4_t __p0) {
	39365	poly64x1_t __ret;
	39366	__ret = (poly64x1_t)(__p0);
	39367	return __ret;
	39368	}
	39369	__ai __attribute__((target("bf16"))) poly16x4_t vreinterpret_p16_bf16(bfloat16x4_t __p0) {
	39370	poly16x4_t __ret;
	39371	__ret = (poly16x4_t)(__p0);
	39372	return __ret;
	39373	}
	39374	__ai __attribute__((target("bf16"))) poly8x16_t vreinterpretq_p8_bf16(bfloat16x8_t __p0) {
	39375	poly8x16_t __ret;
	39376	__ret = (poly8x16_t)(__p0);
	39377	return __ret;
	39378	}
	39379	__ai __attribute__((target("bf16"))) poly64x2_t vreinterpretq_p64_bf16(bfloat16x8_t __p0) {
	39380	poly64x2_t __ret;
	39381	__ret = (poly64x2_t)(__p0);
	39382	return __ret;
	39383	}
	39384	__ai __attribute__((target("bf16"))) poly16x8_t vreinterpretq_p16_bf16(bfloat16x8_t __p0) {
	39385	poly16x8_t __ret;
	39386	__ret = (poly16x8_t)(__p0);
	39387	return __ret;
	39388	}
	39389	__ai __attribute__((target("bf16"))) uint8x16_t vreinterpretq_u8_bf16(bfloat16x8_t __p0) {
	39390	uint8x16_t __ret;
	39391	__ret = (uint8x16_t)(__p0);
	39392	return __ret;
	39393	}
	39394	__ai __attribute__((target("bf16"))) uint32x4_t vreinterpretq_u32_bf16(bfloat16x8_t __p0) {
	39395	uint32x4_t __ret;
	39396	__ret = (uint32x4_t)(__p0);
	39397	return __ret;
	39398	}
	39399	__ai __attribute__((target("bf16"))) uint64x2_t vreinterpretq_u64_bf16(bfloat16x8_t __p0) {
	39400	uint64x2_t __ret;
	39401	__ret = (uint64x2_t)(__p0);
	39402	return __ret;
	39403	}
	39404	__ai __attribute__((target("bf16"))) uint16x8_t vreinterpretq_u16_bf16(bfloat16x8_t __p0) {
	39405	uint16x8_t __ret;
	39406	__ret = (uint16x8_t)(__p0);
	39407	return __ret;
	39408	}
	39409	__ai __attribute__((target("bf16"))) int8x16_t vreinterpretq_s8_bf16(bfloat16x8_t __p0) {
	39410	int8x16_t __ret;
	39411	__ret = (int8x16_t)(__p0);
	39412	return __ret;
	39413	}
	39414	__ai __attribute__((target("bf16"))) float32x4_t vreinterpretq_f32_bf16(bfloat16x8_t __p0) {
	39415	float32x4_t __ret;
	39416	__ret = (float32x4_t)(__p0);
	39417	return __ret;
	39418	}
	39419	__ai __attribute__((target("bf16"))) float16x8_t vreinterpretq_f16_bf16(bfloat16x8_t __p0) {
	39420	float16x8_t __ret;
	39421	__ret = (float16x8_t)(__p0);
	39422	return __ret;
	39423	}
	39424	__ai __attribute__((target("bf16"))) int32x4_t vreinterpretq_s32_bf16(bfloat16x8_t __p0) {
	39425	int32x4_t __ret;
	39426	__ret = (int32x4_t)(__p0);
38331	39427	return __ret;
38332	39428	}
38333		__ai float16x4_t vreinterpret_f16_f64(float64x1_t __p0) {
38334		float16x4_t __ret;
38335		__ret = (float16x4_t)(__p0);
	39429	__ai __attribute__((target("bf16"))) int64x2_t vreinterpretq_s64_bf16(bfloat16x8_t __p0) {
	39430	int64x2_t __ret;
	39431	__ret = (int64x2_t)(__p0);
38336	39432	return __ret;
38337	39433	}
38338		__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
38339		float16x4_t __ret;
38340		__ret = (float16x4_t)(__p0);
	39434	__ai __attribute__((target("bf16"))) int16x8_t vreinterpretq_s16_bf16(bfloat16x8_t __p0) {
	39435	int16x8_t __ret;
	39436	__ret = (int16x8_t)(__p0);
38341	39437	return __ret;
38342	39438	}
38343		__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
38344		float16x4_t __ret;
38345		__ret = (float16x4_t)(__p0);
	39439	__ai __attribute__((target("bf16"))) uint8x8_t vreinterpret_u8_bf16(bfloat16x4_t __p0) {
	39440	uint8x8_t __ret;
	39441	__ret = (uint8x8_t)(__p0);
38346	39442	return __ret;
38347	39443	}
38348		__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
38349		float16x4_t __ret;
38350		__ret = (float16x4_t)(__p0);
	39444	__ai __attribute__((target("bf16"))) uint32x2_t vreinterpret_u32_bf16(bfloat16x4_t __p0) {
	39445	uint32x2_t __ret;
	39446	__ret = (uint32x2_t)(__p0);
38351	39447	return __ret;
38352	39448	}
38353		__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
38354		float16x4_t __ret;
38355		__ret = (float16x4_t)(__p0);
	39449	__ai __attribute__((target("bf16"))) uint64x1_t vreinterpret_u64_bf16(bfloat16x4_t __p0) {
	39450	uint64x1_t __ret;
	39451	__ret = (uint64x1_t)(__p0);
38356	39452	return __ret;
38357	39453	}
38358		__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
38359		int32x2_t __ret;
38360		__ret = (int32x2_t)(__p0);
	39454	__ai __attribute__((target("bf16"))) uint16x4_t vreinterpret_u16_bf16(bfloat16x4_t __p0) {
	39455	uint16x4_t __ret;
	39456	__ret = (uint16x4_t)(__p0);
38361	39457	return __ret;
38362	39458	}
38363		__ai int32x2_t vreinterpret_s32_p64(poly64x1_t __p0) {
38364		int32x2_t __ret;
38365		__ret = (int32x2_t)(__p0);
	39459	__ai __attribute__((target("bf16"))) int8x8_t vreinterpret_s8_bf16(bfloat16x4_t __p0) {
	39460	int8x8_t __ret;
	39461	__ret = (int8x8_t)(__p0);
38366	39462	return __ret;
38367	39463	}
38368		__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
38369		int32x2_t __ret;
38370		__ret = (int32x2_t)(__p0);
	39464	__ai __attribute__((target("bf16"))) float32x2_t vreinterpret_f32_bf16(bfloat16x4_t __p0) {
	39465	float32x2_t __ret;
	39466	__ret = (float32x2_t)(__p0);
38371	39467	return __ret;
38372	39468	}
38373		__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
38374		int32x2_t __ret;
38375		__ret = (int32x2_t)(__p0);
	39469	__ai __attribute__((target("bf16"))) float16x4_t vreinterpret_f16_bf16(bfloat16x4_t __p0) {
	39470	float16x4_t __ret;
	39471	__ret = (float16x4_t)(__p0);
38376	39472	return __ret;
38377	39473	}
38378		__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
	39474	__ai __attribute__((target("bf16"))) int32x2_t vreinterpret_s32_bf16(bfloat16x4_t __p0) {
38379	39475	int32x2_t __ret;
38380	39476	__ret = (int32x2_t)(__p0);
38381	39477	return __ret;
38382	39478	}
38383		__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
38384		int32x2_t __ret;
38385		__ret = (int32x2_t)(__p0);
	39479	__ai __attribute__((target("bf16"))) int64x1_t vreinterpret_s64_bf16(bfloat16x4_t __p0) {
	39480	int64x1_t __ret;
	39481	__ret = (int64x1_t)(__p0);
38386	39482	return __ret;
38387	39483	}
38388		__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
38389		int32x2_t __ret;
38390		__ret = (int32x2_t)(__p0);
	39484	__ai __attribute__((target("bf16"))) int16x4_t vreinterpret_s16_bf16(bfloat16x4_t __p0) {
	39485	int16x4_t __ret;
	39486	__ret = (int16x4_t)(__p0);
38391	39487	return __ret;
38392	39488	}
38393		__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
38394		int32x2_t __ret;
38395		__ret = (int32x2_t)(__p0);
	39489	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p8(poly8x16_t __p0) {
	39490	bfloat16x8_t __ret;
	39491	__ret = (bfloat16x8_t)(__p0);
38396	39492	return __ret;
38397	39493	}
38398		__ai int32x2_t vreinterpret_s32_f64(float64x1_t __p0) {
38399		int32x2_t __ret;
38400		__ret = (int32x2_t)(__p0);
	39494	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p64(poly64x2_t __p0) {
	39495	bfloat16x8_t __ret;
	39496	__ret = (bfloat16x8_t)(__p0);
38401	39497	return __ret;
38402	39498	}
38403		__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
38404		int32x2_t __ret;
38405		__ret = (int32x2_t)(__p0);
	39499	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p16(poly16x8_t __p0) {
	39500	bfloat16x8_t __ret;
	39501	__ret = (bfloat16x8_t)(__p0);
38406	39502	return __ret;
38407	39503	}
38408		__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
38409		int32x2_t __ret;
38410		__ret = (int32x2_t)(__p0);
	39504	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u8(uint8x16_t __p0) {
	39505	bfloat16x8_t __ret;
	39506	__ret = (bfloat16x8_t)(__p0);
38411	39507	return __ret;
38412	39508	}
38413		__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
38414		int32x2_t __ret;
38415		__ret = (int32x2_t)(__p0);
	39509	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u32(uint32x4_t __p0) {
	39510	bfloat16x8_t __ret;
	39511	__ret = (bfloat16x8_t)(__p0);
38416	39512	return __ret;
38417	39513	}
38418		__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
38419		int32x2_t __ret;
38420		__ret = (int32x2_t)(__p0);
	39514	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u64(uint64x2_t __p0) {
	39515	bfloat16x8_t __ret;
	39516	__ret = (bfloat16x8_t)(__p0);
38421	39517	return __ret;
38422	39518	}
38423		__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
38424		int64x1_t __ret;
38425		__ret = (int64x1_t)(__p0);
	39519	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u16(uint16x8_t __p0) {
	39520	bfloat16x8_t __ret;
	39521	__ret = (bfloat16x8_t)(__p0);
38426	39522	return __ret;
38427	39523	}
38428		__ai int64x1_t vreinterpret_s64_p64(poly64x1_t __p0) {
38429		int64x1_t __ret;
38430		__ret = (int64x1_t)(__p0);
	39524	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s8(int8x16_t __p0) {
	39525	bfloat16x8_t __ret;
	39526	__ret = (bfloat16x8_t)(__p0);
38431	39527	return __ret;
38432	39528	}
38433		__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
38434		int64x1_t __ret;
38435		__ret = (int64x1_t)(__p0);
	39529	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_f32(float32x4_t __p0) {
	39530	bfloat16x8_t __ret;
	39531	__ret = (bfloat16x8_t)(__p0);
38436	39532	return __ret;
38437	39533	}
38438		__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
38439		int64x1_t __ret;
38440		__ret = (int64x1_t)(__p0);
	39534	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_f16(float16x8_t __p0) {
	39535	bfloat16x8_t __ret;
	39536	__ret = (bfloat16x8_t)(__p0);
38441	39537	return __ret;
38442	39538	}
38443		__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
38444		int64x1_t __ret;
38445		__ret = (int64x1_t)(__p0);
	39539	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s32(int32x4_t __p0) {
	39540	bfloat16x8_t __ret;
	39541	__ret = (bfloat16x8_t)(__p0);
38446	39542	return __ret;
38447	39543	}
38448		__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
38449		int64x1_t __ret;
38450		__ret = (int64x1_t)(__p0);
	39544	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s64(int64x2_t __p0) {
	39545	bfloat16x8_t __ret;
	39546	__ret = (bfloat16x8_t)(__p0);
38451	39547	return __ret;
38452	39548	}
38453		__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
38454		int64x1_t __ret;
38455		__ret = (int64x1_t)(__p0);
	39549	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s16(int16x8_t __p0) {
	39550	bfloat16x8_t __ret;
	39551	__ret = (bfloat16x8_t)(__p0);
38456	39552	return __ret;
38457	39553	}
38458		__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
38459		int64x1_t __ret;
38460		__ret = (int64x1_t)(__p0);
	39554	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_p8(poly8x8_t __p0) {
	39555	bfloat16x4_t __ret;
	39556	__ret = (bfloat16x4_t)(__p0);
38461	39557	return __ret;
38462	39558	}
38463		__ai int64x1_t vreinterpret_s64_f64(float64x1_t __p0) {
38464		int64x1_t __ret;
38465		__ret = (int64x1_t)(__p0);
	39559	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_p64(poly64x1_t __p0) {
	39560	bfloat16x4_t __ret;
	39561	__ret = (bfloat16x4_t)(__p0);
38466	39562	return __ret;
38467	39563	}
38468		__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
38469		int64x1_t __ret;
38470		__ret = (int64x1_t)(__p0);
	39564	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_p16(poly16x4_t __p0) {
	39565	bfloat16x4_t __ret;
	39566	__ret = (bfloat16x4_t)(__p0);
38471	39567	return __ret;
38472	39568	}
38473		__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
38474		int64x1_t __ret;
38475		__ret = (int64x1_t)(__p0);
	39569	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u8(uint8x8_t __p0) {
	39570	bfloat16x4_t __ret;
	39571	__ret = (bfloat16x4_t)(__p0);
38476	39572	return __ret;
38477	39573	}
38478		__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
38479		int64x1_t __ret;
38480		__ret = (int64x1_t)(__p0);
	39574	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u32(uint32x2_t __p0) {
	39575	bfloat16x4_t __ret;
	39576	__ret = (bfloat16x4_t)(__p0);
38481	39577	return __ret;
38482	39578	}
38483		__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
38484		int64x1_t __ret;
38485		__ret = (int64x1_t)(__p0);
	39579	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u64(uint64x1_t __p0) {
	39580	bfloat16x4_t __ret;
	39581	__ret = (bfloat16x4_t)(__p0);
38486	39582	return __ret;
38487	39583	}
38488		__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
38489		int16x4_t __ret;
38490		__ret = (int16x4_t)(__p0);
	39584	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u16(uint16x4_t __p0) {
	39585	bfloat16x4_t __ret;
	39586	__ret = (bfloat16x4_t)(__p0);
38491	39587	return __ret;
38492	39588	}
38493		__ai int16x4_t vreinterpret_s16_p64(poly64x1_t __p0) {
38494		int16x4_t __ret;
38495		__ret = (int16x4_t)(__p0);
	39589	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s8(int8x8_t __p0) {
	39590	bfloat16x4_t __ret;
	39591	__ret = (bfloat16x4_t)(__p0);
38496	39592	return __ret;
38497	39593	}
38498		__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
38499		int16x4_t __ret;
38500		__ret = (int16x4_t)(__p0);
	39594	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_f32(float32x2_t __p0) {
	39595	bfloat16x4_t __ret;
	39596	__ret = (bfloat16x4_t)(__p0);
38501	39597	return __ret;
38502	39598	}
38503		__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
38504		int16x4_t __ret;
38505		__ret = (int16x4_t)(__p0);
	39599	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_f16(float16x4_t __p0) {
	39600	bfloat16x4_t __ret;
	39601	__ret = (bfloat16x4_t)(__p0);
38506	39602	return __ret;
38507	39603	}
38508		__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
38509		int16x4_t __ret;
38510		__ret = (int16x4_t)(__p0);
	39604	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s32(int32x2_t __p0) {
	39605	bfloat16x4_t __ret;
	39606	__ret = (bfloat16x4_t)(__p0);
38511	39607	return __ret;
38512	39608	}
38513		__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
38514		int16x4_t __ret;
38515		__ret = (int16x4_t)(__p0);
	39609	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s64(int64x1_t __p0) {
	39610	bfloat16x4_t __ret;
	39611	__ret = (bfloat16x4_t)(__p0);
38516	39612	return __ret;
38517	39613	}
38518		__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
38519		int16x4_t __ret;
38520		__ret = (int16x4_t)(__p0);
	39614	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
	39615	bfloat16x4_t __ret;
	39616	__ret = (bfloat16x4_t)(__p0);
38521	39617	return __ret;
38522	39618	}
38523		__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
38524		int16x4_t __ret;
38525		__ret = (int16x4_t)(__p0);
	39619	#endif
	39620	#if (__ARM_FP & 2)
	39621	#ifdef __LITTLE_ENDIAN__
	39622	__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
	39623	float16x4_t __ret;
	39624	__ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 41);
38526	39625	return __ret;
38527	39626	}
38528		__ai int16x4_t vreinterpret_s16_f64(float64x1_t __p0) {
38529		int16x4_t __ret;
38530		__ret = (int16x4_t)(__p0);
	39627	#else
	39628	__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
	39629	float16x4_t __ret;
	39630	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	39631	__ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__rev0, 41);
	39632	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38531	39633	return __ret;
38532	39634	}
38533		__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
38534		int16x4_t __ret;
38535		__ret = (int16x4_t)(__p0);
	39635	__ai float16x4_t __noswap_vcvt_f16_f32(float32x4_t __p0) {
	39636	float16x4_t __ret;
	39637	__ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__p0, 41);
38536	39638	return __ret;
38537	39639	}
38538		__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
38539		int16x4_t __ret;
38540		__ret = (int16x4_t)(__p0);
	39640	#endif
	39641
	39642	#ifdef __LITTLE_ENDIAN__
	39643	__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
	39644	float32x4_t __ret;
	39645	__ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 8);
38541	39646	return __ret;
38542	39647	}
38543		__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
38544		int16x4_t __ret;
38545		__ret = (int16x4_t)(__p0);
	39648	#else
	39649	__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
	39650	float32x4_t __ret;
	39651	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	39652	__ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__rev0, 8);
	39653	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38546	39654	return __ret;
38547	39655	}
38548		__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
38549		int16x4_t __ret;
38550		__ret = (int16x4_t)(__p0);
	39656	__ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
	39657	float32x4_t __ret;
	39658	__ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__p0, 8);
38551	39659	return __ret;
38552	39660	}
38553	39661	#endif
38554		#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_DIRECTED_ROUNDING)
	39662
	39663	#ifdef __LITTLE_ENDIAN__
	39664	#define vld1q_f16(__p0) __extension__ ({ \
	39665	float16x8_t __ret; \
	39666	__ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
	39667	__ret; \
	39668	})
	39669	#else
	39670	#define vld1q_f16(__p0) __extension__ ({ \
	39671	float16x8_t __ret; \
	39672	__ret = (float16x8_t) __builtin_neon_vld1q_v(__p0, 40); \
	39673	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	39674	__ret; \
	39675	})
	39676	#endif
	39677
	39678	#ifdef __LITTLE_ENDIAN__
	39679	#define vld1_f16(__p0) __extension__ ({ \
	39680	float16x4_t __ret; \
	39681	__ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
	39682	__ret; \
	39683	})
	39684	#else
	39685	#define vld1_f16(__p0) __extension__ ({ \
	39686	float16x4_t __ret; \
	39687	__ret = (float16x4_t) __builtin_neon_vld1_v(__p0, 8); \
	39688	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	39689	__ret; \
	39690	})
	39691	#endif
	39692
	39693	#ifdef __LITTLE_ENDIAN__
	39694	#define vld1q_dup_f16(__p0) __extension__ ({ \
	39695	float16x8_t __ret; \
	39696	__ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
	39697	__ret; \
	39698	})
	39699	#else
	39700	#define vld1q_dup_f16(__p0) __extension__ ({ \
	39701	float16x8_t __ret; \
	39702	__ret = (float16x8_t) __builtin_neon_vld1q_dup_v(__p0, 40); \
	39703	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	39704	__ret; \
	39705	})
	39706	#endif
	39707
	39708	#ifdef __LITTLE_ENDIAN__
	39709	#define vld1_dup_f16(__p0) __extension__ ({ \
	39710	float16x4_t __ret; \
	39711	__ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
	39712	__ret; \
	39713	})
	39714	#else
	39715	#define vld1_dup_f16(__p0) __extension__ ({ \
	39716	float16x4_t __ret; \
	39717	__ret = (float16x4_t) __builtin_neon_vld1_dup_v(__p0, 8); \
	39718	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	39719	__ret; \
	39720	})
	39721	#endif
	39722
	39723	#ifdef __LITTLE_ENDIAN__
	39724	#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39725	float16x8_t __ret; \
	39726	float16x8_t __s1 = __p1; \
	39727	__ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
	39728	__ret; \
	39729	})
	39730	#else
	39731	#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39732	float16x8_t __ret; \
	39733	float16x8_t __s1 = __p1; \
	39734	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	39735	__ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
	39736	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	39737	__ret; \
	39738	})
	39739	#endif
	39740
	39741	#ifdef __LITTLE_ENDIAN__
	39742	#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39743	float16x4_t __ret; \
	39744	float16x4_t __s1 = __p1; \
	39745	__ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
	39746	__ret; \
	39747	})
	39748	#else
	39749	#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39750	float16x4_t __ret; \
	39751	float16x4_t __s1 = __p1; \
	39752	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	39753	__ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
	39754	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	39755	__ret; \
	39756	})
	39757	#endif
	39758
	39759	#ifdef __LITTLE_ENDIAN__
	39760	#define vld1q_f16_x2(__p0) __extension__ ({ \
	39761	float16x8x2_t __ret; \
	39762	__builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
	39763	__ret; \
	39764	})
	39765	#else
	39766	#define vld1q_f16_x2(__p0) __extension__ ({ \
	39767	float16x8x2_t __ret; \
	39768	__builtin_neon_vld1q_x2_v(&__ret, __p0, 40); \
	39769	\
	39770	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39771	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39772	__ret; \
	39773	})
	39774	#endif
	39775
	39776	#ifdef __LITTLE_ENDIAN__
	39777	#define vld1_f16_x2(__p0) __extension__ ({ \
	39778	float16x4x2_t __ret; \
	39779	__builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
	39780	__ret; \
	39781	})
	39782	#else
	39783	#define vld1_f16_x2(__p0) __extension__ ({ \
	39784	float16x4x2_t __ret; \
	39785	__builtin_neon_vld1_x2_v(&__ret, __p0, 8); \
	39786	\
	39787	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	39788	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	39789	__ret; \
	39790	})
	39791	#endif
	39792
	39793	#ifdef __LITTLE_ENDIAN__
	39794	#define vld1q_f16_x3(__p0) __extension__ ({ \
	39795	float16x8x3_t __ret; \
	39796	__builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
	39797	__ret; \
	39798	})
	39799	#else
	39800	#define vld1q_f16_x3(__p0) __extension__ ({ \
	39801	float16x8x3_t __ret; \
	39802	__builtin_neon_vld1q_x3_v(&__ret, __p0, 40); \
	39803	\
	39804	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39805	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39806	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	39807	__ret; \
	39808	})
	39809	#endif
	39810
	39811	#ifdef __LITTLE_ENDIAN__
	39812	#define vld1_f16_x3(__p0) __extension__ ({ \
	39813	float16x4x3_t __ret; \
	39814	__builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
	39815	__ret; \
	39816	})
	39817	#else
	39818	#define vld1_f16_x3(__p0) __extension__ ({ \
	39819	float16x4x3_t __ret; \
	39820	__builtin_neon_vld1_x3_v(&__ret, __p0, 8); \
	39821	\
	39822	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	39823	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	39824	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	39825	__ret; \
	39826	})
	39827	#endif
	39828
	39829	#ifdef __LITTLE_ENDIAN__
	39830	#define vld1q_f16_x4(__p0) __extension__ ({ \
	39831	float16x8x4_t __ret; \
	39832	__builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
	39833	__ret; \
	39834	})
	39835	#else
	39836	#define vld1q_f16_x4(__p0) __extension__ ({ \
	39837	float16x8x4_t __ret; \
	39838	__builtin_neon_vld1q_x4_v(&__ret, __p0, 40); \
	39839	\
	39840	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39841	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39842	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	39843	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	39844	__ret; \
	39845	})
	39846	#endif
	39847
	39848	#ifdef __LITTLE_ENDIAN__
	39849	#define vld1_f16_x4(__p0) __extension__ ({ \
	39850	float16x4x4_t __ret; \
	39851	__builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
	39852	__ret; \
	39853	})
	39854	#else
	39855	#define vld1_f16_x4(__p0) __extension__ ({ \
	39856	float16x4x4_t __ret; \
	39857	__builtin_neon_vld1_x4_v(&__ret, __p0, 8); \
	39858	\
	39859	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	39860	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	39861	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	39862	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	39863	__ret; \
	39864	})
	39865	#endif
	39866
	39867	#ifdef __LITTLE_ENDIAN__
	39868	#define vld2q_f16(__p0) __extension__ ({ \
	39869	float16x8x2_t __ret; \
	39870	__builtin_neon_vld2q_v(&__ret, __p0, 40); \
	39871	__ret; \
	39872	})
	39873	#else
	39874	#define vld2q_f16(__p0) __extension__ ({ \
	39875	float16x8x2_t __ret; \
	39876	__builtin_neon_vld2q_v(&__ret, __p0, 40); \
	39877	\
	39878	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39879	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39880	__ret; \
	39881	})
	39882	#endif
	39883
	39884	#ifdef __LITTLE_ENDIAN__
	39885	#define vld2_f16(__p0) __extension__ ({ \
	39886	float16x4x2_t __ret; \
	39887	__builtin_neon_vld2_v(&__ret, __p0, 8); \
	39888	__ret; \
	39889	})
	39890	#else
	39891	#define vld2_f16(__p0) __extension__ ({ \
	39892	float16x4x2_t __ret; \
	39893	__builtin_neon_vld2_v(&__ret, __p0, 8); \
	39894	\
	39895	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	39896	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	39897	__ret; \
	39898	})
	39899	#endif
	39900
	39901	#ifdef __LITTLE_ENDIAN__
	39902	#define vld2q_dup_f16(__p0) __extension__ ({ \
	39903	float16x8x2_t __ret; \
	39904	__builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
	39905	__ret; \
	39906	})
	39907	#else
	39908	#define vld2q_dup_f16(__p0) __extension__ ({ \
	39909	float16x8x2_t __ret; \
	39910	__builtin_neon_vld2q_dup_v(&__ret, __p0, 40); \
	39911	\
	39912	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39913	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39914	__ret; \
	39915	})
	39916	#endif
	39917
	39918	#ifdef __LITTLE_ENDIAN__
	39919	#define vld2_dup_f16(__p0) __extension__ ({ \
	39920	float16x4x2_t __ret; \
	39921	__builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
	39922	__ret; \
	39923	})
	39924	#else
	39925	#define vld2_dup_f16(__p0) __extension__ ({ \
	39926	float16x4x2_t __ret; \
	39927	__builtin_neon_vld2_dup_v(&__ret, __p0, 8); \
	39928	\
	39929	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	39930	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	39931	__ret; \
	39932	})
	39933	#endif
	39934
38555	39935	#ifdef __LITTLE_ENDIAN__
38556		__ai float64x2_t vrndq_f64(float64x2_t __p0) {
38557		float64x2_t __ret;
38558		__ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 42);
38559		return __ret;
38560		}
	39936	#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39937	float16x8x2_t __ret; \
	39938	float16x8x2_t __s1 = __p1; \
	39939	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
	39940	__ret; \
	39941	})
38561	39942	#else
38562		__ai float64x2_t vrndq_f64(float64x2_t __p0) {
38563		float64x2_t __ret;
38564		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38565		__ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 42);
38566		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38567		return __ret;
38568		}
	39943	#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39944	float16x8x2_t __ret; \
	39945	float16x8x2_t __s1 = __p1; \
	39946	float16x8x2_t __rev1; \
	39947	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39948	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39949	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
	39950	\
	39951	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39952	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39953	__ret; \
	39954	})
38569	39955	#endif
38570	39956
38571		__ai float64x1_t vrnd_f64(float64x1_t __p0) {
38572		float64x1_t __ret;
38573		__ret = (float64x1_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 10);
38574		return __ret;
38575		}
38576	39957	#ifdef __LITTLE_ENDIAN__
38577		__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
38578		float64x2_t __ret;
38579		__ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 42);
38580		return __ret;
38581		}
	39958	#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39959	float16x4x2_t __ret; \
	39960	float16x4x2_t __s1 = __p1; \
	39961	__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
	39962	__ret; \
	39963	})
38582	39964	#else
38583		__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
38584		float64x2_t __ret;
38585		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38586		__ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 42);
38587		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38588		return __ret;
38589		}
	39965	#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	39966	float16x4x2_t __ret; \
	39967	float16x4x2_t __s1 = __p1; \
	39968	float16x4x2_t __rev1; \
	39969	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	39970	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	39971	__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
	39972	\
	39973	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	39974	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	39975	__ret; \
	39976	})
38590	39977	#endif
38591	39978
38592		__ai float64x1_t vrnda_f64(float64x1_t __p0) {
38593		float64x1_t __ret;
38594		__ret = (float64x1_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 10);
38595		return __ret;
38596		}
38597	39979	#ifdef __LITTLE_ENDIAN__
38598		__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
38599		float64x2_t __ret;
38600		__ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 42);
38601		return __ret;
38602		}
	39980	#define vld3q_f16(__p0) __extension__ ({ \
	39981	float16x8x3_t __ret; \
	39982	__builtin_neon_vld3q_v(&__ret, __p0, 40); \
	39983	__ret; \
	39984	})
38603	39985	#else
38604		__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
38605		float64x2_t __ret;
38606		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38607		__ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 42);
38608		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38609		return __ret;
38610		}
	39986	#define vld3q_f16(__p0) __extension__ ({ \
	39987	float16x8x3_t __ret; \
	39988	__builtin_neon_vld3q_v(&__ret, __p0, 40); \
	39989	\
	39990	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	39991	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	39992	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	39993	__ret; \
	39994	})
38611	39995	#endif
38612	39996
38613		__ai float64x1_t vrndi_f64(float64x1_t __p0) {
38614		float64x1_t __ret;
38615		__ret = (float64x1_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 10);
38616		return __ret;
38617		}
38618	39997	#ifdef __LITTLE_ENDIAN__
38619		__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
38620		float64x2_t __ret;
38621		__ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 42);
38622		return __ret;
38623		}
	39998	#define vld3_f16(__p0) __extension__ ({ \
	39999	float16x4x3_t __ret; \
	40000	__builtin_neon_vld3_v(&__ret, __p0, 8); \
	40001	__ret; \
	40002	})
38624	40003	#else
38625		__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
38626		float64x2_t __ret;
38627		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38628		__ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 42);
38629		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38630		return __ret;
38631		}
	40004	#define vld3_f16(__p0) __extension__ ({ \
	40005	float16x4x3_t __ret; \
	40006	__builtin_neon_vld3_v(&__ret, __p0, 8); \
	40007	\
	40008	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	40009	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	40010	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	40011	__ret; \
	40012	})
38632	40013	#endif
38633	40014
38634		__ai float64x1_t vrndm_f64(float64x1_t __p0) {
38635		float64x1_t __ret;
38636		__ret = (float64x1_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 10);
38637		return __ret;
38638		}
38639	40015	#ifdef __LITTLE_ENDIAN__
38640		__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
38641		float64x2_t __ret;
38642		__ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 42);
38643		return __ret;
38644		}
	40016	#define vld3q_dup_f16(__p0) __extension__ ({ \
	40017	float16x8x3_t __ret; \
	40018	__builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
	40019	__ret; \
	40020	})
38645	40021	#else
38646		__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
38647		float64x2_t __ret;
38648		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38649		__ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 42);
38650		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38651		return __ret;
38652		}
	40022	#define vld3q_dup_f16(__p0) __extension__ ({ \
	40023	float16x8x3_t __ret; \
	40024	__builtin_neon_vld3q_dup_v(&__ret, __p0, 40); \
	40025	\
	40026	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40027	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40028	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40029	__ret; \
	40030	})
	40031	#endif
	40032
	40033	#ifdef __LITTLE_ENDIAN__
	40034	#define vld3_dup_f16(__p0) __extension__ ({ \
	40035	float16x4x3_t __ret; \
	40036	__builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
	40037	__ret; \
	40038	})
	40039	#else
	40040	#define vld3_dup_f16(__p0) __extension__ ({ \
	40041	float16x4x3_t __ret; \
	40042	__builtin_neon_vld3_dup_v(&__ret, __p0, 8); \
	40043	\
	40044	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	40045	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	40046	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	40047	__ret; \
	40048	})
	40049	#endif
	40050
	40051	#ifdef __LITTLE_ENDIAN__
	40052	#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40053	float16x8x3_t __ret; \
	40054	float16x8x3_t __s1 = __p1; \
	40055	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
	40056	__ret; \
	40057	})
	40058	#else
	40059	#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40060	float16x8x3_t __ret; \
	40061	float16x8x3_t __s1 = __p1; \
	40062	float16x8x3_t __rev1; \
	40063	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40064	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40065	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40066	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
	40067	\
	40068	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40069	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40070	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40071	__ret; \
	40072	})
	40073	#endif
	40074
	40075	#ifdef __LITTLE_ENDIAN__
	40076	#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40077	float16x4x3_t __ret; \
	40078	float16x4x3_t __s1 = __p1; \
	40079	__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
	40080	__ret; \
	40081	})
	40082	#else
	40083	#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40084	float16x4x3_t __ret; \
	40085	float16x4x3_t __s1 = __p1; \
	40086	float16x4x3_t __rev1; \
	40087	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40088	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40089	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40090	__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
	40091	\
	40092	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	40093	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	40094	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	40095	__ret; \
	40096	})
	40097	#endif
	40098
	40099	#ifdef __LITTLE_ENDIAN__
	40100	#define vld4q_f16(__p0) __extension__ ({ \
	40101	float16x8x4_t __ret; \
	40102	__builtin_neon_vld4q_v(&__ret, __p0, 40); \
	40103	__ret; \
	40104	})
	40105	#else
	40106	#define vld4q_f16(__p0) __extension__ ({ \
	40107	float16x8x4_t __ret; \
	40108	__builtin_neon_vld4q_v(&__ret, __p0, 40); \
	40109	\
	40110	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40111	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40112	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40113	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40114	__ret; \
	40115	})
	40116	#endif
	40117
	40118	#ifdef __LITTLE_ENDIAN__
	40119	#define vld4_f16(__p0) __extension__ ({ \
	40120	float16x4x4_t __ret; \
	40121	__builtin_neon_vld4_v(&__ret, __p0, 8); \
	40122	__ret; \
	40123	})
	40124	#else
	40125	#define vld4_f16(__p0) __extension__ ({ \
	40126	float16x4x4_t __ret; \
	40127	__builtin_neon_vld4_v(&__ret, __p0, 8); \
	40128	\
	40129	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	40130	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	40131	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	40132	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	40133	__ret; \
	40134	})
	40135	#endif
	40136
	40137	#ifdef __LITTLE_ENDIAN__
	40138	#define vld4q_dup_f16(__p0) __extension__ ({ \
	40139	float16x8x4_t __ret; \
	40140	__builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
	40141	__ret; \
	40142	})
	40143	#else
	40144	#define vld4q_dup_f16(__p0) __extension__ ({ \
	40145	float16x8x4_t __ret; \
	40146	__builtin_neon_vld4q_dup_v(&__ret, __p0, 40); \
	40147	\
	40148	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40149	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40150	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40151	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40152	__ret; \
	40153	})
	40154	#endif
	40155
	40156	#ifdef __LITTLE_ENDIAN__
	40157	#define vld4_dup_f16(__p0) __extension__ ({ \
	40158	float16x4x4_t __ret; \
	40159	__builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
	40160	__ret; \
	40161	})
	40162	#else
	40163	#define vld4_dup_f16(__p0) __extension__ ({ \
	40164	float16x4x4_t __ret; \
	40165	__builtin_neon_vld4_dup_v(&__ret, __p0, 8); \
	40166	\
	40167	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	40168	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	40169	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	40170	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	40171	__ret; \
	40172	})
	40173	#endif
	40174
	40175	#ifdef __LITTLE_ENDIAN__
	40176	#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40177	float16x8x4_t __ret; \
	40178	float16x8x4_t __s1 = __p1; \
	40179	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 40); \
	40180	__ret; \
	40181	})
	40182	#else
	40183	#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40184	float16x8x4_t __ret; \
	40185	float16x8x4_t __s1 = __p1; \
	40186	float16x8x4_t __rev1; \
	40187	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40188	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40189	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40190	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40191	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 40); \
	40192	\
	40193	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40194	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40195	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40196	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40197	__ret; \
	40198	})
	40199	#endif
	40200
	40201	#ifdef __LITTLE_ENDIAN__
	40202	#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40203	float16x4x4_t __ret; \
	40204	float16x4x4_t __s1 = __p1; \
	40205	__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 8); \
	40206	__ret; \
	40207	})
	40208	#else
	40209	#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40210	float16x4x4_t __ret; \
	40211	float16x4x4_t __s1 = __p1; \
	40212	float16x4x4_t __rev1; \
	40213	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40214	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40215	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40216	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	40217	__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 8); \
	40218	\
	40219	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
	40220	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
	40221	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
	40222	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
	40223	__ret; \
	40224	})
	40225	#endif
	40226
	40227	#ifdef __LITTLE_ENDIAN__
	40228	#define vst1q_f16(__p0, __p1) __extension__ ({ \
	40229	float16x8_t __s1 = __p1; \
	40230	__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 40); \
	40231	})
	40232	#else
	40233	#define vst1q_f16(__p0, __p1) __extension__ ({ \
	40234	float16x8_t __s1 = __p1; \
	40235	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	40236	__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 40); \
	40237	})
	40238	#endif
	40239
	40240	#ifdef __LITTLE_ENDIAN__
	40241	#define vst1_f16(__p0, __p1) __extension__ ({ \
	40242	float16x4_t __s1 = __p1; \
	40243	__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 8); \
	40244	})
	40245	#else
	40246	#define vst1_f16(__p0, __p1) __extension__ ({ \
	40247	float16x4_t __s1 = __p1; \
	40248	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	40249	__builtin_neon_vst1_v(__p0, (int8x8_t)__rev1, 8); \
	40250	})
	40251	#endif
	40252
	40253	#ifdef __LITTLE_ENDIAN__
	40254	#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40255	float16x8_t __s1 = __p1; \
	40256	__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
	40257	})
	40258	#else
	40259	#define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40260	float16x8_t __s1 = __p1; \
	40261	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	40262	__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
	40263	})
	40264	#endif
	40265
	40266	#ifdef __LITTLE_ENDIAN__
	40267	#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40268	float16x4_t __s1 = __p1; \
	40269	__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
	40270	})
	40271	#else
	40272	#define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40273	float16x4_t __s1 = __p1; \
	40274	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	40275	__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
	40276	})
	40277	#endif
	40278
	40279	#ifdef __LITTLE_ENDIAN__
	40280	#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
	40281	float16x8x2_t __s1 = __p1; \
	40282	__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 40); \
	40283	})
	40284	#else
	40285	#define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
	40286	float16x8x2_t __s1 = __p1; \
	40287	float16x8x2_t __rev1; \
	40288	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40289	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40290	__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 40); \
	40291	})
	40292	#endif
	40293
	40294	#ifdef __LITTLE_ENDIAN__
	40295	#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
	40296	float16x4x2_t __s1 = __p1; \
	40297	__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 8); \
	40298	})
	40299	#else
	40300	#define vst1_f16_x2(__p0, __p1) __extension__ ({ \
	40301	float16x4x2_t __s1 = __p1; \
	40302	float16x4x2_t __rev1; \
	40303	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40304	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40305	__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 8); \
	40306	})
	40307	#endif
	40308
	40309	#ifdef __LITTLE_ENDIAN__
	40310	#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
	40311	float16x8x3_t __s1 = __p1; \
	40312	__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 40); \
	40313	})
	40314	#else
	40315	#define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
	40316	float16x8x3_t __s1 = __p1; \
	40317	float16x8x3_t __rev1; \
	40318	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40319	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40320	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40321	__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 40); \
	40322	})
	40323	#endif
	40324
	40325	#ifdef __LITTLE_ENDIAN__
	40326	#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
	40327	float16x4x3_t __s1 = __p1; \
	40328	__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 8); \
	40329	})
	40330	#else
	40331	#define vst1_f16_x3(__p0, __p1) __extension__ ({ \
	40332	float16x4x3_t __s1 = __p1; \
	40333	float16x4x3_t __rev1; \
	40334	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40335	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40336	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40337	__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 8); \
	40338	})
	40339	#endif
	40340
	40341	#ifdef __LITTLE_ENDIAN__
	40342	#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
	40343	float16x8x4_t __s1 = __p1; \
	40344	__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 40); \
	40345	})
	40346	#else
	40347	#define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
	40348	float16x8x4_t __s1 = __p1; \
	40349	float16x8x4_t __rev1; \
	40350	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40351	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40352	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40353	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40354	__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 40); \
	40355	})
	40356	#endif
	40357
	40358	#ifdef __LITTLE_ENDIAN__
	40359	#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
	40360	float16x4x4_t __s1 = __p1; \
	40361	__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 8); \
	40362	})
	40363	#else
	40364	#define vst1_f16_x4(__p0, __p1) __extension__ ({ \
	40365	float16x4x4_t __s1 = __p1; \
	40366	float16x4x4_t __rev1; \
	40367	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40368	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40369	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40370	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	40371	__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 8); \
	40372	})
	40373	#endif
	40374
	40375	#ifdef __LITTLE_ENDIAN__
	40376	#define vst2q_f16(__p0, __p1) __extension__ ({ \
	40377	float16x8x2_t __s1 = __p1; \
	40378	__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 40); \
	40379	})
	40380	#else
	40381	#define vst2q_f16(__p0, __p1) __extension__ ({ \
	40382	float16x8x2_t __s1 = __p1; \
	40383	float16x8x2_t __rev1; \
	40384	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40385	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40386	__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 40); \
	40387	})
38653	40388	#endif
38654	40389
38655		__ai float64x1_t vrndn_f64(float64x1_t __p0) {
38656		float64x1_t __ret;
38657		__ret = (float64x1_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 10);
38658		return __ret;
38659		}
38660	40390	#ifdef __LITTLE_ENDIAN__
38661		__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
38662		float64x2_t __ret;
38663		__ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 42);
38664		return __ret;
38665		}
	40391	#define vst2_f16(__p0, __p1) __extension__ ({ \
	40392	float16x4x2_t __s1 = __p1; \
	40393	__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 8); \
	40394	})
38666	40395	#else
38667		__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
38668		float64x2_t __ret;
38669		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38670		__ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 42);
38671		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38672		return __ret;
38673		}
	40396	#define vst2_f16(__p0, __p1) __extension__ ({ \
	40397	float16x4x2_t __s1 = __p1; \
	40398	float16x4x2_t __rev1; \
	40399	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40400	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40401	__builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 8); \
	40402	})
38674	40403	#endif
38675	40404
38676		__ai float64x1_t vrndp_f64(float64x1_t __p0) {
38677		float64x1_t __ret;
38678		__ret = (float64x1_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 10);
38679		return __ret;
38680		}
38681	40405	#ifdef __LITTLE_ENDIAN__
38682		__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
38683		float64x2_t __ret;
38684		__ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 42);
38685		return __ret;
38686		}
	40406	#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40407	float16x8x2_t __s1 = __p1; \
	40408	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
	40409	})
38687	40410	#else
38688		__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
38689		float64x2_t __ret;
38690		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38691		__ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 42);
38692		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38693		return __ret;
38694		}
	40411	#define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40412	float16x8x2_t __s1 = __p1; \
	40413	float16x8x2_t __rev1; \
	40414	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40415	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40416	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
	40417	})
38695	40418	#endif
38696	40419
38697		__ai float64x1_t vrndx_f64(float64x1_t __p0) {
38698		float64x1_t __ret;
38699		__ret = (float64x1_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 10);
38700		return __ret;
38701		}
	40420	#ifdef __LITTLE_ENDIAN__
	40421	#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40422	float16x4x2_t __s1 = __p1; \
	40423	__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
	40424	})
	40425	#else
	40426	#define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40427	float16x4x2_t __s1 = __p1; \
	40428	float16x4x2_t __rev1; \
	40429	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40430	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40431	__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
	40432	})
38702	40433	#endif
38703		#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_FRINT)
	40434
38704	40435	#ifdef __LITTLE_ENDIAN__
38705		__ai float32x4_t vrnd32xq_f32(float32x4_t __p0) {
38706		float32x4_t __ret;
38707		__ret = (float32x4_t) __builtin_neon_vrnd32xq_v((int8x16_t)__p0, 41);
38708		return __ret;
38709		}
	40436	#define vst3q_f16(__p0, __p1) __extension__ ({ \
	40437	float16x8x3_t __s1 = __p1; \
	40438	__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 40); \
	40439	})
38710	40440	#else
38711		__ai float32x4_t vrnd32xq_f32(float32x4_t __p0) {
38712		float32x4_t __ret;
38713		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
38714		__ret = (float32x4_t) __builtin_neon_vrnd32xq_v((int8x16_t)__rev0, 41);
38715		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38716		return __ret;
38717		}
	40441	#define vst3q_f16(__p0, __p1) __extension__ ({ \
	40442	float16x8x3_t __s1 = __p1; \
	40443	float16x8x3_t __rev1; \
	40444	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40445	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40446	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40447	__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 40); \
	40448	})
38718	40449	#endif
38719	40450
38720	40451	#ifdef __LITTLE_ENDIAN__
38721		__ai float32x2_t vrnd32x_f32(float32x2_t __p0) {
38722		float32x2_t __ret;
38723		__ret = (float32x2_t) __builtin_neon_vrnd32x_v((int8x8_t)__p0, 9);
38724		return __ret;
38725		}
	40452	#define vst3_f16(__p0, __p1) __extension__ ({ \
	40453	float16x4x3_t __s1 = __p1; \
	40454	__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 8); \
	40455	})
38726	40456	#else
38727		__ai float32x2_t vrnd32x_f32(float32x2_t __p0) {
38728		float32x2_t __ret;
38729		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38730		__ret = (float32x2_t) __builtin_neon_vrnd32x_v((int8x8_t)__rev0, 9);
38731		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38732		return __ret;
38733		}
	40457	#define vst3_f16(__p0, __p1) __extension__ ({ \
	40458	float16x4x3_t __s1 = __p1; \
	40459	float16x4x3_t __rev1; \
	40460	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40461	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40462	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40463	__builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 8); \
	40464	})
38734	40465	#endif
38735	40466
38736	40467	#ifdef __LITTLE_ENDIAN__
38737		__ai float32x4_t vrnd32zq_f32(float32x4_t __p0) {
38738		float32x4_t __ret;
38739		__ret = (float32x4_t) __builtin_neon_vrnd32zq_v((int8x16_t)__p0, 41);
38740		return __ret;
38741		}
	40468	#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40469	float16x8x3_t __s1 = __p1; \
	40470	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
	40471	})
38742	40472	#else
38743		__ai float32x4_t vrnd32zq_f32(float32x4_t __p0) {
38744		float32x4_t __ret;
38745		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
38746		__ret = (float32x4_t) __builtin_neon_vrnd32zq_v((int8x16_t)__rev0, 41);
38747		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38748		return __ret;
38749		}
	40473	#define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40474	float16x8x3_t __s1 = __p1; \
	40475	float16x8x3_t __rev1; \
	40476	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40477	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40478	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40479	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
	40480	})
38750	40481	#endif
38751	40482
38752	40483	#ifdef __LITTLE_ENDIAN__
38753		__ai float32x2_t vrnd32z_f32(float32x2_t __p0) {
38754		float32x2_t __ret;
38755		__ret = (float32x2_t) __builtin_neon_vrnd32z_v((int8x8_t)__p0, 9);
38756		return __ret;
38757		}
	40484	#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40485	float16x4x3_t __s1 = __p1; \
	40486	__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
	40487	})
38758	40488	#else
38759		__ai float32x2_t vrnd32z_f32(float32x2_t __p0) {
38760		float32x2_t __ret;
38761		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38762		__ret = (float32x2_t) __builtin_neon_vrnd32z_v((int8x8_t)__rev0, 9);
38763		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38764		return __ret;
38765		}
	40489	#define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40490	float16x4x3_t __s1 = __p1; \
	40491	float16x4x3_t __rev1; \
	40492	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40493	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40494	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40495	__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
	40496	})
38766	40497	#endif
38767	40498
38768	40499	#ifdef __LITTLE_ENDIAN__
38769		__ai float32x4_t vrnd64xq_f32(float32x4_t __p0) {
38770		float32x4_t __ret;
38771		__ret = (float32x4_t) __builtin_neon_vrnd64xq_v((int8x16_t)__p0, 41);
38772		return __ret;
38773		}
	40500	#define vst4q_f16(__p0, __p1) __extension__ ({ \
	40501	float16x8x4_t __s1 = __p1; \
	40502	__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 40); \
	40503	})
38774	40504	#else
38775		__ai float32x4_t vrnd64xq_f32(float32x4_t __p0) {
38776		float32x4_t __ret;
38777		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
38778		__ret = (float32x4_t) __builtin_neon_vrnd64xq_v((int8x16_t)__rev0, 41);
38779		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38780		return __ret;
38781		}
	40505	#define vst4q_f16(__p0, __p1) __extension__ ({ \
	40506	float16x8x4_t __s1 = __p1; \
	40507	float16x8x4_t __rev1; \
	40508	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40509	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40510	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40511	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40512	__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 40); \
	40513	})
38782	40514	#endif
38783	40515
38784	40516	#ifdef __LITTLE_ENDIAN__
38785		__ai float32x2_t vrnd64x_f32(float32x2_t __p0) {
38786		float32x2_t __ret;
38787		__ret = (float32x2_t) __builtin_neon_vrnd64x_v((int8x8_t)__p0, 9);
38788		return __ret;
38789		}
	40517	#define vst4_f16(__p0, __p1) __extension__ ({ \
	40518	float16x4x4_t __s1 = __p1; \
	40519	__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 8); \
	40520	})
38790	40521	#else
38791		__ai float32x2_t vrnd64x_f32(float32x2_t __p0) {
38792		float32x2_t __ret;
38793		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38794		__ret = (float32x2_t) __builtin_neon_vrnd64x_v((int8x8_t)__rev0, 9);
38795		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38796		return __ret;
38797		}
	40522	#define vst4_f16(__p0, __p1) __extension__ ({ \
	40523	float16x4x4_t __s1 = __p1; \
	40524	float16x4x4_t __rev1; \
	40525	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40526	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40527	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40528	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	40529	__builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 8); \
	40530	})
38798	40531	#endif
38799	40532
38800	40533	#ifdef __LITTLE_ENDIAN__
38801		__ai float32x4_t vrnd64zq_f32(float32x4_t __p0) {
38802		float32x4_t __ret;
38803		__ret = (float32x4_t) __builtin_neon_vrnd64zq_v((int8x16_t)__p0, 41);
38804		return __ret;
38805		}
	40534	#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40535	float16x8x4_t __s1 = __p1; \
	40536	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 40); \
	40537	})
38806	40538	#else
38807		__ai float32x4_t vrnd64zq_f32(float32x4_t __p0) {
38808		float32x4_t __ret;
38809		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
38810		__ret = (float32x4_t) __builtin_neon_vrnd64zq_v((int8x16_t)__rev0, 41);
38811		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38812		return __ret;
38813		}
	40539	#define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40540	float16x8x4_t __s1 = __p1; \
	40541	float16x8x4_t __rev1; \
	40542	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
	40543	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
	40544	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
	40545	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
	40546	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 40); \
	40547	})
38814	40548	#endif
38815	40549
38816	40550	#ifdef __LITTLE_ENDIAN__
38817		__ai float32x2_t vrnd64z_f32(float32x2_t __p0) {
38818		float32x2_t __ret;
38819		__ret = (float32x2_t) __builtin_neon_vrnd64z_v((int8x8_t)__p0, 9);
38820		return __ret;
38821		}
	40551	#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40552	float16x4x4_t __s1 = __p1; \
	40553	__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 8); \
	40554	})
38822	40555	#else
38823		__ai float32x2_t vrnd64z_f32(float32x2_t __p0) {
38824		float32x2_t __ret;
38825		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38826		__ret = (float32x2_t) __builtin_neon_vrnd64z_v((int8x8_t)__rev0, 9);
38827		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38828		return __ret;
38829		}
	40556	#define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	40557	float16x4x4_t __s1 = __p1; \
	40558	float16x4x4_t __rev1; \
	40559	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
	40560	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
	40561	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
	40562	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
	40563	__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 8); \
	40564	})
38830	40565	#endif
38831	40566
38832	40567	#endif
38833		#if __ARM_ARCH >= 8 && defined(__aarch64__) && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
	40568	#if __ARM_ARCH >= 8
38834	40569	#ifdef __LITTLE_ENDIAN__
38835		__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
38836		float64x2_t __ret;
38837		__ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	40570	__ai int32x4_t vcvtaq_s32_f32(float32x4_t __p0) {
	40571	int32x4_t __ret;
	40572	__ret = (int32x4_t) __builtin_neon_vcvtaq_s32_v((int8x16_t)__p0, 34);
38838	40573	return __ret;
38839	40574	}
38840	40575	#else
38841		__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
38842		float64x2_t __ret;
38843		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38844		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
38845		__ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
38846		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	40576	__ai int32x4_t vcvtaq_s32_f32(float32x4_t __p0) {
	40577	int32x4_t __ret;
	40578	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40579	__ret = (int32x4_t) __builtin_neon_vcvtaq_s32_v((int8x16_t)__rev0, 34);
	40580	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38847	40581	return __ret;
38848	40582	}
38849	40583	#endif
38850	40584
38851		__ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
38852		float64x1_t __ret;
38853		__ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
38854		return __ret;
38855		}
38856	40585	#ifdef __LITTLE_ENDIAN__
38857		__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
38858		float64x2_t __ret;
38859		__ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	40586	__ai int32x2_t vcvta_s32_f32(float32x2_t __p0) {
	40587	int32x2_t __ret;
	40588	__ret = (int32x2_t) __builtin_neon_vcvta_s32_v((int8x8_t)__p0, 2);
38860	40589	return __ret;
38861	40590	}
38862	40591	#else
38863		__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
38864		float64x2_t __ret;
38865		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38866		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
38867		__ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	40592	__ai int32x2_t vcvta_s32_f32(float32x2_t __p0) {
	40593	int32x2_t __ret;
	40594	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40595	__ret = (int32x2_t) __builtin_neon_vcvta_s32_v((int8x8_t)__rev0, 2);
38868	40596	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38869	40597	return __ret;
38870	40598	}
38871	40599	#endif
38872	40600
38873		__ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
38874		float64x1_t __ret;
38875		__ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
38876		return __ret;
38877		}
38878		#endif
38879		#if defined(__ARM_FEATURE_BF16) && !defined(__aarch64__)
38880		__ai poly8x8_t vreinterpret_p8_bf16(bfloat16x4_t __p0) {
38881		poly8x8_t __ret;
38882		__ret = (poly8x8_t)(__p0);
38883		return __ret;
38884		}
38885		__ai poly64x1_t vreinterpret_p64_bf16(bfloat16x4_t __p0) {
38886		poly64x1_t __ret;
38887		__ret = (poly64x1_t)(__p0);
38888		return __ret;
38889		}
38890		__ai poly16x4_t vreinterpret_p16_bf16(bfloat16x4_t __p0) {
38891		poly16x4_t __ret;
38892		__ret = (poly16x4_t)(__p0);
38893		return __ret;
38894		}
38895		__ai poly8x16_t vreinterpretq_p8_bf16(bfloat16x8_t __p0) {
38896		poly8x16_t __ret;
38897		__ret = (poly8x16_t)(__p0);
38898		return __ret;
38899		}
38900		__ai poly64x2_t vreinterpretq_p64_bf16(bfloat16x8_t __p0) {
38901		poly64x2_t __ret;
38902		__ret = (poly64x2_t)(__p0);
38903		return __ret;
38904		}
38905		__ai poly16x8_t vreinterpretq_p16_bf16(bfloat16x8_t __p0) {
38906		poly16x8_t __ret;
38907		__ret = (poly16x8_t)(__p0);
38908		return __ret;
38909		}
38910		__ai uint8x16_t vreinterpretq_u8_bf16(bfloat16x8_t __p0) {
38911		uint8x16_t __ret;
38912		__ret = (uint8x16_t)(__p0);
38913		return __ret;
38914		}
38915		__ai uint32x4_t vreinterpretq_u32_bf16(bfloat16x8_t __p0) {
	40601	#ifdef __LITTLE_ENDIAN__
	40602	__ai uint32x4_t vcvtaq_u32_f32(float32x4_t __p0) {
38916	40603	uint32x4_t __ret;
38917		__ret = (uint32x4_t)(__p0);
38918		return __ret;
38919		}
38920		__ai uint64x2_t vreinterpretq_u64_bf16(bfloat16x8_t __p0) {
38921		uint64x2_t __ret;
38922		__ret = (uint64x2_t)(__p0);
38923		return __ret;
38924		}
38925		__ai uint16x8_t vreinterpretq_u16_bf16(bfloat16x8_t __p0) {
38926		uint16x8_t __ret;
38927		__ret = (uint16x8_t)(__p0);
38928		return __ret;
38929		}
38930		__ai int8x16_t vreinterpretq_s8_bf16(bfloat16x8_t __p0) {
38931		int8x16_t __ret;
38932		__ret = (int8x16_t)(__p0);
38933		return __ret;
38934		}
38935		__ai float32x4_t vreinterpretq_f32_bf16(bfloat16x8_t __p0) {
38936		float32x4_t __ret;
38937		__ret = (float32x4_t)(__p0);
38938		return __ret;
38939		}
38940		__ai float16x8_t vreinterpretq_f16_bf16(bfloat16x8_t __p0) {
38941		float16x8_t __ret;
38942		__ret = (float16x8_t)(__p0);
38943		return __ret;
38944		}
38945		__ai int32x4_t vreinterpretq_s32_bf16(bfloat16x8_t __p0) {
38946		int32x4_t __ret;
38947		__ret = (int32x4_t)(__p0);
38948		return __ret;
38949		}
38950		__ai int64x2_t vreinterpretq_s64_bf16(bfloat16x8_t __p0) {
38951		int64x2_t __ret;
38952		__ret = (int64x2_t)(__p0);
38953		return __ret;
38954		}
38955		__ai int16x8_t vreinterpretq_s16_bf16(bfloat16x8_t __p0) {
38956		int16x8_t __ret;
38957		__ret = (int16x8_t)(__p0);
38958		return __ret;
38959		}
38960		__ai uint8x8_t vreinterpret_u8_bf16(bfloat16x4_t __p0) {
38961		uint8x8_t __ret;
38962		__ret = (uint8x8_t)(__p0);
38963		return __ret;
38964		}
38965		__ai uint32x2_t vreinterpret_u32_bf16(bfloat16x4_t __p0) {
38966		uint32x2_t __ret;
38967		__ret = (uint32x2_t)(__p0);
38968		return __ret;
38969		}
38970		__ai uint64x1_t vreinterpret_u64_bf16(bfloat16x4_t __p0) {
38971		uint64x1_t __ret;
38972		__ret = (uint64x1_t)(__p0);
38973		return __ret;
38974		}
38975		__ai uint16x4_t vreinterpret_u16_bf16(bfloat16x4_t __p0) {
38976		uint16x4_t __ret;
38977		__ret = (uint16x4_t)(__p0);
38978		return __ret;
38979		}
38980		__ai int8x8_t vreinterpret_s8_bf16(bfloat16x4_t __p0) {
38981		int8x8_t __ret;
38982		__ret = (int8x8_t)(__p0);
38983		return __ret;
38984		}
38985		__ai float32x2_t vreinterpret_f32_bf16(bfloat16x4_t __p0) {
38986		float32x2_t __ret;
38987		__ret = (float32x2_t)(__p0);
38988		return __ret;
38989		}
38990		__ai float16x4_t vreinterpret_f16_bf16(bfloat16x4_t __p0) {
38991		float16x4_t __ret;
38992		__ret = (float16x4_t)(__p0);
38993		return __ret;
38994		}
38995		__ai int32x2_t vreinterpret_s32_bf16(bfloat16x4_t __p0) {
38996		int32x2_t __ret;
38997		__ret = (int32x2_t)(__p0);
38998		return __ret;
38999		}
39000		__ai int64x1_t vreinterpret_s64_bf16(bfloat16x4_t __p0) {
39001		int64x1_t __ret;
39002		__ret = (int64x1_t)(__p0);
39003		return __ret;
39004		}
39005		__ai int16x4_t vreinterpret_s16_bf16(bfloat16x4_t __p0) {
39006		int16x4_t __ret;
39007		__ret = (int16x4_t)(__p0);
39008		return __ret;
39009		}
39010		__ai bfloat16x8_t vreinterpretq_bf16_p8(poly8x16_t __p0) {
39011		bfloat16x8_t __ret;
39012		__ret = (bfloat16x8_t)(__p0);
39013		return __ret;
39014		}
39015		__ai bfloat16x8_t vreinterpretq_bf16_p64(poly64x2_t __p0) {
39016		bfloat16x8_t __ret;
39017		__ret = (bfloat16x8_t)(__p0);
39018		return __ret;
39019		}
39020		__ai bfloat16x8_t vreinterpretq_bf16_p16(poly16x8_t __p0) {
39021		bfloat16x8_t __ret;
39022		__ret = (bfloat16x8_t)(__p0);
39023		return __ret;
39024		}
39025		__ai bfloat16x8_t vreinterpretq_bf16_u8(uint8x16_t __p0) {
39026		bfloat16x8_t __ret;
39027		__ret = (bfloat16x8_t)(__p0);
39028		return __ret;
39029		}
39030		__ai bfloat16x8_t vreinterpretq_bf16_u32(uint32x4_t __p0) {
39031		bfloat16x8_t __ret;
39032		__ret = (bfloat16x8_t)(__p0);
39033		return __ret;
39034		}
39035		__ai bfloat16x8_t vreinterpretq_bf16_u64(uint64x2_t __p0) {
39036		bfloat16x8_t __ret;
39037		__ret = (bfloat16x8_t)(__p0);
39038		return __ret;
39039		}
39040		__ai bfloat16x8_t vreinterpretq_bf16_u16(uint16x8_t __p0) {
39041		bfloat16x8_t __ret;
39042		__ret = (bfloat16x8_t)(__p0);
39043		return __ret;
39044		}
39045		__ai bfloat16x8_t vreinterpretq_bf16_s8(int8x16_t __p0) {
39046		bfloat16x8_t __ret;
39047		__ret = (bfloat16x8_t)(__p0);
39048		return __ret;
39049		}
39050		__ai bfloat16x8_t vreinterpretq_bf16_f32(float32x4_t __p0) {
39051		bfloat16x8_t __ret;
39052		__ret = (bfloat16x8_t)(__p0);
39053		return __ret;
39054		}
39055		__ai bfloat16x8_t vreinterpretq_bf16_f16(float16x8_t __p0) {
39056		bfloat16x8_t __ret;
39057		__ret = (bfloat16x8_t)(__p0);
39058		return __ret;
39059		}
39060		__ai bfloat16x8_t vreinterpretq_bf16_s32(int32x4_t __p0) {
39061		bfloat16x8_t __ret;
39062		__ret = (bfloat16x8_t)(__p0);
39063		return __ret;
39064		}
39065		__ai bfloat16x8_t vreinterpretq_bf16_s64(int64x2_t __p0) {
39066		bfloat16x8_t __ret;
39067		__ret = (bfloat16x8_t)(__p0);
39068		return __ret;
39069		}
39070		__ai bfloat16x8_t vreinterpretq_bf16_s16(int16x8_t __p0) {
39071		bfloat16x8_t __ret;
39072		__ret = (bfloat16x8_t)(__p0);
39073		return __ret;
39074		}
39075		__ai bfloat16x4_t vreinterpret_bf16_p8(poly8x8_t __p0) {
39076		bfloat16x4_t __ret;
39077		__ret = (bfloat16x4_t)(__p0);
39078		return __ret;
39079		}
39080		__ai bfloat16x4_t vreinterpret_bf16_p64(poly64x1_t __p0) {
39081		bfloat16x4_t __ret;
39082		__ret = (bfloat16x4_t)(__p0);
39083		return __ret;
39084		}
39085		__ai bfloat16x4_t vreinterpret_bf16_p16(poly16x4_t __p0) {
39086		bfloat16x4_t __ret;
39087		__ret = (bfloat16x4_t)(__p0);
39088		return __ret;
39089		}
39090		__ai bfloat16x4_t vreinterpret_bf16_u8(uint8x8_t __p0) {
39091		bfloat16x4_t __ret;
39092		__ret = (bfloat16x4_t)(__p0);
39093		return __ret;
39094		}
39095		__ai bfloat16x4_t vreinterpret_bf16_u32(uint32x2_t __p0) {
39096		bfloat16x4_t __ret;
39097		__ret = (bfloat16x4_t)(__p0);
39098		return __ret;
39099		}
39100		__ai bfloat16x4_t vreinterpret_bf16_u64(uint64x1_t __p0) {
39101		bfloat16x4_t __ret;
39102		__ret = (bfloat16x4_t)(__p0);
39103		return __ret;
39104		}
39105		__ai bfloat16x4_t vreinterpret_bf16_u16(uint16x4_t __p0) {
39106		bfloat16x4_t __ret;
39107		__ret = (bfloat16x4_t)(__p0);
39108		return __ret;
39109		}
39110		__ai bfloat16x4_t vreinterpret_bf16_s8(int8x8_t __p0) {
39111		bfloat16x4_t __ret;
39112		__ret = (bfloat16x4_t)(__p0);
39113		return __ret;
39114		}
39115		__ai bfloat16x4_t vreinterpret_bf16_f32(float32x2_t __p0) {
39116		bfloat16x4_t __ret;
39117		__ret = (bfloat16x4_t)(__p0);
39118		return __ret;
39119		}
39120		__ai bfloat16x4_t vreinterpret_bf16_f16(float16x4_t __p0) {
39121		bfloat16x4_t __ret;
39122		__ret = (bfloat16x4_t)(__p0);
39123		return __ret;
39124		}
39125		__ai bfloat16x4_t vreinterpret_bf16_s32(int32x2_t __p0) {
39126		bfloat16x4_t __ret;
39127		__ret = (bfloat16x4_t)(__p0);
39128		return __ret;
39129		}
39130		__ai bfloat16x4_t vreinterpret_bf16_s64(int64x1_t __p0) {
39131		bfloat16x4_t __ret;
39132		__ret = (bfloat16x4_t)(__p0);
39133		return __ret;
39134		}
39135		__ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
39136		bfloat16x4_t __ret;
39137		__ret = (bfloat16x4_t)(__p0);
39138		return __ret;
39139		}
39140		#endif
39141		#if defined(__ARM_FEATURE_BF16) && defined(__aarch64__)
39142		__ai poly8x8_t vreinterpret_p8_bf16(bfloat16x4_t __p0) {
39143		poly8x8_t __ret;
39144		__ret = (poly8x8_t)(__p0);
39145		return __ret;
39146		}
39147		__ai poly64x1_t vreinterpret_p64_bf16(bfloat16x4_t __p0) {
39148		poly64x1_t __ret;
39149		__ret = (poly64x1_t)(__p0);
39150		return __ret;
39151		}
39152		__ai poly16x4_t vreinterpret_p16_bf16(bfloat16x4_t __p0) {
39153		poly16x4_t __ret;
39154		__ret = (poly16x4_t)(__p0);
39155		return __ret;
39156		}
39157		__ai poly8x16_t vreinterpretq_p8_bf16(bfloat16x8_t __p0) {
39158		poly8x16_t __ret;
39159		__ret = (poly8x16_t)(__p0);
	40604	__ret = (uint32x4_t) __builtin_neon_vcvtaq_u32_v((int8x16_t)__p0, 50);
39160	40605	return __ret;
39161	40606	}
39162		__ai poly128_t vreinterpretq_p128_bf16(bfloat16x8_t __p0) {
39163		poly128_t __ret;
39164		__ret = (poly128_t)(__p0);
	40607	#else
	40608	__ai uint32x4_t vcvtaq_u32_f32(float32x4_t __p0) {
	40609	uint32x4_t __ret;
	40610	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40611	__ret = (uint32x4_t) __builtin_neon_vcvtaq_u32_v((int8x16_t)__rev0, 50);
	40612	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39165	40613	return __ret;
39166	40614	}
39167		__ai poly64x2_t vreinterpretq_p64_bf16(bfloat16x8_t __p0) {
39168		poly64x2_t __ret;
39169		__ret = (poly64x2_t)(__p0);
	40615	#endif
	40616
	40617	#ifdef __LITTLE_ENDIAN__
	40618	__ai uint32x2_t vcvta_u32_f32(float32x2_t __p0) {
	40619	uint32x2_t __ret;
	40620	__ret = (uint32x2_t) __builtin_neon_vcvta_u32_v((int8x8_t)__p0, 18);
39170	40621	return __ret;
39171	40622	}
39172		__ai poly16x8_t vreinterpretq_p16_bf16(bfloat16x8_t __p0) {
39173		poly16x8_t __ret;
39174		__ret = (poly16x8_t)(__p0);
	40623	#else
	40624	__ai uint32x2_t vcvta_u32_f32(float32x2_t __p0) {
	40625	uint32x2_t __ret;
	40626	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40627	__ret = (uint32x2_t) __builtin_neon_vcvta_u32_v((int8x8_t)__rev0, 18);
	40628	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39175	40629	return __ret;
39176	40630	}
39177		__ai uint8x16_t vreinterpretq_u8_bf16(bfloat16x8_t __p0) {
39178		uint8x16_t __ret;
39179		__ret = (uint8x16_t)(__p0);
	40631	#endif
	40632
	40633	#ifdef __LITTLE_ENDIAN__
	40634	__ai int32x4_t vcvtmq_s32_f32(float32x4_t __p0) {
	40635	int32x4_t __ret;
	40636	__ret = (int32x4_t) __builtin_neon_vcvtmq_s32_v((int8x16_t)__p0, 34);
39180	40637	return __ret;
39181	40638	}
39182		__ai uint32x4_t vreinterpretq_u32_bf16(bfloat16x8_t __p0) {
39183		uint32x4_t __ret;
39184		__ret = (uint32x4_t)(__p0);
	40639	#else
	40640	__ai int32x4_t vcvtmq_s32_f32(float32x4_t __p0) {
	40641	int32x4_t __ret;
	40642	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40643	__ret = (int32x4_t) __builtin_neon_vcvtmq_s32_v((int8x16_t)__rev0, 34);
	40644	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39185	40645	return __ret;
39186	40646	}
39187		__ai uint64x2_t vreinterpretq_u64_bf16(bfloat16x8_t __p0) {
39188		uint64x2_t __ret;
39189		__ret = (uint64x2_t)(__p0);
	40647	#endif
	40648
	40649	#ifdef __LITTLE_ENDIAN__
	40650	__ai int32x2_t vcvtm_s32_f32(float32x2_t __p0) {
	40651	int32x2_t __ret;
	40652	__ret = (int32x2_t) __builtin_neon_vcvtm_s32_v((int8x8_t)__p0, 2);
39190	40653	return __ret;
39191	40654	}
39192		__ai uint16x8_t vreinterpretq_u16_bf16(bfloat16x8_t __p0) {
39193		uint16x8_t __ret;
39194		__ret = (uint16x8_t)(__p0);
	40655	#else
	40656	__ai int32x2_t vcvtm_s32_f32(float32x2_t __p0) {
	40657	int32x2_t __ret;
	40658	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40659	__ret = (int32x2_t) __builtin_neon_vcvtm_s32_v((int8x8_t)__rev0, 2);
	40660	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39195	40661	return __ret;
39196	40662	}
39197		__ai int8x16_t vreinterpretq_s8_bf16(bfloat16x8_t __p0) {
39198		int8x16_t __ret;
39199		__ret = (int8x16_t)(__p0);
	40663	#endif
	40664
	40665	#ifdef __LITTLE_ENDIAN__
	40666	__ai uint32x4_t vcvtmq_u32_f32(float32x4_t __p0) {
	40667	uint32x4_t __ret;
	40668	__ret = (uint32x4_t) __builtin_neon_vcvtmq_u32_v((int8x16_t)__p0, 50);
39200	40669	return __ret;
39201	40670	}
39202		__ai float64x2_t vreinterpretq_f64_bf16(bfloat16x8_t __p0) {
39203		float64x2_t __ret;
39204		__ret = (float64x2_t)(__p0);
	40671	#else
	40672	__ai uint32x4_t vcvtmq_u32_f32(float32x4_t __p0) {
	40673	uint32x4_t __ret;
	40674	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40675	__ret = (uint32x4_t) __builtin_neon_vcvtmq_u32_v((int8x16_t)__rev0, 50);
	40676	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39205	40677	return __ret;
39206	40678	}
39207		__ai float32x4_t vreinterpretq_f32_bf16(bfloat16x8_t __p0) {
39208		float32x4_t __ret;
39209		__ret = (float32x4_t)(__p0);
	40679	#endif
	40680
	40681	#ifdef __LITTLE_ENDIAN__
	40682	__ai uint32x2_t vcvtm_u32_f32(float32x2_t __p0) {
	40683	uint32x2_t __ret;
	40684	__ret = (uint32x2_t) __builtin_neon_vcvtm_u32_v((int8x8_t)__p0, 18);
39210	40685	return __ret;
39211	40686	}
39212		__ai float16x8_t vreinterpretq_f16_bf16(bfloat16x8_t __p0) {
39213		float16x8_t __ret;
39214		__ret = (float16x8_t)(__p0);
	40687	#else
	40688	__ai uint32x2_t vcvtm_u32_f32(float32x2_t __p0) {
	40689	uint32x2_t __ret;
	40690	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40691	__ret = (uint32x2_t) __builtin_neon_vcvtm_u32_v((int8x8_t)__rev0, 18);
	40692	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39215	40693	return __ret;
39216	40694	}
39217		__ai int32x4_t vreinterpretq_s32_bf16(bfloat16x8_t __p0) {
	40695	#endif
	40696
	40697	#ifdef __LITTLE_ENDIAN__
	40698	__ai int32x4_t vcvtnq_s32_f32(float32x4_t __p0) {
39218	40699	int32x4_t __ret;
39219		__ret = (int32x4_t)(__p0);
	40700	__ret = (int32x4_t) __builtin_neon_vcvtnq_s32_v((int8x16_t)__p0, 34);
39220	40701	return __ret;
39221	40702	}
39222		__ai int64x2_t vreinterpretq_s64_bf16(bfloat16x8_t __p0) {
39223		int64x2_t __ret;
39224		__ret = (int64x2_t)(__p0);
	40703	#else
	40704	__ai int32x4_t vcvtnq_s32_f32(float32x4_t __p0) {
	40705	int32x4_t __ret;
	40706	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40707	__ret = (int32x4_t) __builtin_neon_vcvtnq_s32_v((int8x16_t)__rev0, 34);
	40708	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39225	40709	return __ret;
39226	40710	}
39227		__ai int16x8_t vreinterpretq_s16_bf16(bfloat16x8_t __p0) {
39228		int16x8_t __ret;
39229		__ret = (int16x8_t)(__p0);
	40711	#endif
	40712
	40713	#ifdef __LITTLE_ENDIAN__
	40714	__ai int32x2_t vcvtn_s32_f32(float32x2_t __p0) {
	40715	int32x2_t __ret;
	40716	__ret = (int32x2_t) __builtin_neon_vcvtn_s32_v((int8x8_t)__p0, 2);
39230	40717	return __ret;
39231	40718	}
39232		__ai uint8x8_t vreinterpret_u8_bf16(bfloat16x4_t __p0) {
39233		uint8x8_t __ret;
39234		__ret = (uint8x8_t)(__p0);
	40719	#else
	40720	__ai int32x2_t vcvtn_s32_f32(float32x2_t __p0) {
	40721	int32x2_t __ret;
	40722	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40723	__ret = (int32x2_t) __builtin_neon_vcvtn_s32_v((int8x8_t)__rev0, 2);
	40724	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39235	40725	return __ret;
39236	40726	}
39237		__ai uint32x2_t vreinterpret_u32_bf16(bfloat16x4_t __p0) {
39238		uint32x2_t __ret;
39239		__ret = (uint32x2_t)(__p0);
	40727	#endif
	40728
	40729	#ifdef __LITTLE_ENDIAN__
	40730	__ai uint32x4_t vcvtnq_u32_f32(float32x4_t __p0) {
	40731	uint32x4_t __ret;
	40732	__ret = (uint32x4_t) __builtin_neon_vcvtnq_u32_v((int8x16_t)__p0, 50);
39240	40733	return __ret;
39241	40734	}
39242		__ai uint64x1_t vreinterpret_u64_bf16(bfloat16x4_t __p0) {
39243		uint64x1_t __ret;
39244		__ret = (uint64x1_t)(__p0);
	40735	#else
	40736	__ai uint32x4_t vcvtnq_u32_f32(float32x4_t __p0) {
	40737	uint32x4_t __ret;
	40738	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40739	__ret = (uint32x4_t) __builtin_neon_vcvtnq_u32_v((int8x16_t)__rev0, 50);
	40740	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39245	40741	return __ret;
39246	40742	}
39247		__ai uint16x4_t vreinterpret_u16_bf16(bfloat16x4_t __p0) {
39248		uint16x4_t __ret;
39249		__ret = (uint16x4_t)(__p0);
	40743	#endif
	40744
	40745	#ifdef __LITTLE_ENDIAN__
	40746	__ai uint32x2_t vcvtn_u32_f32(float32x2_t __p0) {
	40747	uint32x2_t __ret;
	40748	__ret = (uint32x2_t) __builtin_neon_vcvtn_u32_v((int8x8_t)__p0, 18);
39250	40749	return __ret;
39251	40750	}
39252		__ai int8x8_t vreinterpret_s8_bf16(bfloat16x4_t __p0) {
39253		int8x8_t __ret;
39254		__ret = (int8x8_t)(__p0);
	40751	#else
	40752	__ai uint32x2_t vcvtn_u32_f32(float32x2_t __p0) {
	40753	uint32x2_t __ret;
	40754	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40755	__ret = (uint32x2_t) __builtin_neon_vcvtn_u32_v((int8x8_t)__rev0, 18);
	40756	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39255	40757	return __ret;
39256	40758	}
39257		__ai float64x1_t vreinterpret_f64_bf16(bfloat16x4_t __p0) {
39258		float64x1_t __ret;
39259		__ret = (float64x1_t)(__p0);
	40759	#endif
	40760
	40761	#ifdef __LITTLE_ENDIAN__
	40762	__ai int32x4_t vcvtpq_s32_f32(float32x4_t __p0) {
	40763	int32x4_t __ret;
	40764	__ret = (int32x4_t) __builtin_neon_vcvtpq_s32_v((int8x16_t)__p0, 34);
39260	40765	return __ret;
39261	40766	}
39262		__ai float32x2_t vreinterpret_f32_bf16(bfloat16x4_t __p0) {
39263		float32x2_t __ret;
39264		__ret = (float32x2_t)(__p0);
	40767	#else
	40768	__ai int32x4_t vcvtpq_s32_f32(float32x4_t __p0) {
	40769	int32x4_t __ret;
	40770	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40771	__ret = (int32x4_t) __builtin_neon_vcvtpq_s32_v((int8x16_t)__rev0, 34);
	40772	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39265	40773	return __ret;
39266	40774	}
39267		__ai float16x4_t vreinterpret_f16_bf16(bfloat16x4_t __p0) {
39268		float16x4_t __ret;
39269		__ret = (float16x4_t)(__p0);
	40775	#endif
	40776
	40777	#ifdef __LITTLE_ENDIAN__
	40778	__ai int32x2_t vcvtp_s32_f32(float32x2_t __p0) {
	40779	int32x2_t __ret;
	40780	__ret = (int32x2_t) __builtin_neon_vcvtp_s32_v((int8x8_t)__p0, 2);
39270	40781	return __ret;
39271	40782	}
39272		__ai int32x2_t vreinterpret_s32_bf16(bfloat16x4_t __p0) {
	40783	#else
	40784	__ai int32x2_t vcvtp_s32_f32(float32x2_t __p0) {
39273	40785	int32x2_t __ret;
39274		__ret = (int32x2_t)(__p0);
	40786	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40787	__ret = (int32x2_t) __builtin_neon_vcvtp_s32_v((int8x8_t)__rev0, 2);
	40788	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39275	40789	return __ret;
39276	40790	}
39277		__ai int64x1_t vreinterpret_s64_bf16(bfloat16x4_t __p0) {
39278		int64x1_t __ret;
39279		__ret = (int64x1_t)(__p0);
	40791	#endif
	40792
	40793	#ifdef __LITTLE_ENDIAN__
	40794	__ai uint32x4_t vcvtpq_u32_f32(float32x4_t __p0) {
	40795	uint32x4_t __ret;
	40796	__ret = (uint32x4_t) __builtin_neon_vcvtpq_u32_v((int8x16_t)__p0, 50);
39280	40797	return __ret;
39281	40798	}
39282		__ai int16x4_t vreinterpret_s16_bf16(bfloat16x4_t __p0) {
39283		int16x4_t __ret;
39284		__ret = (int16x4_t)(__p0);
	40799	#else
	40800	__ai uint32x4_t vcvtpq_u32_f32(float32x4_t __p0) {
	40801	uint32x4_t __ret;
	40802	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40803	__ret = (uint32x4_t) __builtin_neon_vcvtpq_u32_v((int8x16_t)__rev0, 50);
	40804	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39285	40805	return __ret;
39286	40806	}
39287		__ai bfloat16x8_t vreinterpretq_bf16_p8(poly8x16_t __p0) {
39288		bfloat16x8_t __ret;
39289		__ret = (bfloat16x8_t)(__p0);
	40807	#endif
	40808
	40809	#ifdef __LITTLE_ENDIAN__
	40810	__ai uint32x2_t vcvtp_u32_f32(float32x2_t __p0) {
	40811	uint32x2_t __ret;
	40812	__ret = (uint32x2_t) __builtin_neon_vcvtp_u32_v((int8x8_t)__p0, 18);
39290	40813	return __ret;
39291	40814	}
39292		__ai bfloat16x8_t vreinterpretq_bf16_p128(poly128_t __p0) {
39293		bfloat16x8_t __ret;
39294		__ret = (bfloat16x8_t)(__p0);
	40815	#else
	40816	__ai uint32x2_t vcvtp_u32_f32(float32x2_t __p0) {
	40817	uint32x2_t __ret;
	40818	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	40819	__ret = (uint32x2_t) __builtin_neon_vcvtp_u32_v((int8x8_t)__rev0, 18);
	40820	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39295	40821	return __ret;
39296	40822	}
39297		__ai bfloat16x8_t vreinterpretq_bf16_p64(poly64x2_t __p0) {
39298		bfloat16x8_t __ret;
39299		__ret = (bfloat16x8_t)(__p0);
	40823	#endif
	40824
	40825	#ifdef __LITTLE_ENDIAN__
	40826	__ai __attribute__((target("aes"))) uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	40827	uint8x16_t __ret;
	40828	__ret = (uint8x16_t) __builtin_neon_vaesdq_u8((int8x16_t)__p0, (int8x16_t)__p1, 48);
39300	40829	return __ret;
39301	40830	}
39302		__ai bfloat16x8_t vreinterpretq_bf16_p16(poly16x8_t __p0) {
39303		bfloat16x8_t __ret;
39304		__ret = (bfloat16x8_t)(__p0);
	40831	#else
	40832	__ai __attribute__((target("aes"))) uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	40833	uint8x16_t __ret;
	40834	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	40835	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	40836	__ret = (uint8x16_t) __builtin_neon_vaesdq_u8((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	40837	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39305	40838	return __ret;
39306	40839	}
39307		__ai bfloat16x8_t vreinterpretq_bf16_u8(uint8x16_t __p0) {
39308		bfloat16x8_t __ret;
39309		__ret = (bfloat16x8_t)(__p0);
	40840	#endif
	40841
	40842	#ifdef __LITTLE_ENDIAN__
	40843	__ai __attribute__((target("aes"))) uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	40844	uint8x16_t __ret;
	40845	__ret = (uint8x16_t) __builtin_neon_vaeseq_u8((int8x16_t)__p0, (int8x16_t)__p1, 48);
39310	40846	return __ret;
39311	40847	}
39312		__ai bfloat16x8_t vreinterpretq_bf16_u32(uint32x4_t __p0) {
39313		bfloat16x8_t __ret;
39314		__ret = (bfloat16x8_t)(__p0);
	40848	#else
	40849	__ai __attribute__((target("aes"))) uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	40850	uint8x16_t __ret;
	40851	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	40852	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	40853	__ret = (uint8x16_t) __builtin_neon_vaeseq_u8((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	40854	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39315	40855	return __ret;
39316	40856	}
39317		__ai bfloat16x8_t vreinterpretq_bf16_u64(uint64x2_t __p0) {
39318		bfloat16x8_t __ret;
39319		__ret = (bfloat16x8_t)(__p0);
	40857	#endif
	40858
	40859	#ifdef __LITTLE_ENDIAN__
	40860	__ai __attribute__((target("aes"))) uint8x16_t vaesimcq_u8(uint8x16_t __p0) {
	40861	uint8x16_t __ret;
	40862	__ret = (uint8x16_t) __builtin_neon_vaesimcq_u8((int8x16_t)__p0, 48);
39320	40863	return __ret;
39321	40864	}
39322		__ai bfloat16x8_t vreinterpretq_bf16_u16(uint16x8_t __p0) {
39323		bfloat16x8_t __ret;
39324		__ret = (bfloat16x8_t)(__p0);
	40865	#else
	40866	__ai __attribute__((target("aes"))) uint8x16_t vaesimcq_u8(uint8x16_t __p0) {
	40867	uint8x16_t __ret;
	40868	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	40869	__ret = (uint8x16_t) __builtin_neon_vaesimcq_u8((int8x16_t)__rev0, 48);
	40870	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39325	40871	return __ret;
39326	40872	}
39327		__ai bfloat16x8_t vreinterpretq_bf16_s8(int8x16_t __p0) {
39328		bfloat16x8_t __ret;
39329		__ret = (bfloat16x8_t)(__p0);
	40873	#endif
	40874
	40875	#ifdef __LITTLE_ENDIAN__
	40876	__ai __attribute__((target("aes"))) uint8x16_t vaesmcq_u8(uint8x16_t __p0) {
	40877	uint8x16_t __ret;
	40878	__ret = (uint8x16_t) __builtin_neon_vaesmcq_u8((int8x16_t)__p0, 48);
39330	40879	return __ret;
39331	40880	}
39332		__ai bfloat16x8_t vreinterpretq_bf16_f64(float64x2_t __p0) {
39333		bfloat16x8_t __ret;
39334		__ret = (bfloat16x8_t)(__p0);
	40881	#else
	40882	__ai __attribute__((target("aes"))) uint8x16_t vaesmcq_u8(uint8x16_t __p0) {
	40883	uint8x16_t __ret;
	40884	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	40885	__ret = (uint8x16_t) __builtin_neon_vaesmcq_u8((int8x16_t)__rev0, 48);
	40886	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39335	40887	return __ret;
39336	40888	}
39337		__ai bfloat16x8_t vreinterpretq_bf16_f32(float32x4_t __p0) {
39338		bfloat16x8_t __ret;
39339		__ret = (bfloat16x8_t)(__p0);
	40889	#endif
	40890
	40891	#ifdef __LITTLE_ENDIAN__
	40892	__ai __attribute__((target("sha2"))) uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
	40893	uint32x4_t __ret;
	40894	__ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__p0, __p1, __p2);
39340	40895	return __ret;
39341	40896	}
39342		__ai bfloat16x8_t vreinterpretq_bf16_f16(float16x8_t __p0) {
39343		bfloat16x8_t __ret;
39344		__ret = (bfloat16x8_t)(__p0);
	40897	#else
	40898	__ai __attribute__((target("sha2"))) uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
	40899	uint32x4_t __ret;
	40900	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40901	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	40902	__ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__rev0, __p1, __rev2);
	40903	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39345	40904	return __ret;
39346	40905	}
39347		__ai bfloat16x8_t vreinterpretq_bf16_s32(int32x4_t __p0) {
39348		bfloat16x8_t __ret;
39349		__ret = (bfloat16x8_t)(__p0);
	40906	#endif
	40907
	40908	__ai __attribute__((target("sha2"))) uint32_t vsha1h_u32(uint32_t __p0) {
	40909	uint32_t __ret;
	40910	__ret = (uint32_t) __builtin_neon_vsha1h_u32(__p0);
39350	40911	return __ret;
39351	40912	}
39352		__ai bfloat16x8_t vreinterpretq_bf16_s64(int64x2_t __p0) {
39353		bfloat16x8_t __ret;
39354		__ret = (bfloat16x8_t)(__p0);
	40913	#ifdef __LITTLE_ENDIAN__
	40914	__ai __attribute__((target("sha2"))) uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
	40915	uint32x4_t __ret;
	40916	__ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__p0, __p1, __p2);
39355	40917	return __ret;
39356	40918	}
39357		__ai bfloat16x8_t vreinterpretq_bf16_s16(int16x8_t __p0) {
39358		bfloat16x8_t __ret;
39359		__ret = (bfloat16x8_t)(__p0);
	40919	#else
	40920	__ai __attribute__((target("sha2"))) uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
	40921	uint32x4_t __ret;
	40922	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40923	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	40924	__ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__rev0, __p1, __rev2);
	40925	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39360	40926	return __ret;
39361	40927	}
39362		__ai bfloat16x4_t vreinterpret_bf16_p8(poly8x8_t __p0) {
39363		bfloat16x4_t __ret;
39364		__ret = (bfloat16x4_t)(__p0);
	40928	#endif
	40929
	40930	#ifdef __LITTLE_ENDIAN__
	40931	__ai __attribute__((target("sha2"))) uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
	40932	uint32x4_t __ret;
	40933	__ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__p0, __p1, __p2);
39365	40934	return __ret;
39366	40935	}
39367		__ai bfloat16x4_t vreinterpret_bf16_p64(poly64x1_t __p0) {
39368		bfloat16x4_t __ret;
39369		__ret = (bfloat16x4_t)(__p0);
	40936	#else
	40937	__ai __attribute__((target("sha2"))) uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
	40938	uint32x4_t __ret;
	40939	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40940	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	40941	__ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__rev0, __p1, __rev2);
	40942	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39370	40943	return __ret;
39371	40944	}
39372		__ai bfloat16x4_t vreinterpret_bf16_p16(poly16x4_t __p0) {
39373		bfloat16x4_t __ret;
39374		__ret = (bfloat16x4_t)(__p0);
	40945	#endif
	40946
	40947	#ifdef __LITTLE_ENDIAN__
	40948	__ai __attribute__((target("sha2"))) uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	40949	uint32x4_t __ret;
	40950	__ret = (uint32x4_t) __builtin_neon_vsha1su0q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39375	40951	return __ret;
39376	40952	}
39377		__ai bfloat16x4_t vreinterpret_bf16_u8(uint8x8_t __p0) {
39378		bfloat16x4_t __ret;
39379		__ret = (bfloat16x4_t)(__p0);
	40953	#else
	40954	__ai __attribute__((target("sha2"))) uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	40955	uint32x4_t __ret;
	40956	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40957	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	40958	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	40959	__ret = (uint32x4_t) __builtin_neon_vsha1su0q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	40960	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39380	40961	return __ret;
39381	40962	}
39382		__ai bfloat16x4_t vreinterpret_bf16_u32(uint32x2_t __p0) {
39383		bfloat16x4_t __ret;
39384		__ret = (bfloat16x4_t)(__p0);
	40963	#endif
	40964
	40965	#ifdef __LITTLE_ENDIAN__
	40966	__ai __attribute__((target("sha2"))) uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	40967	uint32x4_t __ret;
	40968	__ret = (uint32x4_t) __builtin_neon_vsha1su1q_u32((int8x16_t)__p0, (int8x16_t)__p1, 50);
39385	40969	return __ret;
39386	40970	}
39387		__ai bfloat16x4_t vreinterpret_bf16_u64(uint64x1_t __p0) {
39388		bfloat16x4_t __ret;
39389		__ret = (bfloat16x4_t)(__p0);
	40971	#else
	40972	__ai __attribute__((target("sha2"))) uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	40973	uint32x4_t __ret;
	40974	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40975	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	40976	__ret = (uint32x4_t) __builtin_neon_vsha1su1q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	40977	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39390	40978	return __ret;
39391	40979	}
39392		__ai bfloat16x4_t vreinterpret_bf16_u16(uint16x4_t __p0) {
39393		bfloat16x4_t __ret;
39394		__ret = (bfloat16x4_t)(__p0);
	40980	#endif
	40981
	40982	#ifdef __LITTLE_ENDIAN__
	40983	__ai __attribute__((target("sha2"))) uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	40984	uint32x4_t __ret;
	40985	__ret = (uint32x4_t) __builtin_neon_vsha256hq_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39395	40986	return __ret;
39396	40987	}
39397		__ai bfloat16x4_t vreinterpret_bf16_s8(int8x8_t __p0) {
39398		bfloat16x4_t __ret;
39399		__ret = (bfloat16x4_t)(__p0);
	40988	#else
	40989	__ai __attribute__((target("sha2"))) uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	40990	uint32x4_t __ret;
	40991	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	40992	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	40993	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	40994	__ret = (uint32x4_t) __builtin_neon_vsha256hq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	40995	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39400	40996	return __ret;
39401	40997	}
39402		__ai bfloat16x4_t vreinterpret_bf16_f64(float64x1_t __p0) {
39403		bfloat16x4_t __ret;
39404		__ret = (bfloat16x4_t)(__p0);
	40998	#endif
	40999
	41000	#ifdef __LITTLE_ENDIAN__
	41001	__ai __attribute__((target("sha2"))) uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	41002	uint32x4_t __ret;
	41003	__ret = (uint32x4_t) __builtin_neon_vsha256h2q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39405	41004	return __ret;
39406	41005	}
39407		__ai bfloat16x4_t vreinterpret_bf16_f32(float32x2_t __p0) {
39408		bfloat16x4_t __ret;
39409		__ret = (bfloat16x4_t)(__p0);
	41006	#else
	41007	__ai __attribute__((target("sha2"))) uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	41008	uint32x4_t __ret;
	41009	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41010	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41011	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	41012	__ret = (uint32x4_t) __builtin_neon_vsha256h2q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	41013	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39410	41014	return __ret;
39411	41015	}
39412		__ai bfloat16x4_t vreinterpret_bf16_f16(float16x4_t __p0) {
39413		bfloat16x4_t __ret;
39414		__ret = (bfloat16x4_t)(__p0);
	41016	#endif
	41017
	41018	#ifdef __LITTLE_ENDIAN__
	41019	__ai __attribute__((target("sha2"))) uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	41020	uint32x4_t __ret;
	41021	__ret = (uint32x4_t) __builtin_neon_vsha256su0q_u32((int8x16_t)__p0, (int8x16_t)__p1, 50);
39415	41022	return __ret;
39416	41023	}
39417		__ai bfloat16x4_t vreinterpret_bf16_s32(int32x2_t __p0) {
39418		bfloat16x4_t __ret;
39419		__ret = (bfloat16x4_t)(__p0);
	41024	#else
	41025	__ai __attribute__((target("sha2"))) uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	41026	uint32x4_t __ret;
	41027	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41028	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41029	__ret = (uint32x4_t) __builtin_neon_vsha256su0q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	41030	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39420	41031	return __ret;
39421	41032	}
39422		__ai bfloat16x4_t vreinterpret_bf16_s64(int64x1_t __p0) {
39423		bfloat16x4_t __ret;
39424		__ret = (bfloat16x4_t)(__p0);
	41033	#endif
	41034
	41035	#ifdef __LITTLE_ENDIAN__
	41036	__ai __attribute__((target("sha2"))) uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	41037	uint32x4_t __ret;
	41038	__ret = (uint32x4_t) __builtin_neon_vsha256su1q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39425	41039	return __ret;
39426	41040	}
39427		__ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
39428		bfloat16x4_t __ret;
39429		__ret = (bfloat16x4_t)(__p0);
	41041	#else
	41042	__ai __attribute__((target("sha2"))) uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	41043	uint32x4_t __ret;
	41044	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41045	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41046	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	41047	__ret = (uint32x4_t) __builtin_neon_vsha256su1q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	41048	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39430	41049	return __ret;
39431	41050	}
39432	41051	#endif
39433		#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC)
	41052
	41053	#endif
	41054	#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_DIRECTED_ROUNDING)
39434	41055	#ifdef __LITTLE_ENDIAN__
39435		#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
39436		bfloat16x8_t __ret; \
39437		bfloat16x4_t __s0 = __p0; \
39438		__ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 11); \
39439		__ret; \
39440		})
	41056	__ai float32x4_t vrndq_f32(float32x4_t __p0) {
	41057	float32x4_t __ret;
	41058	__ret = (float32x4_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 41);
	41059	return __ret;
	41060	}
39441	41061	#else
39442		#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
39443		bfloat16x8_t __ret; \
39444		bfloat16x4_t __s0 = __p0; \
39445		bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
39446		__ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 11); \
39447		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
39448		__ret; \
39449		})
39450		#define __noswap_splatq_lane_bf16(__p0, __p1) __extension__ ({ \
39451		bfloat16x8_t __ret; \
39452		bfloat16x4_t __s0 = __p0; \
39453		__ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 11); \
39454		__ret; \
39455		})
	41062	__ai float32x4_t vrndq_f32(float32x4_t __p0) {
	41063	float32x4_t __ret;
	41064	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41065	__ret = (float32x4_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 41);
	41066	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41067	return __ret;
	41068	}
39456	41069	#endif
39457	41070
39458	41071	#ifdef __LITTLE_ENDIAN__
39459		#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
39460		bfloat16x4_t __ret; \
39461		bfloat16x4_t __s0 = __p0; \
39462		__ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 11); \
39463		__ret; \
39464		})
	41072	__ai float32x2_t vrnd_f32(float32x2_t __p0) {
	41073	float32x2_t __ret;
	41074	__ret = (float32x2_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 9);
	41075	return __ret;
	41076	}
39465	41077	#else
39466		#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
39467		bfloat16x4_t __ret; \
39468		bfloat16x4_t __s0 = __p0; \
39469		bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
39470		__ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 11); \
39471		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
39472		__ret; \
39473		})
39474		#define __noswap_splat_lane_bf16(__p0, __p1) __extension__ ({ \
39475		bfloat16x4_t __ret; \
39476		bfloat16x4_t __s0 = __p0; \
39477		__ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 11); \
39478		__ret; \
39479		})
	41078	__ai float32x2_t vrnd_f32(float32x2_t __p0) {
	41079	float32x2_t __ret;
	41080	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41081	__ret = (float32x2_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 9);
	41082	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41083	return __ret;
	41084	}
39480	41085	#endif
39481	41086
39482	41087	#ifdef __LITTLE_ENDIAN__
39483		#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
39484		bfloat16x8_t __ret; \
39485		bfloat16x8_t __s0 = __p0; \
39486		__ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 43); \
39487		__ret; \
39488		})
	41088	__ai float32x4_t vrndaq_f32(float32x4_t __p0) {
	41089	float32x4_t __ret;
	41090	__ret = (float32x4_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 41);
	41091	return __ret;
	41092	}
39489	41093	#else
39490		#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
39491		bfloat16x8_t __ret; \
39492		bfloat16x8_t __s0 = __p0; \
39493		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39494		__ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 43); \
39495		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
39496		__ret; \
39497		})
39498		#define __noswap_splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
39499		bfloat16x8_t __ret; \
39500		bfloat16x8_t __s0 = __p0; \
39501		__ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 43); \
39502		__ret; \
39503		})
	41094	__ai float32x4_t vrndaq_f32(float32x4_t __p0) {
	41095	float32x4_t __ret;
	41096	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41097	__ret = (float32x4_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 41);
	41098	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41099	return __ret;
	41100	}
39504	41101	#endif
39505	41102
39506	41103	#ifdef __LITTLE_ENDIAN__
39507		#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
39508		bfloat16x4_t __ret; \
39509		bfloat16x8_t __s0 = __p0; \
39510		__ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 43); \
39511		__ret; \
39512		})
	41104	__ai float32x2_t vrnda_f32(float32x2_t __p0) {
	41105	float32x2_t __ret;
	41106	__ret = (float32x2_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 9);
	41107	return __ret;
	41108	}
39513	41109	#else
39514		#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
39515		bfloat16x4_t __ret; \
39516		bfloat16x8_t __s0 = __p0; \
39517		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39518		__ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 43); \
39519		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
39520		__ret; \
39521		})
39522		#define __noswap_splat_laneq_bf16(__p0, __p1) __extension__ ({ \
39523		bfloat16x4_t __ret; \
39524		bfloat16x8_t __s0 = __p0; \
39525		__ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 43); \
39526		__ret; \
39527		})
	41110	__ai float32x2_t vrnda_f32(float32x2_t __p0) {
	41111	float32x2_t __ret;
	41112	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41113	__ret = (float32x2_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 9);
	41114	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41115	return __ret;
	41116	}
39528	41117	#endif
39529	41118
39530	41119	#ifdef __LITTLE_ENDIAN__
39531		__ai float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	41120	__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
39532	41121	float32x4_t __ret;
39533		__ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41122	__ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 41);
39534	41123	return __ret;
39535	41124	}
39536	41125	#else
39537		__ai float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	41126	__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
39538	41127	float32x4_t __ret;
39539	41128	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39540		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
39541		bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39542		__ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	41129	__ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 41);
39543	41130	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39544	41131	return __ret;
39545	41132	}
39546		__ai float32x4_t __noswap_vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39547		float32x4_t __ret;
39548		__ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
39549		return __ret;
39550		}
39551	41133	#endif
39552	41134
39553	41135	#ifdef __LITTLE_ENDIAN__
39554		__ai float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
	41136	__ai float32x2_t vrndi_f32(float32x2_t __p0) {
39555	41137	float32x2_t __ret;
39556		__ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	41138	__ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 9);
39557	41139	return __ret;
39558	41140	}
39559	41141	#else
39560		__ai float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
	41142	__ai float32x2_t vrndi_f32(float32x2_t __p0) {
39561	41143	float32x2_t __ret;
39562	41144	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
39563		bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39564		bfloat16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
39565		__ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	41145	__ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 9);
39566	41146	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39567	41147	return __ret;
39568	41148	}
39569		__ai float32x2_t __noswap_vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
39570		float32x2_t __ret;
39571		__ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
39572		return __ret;
39573		}
39574		#endif
39575
39576		#ifdef __LITTLE_ENDIAN__
39577		#define vbfdotq_lane_f32(__p0_142, __p1_142, __p2_142, __p3_142) __extension__ ({ \
39578		float32x4_t __ret_142; \
39579		float32x4_t __s0_142 = __p0_142; \
39580		bfloat16x8_t __s1_142 = __p1_142; \
39581		bfloat16x4_t __s2_142 = __p2_142; \
39582		bfloat16x4_t __reint_142 = __s2_142; \
39583		float32x4_t __reint1_142 = splatq_lane_f32((float32x2_t ) &__reint_142, __p3_142); \
39584		__ret_142 = vbfdotq_f32(__s0_142, __s1_142, (bfloat16x8_t ) &__reint1_142); \
39585		__ret_142; \
39586		})
39587		#else
39588		#define vbfdotq_lane_f32(__p0_143, __p1_143, __p2_143, __p3_143) __extension__ ({ \
39589		float32x4_t __ret_143; \
39590		float32x4_t __s0_143 = __p0_143; \
39591		bfloat16x8_t __s1_143 = __p1_143; \
39592		bfloat16x4_t __s2_143 = __p2_143; \
39593		float32x4_t __rev0_143; __rev0_143 = __builtin_shufflevector(__s0_143, __s0_143, 3, 2, 1, 0); \
39594		bfloat16x8_t __rev1_143; __rev1_143 = __builtin_shufflevector(__s1_143, __s1_143, 7, 6, 5, 4, 3, 2, 1, 0); \
39595		bfloat16x4_t __rev2_143; __rev2_143 = __builtin_shufflevector(__s2_143, __s2_143, 3, 2, 1, 0); \
39596		bfloat16x4_t __reint_143 = __rev2_143; \
39597		float32x4_t __reint1_143 = __noswap_splatq_lane_f32((float32x2_t ) &__reint_143, __p3_143); \
39598		__ret_143 = __noswap_vbfdotq_f32(__rev0_143, __rev1_143, (bfloat16x8_t ) &__reint1_143); \
39599		__ret_143 = __builtin_shufflevector(__ret_143, __ret_143, 3, 2, 1, 0); \
39600		__ret_143; \
39601		})
39602		#endif
39603
39604		#ifdef __LITTLE_ENDIAN__
39605		#define vbfdot_lane_f32(__p0_144, __p1_144, __p2_144, __p3_144) __extension__ ({ \
39606		float32x2_t __ret_144; \
39607		float32x2_t __s0_144 = __p0_144; \
39608		bfloat16x4_t __s1_144 = __p1_144; \
39609		bfloat16x4_t __s2_144 = __p2_144; \
39610		bfloat16x4_t __reint_144 = __s2_144; \
39611		float32x2_t __reint1_144 = splat_lane_f32((float32x2_t ) &__reint_144, __p3_144); \
39612		__ret_144 = vbfdot_f32(__s0_144, __s1_144, (bfloat16x4_t ) &__reint1_144); \
39613		__ret_144; \
39614		})
39615		#else
39616		#define vbfdot_lane_f32(__p0_145, __p1_145, __p2_145, __p3_145) __extension__ ({ \
39617		float32x2_t __ret_145; \
39618		float32x2_t __s0_145 = __p0_145; \
39619		bfloat16x4_t __s1_145 = __p1_145; \
39620		bfloat16x4_t __s2_145 = __p2_145; \
39621		float32x2_t __rev0_145; __rev0_145 = __builtin_shufflevector(__s0_145, __s0_145, 1, 0); \
39622		bfloat16x4_t __rev1_145; __rev1_145 = __builtin_shufflevector(__s1_145, __s1_145, 3, 2, 1, 0); \
39623		bfloat16x4_t __rev2_145; __rev2_145 = __builtin_shufflevector(__s2_145, __s2_145, 3, 2, 1, 0); \
39624		bfloat16x4_t __reint_145 = __rev2_145; \
39625		float32x2_t __reint1_145 = __noswap_splat_lane_f32((float32x2_t ) &__reint_145, __p3_145); \
39626		__ret_145 = __noswap_vbfdot_f32(__rev0_145, __rev1_145, (bfloat16x4_t ) &__reint1_145); \
39627		__ret_145 = __builtin_shufflevector(__ret_145, __ret_145, 1, 0); \
39628		__ret_145; \
39629		})
39630		#endif
39631
39632		#ifdef __LITTLE_ENDIAN__
39633		#define vbfdotq_laneq_f32(__p0_146, __p1_146, __p2_146, __p3_146) __extension__ ({ \
39634		float32x4_t __ret_146; \
39635		float32x4_t __s0_146 = __p0_146; \
39636		bfloat16x8_t __s1_146 = __p1_146; \
39637		bfloat16x8_t __s2_146 = __p2_146; \
39638		bfloat16x8_t __reint_146 = __s2_146; \
39639		float32x4_t __reint1_146 = splatq_laneq_f32((float32x4_t ) &__reint_146, __p3_146); \
39640		__ret_146 = vbfdotq_f32(__s0_146, __s1_146, (bfloat16x8_t ) &__reint1_146); \
39641		__ret_146; \
39642		})
39643		#else
39644		#define vbfdotq_laneq_f32(__p0_147, __p1_147, __p2_147, __p3_147) __extension__ ({ \
39645		float32x4_t __ret_147; \
39646		float32x4_t __s0_147 = __p0_147; \
39647		bfloat16x8_t __s1_147 = __p1_147; \
39648		bfloat16x8_t __s2_147 = __p2_147; \
39649		float32x4_t __rev0_147; __rev0_147 = __builtin_shufflevector(__s0_147, __s0_147, 3, 2, 1, 0); \
39650		bfloat16x8_t __rev1_147; __rev1_147 = __builtin_shufflevector(__s1_147, __s1_147, 7, 6, 5, 4, 3, 2, 1, 0); \
39651		bfloat16x8_t __rev2_147; __rev2_147 = __builtin_shufflevector(__s2_147, __s2_147, 7, 6, 5, 4, 3, 2, 1, 0); \
39652		bfloat16x8_t __reint_147 = __rev2_147; \
39653		float32x4_t __reint1_147 = __noswap_splatq_laneq_f32((float32x4_t ) &__reint_147, __p3_147); \
39654		__ret_147 = __noswap_vbfdotq_f32(__rev0_147, __rev1_147, (bfloat16x8_t ) &__reint1_147); \
39655		__ret_147 = __builtin_shufflevector(__ret_147, __ret_147, 3, 2, 1, 0); \
39656		__ret_147; \
39657		})
39658		#endif
39659
39660		#ifdef __LITTLE_ENDIAN__
39661		#define vbfdot_laneq_f32(__p0_148, __p1_148, __p2_148, __p3_148) __extension__ ({ \
39662		float32x2_t __ret_148; \
39663		float32x2_t __s0_148 = __p0_148; \
39664		bfloat16x4_t __s1_148 = __p1_148; \
39665		bfloat16x8_t __s2_148 = __p2_148; \
39666		bfloat16x8_t __reint_148 = __s2_148; \
39667		float32x2_t __reint1_148 = splat_laneq_f32((float32x4_t ) &__reint_148, __p3_148); \
39668		__ret_148 = vbfdot_f32(__s0_148, __s1_148, (bfloat16x4_t ) &__reint1_148); \
39669		__ret_148; \
39670		})
39671		#else
39672		#define vbfdot_laneq_f32(__p0_149, __p1_149, __p2_149, __p3_149) __extension__ ({ \
39673		float32x2_t __ret_149; \
39674		float32x2_t __s0_149 = __p0_149; \
39675		bfloat16x4_t __s1_149 = __p1_149; \
39676		bfloat16x8_t __s2_149 = __p2_149; \
39677		float32x2_t __rev0_149; __rev0_149 = __builtin_shufflevector(__s0_149, __s0_149, 1, 0); \
39678		bfloat16x4_t __rev1_149; __rev1_149 = __builtin_shufflevector(__s1_149, __s1_149, 3, 2, 1, 0); \
39679		bfloat16x8_t __rev2_149; __rev2_149 = __builtin_shufflevector(__s2_149, __s2_149, 7, 6, 5, 4, 3, 2, 1, 0); \
39680		bfloat16x8_t __reint_149 = __rev2_149; \
39681		float32x2_t __reint1_149 = __noswap_splat_laneq_f32((float32x4_t ) &__reint_149, __p3_149); \
39682		__ret_149 = __noswap_vbfdot_f32(__rev0_149, __rev1_149, (bfloat16x4_t ) &__reint1_149); \
39683		__ret_149 = __builtin_shufflevector(__ret_149, __ret_149, 1, 0); \
39684		__ret_149; \
39685		})
39686	41149	#endif
39687	41150
39688	41151	#ifdef __LITTLE_ENDIAN__
39689		__ai float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	41152	__ai float32x4_t vrndmq_f32(float32x4_t __p0) {
39690	41153	float32x4_t __ret;
39691		__ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41154	__ret = (float32x4_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 41);
39692	41155	return __ret;
39693	41156	}
39694	41157	#else
39695		__ai float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	41158	__ai float32x4_t vrndmq_f32(float32x4_t __p0) {
39696	41159	float32x4_t __ret;
39697	41160	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39698		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
39699		bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39700		__ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	41161	__ret = (float32x4_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 41);
39701	41162	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39702	41163	return __ret;
39703	41164	}
39704		__ai float32x4_t __noswap_vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39705		float32x4_t __ret;
39706		__ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
39707		return __ret;
39708		}
39709	41165	#endif
39710	41166
39711	41167	#ifdef __LITTLE_ENDIAN__
39712		__ai float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39713		float32x4_t __ret;
39714		__ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41168	__ai float32x2_t vrndm_f32(float32x2_t __p0) {
	41169	float32x2_t __ret;
	41170	__ret = (float32x2_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 9);
39715	41171	return __ret;
39716	41172	}
39717	41173	#else
39718		__ai float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39719		float32x4_t __ret;
39720		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39721		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
39722		bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39723		__ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
39724		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39725		return __ret;
39726		}
39727		__ai float32x4_t __noswap_vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39728		float32x4_t __ret;
39729		__ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41174	__ai float32x2_t vrndm_f32(float32x2_t __p0) {
	41175	float32x2_t __ret;
	41176	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41177	__ret = (float32x2_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 9);
	41178	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39730	41179	return __ret;
39731	41180	}
39732	41181	#endif
39733	41182
39734	41183	#ifdef __LITTLE_ENDIAN__
39735		__ai float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	41184	__ai float32x4_t vrndnq_f32(float32x4_t __p0) {
39736	41185	float32x4_t __ret;
39737		__ret = (float32x4_t) __builtin_neon_vbfmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41186	__ret = (float32x4_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 41);
39738	41187	return __ret;
39739	41188	}
39740	41189	#else
39741		__ai float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
	41190	__ai float32x4_t vrndnq_f32(float32x4_t __p0) {
39742	41191	float32x4_t __ret;
39743	41192	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39744		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
39745		bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39746		__ret = (float32x4_t) __builtin_neon_vbfmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	41193	__ret = (float32x4_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 41);
39747	41194	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39748	41195	return __ret;
39749	41196	}
39750	41197	#endif
39751	41198
39752	41199	#ifdef __LITTLE_ENDIAN__
39753		__ai bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
39754		bfloat16x8_t __ret;
39755		__ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
	41200	__ai float32x2_t vrndn_f32(float32x2_t __p0) {
	41201	float32x2_t __ret;
	41202	__ret = (float32x2_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 9);
39756	41203	return __ret;
39757	41204	}
39758	41205	#else
39759		__ai bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
39760		bfloat16x8_t __ret;
39761		bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39762		bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39763		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
39764		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
39765		return __ret;
39766		}
39767		__ai bfloat16x8_t __noswap_vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
39768		bfloat16x8_t __ret;
39769		__ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
	41206	__ai float32x2_t vrndn_f32(float32x2_t __p0) {
	41207	float32x2_t __ret;
	41208	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41209	__ret = (float32x2_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 9);
	41210	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39770	41211	return __ret;
39771	41212	}
39772	41213	#endif
39773	41214
39774		#define vcreate_bf16(__p0) __extension__ ({ \
39775		bfloat16x4_t __ret; \
39776		uint64_t __promote = __p0; \
39777		__ret = (bfloat16x4_t)(__promote); \
39778		__ret; \
39779		})
39780		#ifdef __LITTLE_ENDIAN__
39781		__ai float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_150) {
39782		float32x4_t __ret_150;
39783		bfloat16x4_t __reint_150 = __p0_150;
39784		int32x4_t __reint1_150 = vshll_n_s16((int16x4_t ) &__reint_150, 16);
39785		__ret_150 = (float32x4_t ) &__reint1_150;
39786		return __ret_150;
39787		}
39788		#else
39789		__ai float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_151) {
39790		float32x4_t __ret_151;
39791		bfloat16x4_t __rev0_151; __rev0_151 = __builtin_shufflevector(__p0_151, __p0_151, 3, 2, 1, 0);
39792		bfloat16x4_t __reint_151 = __rev0_151;
39793		int32x4_t __reint1_151 = __noswap_vshll_n_s16((int16x4_t ) &__reint_151, 16);
39794		__ret_151 = (float32x4_t ) &__reint1_151;
39795		__ret_151 = __builtin_shufflevector(__ret_151, __ret_151, 3, 2, 1, 0);
39796		return __ret_151;
39797		}
39798		__ai float32x4_t __noswap_vcvt_f32_bf16(bfloat16x4_t __p0_152) {
39799		float32x4_t __ret_152;
39800		bfloat16x4_t __reint_152 = __p0_152;
39801		int32x4_t __reint1_152 = __noswap_vshll_n_s16((int16x4_t ) &__reint_152, 16);
39802		__ret_152 = (float32x4_t ) &__reint1_152;
39803		return __ret_152;
39804		}
39805		#endif
39806
39807		__ai float32_t vcvtah_f32_bf16(bfloat16_t __p0) {
	41215	__ai float32_t vrndns_f32(float32_t __p0) {
39808	41216	float32_t __ret;
39809		bfloat16_t __reint = __p0;
39810		int32_t __reint1 = (int32_t ) &__reint << 16;
39811		__ret = (float32_t ) &__reint1;
	41217	__ret = (float32_t) __builtin_neon_vrndns_f32(__p0);
39812	41218	return __ret;
39813	41219	}
39814		__ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
39815		bfloat16_t __ret;
39816		__ret = (bfloat16_t) __builtin_neon_vcvth_bf16_f32(__p0);
	41220	#ifdef __LITTLE_ENDIAN__
	41221	__ai float32x4_t vrndpq_f32(float32x4_t __p0) {
	41222	float32x4_t __ret;
	41223	__ret = (float32x4_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 41);
39817	41224	return __ret;
39818	41225	}
39819		#ifdef __LITTLE_ENDIAN__
39820		#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
39821		bfloat16_t __ret; \
39822		bfloat16x4_t __s0 = __p0; \
39823		__ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__s0, __p1); \
39824		__ret; \
39825		})
39826		#else
39827		#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
39828		bfloat16_t __ret; \
39829		bfloat16x4_t __s0 = __p0; \
39830		bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
39831		__ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__rev0, __p1); \
39832		__ret; \
39833		})
39834		#endif
39835
39836		#ifdef __LITTLE_ENDIAN__
39837		#define vdupq_lane_bf16(__p0_153, __p1_153) __extension__ ({ \
39838		bfloat16x8_t __ret_153; \
39839		bfloat16x4_t __s0_153 = __p0_153; \
39840		__ret_153 = splatq_lane_bf16(__s0_153, __p1_153); \
39841		__ret_153; \
39842		})
39843		#else
39844		#define vdupq_lane_bf16(__p0_154, __p1_154) __extension__ ({ \
39845		bfloat16x8_t __ret_154; \
39846		bfloat16x4_t __s0_154 = __p0_154; \
39847		bfloat16x4_t __rev0_154; __rev0_154 = __builtin_shufflevector(__s0_154, __s0_154, 3, 2, 1, 0); \
39848		__ret_154 = __noswap_splatq_lane_bf16(__rev0_154, __p1_154); \
39849		__ret_154 = __builtin_shufflevector(__ret_154, __ret_154, 7, 6, 5, 4, 3, 2, 1, 0); \
39850		__ret_154; \
39851		})
39852		#endif
39853
39854		#ifdef __LITTLE_ENDIAN__
39855		#define vdup_lane_bf16(__p0_155, __p1_155) __extension__ ({ \
39856		bfloat16x4_t __ret_155; \
39857		bfloat16x4_t __s0_155 = __p0_155; \
39858		__ret_155 = splat_lane_bf16(__s0_155, __p1_155); \
39859		__ret_155; \
39860		})
39861	41226	#else
39862		#define vdup_lane_bf16(__p0_156, __p1_156) __extension__ ({ \
39863		bfloat16x4_t __ret_156; \
39864		bfloat16x4_t __s0_156 = __p0_156; \
39865		bfloat16x4_t __rev0_156; __rev0_156 = __builtin_shufflevector(__s0_156, __s0_156, 3, 2, 1, 0); \
39866		__ret_156 = __noswap_splat_lane_bf16(__rev0_156, __p1_156); \
39867		__ret_156 = __builtin_shufflevector(__ret_156, __ret_156, 3, 2, 1, 0); \
39868		__ret_156; \
39869		})
	41227	__ai float32x4_t vrndpq_f32(float32x4_t __p0) {
	41228	float32x4_t __ret;
	41229	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41230	__ret = (float32x4_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 41);
	41231	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41232	return __ret;
	41233	}
39870	41234	#endif
39871	41235
39872	41236	#ifdef __LITTLE_ENDIAN__
39873		#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
39874		bfloat16_t __ret; \
39875		bfloat16x8_t __s0 = __p0; \
39876		__ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__s0, __p1); \
39877		__ret; \
39878		})
	41237	__ai float32x2_t vrndp_f32(float32x2_t __p0) {
	41238	float32x2_t __ret;
	41239	__ret = (float32x2_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 9);
	41240	return __ret;
	41241	}
39879	41242	#else
39880		#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
39881		bfloat16_t __ret; \
39882		bfloat16x8_t __s0 = __p0; \
39883		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39884		__ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__rev0, __p1); \
39885		__ret; \
39886		})
	41243	__ai float32x2_t vrndp_f32(float32x2_t __p0) {
	41244	float32x2_t __ret;
	41245	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41246	__ret = (float32x2_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 9);
	41247	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41248	return __ret;
	41249	}
39887	41250	#endif
39888	41251
39889	41252	#ifdef __LITTLE_ENDIAN__
39890		#define vdupq_laneq_bf16(__p0_157, __p1_157) __extension__ ({ \
39891		bfloat16x8_t __ret_157; \
39892		bfloat16x8_t __s0_157 = __p0_157; \
39893		__ret_157 = splatq_laneq_bf16(__s0_157, __p1_157); \
39894		__ret_157; \
39895		})
	41253	__ai float32x4_t vrndxq_f32(float32x4_t __p0) {
	41254	float32x4_t __ret;
	41255	__ret = (float32x4_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 41);
	41256	return __ret;
	41257	}
39896	41258	#else
39897		#define vdupq_laneq_bf16(__p0_158, __p1_158) __extension__ ({ \
39898		bfloat16x8_t __ret_158; \
39899		bfloat16x8_t __s0_158 = __p0_158; \
39900		bfloat16x8_t __rev0_158; __rev0_158 = __builtin_shufflevector(__s0_158, __s0_158, 7, 6, 5, 4, 3, 2, 1, 0); \
39901		__ret_158 = __noswap_splatq_laneq_bf16(__rev0_158, __p1_158); \
39902		__ret_158 = __builtin_shufflevector(__ret_158, __ret_158, 7, 6, 5, 4, 3, 2, 1, 0); \
39903		__ret_158; \
39904		})
	41259	__ai float32x4_t vrndxq_f32(float32x4_t __p0) {
	41260	float32x4_t __ret;
	41261	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41262	__ret = (float32x4_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 41);
	41263	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41264	return __ret;
	41265	}
39905	41266	#endif
39906	41267
39907	41268	#ifdef __LITTLE_ENDIAN__
39908		#define vdup_laneq_bf16(__p0_159, __p1_159) __extension__ ({ \
39909		bfloat16x4_t __ret_159; \
39910		bfloat16x8_t __s0_159 = __p0_159; \
39911		__ret_159 = splat_laneq_bf16(__s0_159, __p1_159); \
39912		__ret_159; \
39913		})
	41269	__ai float32x2_t vrndx_f32(float32x2_t __p0) {
	41270	float32x2_t __ret;
	41271	__ret = (float32x2_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 9);
	41272	return __ret;
	41273	}
39914	41274	#else
39915		#define vdup_laneq_bf16(__p0_160, __p1_160) __extension__ ({ \
39916		bfloat16x4_t __ret_160; \
39917		bfloat16x8_t __s0_160 = __p0_160; \
39918		bfloat16x8_t __rev0_160; __rev0_160 = __builtin_shufflevector(__s0_160, __s0_160, 7, 6, 5, 4, 3, 2, 1, 0); \
39919		__ret_160 = __noswap_splat_laneq_bf16(__rev0_160, __p1_160); \
39920		__ret_160 = __builtin_shufflevector(__ret_160, __ret_160, 3, 2, 1, 0); \
39921		__ret_160; \
39922		})
	41275	__ai float32x2_t vrndx_f32(float32x2_t __p0) {
	41276	float32x2_t __ret;
	41277	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41278	__ret = (float32x2_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 9);
	41279	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41280	return __ret;
	41281	}
39923	41282	#endif
39924	41283
39925	41284	#ifdef __LITTLE_ENDIAN__
39926		__ai bfloat16x8_t vdupq_n_bf16(bfloat16_t __p0) {
39927		bfloat16x8_t __ret;
39928		__ret = (bfloat16x8_t) {__p0, __p0, __p0, __p0, __p0, __p0, __p0, __p0};
	41285	__ai __attribute__((target("fullfp16"))) float16x8_t vrndq_f16(float16x8_t __p0) {
	41286	float16x8_t __ret;
	41287	__ret = (float16x8_t) __builtin_neon_vrndq_f16((int8x16_t)__p0, 40);
39929	41288	return __ret;
39930	41289	}
39931	41290	#else
39932		__ai bfloat16x8_t vdupq_n_bf16(bfloat16_t __p0) {
39933		bfloat16x8_t __ret;
39934		__ret = (bfloat16x8_t) {__p0, __p0, __p0, __p0, __p0, __p0, __p0, __p0};
	41291	__ai __attribute__((target("fullfp16"))) float16x8_t vrndq_f16(float16x8_t __p0) {
	41292	float16x8_t __ret;
	41293	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41294	__ret = (float16x8_t) __builtin_neon_vrndq_f16((int8x16_t)__rev0, 40);
39935	41295	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
39936	41296	return __ret;
39937	41297	}
39938	41298	#endif
39939	41299
39940	41300	#ifdef __LITTLE_ENDIAN__
39941		__ai bfloat16x4_t vdup_n_bf16(bfloat16_t __p0) {
39942		bfloat16x4_t __ret;
39943		__ret = (bfloat16x4_t) {__p0, __p0, __p0, __p0};
	41301	__ai __attribute__((target("fullfp16"))) float16x4_t vrnd_f16(float16x4_t __p0) {
	41302	float16x4_t __ret;
	41303	__ret = (float16x4_t) __builtin_neon_vrnd_f16((int8x8_t)__p0, 8);
39944	41304	return __ret;
39945	41305	}
39946	41306	#else
39947		__ai bfloat16x4_t vdup_n_bf16(bfloat16_t __p0) {
39948		bfloat16x4_t __ret;
39949		__ret = (bfloat16x4_t) {__p0, __p0, __p0, __p0};
	41307	__ai __attribute__((target("fullfp16"))) float16x4_t vrnd_f16(float16x4_t __p0) {
	41308	float16x4_t __ret;
	41309	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41310	__ret = (float16x4_t) __builtin_neon_vrnd_f16((int8x8_t)__rev0, 8);
39950	41311	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39951	41312	return __ret;
39952	41313	}
39953	41314	#endif
39954	41315
39955	41316	#ifdef __LITTLE_ENDIAN__
39956		__ai bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
39957		bfloat16x4_t __ret;
39958		__ret = __builtin_shufflevector(__p0, __p0, 4, 5, 6, 7);
	41317	__ai __attribute__((target("fullfp16"))) float16x8_t vrndaq_f16(float16x8_t __p0) {
	41318	float16x8_t __ret;
	41319	__ret = (float16x8_t) __builtin_neon_vrndaq_f16((int8x16_t)__p0, 40);
39959	41320	return __ret;
39960	41321	}
39961	41322	#else
39962		__ai bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
39963		bfloat16x4_t __ret;
39964		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
39965		__ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
39966		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39967		return __ret;
39968		}
39969		__ai bfloat16x4_t __noswap_vget_high_bf16(bfloat16x8_t __p0) {
39970		bfloat16x4_t __ret;
39971		__ret = __builtin_shufflevector(__p0, __p0, 4, 5, 6, 7);
	41323	__ai __attribute__((target("fullfp16"))) float16x8_t vrndaq_f16(float16x8_t __p0) {
	41324	float16x8_t __ret;
	41325	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41326	__ret = (float16x8_t) __builtin_neon_vrndaq_f16((int8x16_t)__rev0, 40);
	41327	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
39972	41328	return __ret;
39973	41329	}
39974	41330	#endif
39975	41331
39976	41332	#ifdef __LITTLE_ENDIAN__
39977		#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
39978		bfloat16_t __ret; \
39979		bfloat16x8_t __s0 = __p0; \
39980		__ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
39981		__ret; \
39982		})
39983		#else
39984		#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
39985		bfloat16_t __ret; \
39986		bfloat16x8_t __s0 = __p0; \
39987		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39988		__ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__rev0, __p1); \
39989		__ret; \
39990		})
39991		#define __noswap_vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
39992		bfloat16_t __ret; \
39993		bfloat16x8_t __s0 = __p0; \
39994		__ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
39995		__ret; \
39996		})
39997		#endif
39998
39999		#ifdef __LITTLE_ENDIAN__
40000		#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
40001		bfloat16_t __ret; \
40002		bfloat16x4_t __s0 = __p0; \
40003		__ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
40004		__ret; \
40005		})
40006		#else
40007		#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
40008		bfloat16_t __ret; \
40009		bfloat16x4_t __s0 = __p0; \
40010		bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
40011		__ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__rev0, __p1); \
40012		__ret; \
40013		})
40014		#define __noswap_vget_lane_bf16(__p0, __p1) __extension__ ({ \
40015		bfloat16_t __ret; \
40016		bfloat16x4_t __s0 = __p0; \
40017		__ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
40018		__ret; \
40019		})
40020		#endif
40021
40022		#ifdef __LITTLE_ENDIAN__
40023		__ai bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
40024		bfloat16x4_t __ret;
40025		__ret = __builtin_shufflevector(__p0, __p0, 0, 1, 2, 3);
	41333	__ai __attribute__((target("fullfp16"))) float16x4_t vrnda_f16(float16x4_t __p0) {
	41334	float16x4_t __ret;
	41335	__ret = (float16x4_t) __builtin_neon_vrnda_f16((int8x8_t)__p0, 8);
40026	41336	return __ret;
40027	41337	}
40028	41338	#else
40029		__ai bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
40030		bfloat16x4_t __ret;
40031		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
40032		__ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
	41339	__ai __attribute__((target("fullfp16"))) float16x4_t vrnda_f16(float16x4_t __p0) {
	41340	float16x4_t __ret;
	41341	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41342	__ret = (float16x4_t) __builtin_neon_vrnda_f16((int8x8_t)__rev0, 8);
40033	41343	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
40034	41344	return __ret;
40035	41345	}
40036		__ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
40037		bfloat16x4_t __ret;
40038		__ret = __builtin_shufflevector(__p0, __p0, 0, 1, 2, 3);
40039		return __ret;
40040		}
40041		#endif
40042
40043		#ifdef __LITTLE_ENDIAN__
40044		#define vld1q_bf16(__p0) __extension__ ({ \
40045		bfloat16x8_t __ret; \
40046		__ret = (bfloat16x8_t) __builtin_neon_vld1q_v(__p0, 43); \
40047		__ret; \
40048		})
40049		#else
40050		#define vld1q_bf16(__p0) __extension__ ({ \
40051		bfloat16x8_t __ret; \
40052		__ret = (bfloat16x8_t) __builtin_neon_vld1q_v(__p0, 43); \
40053		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
40054		__ret; \
40055		})
40056	41346	#endif
40057	41347
40058	41348	#ifdef __LITTLE_ENDIAN__
40059		#define vld1_bf16(__p0) __extension__ ({ \
40060		bfloat16x4_t __ret; \
40061		__ret = (bfloat16x4_t) __builtin_neon_vld1_v(__p0, 11); \
40062		__ret; \
40063		})
40064		#else
40065		#define vld1_bf16(__p0) __extension__ ({ \
40066		bfloat16x4_t __ret; \
40067		__ret = (bfloat16x4_t) __builtin_neon_vld1_v(__p0, 11); \
40068		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
40069		__ret; \
40070		})
40071		#endif
40072
40073		#ifdef __LITTLE_ENDIAN__
40074		#define vld1q_dup_bf16(__p0) __extension__ ({ \
40075		bfloat16x8_t __ret; \
40076		__ret = (bfloat16x8_t) __builtin_neon_vld1q_dup_v(__p0, 43); \
40077		__ret; \
40078		})
40079		#else
40080		#define vld1q_dup_bf16(__p0) __extension__ ({ \
40081		bfloat16x8_t __ret; \
40082		__ret = (bfloat16x8_t) __builtin_neon_vld1q_dup_v(__p0, 43); \
40083		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
40084		__ret; \
40085		})
40086		#endif
40087
40088		#ifdef __LITTLE_ENDIAN__
40089		#define vld1_dup_bf16(__p0) __extension__ ({ \
40090		bfloat16x4_t __ret; \
40091		__ret = (bfloat16x4_t) __builtin_neon_vld1_dup_v(__p0, 11); \
40092		__ret; \
40093		})
40094		#else
40095		#define vld1_dup_bf16(__p0) __extension__ ({ \
40096		bfloat16x4_t __ret; \
40097		__ret = (bfloat16x4_t) __builtin_neon_vld1_dup_v(__p0, 11); \
40098		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
40099		__ret; \
40100		})
40101		#endif
40102
40103		#ifdef __LITTLE_ENDIAN__
40104		#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40105		bfloat16x8_t __ret; \
40106		bfloat16x8_t __s1 = __p1; \
40107		__ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 43); \
40108		__ret; \
40109		})
40110		#else
40111		#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40112		bfloat16x8_t __ret; \
40113		bfloat16x8_t __s1 = __p1; \
40114		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
40115		__ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 43); \
40116		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
40117		__ret; \
40118		})
40119		#endif
40120
40121		#ifdef __LITTLE_ENDIAN__
40122		#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40123		bfloat16x4_t __ret; \
40124		bfloat16x4_t __s1 = __p1; \
40125		__ret = (bfloat16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 11); \
40126		__ret; \
40127		})
	41349	__ai __attribute__((target("fullfp16"))) float16x8_t vrndmq_f16(float16x8_t __p0) {
	41350	float16x8_t __ret;
	41351	__ret = (float16x8_t) __builtin_neon_vrndmq_f16((int8x16_t)__p0, 40);
	41352	return __ret;
	41353	}
40128	41354	#else
40129		#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40130		bfloat16x4_t __ret; \
40131		bfloat16x4_t __s1 = __p1; \
40132		bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
40133		__ret = (bfloat16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 11); \
40134		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
40135		__ret; \
40136		})
	41355	__ai __attribute__((target("fullfp16"))) float16x8_t vrndmq_f16(float16x8_t __p0) {
	41356	float16x8_t __ret;
	41357	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41358	__ret = (float16x8_t) __builtin_neon_vrndmq_f16((int8x16_t)__rev0, 40);
	41359	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41360	return __ret;
	41361	}
40137	41362	#endif
40138	41363
40139	41364	#ifdef __LITTLE_ENDIAN__
40140		#define vld1q_bf16_x2(__p0) __extension__ ({ \
40141		bfloat16x8x2_t __ret; \
40142		__builtin_neon_vld1q_x2_v(&__ret, __p0, 43); \
40143		__ret; \
40144		})
	41365	__ai __attribute__((target("fullfp16"))) float16x4_t vrndm_f16(float16x4_t __p0) {
	41366	float16x4_t __ret;
	41367	__ret = (float16x4_t) __builtin_neon_vrndm_f16((int8x8_t)__p0, 8);
	41368	return __ret;
	41369	}
40145	41370	#else
40146		#define vld1q_bf16_x2(__p0) __extension__ ({ \
40147		bfloat16x8x2_t __ret; \
40148		__builtin_neon_vld1q_x2_v(&__ret, __p0, 43); \
40149		\
40150		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40151		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40152		__ret; \
40153		})
	41371	__ai __attribute__((target("fullfp16"))) float16x4_t vrndm_f16(float16x4_t __p0) {
	41372	float16x4_t __ret;
	41373	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41374	__ret = (float16x4_t) __builtin_neon_vrndm_f16((int8x8_t)__rev0, 8);
	41375	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41376	return __ret;
	41377	}
40154	41378	#endif
40155	41379
40156	41380	#ifdef __LITTLE_ENDIAN__
40157		#define vld1_bf16_x2(__p0) __extension__ ({ \
40158		bfloat16x4x2_t __ret; \
40159		__builtin_neon_vld1_x2_v(&__ret, __p0, 11); \
40160		__ret; \
40161		})
	41381	__ai __attribute__((target("fullfp16"))) float16x8_t vrndnq_f16(float16x8_t __p0) {
	41382	float16x8_t __ret;
	41383	__ret = (float16x8_t) __builtin_neon_vrndnq_f16((int8x16_t)__p0, 40);
	41384	return __ret;
	41385	}
40162	41386	#else
40163		#define vld1_bf16_x2(__p0) __extension__ ({ \
40164		bfloat16x4x2_t __ret; \
40165		__builtin_neon_vld1_x2_v(&__ret, __p0, 11); \
40166		\
40167		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40168		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40169		__ret; \
40170		})
	41387	__ai __attribute__((target("fullfp16"))) float16x8_t vrndnq_f16(float16x8_t __p0) {
	41388	float16x8_t __ret;
	41389	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41390	__ret = (float16x8_t) __builtin_neon_vrndnq_f16((int8x16_t)__rev0, 40);
	41391	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41392	return __ret;
	41393	}
40171	41394	#endif
40172	41395
40173	41396	#ifdef __LITTLE_ENDIAN__
40174		#define vld1q_bf16_x3(__p0) __extension__ ({ \
40175		bfloat16x8x3_t __ret; \
40176		__builtin_neon_vld1q_x3_v(&__ret, __p0, 43); \
40177		__ret; \
40178		})
	41397	__ai __attribute__((target("fullfp16"))) float16x4_t vrndn_f16(float16x4_t __p0) {
	41398	float16x4_t __ret;
	41399	__ret = (float16x4_t) __builtin_neon_vrndn_f16((int8x8_t)__p0, 8);
	41400	return __ret;
	41401	}
40179	41402	#else
40180		#define vld1q_bf16_x3(__p0) __extension__ ({ \
40181		bfloat16x8x3_t __ret; \
40182		__builtin_neon_vld1q_x3_v(&__ret, __p0, 43); \
40183		\
40184		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40185		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40186		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40187		__ret; \
40188		})
	41403	__ai __attribute__((target("fullfp16"))) float16x4_t vrndn_f16(float16x4_t __p0) {
	41404	float16x4_t __ret;
	41405	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41406	__ret = (float16x4_t) __builtin_neon_vrndn_f16((int8x8_t)__rev0, 8);
	41407	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41408	return __ret;
	41409	}
40189	41410	#endif
40190	41411
40191	41412	#ifdef __LITTLE_ENDIAN__
40192		#define vld1_bf16_x3(__p0) __extension__ ({ \
40193		bfloat16x4x3_t __ret; \
40194		__builtin_neon_vld1_x3_v(&__ret, __p0, 11); \
40195		__ret; \
40196		})
	41413	__ai __attribute__((target("fullfp16"))) float16x8_t vrndpq_f16(float16x8_t __p0) {
	41414	float16x8_t __ret;
	41415	__ret = (float16x8_t) __builtin_neon_vrndpq_f16((int8x16_t)__p0, 40);
	41416	return __ret;
	41417	}
40197	41418	#else
40198		#define vld1_bf16_x3(__p0) __extension__ ({ \
40199		bfloat16x4x3_t __ret; \
40200		__builtin_neon_vld1_x3_v(&__ret, __p0, 11); \
40201		\
40202		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40203		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40204		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40205		__ret; \
40206		})
	41419	__ai __attribute__((target("fullfp16"))) float16x8_t vrndpq_f16(float16x8_t __p0) {
	41420	float16x8_t __ret;
	41421	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41422	__ret = (float16x8_t) __builtin_neon_vrndpq_f16((int8x16_t)__rev0, 40);
	41423	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41424	return __ret;
	41425	}
40207	41426	#endif
40208	41427
40209	41428	#ifdef __LITTLE_ENDIAN__
40210		#define vld1q_bf16_x4(__p0) __extension__ ({ \
40211		bfloat16x8x4_t __ret; \
40212		__builtin_neon_vld1q_x4_v(&__ret, __p0, 43); \
40213		__ret; \
40214		})
	41429	__ai __attribute__((target("fullfp16"))) float16x4_t vrndp_f16(float16x4_t __p0) {
	41430	float16x4_t __ret;
	41431	__ret = (float16x4_t) __builtin_neon_vrndp_f16((int8x8_t)__p0, 8);
	41432	return __ret;
	41433	}
40215	41434	#else
40216		#define vld1q_bf16_x4(__p0) __extension__ ({ \
40217		bfloat16x8x4_t __ret; \
40218		__builtin_neon_vld1q_x4_v(&__ret, __p0, 43); \
40219		\
40220		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40221		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40222		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40223		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40224		__ret; \
40225		})
	41435	__ai __attribute__((target("fullfp16"))) float16x4_t vrndp_f16(float16x4_t __p0) {
	41436	float16x4_t __ret;
	41437	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41438	__ret = (float16x4_t) __builtin_neon_vrndp_f16((int8x8_t)__rev0, 8);
	41439	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41440	return __ret;
	41441	}
40226	41442	#endif
40227	41443
40228	41444	#ifdef __LITTLE_ENDIAN__
40229		#define vld1_bf16_x4(__p0) __extension__ ({ \
40230		bfloat16x4x4_t __ret; \
40231		__builtin_neon_vld1_x4_v(&__ret, __p0, 11); \
40232		__ret; \
40233		})
	41445	__ai __attribute__((target("fullfp16"))) float16x8_t vrndxq_f16(float16x8_t __p0) {
	41446	float16x8_t __ret;
	41447	__ret = (float16x8_t) __builtin_neon_vrndxq_f16((int8x16_t)__p0, 40);
	41448	return __ret;
	41449	}
40234	41450	#else
40235		#define vld1_bf16_x4(__p0) __extension__ ({ \
40236		bfloat16x4x4_t __ret; \
40237		__builtin_neon_vld1_x4_v(&__ret, __p0, 11); \
40238		\
40239		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40240		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40241		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40242		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
40243		__ret; \
40244		})
	41451	__ai __attribute__((target("fullfp16"))) float16x8_t vrndxq_f16(float16x8_t __p0) {
	41452	float16x8_t __ret;
	41453	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41454	__ret = (float16x8_t) __builtin_neon_vrndxq_f16((int8x16_t)__rev0, 40);
	41455	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41456	return __ret;
	41457	}
40245	41458	#endif
40246	41459
40247	41460	#ifdef __LITTLE_ENDIAN__
40248		#define vld2q_bf16(__p0) __extension__ ({ \
40249		bfloat16x8x2_t __ret; \
40250		__builtin_neon_vld2q_v(&__ret, __p0, 43); \
40251		__ret; \
40252		})
	41461	__ai __attribute__((target("fullfp16"))) float16x4_t vrndx_f16(float16x4_t __p0) {
	41462	float16x4_t __ret;
	41463	__ret = (float16x4_t) __builtin_neon_vrndx_f16((int8x8_t)__p0, 8);
	41464	return __ret;
	41465	}
40253	41466	#else
40254		#define vld2q_bf16(__p0) __extension__ ({ \
40255		bfloat16x8x2_t __ret; \
40256		__builtin_neon_vld2q_v(&__ret, __p0, 43); \
40257		\
40258		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40259		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40260		__ret; \
40261		})
	41467	__ai __attribute__((target("fullfp16"))) float16x4_t vrndx_f16(float16x4_t __p0) {
	41468	float16x4_t __ret;
	41469	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41470	__ret = (float16x4_t) __builtin_neon_vrndx_f16((int8x8_t)__rev0, 8);
	41471	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41472	return __ret;
	41473	}
40262	41474	#endif
40263	41475
	41476	#endif
	41477	#if __ARM_ARCH >= 8 && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
40264	41478	#ifdef __LITTLE_ENDIAN__
40265		#define vld2_bf16(__p0) __extension__ ({ \
40266		bfloat16x4x2_t __ret; \
40267		__builtin_neon_vld2_v(&__ret, __p0, 11); \
40268		__ret; \
40269		})
	41479	__ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	41480	float32x4_t __ret;
	41481	__ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	41482	return __ret;
	41483	}
40270	41484	#else
40271		#define vld2_bf16(__p0) __extension__ ({ \
40272		bfloat16x4x2_t __ret; \
40273		__builtin_neon_vld2_v(&__ret, __p0, 11); \
40274		\
40275		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40276		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40277		__ret; \
40278		})
	41485	__ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	41486	float32x4_t __ret;
	41487	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41488	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41489	__ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	41490	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41491	return __ret;
	41492	}
40279	41493	#endif
40280	41494
40281	41495	#ifdef __LITTLE_ENDIAN__
40282		#define vld2q_dup_bf16(__p0) __extension__ ({ \
40283		bfloat16x8x2_t __ret; \
40284		__builtin_neon_vld2q_dup_v(&__ret, __p0, 43); \
40285		__ret; \
40286		})
	41496	__ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
	41497	float32x2_t __ret;
	41498	__ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	41499	return __ret;
	41500	}
40287	41501	#else
40288		#define vld2q_dup_bf16(__p0) __extension__ ({ \
40289		bfloat16x8x2_t __ret; \
40290		__builtin_neon_vld2q_dup_v(&__ret, __p0, 43); \
40291		\
40292		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40293		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40294		__ret; \
40295		})
	41502	__ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
	41503	float32x2_t __ret;
	41504	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41505	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41506	__ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
	41507	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41508	return __ret;
	41509	}
40296	41510	#endif
40297	41511
40298	41512	#ifdef __LITTLE_ENDIAN__
40299		#define vld2_dup_bf16(__p0) __extension__ ({ \
40300		bfloat16x4x2_t __ret; \
40301		__builtin_neon_vld2_dup_v(&__ret, __p0, 11); \
40302		__ret; \
40303		})
	41513	__ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	41514	float32x4_t __ret;
	41515	__ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	41516	return __ret;
	41517	}
40304	41518	#else
40305		#define vld2_dup_bf16(__p0) __extension__ ({ \
40306		bfloat16x4x2_t __ret; \
40307		__builtin_neon_vld2_dup_v(&__ret, __p0, 11); \
40308		\
40309		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40310		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40311		__ret; \
40312		})
	41519	__ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	41520	float32x4_t __ret;
	41521	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41522	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41523	__ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	41524	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41525	return __ret;
	41526	}
40313	41527	#endif
40314	41528
40315	41529	#ifdef __LITTLE_ENDIAN__
40316		#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40317		bfloat16x8x2_t __ret; \
40318		bfloat16x8x2_t __s1 = __p1; \
40319		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
40320		__ret; \
40321		})
	41530	__ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
	41531	float32x2_t __ret;
	41532	__ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	41533	return __ret;
	41534	}
40322	41535	#else
40323		#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40324		bfloat16x8x2_t __ret; \
40325		bfloat16x8x2_t __s1 = __p1; \
40326		bfloat16x8x2_t __rev1; \
40327		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40328		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40329		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
40330		\
40331		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40332		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40333		__ret; \
40334		})
	41536	__ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
	41537	float32x2_t __ret;
	41538	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41539	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41540	__ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
	41541	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41542	return __ret;
	41543	}
40335	41544	#endif
40336	41545
40337	41546	#ifdef __LITTLE_ENDIAN__
40338		#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40339		bfloat16x4x2_t __ret; \
40340		bfloat16x4x2_t __s1 = __p1; \
40341		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
40342		__ret; \
40343		})
	41547	__ai __attribute__((target("fullfp16"))) float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	41548	float16x8_t __ret;
	41549	__ret = (float16x8_t) __builtin_neon_vmaxnmq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
	41550	return __ret;
	41551	}
40344	41552	#else
40345		#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40346		bfloat16x4x2_t __ret; \
40347		bfloat16x4x2_t __s1 = __p1; \
40348		bfloat16x4x2_t __rev1; \
40349		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40350		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40351		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
40352		\
40353		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40354		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40355		__ret; \
40356		})
	41553	__ai __attribute__((target("fullfp16"))) float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	41554	float16x8_t __ret;
	41555	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41556	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	41557	__ret = (float16x8_t) __builtin_neon_vmaxnmq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	41558	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41559	return __ret;
	41560	}
40357	41561	#endif
40358	41562
40359	41563	#ifdef __LITTLE_ENDIAN__
40360		#define vld3q_bf16(__p0) __extension__ ({ \
40361		bfloat16x8x3_t __ret; \
40362		__builtin_neon_vld3q_v(&__ret, __p0, 43); \
40363		__ret; \
40364		})
	41564	__ai __attribute__((target("fullfp16"))) float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
	41565	float16x4_t __ret;
	41566	__ret = (float16x4_t) __builtin_neon_vmaxnm_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	41567	return __ret;
	41568	}
40365	41569	#else
40366		#define vld3q_bf16(__p0) __extension__ ({ \
40367		bfloat16x8x3_t __ret; \
40368		__builtin_neon_vld3q_v(&__ret, __p0, 43); \
40369		\
40370		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40371		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40372		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40373		__ret; \
40374		})
	41570	__ai __attribute__((target("fullfp16"))) float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
	41571	float16x4_t __ret;
	41572	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41573	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41574	__ret = (float16x4_t) __builtin_neon_vmaxnm_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	41575	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41576	return __ret;
	41577	}
40375	41578	#endif
40376	41579
40377	41580	#ifdef __LITTLE_ENDIAN__
40378		#define vld3_bf16(__p0) __extension__ ({ \
40379		bfloat16x4x3_t __ret; \
40380		__builtin_neon_vld3_v(&__ret, __p0, 11); \
40381		__ret; \
40382		})
	41581	__ai __attribute__((target("fullfp16"))) float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	41582	float16x8_t __ret;
	41583	__ret = (float16x8_t) __builtin_neon_vminnmq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
	41584	return __ret;
	41585	}
40383	41586	#else
40384		#define vld3_bf16(__p0) __extension__ ({ \
40385		bfloat16x4x3_t __ret; \
40386		__builtin_neon_vld3_v(&__ret, __p0, 11); \
40387		\
40388		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40389		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40390		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40391		__ret; \
40392		})
	41587	__ai __attribute__((target("fullfp16"))) float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	41588	float16x8_t __ret;
	41589	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41590	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	41591	__ret = (float16x8_t) __builtin_neon_vminnmq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	41592	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41593	return __ret;
	41594	}
40393	41595	#endif
40394	41596
40395	41597	#ifdef __LITTLE_ENDIAN__
40396		#define vld3q_dup_bf16(__p0) __extension__ ({ \
40397		bfloat16x8x3_t __ret; \
40398		__builtin_neon_vld3q_dup_v(&__ret, __p0, 43); \
40399		__ret; \
40400		})
	41598	__ai __attribute__((target("fullfp16"))) float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
	41599	float16x4_t __ret;
	41600	__ret = (float16x4_t) __builtin_neon_vminnm_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
	41601	return __ret;
	41602	}
40401	41603	#else
40402		#define vld3q_dup_bf16(__p0) __extension__ ({ \
40403		bfloat16x8x3_t __ret; \
40404		__builtin_neon_vld3q_dup_v(&__ret, __p0, 43); \
40405		\
40406		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40407		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40408		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40409		__ret; \
40410		})
	41604	__ai __attribute__((target("fullfp16"))) float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
	41605	float16x4_t __ret;
	41606	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41607	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41608	__ret = (float16x4_t) __builtin_neon_vminnm_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	41609	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41610	return __ret;
	41611	}
40411	41612	#endif
40412	41613
	41614	#endif
	41615	#if defined(__ARM_FEATURE_FMA)
40413	41616	#ifdef __LITTLE_ENDIAN__
40414		#define vld3_dup_bf16(__p0) __extension__ ({ \
40415		bfloat16x4x3_t __ret; \
40416		__builtin_neon_vld3_dup_v(&__ret, __p0, 11); \
40417		__ret; \
40418		})
	41617	__ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	41618	float32x4_t __ret;
	41619	__ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41620	return __ret;
	41621	}
40419	41622	#else
40420		#define vld3_dup_bf16(__p0) __extension__ ({ \
40421		bfloat16x4x3_t __ret; \
40422		__builtin_neon_vld3_dup_v(&__ret, __p0, 11); \
40423		\
40424		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40425		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40426		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40427		__ret; \
40428		})
	41623	__ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	41624	float32x4_t __ret;
	41625	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41626	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41627	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	41628	__ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	41629	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41630	return __ret;
	41631	}
	41632	__ai float32x4_t __noswap_vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	41633	float32x4_t __ret;
	41634	__ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	41635	return __ret;
	41636	}
40429	41637	#endif
40430	41638
40431	41639	#ifdef __LITTLE_ENDIAN__
40432		#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40433		bfloat16x8x3_t __ret; \
40434		bfloat16x8x3_t __s1 = __p1; \
40435		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
40436		__ret; \
40437		})
	41640	__ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	41641	float32x2_t __ret;
	41642	__ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	41643	return __ret;
	41644	}
40438	41645	#else
40439		#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40440		bfloat16x8x3_t __ret; \
40441		bfloat16x8x3_t __s1 = __p1; \
40442		bfloat16x8x3_t __rev1; \
40443		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40444		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40445		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40446		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
40447		\
40448		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40449		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40450		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40451		__ret; \
40452		})
	41646	__ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	41647	float32x2_t __ret;
	41648	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41649	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41650	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	41651	__ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	41652	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41653	return __ret;
	41654	}
	41655	__ai float32x2_t __noswap_vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	41656	float32x2_t __ret;
	41657	__ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	41658	return __ret;
	41659	}
40453	41660	#endif
40454	41661
40455	41662	#ifdef __LITTLE_ENDIAN__
40456		#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40457		bfloat16x4x3_t __ret; \
40458		bfloat16x4x3_t __s1 = __p1; \
40459		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
40460		__ret; \
40461		})
	41663	__ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
	41664	float32x4_t __ret;
	41665	__ret = vfmaq_f32(__p0, __p1, (float32x4_t) {__p2, __p2, __p2, __p2});
	41666	return __ret;
	41667	}
40462	41668	#else
40463		#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40464		bfloat16x4x3_t __ret; \
40465		bfloat16x4x3_t __s1 = __p1; \
40466		bfloat16x4x3_t __rev1; \
40467		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40468		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40469		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40470		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
40471		\
40472		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40473		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40474		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40475		__ret; \
40476		})
	41669	__ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
	41670	float32x4_t __ret;
	41671	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41672	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41673	__ret = __noswap_vfmaq_f32(__rev0, __rev1, (float32x4_t) {__p2, __p2, __p2, __p2});
	41674	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41675	return __ret;
	41676	}
40477	41677	#endif
40478	41678
40479	41679	#ifdef __LITTLE_ENDIAN__
40480		#define vld4q_bf16(__p0) __extension__ ({ \
40481		bfloat16x8x4_t __ret; \
40482		__builtin_neon_vld4q_v(&__ret, __p0, 43); \
40483		__ret; \
40484		})
	41680	__ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
	41681	float32x2_t __ret;
	41682	__ret = vfma_f32(__p0, __p1, (float32x2_t) {__p2, __p2});
	41683	return __ret;
	41684	}
40485	41685	#else
40486		#define vld4q_bf16(__p0) __extension__ ({ \
40487		bfloat16x8x4_t __ret; \
40488		__builtin_neon_vld4q_v(&__ret, __p0, 43); \
40489		\
40490		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40491		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40492		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40493		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40494		__ret; \
40495		})
	41686	__ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
	41687	float32x2_t __ret;
	41688	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41689	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41690	__ret = __noswap_vfma_f32(__rev0, __rev1, (float32x2_t) {__p2, __p2});
	41691	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41692	return __ret;
	41693	}
40496	41694	#endif
40497	41695
40498	41696	#ifdef __LITTLE_ENDIAN__
40499		#define vld4_bf16(__p0) __extension__ ({ \
40500		bfloat16x4x4_t __ret; \
40501		__builtin_neon_vld4_v(&__ret, __p0, 11); \
40502		__ret; \
40503		})
	41697	__ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	41698	float32x4_t __ret;
	41699	__ret = vfmaq_f32(__p0, -__p1, __p2);
	41700	return __ret;
	41701	}
40504	41702	#else
40505		#define vld4_bf16(__p0) __extension__ ({ \
40506		bfloat16x4x4_t __ret; \
40507		__builtin_neon_vld4_v(&__ret, __p0, 11); \
40508		\
40509		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40510		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40511		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40512		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
40513		__ret; \
40514		})
	41703	__ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
	41704	float32x4_t __ret;
	41705	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41706	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41707	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	41708	__ret = __noswap_vfmaq_f32(__rev0, -__rev1, __rev2);
	41709	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41710	return __ret;
	41711	}
40515	41712	#endif
40516	41713
40517	41714	#ifdef __LITTLE_ENDIAN__
40518		#define vld4q_dup_bf16(__p0) __extension__ ({ \
40519		bfloat16x8x4_t __ret; \
40520		__builtin_neon_vld4q_dup_v(&__ret, __p0, 43); \
40521		__ret; \
40522		})
	41715	__ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	41716	float32x2_t __ret;
	41717	__ret = vfma_f32(__p0, -__p1, __p2);
	41718	return __ret;
	41719	}
40523	41720	#else
40524		#define vld4q_dup_bf16(__p0) __extension__ ({ \
40525		bfloat16x8x4_t __ret; \
40526		__builtin_neon_vld4q_dup_v(&__ret, __p0, 43); \
40527		\
40528		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40529		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40530		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40531		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40532		__ret; \
40533		})
	41721	__ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
	41722	float32x2_t __ret;
	41723	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41724	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41725	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	41726	__ret = __noswap_vfma_f32(__rev0, -__rev1, __rev2);
	41727	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41728	return __ret;
	41729	}
40534	41730	#endif
40535	41731
	41732	#endif
	41733	#if defined(__aarch64__)
40536	41734	#ifdef __LITTLE_ENDIAN__
40537		#define vld4_dup_bf16(__p0) __extension__ ({ \
40538		bfloat16x4x4_t __ret; \
40539		__builtin_neon_vld4_dup_v(&__ret, __p0, 11); \
40540		__ret; \
40541		})
	41735	__ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
	41736	float64x2_t __ret;
	41737	__ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	41738	return __ret;
	41739	}
40542	41740	#else
40543		#define vld4_dup_bf16(__p0) __extension__ ({ \
40544		bfloat16x4x4_t __ret; \
40545		__builtin_neon_vld4_dup_v(&__ret, __p0, 11); \
40546		\
40547		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40548		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40549		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40550		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
40551		__ret; \
40552		})
	41741	__ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
	41742	float64x2_t __ret;
	41743	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41744	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41745	__ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	41746	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41747	return __ret;
	41748	}
40553	41749	#endif
40554	41750
	41751	__ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
	41752	float64x1_t __ret;
	41753	__ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	41754	return __ret;
	41755	}
	41756	__ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
	41757	float64_t __ret;
	41758	__ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
	41759	return __ret;
	41760	}
	41761	__ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
	41762	float32_t __ret;
	41763	__ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
	41764	return __ret;
	41765	}
40555	41766	#ifdef __LITTLE_ENDIAN__
40556		#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40557		bfloat16x8x4_t __ret; \
40558		bfloat16x8x4_t __s1 = __p1; \
40559		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 43); \
40560		__ret; \
40561		})
	41767	__ai float64x2_t vabsq_f64(float64x2_t __p0) {
	41768	float64x2_t __ret;
	41769	__ret = (float64x2_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 42);
	41770	return __ret;
	41771	}
40562	41772	#else
40563		#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40564		bfloat16x8x4_t __ret; \
40565		bfloat16x8x4_t __s1 = __p1; \
40566		bfloat16x8x4_t __rev1; \
40567		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40568		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40569		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40570		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40571		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 43); \
40572		\
40573		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40574		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40575		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40576		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40577		__ret; \
40578		})
	41773	__ai float64x2_t vabsq_f64(float64x2_t __p0) {
	41774	float64x2_t __ret;
	41775	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41776	__ret = (float64x2_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 42);
	41777	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41778	return __ret;
	41779	}
40579	41780	#endif
40580	41781
40581	41782	#ifdef __LITTLE_ENDIAN__
40582		#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40583		bfloat16x4x4_t __ret; \
40584		bfloat16x4x4_t __s1 = __p1; \
40585		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 11); \
40586		__ret; \
40587		})
	41783	__ai int64x2_t vabsq_s64(int64x2_t __p0) {
	41784	int64x2_t __ret;
	41785	__ret = (int64x2_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 35);
	41786	return __ret;
	41787	}
40588	41788	#else
40589		#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40590		bfloat16x4x4_t __ret; \
40591		bfloat16x4x4_t __s1 = __p1; \
40592		bfloat16x4x4_t __rev1; \
40593		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40594		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40595		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40596		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
40597		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 11); \
40598		\
40599		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
40600		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0); \
40601		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 3, 2, 1, 0); \
40602		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 3, 2, 1, 0); \
40603		__ret; \
40604		})
	41789	__ai int64x2_t vabsq_s64(int64x2_t __p0) {
	41790	int64x2_t __ret;
	41791	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41792	__ret = (int64x2_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 35);
	41793	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41794	return __ret;
	41795	}
40605	41796	#endif
40606	41797
	41798	__ai float64x1_t vabs_f64(float64x1_t __p0) {
	41799	float64x1_t __ret;
	41800	__ret = (float64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 10);
	41801	return __ret;
	41802	}
	41803	__ai int64x1_t vabs_s64(int64x1_t __p0) {
	41804	int64x1_t __ret;
	41805	__ret = (int64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 3);
	41806	return __ret;
	41807	}
	41808	__ai int64_t vabsd_s64(int64_t __p0) {
	41809	int64_t __ret;
	41810	__ret = (int64_t) __builtin_neon_vabsd_s64(__p0);
	41811	return __ret;
	41812	}
40607	41813	#ifdef __LITTLE_ENDIAN__
40608		#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40609		bfloat16x8_t __ret; \
40610		bfloat16_t __s0 = __p0; \
40611		bfloat16x8_t __s1 = __p1; \
40612		__ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
40613		__ret; \
40614		})
	41814	__ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
	41815	float64x2_t __ret;
	41816	__ret = __p0 + __p1;
	41817	return __ret;
	41818	}
40615	41819	#else
40616		#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40617		bfloat16x8_t __ret; \
40618		bfloat16_t __s0 = __p0; \
40619		bfloat16x8_t __s1 = __p1; \
40620		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
40621		__ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__rev1, __p2); \
40622		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
40623		__ret; \
40624		})
40625		#define __noswap_vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40626		bfloat16x8_t __ret; \
40627		bfloat16_t __s0 = __p0; \
40628		bfloat16x8_t __s1 = __p1; \
40629		__ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
40630		__ret; \
40631		})
	41820	__ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
	41821	float64x2_t __ret;
	41822	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41823	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41824	__ret = __rev0 + __rev1;
	41825	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	41826	return __ret;
	41827	}
40632	41828	#endif
40633	41829
	41830	__ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
	41831	float64x1_t __ret;
	41832	__ret = __p0 + __p1;
	41833	return __ret;
	41834	}
	41835	__ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
	41836	uint64_t __ret;
	41837	__ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
	41838	return __ret;
	41839	}
	41840	__ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
	41841	int64_t __ret;
	41842	__ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
	41843	return __ret;
	41844	}
	41845	__ai poly128_t vaddq_p128(poly128_t __p0, poly128_t __p1) {
	41846	poly128_t __ret;
	41847	__ret = (poly128_t) __builtin_neon_vaddq_p128(__p0, __p1);
	41848	return __ret;
	41849	}
40634	41850	#ifdef __LITTLE_ENDIAN__
40635		#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40636		bfloat16x4_t __ret; \
40637		bfloat16_t __s0 = __p0; \
40638		bfloat16x4_t __s1 = __p1; \
40639		__ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
40640		__ret; \
40641		})
	41851	__ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	41852	uint16x8_t __ret;
	41853	__ret = vcombine_u16(__p0, vaddhn_u32(__p1, __p2));
	41854	return __ret;
	41855	}
40642	41856	#else
40643		#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40644		bfloat16x4_t __ret; \
40645		bfloat16_t __s0 = __p0; \
40646		bfloat16x4_t __s1 = __p1; \
40647		bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
40648		__ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__rev1, __p2); \
40649		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
40650		__ret; \
40651		})
40652		#define __noswap_vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40653		bfloat16x4_t __ret; \
40654		bfloat16_t __s0 = __p0; \
40655		bfloat16x4_t __s1 = __p1; \
40656		__ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
40657		__ret; \
40658		})
	41857	__ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	41858	uint16x8_t __ret;
	41859	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41860	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41861	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	41862	__ret = __noswap_vcombine_u16(__rev0, __noswap_vaddhn_u32(__rev1, __rev2));
	41863	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41864	return __ret;
	41865	}
40659	41866	#endif
40660	41867
40661	41868	#ifdef __LITTLE_ENDIAN__
40662		#define vst1q_bf16(__p0, __p1) __extension__ ({ \
40663		bfloat16x8_t __s1 = __p1; \
40664		__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 43); \
40665		})
	41869	__ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	41870	uint32x4_t __ret;
	41871	__ret = vcombine_u32(__p0, vaddhn_u64(__p1, __p2));
	41872	return __ret;
	41873	}
40666	41874	#else
40667		#define vst1q_bf16(__p0, __p1) __extension__ ({ \
40668		bfloat16x8_t __s1 = __p1; \
40669		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
40670		__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 43); \
40671		})
	41875	__ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	41876	uint32x4_t __ret;
	41877	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41878	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41879	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	41880	__ret = __noswap_vcombine_u32(__rev0, __noswap_vaddhn_u64(__rev1, __rev2));
	41881	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41882	return __ret;
	41883	}
40672	41884	#endif
40673	41885
40674	41886	#ifdef __LITTLE_ENDIAN__
40675		#define vst1_bf16(__p0, __p1) __extension__ ({ \
40676		bfloat16x4_t __s1 = __p1; \
40677		__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 11); \
40678		})
	41887	__ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	41888	uint8x16_t __ret;
	41889	__ret = vcombine_u8(__p0, vaddhn_u16(__p1, __p2));
	41890	return __ret;
	41891	}
40679	41892	#else
40680		#define vst1_bf16(__p0, __p1) __extension__ ({ \
40681		bfloat16x4_t __s1 = __p1; \
40682		bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
40683		__builtin_neon_vst1_v(__p0, (int8x8_t)__rev1, 11); \
40684		})
	41893	__ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	41894	uint8x16_t __ret;
	41895	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41896	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	41897	uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	41898	__ret = __noswap_vcombine_u8(__rev0, __noswap_vaddhn_u16(__rev1, __rev2));
	41899	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	41900	return __ret;
	41901	}
40685	41902	#endif
40686	41903
40687	41904	#ifdef __LITTLE_ENDIAN__
40688		#define vst1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40689		bfloat16x8_t __s1 = __p1; \
40690		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 43); \
40691		})
	41905	__ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	41906	int16x8_t __ret;
	41907	__ret = vcombine_s16(__p0, vaddhn_s32(__p1, __p2));
	41908	return __ret;
	41909	}
40692	41910	#else
40693		#define vst1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40694		bfloat16x8_t __s1 = __p1; \
40695		bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
40696		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 43); \
40697		})
	41911	__ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	41912	int16x8_t __ret;
	41913	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41914	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	41915	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	41916	__ret = __noswap_vcombine_s16(__rev0, __noswap_vaddhn_s32(__rev1, __rev2));
	41917	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	41918	return __ret;
	41919	}
40698	41920	#endif
40699	41921
40700	41922	#ifdef __LITTLE_ENDIAN__
40701		#define vst1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40702		bfloat16x4_t __s1 = __p1; \
40703		__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 11); \
40704		})
	41923	__ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	41924	int32x4_t __ret;
	41925	__ret = vcombine_s32(__p0, vaddhn_s64(__p1, __p2));
	41926	return __ret;
	41927	}
40705	41928	#else
40706		#define vst1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40707		bfloat16x4_t __s1 = __p1; \
40708		bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
40709		__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__rev1, __p2, 11); \
40710		})
	41929	__ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	41930	int32x4_t __ret;
	41931	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	41932	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	41933	int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	41934	__ret = __noswap_vcombine_s32(__rev0, __noswap_vaddhn_s64(__rev1, __rev2));
	41935	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	41936	return __ret;
	41937	}
40711	41938	#endif
40712	41939
40713	41940	#ifdef __LITTLE_ENDIAN__
40714		#define vst1q_bf16_x2(__p0, __p1) __extension__ ({ \
40715		bfloat16x8x2_t __s1 = __p1; \
40716		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 43); \
40717		})
	41941	__ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	41942	int8x16_t __ret;
	41943	__ret = vcombine_s8(__p0, vaddhn_s16(__p1, __p2));
	41944	return __ret;
	41945	}
40718	41946	#else
40719		#define vst1q_bf16_x2(__p0, __p1) __extension__ ({ \
40720		bfloat16x8x2_t __s1 = __p1; \
40721		bfloat16x8x2_t __rev1; \
40722		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40723		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40724		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 43); \
40725		})
	41947	__ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	41948	int8x16_t __ret;
	41949	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41950	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	41951	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	41952	__ret = __noswap_vcombine_s8(__rev0, __noswap_vaddhn_s16(__rev1, __rev2));
	41953	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	41954	return __ret;
	41955	}
40726	41956	#endif
40727	41957
40728	41958	#ifdef __LITTLE_ENDIAN__
40729		#define vst1_bf16_x2(__p0, __p1) __extension__ ({ \
40730		bfloat16x4x2_t __s1 = __p1; \
40731		__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 11); \
40732		})
	41959	__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
	41960	uint16_t __ret;
	41961	__ret = (uint16_t) __builtin_neon_vaddlvq_u8(__p0);
	41962	return __ret;
	41963	}
40733	41964	#else
40734		#define vst1_bf16_x2(__p0, __p1) __extension__ ({ \
40735		bfloat16x4x2_t __s1 = __p1; \
40736		bfloat16x4x2_t __rev1; \
40737		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40738		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40739		__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 11); \
40740		})
	41965	__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
	41966	uint16_t __ret;
	41967	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	41968	__ret = (uint16_t) __builtin_neon_vaddlvq_u8(__rev0);
	41969	return __ret;
	41970	}
40741	41971	#endif
40742	41972
40743	41973	#ifdef __LITTLE_ENDIAN__
40744		#define vst1q_bf16_x3(__p0, __p1) __extension__ ({ \
40745		bfloat16x8x3_t __s1 = __p1; \
40746		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 43); \
40747		})
	41974	__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
	41975	uint64_t __ret;
	41976	__ret = (uint64_t) __builtin_neon_vaddlvq_u32(__p0);
	41977	return __ret;
	41978	}
40748	41979	#else
40749		#define vst1q_bf16_x3(__p0, __p1) __extension__ ({ \
40750		bfloat16x8x3_t __s1 = __p1; \
40751		bfloat16x8x3_t __rev1; \
40752		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40753		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40754		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40755		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 43); \
40756		})
	41980	__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
	41981	uint64_t __ret;
	41982	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	41983	__ret = (uint64_t) __builtin_neon_vaddlvq_u32(__rev0);
	41984	return __ret;
	41985	}
40757	41986	#endif
40758	41987
40759	41988	#ifdef __LITTLE_ENDIAN__
40760		#define vst1_bf16_x3(__p0, __p1) __extension__ ({ \
40761		bfloat16x4x3_t __s1 = __p1; \
40762		__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 11); \
40763		})
	41989	__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
	41990	uint32_t __ret;
	41991	__ret = (uint32_t) __builtin_neon_vaddlvq_u16(__p0);
	41992	return __ret;
	41993	}
40764	41994	#else
40765		#define vst1_bf16_x3(__p0, __p1) __extension__ ({ \
40766		bfloat16x4x3_t __s1 = __p1; \
40767		bfloat16x4x3_t __rev1; \
40768		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40769		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40770		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40771		__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 11); \
40772		})
	41995	__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
	41996	uint32_t __ret;
	41997	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	41998	__ret = (uint32_t) __builtin_neon_vaddlvq_u16(__rev0);
	41999	return __ret;
	42000	}
40773	42001	#endif
40774	42002
40775	42003	#ifdef __LITTLE_ENDIAN__
40776		#define vst1q_bf16_x4(__p0, __p1) __extension__ ({ \
40777		bfloat16x8x4_t __s1 = __p1; \
40778		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 43); \
40779		})
	42004	__ai int16_t vaddlvq_s8(int8x16_t __p0) {
	42005	int16_t __ret;
	42006	__ret = (int16_t) __builtin_neon_vaddlvq_s8(__p0);
	42007	return __ret;
	42008	}
40780	42009	#else
40781		#define vst1q_bf16_x4(__p0, __p1) __extension__ ({ \
40782		bfloat16x8x4_t __s1 = __p1; \
40783		bfloat16x8x4_t __rev1; \
40784		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40785		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40786		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40787		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40788		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 43); \
40789		})
	42010	__ai int16_t vaddlvq_s8(int8x16_t __p0) {
	42011	int16_t __ret;
	42012	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42013	__ret = (int16_t) __builtin_neon_vaddlvq_s8(__rev0);
	42014	return __ret;
	42015	}
40790	42016	#endif
40791	42017
40792	42018	#ifdef __LITTLE_ENDIAN__
40793		#define vst1_bf16_x4(__p0, __p1) __extension__ ({ \
40794		bfloat16x4x4_t __s1 = __p1; \
40795		__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 11); \
40796		})
	42019	__ai int64_t vaddlvq_s32(int32x4_t __p0) {
	42020	int64_t __ret;
	42021	__ret = (int64_t) __builtin_neon_vaddlvq_s32(__p0);
	42022	return __ret;
	42023	}
40797	42024	#else
40798		#define vst1_bf16_x4(__p0, __p1) __extension__ ({ \
40799		bfloat16x4x4_t __s1 = __p1; \
40800		bfloat16x4x4_t __rev1; \
40801		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40802		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40803		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40804		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
40805		__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 11); \
40806		})
	42025	__ai int64_t vaddlvq_s32(int32x4_t __p0) {
	42026	int64_t __ret;
	42027	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42028	__ret = (int64_t) __builtin_neon_vaddlvq_s32(__rev0);
	42029	return __ret;
	42030	}
40807	42031	#endif
40808	42032
40809	42033	#ifdef __LITTLE_ENDIAN__
40810		#define vst2q_bf16(__p0, __p1) __extension__ ({ \
40811		bfloat16x8x2_t __s1 = __p1; \
40812		__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 43); \
40813		})
	42034	__ai int32_t vaddlvq_s16(int16x8_t __p0) {
	42035	int32_t __ret;
	42036	__ret = (int32_t) __builtin_neon_vaddlvq_s16(__p0);
	42037	return __ret;
	42038	}
40814	42039	#else
40815		#define vst2q_bf16(__p0, __p1) __extension__ ({ \
40816		bfloat16x8x2_t __s1 = __p1; \
40817		bfloat16x8x2_t __rev1; \
40818		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40819		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40820		__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 43); \
40821		})
	42040	__ai int32_t vaddlvq_s16(int16x8_t __p0) {
	42041	int32_t __ret;
	42042	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42043	__ret = (int32_t) __builtin_neon_vaddlvq_s16(__rev0);
	42044	return __ret;
	42045	}
40822	42046	#endif
40823	42047
40824	42048	#ifdef __LITTLE_ENDIAN__
40825		#define vst2_bf16(__p0, __p1) __extension__ ({ \
40826		bfloat16x4x2_t __s1 = __p1; \
40827		__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 11); \
40828		})
	42049	__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
	42050	uint16_t __ret;
	42051	__ret = (uint16_t) __builtin_neon_vaddlv_u8(__p0);
	42052	return __ret;
	42053	}
40829	42054	#else
40830		#define vst2_bf16(__p0, __p1) __extension__ ({ \
40831		bfloat16x4x2_t __s1 = __p1; \
40832		bfloat16x4x2_t __rev1; \
40833		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40834		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40835		__builtin_neon_vst2_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], 11); \
40836		})
	42055	__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
	42056	uint16_t __ret;
	42057	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42058	__ret = (uint16_t) __builtin_neon_vaddlv_u8(__rev0);
	42059	return __ret;
	42060	}
40837	42061	#endif
40838	42062
40839	42063	#ifdef __LITTLE_ENDIAN__
40840		#define vst2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40841		bfloat16x8x2_t __s1 = __p1; \
40842		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
40843		})
	42064	__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
	42065	uint64_t __ret;
	42066	__ret = (uint64_t) __builtin_neon_vaddlv_u32(__p0);
	42067	return __ret;
	42068	}
40844	42069	#else
40845		#define vst2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40846		bfloat16x8x2_t __s1 = __p1; \
40847		bfloat16x8x2_t __rev1; \
40848		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40849		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40850		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
40851		})
	42070	__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
	42071	uint64_t __ret;
	42072	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42073	__ret = (uint64_t) __builtin_neon_vaddlv_u32(__rev0);
	42074	return __ret;
	42075	}
40852	42076	#endif
40853	42077
40854	42078	#ifdef __LITTLE_ENDIAN__
40855		#define vst2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40856		bfloat16x4x2_t __s1 = __p1; \
40857		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
40858		})
	42079	__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
	42080	uint32_t __ret;
	42081	__ret = (uint32_t) __builtin_neon_vaddlv_u16(__p0);
	42082	return __ret;
	42083	}
40859	42084	#else
40860		#define vst2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40861		bfloat16x4x2_t __s1 = __p1; \
40862		bfloat16x4x2_t __rev1; \
40863		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40864		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40865		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
40866		})
	42085	__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
	42086	uint32_t __ret;
	42087	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42088	__ret = (uint32_t) __builtin_neon_vaddlv_u16(__rev0);
	42089	return __ret;
	42090	}
40867	42091	#endif
40868	42092
40869	42093	#ifdef __LITTLE_ENDIAN__
40870		#define vst3q_bf16(__p0, __p1) __extension__ ({ \
40871		bfloat16x8x3_t __s1 = __p1; \
40872		__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 43); \
40873		})
	42094	__ai int16_t vaddlv_s8(int8x8_t __p0) {
	42095	int16_t __ret;
	42096	__ret = (int16_t) __builtin_neon_vaddlv_s8(__p0);
	42097	return __ret;
	42098	}
40874	42099	#else
40875		#define vst3q_bf16(__p0, __p1) __extension__ ({ \
40876		bfloat16x8x3_t __s1 = __p1; \
40877		bfloat16x8x3_t __rev1; \
40878		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40879		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40880		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40881		__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 43); \
40882		})
	42100	__ai int16_t vaddlv_s8(int8x8_t __p0) {
	42101	int16_t __ret;
	42102	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42103	__ret = (int16_t) __builtin_neon_vaddlv_s8(__rev0);
	42104	return __ret;
	42105	}
40883	42106	#endif
40884	42107
40885	42108	#ifdef __LITTLE_ENDIAN__
40886		#define vst3_bf16(__p0, __p1) __extension__ ({ \
40887		bfloat16x4x3_t __s1 = __p1; \
40888		__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 11); \
40889		})
	42109	__ai int64_t vaddlv_s32(int32x2_t __p0) {
	42110	int64_t __ret;
	42111	__ret = (int64_t) __builtin_neon_vaddlv_s32(__p0);
	42112	return __ret;
	42113	}
40890	42114	#else
40891		#define vst3_bf16(__p0, __p1) __extension__ ({ \
40892		bfloat16x4x3_t __s1 = __p1; \
40893		bfloat16x4x3_t __rev1; \
40894		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40895		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40896		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40897		__builtin_neon_vst3_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], 11); \
40898		})
	42115	__ai int64_t vaddlv_s32(int32x2_t __p0) {
	42116	int64_t __ret;
	42117	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42118	__ret = (int64_t) __builtin_neon_vaddlv_s32(__rev0);
	42119	return __ret;
	42120	}
40899	42121	#endif
40900	42122
40901	42123	#ifdef __LITTLE_ENDIAN__
40902		#define vst3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40903		bfloat16x8x3_t __s1 = __p1; \
40904		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
40905		})
	42124	__ai int32_t vaddlv_s16(int16x4_t __p0) {
	42125	int32_t __ret;
	42126	__ret = (int32_t) __builtin_neon_vaddlv_s16(__p0);
	42127	return __ret;
	42128	}
40906	42129	#else
40907		#define vst3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40908		bfloat16x8x3_t __s1 = __p1; \
40909		bfloat16x8x3_t __rev1; \
40910		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40911		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40912		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40913		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
40914		})
	42130	__ai int32_t vaddlv_s16(int16x4_t __p0) {
	42131	int32_t __ret;
	42132	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42133	__ret = (int32_t) __builtin_neon_vaddlv_s16(__rev0);
	42134	return __ret;
	42135	}
40915	42136	#endif
40916	42137
40917	42138	#ifdef __LITTLE_ENDIAN__
40918		#define vst3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40919		bfloat16x4x3_t __s1 = __p1; \
40920		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
40921		})
	42139	__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
	42140	uint8_t __ret;
	42141	__ret = (uint8_t) __builtin_neon_vaddvq_u8(__p0);
	42142	return __ret;
	42143	}
40922	42144	#else
40923		#define vst3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40924		bfloat16x4x3_t __s1 = __p1; \
40925		bfloat16x4x3_t __rev1; \
40926		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40927		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40928		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40929		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
40930		})
	42145	__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
	42146	uint8_t __ret;
	42147	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42148	__ret = (uint8_t) __builtin_neon_vaddvq_u8(__rev0);
	42149	return __ret;
	42150	}
40931	42151	#endif
40932	42152
40933	42153	#ifdef __LITTLE_ENDIAN__
40934		#define vst4q_bf16(__p0, __p1) __extension__ ({ \
40935		bfloat16x8x4_t __s1 = __p1; \
40936		__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 43); \
40937		})
	42154	__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
	42155	uint32_t __ret;
	42156	__ret = (uint32_t) __builtin_neon_vaddvq_u32(__p0);
	42157	return __ret;
	42158	}
40938	42159	#else
40939		#define vst4q_bf16(__p0, __p1) __extension__ ({ \
40940		bfloat16x8x4_t __s1 = __p1; \
40941		bfloat16x8x4_t __rev1; \
40942		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40943		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40944		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40945		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40946		__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 43); \
40947		})
	42160	__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
	42161	uint32_t __ret;
	42162	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42163	__ret = (uint32_t) __builtin_neon_vaddvq_u32(__rev0);
	42164	return __ret;
	42165	}
40948	42166	#endif
40949	42167
40950	42168	#ifdef __LITTLE_ENDIAN__
40951		#define vst4_bf16(__p0, __p1) __extension__ ({ \
40952		bfloat16x4x4_t __s1 = __p1; \
40953		__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 11); \
40954		})
40955		#else
40956		#define vst4_bf16(__p0, __p1) __extension__ ({ \
40957		bfloat16x4x4_t __s1 = __p1; \
40958		bfloat16x4x4_t __rev1; \
40959		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40960		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40961		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40962		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
40963		__builtin_neon_vst4_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], 11); \
40964		})
	42169	__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
	42170	uint64_t __ret;
	42171	__ret = (uint64_t) __builtin_neon_vaddvq_u64(__p0);
	42172	return __ret;
	42173	}
	42174	#else
	42175	__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
	42176	uint64_t __ret;
	42177	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42178	__ret = (uint64_t) __builtin_neon_vaddvq_u64(__rev0);
	42179	return __ret;
	42180	}
40965	42181	#endif
40966	42182
40967	42183	#ifdef __LITTLE_ENDIAN__
40968		#define vst4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40969		bfloat16x8x4_t __s1 = __p1; \
40970		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 43); \
40971		})
	42184	__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
	42185	uint16_t __ret;
	42186	__ret = (uint16_t) __builtin_neon_vaddvq_u16(__p0);
	42187	return __ret;
	42188	}
40972	42189	#else
40973		#define vst4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40974		bfloat16x8x4_t __s1 = __p1; \
40975		bfloat16x8x4_t __rev1; \
40976		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
40977		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40978		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40979		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40980		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 43); \
40981		})
	42190	__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
	42191	uint16_t __ret;
	42192	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42193	__ret = (uint16_t) __builtin_neon_vaddvq_u16(__rev0);
	42194	return __ret;
	42195	}
40982	42196	#endif
40983	42197
40984	42198	#ifdef __LITTLE_ENDIAN__
40985		#define vst4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40986		bfloat16x4x4_t __s1 = __p1; \
40987		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 11); \
40988		})
	42199	__ai int8_t vaddvq_s8(int8x16_t __p0) {
	42200	int8_t __ret;
	42201	__ret = (int8_t) __builtin_neon_vaddvq_s8(__p0);
	42202	return __ret;
	42203	}
40989	42204	#else
40990		#define vst4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40991		bfloat16x4x4_t __s1 = __p1; \
40992		bfloat16x4x4_t __rev1; \
40993		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
40994		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40995		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40996		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
40997		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], __p2, 11); \
40998		})
	42205	__ai int8_t vaddvq_s8(int8x16_t __p0) {
	42206	int8_t __ret;
	42207	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42208	__ret = (int8_t) __builtin_neon_vaddvq_s8(__rev0);
	42209	return __ret;
	42210	}
40999	42211	#endif
41000	42212
41001		#endif
41002		#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC) && !defined(__aarch64__)
41003	42213	#ifdef __LITTLE_ENDIAN__
41004		__ai bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
41005		bfloat16x4_t __ret;
41006		__ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__p0, 11);
	42214	__ai float64_t vaddvq_f64(float64x2_t __p0) {
	42215	float64_t __ret;
	42216	__ret = (float64_t) __builtin_neon_vaddvq_f64(__p0);
41007	42217	return __ret;
41008	42218	}
41009	42219	#else
41010		__ai bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
41011		bfloat16x4_t __ret;
41012		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41013		__ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__rev0, 11);
41014		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
41015		return __ret;
41016		}
41017		__ai bfloat16x4_t __noswap___a32_vcvt_bf16_f32(float32x4_t __p0) {
41018		bfloat16x4_t __ret;
41019		__ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__p0, 11);
	42220	__ai float64_t vaddvq_f64(float64x2_t __p0) {
	42221	float64_t __ret;
	42222	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42223	__ret = (float64_t) __builtin_neon_vaddvq_f64(__rev0);
41020	42224	return __ret;
41021	42225	}
41022	42226	#endif
41023	42227
41024	42228	#ifdef __LITTLE_ENDIAN__
41025		__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
41026		bfloat16x4_t __ret;
41027		__ret = __a32_vcvt_bf16_f32(__p0);
	42229	__ai float32_t vaddvq_f32(float32x4_t __p0) {
	42230	float32_t __ret;
	42231	__ret = (float32_t) __builtin_neon_vaddvq_f32(__p0);
41028	42232	return __ret;
41029	42233	}
41030	42234	#else
41031		__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
41032		bfloat16x4_t __ret;
	42235	__ai float32_t vaddvq_f32(float32x4_t __p0) {
	42236	float32_t __ret;
41033	42237	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41034		__ret = __noswap___a32_vcvt_bf16_f32(__rev0);
41035		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42238	__ret = (float32_t) __builtin_neon_vaddvq_f32(__rev0);
41036	42239	return __ret;
41037	42240	}
41038	42241	#endif
41039	42242
41040	42243	#ifdef __LITTLE_ENDIAN__
41041		__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
41042		bfloat16x8_t __ret;
41043		__ret = vcombine_bf16(__a32_vcvt_bf16_f32(__p1), vget_low_bf16(__p0));
	42244	__ai int32_t vaddvq_s32(int32x4_t __p0) {
	42245	int32_t __ret;
	42246	__ret = (int32_t) __builtin_neon_vaddvq_s32(__p0);
41044	42247	return __ret;
41045	42248	}
41046	42249	#else
41047		__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
41048		bfloat16x8_t __ret;
41049		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
41050		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41051		__ret = __noswap_vcombine_bf16(__noswap___a32_vcvt_bf16_f32(__rev1), __noswap_vget_low_bf16(__rev0));
41052		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42250	__ai int32_t vaddvq_s32(int32x4_t __p0) {
	42251	int32_t __ret;
	42252	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42253	__ret = (int32_t) __builtin_neon_vaddvq_s32(__rev0);
41053	42254	return __ret;
41054	42255	}
41055	42256	#endif
41056	42257
41057	42258	#ifdef __LITTLE_ENDIAN__
41058		__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
41059		bfloat16x8_t __ret;
41060		__ret = vcombine_bf16((bfloat16x4_t)(0ULL), __a32_vcvt_bf16_f32(__p0));
	42259	__ai int64_t vaddvq_s64(int64x2_t __p0) {
	42260	int64_t __ret;
	42261	__ret = (int64_t) __builtin_neon_vaddvq_s64(__p0);
41061	42262	return __ret;
41062	42263	}
41063	42264	#else
41064		__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
41065		bfloat16x8_t __ret;
41066		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41067		__ret = __noswap_vcombine_bf16((bfloat16x4_t)(0ULL), __noswap___a32_vcvt_bf16_f32(__rev0));
41068		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42265	__ai int64_t vaddvq_s64(int64x2_t __p0) {
	42266	int64_t __ret;
	42267	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42268	__ret = (int64_t) __builtin_neon_vaddvq_s64(__rev0);
41069	42269	return __ret;
41070	42270	}
41071	42271	#endif
41072	42272
41073		#endif
41074		#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC) && defined(__aarch64__)
41075	42273	#ifdef __LITTLE_ENDIAN__
41076		__ai bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
41077		bfloat16x8_t __ret;
41078		__ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__p0, 43);
	42274	__ai int16_t vaddvq_s16(int16x8_t __p0) {
	42275	int16_t __ret;
	42276	__ret = (int16_t) __builtin_neon_vaddvq_s16(__p0);
41079	42277	return __ret;
41080	42278	}
41081	42279	#else
41082		__ai bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
41083		bfloat16x8_t __ret;
41084		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41085		__ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__rev0, 43);
41086		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
41087		return __ret;
41088		}
41089		__ai bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
41090		bfloat16x8_t __ret;
41091		__ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__p0, 43);
	42280	__ai int16_t vaddvq_s16(int16x8_t __p0) {
	42281	int16_t __ret;
	42282	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42283	__ret = (int16_t) __builtin_neon_vaddvq_s16(__rev0);
41092	42284	return __ret;
41093	42285	}
41094	42286	#endif
41095	42287
41096	42288	#ifdef __LITTLE_ENDIAN__
41097		#define vcopyq_lane_bf16(__p0_161, __p1_161, __p2_161, __p3_161) __extension__ ({ \
41098		bfloat16x8_t __ret_161; \
41099		bfloat16x8_t __s0_161 = __p0_161; \
41100		bfloat16x4_t __s2_161 = __p2_161; \
41101		__ret_161 = vsetq_lane_bf16(vget_lane_bf16(__s2_161, __p3_161), __s0_161, __p1_161); \
41102		__ret_161; \
41103		})
	42289	__ai uint8_t vaddv_u8(uint8x8_t __p0) {
	42290	uint8_t __ret;
	42291	__ret = (uint8_t) __builtin_neon_vaddv_u8(__p0);
	42292	return __ret;
	42293	}
41104	42294	#else
41105		#define vcopyq_lane_bf16(__p0_162, __p1_162, __p2_162, __p3_162) __extension__ ({ \
41106		bfloat16x8_t __ret_162; \
41107		bfloat16x8_t __s0_162 = __p0_162; \
41108		bfloat16x4_t __s2_162 = __p2_162; \
41109		bfloat16x8_t __rev0_162; __rev0_162 = __builtin_shufflevector(__s0_162, __s0_162, 7, 6, 5, 4, 3, 2, 1, 0); \
41110		bfloat16x4_t __rev2_162; __rev2_162 = __builtin_shufflevector(__s2_162, __s2_162, 3, 2, 1, 0); \
41111		__ret_162 = __noswap_vsetq_lane_bf16(__noswap_vget_lane_bf16(__rev2_162, __p3_162), __rev0_162, __p1_162); \
41112		__ret_162 = __builtin_shufflevector(__ret_162, __ret_162, 7, 6, 5, 4, 3, 2, 1, 0); \
41113		__ret_162; \
41114		})
	42295	__ai uint8_t vaddv_u8(uint8x8_t __p0) {
	42296	uint8_t __ret;
	42297	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42298	__ret = (uint8_t) __builtin_neon_vaddv_u8(__rev0);
	42299	return __ret;
	42300	}
41115	42301	#endif
41116	42302
41117	42303	#ifdef __LITTLE_ENDIAN__
41118		#define vcopy_lane_bf16(__p0_163, __p1_163, __p2_163, __p3_163) __extension__ ({ \
41119		bfloat16x4_t __ret_163; \
41120		bfloat16x4_t __s0_163 = __p0_163; \
41121		bfloat16x4_t __s2_163 = __p2_163; \
41122		__ret_163 = vset_lane_bf16(vget_lane_bf16(__s2_163, __p3_163), __s0_163, __p1_163); \
41123		__ret_163; \
41124		})
	42304	__ai uint32_t vaddv_u32(uint32x2_t __p0) {
	42305	uint32_t __ret;
	42306	__ret = (uint32_t) __builtin_neon_vaddv_u32(__p0);
	42307	return __ret;
	42308	}
41125	42309	#else
41126		#define vcopy_lane_bf16(__p0_164, __p1_164, __p2_164, __p3_164) __extension__ ({ \
41127		bfloat16x4_t __ret_164; \
41128		bfloat16x4_t __s0_164 = __p0_164; \
41129		bfloat16x4_t __s2_164 = __p2_164; \
41130		bfloat16x4_t __rev0_164; __rev0_164 = __builtin_shufflevector(__s0_164, __s0_164, 3, 2, 1, 0); \
41131		bfloat16x4_t __rev2_164; __rev2_164 = __builtin_shufflevector(__s2_164, __s2_164, 3, 2, 1, 0); \
41132		__ret_164 = __noswap_vset_lane_bf16(__noswap_vget_lane_bf16(__rev2_164, __p3_164), __rev0_164, __p1_164); \
41133		__ret_164 = __builtin_shufflevector(__ret_164, __ret_164, 3, 2, 1, 0); \
41134		__ret_164; \
41135		})
	42310	__ai uint32_t vaddv_u32(uint32x2_t __p0) {
	42311	uint32_t __ret;
	42312	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42313	__ret = (uint32_t) __builtin_neon_vaddv_u32(__rev0);
	42314	return __ret;
	42315	}
41136	42316	#endif
41137	42317
41138	42318	#ifdef __LITTLE_ENDIAN__
41139		#define vcopyq_laneq_bf16(__p0_165, __p1_165, __p2_165, __p3_165) __extension__ ({ \
41140		bfloat16x8_t __ret_165; \
41141		bfloat16x8_t __s0_165 = __p0_165; \
41142		bfloat16x8_t __s2_165 = __p2_165; \
41143		__ret_165 = vsetq_lane_bf16(vgetq_lane_bf16(__s2_165, __p3_165), __s0_165, __p1_165); \
41144		__ret_165; \
41145		})
	42319	__ai uint16_t vaddv_u16(uint16x4_t __p0) {
	42320	uint16_t __ret;
	42321	__ret = (uint16_t) __builtin_neon_vaddv_u16(__p0);
	42322	return __ret;
	42323	}
41146	42324	#else
41147		#define vcopyq_laneq_bf16(__p0_166, __p1_166, __p2_166, __p3_166) __extension__ ({ \
41148		bfloat16x8_t __ret_166; \
41149		bfloat16x8_t __s0_166 = __p0_166; \
41150		bfloat16x8_t __s2_166 = __p2_166; \
41151		bfloat16x8_t __rev0_166; __rev0_166 = __builtin_shufflevector(__s0_166, __s0_166, 7, 6, 5, 4, 3, 2, 1, 0); \
41152		bfloat16x8_t __rev2_166; __rev2_166 = __builtin_shufflevector(__s2_166, __s2_166, 7, 6, 5, 4, 3, 2, 1, 0); \
41153		__ret_166 = __noswap_vsetq_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_166, __p3_166), __rev0_166, __p1_166); \
41154		__ret_166 = __builtin_shufflevector(__ret_166, __ret_166, 7, 6, 5, 4, 3, 2, 1, 0); \
41155		__ret_166; \
41156		})
	42325	__ai uint16_t vaddv_u16(uint16x4_t __p0) {
	42326	uint16_t __ret;
	42327	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42328	__ret = (uint16_t) __builtin_neon_vaddv_u16(__rev0);
	42329	return __ret;
	42330	}
41157	42331	#endif
41158	42332
41159	42333	#ifdef __LITTLE_ENDIAN__
41160		#define vcopy_laneq_bf16(__p0_167, __p1_167, __p2_167, __p3_167) __extension__ ({ \
41161		bfloat16x4_t __ret_167; \
41162		bfloat16x4_t __s0_167 = __p0_167; \
41163		bfloat16x8_t __s2_167 = __p2_167; \
41164		__ret_167 = vset_lane_bf16(vgetq_lane_bf16(__s2_167, __p3_167), __s0_167, __p1_167); \
41165		__ret_167; \
41166		})
	42334	__ai int8_t vaddv_s8(int8x8_t __p0) {
	42335	int8_t __ret;
	42336	__ret = (int8_t) __builtin_neon_vaddv_s8(__p0);
	42337	return __ret;
	42338	}
41167	42339	#else
41168		#define vcopy_laneq_bf16(__p0_168, __p1_168, __p2_168, __p3_168) __extension__ ({ \
41169		bfloat16x4_t __ret_168; \
41170		bfloat16x4_t __s0_168 = __p0_168; \
41171		bfloat16x8_t __s2_168 = __p2_168; \
41172		bfloat16x4_t __rev0_168; __rev0_168 = __builtin_shufflevector(__s0_168, __s0_168, 3, 2, 1, 0); \
41173		bfloat16x8_t __rev2_168; __rev2_168 = __builtin_shufflevector(__s2_168, __s2_168, 7, 6, 5, 4, 3, 2, 1, 0); \
41174		__ret_168 = __noswap_vset_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_168, __p3_168), __rev0_168, __p1_168); \
41175		__ret_168 = __builtin_shufflevector(__ret_168, __ret_168, 3, 2, 1, 0); \
41176		__ret_168; \
41177		})
	42340	__ai int8_t vaddv_s8(int8x8_t __p0) {
	42341	int8_t __ret;
	42342	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42343	__ret = (int8_t) __builtin_neon_vaddv_s8(__rev0);
	42344	return __ret;
	42345	}
41178	42346	#endif
41179	42347
41180	42348	#ifdef __LITTLE_ENDIAN__
41181		__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
41182		bfloat16x4_t __ret;
41183		__ret = vget_low_bf16(__a64_vcvtq_low_bf16_f32(__p0));
	42349	__ai float32_t vaddv_f32(float32x2_t __p0) {
	42350	float32_t __ret;
	42351	__ret = (float32_t) __builtin_neon_vaddv_f32(__p0);
41184	42352	return __ret;
41185	42353	}
41186	42354	#else
41187		__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
41188		bfloat16x4_t __ret;
41189		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41190		__ret = __noswap_vget_low_bf16(__noswap___a64_vcvtq_low_bf16_f32(__rev0));
41191		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42355	__ai float32_t vaddv_f32(float32x2_t __p0) {
	42356	float32_t __ret;
	42357	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42358	__ret = (float32_t) __builtin_neon_vaddv_f32(__rev0);
41192	42359	return __ret;
41193	42360	}
41194	42361	#endif
41195	42362
41196	42363	#ifdef __LITTLE_ENDIAN__
41197		__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
41198		bfloat16x8_t __ret;
41199		__ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_v((int8x16_t)__p0, (int8x16_t)__p1, 43);
	42364	__ai int32_t vaddv_s32(int32x2_t __p0) {
	42365	int32_t __ret;
	42366	__ret = (int32_t) __builtin_neon_vaddv_s32(__p0);
41200	42367	return __ret;
41201	42368	}
41202	42369	#else
41203		__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
41204		bfloat16x8_t __ret;
41205		bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
41206		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41207		__ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_v((int8x16_t)__rev0, (int8x16_t)__rev1, 43);
41208		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42370	__ai int32_t vaddv_s32(int32x2_t __p0) {
	42371	int32_t __ret;
	42372	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42373	__ret = (int32_t) __builtin_neon_vaddv_s32(__rev0);
41209	42374	return __ret;
41210	42375	}
41211	42376	#endif
41212	42377
41213	42378	#ifdef __LITTLE_ENDIAN__
41214		__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
41215		bfloat16x8_t __ret;
41216		__ret = __a64_vcvtq_low_bf16_f32(__p0);
	42379	__ai int16_t vaddv_s16(int16x4_t __p0) {
	42380	int16_t __ret;
	42381	__ret = (int16_t) __builtin_neon_vaddv_s16(__p0);
41217	42382	return __ret;
41218	42383	}
41219	42384	#else
41220		__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
41221		bfloat16x8_t __ret;
41222		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41223		__ret = __noswap___a64_vcvtq_low_bf16_f32(__rev0);
41224		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42385	__ai int16_t vaddv_s16(int16x4_t __p0) {
	42386	int16_t __ret;
	42387	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42388	__ret = (int16_t) __builtin_neon_vaddv_s16(__rev0);
41225	42389	return __ret;
41226	42390	}
41227	42391	#endif
41228	42392
41229		#endif
41230		#if defined(__ARM_FEATURE_COMPLEX)
	42393	__ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
	42394	poly64x1_t __ret;
	42395	__ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
	42396	return __ret;
	42397	}
41231	42398	#ifdef __LITTLE_ENDIAN__
41232		__ai float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
41233		float32x2_t __ret;
41234		__ret = (float32x2_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	42399	__ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
	42400	poly64x2_t __ret;
	42401	__ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 38);
41235	42402	return __ret;
41236	42403	}
41237	42404	#else
41238		__ai float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
41239		float32x2_t __ret;
41240		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41241		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41242		__ret = (float32x2_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
	42405	__ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
	42406	poly64x2_t __ret;
	42407	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42408	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42409	poly64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	42410	__ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 38);
41243	42411	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41244	42412	return __ret;
41245	42413	}
41246	42414	#endif
41247	42415
41248	42416	#ifdef __LITTLE_ENDIAN__
41249		__ai float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
41250		float32x2_t __ret;
41251		__ret = (float32x2_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	42417	__ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	42418	float64x2_t __ret;
	42419	__ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
41252	42420	return __ret;
41253	42421	}
41254	42422	#else
41255		__ai float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
41256		float32x2_t __ret;
41257		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41258		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41259		__ret = (float32x2_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
	42423	__ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	42424	float64x2_t __ret;
	42425	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42426	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42427	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	42428	__ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
41260	42429	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41261	42430	return __ret;
41262	42431	}
41263	42432	#endif
41264	42433
	42434	__ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	42435	float64x1_t __ret;
	42436	__ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	42437	return __ret;
	42438	}
41265	42439	#ifdef __LITTLE_ENDIAN__
41266		__ai float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
41267		float32x4_t __ret;
41268		__ret = (float32x4_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	42440	__ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
	42441	uint64x2_t __ret;
	42442	__ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41269	42443	return __ret;
41270	42444	}
41271	42445	#else
41272		__ai float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
41273		float32x4_t __ret;
41274		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41275		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41276		__ret = (float32x4_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
41277		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42446	__ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
	42447	uint64x2_t __ret;
	42448	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42449	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42450	__ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	42451	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41278	42452	return __ret;
41279	42453	}
41280	42454	#endif
41281	42455
	42456	__ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
	42457	uint64x1_t __ret;
	42458	__ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
	42459	return __ret;
	42460	}
	42461	__ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
	42462	uint64_t __ret;
	42463	__ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
	42464	return __ret;
	42465	}
	42466	__ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
	42467	uint32_t __ret;
	42468	__ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
	42469	return __ret;
	42470	}
41282	42471	#ifdef __LITTLE_ENDIAN__
41283		__ai float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
41284		float32x4_t __ret;
41285		__ret = (float32x4_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	42472	__ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
	42473	uint64x2_t __ret;
	42474	__ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41286	42475	return __ret;
41287	42476	}
41288	42477	#else
41289		__ai float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
41290		float32x4_t __ret;
41291		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41292		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41293		__ret = (float32x4_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
41294		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42478	__ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
	42479	uint64x2_t __ret;
	42480	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42481	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42482	__ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	42483	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41295	42484	return __ret;
41296	42485	}
41297	42486	#endif
41298	42487
41299		#ifdef __LITTLE_ENDIAN__
41300		__ai float32x4_t vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41301		float32x4_t __ret;
41302		__ret = (float32x4_t) __builtin_neon_vcmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42488	__ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
	42489	uint64x1_t __ret;
	42490	__ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41303	42491	return __ret;
41304	42492	}
41305		#else
41306		__ai float32x4_t vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41307		float32x4_t __ret;
41308		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41309		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41310		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41311		__ret = (float32x4_t) __builtin_neon_vcmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
41312		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42493	__ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
	42494	uint64_t __ret;
	42495	__ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
41313	42496	return __ret;
41314	42497	}
41315		__ai float32x4_t __noswap_vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41316		float32x4_t __ret;
41317		__ret = (float32x4_t) __builtin_neon_vcmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42498	__ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
	42499	uint32_t __ret;
	42500	__ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
41318	42501	return __ret;
41319	42502	}
41320		#endif
41321
41322	42503	#ifdef __LITTLE_ENDIAN__
41323		__ai float32x2_t vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41324		float32x2_t __ret;
41325		__ret = (float32x2_t) __builtin_neon_vcmla_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	42504	__ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
	42505	uint64x2_t __ret;
	42506	__ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41326	42507	return __ret;
41327	42508	}
41328	42509	#else
41329		__ai float32x2_t vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41330		float32x2_t __ret;
41331		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41332		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41333		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41334		__ret = (float32x2_t) __builtin_neon_vcmla_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	42510	__ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
	42511	uint64x2_t __ret;
	42512	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42513	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42514	__ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
41335	42515	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41336	42516	return __ret;
41337	42517	}
41338		__ai float32x2_t __noswap_vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41339		float32x2_t __ret;
41340		__ret = (float32x2_t) __builtin_neon_vcmla_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	42518	#endif
	42519
	42520	__ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
	42521	uint64x1_t __ret;
	42522	__ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
	42523	return __ret;
	42524	}
	42525	__ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
	42526	uint64_t __ret;
	42527	__ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
	42528	return __ret;
	42529	}
	42530	__ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
	42531	uint32_t __ret;
	42532	__ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
	42533	return __ret;
	42534	}
	42535	#ifdef __LITTLE_ENDIAN__
	42536	__ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
	42537	uint64x2_t __ret;
	42538	__ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	42539	return __ret;
	42540	}
	42541	#else
	42542	__ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
	42543	uint64x2_t __ret;
	42544	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42545	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42546	__ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	42547	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41341	42548	return __ret;
41342	42549	}
41343	42550	#endif
41344	42551
	42552	__ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
	42553	uint64x1_t __ret;
	42554	__ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
	42555	return __ret;
	42556	}
	42557	__ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
	42558	uint64_t __ret;
	42559	__ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
	42560	return __ret;
	42561	}
	42562	__ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
	42563	uint32_t __ret;
	42564	__ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
	42565	return __ret;
	42566	}
	42567	__ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
	42568	uint64x1_t __ret;
	42569	__ret = (uint64x1_t)(__p0 == __p1);
	42570	return __ret;
	42571	}
41345	42572	#ifdef __LITTLE_ENDIAN__
41346		#define vcmla_lane_f32(__p0_169, __p1_169, __p2_169, __p3_169) __extension__ ({ \
41347		float32x2_t __ret_169; \
41348		float32x2_t __s0_169 = __p0_169; \
41349		float32x2_t __s1_169 = __p1_169; \
41350		float32x2_t __s2_169 = __p2_169; \
41351		float32x2_t __reint_169 = __s2_169; \
41352		uint64x1_t __reint1_169 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_169, __p3_169)}; \
41353		__ret_169 = vcmla_f32(__s0_169, __s1_169, (float32x2_t ) &__reint1_169); \
41354		__ret_169; \
41355		})
	42573	__ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
	42574	uint64x2_t __ret;
	42575	__ret = (uint64x2_t)(__p0 == __p1);
	42576	return __ret;
	42577	}
41356	42578	#else
41357		#define vcmla_lane_f32(__p0_170, __p1_170, __p2_170, __p3_170) __extension__ ({ \
41358		float32x2_t __ret_170; \
41359		float32x2_t __s0_170 = __p0_170; \
41360		float32x2_t __s1_170 = __p1_170; \
41361		float32x2_t __s2_170 = __p2_170; \
41362		float32x2_t __rev0_170; __rev0_170 = __builtin_shufflevector(__s0_170, __s0_170, 1, 0); \
41363		float32x2_t __rev1_170; __rev1_170 = __builtin_shufflevector(__s1_170, __s1_170, 1, 0); \
41364		float32x2_t __rev2_170; __rev2_170 = __builtin_shufflevector(__s2_170, __s2_170, 1, 0); \
41365		float32x2_t __reint_170 = __rev2_170; \
41366		uint64x1_t __reint1_170 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_170, __p3_170)}; \
41367		__ret_170 = __noswap_vcmla_f32(__rev0_170, __rev1_170, (float32x2_t ) &__reint1_170); \
41368		__ret_170 = __builtin_shufflevector(__ret_170, __ret_170, 1, 0); \
41369		__ret_170; \
41370		})
	42579	__ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
	42580	uint64x2_t __ret;
	42581	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42582	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42583	__ret = (uint64x2_t)(__rev0 == __rev1);
	42584	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42585	return __ret;
	42586	}
41371	42587	#endif
41372	42588
41373	42589	#ifdef __LITTLE_ENDIAN__
41374		#define vcmlaq_lane_f32(__p0_171, __p1_171, __p2_171, __p3_171) __extension__ ({ \
41375		float32x4_t __ret_171; \
41376		float32x4_t __s0_171 = __p0_171; \
41377		float32x4_t __s1_171 = __p1_171; \
41378		float32x2_t __s2_171 = __p2_171; \
41379		float32x2_t __reint_171 = __s2_171; \
41380		uint64x2_t __reint1_171 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_171, __p3_171), vget_lane_u64((uint64x1_t ) &__reint_171, __p3_171)}; \
41381		__ret_171 = vcmlaq_f32(__s0_171, __s1_171, (float32x4_t ) &__reint1_171); \
41382		__ret_171; \
41383		})
	42590	__ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	42591	uint64x2_t __ret;
	42592	__ret = (uint64x2_t)(__p0 == __p1);
	42593	return __ret;
	42594	}
41384	42595	#else
41385		#define vcmlaq_lane_f32(__p0_172, __p1_172, __p2_172, __p3_172) __extension__ ({ \
41386		float32x4_t __ret_172; \
41387		float32x4_t __s0_172 = __p0_172; \
41388		float32x4_t __s1_172 = __p1_172; \
41389		float32x2_t __s2_172 = __p2_172; \
41390		float32x4_t __rev0_172; __rev0_172 = __builtin_shufflevector(__s0_172, __s0_172, 3, 2, 1, 0); \
41391		float32x4_t __rev1_172; __rev1_172 = __builtin_shufflevector(__s1_172, __s1_172, 3, 2, 1, 0); \
41392		float32x2_t __rev2_172; __rev2_172 = __builtin_shufflevector(__s2_172, __s2_172, 1, 0); \
41393		float32x2_t __reint_172 = __rev2_172; \
41394		uint64x2_t __reint1_172 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_172, __p3_172), vget_lane_u64((uint64x1_t ) &__reint_172, __p3_172)}; \
41395		__ret_172 = __noswap_vcmlaq_f32(__rev0_172, __rev1_172, (float32x4_t ) &__reint1_172); \
41396		__ret_172 = __builtin_shufflevector(__ret_172, __ret_172, 3, 2, 1, 0); \
41397		__ret_172; \
41398		})
	42596	__ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	42597	uint64x2_t __ret;
	42598	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42599	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42600	__ret = (uint64x2_t)(__rev0 == __rev1);
	42601	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42602	return __ret;
	42603	}
41399	42604	#endif
41400	42605
41401	42606	#ifdef __LITTLE_ENDIAN__
41402		#define vcmla_laneq_f32(__p0_173, __p1_173, __p2_173, __p3_173) __extension__ ({ \
41403		float32x2_t __ret_173; \
41404		float32x2_t __s0_173 = __p0_173; \
41405		float32x2_t __s1_173 = __p1_173; \
41406		float32x4_t __s2_173 = __p2_173; \
41407		float32x4_t __reint_173 = __s2_173; \
41408		uint64x1_t __reint1_173 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_173, __p3_173)}; \
41409		__ret_173 = vcmla_f32(__s0_173, __s1_173, (float32x2_t ) &__reint1_173); \
41410		__ret_173; \
41411		})
	42607	__ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
	42608	uint64x2_t __ret;
	42609	__ret = (uint64x2_t)(__p0 == __p1);
	42610	return __ret;
	42611	}
41412	42612	#else
41413		#define vcmla_laneq_f32(__p0_174, __p1_174, __p2_174, __p3_174) __extension__ ({ \
41414		float32x2_t __ret_174; \
41415		float32x2_t __s0_174 = __p0_174; \
41416		float32x2_t __s1_174 = __p1_174; \
41417		float32x4_t __s2_174 = __p2_174; \
41418		float32x2_t __rev0_174; __rev0_174 = __builtin_shufflevector(__s0_174, __s0_174, 1, 0); \
41419		float32x2_t __rev1_174; __rev1_174 = __builtin_shufflevector(__s1_174, __s1_174, 1, 0); \
41420		float32x4_t __rev2_174; __rev2_174 = __builtin_shufflevector(__s2_174, __s2_174, 3, 2, 1, 0); \
41421		float32x4_t __reint_174 = __rev2_174; \
41422		uint64x1_t __reint1_174 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_174, __p3_174)}; \
41423		__ret_174 = __noswap_vcmla_f32(__rev0_174, __rev1_174, (float32x2_t ) &__reint1_174); \
41424		__ret_174 = __builtin_shufflevector(__ret_174, __ret_174, 1, 0); \
41425		__ret_174; \
41426		})
	42613	__ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
	42614	uint64x2_t __ret;
	42615	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42616	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42617	__ret = (uint64x2_t)(__rev0 == __rev1);
	42618	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42619	return __ret;
	42620	}
41427	42621	#endif
41428	42622
41429	42623	#ifdef __LITTLE_ENDIAN__
41430		#define vcmlaq_laneq_f32(__p0_175, __p1_175, __p2_175, __p3_175) __extension__ ({ \
41431		float32x4_t __ret_175; \
41432		float32x4_t __s0_175 = __p0_175; \
41433		float32x4_t __s1_175 = __p1_175; \
41434		float32x4_t __s2_175 = __p2_175; \
41435		float32x4_t __reint_175 = __s2_175; \
41436		uint64x2_t __reint1_175 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_175, __p3_175), vgetq_lane_u64((uint64x2_t ) &__reint_175, __p3_175)}; \
41437		__ret_175 = vcmlaq_f32(__s0_175, __s1_175, (float32x4_t ) &__reint1_175); \
41438		__ret_175; \
41439		})
	42624	__ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
	42625	uint64x2_t __ret;
	42626	__ret = (uint64x2_t)(__p0 == __p1);
	42627	return __ret;
	42628	}
41440	42629	#else
41441		#define vcmlaq_laneq_f32(__p0_176, __p1_176, __p2_176, __p3_176) __extension__ ({ \
41442		float32x4_t __ret_176; \
41443		float32x4_t __s0_176 = __p0_176; \
41444		float32x4_t __s1_176 = __p1_176; \
41445		float32x4_t __s2_176 = __p2_176; \
41446		float32x4_t __rev0_176; __rev0_176 = __builtin_shufflevector(__s0_176, __s0_176, 3, 2, 1, 0); \
41447		float32x4_t __rev1_176; __rev1_176 = __builtin_shufflevector(__s1_176, __s1_176, 3, 2, 1, 0); \
41448		float32x4_t __rev2_176; __rev2_176 = __builtin_shufflevector(__s2_176, __s2_176, 3, 2, 1, 0); \
41449		float32x4_t __reint_176 = __rev2_176; \
41450		uint64x2_t __reint1_176 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_176, __p3_176), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_176, __p3_176)}; \
41451		__ret_176 = __noswap_vcmlaq_f32(__rev0_176, __rev1_176, (float32x4_t ) &__reint1_176); \
41452		__ret_176 = __builtin_shufflevector(__ret_176, __ret_176, 3, 2, 1, 0); \
41453		__ret_176; \
41454		})
	42630	__ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
	42631	uint64x2_t __ret;
	42632	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42633	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	42634	__ret = (uint64x2_t)(__rev0 == __rev1);
	42635	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42636	return __ret;
	42637	}
41455	42638	#endif
41456	42639
	42640	__ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
	42641	uint64x1_t __ret;
	42642	__ret = (uint64x1_t)(__p0 == __p1);
	42643	return __ret;
	42644	}
	42645	__ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
	42646	uint64x1_t __ret;
	42647	__ret = (uint64x1_t)(__p0 == __p1);
	42648	return __ret;
	42649	}
	42650	__ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
	42651	uint64x1_t __ret;
	42652	__ret = (uint64x1_t)(__p0 == __p1);
	42653	return __ret;
	42654	}
	42655	__ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
	42656	uint64_t __ret;
	42657	__ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
	42658	return __ret;
	42659	}
	42660	__ai uint64_t vceqd_s64(int64_t __p0, int64_t __p1) {
	42661	uint64_t __ret;
	42662	__ret = (uint64_t) __builtin_neon_vceqd_s64(__p0, __p1);
	42663	return __ret;
	42664	}
	42665	__ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
	42666	uint64_t __ret;
	42667	__ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
	42668	return __ret;
	42669	}
	42670	__ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
	42671	uint32_t __ret;
	42672	__ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
	42673	return __ret;
	42674	}
41457	42675	#ifdef __LITTLE_ENDIAN__
41458		__ai float32x4_t vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41459		float32x4_t __ret;
41460		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42676	__ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
	42677	uint8x8_t __ret;
	42678	__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 16);
41461	42679	return __ret;
41462	42680	}
41463	42681	#else
41464		__ai float32x4_t vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41465		float32x4_t __ret;
41466		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41467		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41468		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41469		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
41470		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42682	__ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
	42683	uint8x8_t __ret;
	42684	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42685	__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
	42686	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42687	return __ret;
	42688	}
	42689	#endif
	42690
	42691	__ai uint64x1_t vceqz_p64(poly64x1_t __p0) {
	42692	uint64x1_t __ret;
	42693	__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
	42694	return __ret;
	42695	}
	42696	#ifdef __LITTLE_ENDIAN__
	42697	__ai uint8x16_t vceqzq_p8(poly8x16_t __p0) {
	42698	uint8x16_t __ret;
	42699	__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 48);
41471	42700	return __ret;
41472	42701	}
41473		__ai float32x4_t __noswap_vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41474		float32x4_t __ret;
41475		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42702	#else
	42703	__ai uint8x16_t vceqzq_p8(poly8x16_t __p0) {
	42704	uint8x16_t __ret;
	42705	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42706	__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
	42707	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
41476	42708	return __ret;
41477	42709	}
41478	42710	#endif
41479	42711
41480	42712	#ifdef __LITTLE_ENDIAN__
41481		__ai float32x2_t vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41482		float32x2_t __ret;
41483		__ret = (float32x2_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	42713	__ai uint64x2_t vceqzq_p64(poly64x2_t __p0) {
	42714	uint64x2_t __ret;
	42715	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
41484	42716	return __ret;
41485	42717	}
41486	42718	#else
41487		__ai float32x2_t vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41488		float32x2_t __ret;
41489		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41490		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41491		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41492		__ret = (float32x2_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	42719	__ai uint64x2_t vceqzq_p64(poly64x2_t __p0) {
	42720	uint64x2_t __ret;
	42721	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42722	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
41493	42723	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41494	42724	return __ret;
41495	42725	}
41496		__ai float32x2_t __noswap_vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41497		float32x2_t __ret;
41498		__ret = (float32x2_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
41499		return __ret;
41500		}
41501	42726	#endif
41502	42727
41503	42728	#ifdef __LITTLE_ENDIAN__
41504		#define vcmla_rot180_lane_f32(__p0_177, __p1_177, __p2_177, __p3_177) __extension__ ({ \
41505		float32x2_t __ret_177; \
41506		float32x2_t __s0_177 = __p0_177; \
41507		float32x2_t __s1_177 = __p1_177; \
41508		float32x2_t __s2_177 = __p2_177; \
41509		float32x2_t __reint_177 = __s2_177; \
41510		uint64x1_t __reint1_177 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_177, __p3_177)}; \
41511		__ret_177 = vcmla_rot180_f32(__s0_177, __s1_177, (float32x2_t ) &__reint1_177); \
41512		__ret_177; \
41513		})
	42729	__ai uint8x16_t vceqzq_u8(uint8x16_t __p0) {
	42730	uint8x16_t __ret;
	42731	__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 48);
	42732	return __ret;
	42733	}
41514	42734	#else
41515		#define vcmla_rot180_lane_f32(__p0_178, __p1_178, __p2_178, __p3_178) __extension__ ({ \
41516		float32x2_t __ret_178; \
41517		float32x2_t __s0_178 = __p0_178; \
41518		float32x2_t __s1_178 = __p1_178; \
41519		float32x2_t __s2_178 = __p2_178; \
41520		float32x2_t __rev0_178; __rev0_178 = __builtin_shufflevector(__s0_178, __s0_178, 1, 0); \
41521		float32x2_t __rev1_178; __rev1_178 = __builtin_shufflevector(__s1_178, __s1_178, 1, 0); \
41522		float32x2_t __rev2_178; __rev2_178 = __builtin_shufflevector(__s2_178, __s2_178, 1, 0); \
41523		float32x2_t __reint_178 = __rev2_178; \
41524		uint64x1_t __reint1_178 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_178, __p3_178)}; \
41525		__ret_178 = __noswap_vcmla_rot180_f32(__rev0_178, __rev1_178, (float32x2_t ) &__reint1_178); \
41526		__ret_178 = __builtin_shufflevector(__ret_178, __ret_178, 1, 0); \
41527		__ret_178; \
41528		})
	42735	__ai uint8x16_t vceqzq_u8(uint8x16_t __p0) {
	42736	uint8x16_t __ret;
	42737	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42738	__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
	42739	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42740	return __ret;
	42741	}
41529	42742	#endif
41530	42743
41531	42744	#ifdef __LITTLE_ENDIAN__
41532		#define vcmlaq_rot180_lane_f32(__p0_179, __p1_179, __p2_179, __p3_179) __extension__ ({ \
41533		float32x4_t __ret_179; \
41534		float32x4_t __s0_179 = __p0_179; \
41535		float32x4_t __s1_179 = __p1_179; \
41536		float32x2_t __s2_179 = __p2_179; \
41537		float32x2_t __reint_179 = __s2_179; \
41538		uint64x2_t __reint1_179 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_179, __p3_179), vget_lane_u64((uint64x1_t ) &__reint_179, __p3_179)}; \
41539		__ret_179 = vcmlaq_rot180_f32(__s0_179, __s1_179, (float32x4_t ) &__reint1_179); \
41540		__ret_179; \
41541		})
	42745	__ai uint32x4_t vceqzq_u32(uint32x4_t __p0) {
	42746	uint32x4_t __ret;
	42747	__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 50);
	42748	return __ret;
	42749	}
41542	42750	#else
41543		#define vcmlaq_rot180_lane_f32(__p0_180, __p1_180, __p2_180, __p3_180) __extension__ ({ \
41544		float32x4_t __ret_180; \
41545		float32x4_t __s0_180 = __p0_180; \
41546		float32x4_t __s1_180 = __p1_180; \
41547		float32x2_t __s2_180 = __p2_180; \
41548		float32x4_t __rev0_180; __rev0_180 = __builtin_shufflevector(__s0_180, __s0_180, 3, 2, 1, 0); \
41549		float32x4_t __rev1_180; __rev1_180 = __builtin_shufflevector(__s1_180, __s1_180, 3, 2, 1, 0); \
41550		float32x2_t __rev2_180; __rev2_180 = __builtin_shufflevector(__s2_180, __s2_180, 1, 0); \
41551		float32x2_t __reint_180 = __rev2_180; \
41552		uint64x2_t __reint1_180 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_180, __p3_180), vget_lane_u64((uint64x1_t ) &__reint_180, __p3_180)}; \
41553		__ret_180 = __noswap_vcmlaq_rot180_f32(__rev0_180, __rev1_180, (float32x4_t ) &__reint1_180); \
41554		__ret_180 = __builtin_shufflevector(__ret_180, __ret_180, 3, 2, 1, 0); \
41555		__ret_180; \
41556		})
	42751	__ai uint32x4_t vceqzq_u32(uint32x4_t __p0) {
	42752	uint32x4_t __ret;
	42753	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42754	__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
	42755	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42756	return __ret;
	42757	}
41557	42758	#endif
41558	42759
41559	42760	#ifdef __LITTLE_ENDIAN__
41560		#define vcmla_rot180_laneq_f32(__p0_181, __p1_181, __p2_181, __p3_181) __extension__ ({ \
41561		float32x2_t __ret_181; \
41562		float32x2_t __s0_181 = __p0_181; \
41563		float32x2_t __s1_181 = __p1_181; \
41564		float32x4_t __s2_181 = __p2_181; \
41565		float32x4_t __reint_181 = __s2_181; \
41566		uint64x1_t __reint1_181 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_181, __p3_181)}; \
41567		__ret_181 = vcmla_rot180_f32(__s0_181, __s1_181, (float32x2_t ) &__reint1_181); \
41568		__ret_181; \
41569		})
	42761	__ai uint64x2_t vceqzq_u64(uint64x2_t __p0) {
	42762	uint64x2_t __ret;
	42763	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
	42764	return __ret;
	42765	}
41570	42766	#else
41571		#define vcmla_rot180_laneq_f32(__p0_182, __p1_182, __p2_182, __p3_182) __extension__ ({ \
41572		float32x2_t __ret_182; \
41573		float32x2_t __s0_182 = __p0_182; \
41574		float32x2_t __s1_182 = __p1_182; \
41575		float32x4_t __s2_182 = __p2_182; \
41576		float32x2_t __rev0_182; __rev0_182 = __builtin_shufflevector(__s0_182, __s0_182, 1, 0); \
41577		float32x2_t __rev1_182; __rev1_182 = __builtin_shufflevector(__s1_182, __s1_182, 1, 0); \
41578		float32x4_t __rev2_182; __rev2_182 = __builtin_shufflevector(__s2_182, __s2_182, 3, 2, 1, 0); \
41579		float32x4_t __reint_182 = __rev2_182; \
41580		uint64x1_t __reint1_182 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_182, __p3_182)}; \
41581		__ret_182 = __noswap_vcmla_rot180_f32(__rev0_182, __rev1_182, (float32x2_t ) &__reint1_182); \
41582		__ret_182 = __builtin_shufflevector(__ret_182, __ret_182, 1, 0); \
41583		__ret_182; \
41584		})
	42767	__ai uint64x2_t vceqzq_u64(uint64x2_t __p0) {
	42768	uint64x2_t __ret;
	42769	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42770	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
	42771	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42772	return __ret;
	42773	}
41585	42774	#endif
41586	42775
41587	42776	#ifdef __LITTLE_ENDIAN__
41588		#define vcmlaq_rot180_laneq_f32(__p0_183, __p1_183, __p2_183, __p3_183) __extension__ ({ \
41589		float32x4_t __ret_183; \
41590		float32x4_t __s0_183 = __p0_183; \
41591		float32x4_t __s1_183 = __p1_183; \
41592		float32x4_t __s2_183 = __p2_183; \
41593		float32x4_t __reint_183 = __s2_183; \
41594		uint64x2_t __reint1_183 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_183, __p3_183), vgetq_lane_u64((uint64x2_t ) &__reint_183, __p3_183)}; \
41595		__ret_183 = vcmlaq_rot180_f32(__s0_183, __s1_183, (float32x4_t ) &__reint1_183); \
41596		__ret_183; \
41597		})
	42777	__ai uint16x8_t vceqzq_u16(uint16x8_t __p0) {
	42778	uint16x8_t __ret;
	42779	__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 49);
	42780	return __ret;
	42781	}
41598	42782	#else
41599		#define vcmlaq_rot180_laneq_f32(__p0_184, __p1_184, __p2_184, __p3_184) __extension__ ({ \
41600		float32x4_t __ret_184; \
41601		float32x4_t __s0_184 = __p0_184; \
41602		float32x4_t __s1_184 = __p1_184; \
41603		float32x4_t __s2_184 = __p2_184; \
41604		float32x4_t __rev0_184; __rev0_184 = __builtin_shufflevector(__s0_184, __s0_184, 3, 2, 1, 0); \
41605		float32x4_t __rev1_184; __rev1_184 = __builtin_shufflevector(__s1_184, __s1_184, 3, 2, 1, 0); \
41606		float32x4_t __rev2_184; __rev2_184 = __builtin_shufflevector(__s2_184, __s2_184, 3, 2, 1, 0); \
41607		float32x4_t __reint_184 = __rev2_184; \
41608		uint64x2_t __reint1_184 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_184, __p3_184), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_184, __p3_184)}; \
41609		__ret_184 = __noswap_vcmlaq_rot180_f32(__rev0_184, __rev1_184, (float32x4_t ) &__reint1_184); \
41610		__ret_184 = __builtin_shufflevector(__ret_184, __ret_184, 3, 2, 1, 0); \
41611		__ret_184; \
41612		})
	42783	__ai uint16x8_t vceqzq_u16(uint16x8_t __p0) {
	42784	uint16x8_t __ret;
	42785	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42786	__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
	42787	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42788	return __ret;
	42789	}
41613	42790	#endif
41614	42791
41615	42792	#ifdef __LITTLE_ENDIAN__
41616		__ai float32x4_t vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41617		float32x4_t __ret;
41618		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42793	__ai uint8x16_t vceqzq_s8(int8x16_t __p0) {
	42794	uint8x16_t __ret;
	42795	__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 48);
41619	42796	return __ret;
41620	42797	}
41621	42798	#else
41622		__ai float32x4_t vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41623		float32x4_t __ret;
41624		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41625		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41626		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41627		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
41628		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
41629		return __ret;
41630		}
41631		__ai float32x4_t __noswap_vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41632		float32x4_t __ret;
41633		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42799	__ai uint8x16_t vceqzq_s8(int8x16_t __p0) {
	42800	uint8x16_t __ret;
	42801	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	42802	__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
	42803	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
41634	42804	return __ret;
41635	42805	}
41636	42806	#endif
41637	42807
41638	42808	#ifdef __LITTLE_ENDIAN__
41639		__ai float32x2_t vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41640		float32x2_t __ret;
41641		__ret = (float32x2_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	42809	__ai uint64x2_t vceqzq_f64(float64x2_t __p0) {
	42810	uint64x2_t __ret;
	42811	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
41642	42812	return __ret;
41643	42813	}
41644	42814	#else
41645		__ai float32x2_t vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41646		float32x2_t __ret;
41647		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41648		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41649		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41650		__ret = (float32x2_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	42815	__ai uint64x2_t vceqzq_f64(float64x2_t __p0) {
	42816	uint64x2_t __ret;
	42817	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42818	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
41651	42819	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41652	42820	return __ret;
41653	42821	}
41654		__ai float32x2_t __noswap_vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41655		float32x2_t __ret;
41656		__ret = (float32x2_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
41657		return __ret;
41658		}
41659	42822	#endif
41660	42823
41661	42824	#ifdef __LITTLE_ENDIAN__
41662		#define vcmla_rot270_lane_f32(__p0_185, __p1_185, __p2_185, __p3_185) __extension__ ({ \
41663		float32x2_t __ret_185; \
41664		float32x2_t __s0_185 = __p0_185; \
41665		float32x2_t __s1_185 = __p1_185; \
41666		float32x2_t __s2_185 = __p2_185; \
41667		float32x2_t __reint_185 = __s2_185; \
41668		uint64x1_t __reint1_185 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_185, __p3_185)}; \
41669		__ret_185 = vcmla_rot270_f32(__s0_185, __s1_185, (float32x2_t ) &__reint1_185); \
41670		__ret_185; \
41671		})
	42825	__ai uint32x4_t vceqzq_f32(float32x4_t __p0) {
	42826	uint32x4_t __ret;
	42827	__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 50);
	42828	return __ret;
	42829	}
41672	42830	#else
41673		#define vcmla_rot270_lane_f32(__p0_186, __p1_186, __p2_186, __p3_186) __extension__ ({ \
41674		float32x2_t __ret_186; \
41675		float32x2_t __s0_186 = __p0_186; \
41676		float32x2_t __s1_186 = __p1_186; \
41677		float32x2_t __s2_186 = __p2_186; \
41678		float32x2_t __rev0_186; __rev0_186 = __builtin_shufflevector(__s0_186, __s0_186, 1, 0); \
41679		float32x2_t __rev1_186; __rev1_186 = __builtin_shufflevector(__s1_186, __s1_186, 1, 0); \
41680		float32x2_t __rev2_186; __rev2_186 = __builtin_shufflevector(__s2_186, __s2_186, 1, 0); \
41681		float32x2_t __reint_186 = __rev2_186; \
41682		uint64x1_t __reint1_186 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_186, __p3_186)}; \
41683		__ret_186 = __noswap_vcmla_rot270_f32(__rev0_186, __rev1_186, (float32x2_t ) &__reint1_186); \
41684		__ret_186 = __builtin_shufflevector(__ret_186, __ret_186, 1, 0); \
41685		__ret_186; \
41686		})
	42831	__ai uint32x4_t vceqzq_f32(float32x4_t __p0) {
	42832	uint32x4_t __ret;
	42833	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42834	__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
	42835	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42836	return __ret;
	42837	}
41687	42838	#endif
41688	42839
41689	42840	#ifdef __LITTLE_ENDIAN__
41690		#define vcmlaq_rot270_lane_f32(__p0_187, __p1_187, __p2_187, __p3_187) __extension__ ({ \
41691		float32x4_t __ret_187; \
41692		float32x4_t __s0_187 = __p0_187; \
41693		float32x4_t __s1_187 = __p1_187; \
41694		float32x2_t __s2_187 = __p2_187; \
41695		float32x2_t __reint_187 = __s2_187; \
41696		uint64x2_t __reint1_187 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_187, __p3_187), vget_lane_u64((uint64x1_t ) &__reint_187, __p3_187)}; \
41697		__ret_187 = vcmlaq_rot270_f32(__s0_187, __s1_187, (float32x4_t ) &__reint1_187); \
41698		__ret_187; \
41699		})
	42841	__ai uint32x4_t vceqzq_s32(int32x4_t __p0) {
	42842	uint32x4_t __ret;
	42843	__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 50);
	42844	return __ret;
	42845	}
41700	42846	#else
41701		#define vcmlaq_rot270_lane_f32(__p0_188, __p1_188, __p2_188, __p3_188) __extension__ ({ \
41702		float32x4_t __ret_188; \
41703		float32x4_t __s0_188 = __p0_188; \
41704		float32x4_t __s1_188 = __p1_188; \
41705		float32x2_t __s2_188 = __p2_188; \
41706		float32x4_t __rev0_188; __rev0_188 = __builtin_shufflevector(__s0_188, __s0_188, 3, 2, 1, 0); \
41707		float32x4_t __rev1_188; __rev1_188 = __builtin_shufflevector(__s1_188, __s1_188, 3, 2, 1, 0); \
41708		float32x2_t __rev2_188; __rev2_188 = __builtin_shufflevector(__s2_188, __s2_188, 1, 0); \
41709		float32x2_t __reint_188 = __rev2_188; \
41710		uint64x2_t __reint1_188 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_188, __p3_188), vget_lane_u64((uint64x1_t ) &__reint_188, __p3_188)}; \
41711		__ret_188 = __noswap_vcmlaq_rot270_f32(__rev0_188, __rev1_188, (float32x4_t ) &__reint1_188); \
41712		__ret_188 = __builtin_shufflevector(__ret_188, __ret_188, 3, 2, 1, 0); \
41713		__ret_188; \
41714		})
	42847	__ai uint32x4_t vceqzq_s32(int32x4_t __p0) {
	42848	uint32x4_t __ret;
	42849	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42850	__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
	42851	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42852	return __ret;
	42853	}
41715	42854	#endif
41716	42855
41717	42856	#ifdef __LITTLE_ENDIAN__
41718		#define vcmla_rot270_laneq_f32(__p0_189, __p1_189, __p2_189, __p3_189) __extension__ ({ \
41719		float32x2_t __ret_189; \
41720		float32x2_t __s0_189 = __p0_189; \
41721		float32x2_t __s1_189 = __p1_189; \
41722		float32x4_t __s2_189 = __p2_189; \
41723		float32x4_t __reint_189 = __s2_189; \
41724		uint64x1_t __reint1_189 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_189, __p3_189)}; \
41725		__ret_189 = vcmla_rot270_f32(__s0_189, __s1_189, (float32x2_t ) &__reint1_189); \
41726		__ret_189; \
41727		})
	42857	__ai uint64x2_t vceqzq_s64(int64x2_t __p0) {
	42858	uint64x2_t __ret;
	42859	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
	42860	return __ret;
	42861	}
41728	42862	#else
41729		#define vcmla_rot270_laneq_f32(__p0_190, __p1_190, __p2_190, __p3_190) __extension__ ({ \
41730		float32x2_t __ret_190; \
41731		float32x2_t __s0_190 = __p0_190; \
41732		float32x2_t __s1_190 = __p1_190; \
41733		float32x4_t __s2_190 = __p2_190; \
41734		float32x2_t __rev0_190; __rev0_190 = __builtin_shufflevector(__s0_190, __s0_190, 1, 0); \
41735		float32x2_t __rev1_190; __rev1_190 = __builtin_shufflevector(__s1_190, __s1_190, 1, 0); \
41736		float32x4_t __rev2_190; __rev2_190 = __builtin_shufflevector(__s2_190, __s2_190, 3, 2, 1, 0); \
41737		float32x4_t __reint_190 = __rev2_190; \
41738		uint64x1_t __reint1_190 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_190, __p3_190)}; \
41739		__ret_190 = __noswap_vcmla_rot270_f32(__rev0_190, __rev1_190, (float32x2_t ) &__reint1_190); \
41740		__ret_190 = __builtin_shufflevector(__ret_190, __ret_190, 1, 0); \
41741		__ret_190; \
41742		})
	42863	__ai uint64x2_t vceqzq_s64(int64x2_t __p0) {
	42864	uint64x2_t __ret;
	42865	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42866	__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
	42867	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42868	return __ret;
	42869	}
41743	42870	#endif
41744	42871
41745	42872	#ifdef __LITTLE_ENDIAN__
41746		#define vcmlaq_rot270_laneq_f32(__p0_191, __p1_191, __p2_191, __p3_191) __extension__ ({ \
41747		float32x4_t __ret_191; \
41748		float32x4_t __s0_191 = __p0_191; \
41749		float32x4_t __s1_191 = __p1_191; \
41750		float32x4_t __s2_191 = __p2_191; \
41751		float32x4_t __reint_191 = __s2_191; \
41752		uint64x2_t __reint1_191 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_191, __p3_191), vgetq_lane_u64((uint64x2_t ) &__reint_191, __p3_191)}; \
41753		__ret_191 = vcmlaq_rot270_f32(__s0_191, __s1_191, (float32x4_t ) &__reint1_191); \
41754		__ret_191; \
41755		})
	42873	__ai uint16x8_t vceqzq_s16(int16x8_t __p0) {
	42874	uint16x8_t __ret;
	42875	__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 49);
	42876	return __ret;
	42877	}
41756	42878	#else
41757		#define vcmlaq_rot270_laneq_f32(__p0_192, __p1_192, __p2_192, __p3_192) __extension__ ({ \
41758		float32x4_t __ret_192; \
41759		float32x4_t __s0_192 = __p0_192; \
41760		float32x4_t __s1_192 = __p1_192; \
41761		float32x4_t __s2_192 = __p2_192; \
41762		float32x4_t __rev0_192; __rev0_192 = __builtin_shufflevector(__s0_192, __s0_192, 3, 2, 1, 0); \
41763		float32x4_t __rev1_192; __rev1_192 = __builtin_shufflevector(__s1_192, __s1_192, 3, 2, 1, 0); \
41764		float32x4_t __rev2_192; __rev2_192 = __builtin_shufflevector(__s2_192, __s2_192, 3, 2, 1, 0); \
41765		float32x4_t __reint_192 = __rev2_192; \
41766		uint64x2_t __reint1_192 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_192, __p3_192), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_192, __p3_192)}; \
41767		__ret_192 = __noswap_vcmlaq_rot270_f32(__rev0_192, __rev1_192, (float32x4_t ) &__reint1_192); \
41768		__ret_192 = __builtin_shufflevector(__ret_192, __ret_192, 3, 2, 1, 0); \
41769		__ret_192; \
41770		})
	42879	__ai uint16x8_t vceqzq_s16(int16x8_t __p0) {
	42880	uint16x8_t __ret;
	42881	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42882	__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
	42883	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42884	return __ret;
	42885	}
41771	42886	#endif
41772	42887
41773	42888	#ifdef __LITTLE_ENDIAN__
41774		__ai float32x4_t vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41775		float32x4_t __ret;
41776		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42889	__ai uint8x8_t vceqz_u8(uint8x8_t __p0) {
	42890	uint8x8_t __ret;
	42891	__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 16);
41777	42892	return __ret;
41778	42893	}
41779	42894	#else
41780		__ai float32x4_t vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41781		float32x4_t __ret;
41782		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41783		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41784		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41785		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
41786		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
41787		return __ret;
41788		}
41789		__ai float32x4_t __noswap_vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41790		float32x4_t __ret;
41791		__ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	42895	__ai uint8x8_t vceqz_u8(uint8x8_t __p0) {
	42896	uint8x8_t __ret;
	42897	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42898	__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
	42899	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
41792	42900	return __ret;
41793	42901	}
41794	42902	#endif
41795	42903
41796	42904	#ifdef __LITTLE_ENDIAN__
41797		__ai float32x2_t vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41798		float32x2_t __ret;
41799		__ret = (float32x2_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	42905	__ai uint32x2_t vceqz_u32(uint32x2_t __p0) {
	42906	uint32x2_t __ret;
	42907	__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 18);
41800	42908	return __ret;
41801	42909	}
41802	42910	#else
41803		__ai float32x2_t vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41804		float32x2_t __ret;
41805		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41806		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41807		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41808		__ret = (float32x2_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	42911	__ai uint32x2_t vceqz_u32(uint32x2_t __p0) {
	42912	uint32x2_t __ret;
	42913	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42914	__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
41809	42915	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41810	42916	return __ret;
41811	42917	}
41812		__ai float32x2_t __noswap_vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41813		float32x2_t __ret;
41814		__ret = (float32x2_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
41815		return __ret;
41816		}
41817	42918	#endif
41818	42919
	42920	__ai uint64x1_t vceqz_u64(uint64x1_t __p0) {
	42921	uint64x1_t __ret;
	42922	__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
	42923	return __ret;
	42924	}
41819	42925	#ifdef __LITTLE_ENDIAN__
41820		#define vcmla_rot90_lane_f32(__p0_193, __p1_193, __p2_193, __p3_193) __extension__ ({ \
41821		float32x2_t __ret_193; \
41822		float32x2_t __s0_193 = __p0_193; \
41823		float32x2_t __s1_193 = __p1_193; \
41824		float32x2_t __s2_193 = __p2_193; \
41825		float32x2_t __reint_193 = __s2_193; \
41826		uint64x1_t __reint1_193 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_193, __p3_193)}; \
41827		__ret_193 = vcmla_rot90_f32(__s0_193, __s1_193, (float32x2_t ) &__reint1_193); \
41828		__ret_193; \
41829		})
	42926	__ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
	42927	uint16x4_t __ret;
	42928	__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 17);
	42929	return __ret;
	42930	}
41830	42931	#else
41831		#define vcmla_rot90_lane_f32(__p0_194, __p1_194, __p2_194, __p3_194) __extension__ ({ \
41832		float32x2_t __ret_194; \
41833		float32x2_t __s0_194 = __p0_194; \
41834		float32x2_t __s1_194 = __p1_194; \
41835		float32x2_t __s2_194 = __p2_194; \
41836		float32x2_t __rev0_194; __rev0_194 = __builtin_shufflevector(__s0_194, __s0_194, 1, 0); \
41837		float32x2_t __rev1_194; __rev1_194 = __builtin_shufflevector(__s1_194, __s1_194, 1, 0); \
41838		float32x2_t __rev2_194; __rev2_194 = __builtin_shufflevector(__s2_194, __s2_194, 1, 0); \
41839		float32x2_t __reint_194 = __rev2_194; \
41840		uint64x1_t __reint1_194 = (uint64x1_t) {vget_lane_u64((uint64x1_t ) &__reint_194, __p3_194)}; \
41841		__ret_194 = __noswap_vcmla_rot90_f32(__rev0_194, __rev1_194, (float32x2_t ) &__reint1_194); \
41842		__ret_194 = __builtin_shufflevector(__ret_194, __ret_194, 1, 0); \
41843		__ret_194; \
41844		})
	42932	__ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
	42933	uint16x4_t __ret;
	42934	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	42935	__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
	42936	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	42937	return __ret;
	42938	}
41845	42939	#endif
41846	42940
41847	42941	#ifdef __LITTLE_ENDIAN__
41848		#define vcmlaq_rot90_lane_f32(__p0_195, __p1_195, __p2_195, __p3_195) __extension__ ({ \
41849		float32x4_t __ret_195; \
41850		float32x4_t __s0_195 = __p0_195; \
41851		float32x4_t __s1_195 = __p1_195; \
41852		float32x2_t __s2_195 = __p2_195; \
41853		float32x2_t __reint_195 = __s2_195; \
41854		uint64x2_t __reint1_195 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_195, __p3_195), vget_lane_u64((uint64x1_t ) &__reint_195, __p3_195)}; \
41855		__ret_195 = vcmlaq_rot90_f32(__s0_195, __s1_195, (float32x4_t ) &__reint1_195); \
41856		__ret_195; \
41857		})
	42942	__ai uint8x8_t vceqz_s8(int8x8_t __p0) {
	42943	uint8x8_t __ret;
	42944	__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 16);
	42945	return __ret;
	42946	}
41858	42947	#else
41859		#define vcmlaq_rot90_lane_f32(__p0_196, __p1_196, __p2_196, __p3_196) __extension__ ({ \
41860		float32x4_t __ret_196; \
41861		float32x4_t __s0_196 = __p0_196; \
41862		float32x4_t __s1_196 = __p1_196; \
41863		float32x2_t __s2_196 = __p2_196; \
41864		float32x4_t __rev0_196; __rev0_196 = __builtin_shufflevector(__s0_196, __s0_196, 3, 2, 1, 0); \
41865		float32x4_t __rev1_196; __rev1_196 = __builtin_shufflevector(__s1_196, __s1_196, 3, 2, 1, 0); \
41866		float32x2_t __rev2_196; __rev2_196 = __builtin_shufflevector(__s2_196, __s2_196, 1, 0); \
41867		float32x2_t __reint_196 = __rev2_196; \
41868		uint64x2_t __reint1_196 = (uint64x2_t) {vget_lane_u64((uint64x1_t ) &__reint_196, __p3_196), vget_lane_u64((uint64x1_t ) &__reint_196, __p3_196)}; \
41869		__ret_196 = __noswap_vcmlaq_rot90_f32(__rev0_196, __rev1_196, (float32x4_t ) &__reint1_196); \
41870		__ret_196 = __builtin_shufflevector(__ret_196, __ret_196, 3, 2, 1, 0); \
41871		__ret_196; \
41872		})
	42948	__ai uint8x8_t vceqz_s8(int8x8_t __p0) {
	42949	uint8x8_t __ret;
	42950	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	42951	__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
	42952	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	42953	return __ret;
	42954	}
41873	42955	#endif
41874	42956
	42957	__ai uint64x1_t vceqz_f64(float64x1_t __p0) {
	42958	uint64x1_t __ret;
	42959	__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
	42960	return __ret;
	42961	}
41875	42962	#ifdef __LITTLE_ENDIAN__
41876		#define vcmla_rot90_laneq_f32(__p0_197, __p1_197, __p2_197, __p3_197) __extension__ ({ \
41877		float32x2_t __ret_197; \
41878		float32x2_t __s0_197 = __p0_197; \
41879		float32x2_t __s1_197 = __p1_197; \
41880		float32x4_t __s2_197 = __p2_197; \
41881		float32x4_t __reint_197 = __s2_197; \
41882		uint64x1_t __reint1_197 = (uint64x1_t) {vgetq_lane_u64((uint64x2_t ) &__reint_197, __p3_197)}; \
41883		__ret_197 = vcmla_rot90_f32(__s0_197, __s1_197, (float32x2_t ) &__reint1_197); \
41884		__ret_197; \
41885		})
	42963	__ai uint32x2_t vceqz_f32(float32x2_t __p0) {
	42964	uint32x2_t __ret;
	42965	__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 18);
	42966	return __ret;
	42967	}
41886	42968	#else
41887		#define vcmla_rot90_laneq_f32(__p0_198, __p1_198, __p2_198, __p3_198) __extension__ ({ \
41888		float32x2_t __ret_198; \
41889		float32x2_t __s0_198 = __p0_198; \
41890		float32x2_t __s1_198 = __p1_198; \
41891		float32x4_t __s2_198 = __p2_198; \
41892		float32x2_t __rev0_198; __rev0_198 = __builtin_shufflevector(__s0_198, __s0_198, 1, 0); \
41893		float32x2_t __rev1_198; __rev1_198 = __builtin_shufflevector(__s1_198, __s1_198, 1, 0); \
41894		float32x4_t __rev2_198; __rev2_198 = __builtin_shufflevector(__s2_198, __s2_198, 3, 2, 1, 0); \
41895		float32x4_t __reint_198 = __rev2_198; \
41896		uint64x1_t __reint1_198 = (uint64x1_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_198, __p3_198)}; \
41897		__ret_198 = __noswap_vcmla_rot90_f32(__rev0_198, __rev1_198, (float32x2_t ) &__reint1_198); \
41898		__ret_198 = __builtin_shufflevector(__ret_198, __ret_198, 1, 0); \
41899		__ret_198; \
41900		})
	42969	__ai uint32x2_t vceqz_f32(float32x2_t __p0) {
	42970	uint32x2_t __ret;
	42971	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42972	__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
	42973	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42974	return __ret;
	42975	}
41901	42976	#endif
41902	42977
41903	42978	#ifdef __LITTLE_ENDIAN__
41904		#define vcmlaq_rot90_laneq_f32(__p0_199, __p1_199, __p2_199, __p3_199) __extension__ ({ \
41905		float32x4_t __ret_199; \
41906		float32x4_t __s0_199 = __p0_199; \
41907		float32x4_t __s1_199 = __p1_199; \
41908		float32x4_t __s2_199 = __p2_199; \
41909		float32x4_t __reint_199 = __s2_199; \
41910		uint64x2_t __reint1_199 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_199, __p3_199), vgetq_lane_u64((uint64x2_t ) &__reint_199, __p3_199)}; \
41911		__ret_199 = vcmlaq_rot90_f32(__s0_199, __s1_199, (float32x4_t ) &__reint1_199); \
41912		__ret_199; \
41913		})
	42979	__ai uint32x2_t vceqz_s32(int32x2_t __p0) {
	42980	uint32x2_t __ret;
	42981	__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 18);
	42982	return __ret;
	42983	}
41914	42984	#else
41915		#define vcmlaq_rot90_laneq_f32(__p0_200, __p1_200, __p2_200, __p3_200) __extension__ ({ \
41916		float32x4_t __ret_200; \
41917		float32x4_t __s0_200 = __p0_200; \
41918		float32x4_t __s1_200 = __p1_200; \
41919		float32x4_t __s2_200 = __p2_200; \
41920		float32x4_t __rev0_200; __rev0_200 = __builtin_shufflevector(__s0_200, __s0_200, 3, 2, 1, 0); \
41921		float32x4_t __rev1_200; __rev1_200 = __builtin_shufflevector(__s1_200, __s1_200, 3, 2, 1, 0); \
41922		float32x4_t __rev2_200; __rev2_200 = __builtin_shufflevector(__s2_200, __s2_200, 3, 2, 1, 0); \
41923		float32x4_t __reint_200 = __rev2_200; \
41924		uint64x2_t __reint1_200 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_200, __p3_200), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_200, __p3_200)}; \
41925		__ret_200 = __noswap_vcmlaq_rot90_f32(__rev0_200, __rev1_200, (float32x4_t ) &__reint1_200); \
41926		__ret_200 = __builtin_shufflevector(__ret_200, __ret_200, 3, 2, 1, 0); \
41927		__ret_200; \
41928		})
	42985	__ai uint32x2_t vceqz_s32(int32x2_t __p0) {
	42986	uint32x2_t __ret;
	42987	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	42988	__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
	42989	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	42990	return __ret;
	42991	}
41929	42992	#endif
41930	42993
41931		#endif
41932		#if defined(__ARM_FEATURE_COMPLEX) && defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
	42994	__ai uint64x1_t vceqz_s64(int64x1_t __p0) {
	42995	uint64x1_t __ret;
	42996	__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
	42997	return __ret;
	42998	}
41933	42999	#ifdef __LITTLE_ENDIAN__
41934		__ai float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
41935		float16x4_t __ret;
41936		__ret = (float16x4_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	43000	__ai uint16x4_t vceqz_s16(int16x4_t __p0) {
	43001	uint16x4_t __ret;
	43002	__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 17);
41937	43003	return __ret;
41938	43004	}
41939	43005	#else
41940		__ai float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
41941		float16x4_t __ret;
41942		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41943		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41944		__ret = (float16x4_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	43006	__ai uint16x4_t vceqz_s16(int16x4_t __p0) {
	43007	uint16x4_t __ret;
	43008	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43009	__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
41945	43010	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
41946	43011	return __ret;
41947	43012	}
41948		#endif
41949
41950		#ifdef __LITTLE_ENDIAN__
41951		__ai float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
41952		float16x4_t __ret;
41953		__ret = (float16x4_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	43013	#endif
	43014
	43015	__ai uint64_t vceqzd_u64(uint64_t __p0) {
	43016	uint64_t __ret;
	43017	__ret = (uint64_t) __builtin_neon_vceqzd_u64(__p0);
	43018	return __ret;
	43019	}
	43020	__ai uint64_t vceqzd_s64(int64_t __p0) {
	43021	uint64_t __ret;
	43022	__ret = (uint64_t) __builtin_neon_vceqzd_s64(__p0);
	43023	return __ret;
	43024	}
	43025	__ai uint64_t vceqzd_f64(float64_t __p0) {
	43026	uint64_t __ret;
	43027	__ret = (uint64_t) __builtin_neon_vceqzd_f64(__p0);
41954	43028	return __ret;
41955	43029	}
41956		#else
41957		__ai float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
41958		float16x4_t __ret;
41959		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41960		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41961		__ret = (float16x4_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
41962		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43030	__ai uint32_t vceqzs_f32(float32_t __p0) {
	43031	uint32_t __ret;
	43032	__ret = (uint32_t) __builtin_neon_vceqzs_f32(__p0);
41963	43033	return __ret;
41964	43034	}
41965		#endif
41966
41967	43035	#ifdef __LITTLE_ENDIAN__
41968		__ai float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
41969		float16x8_t __ret;
41970		__ret = (float16x8_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	43036	__ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43037	uint64x2_t __ret;
	43038	__ret = (uint64x2_t)(__p0 >= __p1);
41971	43039	return __ret;
41972	43040	}
41973	43041	#else
41974		__ai float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
41975		float16x8_t __ret;
41976		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
41977		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
41978		__ret = (float16x8_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
41979		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	43042	__ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43043	uint64x2_t __ret;
	43044	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43045	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43046	__ret = (uint64x2_t)(__rev0 >= __rev1);
	43047	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41980	43048	return __ret;
41981	43049	}
41982	43050	#endif
41983	43051
41984	43052	#ifdef __LITTLE_ENDIAN__
41985		__ai float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
41986		float16x8_t __ret;
41987		__ret = (float16x8_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	43053	__ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
	43054	uint64x2_t __ret;
	43055	__ret = (uint64x2_t)(__p0 >= __p1);
41988	43056	return __ret;
41989	43057	}
41990	43058	#else
41991		__ai float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
41992		float16x8_t __ret;
41993		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
41994		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
41995		__ret = (float16x8_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
41996		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	43059	__ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
	43060	uint64x2_t __ret;
	43061	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43062	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43063	__ret = (uint64x2_t)(__rev0 >= __rev1);
	43064	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41997	43065	return __ret;
41998	43066	}
41999	43067	#endif
42000	43068
42001	43069	#ifdef __LITTLE_ENDIAN__
42002		__ai float16x8_t vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42003		float16x8_t __ret;
42004		__ret = (float16x8_t) __builtin_neon_vcmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43070	__ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
	43071	uint64x2_t __ret;
	43072	__ret = (uint64x2_t)(__p0 >= __p1);
42005	43073	return __ret;
42006	43074	}
42007	43075	#else
42008		__ai float16x8_t vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42009		float16x8_t __ret;
42010		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42011		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
42012		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42013		__ret = (float16x8_t) __builtin_neon_vcmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
42014		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
42015		return __ret;
42016		}
42017		__ai float16x8_t __noswap_vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42018		float16x8_t __ret;
42019		__ret = (float16x8_t) __builtin_neon_vcmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43076	__ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
	43077	uint64x2_t __ret;
	43078	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43079	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43080	__ret = (uint64x2_t)(__rev0 >= __rev1);
	43081	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
42020	43082	return __ret;
42021	43083	}
42022	43084	#endif
42023	43085
42024		#ifdef __LITTLE_ENDIAN__
42025		__ai float16x4_t vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42026		float16x4_t __ret;
42027		__ret = (float16x4_t) __builtin_neon_vcmla_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43086	__ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
	43087	uint64x1_t __ret;
	43088	__ret = (uint64x1_t)(__p0 >= __p1);
42028	43089	return __ret;
42029	43090	}
42030		#else
42031		__ai float16x4_t vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42032		float16x4_t __ret;
42033		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
42034		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
42035		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42036		__ret = (float16x4_t) __builtin_neon_vcmla_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
42037		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43091	__ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
	43092	uint64x1_t __ret;
	43093	__ret = (uint64x1_t)(__p0 >= __p1);
42038	43094	return __ret;
42039	43095	}
42040		__ai float16x4_t __noswap_vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42041		float16x4_t __ret;
42042		__ret = (float16x4_t) __builtin_neon_vcmla_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43096	__ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
	43097	uint64x1_t __ret;
	43098	__ret = (uint64x1_t)(__p0 >= __p1);
42043	43099	return __ret;
42044	43100	}
42045		#endif
42046
42047		#ifdef __LITTLE_ENDIAN__
42048		#define vcmla_lane_f16(__p0_201, __p1_201, __p2_201, __p3_201) __extension__ ({ \
42049		float16x4_t __ret_201; \
42050		float16x4_t __s0_201 = __p0_201; \
42051		float16x4_t __s1_201 = __p1_201; \
42052		float16x4_t __s2_201 = __p2_201; \
42053		float16x4_t __reint_201 = __s2_201; \
42054		uint32x2_t __reint1_201 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_201, __p3_201), vget_lane_u32((uint32x2_t ) &__reint_201, __p3_201)}; \
42055		__ret_201 = vcmla_f16(__s0_201, __s1_201, (float16x4_t ) &__reint1_201); \
42056		__ret_201; \
42057		})
42058		#else
42059		#define vcmla_lane_f16(__p0_202, __p1_202, __p2_202, __p3_202) __extension__ ({ \
42060		float16x4_t __ret_202; \
42061		float16x4_t __s0_202 = __p0_202; \
42062		float16x4_t __s1_202 = __p1_202; \
42063		float16x4_t __s2_202 = __p2_202; \
42064		float16x4_t __rev0_202; __rev0_202 = __builtin_shufflevector(__s0_202, __s0_202, 3, 2, 1, 0); \
42065		float16x4_t __rev1_202; __rev1_202 = __builtin_shufflevector(__s1_202, __s1_202, 3, 2, 1, 0); \
42066		float16x4_t __rev2_202; __rev2_202 = __builtin_shufflevector(__s2_202, __s2_202, 3, 2, 1, 0); \
42067		float16x4_t __reint_202 = __rev2_202; \
42068		uint32x2_t __reint1_202 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_202, __p3_202), __noswap_vget_lane_u32((uint32x2_t ) &__reint_202, __p3_202)}; \
42069		__ret_202 = __noswap_vcmla_f16(__rev0_202, __rev1_202, (float16x4_t ) &__reint1_202); \
42070		__ret_202 = __builtin_shufflevector(__ret_202, __ret_202, 3, 2, 1, 0); \
42071		__ret_202; \
42072		})
42073		#endif
42074
42075		#ifdef __LITTLE_ENDIAN__
42076		#define vcmlaq_lane_f16(__p0_203, __p1_203, __p2_203, __p3_203) __extension__ ({ \
42077		float16x8_t __ret_203; \
42078		float16x8_t __s0_203 = __p0_203; \
42079		float16x8_t __s1_203 = __p1_203; \
42080		float16x4_t __s2_203 = __p2_203; \
42081		float16x4_t __reint_203 = __s2_203; \
42082		uint32x4_t __reint1_203 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_203, __p3_203), vget_lane_u32((uint32x2_t ) &__reint_203, __p3_203), vget_lane_u32((uint32x2_t ) &__reint_203, __p3_203), vget_lane_u32((uint32x2_t ) &__reint_203, __p3_203)}; \
42083		__ret_203 = vcmlaq_f16(__s0_203, __s1_203, (float16x8_t ) &__reint1_203); \
42084		__ret_203; \
42085		})
42086		#else
42087		#define vcmlaq_lane_f16(__p0_204, __p1_204, __p2_204, __p3_204) __extension__ ({ \
42088		float16x8_t __ret_204; \
42089		float16x8_t __s0_204 = __p0_204; \
42090		float16x8_t __s1_204 = __p1_204; \
42091		float16x4_t __s2_204 = __p2_204; \
42092		float16x8_t __rev0_204; __rev0_204 = __builtin_shufflevector(__s0_204, __s0_204, 7, 6, 5, 4, 3, 2, 1, 0); \
42093		float16x8_t __rev1_204; __rev1_204 = __builtin_shufflevector(__s1_204, __s1_204, 7, 6, 5, 4, 3, 2, 1, 0); \
42094		float16x4_t __rev2_204; __rev2_204 = __builtin_shufflevector(__s2_204, __s2_204, 3, 2, 1, 0); \
42095		float16x4_t __reint_204 = __rev2_204; \
42096		uint32x4_t __reint1_204 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_204, __p3_204), __noswap_vget_lane_u32((uint32x2_t ) &__reint_204, __p3_204), __noswap_vget_lane_u32((uint32x2_t ) &__reint_204, __p3_204), __noswap_vget_lane_u32((uint32x2_t ) &__reint_204, __p3_204)}; \
42097		__ret_204 = __noswap_vcmlaq_f16(__rev0_204, __rev1_204, (float16x8_t ) &__reint1_204); \
42098		__ret_204 = __builtin_shufflevector(__ret_204, __ret_204, 7, 6, 5, 4, 3, 2, 1, 0); \
42099		__ret_204; \
42100		})
42101		#endif
42102
42103		#ifdef __LITTLE_ENDIAN__
42104		#define vcmla_laneq_f16(__p0_205, __p1_205, __p2_205, __p3_205) __extension__ ({ \
42105		float16x4_t __ret_205; \
42106		float16x4_t __s0_205 = __p0_205; \
42107		float16x4_t __s1_205 = __p1_205; \
42108		float16x8_t __s2_205 = __p2_205; \
42109		float16x8_t __reint_205 = __s2_205; \
42110		uint32x2_t __reint1_205 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_205, __p3_205), vgetq_lane_u32((uint32x4_t ) &__reint_205, __p3_205)}; \
42111		__ret_205 = vcmla_f16(__s0_205, __s1_205, (float16x4_t ) &__reint1_205); \
42112		__ret_205; \
42113		})
42114		#else
42115		#define vcmla_laneq_f16(__p0_206, __p1_206, __p2_206, __p3_206) __extension__ ({ \
42116		float16x4_t __ret_206; \
42117		float16x4_t __s0_206 = __p0_206; \
42118		float16x4_t __s1_206 = __p1_206; \
42119		float16x8_t __s2_206 = __p2_206; \
42120		float16x4_t __rev0_206; __rev0_206 = __builtin_shufflevector(__s0_206, __s0_206, 3, 2, 1, 0); \
42121		float16x4_t __rev1_206; __rev1_206 = __builtin_shufflevector(__s1_206, __s1_206, 3, 2, 1, 0); \
42122		float16x8_t __rev2_206; __rev2_206 = __builtin_shufflevector(__s2_206, __s2_206, 7, 6, 5, 4, 3, 2, 1, 0); \
42123		float16x8_t __reint_206 = __rev2_206; \
42124		uint32x2_t __reint1_206 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_206, __p3_206), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_206, __p3_206)}; \
42125		__ret_206 = __noswap_vcmla_f16(__rev0_206, __rev1_206, (float16x4_t ) &__reint1_206); \
42126		__ret_206 = __builtin_shufflevector(__ret_206, __ret_206, 3, 2, 1, 0); \
42127		__ret_206; \
42128		})
42129		#endif
42130
42131		#ifdef __LITTLE_ENDIAN__
42132		#define vcmlaq_laneq_f16(__p0_207, __p1_207, __p2_207, __p3_207) __extension__ ({ \
42133		float16x8_t __ret_207; \
42134		float16x8_t __s0_207 = __p0_207; \
42135		float16x8_t __s1_207 = __p1_207; \
42136		float16x8_t __s2_207 = __p2_207; \
42137		float16x8_t __reint_207 = __s2_207; \
42138		uint32x4_t __reint1_207 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_207, __p3_207), vgetq_lane_u32((uint32x4_t ) &__reint_207, __p3_207), vgetq_lane_u32((uint32x4_t ) &__reint_207, __p3_207), vgetq_lane_u32((uint32x4_t ) &__reint_207, __p3_207)}; \
42139		__ret_207 = vcmlaq_f16(__s0_207, __s1_207, (float16x8_t ) &__reint1_207); \
42140		__ret_207; \
42141		})
42142		#else
42143		#define vcmlaq_laneq_f16(__p0_208, __p1_208, __p2_208, __p3_208) __extension__ ({ \
42144		float16x8_t __ret_208; \
42145		float16x8_t __s0_208 = __p0_208; \
42146		float16x8_t __s1_208 = __p1_208; \
42147		float16x8_t __s2_208 = __p2_208; \
42148		float16x8_t __rev0_208; __rev0_208 = __builtin_shufflevector(__s0_208, __s0_208, 7, 6, 5, 4, 3, 2, 1, 0); \
42149		float16x8_t __rev1_208; __rev1_208 = __builtin_shufflevector(__s1_208, __s1_208, 7, 6, 5, 4, 3, 2, 1, 0); \
42150		float16x8_t __rev2_208; __rev2_208 = __builtin_shufflevector(__s2_208, __s2_208, 7, 6, 5, 4, 3, 2, 1, 0); \
42151		float16x8_t __reint_208 = __rev2_208; \
42152		uint32x4_t __reint1_208 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_208, __p3_208), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_208, __p3_208), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_208, __p3_208), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_208, __p3_208)}; \
42153		__ret_208 = __noswap_vcmlaq_f16(__rev0_208, __rev1_208, (float16x8_t ) &__reint1_208); \
42154		__ret_208 = __builtin_shufflevector(__ret_208, __ret_208, 7, 6, 5, 4, 3, 2, 1, 0); \
42155		__ret_208; \
42156		})
42157		#endif
42158
42159		#ifdef __LITTLE_ENDIAN__
42160		__ai float16x8_t vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42161		float16x8_t __ret;
42162		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43101	__ai uint64_t vcged_s64(int64_t __p0, int64_t __p1) {
	43102	uint64_t __ret;
	43103	__ret = (uint64_t) __builtin_neon_vcged_s64(__p0, __p1);
42163	43104	return __ret;
42164	43105	}
42165		#else
42166		__ai float16x8_t vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42167		float16x8_t __ret;
42168		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42169		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
42170		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42171		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
42172		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	43106	__ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
	43107	uint64_t __ret;
	43108	__ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
42173	43109	return __ret;
42174	43110	}
42175		__ai float16x8_t __noswap_vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42176		float16x8_t __ret;
42177		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43111	__ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
	43112	uint64_t __ret;
	43113	__ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
42178	43114	return __ret;
42179	43115	}
42180		#endif
42181
42182		#ifdef __LITTLE_ENDIAN__
42183		__ai float16x4_t vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42184		float16x4_t __ret;
42185		__ret = (float16x4_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43116	__ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
	43117	uint32_t __ret;
	43118	__ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
42186	43119	return __ret;
42187	43120	}
42188		#else
42189		__ai float16x4_t vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42190		float16x4_t __ret;
42191		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
42192		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
42193		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42194		__ret = (float16x4_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
42195		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43121	#ifdef __LITTLE_ENDIAN__
	43122	__ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
	43123	uint8x16_t __ret;
	43124	__ret = (uint8x16_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 48);
42196	43125	return __ret;
42197	43126	}
42198		__ai float16x4_t __noswap_vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42199		float16x4_t __ret;
42200		__ret = (float16x4_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43127	#else
	43128	__ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
	43129	uint8x16_t __ret;
	43130	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	43131	__ret = (uint8x16_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 48);
	43132	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
42201	43133	return __ret;
42202	43134	}
42203	43135	#endif
42204	43136
42205	43137	#ifdef __LITTLE_ENDIAN__
42206		#define vcmla_rot180_lane_f16(__p0_209, __p1_209, __p2_209, __p3_209) __extension__ ({ \
42207		float16x4_t __ret_209; \
42208		float16x4_t __s0_209 = __p0_209; \
42209		float16x4_t __s1_209 = __p1_209; \
42210		float16x4_t __s2_209 = __p2_209; \
42211		float16x4_t __reint_209 = __s2_209; \
42212		uint32x2_t __reint1_209 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_209, __p3_209), vget_lane_u32((uint32x2_t ) &__reint_209, __p3_209)}; \
42213		__ret_209 = vcmla_rot180_f16(__s0_209, __s1_209, (float16x4_t ) &__reint1_209); \
42214		__ret_209; \
42215		})
42216		#else
42217		#define vcmla_rot180_lane_f16(__p0_210, __p1_210, __p2_210, __p3_210) __extension__ ({ \
42218		float16x4_t __ret_210; \
42219		float16x4_t __s0_210 = __p0_210; \
42220		float16x4_t __s1_210 = __p1_210; \
42221		float16x4_t __s2_210 = __p2_210; \
42222		float16x4_t __rev0_210; __rev0_210 = __builtin_shufflevector(__s0_210, __s0_210, 3, 2, 1, 0); \
42223		float16x4_t __rev1_210; __rev1_210 = __builtin_shufflevector(__s1_210, __s1_210, 3, 2, 1, 0); \
42224		float16x4_t __rev2_210; __rev2_210 = __builtin_shufflevector(__s2_210, __s2_210, 3, 2, 1, 0); \
42225		float16x4_t __reint_210 = __rev2_210; \
42226		uint32x2_t __reint1_210 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_210, __p3_210), __noswap_vget_lane_u32((uint32x2_t ) &__reint_210, __p3_210)}; \
42227		__ret_210 = __noswap_vcmla_rot180_f16(__rev0_210, __rev1_210, (float16x4_t ) &__reint1_210); \
42228		__ret_210 = __builtin_shufflevector(__ret_210, __ret_210, 3, 2, 1, 0); \
42229		__ret_210; \
42230		})
42231		#endif
42232
42233		#ifdef __LITTLE_ENDIAN__
42234		#define vcmlaq_rot180_lane_f16(__p0_211, __p1_211, __p2_211, __p3_211) __extension__ ({ \
42235		float16x8_t __ret_211; \
42236		float16x8_t __s0_211 = __p0_211; \
42237		float16x8_t __s1_211 = __p1_211; \
42238		float16x4_t __s2_211 = __p2_211; \
42239		float16x4_t __reint_211 = __s2_211; \
42240		uint32x4_t __reint1_211 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211), vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211), vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211), vget_lane_u32((uint32x2_t ) &__reint_211, __p3_211)}; \
42241		__ret_211 = vcmlaq_rot180_f16(__s0_211, __s1_211, (float16x8_t ) &__reint1_211); \
42242		__ret_211; \
42243		})
42244		#else
42245		#define vcmlaq_rot180_lane_f16(__p0_212, __p1_212, __p2_212, __p3_212) __extension__ ({ \
42246		float16x8_t __ret_212; \
42247		float16x8_t __s0_212 = __p0_212; \
42248		float16x8_t __s1_212 = __p1_212; \
42249		float16x4_t __s2_212 = __p2_212; \
42250		float16x8_t __rev0_212; __rev0_212 = __builtin_shufflevector(__s0_212, __s0_212, 7, 6, 5, 4, 3, 2, 1, 0); \
42251		float16x8_t __rev1_212; __rev1_212 = __builtin_shufflevector(__s1_212, __s1_212, 7, 6, 5, 4, 3, 2, 1, 0); \
42252		float16x4_t __rev2_212; __rev2_212 = __builtin_shufflevector(__s2_212, __s2_212, 3, 2, 1, 0); \
42253		float16x4_t __reint_212 = __rev2_212; \
42254		uint32x4_t __reint1_212 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212), __noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212), __noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212), __noswap_vget_lane_u32((uint32x2_t ) &__reint_212, __p3_212)}; \
42255		__ret_212 = __noswap_vcmlaq_rot180_f16(__rev0_212, __rev1_212, (float16x8_t ) &__reint1_212); \
42256		__ret_212 = __builtin_shufflevector(__ret_212, __ret_212, 7, 6, 5, 4, 3, 2, 1, 0); \
42257		__ret_212; \
42258		})
42259		#endif
42260
42261		#ifdef __LITTLE_ENDIAN__
42262		#define vcmla_rot180_laneq_f16(__p0_213, __p1_213, __p2_213, __p3_213) __extension__ ({ \
42263		float16x4_t __ret_213; \
42264		float16x4_t __s0_213 = __p0_213; \
42265		float16x4_t __s1_213 = __p1_213; \
42266		float16x8_t __s2_213 = __p2_213; \
42267		float16x8_t __reint_213 = __s2_213; \
42268		uint32x2_t __reint1_213 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_213, __p3_213), vgetq_lane_u32((uint32x4_t ) &__reint_213, __p3_213)}; \
42269		__ret_213 = vcmla_rot180_f16(__s0_213, __s1_213, (float16x4_t ) &__reint1_213); \
42270		__ret_213; \
42271		})
42272		#else
42273		#define vcmla_rot180_laneq_f16(__p0_214, __p1_214, __p2_214, __p3_214) __extension__ ({ \
42274		float16x4_t __ret_214; \
42275		float16x4_t __s0_214 = __p0_214; \
42276		float16x4_t __s1_214 = __p1_214; \
42277		float16x8_t __s2_214 = __p2_214; \
42278		float16x4_t __rev0_214; __rev0_214 = __builtin_shufflevector(__s0_214, __s0_214, 3, 2, 1, 0); \
42279		float16x4_t __rev1_214; __rev1_214 = __builtin_shufflevector(__s1_214, __s1_214, 3, 2, 1, 0); \
42280		float16x8_t __rev2_214; __rev2_214 = __builtin_shufflevector(__s2_214, __s2_214, 7, 6, 5, 4, 3, 2, 1, 0); \
42281		float16x8_t __reint_214 = __rev2_214; \
42282		uint32x2_t __reint1_214 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_214, __p3_214), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_214, __p3_214)}; \
42283		__ret_214 = __noswap_vcmla_rot180_f16(__rev0_214, __rev1_214, (float16x4_t ) &__reint1_214); \
42284		__ret_214 = __builtin_shufflevector(__ret_214, __ret_214, 3, 2, 1, 0); \
42285		__ret_214; \
42286		})
42287		#endif
42288
42289		#ifdef __LITTLE_ENDIAN__
42290		#define vcmlaq_rot180_laneq_f16(__p0_215, __p1_215, __p2_215, __p3_215) __extension__ ({ \
42291		float16x8_t __ret_215; \
42292		float16x8_t __s0_215 = __p0_215; \
42293		float16x8_t __s1_215 = __p1_215; \
42294		float16x8_t __s2_215 = __p2_215; \
42295		float16x8_t __reint_215 = __s2_215; \
42296		uint32x4_t __reint1_215 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215), vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215), vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215), vgetq_lane_u32((uint32x4_t ) &__reint_215, __p3_215)}; \
42297		__ret_215 = vcmlaq_rot180_f16(__s0_215, __s1_215, (float16x8_t ) &__reint1_215); \
42298		__ret_215; \
42299		})
42300		#else
42301		#define vcmlaq_rot180_laneq_f16(__p0_216, __p1_216, __p2_216, __p3_216) __extension__ ({ \
42302		float16x8_t __ret_216; \
42303		float16x8_t __s0_216 = __p0_216; \
42304		float16x8_t __s1_216 = __p1_216; \
42305		float16x8_t __s2_216 = __p2_216; \
42306		float16x8_t __rev0_216; __rev0_216 = __builtin_shufflevector(__s0_216, __s0_216, 7, 6, 5, 4, 3, 2, 1, 0); \
42307		float16x8_t __rev1_216; __rev1_216 = __builtin_shufflevector(__s1_216, __s1_216, 7, 6, 5, 4, 3, 2, 1, 0); \
42308		float16x8_t __rev2_216; __rev2_216 = __builtin_shufflevector(__s2_216, __s2_216, 7, 6, 5, 4, 3, 2, 1, 0); \
42309		float16x8_t __reint_216 = __rev2_216; \
42310		uint32x4_t __reint1_216 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_216, __p3_216)}; \
42311		__ret_216 = __noswap_vcmlaq_rot180_f16(__rev0_216, __rev1_216, (float16x8_t ) &__reint1_216); \
42312		__ret_216 = __builtin_shufflevector(__ret_216, __ret_216, 7, 6, 5, 4, 3, 2, 1, 0); \
42313		__ret_216; \
42314		})
42315		#endif
42316
42317		#ifdef __LITTLE_ENDIAN__
42318		__ai float16x8_t vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42319		float16x8_t __ret;
42320		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43138	__ai uint64x2_t vcgezq_f64(float64x2_t __p0) {
	43139	uint64x2_t __ret;
	43140	__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 51);
42321	43141	return __ret;
42322	43142	}
42323	43143	#else
42324		__ai float16x8_t vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42325		float16x8_t __ret;
42326		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42327		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
42328		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42329		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
42330		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
42331		return __ret;
42332		}
42333		__ai float16x8_t __noswap_vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42334		float16x8_t __ret;
42335		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43144	__ai uint64x2_t vcgezq_f64(float64x2_t __p0) {
	43145	uint64x2_t __ret;
	43146	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43147	__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 51);
	43148	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
42336	43149	return __ret;
42337	43150	}
42338	43151	#endif
42339	43152
42340	43153	#ifdef __LITTLE_ENDIAN__
42341		__ai float16x4_t vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42342		float16x4_t __ret;
42343		__ret = (float16x4_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43154	__ai uint32x4_t vcgezq_f32(float32x4_t __p0) {
	43155	uint32x4_t __ret;
	43156	__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 50);
42344	43157	return __ret;
42345	43158	}
42346	43159	#else
42347		__ai float16x4_t vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42348		float16x4_t __ret;
42349		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
42350		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
42351		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42352		__ret = (float16x4_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
	43160	__ai uint32x4_t vcgezq_f32(float32x4_t __p0) {
	43161	uint32x4_t __ret;
	43162	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43163	__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 50);
42353	43164	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
42354	43165	return __ret;
42355	43166	}
42356		__ai float16x4_t __noswap_vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42357		float16x4_t __ret;
42358		__ret = (float16x4_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
42359		return __ret;
42360		}
42361		#endif
42362
42363		#ifdef __LITTLE_ENDIAN__
42364		#define vcmla_rot270_lane_f16(__p0_217, __p1_217, __p2_217, __p3_217) __extension__ ({ \
42365		float16x4_t __ret_217; \
42366		float16x4_t __s0_217 = __p0_217; \
42367		float16x4_t __s1_217 = __p1_217; \
42368		float16x4_t __s2_217 = __p2_217; \
42369		float16x4_t __reint_217 = __s2_217; \
42370		uint32x2_t __reint1_217 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_217, __p3_217), vget_lane_u32((uint32x2_t ) &__reint_217, __p3_217)}; \
42371		__ret_217 = vcmla_rot270_f16(__s0_217, __s1_217, (float16x4_t ) &__reint1_217); \
42372		__ret_217; \
42373		})
42374		#else
42375		#define vcmla_rot270_lane_f16(__p0_218, __p1_218, __p2_218, __p3_218) __extension__ ({ \
42376		float16x4_t __ret_218; \
42377		float16x4_t __s0_218 = __p0_218; \
42378		float16x4_t __s1_218 = __p1_218; \
42379		float16x4_t __s2_218 = __p2_218; \
42380		float16x4_t __rev0_218; __rev0_218 = __builtin_shufflevector(__s0_218, __s0_218, 3, 2, 1, 0); \
42381		float16x4_t __rev1_218; __rev1_218 = __builtin_shufflevector(__s1_218, __s1_218, 3, 2, 1, 0); \
42382		float16x4_t __rev2_218; __rev2_218 = __builtin_shufflevector(__s2_218, __s2_218, 3, 2, 1, 0); \
42383		float16x4_t __reint_218 = __rev2_218; \
42384		uint32x2_t __reint1_218 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_218, __p3_218), __noswap_vget_lane_u32((uint32x2_t ) &__reint_218, __p3_218)}; \
42385		__ret_218 = __noswap_vcmla_rot270_f16(__rev0_218, __rev1_218, (float16x4_t ) &__reint1_218); \
42386		__ret_218 = __builtin_shufflevector(__ret_218, __ret_218, 3, 2, 1, 0); \
42387		__ret_218; \
42388		})
42389		#endif
42390
42391		#ifdef __LITTLE_ENDIAN__
42392		#define vcmlaq_rot270_lane_f16(__p0_219, __p1_219, __p2_219, __p3_219) __extension__ ({ \
42393		float16x8_t __ret_219; \
42394		float16x8_t __s0_219 = __p0_219; \
42395		float16x8_t __s1_219 = __p1_219; \
42396		float16x4_t __s2_219 = __p2_219; \
42397		float16x4_t __reint_219 = __s2_219; \
42398		uint32x4_t __reint1_219 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219), vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219), vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219), vget_lane_u32((uint32x2_t ) &__reint_219, __p3_219)}; \
42399		__ret_219 = vcmlaq_rot270_f16(__s0_219, __s1_219, (float16x8_t ) &__reint1_219); \
42400		__ret_219; \
42401		})
42402		#else
42403		#define vcmlaq_rot270_lane_f16(__p0_220, __p1_220, __p2_220, __p3_220) __extension__ ({ \
42404		float16x8_t __ret_220; \
42405		float16x8_t __s0_220 = __p0_220; \
42406		float16x8_t __s1_220 = __p1_220; \
42407		float16x4_t __s2_220 = __p2_220; \
42408		float16x8_t __rev0_220; __rev0_220 = __builtin_shufflevector(__s0_220, __s0_220, 7, 6, 5, 4, 3, 2, 1, 0); \
42409		float16x8_t __rev1_220; __rev1_220 = __builtin_shufflevector(__s1_220, __s1_220, 7, 6, 5, 4, 3, 2, 1, 0); \
42410		float16x4_t __rev2_220; __rev2_220 = __builtin_shufflevector(__s2_220, __s2_220, 3, 2, 1, 0); \
42411		float16x4_t __reint_220 = __rev2_220; \
42412		uint32x4_t __reint1_220 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220), __noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220), __noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220), __noswap_vget_lane_u32((uint32x2_t ) &__reint_220, __p3_220)}; \
42413		__ret_220 = __noswap_vcmlaq_rot270_f16(__rev0_220, __rev1_220, (float16x8_t ) &__reint1_220); \
42414		__ret_220 = __builtin_shufflevector(__ret_220, __ret_220, 7, 6, 5, 4, 3, 2, 1, 0); \
42415		__ret_220; \
42416		})
42417		#endif
42418
42419		#ifdef __LITTLE_ENDIAN__
42420		#define vcmla_rot270_laneq_f16(__p0_221, __p1_221, __p2_221, __p3_221) __extension__ ({ \
42421		float16x4_t __ret_221; \
42422		float16x4_t __s0_221 = __p0_221; \
42423		float16x4_t __s1_221 = __p1_221; \
42424		float16x8_t __s2_221 = __p2_221; \
42425		float16x8_t __reint_221 = __s2_221; \
42426		uint32x2_t __reint1_221 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_221, __p3_221), vgetq_lane_u32((uint32x4_t ) &__reint_221, __p3_221)}; \
42427		__ret_221 = vcmla_rot270_f16(__s0_221, __s1_221, (float16x4_t ) &__reint1_221); \
42428		__ret_221; \
42429		})
42430		#else
42431		#define vcmla_rot270_laneq_f16(__p0_222, __p1_222, __p2_222, __p3_222) __extension__ ({ \
42432		float16x4_t __ret_222; \
42433		float16x4_t __s0_222 = __p0_222; \
42434		float16x4_t __s1_222 = __p1_222; \
42435		float16x8_t __s2_222 = __p2_222; \
42436		float16x4_t __rev0_222; __rev0_222 = __builtin_shufflevector(__s0_222, __s0_222, 3, 2, 1, 0); \
42437		float16x4_t __rev1_222; __rev1_222 = __builtin_shufflevector(__s1_222, __s1_222, 3, 2, 1, 0); \
42438		float16x8_t __rev2_222; __rev2_222 = __builtin_shufflevector(__s2_222, __s2_222, 7, 6, 5, 4, 3, 2, 1, 0); \
42439		float16x8_t __reint_222 = __rev2_222; \
42440		uint32x2_t __reint1_222 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_222, __p3_222), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_222, __p3_222)}; \
42441		__ret_222 = __noswap_vcmla_rot270_f16(__rev0_222, __rev1_222, (float16x4_t ) &__reint1_222); \
42442		__ret_222 = __builtin_shufflevector(__ret_222, __ret_222, 3, 2, 1, 0); \
42443		__ret_222; \
42444		})
42445		#endif
42446
42447		#ifdef __LITTLE_ENDIAN__
42448		#define vcmlaq_rot270_laneq_f16(__p0_223, __p1_223, __p2_223, __p3_223) __extension__ ({ \
42449		float16x8_t __ret_223; \
42450		float16x8_t __s0_223 = __p0_223; \
42451		float16x8_t __s1_223 = __p1_223; \
42452		float16x8_t __s2_223 = __p2_223; \
42453		float16x8_t __reint_223 = __s2_223; \
42454		uint32x4_t __reint1_223 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223), vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223), vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223), vgetq_lane_u32((uint32x4_t ) &__reint_223, __p3_223)}; \
42455		__ret_223 = vcmlaq_rot270_f16(__s0_223, __s1_223, (float16x8_t ) &__reint1_223); \
42456		__ret_223; \
42457		})
42458		#else
42459		#define vcmlaq_rot270_laneq_f16(__p0_224, __p1_224, __p2_224, __p3_224) __extension__ ({ \
42460		float16x8_t __ret_224; \
42461		float16x8_t __s0_224 = __p0_224; \
42462		float16x8_t __s1_224 = __p1_224; \
42463		float16x8_t __s2_224 = __p2_224; \
42464		float16x8_t __rev0_224; __rev0_224 = __builtin_shufflevector(__s0_224, __s0_224, 7, 6, 5, 4, 3, 2, 1, 0); \
42465		float16x8_t __rev1_224; __rev1_224 = __builtin_shufflevector(__s1_224, __s1_224, 7, 6, 5, 4, 3, 2, 1, 0); \
42466		float16x8_t __rev2_224; __rev2_224 = __builtin_shufflevector(__s2_224, __s2_224, 7, 6, 5, 4, 3, 2, 1, 0); \
42467		float16x8_t __reint_224 = __rev2_224; \
42468		uint32x4_t __reint1_224 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_224, __p3_224)}; \
42469		__ret_224 = __noswap_vcmlaq_rot270_f16(__rev0_224, __rev1_224, (float16x8_t ) &__reint1_224); \
42470		__ret_224 = __builtin_shufflevector(__ret_224, __ret_224, 7, 6, 5, 4, 3, 2, 1, 0); \
42471		__ret_224; \
42472		})
42473	43167	#endif
42474	43168
42475	43169	#ifdef __LITTLE_ENDIAN__
42476		__ai float16x8_t vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42477		float16x8_t __ret;
42478		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43170	__ai uint32x4_t vcgezq_s32(int32x4_t __p0) {
	43171	uint32x4_t __ret;
	43172	__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 50);
42479	43173	return __ret;
42480	43174	}
42481	43175	#else
42482		__ai float16x8_t vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42483		float16x8_t __ret;
42484		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42485		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
42486		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42487		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
42488		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
42489		return __ret;
42490		}
42491		__ai float16x8_t __noswap_vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42492		float16x8_t __ret;
42493		__ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	43176	__ai uint32x4_t vcgezq_s32(int32x4_t __p0) {
	43177	uint32x4_t __ret;
	43178	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43179	__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 50);
	43180	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
42494	43181	return __ret;
42495	43182	}
42496	43183	#endif
42497	43184
42498	43185	#ifdef __LITTLE_ENDIAN__
42499		__ai float16x4_t vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42500		float16x4_t __ret;
42501		__ret = (float16x4_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43186	__ai uint64x2_t vcgezq_s64(int64x2_t __p0) {
	43187	uint64x2_t __ret;
	43188	__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 51);
42502	43189	return __ret;
42503	43190	}
42504	43191	#else
42505		__ai float16x4_t vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42506		float16x4_t __ret;
42507		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
42508		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
42509		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42510		__ret = (float16x4_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
42511		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
42512		return __ret;
42513		}
42514		__ai float16x4_t __noswap_vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42515		float16x4_t __ret;
42516		__ret = (float16x4_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	43192	__ai uint64x2_t vcgezq_s64(int64x2_t __p0) {
	43193	uint64x2_t __ret;
	43194	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43195	__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 51);
	43196	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
42517	43197	return __ret;
42518	43198	}
42519	43199	#endif
42520	43200
42521	43201	#ifdef __LITTLE_ENDIAN__
42522		#define vcmla_rot90_lane_f16(__p0_225, __p1_225, __p2_225, __p3_225) __extension__ ({ \
42523		float16x4_t __ret_225; \
42524		float16x4_t __s0_225 = __p0_225; \
42525		float16x4_t __s1_225 = __p1_225; \
42526		float16x4_t __s2_225 = __p2_225; \
42527		float16x4_t __reint_225 = __s2_225; \
42528		uint32x2_t __reint1_225 = (uint32x2_t) {vget_lane_u32((uint32x2_t ) &__reint_225, __p3_225), vget_lane_u32((uint32x2_t ) &__reint_225, __p3_225)}; \
42529		__ret_225 = vcmla_rot90_f16(__s0_225, __s1_225, (float16x4_t ) &__reint1_225); \
42530		__ret_225; \
42531		})
42532		#else
42533		#define vcmla_rot90_lane_f16(__p0_226, __p1_226, __p2_226, __p3_226) __extension__ ({ \
42534		float16x4_t __ret_226; \
42535		float16x4_t __s0_226 = __p0_226; \
42536		float16x4_t __s1_226 = __p1_226; \
42537		float16x4_t __s2_226 = __p2_226; \
42538		float16x4_t __rev0_226; __rev0_226 = __builtin_shufflevector(__s0_226, __s0_226, 3, 2, 1, 0); \
42539		float16x4_t __rev1_226; __rev1_226 = __builtin_shufflevector(__s1_226, __s1_226, 3, 2, 1, 0); \
42540		float16x4_t __rev2_226; __rev2_226 = __builtin_shufflevector(__s2_226, __s2_226, 3, 2, 1, 0); \
42541		float16x4_t __reint_226 = __rev2_226; \
42542		uint32x2_t __reint1_226 = (uint32x2_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_226, __p3_226), __noswap_vget_lane_u32((uint32x2_t ) &__reint_226, __p3_226)}; \
42543		__ret_226 = __noswap_vcmla_rot90_f16(__rev0_226, __rev1_226, (float16x4_t ) &__reint1_226); \
42544		__ret_226 = __builtin_shufflevector(__ret_226, __ret_226, 3, 2, 1, 0); \
42545		__ret_226; \
42546		})
42547		#endif
42548
42549		#ifdef __LITTLE_ENDIAN__
42550		#define vcmlaq_rot90_lane_f16(__p0_227, __p1_227, __p2_227, __p3_227) __extension__ ({ \
42551		float16x8_t __ret_227; \
42552		float16x8_t __s0_227 = __p0_227; \
42553		float16x8_t __s1_227 = __p1_227; \
42554		float16x4_t __s2_227 = __p2_227; \
42555		float16x4_t __reint_227 = __s2_227; \
42556		uint32x4_t __reint1_227 = (uint32x4_t) {vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227), vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227), vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227), vget_lane_u32((uint32x2_t ) &__reint_227, __p3_227)}; \
42557		__ret_227 = vcmlaq_rot90_f16(__s0_227, __s1_227, (float16x8_t ) &__reint1_227); \
42558		__ret_227; \
42559		})
42560		#else
42561		#define vcmlaq_rot90_lane_f16(__p0_228, __p1_228, __p2_228, __p3_228) __extension__ ({ \
42562		float16x8_t __ret_228; \
42563		float16x8_t __s0_228 = __p0_228; \
42564		float16x8_t __s1_228 = __p1_228; \
42565		float16x4_t __s2_228 = __p2_228; \
42566		float16x8_t __rev0_228; __rev0_228 = __builtin_shufflevector(__s0_228, __s0_228, 7, 6, 5, 4, 3, 2, 1, 0); \
42567		float16x8_t __rev1_228; __rev1_228 = __builtin_shufflevector(__s1_228, __s1_228, 7, 6, 5, 4, 3, 2, 1, 0); \
42568		float16x4_t __rev2_228; __rev2_228 = __builtin_shufflevector(__s2_228, __s2_228, 3, 2, 1, 0); \
42569		float16x4_t __reint_228 = __rev2_228; \
42570		uint32x4_t __reint1_228 = (uint32x4_t) {__noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228), __noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228), __noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228), __noswap_vget_lane_u32((uint32x2_t ) &__reint_228, __p3_228)}; \
42571		__ret_228 = __noswap_vcmlaq_rot90_f16(__rev0_228, __rev1_228, (float16x8_t ) &__reint1_228); \
42572		__ret_228 = __builtin_shufflevector(__ret_228, __ret_228, 7, 6, 5, 4, 3, 2, 1, 0); \
42573		__ret_228; \
42574		})
42575		#endif
42576
42577		#ifdef __LITTLE_ENDIAN__
42578		#define vcmla_rot90_laneq_f16(__p0_229, __p1_229, __p2_229, __p3_229) __extension__ ({ \
42579		float16x4_t __ret_229; \
42580		float16x4_t __s0_229 = __p0_229; \
42581		float16x4_t __s1_229 = __p1_229; \
42582		float16x8_t __s2_229 = __p2_229; \
42583		float16x8_t __reint_229 = __s2_229; \
42584		uint32x2_t __reint1_229 = (uint32x2_t) {vgetq_lane_u32((uint32x4_t ) &__reint_229, __p3_229), vgetq_lane_u32((uint32x4_t ) &__reint_229, __p3_229)}; \
42585		__ret_229 = vcmla_rot90_f16(__s0_229, __s1_229, (float16x4_t ) &__reint1_229); \
42586		__ret_229; \
42587		})
42588		#else
42589		#define vcmla_rot90_laneq_f16(__p0_230, __p1_230, __p2_230, __p3_230) __extension__ ({ \
42590		float16x4_t __ret_230; \
42591		float16x4_t __s0_230 = __p0_230; \
42592		float16x4_t __s1_230 = __p1_230; \
42593		float16x8_t __s2_230 = __p2_230; \
42594		float16x4_t __rev0_230; __rev0_230 = __builtin_shufflevector(__s0_230, __s0_230, 3, 2, 1, 0); \
42595		float16x4_t __rev1_230; __rev1_230 = __builtin_shufflevector(__s1_230, __s1_230, 3, 2, 1, 0); \
42596		float16x8_t __rev2_230; __rev2_230 = __builtin_shufflevector(__s2_230, __s2_230, 7, 6, 5, 4, 3, 2, 1, 0); \
42597		float16x8_t __reint_230 = __rev2_230; \
42598		uint32x2_t __reint1_230 = (uint32x2_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_230, __p3_230), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_230, __p3_230)}; \
42599		__ret_230 = __noswap_vcmla_rot90_f16(__rev0_230, __rev1_230, (float16x4_t ) &__reint1_230); \
42600		__ret_230 = __builtin_shufflevector(__ret_230, __ret_230, 3, 2, 1, 0); \
42601		__ret_230; \
42602		})
42603		#endif
42604
42605		#ifdef __LITTLE_ENDIAN__
42606		#define vcmlaq_rot90_laneq_f16(__p0_231, __p1_231, __p2_231, __p3_231) __extension__ ({ \
42607		float16x8_t __ret_231; \
42608		float16x8_t __s0_231 = __p0_231; \
42609		float16x8_t __s1_231 = __p1_231; \
42610		float16x8_t __s2_231 = __p2_231; \
42611		float16x8_t __reint_231 = __s2_231; \
42612		uint32x4_t __reint1_231 = (uint32x4_t) {vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231), vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231), vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231), vgetq_lane_u32((uint32x4_t ) &__reint_231, __p3_231)}; \
42613		__ret_231 = vcmlaq_rot90_f16(__s0_231, __s1_231, (float16x8_t ) &__reint1_231); \
42614		__ret_231; \
42615		})
	43202	__ai uint16x8_t vcgezq_s16(int16x8_t __p0) {
	43203	uint16x8_t __ret;
	43204	__ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 49);
	43205	return __ret;
	43206	}
42616	43207	#else
42617		#define vcmlaq_rot90_laneq_f16(__p0_232, __p1_232, __p2_232, __p3_232) __extension__ ({ \
42618		float16x8_t __ret_232; \
42619		float16x8_t __s0_232 = __p0_232; \
42620		float16x8_t __s1_232 = __p1_232; \
42621		float16x8_t __s2_232 = __p2_232; \
42622		float16x8_t __rev0_232; __rev0_232 = __builtin_shufflevector(__s0_232, __s0_232, 7, 6, 5, 4, 3, 2, 1, 0); \
42623		float16x8_t __rev1_232; __rev1_232 = __builtin_shufflevector(__s1_232, __s1_232, 7, 6, 5, 4, 3, 2, 1, 0); \
42624		float16x8_t __rev2_232; __rev2_232 = __builtin_shufflevector(__s2_232, __s2_232, 7, 6, 5, 4, 3, 2, 1, 0); \
42625		float16x8_t __reint_232 = __rev2_232; \
42626		uint32x4_t __reint1_232 = (uint32x4_t) {__noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232), __noswap_vgetq_lane_u32((uint32x4_t ) &__reint_232, __p3_232)}; \
42627		__ret_232 = __noswap_vcmlaq_rot90_f16(__rev0_232, __rev1_232, (float16x8_t ) &__reint1_232); \
42628		__ret_232 = __builtin_shufflevector(__ret_232, __ret_232, 7, 6, 5, 4, 3, 2, 1, 0); \
42629		__ret_232; \
42630		})
	43208	__ai uint16x8_t vcgezq_s16(int16x8_t __p0) {
	43209	uint16x8_t __ret;
	43210	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	43211	__ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 49);
	43212	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	43213	return __ret;
	43214	}
42631	43215	#endif
42632	43216
42633		#endif
42634		#if defined(__ARM_FEATURE_COMPLEX) && defined(__aarch64__)
42635	43217	#ifdef __LITTLE_ENDIAN__
42636		__ai float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
42637		float64x2_t __ret;
42638		__ret = (float64x2_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	43218	__ai uint8x8_t vcgez_s8(int8x8_t __p0) {
	43219	uint8x8_t __ret;
	43220	__ret = (uint8x8_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 16);
42639	43221	return __ret;
42640	43222	}
42641	43223	#else
42642		__ai float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
42643		float64x2_t __ret;
42644		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
42645		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42646		__ret = (float64x2_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
42647		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43224	__ai uint8x8_t vcgez_s8(int8x8_t __p0) {
	43225	uint8x8_t __ret;
	43226	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	43227	__ret = (uint8x8_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 16);
	43228	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
42648	43229	return __ret;
42649	43230	}
42650	43231	#endif
42651	43232
	43233	__ai uint64x1_t vcgez_f64(float64x1_t __p0) {
	43234	uint64x1_t __ret;
	43235	__ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
	43236	return __ret;
	43237	}
42652	43238	#ifdef __LITTLE_ENDIAN__
42653		__ai float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
42654		float64x2_t __ret;
42655		__ret = (float64x2_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	43239	__ai uint32x2_t vcgez_f32(float32x2_t __p0) {
	43240	uint32x2_t __ret;
	43241	__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 18);
42656	43242	return __ret;
42657	43243	}
42658	43244	#else
42659		__ai float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
42660		float64x2_t __ret;
42661		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
42662		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42663		__ret = (float64x2_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	43245	__ai uint32x2_t vcgez_f32(float32x2_t __p0) {
	43246	uint32x2_t __ret;
	43247	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43248	__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 18);
42664	43249	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
42665	43250	return __ret;
42666	43251	}
42667	43252	#endif
42668	43253
42669	43254	#ifdef __LITTLE_ENDIAN__
42670		__ai float64x2_t vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42671		float64x2_t __ret;
42672		__ret = (float64x2_t) __builtin_neon_vcmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	43255	__ai uint32x2_t vcgez_s32(int32x2_t __p0) {
	43256	uint32x2_t __ret;
	43257	__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 18);
42673	43258	return __ret;
42674	43259	}
42675	43260	#else
42676		__ai float64x2_t vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42677		float64x2_t __ret;
42678		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
42679		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42680		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
42681		__ret = (float64x2_t) __builtin_neon_vcmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	43261	__ai uint32x2_t vcgez_s32(int32x2_t __p0) {
	43262	uint32x2_t __ret;
	43263	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43264	__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 18);
42682	43265	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
42683	43266	return __ret;
42684	43267	}
42685		__ai float64x2_t __noswap_vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42686		float64x2_t __ret;
42687		__ret = (float64x2_t) __builtin_neon_vcmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
42688		return __ret;
42689		}
42690	43268	#endif
42691	43269
42692		__ai float64x1_t vcmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
42693		float64x1_t __ret;
42694		__ret = (float64x1_t) __builtin_neon_vcmla_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	43270	__ai uint64x1_t vcgez_s64(int64x1_t __p0) {
	43271	uint64x1_t __ret;
	43272	__ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
42695	43273	return __ret;
42696	43274	}
42697		#define vcmla_lane_f64(__p0_233, __p1_233, __p2_233, __p3_233) __extension__ ({ \
42698		float64x1_t __ret_233; \
42699		float64x1_t __s0_233 = __p0_233; \
42700		float64x1_t __s1_233 = __p1_233; \
42701		float64x1_t __s2_233 = __p2_233; \
42702		float64x1_t __reint_233 = __s2_233; \
42703		uint64x2_t __reint1_233 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_233, __p3_233), vgetq_lane_u64((uint64x2_t ) &__reint_233, __p3_233)}; \
42704		__ret_233 = vcmla_f64(__s0_233, __s1_233, (float64x1_t ) &__reint1_233); \
42705		__ret_233; \
42706		})
42707		#ifdef __LITTLE_ENDIAN__
42708		#define vcmlaq_lane_f64(__p0_234, __p1_234, __p2_234, __p3_234) __extension__ ({ \
42709		float64x2_t __ret_234; \
42710		float64x2_t __s0_234 = __p0_234; \
42711		float64x2_t __s1_234 = __p1_234; \
42712		float64x1_t __s2_234 = __p2_234; \
42713		float64x1_t __reint_234 = __s2_234; \
42714		uint64x2_t __reint1_234 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_234, __p3_234), vgetq_lane_u64((uint64x2_t ) &__reint_234, __p3_234)}; \
42715		__ret_234 = vcmlaq_f64(__s0_234, __s1_234, (float64x2_t ) &__reint1_234); \
42716		__ret_234; \
42717		})
42718		#else
42719		#define vcmlaq_lane_f64(__p0_235, __p1_235, __p2_235, __p3_235) __extension__ ({ \
42720		float64x2_t __ret_235; \
42721		float64x2_t __s0_235 = __p0_235; \
42722		float64x2_t __s1_235 = __p1_235; \
42723		float64x1_t __s2_235 = __p2_235; \
42724		float64x2_t __rev0_235; __rev0_235 = __builtin_shufflevector(__s0_235, __s0_235, 1, 0); \
42725		float64x2_t __rev1_235; __rev1_235 = __builtin_shufflevector(__s1_235, __s1_235, 1, 0); \
42726		float64x1_t __reint_235 = __s2_235; \
42727		uint64x2_t __reint1_235 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_235, __p3_235), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_235, __p3_235)}; \
42728		__ret_235 = __noswap_vcmlaq_f64(__rev0_235, __rev1_235, (float64x2_t ) &__reint1_235); \
42729		__ret_235 = __builtin_shufflevector(__ret_235, __ret_235, 1, 0); \
42730		__ret_235; \
42731		})
42732		#endif
42733
42734	43275	#ifdef __LITTLE_ENDIAN__
42735		#define vcmla_laneq_f64(__p0_236, __p1_236, __p2_236, __p3_236) __extension__ ({ \
42736		float64x1_t __ret_236; \
42737		float64x1_t __s0_236 = __p0_236; \
42738		float64x1_t __s1_236 = __p1_236; \
42739		float64x2_t __s2_236 = __p2_236; \
42740		float64x2_t __reint_236 = __s2_236; \
42741		uint64x2_t __reint1_236 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_236, __p3_236), vgetq_lane_u64((uint64x2_t ) &__reint_236, __p3_236)}; \
42742		__ret_236 = vcmla_f64(__s0_236, __s1_236, (float64x1_t ) &__reint1_236); \
42743		__ret_236; \
42744		})
	43276	__ai uint16x4_t vcgez_s16(int16x4_t __p0) {
	43277	uint16x4_t __ret;
	43278	__ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 17);
	43279	return __ret;
	43280	}
42745	43281	#else
42746		#define vcmla_laneq_f64(__p0_237, __p1_237, __p2_237, __p3_237) __extension__ ({ \
42747		float64x1_t __ret_237; \
42748		float64x1_t __s0_237 = __p0_237; \
42749		float64x1_t __s1_237 = __p1_237; \
42750		float64x2_t __s2_237 = __p2_237; \
42751		float64x2_t __rev2_237; __rev2_237 = __builtin_shufflevector(__s2_237, __s2_237, 1, 0); \
42752		float64x2_t __reint_237 = __rev2_237; \
42753		uint64x2_t __reint1_237 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_237, __p3_237), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_237, __p3_237)}; \
42754		__ret_237 = vcmla_f64(__s0_237, __s1_237, (float64x1_t ) &__reint1_237); \
42755		__ret_237; \
42756		})
	43282	__ai uint16x4_t vcgez_s16(int16x4_t __p0) {
	43283	uint16x4_t __ret;
	43284	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43285	__ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 17);
	43286	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43287	return __ret;
	43288	}
42757	43289	#endif
42758	43290
	43291	__ai uint64_t vcgezd_s64(int64_t __p0) {
	43292	uint64_t __ret;
	43293	__ret = (uint64_t) __builtin_neon_vcgezd_s64(__p0);
	43294	return __ret;
	43295	}
	43296	__ai uint64_t vcgezd_f64(float64_t __p0) {
	43297	uint64_t __ret;
	43298	__ret = (uint64_t) __builtin_neon_vcgezd_f64(__p0);
	43299	return __ret;
	43300	}
	43301	__ai uint32_t vcgezs_f32(float32_t __p0) {
	43302	uint32_t __ret;
	43303	__ret = (uint32_t) __builtin_neon_vcgezs_f32(__p0);
	43304	return __ret;
	43305	}
42759	43306	#ifdef __LITTLE_ENDIAN__
42760		#define vcmlaq_laneq_f64(__p0_238, __p1_238, __p2_238, __p3_238) __extension__ ({ \
42761		float64x2_t __ret_238; \
42762		float64x2_t __s0_238 = __p0_238; \
42763		float64x2_t __s1_238 = __p1_238; \
42764		float64x2_t __s2_238 = __p2_238; \
42765		float64x2_t __reint_238 = __s2_238; \
42766		uint64x2_t __reint1_238 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_238, __p3_238), vgetq_lane_u64((uint64x2_t ) &__reint_238, __p3_238)}; \
42767		__ret_238 = vcmlaq_f64(__s0_238, __s1_238, (float64x2_t ) &__reint1_238); \
42768		__ret_238; \
42769		})
	43307	__ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43308	uint64x2_t __ret;
	43309	__ret = (uint64x2_t)(__p0 > __p1);
	43310	return __ret;
	43311	}
42770	43312	#else
42771		#define vcmlaq_laneq_f64(__p0_239, __p1_239, __p2_239, __p3_239) __extension__ ({ \
42772		float64x2_t __ret_239; \
42773		float64x2_t __s0_239 = __p0_239; \
42774		float64x2_t __s1_239 = __p1_239; \
42775		float64x2_t __s2_239 = __p2_239; \
42776		float64x2_t __rev0_239; __rev0_239 = __builtin_shufflevector(__s0_239, __s0_239, 1, 0); \
42777		float64x2_t __rev1_239; __rev1_239 = __builtin_shufflevector(__s1_239, __s1_239, 1, 0); \
42778		float64x2_t __rev2_239; __rev2_239 = __builtin_shufflevector(__s2_239, __s2_239, 1, 0); \
42779		float64x2_t __reint_239 = __rev2_239; \
42780		uint64x2_t __reint1_239 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_239, __p3_239), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_239, __p3_239)}; \
42781		__ret_239 = __noswap_vcmlaq_f64(__rev0_239, __rev1_239, (float64x2_t ) &__reint1_239); \
42782		__ret_239 = __builtin_shufflevector(__ret_239, __ret_239, 1, 0); \
42783		__ret_239; \
42784		})
	43313	__ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43314	uint64x2_t __ret;
	43315	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43316	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43317	__ret = (uint64x2_t)(__rev0 > __rev1);
	43318	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43319	return __ret;
	43320	}
42785	43321	#endif
42786	43322
42787	43323	#ifdef __LITTLE_ENDIAN__
42788		__ai float64x2_t vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42789		float64x2_t __ret;
42790		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	43324	__ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
	43325	uint64x2_t __ret;
	43326	__ret = (uint64x2_t)(__p0 > __p1);
42791	43327	return __ret;
42792	43328	}
42793	43329	#else
42794		__ai float64x2_t vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42795		float64x2_t __ret;
	43330	__ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
	43331	uint64x2_t __ret;
42796	43332	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
42797	43333	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42798		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
42799		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	43334	__ret = (uint64x2_t)(__rev0 > __rev1);
42800	43335	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
42801	43336	return __ret;
42802	43337	}
42803		__ai float64x2_t __noswap_vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42804		float64x2_t __ret;
42805		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
42806		return __ret;
42807		}
42808	43338	#endif
42809	43339
42810		__ai float64x1_t vcmla_rot180_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
42811		float64x1_t __ret;
42812		__ret = (float64x1_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	43340	#ifdef __LITTLE_ENDIAN__
	43341	__ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
	43342	uint64x2_t __ret;
	43343	__ret = (uint64x2_t)(__p0 > __p1);
42813	43344	return __ret;
42814	43345	}
42815		#define vcmla_rot180_lane_f64(__p0_240, __p1_240, __p2_240, __p3_240) __extension__ ({ \
42816		float64x1_t __ret_240; \
42817		float64x1_t __s0_240 = __p0_240; \
42818		float64x1_t __s1_240 = __p1_240; \
42819		float64x1_t __s2_240 = __p2_240; \
42820		float64x1_t __reint_240 = __s2_240; \
42821		uint64x2_t __reint1_240 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_240, __p3_240), vgetq_lane_u64((uint64x2_t ) &__reint_240, __p3_240)}; \
42822		__ret_240 = vcmla_rot180_f64(__s0_240, __s1_240, (float64x1_t ) &__reint1_240); \
42823		__ret_240; \
42824		})
42825		#ifdef __LITTLE_ENDIAN__
42826		#define vcmlaq_rot180_lane_f64(__p0_241, __p1_241, __p2_241, __p3_241) __extension__ ({ \
42827		float64x2_t __ret_241; \
42828		float64x2_t __s0_241 = __p0_241; \
42829		float64x2_t __s1_241 = __p1_241; \
42830		float64x1_t __s2_241 = __p2_241; \
42831		float64x1_t __reint_241 = __s2_241; \
42832		uint64x2_t __reint1_241 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_241, __p3_241), vgetq_lane_u64((uint64x2_t ) &__reint_241, __p3_241)}; \
42833		__ret_241 = vcmlaq_rot180_f64(__s0_241, __s1_241, (float64x2_t ) &__reint1_241); \
42834		__ret_241; \
42835		})
42836		#else
42837		#define vcmlaq_rot180_lane_f64(__p0_242, __p1_242, __p2_242, __p3_242) __extension__ ({ \
42838		float64x2_t __ret_242; \
42839		float64x2_t __s0_242 = __p0_242; \
42840		float64x2_t __s1_242 = __p1_242; \
42841		float64x1_t __s2_242 = __p2_242; \
42842		float64x2_t __rev0_242; __rev0_242 = __builtin_shufflevector(__s0_242, __s0_242, 1, 0); \
42843		float64x2_t __rev1_242; __rev1_242 = __builtin_shufflevector(__s1_242, __s1_242, 1, 0); \
42844		float64x1_t __reint_242 = __s2_242; \
42845		uint64x2_t __reint1_242 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_242, __p3_242), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_242, __p3_242)}; \
42846		__ret_242 = __noswap_vcmlaq_rot180_f64(__rev0_242, __rev1_242, (float64x2_t ) &__reint1_242); \
42847		__ret_242 = __builtin_shufflevector(__ret_242, __ret_242, 1, 0); \
42848		__ret_242; \
42849		})
42850		#endif
42851
42852		#ifdef __LITTLE_ENDIAN__
42853		#define vcmla_rot180_laneq_f64(__p0_243, __p1_243, __p2_243, __p3_243) __extension__ ({ \
42854		float64x1_t __ret_243; \
42855		float64x1_t __s0_243 = __p0_243; \
42856		float64x1_t __s1_243 = __p1_243; \
42857		float64x2_t __s2_243 = __p2_243; \
42858		float64x2_t __reint_243 = __s2_243; \
42859		uint64x2_t __reint1_243 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_243, __p3_243), vgetq_lane_u64((uint64x2_t ) &__reint_243, __p3_243)}; \
42860		__ret_243 = vcmla_rot180_f64(__s0_243, __s1_243, (float64x1_t ) &__reint1_243); \
42861		__ret_243; \
42862		})
42863	43346	#else
42864		#define vcmla_rot180_laneq_f64(__p0_244, __p1_244, __p2_244, __p3_244) __extension__ ({ \
42865		float64x1_t __ret_244; \
42866		float64x1_t __s0_244 = __p0_244; \
42867		float64x1_t __s1_244 = __p1_244; \
42868		float64x2_t __s2_244 = __p2_244; \
42869		float64x2_t __rev2_244; __rev2_244 = __builtin_shufflevector(__s2_244, __s2_244, 1, 0); \
42870		float64x2_t __reint_244 = __rev2_244; \
42871		uint64x2_t __reint1_244 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_244, __p3_244), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_244, __p3_244)}; \
42872		__ret_244 = vcmla_rot180_f64(__s0_244, __s1_244, (float64x1_t ) &__reint1_244); \
42873		__ret_244; \
42874		})
42875		#endif
42876
42877		#ifdef __LITTLE_ENDIAN__
42878		#define vcmlaq_rot180_laneq_f64(__p0_245, __p1_245, __p2_245, __p3_245) __extension__ ({ \
42879		float64x2_t __ret_245; \
42880		float64x2_t __s0_245 = __p0_245; \
42881		float64x2_t __s1_245 = __p1_245; \
42882		float64x2_t __s2_245 = __p2_245; \
42883		float64x2_t __reint_245 = __s2_245; \
42884		uint64x2_t __reint1_245 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_245, __p3_245), vgetq_lane_u64((uint64x2_t ) &__reint_245, __p3_245)}; \
42885		__ret_245 = vcmlaq_rot180_f64(__s0_245, __s1_245, (float64x2_t ) &__reint1_245); \
42886		__ret_245; \
42887		})
42888		#else
42889		#define vcmlaq_rot180_laneq_f64(__p0_246, __p1_246, __p2_246, __p3_246) __extension__ ({ \
42890		float64x2_t __ret_246; \
42891		float64x2_t __s0_246 = __p0_246; \
42892		float64x2_t __s1_246 = __p1_246; \
42893		float64x2_t __s2_246 = __p2_246; \
42894		float64x2_t __rev0_246; __rev0_246 = __builtin_shufflevector(__s0_246, __s0_246, 1, 0); \
42895		float64x2_t __rev1_246; __rev1_246 = __builtin_shufflevector(__s1_246, __s1_246, 1, 0); \
42896		float64x2_t __rev2_246; __rev2_246 = __builtin_shufflevector(__s2_246, __s2_246, 1, 0); \
42897		float64x2_t __reint_246 = __rev2_246; \
42898		uint64x2_t __reint1_246 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_246, __p3_246), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_246, __p3_246)}; \
42899		__ret_246 = __noswap_vcmlaq_rot180_f64(__rev0_246, __rev1_246, (float64x2_t ) &__reint1_246); \
42900		__ret_246 = __builtin_shufflevector(__ret_246, __ret_246, 1, 0); \
42901		__ret_246; \
42902		})
	43347	__ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
	43348	uint64x2_t __ret;
	43349	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43350	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43351	__ret = (uint64x2_t)(__rev0 > __rev1);
	43352	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43353	return __ret;
	43354	}
42903	43355	#endif
42904	43356
42905		#ifdef __LITTLE_ENDIAN__
42906		__ai float64x2_t vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42907		float64x2_t __ret;
42908		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	43357	__ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
	43358	uint64x1_t __ret;
	43359	__ret = (uint64x1_t)(__p0 > __p1);
42909	43360	return __ret;
42910	43361	}
42911		#else
42912		__ai float64x2_t vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42913		float64x2_t __ret;
42914		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
42915		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42916		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
42917		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
42918		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43362	__ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
	43363	uint64x1_t __ret;
	43364	__ret = (uint64x1_t)(__p0 > __p1);
42919	43365	return __ret;
42920	43366	}
42921		__ai float64x2_t __noswap_vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42922		float64x2_t __ret;
42923		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	43367	__ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
	43368	uint64x1_t __ret;
	43369	__ret = (uint64x1_t)(__p0 > __p1);
42924	43370	return __ret;
42925	43371	}
42926		#endif
42927
42928		__ai float64x1_t vcmla_rot270_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
42929		float64x1_t __ret;
42930		__ret = (float64x1_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	43372	__ai uint64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
	43373	uint64_t __ret;
	43374	__ret = (uint64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
	43375	return __ret;
	43376	}
	43377	__ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
	43378	uint64_t __ret;
	43379	__ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
	43380	return __ret;
	43381	}
	43382	__ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
	43383	uint64_t __ret;
	43384	__ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
	43385	return __ret;
	43386	}
	43387	__ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
	43388	uint32_t __ret;
	43389	__ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
42931	43390	return __ret;
42932	43391	}
42933		#define vcmla_rot270_lane_f64(__p0_247, __p1_247, __p2_247, __p3_247) __extension__ ({ \
42934		float64x1_t __ret_247; \
42935		float64x1_t __s0_247 = __p0_247; \
42936		float64x1_t __s1_247 = __p1_247; \
42937		float64x1_t __s2_247 = __p2_247; \
42938		float64x1_t __reint_247 = __s2_247; \
42939		uint64x2_t __reint1_247 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_247, __p3_247), vgetq_lane_u64((uint64x2_t ) &__reint_247, __p3_247)}; \
42940		__ret_247 = vcmla_rot270_f64(__s0_247, __s1_247, (float64x1_t ) &__reint1_247); \
42941		__ret_247; \
42942		})
42943		#ifdef __LITTLE_ENDIAN__
42944		#define vcmlaq_rot270_lane_f64(__p0_248, __p1_248, __p2_248, __p3_248) __extension__ ({ \
42945		float64x2_t __ret_248; \
42946		float64x2_t __s0_248 = __p0_248; \
42947		float64x2_t __s1_248 = __p1_248; \
42948		float64x1_t __s2_248 = __p2_248; \
42949		float64x1_t __reint_248 = __s2_248; \
42950		uint64x2_t __reint1_248 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_248, __p3_248), vgetq_lane_u64((uint64x2_t ) &__reint_248, __p3_248)}; \
42951		__ret_248 = vcmlaq_rot270_f64(__s0_248, __s1_248, (float64x2_t ) &__reint1_248); \
42952		__ret_248; \
42953		})
42954		#else
42955		#define vcmlaq_rot270_lane_f64(__p0_249, __p1_249, __p2_249, __p3_249) __extension__ ({ \
42956		float64x2_t __ret_249; \
42957		float64x2_t __s0_249 = __p0_249; \
42958		float64x2_t __s1_249 = __p1_249; \
42959		float64x1_t __s2_249 = __p2_249; \
42960		float64x2_t __rev0_249; __rev0_249 = __builtin_shufflevector(__s0_249, __s0_249, 1, 0); \
42961		float64x2_t __rev1_249; __rev1_249 = __builtin_shufflevector(__s1_249, __s1_249, 1, 0); \
42962		float64x1_t __reint_249 = __s2_249; \
42963		uint64x2_t __reint1_249 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_249, __p3_249), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_249, __p3_249)}; \
42964		__ret_249 = __noswap_vcmlaq_rot270_f64(__rev0_249, __rev1_249, (float64x2_t ) &__reint1_249); \
42965		__ret_249 = __builtin_shufflevector(__ret_249, __ret_249, 1, 0); \
42966		__ret_249; \
42967		})
42968		#endif
42969
42970		#ifdef __LITTLE_ENDIAN__
42971		#define vcmla_rot270_laneq_f64(__p0_250, __p1_250, __p2_250, __p3_250) __extension__ ({ \
42972		float64x1_t __ret_250; \
42973		float64x1_t __s0_250 = __p0_250; \
42974		float64x1_t __s1_250 = __p1_250; \
42975		float64x2_t __s2_250 = __p2_250; \
42976		float64x2_t __reint_250 = __s2_250; \
42977		uint64x2_t __reint1_250 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_250, __p3_250), vgetq_lane_u64((uint64x2_t ) &__reint_250, __p3_250)}; \
42978		__ret_250 = vcmla_rot270_f64(__s0_250, __s1_250, (float64x1_t ) &__reint1_250); \
42979		__ret_250; \
42980		})
42981		#else
42982		#define vcmla_rot270_laneq_f64(__p0_251, __p1_251, __p2_251, __p3_251) __extension__ ({ \
42983		float64x1_t __ret_251; \
42984		float64x1_t __s0_251 = __p0_251; \
42985		float64x1_t __s1_251 = __p1_251; \
42986		float64x2_t __s2_251 = __p2_251; \
42987		float64x2_t __rev2_251; __rev2_251 = __builtin_shufflevector(__s2_251, __s2_251, 1, 0); \
42988		float64x2_t __reint_251 = __rev2_251; \
42989		uint64x2_t __reint1_251 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_251, __p3_251), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_251, __p3_251)}; \
42990		__ret_251 = vcmla_rot270_f64(__s0_251, __s1_251, (float64x1_t ) &__reint1_251); \
42991		__ret_251; \
42992		})
42993		#endif
42994
42995	43392	#ifdef __LITTLE_ENDIAN__
42996		#define vcmlaq_rot270_laneq_f64(__p0_252, __p1_252, __p2_252, __p3_252) __extension__ ({ \
42997		float64x2_t __ret_252; \
42998		float64x2_t __s0_252 = __p0_252; \
42999		float64x2_t __s1_252 = __p1_252; \
43000		float64x2_t __s2_252 = __p2_252; \
43001		float64x2_t __reint_252 = __s2_252; \
43002		uint64x2_t __reint1_252 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_252, __p3_252), vgetq_lane_u64((uint64x2_t ) &__reint_252, __p3_252)}; \
43003		__ret_252 = vcmlaq_rot270_f64(__s0_252, __s1_252, (float64x2_t ) &__reint1_252); \
43004		__ret_252; \
43005		})
	43393	__ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
	43394	uint8x16_t __ret;
	43395	__ret = (uint8x16_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 48);
	43396	return __ret;
	43397	}
43006	43398	#else
43007		#define vcmlaq_rot270_laneq_f64(__p0_253, __p1_253, __p2_253, __p3_253) __extension__ ({ \
43008		float64x2_t __ret_253; \
43009		float64x2_t __s0_253 = __p0_253; \
43010		float64x2_t __s1_253 = __p1_253; \
43011		float64x2_t __s2_253 = __p2_253; \
43012		float64x2_t __rev0_253; __rev0_253 = __builtin_shufflevector(__s0_253, __s0_253, 1, 0); \
43013		float64x2_t __rev1_253; __rev1_253 = __builtin_shufflevector(__s1_253, __s1_253, 1, 0); \
43014		float64x2_t __rev2_253; __rev2_253 = __builtin_shufflevector(__s2_253, __s2_253, 1, 0); \
43015		float64x2_t __reint_253 = __rev2_253; \
43016		uint64x2_t __reint1_253 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_253, __p3_253), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_253, __p3_253)}; \
43017		__ret_253 = __noswap_vcmlaq_rot270_f64(__rev0_253, __rev1_253, (float64x2_t ) &__reint1_253); \
43018		__ret_253 = __builtin_shufflevector(__ret_253, __ret_253, 1, 0); \
43019		__ret_253; \
43020		})
	43399	__ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
	43400	uint8x16_t __ret;
	43401	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	43402	__ret = (uint8x16_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 48);
	43403	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	43404	return __ret;
	43405	}
43021	43406	#endif
43022	43407
43023	43408	#ifdef __LITTLE_ENDIAN__
43024		__ai float64x2_t vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
43025		float64x2_t __ret;
43026		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	43409	__ai uint64x2_t vcgtzq_f64(float64x2_t __p0) {
	43410	uint64x2_t __ret;
	43411	__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 51);
43027	43412	return __ret;
43028	43413	}
43029	43414	#else
43030		__ai float64x2_t vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
43031		float64x2_t __ret;
	43415	__ai uint64x2_t vcgtzq_f64(float64x2_t __p0) {
	43416	uint64x2_t __ret;
43032	43417	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43033		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43034		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
43035		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	43418	__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 51);
43036	43419	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43037	43420	return __ret;
43038	43421	}
43039		__ai float64x2_t __noswap_vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
43040		float64x2_t __ret;
43041		__ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
43042		return __ret;
43043		}
43044	43422	#endif
43045	43423
43046		__ai float64x1_t vcmla_rot90_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
43047		float64x1_t __ret;
43048		__ret = (float64x1_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	43424	#ifdef __LITTLE_ENDIAN__
	43425	__ai uint32x4_t vcgtzq_f32(float32x4_t __p0) {
	43426	uint32x4_t __ret;
	43427	__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 50);
43049	43428	return __ret;
43050	43429	}
43051		#define vcmla_rot90_lane_f64(__p0_254, __p1_254, __p2_254, __p3_254) __extension__ ({ \
43052		float64x1_t __ret_254; \
43053		float64x1_t __s0_254 = __p0_254; \
43054		float64x1_t __s1_254 = __p1_254; \
43055		float64x1_t __s2_254 = __p2_254; \
43056		float64x1_t __reint_254 = __s2_254; \
43057		uint64x2_t __reint1_254 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_254, __p3_254), vgetq_lane_u64((uint64x2_t ) &__reint_254, __p3_254)}; \
43058		__ret_254 = vcmla_rot90_f64(__s0_254, __s1_254, (float64x1_t ) &__reint1_254); \
43059		__ret_254; \
43060		})
43061		#ifdef __LITTLE_ENDIAN__
43062		#define vcmlaq_rot90_lane_f64(__p0_255, __p1_255, __p2_255, __p3_255) __extension__ ({ \
43063		float64x2_t __ret_255; \
43064		float64x2_t __s0_255 = __p0_255; \
43065		float64x2_t __s1_255 = __p1_255; \
43066		float64x1_t __s2_255 = __p2_255; \
43067		float64x1_t __reint_255 = __s2_255; \
43068		uint64x2_t __reint1_255 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_255, __p3_255), vgetq_lane_u64((uint64x2_t ) &__reint_255, __p3_255)}; \
43069		__ret_255 = vcmlaq_rot90_f64(__s0_255, __s1_255, (float64x2_t ) &__reint1_255); \
43070		__ret_255; \
43071		})
43072	43430	#else
43073		#define vcmlaq_rot90_lane_f64(__p0_256, __p1_256, __p2_256, __p3_256) __extension__ ({ \
43074		float64x2_t __ret_256; \
43075		float64x2_t __s0_256 = __p0_256; \
43076		float64x2_t __s1_256 = __p1_256; \
43077		float64x1_t __s2_256 = __p2_256; \
43078		float64x2_t __rev0_256; __rev0_256 = __builtin_shufflevector(__s0_256, __s0_256, 1, 0); \
43079		float64x2_t __rev1_256; __rev1_256 = __builtin_shufflevector(__s1_256, __s1_256, 1, 0); \
43080		float64x1_t __reint_256 = __s2_256; \
43081		uint64x2_t __reint1_256 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_256, __p3_256), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_256, __p3_256)}; \
43082		__ret_256 = __noswap_vcmlaq_rot90_f64(__rev0_256, __rev1_256, (float64x2_t ) &__reint1_256); \
43083		__ret_256 = __builtin_shufflevector(__ret_256, __ret_256, 1, 0); \
43084		__ret_256; \
43085		})
	43431	__ai uint32x4_t vcgtzq_f32(float32x4_t __p0) {
	43432	uint32x4_t __ret;
	43433	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43434	__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 50);
	43435	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43436	return __ret;
	43437	}
43086	43438	#endif
43087	43439
43088	43440	#ifdef __LITTLE_ENDIAN__
43089		#define vcmla_rot90_laneq_f64(__p0_257, __p1_257, __p2_257, __p3_257) __extension__ ({ \
43090		float64x1_t __ret_257; \
43091		float64x1_t __s0_257 = __p0_257; \
43092		float64x1_t __s1_257 = __p1_257; \
43093		float64x2_t __s2_257 = __p2_257; \
43094		float64x2_t __reint_257 = __s2_257; \
43095		uint64x2_t __reint1_257 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_257, __p3_257), vgetq_lane_u64((uint64x2_t ) &__reint_257, __p3_257)}; \
43096		__ret_257 = vcmla_rot90_f64(__s0_257, __s1_257, (float64x1_t ) &__reint1_257); \
43097		__ret_257; \
43098		})
	43441	__ai uint32x4_t vcgtzq_s32(int32x4_t __p0) {
	43442	uint32x4_t __ret;
	43443	__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 50);
	43444	return __ret;
	43445	}
43099	43446	#else
43100		#define vcmla_rot90_laneq_f64(__p0_258, __p1_258, __p2_258, __p3_258) __extension__ ({ \
43101		float64x1_t __ret_258; \
43102		float64x1_t __s0_258 = __p0_258; \
43103		float64x1_t __s1_258 = __p1_258; \
43104		float64x2_t __s2_258 = __p2_258; \
43105		float64x2_t __rev2_258; __rev2_258 = __builtin_shufflevector(__s2_258, __s2_258, 1, 0); \
43106		float64x2_t __reint_258 = __rev2_258; \
43107		uint64x2_t __reint1_258 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_258, __p3_258), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_258, __p3_258)}; \
43108		__ret_258 = vcmla_rot90_f64(__s0_258, __s1_258, (float64x1_t ) &__reint1_258); \
43109		__ret_258; \
43110		})
	43447	__ai uint32x4_t vcgtzq_s32(int32x4_t __p0) {
	43448	uint32x4_t __ret;
	43449	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43450	__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 50);
	43451	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43452	return __ret;
	43453	}
43111	43454	#endif
43112	43455
43113	43456	#ifdef __LITTLE_ENDIAN__
43114		#define vcmlaq_rot90_laneq_f64(__p0_259, __p1_259, __p2_259, __p3_259) __extension__ ({ \
43115		float64x2_t __ret_259; \
43116		float64x2_t __s0_259 = __p0_259; \
43117		float64x2_t __s1_259 = __p1_259; \
43118		float64x2_t __s2_259 = __p2_259; \
43119		float64x2_t __reint_259 = __s2_259; \
43120		uint64x2_t __reint1_259 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_259, __p3_259), vgetq_lane_u64((uint64x2_t ) &__reint_259, __p3_259)}; \
43121		__ret_259 = vcmlaq_rot90_f64(__s0_259, __s1_259, (float64x2_t ) &__reint1_259); \
43122		__ret_259; \
43123		})
	43457	__ai uint64x2_t vcgtzq_s64(int64x2_t __p0) {
	43458	uint64x2_t __ret;
	43459	__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 51);
	43460	return __ret;
	43461	}
43124	43462	#else
43125		#define vcmlaq_rot90_laneq_f64(__p0_260, __p1_260, __p2_260, __p3_260) __extension__ ({ \
43126		float64x2_t __ret_260; \
43127		float64x2_t __s0_260 = __p0_260; \
43128		float64x2_t __s1_260 = __p1_260; \
43129		float64x2_t __s2_260 = __p2_260; \
43130		float64x2_t __rev0_260; __rev0_260 = __builtin_shufflevector(__s0_260, __s0_260, 1, 0); \
43131		float64x2_t __rev1_260; __rev1_260 = __builtin_shufflevector(__s1_260, __s1_260, 1, 0); \
43132		float64x2_t __rev2_260; __rev2_260 = __builtin_shufflevector(__s2_260, __s2_260, 1, 0); \
43133		float64x2_t __reint_260 = __rev2_260; \
43134		uint64x2_t __reint1_260 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_260, __p3_260), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_260, __p3_260)}; \
43135		__ret_260 = __noswap_vcmlaq_rot90_f64(__rev0_260, __rev1_260, (float64x2_t ) &__reint1_260); \
43136		__ret_260 = __builtin_shufflevector(__ret_260, __ret_260, 1, 0); \
43137		__ret_260; \
43138		})
	43463	__ai uint64x2_t vcgtzq_s64(int64x2_t __p0) {
	43464	uint64x2_t __ret;
	43465	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43466	__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 51);
	43467	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43468	return __ret;
	43469	}
43139	43470	#endif
43140	43471
43141		#endif
43142		#if defined(__ARM_FEATURE_DOTPROD)
43143	43472	#ifdef __LITTLE_ENDIAN__
43144		__ai uint32x4_t vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
43145		uint32x4_t __ret;
43146		__ret = (uint32x4_t) __builtin_neon_vdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	43473	__ai uint16x8_t vcgtzq_s16(int16x8_t __p0) {
	43474	uint16x8_t __ret;
	43475	__ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 49);
43147	43476	return __ret;
43148	43477	}
43149	43478	#else
43150		__ai uint32x4_t vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
43151		uint32x4_t __ret;
43152		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43153		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43154		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43155		__ret = (uint32x4_t) __builtin_neon_vdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
43156		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43157		return __ret;
43158		}
43159		__ai uint32x4_t __noswap_vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
43160		uint32x4_t __ret;
43161		__ret = (uint32x4_t) __builtin_neon_vdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	43479	__ai uint16x8_t vcgtzq_s16(int16x8_t __p0) {
	43480	uint16x8_t __ret;
	43481	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	43482	__ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 49);
	43483	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43162	43484	return __ret;
43163	43485	}
43164	43486	#endif
43165	43487
43166	43488	#ifdef __LITTLE_ENDIAN__
43167		__ai int32x4_t vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
43168		int32x4_t __ret;
43169		__ret = (int32x4_t) __builtin_neon_vdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
	43489	__ai uint8x8_t vcgtz_s8(int8x8_t __p0) {
	43490	uint8x8_t __ret;
	43491	__ret = (uint8x8_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 16);
43170	43492	return __ret;
43171	43493	}
43172	43494	#else
43173		__ai int32x4_t vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
43174		int32x4_t __ret;
43175		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43176		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43177		int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43178		__ret = (int32x4_t) __builtin_neon_vdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
43179		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43180		return __ret;
43181		}
43182		__ai int32x4_t __noswap_vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
43183		int32x4_t __ret;
43184		__ret = (int32x4_t) __builtin_neon_vdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
	43495	__ai uint8x8_t vcgtz_s8(int8x8_t __p0) {
	43496	uint8x8_t __ret;
	43497	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	43498	__ret = (uint8x8_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 16);
	43499	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43185	43500	return __ret;
43186	43501	}
43187	43502	#endif
43188	43503
	43504	__ai uint64x1_t vcgtz_f64(float64x1_t __p0) {
	43505	uint64x1_t __ret;
	43506	__ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
	43507	return __ret;
	43508	}
43189	43509	#ifdef __LITTLE_ENDIAN__
43190		__ai uint32x2_t vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
	43510	__ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
43191	43511	uint32x2_t __ret;
43192		__ret = (uint32x2_t) __builtin_neon_vdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 18);
	43512	__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 18);
43193	43513	return __ret;
43194	43514	}
43195	43515	#else
43196		__ai uint32x2_t vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
	43516	__ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
43197	43517	uint32x2_t __ret;
43198		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43199		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43200		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43201		__ret = (uint32x2_t) __builtin_neon_vdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 18);
	43518	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43519	__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 18);
43202	43520	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43203	43521	return __ret;
43204	43522	}
43205		__ai uint32x2_t __noswap_vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
43206		uint32x2_t __ret;
43207		__ret = (uint32x2_t) __builtin_neon_vdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 18);
43208		return __ret;
43209		}
43210	43523	#endif
43211	43524
43212	43525	#ifdef __LITTLE_ENDIAN__
43213		__ai int32x2_t vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
43214		int32x2_t __ret;
43215		__ret = (int32x2_t) __builtin_neon_vdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
	43526	__ai uint32x2_t vcgtz_s32(int32x2_t __p0) {
	43527	uint32x2_t __ret;
	43528	__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 18);
43216	43529	return __ret;
43217	43530	}
43218	43531	#else
43219		__ai int32x2_t vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
43220		int32x2_t __ret;
	43532	__ai uint32x2_t vcgtz_s32(int32x2_t __p0) {
	43533	uint32x2_t __ret;
43221	43534	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43222		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43223		int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43224		__ret = (int32x2_t) __builtin_neon_vdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
	43535	__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 18);
43225	43536	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43226	43537	return __ret;
43227	43538	}
43228		__ai int32x2_t __noswap_vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
43229		int32x2_t __ret;
43230		__ret = (int32x2_t) __builtin_neon_vdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
43231		return __ret;
43232		}
43233		#endif
43234
43235		#ifdef __LITTLE_ENDIAN__
43236		#define vdotq_lane_u32(__p0_261, __p1_261, __p2_261, __p3_261) __extension__ ({ \
43237		uint32x4_t __ret_261; \
43238		uint32x4_t __s0_261 = __p0_261; \
43239		uint8x16_t __s1_261 = __p1_261; \
43240		uint8x8_t __s2_261 = __p2_261; \
43241		uint8x8_t __reint_261 = __s2_261; \
43242		uint32x4_t __reint1_261 = splatq_lane_u32((uint32x2_t ) &__reint_261, __p3_261); \
43243		__ret_261 = vdotq_u32(__s0_261, __s1_261, (uint8x16_t ) &__reint1_261); \
43244		__ret_261; \
43245		})
43246		#else
43247		#define vdotq_lane_u32(__p0_262, __p1_262, __p2_262, __p3_262) __extension__ ({ \
43248		uint32x4_t __ret_262; \
43249		uint32x4_t __s0_262 = __p0_262; \
43250		uint8x16_t __s1_262 = __p1_262; \
43251		uint8x8_t __s2_262 = __p2_262; \
43252		uint32x4_t __rev0_262; __rev0_262 = __builtin_shufflevector(__s0_262, __s0_262, 3, 2, 1, 0); \
43253		uint8x16_t __rev1_262; __rev1_262 = __builtin_shufflevector(__s1_262, __s1_262, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43254		uint8x8_t __rev2_262; __rev2_262 = __builtin_shufflevector(__s2_262, __s2_262, 7, 6, 5, 4, 3, 2, 1, 0); \
43255		uint8x8_t __reint_262 = __rev2_262; \
43256		uint32x4_t __reint1_262 = __noswap_splatq_lane_u32((uint32x2_t ) &__reint_262, __p3_262); \
43257		__ret_262 = __noswap_vdotq_u32(__rev0_262, __rev1_262, (uint8x16_t ) &__reint1_262); \
43258		__ret_262 = __builtin_shufflevector(__ret_262, __ret_262, 3, 2, 1, 0); \
43259		__ret_262; \
43260		})
43261		#endif
43262
43263		#ifdef __LITTLE_ENDIAN__
43264		#define vdotq_lane_s32(__p0_263, __p1_263, __p2_263, __p3_263) __extension__ ({ \
43265		int32x4_t __ret_263; \
43266		int32x4_t __s0_263 = __p0_263; \
43267		int8x16_t __s1_263 = __p1_263; \
43268		int8x8_t __s2_263 = __p2_263; \
43269		int8x8_t __reint_263 = __s2_263; \
43270		int32x4_t __reint1_263 = splatq_lane_s32((int32x2_t ) &__reint_263, __p3_263); \
43271		__ret_263 = vdotq_s32(__s0_263, __s1_263, (int8x16_t ) &__reint1_263); \
43272		__ret_263; \
43273		})
43274		#else
43275		#define vdotq_lane_s32(__p0_264, __p1_264, __p2_264, __p3_264) __extension__ ({ \
43276		int32x4_t __ret_264; \
43277		int32x4_t __s0_264 = __p0_264; \
43278		int8x16_t __s1_264 = __p1_264; \
43279		int8x8_t __s2_264 = __p2_264; \
43280		int32x4_t __rev0_264; __rev0_264 = __builtin_shufflevector(__s0_264, __s0_264, 3, 2, 1, 0); \
43281		int8x16_t __rev1_264; __rev1_264 = __builtin_shufflevector(__s1_264, __s1_264, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43282		int8x8_t __rev2_264; __rev2_264 = __builtin_shufflevector(__s2_264, __s2_264, 7, 6, 5, 4, 3, 2, 1, 0); \
43283		int8x8_t __reint_264 = __rev2_264; \
43284		int32x4_t __reint1_264 = __noswap_splatq_lane_s32((int32x2_t ) &__reint_264, __p3_264); \
43285		__ret_264 = __noswap_vdotq_s32(__rev0_264, __rev1_264, (int8x16_t ) &__reint1_264); \
43286		__ret_264 = __builtin_shufflevector(__ret_264, __ret_264, 3, 2, 1, 0); \
43287		__ret_264; \
43288		})
43289	43539	#endif
43290	43540
	43541	__ai uint64x1_t vcgtz_s64(int64x1_t __p0) {
	43542	uint64x1_t __ret;
	43543	__ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
	43544	return __ret;
	43545	}
43291	43546	#ifdef __LITTLE_ENDIAN__
43292		#define vdot_lane_u32(__p0_265, __p1_265, __p2_265, __p3_265) __extension__ ({ \
43293		uint32x2_t __ret_265; \
43294		uint32x2_t __s0_265 = __p0_265; \
43295		uint8x8_t __s1_265 = __p1_265; \
43296		uint8x8_t __s2_265 = __p2_265; \
43297		uint8x8_t __reint_265 = __s2_265; \
43298		uint32x2_t __reint1_265 = splat_lane_u32((uint32x2_t ) &__reint_265, __p3_265); \
43299		__ret_265 = vdot_u32(__s0_265, __s1_265, (uint8x8_t ) &__reint1_265); \
43300		__ret_265; \
43301		})
	43547	__ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
	43548	uint16x4_t __ret;
	43549	__ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 17);
	43550	return __ret;
	43551	}
43302	43552	#else
43303		#define vdot_lane_u32(__p0_266, __p1_266, __p2_266, __p3_266) __extension__ ({ \
43304		uint32x2_t __ret_266; \
43305		uint32x2_t __s0_266 = __p0_266; \
43306		uint8x8_t __s1_266 = __p1_266; \
43307		uint8x8_t __s2_266 = __p2_266; \
43308		uint32x2_t __rev0_266; __rev0_266 = __builtin_shufflevector(__s0_266, __s0_266, 1, 0); \
43309		uint8x8_t __rev1_266; __rev1_266 = __builtin_shufflevector(__s1_266, __s1_266, 7, 6, 5, 4, 3, 2, 1, 0); \
43310		uint8x8_t __rev2_266; __rev2_266 = __builtin_shufflevector(__s2_266, __s2_266, 7, 6, 5, 4, 3, 2, 1, 0); \
43311		uint8x8_t __reint_266 = __rev2_266; \
43312		uint32x2_t __reint1_266 = __noswap_splat_lane_u32((uint32x2_t ) &__reint_266, __p3_266); \
43313		__ret_266 = __noswap_vdot_u32(__rev0_266, __rev1_266, (uint8x8_t ) &__reint1_266); \
43314		__ret_266 = __builtin_shufflevector(__ret_266, __ret_266, 1, 0); \
43315		__ret_266; \
43316		})
	43553	__ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
	43554	uint16x4_t __ret;
	43555	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43556	__ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 17);
	43557	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43558	return __ret;
	43559	}
43317	43560	#endif
43318	43561
	43562	__ai uint64_t vcgtzd_s64(int64_t __p0) {
	43563	uint64_t __ret;
	43564	__ret = (uint64_t) __builtin_neon_vcgtzd_s64(__p0);
	43565	return __ret;
	43566	}
	43567	__ai uint64_t vcgtzd_f64(float64_t __p0) {
	43568	uint64_t __ret;
	43569	__ret = (uint64_t) __builtin_neon_vcgtzd_f64(__p0);
	43570	return __ret;
	43571	}
	43572	__ai uint32_t vcgtzs_f32(float32_t __p0) {
	43573	uint32_t __ret;
	43574	__ret = (uint32_t) __builtin_neon_vcgtzs_f32(__p0);
	43575	return __ret;
	43576	}
43319	43577	#ifdef __LITTLE_ENDIAN__
43320		#define vdot_lane_s32(__p0_267, __p1_267, __p2_267, __p3_267) __extension__ ({ \
43321		int32x2_t __ret_267; \
43322		int32x2_t __s0_267 = __p0_267; \
43323		int8x8_t __s1_267 = __p1_267; \
43324		int8x8_t __s2_267 = __p2_267; \
43325		int8x8_t __reint_267 = __s2_267; \
43326		int32x2_t __reint1_267 = splat_lane_s32((int32x2_t ) &__reint_267, __p3_267); \
43327		__ret_267 = vdot_s32(__s0_267, __s1_267, (int8x8_t ) &__reint1_267); \
43328		__ret_267; \
43329		})
	43578	__ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43579	uint64x2_t __ret;
	43580	__ret = (uint64x2_t)(__p0 <= __p1);
	43581	return __ret;
	43582	}
43330	43583	#else
43331		#define vdot_lane_s32(__p0_268, __p1_268, __p2_268, __p3_268) __extension__ ({ \
43332		int32x2_t __ret_268; \
43333		int32x2_t __s0_268 = __p0_268; \
43334		int8x8_t __s1_268 = __p1_268; \
43335		int8x8_t __s2_268 = __p2_268; \
43336		int32x2_t __rev0_268; __rev0_268 = __builtin_shufflevector(__s0_268, __s0_268, 1, 0); \
43337		int8x8_t __rev1_268; __rev1_268 = __builtin_shufflevector(__s1_268, __s1_268, 7, 6, 5, 4, 3, 2, 1, 0); \
43338		int8x8_t __rev2_268; __rev2_268 = __builtin_shufflevector(__s2_268, __s2_268, 7, 6, 5, 4, 3, 2, 1, 0); \
43339		int8x8_t __reint_268 = __rev2_268; \
43340		int32x2_t __reint1_268 = __noswap_splat_lane_s32((int32x2_t ) &__reint_268, __p3_268); \
43341		__ret_268 = __noswap_vdot_s32(__rev0_268, __rev1_268, (int8x8_t ) &__reint1_268); \
43342		__ret_268 = __builtin_shufflevector(__ret_268, __ret_268, 1, 0); \
43343		__ret_268; \
43344		})
	43584	__ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43585	uint64x2_t __ret;
	43586	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43587	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43588	__ret = (uint64x2_t)(__rev0 <= __rev1);
	43589	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43590	return __ret;
	43591	}
43345	43592	#endif
43346	43593
43347		#endif
43348		#if defined(__ARM_FEATURE_DOTPROD) && defined(__aarch64__)
43349	43594	#ifdef __LITTLE_ENDIAN__
43350		#define vdotq_laneq_u32(__p0_269, __p1_269, __p2_269, __p3_269) __extension__ ({ \
43351		uint32x4_t __ret_269; \
43352		uint32x4_t __s0_269 = __p0_269; \
43353		uint8x16_t __s1_269 = __p1_269; \
43354		uint8x16_t __s2_269 = __p2_269; \
43355		uint8x16_t __reint_269 = __s2_269; \
43356		uint32x4_t __reint1_269 = splatq_laneq_u32((uint32x4_t ) &__reint_269, __p3_269); \
43357		__ret_269 = vdotq_u32(__s0_269, __s1_269, (uint8x16_t ) &__reint1_269); \
43358		__ret_269; \
43359		})
	43595	__ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
	43596	uint64x2_t __ret;
	43597	__ret = (uint64x2_t)(__p0 <= __p1);
	43598	return __ret;
	43599	}
43360	43600	#else
43361		#define vdotq_laneq_u32(__p0_270, __p1_270, __p2_270, __p3_270) __extension__ ({ \
43362		uint32x4_t __ret_270; \
43363		uint32x4_t __s0_270 = __p0_270; \
43364		uint8x16_t __s1_270 = __p1_270; \
43365		uint8x16_t __s2_270 = __p2_270; \
43366		uint32x4_t __rev0_270; __rev0_270 = __builtin_shufflevector(__s0_270, __s0_270, 3, 2, 1, 0); \
43367		uint8x16_t __rev1_270; __rev1_270 = __builtin_shufflevector(__s1_270, __s1_270, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43368		uint8x16_t __rev2_270; __rev2_270 = __builtin_shufflevector(__s2_270, __s2_270, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43369		uint8x16_t __reint_270 = __rev2_270; \
43370		uint32x4_t __reint1_270 = __noswap_splatq_laneq_u32((uint32x4_t ) &__reint_270, __p3_270); \
43371		__ret_270 = __noswap_vdotq_u32(__rev0_270, __rev1_270, (uint8x16_t ) &__reint1_270); \
43372		__ret_270 = __builtin_shufflevector(__ret_270, __ret_270, 3, 2, 1, 0); \
43373		__ret_270; \
43374		})
	43601	__ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
	43602	uint64x2_t __ret;
	43603	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43604	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43605	__ret = (uint64x2_t)(__rev0 <= __rev1);
	43606	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43607	return __ret;
	43608	}
43375	43609	#endif
43376	43610
43377	43611	#ifdef __LITTLE_ENDIAN__
43378		#define vdotq_laneq_s32(__p0_271, __p1_271, __p2_271, __p3_271) __extension__ ({ \
43379		int32x4_t __ret_271; \
43380		int32x4_t __s0_271 = __p0_271; \
43381		int8x16_t __s1_271 = __p1_271; \
43382		int8x16_t __s2_271 = __p2_271; \
43383		int8x16_t __reint_271 = __s2_271; \
43384		int32x4_t __reint1_271 = splatq_laneq_s32((int32x4_t ) &__reint_271, __p3_271); \
43385		__ret_271 = vdotq_s32(__s0_271, __s1_271, (int8x16_t ) &__reint1_271); \
43386		__ret_271; \
43387		})
	43612	__ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
	43613	uint64x2_t __ret;
	43614	__ret = (uint64x2_t)(__p0 <= __p1);
	43615	return __ret;
	43616	}
43388	43617	#else
43389		#define vdotq_laneq_s32(__p0_272, __p1_272, __p2_272, __p3_272) __extension__ ({ \
43390		int32x4_t __ret_272; \
43391		int32x4_t __s0_272 = __p0_272; \
43392		int8x16_t __s1_272 = __p1_272; \
43393		int8x16_t __s2_272 = __p2_272; \
43394		int32x4_t __rev0_272; __rev0_272 = __builtin_shufflevector(__s0_272, __s0_272, 3, 2, 1, 0); \
43395		int8x16_t __rev1_272; __rev1_272 = __builtin_shufflevector(__s1_272, __s1_272, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43396		int8x16_t __rev2_272; __rev2_272 = __builtin_shufflevector(__s2_272, __s2_272, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43397		int8x16_t __reint_272 = __rev2_272; \
43398		int32x4_t __reint1_272 = __noswap_splatq_laneq_s32((int32x4_t ) &__reint_272, __p3_272); \
43399		__ret_272 = __noswap_vdotq_s32(__rev0_272, __rev1_272, (int8x16_t ) &__reint1_272); \
43400		__ret_272 = __builtin_shufflevector(__ret_272, __ret_272, 3, 2, 1, 0); \
43401		__ret_272; \
43402		})
	43618	__ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
	43619	uint64x2_t __ret;
	43620	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43621	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43622	__ret = (uint64x2_t)(__rev0 <= __rev1);
	43623	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43624	return __ret;
	43625	}
43403	43626	#endif
43404	43627
	43628	__ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
	43629	uint64x1_t __ret;
	43630	__ret = (uint64x1_t)(__p0 <= __p1);
	43631	return __ret;
	43632	}
	43633	__ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
	43634	uint64x1_t __ret;
	43635	__ret = (uint64x1_t)(__p0 <= __p1);
	43636	return __ret;
	43637	}
	43638	__ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
	43639	uint64x1_t __ret;
	43640	__ret = (uint64x1_t)(__p0 <= __p1);
	43641	return __ret;
	43642	}
	43643	__ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
	43644	uint64_t __ret;
	43645	__ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
	43646	return __ret;
	43647	}
	43648	__ai uint64_t vcled_s64(int64_t __p0, int64_t __p1) {
	43649	uint64_t __ret;
	43650	__ret = (uint64_t) __builtin_neon_vcled_s64(__p0, __p1);
	43651	return __ret;
	43652	}
	43653	__ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
	43654	uint64_t __ret;
	43655	__ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
	43656	return __ret;
	43657	}
	43658	__ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
	43659	uint32_t __ret;
	43660	__ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
	43661	return __ret;
	43662	}
43405	43663	#ifdef __LITTLE_ENDIAN__
43406		#define vdot_laneq_u32(__p0_273, __p1_273, __p2_273, __p3_273) __extension__ ({ \
43407		uint32x2_t __ret_273; \
43408		uint32x2_t __s0_273 = __p0_273; \
43409		uint8x8_t __s1_273 = __p1_273; \
43410		uint8x16_t __s2_273 = __p2_273; \
43411		uint8x16_t __reint_273 = __s2_273; \
43412		uint32x2_t __reint1_273 = splat_laneq_u32((uint32x4_t ) &__reint_273, __p3_273); \
43413		__ret_273 = vdot_u32(__s0_273, __s1_273, (uint8x8_t ) &__reint1_273); \
43414		__ret_273; \
43415		})
	43664	__ai uint8x16_t vclezq_s8(int8x16_t __p0) {
	43665	uint8x16_t __ret;
	43666	__ret = (uint8x16_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 48);
	43667	return __ret;
	43668	}
43416	43669	#else
43417		#define vdot_laneq_u32(__p0_274, __p1_274, __p2_274, __p3_274) __extension__ ({ \
43418		uint32x2_t __ret_274; \
43419		uint32x2_t __s0_274 = __p0_274; \
43420		uint8x8_t __s1_274 = __p1_274; \
43421		uint8x16_t __s2_274 = __p2_274; \
43422		uint32x2_t __rev0_274; __rev0_274 = __builtin_shufflevector(__s0_274, __s0_274, 1, 0); \
43423		uint8x8_t __rev1_274; __rev1_274 = __builtin_shufflevector(__s1_274, __s1_274, 7, 6, 5, 4, 3, 2, 1, 0); \
43424		uint8x16_t __rev2_274; __rev2_274 = __builtin_shufflevector(__s2_274, __s2_274, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43425		uint8x16_t __reint_274 = __rev2_274; \
43426		uint32x2_t __reint1_274 = __noswap_splat_laneq_u32((uint32x4_t ) &__reint_274, __p3_274); \
43427		__ret_274 = __noswap_vdot_u32(__rev0_274, __rev1_274, (uint8x8_t ) &__reint1_274); \
43428		__ret_274 = __builtin_shufflevector(__ret_274, __ret_274, 1, 0); \
43429		__ret_274; \
43430		})
	43670	__ai uint8x16_t vclezq_s8(int8x16_t __p0) {
	43671	uint8x16_t __ret;
	43672	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	43673	__ret = (uint8x16_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 48);
	43674	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	43675	return __ret;
	43676	}
43431	43677	#endif
43432	43678
43433	43679	#ifdef __LITTLE_ENDIAN__
43434		#define vdot_laneq_s32(__p0_275, __p1_275, __p2_275, __p3_275) __extension__ ({ \
43435		int32x2_t __ret_275; \
43436		int32x2_t __s0_275 = __p0_275; \
43437		int8x8_t __s1_275 = __p1_275; \
43438		int8x16_t __s2_275 = __p2_275; \
43439		int8x16_t __reint_275 = __s2_275; \
43440		int32x2_t __reint1_275 = splat_laneq_s32((int32x4_t ) &__reint_275, __p3_275); \
43441		__ret_275 = vdot_s32(__s0_275, __s1_275, (int8x8_t ) &__reint1_275); \
43442		__ret_275; \
43443		})
	43680	__ai uint64x2_t vclezq_f64(float64x2_t __p0) {
	43681	uint64x2_t __ret;
	43682	__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 51);
	43683	return __ret;
	43684	}
43444	43685	#else
43445		#define vdot_laneq_s32(__p0_276, __p1_276, __p2_276, __p3_276) __extension__ ({ \
43446		int32x2_t __ret_276; \
43447		int32x2_t __s0_276 = __p0_276; \
43448		int8x8_t __s1_276 = __p1_276; \
43449		int8x16_t __s2_276 = __p2_276; \
43450		int32x2_t __rev0_276; __rev0_276 = __builtin_shufflevector(__s0_276, __s0_276, 1, 0); \
43451		int8x8_t __rev1_276; __rev1_276 = __builtin_shufflevector(__s1_276, __s1_276, 7, 6, 5, 4, 3, 2, 1, 0); \
43452		int8x16_t __rev2_276; __rev2_276 = __builtin_shufflevector(__s2_276, __s2_276, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
43453		int8x16_t __reint_276 = __rev2_276; \
43454		int32x2_t __reint1_276 = __noswap_splat_laneq_s32((int32x4_t ) &__reint_276, __p3_276); \
43455		__ret_276 = __noswap_vdot_s32(__rev0_276, __rev1_276, (int8x8_t ) &__reint1_276); \
43456		__ret_276 = __builtin_shufflevector(__ret_276, __ret_276, 1, 0); \
43457		__ret_276; \
43458		})
	43686	__ai uint64x2_t vclezq_f64(float64x2_t __p0) {
	43687	uint64x2_t __ret;
	43688	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43689	__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 51);
	43690	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43691	return __ret;
	43692	}
43459	43693	#endif
43460	43694
43461		#endif
43462		#if defined(__ARM_FEATURE_FMA)
43463	43695	#ifdef __LITTLE_ENDIAN__
43464		__ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43465		float32x4_t __ret;
43466		__ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43696	__ai uint32x4_t vclezq_f32(float32x4_t __p0) {
	43697	uint32x4_t __ret;
	43698	__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 50);
43467	43699	return __ret;
43468	43700	}
43469	43701	#else
43470		__ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43471		float32x4_t __ret;
	43702	__ai uint32x4_t vclezq_f32(float32x4_t __p0) {
	43703	uint32x4_t __ret;
43472	43704	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43473		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43474		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43475		__ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	43705	__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 50);
43476	43706	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43477	43707	return __ret;
43478	43708	}
43479		__ai float32x4_t __noswap_vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43480		float32x4_t __ret;
43481		__ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43709	#endif
	43710
	43711	#ifdef __LITTLE_ENDIAN__
	43712	__ai uint32x4_t vclezq_s32(int32x4_t __p0) {
	43713	uint32x4_t __ret;
	43714	__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 50);
	43715	return __ret;
	43716	}
	43717	#else
	43718	__ai uint32x4_t vclezq_s32(int32x4_t __p0) {
	43719	uint32x4_t __ret;
	43720	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43721	__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 50);
	43722	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43482	43723	return __ret;
43483	43724	}
43484	43725	#endif
43485	43726
43486	43727	#ifdef __LITTLE_ENDIAN__
43487		__ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43488		float32x2_t __ret;
43489		__ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	43728	__ai uint64x2_t vclezq_s64(int64x2_t __p0) {
	43729	uint64x2_t __ret;
	43730	__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 51);
43490	43731	return __ret;
43491	43732	}
43492	43733	#else
43493		__ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43494		float32x2_t __ret;
43495		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43496		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43497		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
43498		__ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	43734	__ai uint64x2_t vclezq_s64(int64x2_t __p0) {
	43735	uint64x2_t __ret;
	43736	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43737	__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 51);
43499	43738	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43500	43739	return __ret;
43501	43740	}
43502		__ai float32x2_t __noswap_vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43503		float32x2_t __ret;
43504		__ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
43505		return __ret;
43506		}
43507	43741	#endif
43508	43742
43509	43743	#ifdef __LITTLE_ENDIAN__
43510		__ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
43511		float32x4_t __ret;
43512		__ret = vfmaq_f32(__p0, __p1, (float32x4_t) {__p2, __p2, __p2, __p2});
	43744	__ai uint16x8_t vclezq_s16(int16x8_t __p0) {
	43745	uint16x8_t __ret;
	43746	__ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 49);
43513	43747	return __ret;
43514	43748	}
43515	43749	#else
43516		__ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
43517		float32x4_t __ret;
43518		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43519		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43520		__ret = __noswap_vfmaq_f32(__rev0, __rev1, (float32x4_t) {__p2, __p2, __p2, __p2});
43521		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43750	__ai uint16x8_t vclezq_s16(int16x8_t __p0) {
	43751	uint16x8_t __ret;
	43752	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	43753	__ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 49);
	43754	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43522	43755	return __ret;
43523	43756	}
43524	43757	#endif
43525	43758
43526	43759	#ifdef __LITTLE_ENDIAN__
43527		__ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
43528		float32x2_t __ret;
43529		__ret = vfma_f32(__p0, __p1, (float32x2_t) {__p2, __p2});
	43760	__ai uint8x8_t vclez_s8(int8x8_t __p0) {
	43761	uint8x8_t __ret;
	43762	__ret = (uint8x8_t) __builtin_neon_vclez_v((int8x8_t)__p0, 16);
43530	43763	return __ret;
43531	43764	}
43532	43765	#else
43533		__ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
43534		float32x2_t __ret;
43535		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43536		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43537		__ret = __noswap_vfma_f32(__rev0, __rev1, (float32x2_t) {__p2, __p2});
43538		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43766	__ai uint8x8_t vclez_s8(int8x8_t __p0) {
	43767	uint8x8_t __ret;
	43768	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	43769	__ret = (uint8x8_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 16);
	43770	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43539	43771	return __ret;
43540	43772	}
43541	43773	#endif
43542	43774
	43775	__ai uint64x1_t vclez_f64(float64x1_t __p0) {
	43776	uint64x1_t __ret;
	43777	__ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
	43778	return __ret;
	43779	}
43543	43780	#ifdef __LITTLE_ENDIAN__
43544		__ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43545		float32x4_t __ret;
43546		__ret = vfmaq_f32(__p0, -__p1, __p2);
	43781	__ai uint32x2_t vclez_f32(float32x2_t __p0) {
	43782	uint32x2_t __ret;
	43783	__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__p0, 18);
43547	43784	return __ret;
43548	43785	}
43549	43786	#else
43550		__ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43551		float32x4_t __ret;
43552		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43553		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43554		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43555		__ret = __noswap_vfmaq_f32(__rev0, -__rev1, __rev2);
43556		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43787	__ai uint32x2_t vclez_f32(float32x2_t __p0) {
	43788	uint32x2_t __ret;
	43789	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43790	__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 18);
	43791	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43557	43792	return __ret;
43558	43793	}
43559	43794	#endif
43560	43795
43561	43796	#ifdef __LITTLE_ENDIAN__
43562		__ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43563		float32x2_t __ret;
43564		__ret = vfma_f32(__p0, -__p1, __p2);
	43797	__ai uint32x2_t vclez_s32(int32x2_t __p0) {
	43798	uint32x2_t __ret;
	43799	__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__p0, 18);
43565	43800	return __ret;
43566	43801	}
43567	43802	#else
43568		__ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43569		float32x2_t __ret;
43570		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43571		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43572		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
43573		__ret = __noswap_vfma_f32(__rev0, -__rev1, __rev2);
	43803	__ai uint32x2_t vclez_s32(int32x2_t __p0) {
	43804	uint32x2_t __ret;
	43805	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43806	__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 18);
43574	43807	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43575	43808	return __ret;
43576	43809	}
43577	43810	#endif
43578	43811
43579		#endif
43580		#if defined(__ARM_FEATURE_FP16_FML) && defined(__aarch64__)
	43812	__ai uint64x1_t vclez_s64(int64x1_t __p0) {
	43813	uint64x1_t __ret;
	43814	__ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
	43815	return __ret;
	43816	}
43581	43817	#ifdef __LITTLE_ENDIAN__
43582		__ai float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43583		float32x4_t __ret;
43584		__ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43818	__ai uint16x4_t vclez_s16(int16x4_t __p0) {
	43819	uint16x4_t __ret;
	43820	__ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__p0, 17);
43585	43821	return __ret;
43586	43822	}
43587	43823	#else
43588		__ai float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43589		float32x4_t __ret;
43590		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43591		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43592		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43593		__ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	43824	__ai uint16x4_t vclez_s16(int16x4_t __p0) {
	43825	uint16x4_t __ret;
	43826	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43827	__ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 17);
43594	43828	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43595	43829	return __ret;
43596	43830	}
43597		__ai float32x4_t __noswap_vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43598		float32x4_t __ret;
43599		__ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
43600		return __ret;
43601		}
43602	43831	#endif
43603	43832
	43833	__ai uint64_t vclezd_s64(int64_t __p0) {
	43834	uint64_t __ret;
	43835	__ret = (uint64_t) __builtin_neon_vclezd_s64(__p0);
	43836	return __ret;
	43837	}
	43838	__ai uint64_t vclezd_f64(float64_t __p0) {
	43839	uint64_t __ret;
	43840	__ret = (uint64_t) __builtin_neon_vclezd_f64(__p0);
	43841	return __ret;
	43842	}
	43843	__ai uint32_t vclezs_f32(float32_t __p0) {
	43844	uint32_t __ret;
	43845	__ret = (uint32_t) __builtin_neon_vclezs_f32(__p0);
	43846	return __ret;
	43847	}
43604	43848	#ifdef __LITTLE_ENDIAN__
43605		__ai float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43606		float32x2_t __ret;
43607		__ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	43849	__ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43850	uint64x2_t __ret;
	43851	__ret = (uint64x2_t)(__p0 < __p1);
43608	43852	return __ret;
43609	43853	}
43610	43854	#else
43611		__ai float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43612		float32x2_t __ret;
43613		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43614		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43615		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43616		__ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	43855	__ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	43856	uint64x2_t __ret;
	43857	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43858	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43859	__ret = (uint64x2_t)(__rev0 < __rev1);
43617	43860	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43618	43861	return __ret;
43619	43862	}
43620		__ai float32x2_t __noswap_vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43621		float32x2_t __ret;
43622		__ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
43623		return __ret;
43624		}
43625	43863	#endif
43626	43864
43627	43865	#ifdef __LITTLE_ENDIAN__
43628		__ai float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43629		float32x4_t __ret;
43630		__ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43866	__ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
	43867	uint64x2_t __ret;
	43868	__ret = (uint64x2_t)(__p0 < __p1);
43631	43869	return __ret;
43632	43870	}
43633	43871	#else
43634		__ai float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43635		float32x4_t __ret;
43636		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43637		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43638		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43639		__ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
43640		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43641		return __ret;
43642		}
43643		__ai float32x4_t __noswap_vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43644		float32x4_t __ret;
43645		__ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43872	__ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
	43873	uint64x2_t __ret;
	43874	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43875	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43876	__ret = (uint64x2_t)(__rev0 < __rev1);
	43877	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43646	43878	return __ret;
43647	43879	}
43648	43880	#endif
43649	43881
43650	43882	#ifdef __LITTLE_ENDIAN__
43651		__ai float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43652		float32x2_t __ret;
43653		__ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	43883	__ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
	43884	uint64x2_t __ret;
	43885	__ret = (uint64x2_t)(__p0 < __p1);
43654	43886	return __ret;
43655	43887	}
43656	43888	#else
43657		__ai float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43658		float32x2_t __ret;
43659		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43660		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43661		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43662		__ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	43889	__ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
	43890	uint64x2_t __ret;
	43891	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43892	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	43893	__ret = (uint64x2_t)(__rev0 < __rev1);
43663	43894	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43664	43895	return __ret;
43665	43896	}
43666		__ai float32x2_t __noswap_vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43667		float32x2_t __ret;
43668		__ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
43669		return __ret;
43670		}
43671	43897	#endif
43672	43898
43673		#ifdef __LITTLE_ENDIAN__
43674		__ai float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43675		float32x4_t __ret;
43676		__ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43899	__ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
	43900	uint64x1_t __ret;
	43901	__ret = (uint64x1_t)(__p0 < __p1);
43677	43902	return __ret;
43678	43903	}
43679		#else
43680		__ai float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43681		float32x4_t __ret;
43682		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43683		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43684		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43685		__ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
43686		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43904	__ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
	43905	uint64x1_t __ret;
	43906	__ret = (uint64x1_t)(__p0 < __p1);
43687	43907	return __ret;
43688	43908	}
43689		__ai float32x4_t __noswap_vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43690		float32x4_t __ret;
43691		__ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43909	__ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
	43910	uint64x1_t __ret;
	43911	__ret = (uint64x1_t)(__p0 < __p1);
43692	43912	return __ret;
43693	43913	}
43694		#endif
43695
43696		#ifdef __LITTLE_ENDIAN__
43697		__ai float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43698		float32x2_t __ret;
43699		__ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	43914	__ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
	43915	uint64_t __ret;
	43916	__ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
43700	43917	return __ret;
43701	43918	}
43702		#else
43703		__ai float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43704		float32x2_t __ret;
43705		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43706		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43707		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43708		__ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
43709		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	43919	__ai uint64_t vcltd_s64(int64_t __p0, int64_t __p1) {
	43920	uint64_t __ret;
	43921	__ret = (uint64_t) __builtin_neon_vcltd_s64(__p0, __p1);
43710	43922	return __ret;
43711	43923	}
43712		__ai float32x2_t __noswap_vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43713		float32x2_t __ret;
43714		__ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	43924	__ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
	43925	uint64_t __ret;
	43926	__ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
43715	43927	return __ret;
43716	43928	}
43717		#endif
43718
43719		#ifdef __LITTLE_ENDIAN__
43720		__ai float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43721		float32x4_t __ret;
43722		__ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43929	__ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
	43930	uint32_t __ret;
	43931	__ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
43723	43932	return __ret;
43724	43933	}
43725		#else
43726		__ai float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43727		float32x4_t __ret;
43728		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43729		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43730		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43731		__ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
43732		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	43934	#ifdef __LITTLE_ENDIAN__
	43935	__ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
	43936	uint8x16_t __ret;
	43937	__ret = (uint8x16_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 48);
43733	43938	return __ret;
43734	43939	}
43735		__ai float32x4_t __noswap_vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43736		float32x4_t __ret;
43737		__ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	43940	#else
	43941	__ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
	43942	uint8x16_t __ret;
	43943	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	43944	__ret = (uint8x16_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 48);
	43945	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43738	43946	return __ret;
43739	43947	}
43740	43948	#endif
43741	43949
43742	43950	#ifdef __LITTLE_ENDIAN__
43743		__ai float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43744		float32x2_t __ret;
43745		__ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	43951	__ai uint64x2_t vcltzq_f64(float64x2_t __p0) {
	43952	uint64x2_t __ret;
	43953	__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 51);
43746	43954	return __ret;
43747	43955	}
43748	43956	#else
43749		__ai float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43750		float32x2_t __ret;
43751		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43752		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43753		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43754		__ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	43957	__ai uint64x2_t vcltzq_f64(float64x2_t __p0) {
	43958	uint64x2_t __ret;
	43959	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	43960	__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 51);
43755	43961	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43756	43962	return __ret;
43757	43963	}
43758		__ai float32x2_t __noswap_vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43759		float32x2_t __ret;
43760		__ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
43761		return __ret;
43762		}
43763	43964	#endif
43764	43965
43765		#endif
43766		#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
43767	43966	#ifdef __LITTLE_ENDIAN__
43768		__ai float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
43769		float16x8_t __ret;
43770		__ret = (float16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	43967	__ai uint32x4_t vcltzq_f32(float32x4_t __p0) {
	43968	uint32x4_t __ret;
	43969	__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 50);
43771	43970	return __ret;
43772	43971	}
43773	43972	#else
43774		__ai float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
43775		float16x8_t __ret;
43776		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43777		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43778		__ret = (float16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
43779		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	43973	__ai uint32x4_t vcltzq_f32(float32x4_t __p0) {
	43974	uint32x4_t __ret;
	43975	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43976	__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 50);
	43977	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43780	43978	return __ret;
43781	43979	}
43782	43980	#endif
43783	43981
43784	43982	#ifdef __LITTLE_ENDIAN__
43785		__ai float16x4_t vabd_f16(float16x4_t __p0, float16x4_t __p1) {
43786		float16x4_t __ret;
43787		__ret = (float16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	43983	__ai uint32x4_t vcltzq_s32(int32x4_t __p0) {
	43984	uint32x4_t __ret;
	43985	__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 50);
43788	43986	return __ret;
43789	43987	}
43790	43988	#else
43791		__ai float16x4_t vabd_f16(float16x4_t __p0, float16x4_t __p1) {
43792		float16x4_t __ret;
43793		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43794		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43795		__ret = (float16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	43989	__ai uint32x4_t vcltzq_s32(int32x4_t __p0) {
	43990	uint32x4_t __ret;
	43991	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	43992	__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 50);
43796	43993	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43797	43994	return __ret;
43798	43995	}
43799	43996	#endif
43800	43997
43801	43998	#ifdef __LITTLE_ENDIAN__
43802		__ai float16x8_t vabsq_f16(float16x8_t __p0) {
43803		float16x8_t __ret;
43804		__ret = (float16x8_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 40);
	43999	__ai uint64x2_t vcltzq_s64(int64x2_t __p0) {
	44000	uint64x2_t __ret;
	44001	__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 51);
43805	44002	return __ret;
43806	44003	}
43807	44004	#else
43808		__ai float16x8_t vabsq_f16(float16x8_t __p0) {
43809		float16x8_t __ret;
43810		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43811		__ret = (float16x8_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 40);
43812		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	44005	__ai uint64x2_t vcltzq_s64(int64x2_t __p0) {
	44006	uint64x2_t __ret;
	44007	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	44008	__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 51);
	44009	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43813	44010	return __ret;
43814	44011	}
43815	44012	#endif
43816	44013
43817	44014	#ifdef __LITTLE_ENDIAN__
43818		__ai float16x4_t vabs_f16(float16x4_t __p0) {
43819		float16x4_t __ret;
43820		__ret = (float16x4_t) __builtin_neon_vabs_v((int8x8_t)__p0, 8);
	44015	__ai uint16x8_t vcltzq_s16(int16x8_t __p0) {
	44016	uint16x8_t __ret;
	44017	__ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 49);
43821	44018	return __ret;
43822	44019	}
43823	44020	#else
43824		__ai float16x4_t vabs_f16(float16x4_t __p0) {
43825		float16x4_t __ret;
43826		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43827		__ret = (float16x4_t) __builtin_neon_vabs_v((int8x8_t)__rev0, 8);
43828		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	44021	__ai uint16x8_t vcltzq_s16(int16x8_t __p0) {
	44022	uint16x8_t __ret;
	44023	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	44024	__ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 49);
	44025	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43829	44026	return __ret;
43830	44027	}
43831	44028	#endif
43832	44029
43833	44030	#ifdef __LITTLE_ENDIAN__
43834		__ai float16x8_t vaddq_f16(float16x8_t __p0, float16x8_t __p1) {
43835		float16x8_t __ret;
43836		__ret = __p0 + __p1;
	44031	__ai uint8x8_t vcltz_s8(int8x8_t __p0) {
	44032	uint8x8_t __ret;
	44033	__ret = (uint8x8_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 16);
43837	44034	return __ret;
43838	44035	}
43839	44036	#else
43840		__ai float16x8_t vaddq_f16(float16x8_t __p0, float16x8_t __p1) {
43841		float16x8_t __ret;
43842		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43843		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43844		__ret = __rev0 + __rev1;
	44037	__ai uint8x8_t vcltz_s8(int8x8_t __p0) {
	44038	uint8x8_t __ret;
	44039	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	44040	__ret = (uint8x8_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 16);
43845	44041	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43846	44042	return __ret;
43847	44043	}
43848	44044	#endif
43849	44045
	44046	__ai uint64x1_t vcltz_f64(float64x1_t __p0) {
	44047	uint64x1_t __ret;
	44048	__ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
	44049	return __ret;
	44050	}
43850	44051	#ifdef __LITTLE_ENDIAN__
43851		__ai float16x4_t vadd_f16(float16x4_t __p0, float16x4_t __p1) {
43852		float16x4_t __ret;
43853		__ret = __p0 + __p1;
	44052	__ai uint32x2_t vcltz_f32(float32x2_t __p0) {
	44053	uint32x2_t __ret;
	44054	__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 18);
43854	44055	return __ret;
43855	44056	}
43856	44057	#else
43857		__ai float16x4_t vadd_f16(float16x4_t __p0, float16x4_t __p1) {
43858		float16x4_t __ret;
43859		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43860		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43861		__ret = __rev0 + __rev1;
43862		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	44058	__ai uint32x2_t vcltz_f32(float32x2_t __p0) {
	44059	uint32x2_t __ret;
	44060	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	44061	__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 18);
	44062	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43863	44063	return __ret;
43864	44064	}
43865	44065	#endif
43866	44066
43867	44067	#ifdef __LITTLE_ENDIAN__
43868		__ai float16x8_t vbslq_f16(uint16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
43869		float16x8_t __ret;
43870		__ret = (float16x8_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	44068	__ai uint32x2_t vcltz_s32(int32x2_t __p0) {
	44069	uint32x2_t __ret;
	44070	__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 18);
43871	44071	return __ret;
43872	44072	}
43873	44073	#else
43874		__ai float16x8_t vbslq_f16(uint16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
43875		float16x8_t __ret;
43876		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43877		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43878		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43879		__ret = (float16x8_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
43880		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	44074	__ai uint32x2_t vcltz_s32(int32x2_t __p0) {
	44075	uint32x2_t __ret;
	44076	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	44077	__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 18);
	44078	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43881	44079	return __ret;
43882	44080	}
43883	44081	#endif
43884	44082
	44083	__ai uint64x1_t vcltz_s64(int64x1_t __p0) {
	44084	uint64x1_t __ret;
	44085	__ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
	44086	return __ret;
	44087	}
43885	44088	#ifdef __LITTLE_ENDIAN__
43886		__ai float16x4_t vbsl_f16(uint16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
43887		float16x4_t __ret;
43888		__ret = (float16x4_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	44089	__ai uint16x4_t vcltz_s16(int16x4_t __p0) {
	44090	uint16x4_t __ret;
	44091	__ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 17);
43889	44092	return __ret;
43890	44093	}
43891	44094	#else
43892		__ai float16x4_t vbsl_f16(uint16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
43893		float16x4_t __ret;
43894		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43895		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43896		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43897		__ret = (float16x4_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
	44095	__ai uint16x4_t vcltz_s16(int16x4_t __p0) {
	44096	uint16x4_t __ret;
	44097	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	44098	__ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 17);
43898	44099	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43899	44100	return __ret;
43900	44101	}
43901	44102	#endif
43902	44103
	44104	__ai uint64_t vcltzd_s64(int64_t __p0) {
	44105	uint64_t __ret;
	44106	__ret = (uint64_t) __builtin_neon_vcltzd_s64(__p0);
	44107	return __ret;
	44108	}
	44109	__ai uint64_t vcltzd_f64(float64_t __p0) {
	44110	uint64_t __ret;
	44111	__ret = (uint64_t) __builtin_neon_vcltzd_f64(__p0);
	44112	return __ret;
	44113	}
	44114	__ai uint32_t vcltzs_f32(float32_t __p0) {
	44115	uint32_t __ret;
	44116	__ret = (uint32_t) __builtin_neon_vcltzs_f32(__p0);
	44117	return __ret;
	44118	}
43903	44119	#ifdef __LITTLE_ENDIAN__
43904		__ai uint16x8_t vcageq_f16(float16x8_t __p0, float16x8_t __p1) {
43905		uint16x8_t __ret;
43906		__ret = (uint16x8_t) __builtin_neon_vcageq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	44120	__ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
	44121	poly64x2_t __ret;
	44122	__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
43907	44123	return __ret;
43908	44124	}
43909	44125	#else
43910		__ai uint16x8_t vcageq_f16(float16x8_t __p0, float16x8_t __p1) {
43911		uint16x8_t __ret;
43912		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43913		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43914		__ret = (uint16x8_t) __builtin_neon_vcageq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
43915		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	44126	__ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
	44127	poly64x2_t __ret;
	44128	__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
	44129	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43916	44130	return __ret;
43917	44131	}
43918	44132	#endif
43919	44133
43920	44134	#ifdef __LITTLE_ENDIAN__
43921		__ai uint16x4_t vcage_f16(float16x4_t __p0, float16x4_t __p1) {
43922		uint16x4_t __ret;
43923		__ret = (uint16x4_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
	44135	__ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
	44136	float64x2_t __ret;
	44137	__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
43924	44138	return __ret;
43925	44139	}
43926	44140	#else
43927		__ai uint16x4_t vcage_f16(float16x4_t __p0, float16x4_t __p1) {
43928		uint16x4_t __ret;
43929		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43930		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43931		__ret = (uint16x4_t) __builtin_neon_vcage_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
43932		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43933		return __ret;
43934		}
	44141	__ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
	44142	float64x2_t __ret;
	44143	__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
	44144	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	44145	return __ret;
	44146	}
	44147	#endif
	44148
	44149	#ifdef __LITTLE_ENDIAN__
	44150	#define vcopyq_lane_p8(__p0_257, __p1_257, __p2_257, __p3_257) __extension__ ({ \
	44151	poly8x16_t __ret_257; \
	44152	poly8x16_t __s0_257 = __p0_257; \
	44153	poly8x8_t __s2_257 = __p2_257; \
	44154	__ret_257 = vsetq_lane_p8(vget_lane_p8(__s2_257, __p3_257), __s0_257, __p1_257); \
	44155	__ret_257; \
	44156	})
	44157	#else
	44158	#define vcopyq_lane_p8(__p0_258, __p1_258, __p2_258, __p3_258) __extension__ ({ \
	44159	poly8x16_t __ret_258; \
	44160	poly8x16_t __s0_258 = __p0_258; \
	44161	poly8x8_t __s2_258 = __p2_258; \
	44162	poly8x16_t __rev0_258; __rev0_258 = __builtin_shufflevector(__s0_258, __s0_258, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44163	poly8x8_t __rev2_258; __rev2_258 = __builtin_shufflevector(__s2_258, __s2_258, 7, 6, 5, 4, 3, 2, 1, 0); \
	44164	__ret_258 = __noswap_vsetq_lane_p8(__noswap_vget_lane_p8(__rev2_258, __p3_258), __rev0_258, __p1_258); \
	44165	__ret_258 = __builtin_shufflevector(__ret_258, __ret_258, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44166	__ret_258; \
	44167	})
	44168	#endif
	44169
	44170	#ifdef __LITTLE_ENDIAN__
	44171	#define vcopyq_lane_p16(__p0_259, __p1_259, __p2_259, __p3_259) __extension__ ({ \
	44172	poly16x8_t __ret_259; \
	44173	poly16x8_t __s0_259 = __p0_259; \
	44174	poly16x4_t __s2_259 = __p2_259; \
	44175	__ret_259 = vsetq_lane_p16(vget_lane_p16(__s2_259, __p3_259), __s0_259, __p1_259); \
	44176	__ret_259; \
	44177	})
	44178	#else
	44179	#define vcopyq_lane_p16(__p0_260, __p1_260, __p2_260, __p3_260) __extension__ ({ \
	44180	poly16x8_t __ret_260; \
	44181	poly16x8_t __s0_260 = __p0_260; \
	44182	poly16x4_t __s2_260 = __p2_260; \
	44183	poly16x8_t __rev0_260; __rev0_260 = __builtin_shufflevector(__s0_260, __s0_260, 7, 6, 5, 4, 3, 2, 1, 0); \
	44184	poly16x4_t __rev2_260; __rev2_260 = __builtin_shufflevector(__s2_260, __s2_260, 3, 2, 1, 0); \
	44185	__ret_260 = __noswap_vsetq_lane_p16(__noswap_vget_lane_p16(__rev2_260, __p3_260), __rev0_260, __p1_260); \
	44186	__ret_260 = __builtin_shufflevector(__ret_260, __ret_260, 7, 6, 5, 4, 3, 2, 1, 0); \
	44187	__ret_260; \
	44188	})
	44189	#endif
	44190
	44191	#ifdef __LITTLE_ENDIAN__
	44192	#define vcopyq_lane_u8(__p0_261, __p1_261, __p2_261, __p3_261) __extension__ ({ \
	44193	uint8x16_t __ret_261; \
	44194	uint8x16_t __s0_261 = __p0_261; \
	44195	uint8x8_t __s2_261 = __p2_261; \
	44196	__ret_261 = vsetq_lane_u8(vget_lane_u8(__s2_261, __p3_261), __s0_261, __p1_261); \
	44197	__ret_261; \
	44198	})
	44199	#else
	44200	#define vcopyq_lane_u8(__p0_262, __p1_262, __p2_262, __p3_262) __extension__ ({ \
	44201	uint8x16_t __ret_262; \
	44202	uint8x16_t __s0_262 = __p0_262; \
	44203	uint8x8_t __s2_262 = __p2_262; \
	44204	uint8x16_t __rev0_262; __rev0_262 = __builtin_shufflevector(__s0_262, __s0_262, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44205	uint8x8_t __rev2_262; __rev2_262 = __builtin_shufflevector(__s2_262, __s2_262, 7, 6, 5, 4, 3, 2, 1, 0); \
	44206	__ret_262 = __noswap_vsetq_lane_u8(__noswap_vget_lane_u8(__rev2_262, __p3_262), __rev0_262, __p1_262); \
	44207	__ret_262 = __builtin_shufflevector(__ret_262, __ret_262, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44208	__ret_262; \
	44209	})
	44210	#endif
	44211
	44212	#ifdef __LITTLE_ENDIAN__
	44213	#define vcopyq_lane_u32(__p0_263, __p1_263, __p2_263, __p3_263) __extension__ ({ \
	44214	uint32x4_t __ret_263; \
	44215	uint32x4_t __s0_263 = __p0_263; \
	44216	uint32x2_t __s2_263 = __p2_263; \
	44217	__ret_263 = vsetq_lane_u32(vget_lane_u32(__s2_263, __p3_263), __s0_263, __p1_263); \
	44218	__ret_263; \
	44219	})
	44220	#else
	44221	#define vcopyq_lane_u32(__p0_264, __p1_264, __p2_264, __p3_264) __extension__ ({ \
	44222	uint32x4_t __ret_264; \
	44223	uint32x4_t __s0_264 = __p0_264; \
	44224	uint32x2_t __s2_264 = __p2_264; \
	44225	uint32x4_t __rev0_264; __rev0_264 = __builtin_shufflevector(__s0_264, __s0_264, 3, 2, 1, 0); \
	44226	uint32x2_t __rev2_264; __rev2_264 = __builtin_shufflevector(__s2_264, __s2_264, 1, 0); \
	44227	__ret_264 = __noswap_vsetq_lane_u32(__noswap_vget_lane_u32(__rev2_264, __p3_264), __rev0_264, __p1_264); \
	44228	__ret_264 = __builtin_shufflevector(__ret_264, __ret_264, 3, 2, 1, 0); \
	44229	__ret_264; \
	44230	})
	44231	#endif
	44232
	44233	#ifdef __LITTLE_ENDIAN__
	44234	#define vcopyq_lane_u64(__p0_265, __p1_265, __p2_265, __p3_265) __extension__ ({ \
	44235	uint64x2_t __ret_265; \
	44236	uint64x2_t __s0_265 = __p0_265; \
	44237	uint64x1_t __s2_265 = __p2_265; \
	44238	__ret_265 = vsetq_lane_u64(vget_lane_u64(__s2_265, __p3_265), __s0_265, __p1_265); \
	44239	__ret_265; \
	44240	})
	44241	#else
	44242	#define vcopyq_lane_u64(__p0_266, __p1_266, __p2_266, __p3_266) __extension__ ({ \
	44243	uint64x2_t __ret_266; \
	44244	uint64x2_t __s0_266 = __p0_266; \
	44245	uint64x1_t __s2_266 = __p2_266; \
	44246	uint64x2_t __rev0_266; __rev0_266 = __builtin_shufflevector(__s0_266, __s0_266, 1, 0); \
	44247	__ret_266 = __noswap_vsetq_lane_u64(vget_lane_u64(__s2_266, __p3_266), __rev0_266, __p1_266); \
	44248	__ret_266 = __builtin_shufflevector(__ret_266, __ret_266, 1, 0); \
	44249	__ret_266; \
	44250	})
43935	44251	#endif
43936	44252
43937	44253	#ifdef __LITTLE_ENDIAN__
43938		__ai uint16x8_t vcagtq_f16(float16x8_t __p0, float16x8_t __p1) {
43939		uint16x8_t __ret;
43940		__ret = (uint16x8_t) __builtin_neon_vcagtq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
43941		return __ret;
43942		}
	44254	#define vcopyq_lane_u16(__p0_267, __p1_267, __p2_267, __p3_267) __extension__ ({ \
	44255	uint16x8_t __ret_267; \
	44256	uint16x8_t __s0_267 = __p0_267; \
	44257	uint16x4_t __s2_267 = __p2_267; \
	44258	__ret_267 = vsetq_lane_u16(vget_lane_u16(__s2_267, __p3_267), __s0_267, __p1_267); \
	44259	__ret_267; \
	44260	})
43943	44261	#else
43944		__ai uint16x8_t vcagtq_f16(float16x8_t __p0, float16x8_t __p1) {
43945		uint16x8_t __ret;
43946		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43947		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43948		__ret = (uint16x8_t) __builtin_neon_vcagtq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
43949		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43950		return __ret;
43951		}
	44262	#define vcopyq_lane_u16(__p0_268, __p1_268, __p2_268, __p3_268) __extension__ ({ \
	44263	uint16x8_t __ret_268; \
	44264	uint16x8_t __s0_268 = __p0_268; \
	44265	uint16x4_t __s2_268 = __p2_268; \
	44266	uint16x8_t __rev0_268; __rev0_268 = __builtin_shufflevector(__s0_268, __s0_268, 7, 6, 5, 4, 3, 2, 1, 0); \
	44267	uint16x4_t __rev2_268; __rev2_268 = __builtin_shufflevector(__s2_268, __s2_268, 3, 2, 1, 0); \
	44268	__ret_268 = __noswap_vsetq_lane_u16(__noswap_vget_lane_u16(__rev2_268, __p3_268), __rev0_268, __p1_268); \
	44269	__ret_268 = __builtin_shufflevector(__ret_268, __ret_268, 7, 6, 5, 4, 3, 2, 1, 0); \
	44270	__ret_268; \
	44271	})
43952	44272	#endif
43953	44273
43954	44274	#ifdef __LITTLE_ENDIAN__
43955		__ai uint16x4_t vcagt_f16(float16x4_t __p0, float16x4_t __p1) {
43956		uint16x4_t __ret;
43957		__ret = (uint16x4_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
43958		return __ret;
43959		}
	44275	#define vcopyq_lane_s8(__p0_269, __p1_269, __p2_269, __p3_269) __extension__ ({ \
	44276	int8x16_t __ret_269; \
	44277	int8x16_t __s0_269 = __p0_269; \
	44278	int8x8_t __s2_269 = __p2_269; \
	44279	__ret_269 = vsetq_lane_s8(vget_lane_s8(__s2_269, __p3_269), __s0_269, __p1_269); \
	44280	__ret_269; \
	44281	})
43960	44282	#else
43961		__ai uint16x4_t vcagt_f16(float16x4_t __p0, float16x4_t __p1) {
43962		uint16x4_t __ret;
43963		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43964		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43965		__ret = (uint16x4_t) __builtin_neon_vcagt_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
43966		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43967		return __ret;
43968		}
	44283	#define vcopyq_lane_s8(__p0_270, __p1_270, __p2_270, __p3_270) __extension__ ({ \
	44284	int8x16_t __ret_270; \
	44285	int8x16_t __s0_270 = __p0_270; \
	44286	int8x8_t __s2_270 = __p2_270; \
	44287	int8x16_t __rev0_270; __rev0_270 = __builtin_shufflevector(__s0_270, __s0_270, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44288	int8x8_t __rev2_270; __rev2_270 = __builtin_shufflevector(__s2_270, __s2_270, 7, 6, 5, 4, 3, 2, 1, 0); \
	44289	__ret_270 = __noswap_vsetq_lane_s8(__noswap_vget_lane_s8(__rev2_270, __p3_270), __rev0_270, __p1_270); \
	44290	__ret_270 = __builtin_shufflevector(__ret_270, __ret_270, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44291	__ret_270; \
	44292	})
43969	44293	#endif
43970	44294
43971	44295	#ifdef __LITTLE_ENDIAN__
43972		__ai uint16x8_t vcaleq_f16(float16x8_t __p0, float16x8_t __p1) {
43973		uint16x8_t __ret;
43974		__ret = (uint16x8_t) __builtin_neon_vcaleq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
43975		return __ret;
43976		}
	44296	#define vcopyq_lane_f32(__p0_271, __p1_271, __p2_271, __p3_271) __extension__ ({ \
	44297	float32x4_t __ret_271; \
	44298	float32x4_t __s0_271 = __p0_271; \
	44299	float32x2_t __s2_271 = __p2_271; \
	44300	__ret_271 = vsetq_lane_f32(vget_lane_f32(__s2_271, __p3_271), __s0_271, __p1_271); \
	44301	__ret_271; \
	44302	})
43977	44303	#else
43978		__ai uint16x8_t vcaleq_f16(float16x8_t __p0, float16x8_t __p1) {
43979		uint16x8_t __ret;
43980		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43981		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43982		__ret = (uint16x8_t) __builtin_neon_vcaleq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
43983		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43984		return __ret;
43985		}
	44304	#define vcopyq_lane_f32(__p0_272, __p1_272, __p2_272, __p3_272) __extension__ ({ \
	44305	float32x4_t __ret_272; \
	44306	float32x4_t __s0_272 = __p0_272; \
	44307	float32x2_t __s2_272 = __p2_272; \
	44308	float32x4_t __rev0_272; __rev0_272 = __builtin_shufflevector(__s0_272, __s0_272, 3, 2, 1, 0); \
	44309	float32x2_t __rev2_272; __rev2_272 = __builtin_shufflevector(__s2_272, __s2_272, 1, 0); \
	44310	__ret_272 = __noswap_vsetq_lane_f32(__noswap_vget_lane_f32(__rev2_272, __p3_272), __rev0_272, __p1_272); \
	44311	__ret_272 = __builtin_shufflevector(__ret_272, __ret_272, 3, 2, 1, 0); \
	44312	__ret_272; \
	44313	})
43986	44314	#endif
43987	44315
43988	44316	#ifdef __LITTLE_ENDIAN__
43989		__ai uint16x4_t vcale_f16(float16x4_t __p0, float16x4_t __p1) {
43990		uint16x4_t __ret;
43991		__ret = (uint16x4_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
43992		return __ret;
43993		}
	44317	#define vcopyq_lane_s32(__p0_273, __p1_273, __p2_273, __p3_273) __extension__ ({ \
	44318	int32x4_t __ret_273; \
	44319	int32x4_t __s0_273 = __p0_273; \
	44320	int32x2_t __s2_273 = __p2_273; \
	44321	__ret_273 = vsetq_lane_s32(vget_lane_s32(__s2_273, __p3_273), __s0_273, __p1_273); \
	44322	__ret_273; \
	44323	})
43994	44324	#else
43995		__ai uint16x4_t vcale_f16(float16x4_t __p0, float16x4_t __p1) {
43996		uint16x4_t __ret;
43997		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43998		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43999		__ret = (uint16x4_t) __builtin_neon_vcale_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
44000		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44001		return __ret;
44002		}
	44325	#define vcopyq_lane_s32(__p0_274, __p1_274, __p2_274, __p3_274) __extension__ ({ \
	44326	int32x4_t __ret_274; \
	44327	int32x4_t __s0_274 = __p0_274; \
	44328	int32x2_t __s2_274 = __p2_274; \
	44329	int32x4_t __rev0_274; __rev0_274 = __builtin_shufflevector(__s0_274, __s0_274, 3, 2, 1, 0); \
	44330	int32x2_t __rev2_274; __rev2_274 = __builtin_shufflevector(__s2_274, __s2_274, 1, 0); \
	44331	__ret_274 = __noswap_vsetq_lane_s32(__noswap_vget_lane_s32(__rev2_274, __p3_274), __rev0_274, __p1_274); \
	44332	__ret_274 = __builtin_shufflevector(__ret_274, __ret_274, 3, 2, 1, 0); \
	44333	__ret_274; \
	44334	})
44003	44335	#endif
44004	44336
44005	44337	#ifdef __LITTLE_ENDIAN__
44006		__ai uint16x8_t vcaltq_f16(float16x8_t __p0, float16x8_t __p1) {
44007		uint16x8_t __ret;
44008		__ret = (uint16x8_t) __builtin_neon_vcaltq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
44009		return __ret;
44010		}
	44338	#define vcopyq_lane_s64(__p0_275, __p1_275, __p2_275, __p3_275) __extension__ ({ \
	44339	int64x2_t __ret_275; \
	44340	int64x2_t __s0_275 = __p0_275; \
	44341	int64x1_t __s2_275 = __p2_275; \
	44342	__ret_275 = vsetq_lane_s64(vget_lane_s64(__s2_275, __p3_275), __s0_275, __p1_275); \
	44343	__ret_275; \
	44344	})
44011	44345	#else
44012		__ai uint16x8_t vcaltq_f16(float16x8_t __p0, float16x8_t __p1) {
44013		uint16x8_t __ret;
44014		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44015		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44016		__ret = (uint16x8_t) __builtin_neon_vcaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
44017		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44018		return __ret;
44019		}
	44346	#define vcopyq_lane_s64(__p0_276, __p1_276, __p2_276, __p3_276) __extension__ ({ \
	44347	int64x2_t __ret_276; \
	44348	int64x2_t __s0_276 = __p0_276; \
	44349	int64x1_t __s2_276 = __p2_276; \
	44350	int64x2_t __rev0_276; __rev0_276 = __builtin_shufflevector(__s0_276, __s0_276, 1, 0); \
	44351	__ret_276 = __noswap_vsetq_lane_s64(vget_lane_s64(__s2_276, __p3_276), __rev0_276, __p1_276); \
	44352	__ret_276 = __builtin_shufflevector(__ret_276, __ret_276, 1, 0); \
	44353	__ret_276; \
	44354	})
44020	44355	#endif
44021	44356
44022	44357	#ifdef __LITTLE_ENDIAN__
44023		__ai uint16x4_t vcalt_f16(float16x4_t __p0, float16x4_t __p1) {
44024		uint16x4_t __ret;
44025		__ret = (uint16x4_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
44026		return __ret;
44027		}
	44358	#define vcopyq_lane_s16(__p0_277, __p1_277, __p2_277, __p3_277) __extension__ ({ \
	44359	int16x8_t __ret_277; \
	44360	int16x8_t __s0_277 = __p0_277; \
	44361	int16x4_t __s2_277 = __p2_277; \
	44362	__ret_277 = vsetq_lane_s16(vget_lane_s16(__s2_277, __p3_277), __s0_277, __p1_277); \
	44363	__ret_277; \
	44364	})
44028	44365	#else
44029		__ai uint16x4_t vcalt_f16(float16x4_t __p0, float16x4_t __p1) {
44030		uint16x4_t __ret;
44031		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44032		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44033		__ret = (uint16x4_t) __builtin_neon_vcalt_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
44034		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44035		return __ret;
44036		}
	44366	#define vcopyq_lane_s16(__p0_278, __p1_278, __p2_278, __p3_278) __extension__ ({ \
	44367	int16x8_t __ret_278; \
	44368	int16x8_t __s0_278 = __p0_278; \
	44369	int16x4_t __s2_278 = __p2_278; \
	44370	int16x8_t __rev0_278; __rev0_278 = __builtin_shufflevector(__s0_278, __s0_278, 7, 6, 5, 4, 3, 2, 1, 0); \
	44371	int16x4_t __rev2_278; __rev2_278 = __builtin_shufflevector(__s2_278, __s2_278, 3, 2, 1, 0); \
	44372	__ret_278 = __noswap_vsetq_lane_s16(__noswap_vget_lane_s16(__rev2_278, __p3_278), __rev0_278, __p1_278); \
	44373	__ret_278 = __builtin_shufflevector(__ret_278, __ret_278, 7, 6, 5, 4, 3, 2, 1, 0); \
	44374	__ret_278; \
	44375	})
44037	44376	#endif
44038	44377
44039	44378	#ifdef __LITTLE_ENDIAN__
44040		__ai uint16x8_t vceqq_f16(float16x8_t __p0, float16x8_t __p1) {
44041		uint16x8_t __ret;
44042		__ret = (uint16x8_t)(__p0 == __p1);
44043		return __ret;
44044		}
	44379	#define vcopy_lane_p8(__p0_279, __p1_279, __p2_279, __p3_279) __extension__ ({ \
	44380	poly8x8_t __ret_279; \
	44381	poly8x8_t __s0_279 = __p0_279; \
	44382	poly8x8_t __s2_279 = __p2_279; \
	44383	__ret_279 = vset_lane_p8(vget_lane_p8(__s2_279, __p3_279), __s0_279, __p1_279); \
	44384	__ret_279; \
	44385	})
44045	44386	#else
44046		__ai uint16x8_t vceqq_f16(float16x8_t __p0, float16x8_t __p1) {
44047		uint16x8_t __ret;
44048		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44049		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44050		__ret = (uint16x8_t)(__rev0 == __rev1);
44051		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44052		return __ret;
44053		}
	44387	#define vcopy_lane_p8(__p0_280, __p1_280, __p2_280, __p3_280) __extension__ ({ \
	44388	poly8x8_t __ret_280; \
	44389	poly8x8_t __s0_280 = __p0_280; \
	44390	poly8x8_t __s2_280 = __p2_280; \
	44391	poly8x8_t __rev0_280; __rev0_280 = __builtin_shufflevector(__s0_280, __s0_280, 7, 6, 5, 4, 3, 2, 1, 0); \
	44392	poly8x8_t __rev2_280; __rev2_280 = __builtin_shufflevector(__s2_280, __s2_280, 7, 6, 5, 4, 3, 2, 1, 0); \
	44393	__ret_280 = __noswap_vset_lane_p8(__noswap_vget_lane_p8(__rev2_280, __p3_280), __rev0_280, __p1_280); \
	44394	__ret_280 = __builtin_shufflevector(__ret_280, __ret_280, 7, 6, 5, 4, 3, 2, 1, 0); \
	44395	__ret_280; \
	44396	})
44054	44397	#endif
44055	44398
44056	44399	#ifdef __LITTLE_ENDIAN__
44057		__ai uint16x4_t vceq_f16(float16x4_t __p0, float16x4_t __p1) {
44058		uint16x4_t __ret;
44059		__ret = (uint16x4_t)(__p0 == __p1);
44060		return __ret;
44061		}
	44400	#define vcopy_lane_p16(__p0_281, __p1_281, __p2_281, __p3_281) __extension__ ({ \
	44401	poly16x4_t __ret_281; \
	44402	poly16x4_t __s0_281 = __p0_281; \
	44403	poly16x4_t __s2_281 = __p2_281; \
	44404	__ret_281 = vset_lane_p16(vget_lane_p16(__s2_281, __p3_281), __s0_281, __p1_281); \
	44405	__ret_281; \
	44406	})
44062	44407	#else
44063		__ai uint16x4_t vceq_f16(float16x4_t __p0, float16x4_t __p1) {
44064		uint16x4_t __ret;
44065		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44066		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44067		__ret = (uint16x4_t)(__rev0 == __rev1);
44068		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44069		return __ret;
44070		}
	44408	#define vcopy_lane_p16(__p0_282, __p1_282, __p2_282, __p3_282) __extension__ ({ \
	44409	poly16x4_t __ret_282; \
	44410	poly16x4_t __s0_282 = __p0_282; \
	44411	poly16x4_t __s2_282 = __p2_282; \
	44412	poly16x4_t __rev0_282; __rev0_282 = __builtin_shufflevector(__s0_282, __s0_282, 3, 2, 1, 0); \
	44413	poly16x4_t __rev2_282; __rev2_282 = __builtin_shufflevector(__s2_282, __s2_282, 3, 2, 1, 0); \
	44414	__ret_282 = __noswap_vset_lane_p16(__noswap_vget_lane_p16(__rev2_282, __p3_282), __rev0_282, __p1_282); \
	44415	__ret_282 = __builtin_shufflevector(__ret_282, __ret_282, 3, 2, 1, 0); \
	44416	__ret_282; \
	44417	})
44071	44418	#endif
44072	44419
44073	44420	#ifdef __LITTLE_ENDIAN__
44074		__ai uint16x8_t vceqzq_f16(float16x8_t __p0) {
44075		uint16x8_t __ret;
44076		__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 49);
44077		return __ret;
44078		}
	44421	#define vcopy_lane_u8(__p0_283, __p1_283, __p2_283, __p3_283) __extension__ ({ \
	44422	uint8x8_t __ret_283; \
	44423	uint8x8_t __s0_283 = __p0_283; \
	44424	uint8x8_t __s2_283 = __p2_283; \
	44425	__ret_283 = vset_lane_u8(vget_lane_u8(__s2_283, __p3_283), __s0_283, __p1_283); \
	44426	__ret_283; \
	44427	})
44079	44428	#else
44080		__ai uint16x8_t vceqzq_f16(float16x8_t __p0) {
44081		uint16x8_t __ret;
44082		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44083		__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
44084		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44085		return __ret;
44086		}
	44429	#define vcopy_lane_u8(__p0_284, __p1_284, __p2_284, __p3_284) __extension__ ({ \
	44430	uint8x8_t __ret_284; \
	44431	uint8x8_t __s0_284 = __p0_284; \
	44432	uint8x8_t __s2_284 = __p2_284; \
	44433	uint8x8_t __rev0_284; __rev0_284 = __builtin_shufflevector(__s0_284, __s0_284, 7, 6, 5, 4, 3, 2, 1, 0); \
	44434	uint8x8_t __rev2_284; __rev2_284 = __builtin_shufflevector(__s2_284, __s2_284, 7, 6, 5, 4, 3, 2, 1, 0); \
	44435	__ret_284 = __noswap_vset_lane_u8(__noswap_vget_lane_u8(__rev2_284, __p3_284), __rev0_284, __p1_284); \
	44436	__ret_284 = __builtin_shufflevector(__ret_284, __ret_284, 7, 6, 5, 4, 3, 2, 1, 0); \
	44437	__ret_284; \
	44438	})
44087	44439	#endif
44088	44440
44089	44441	#ifdef __LITTLE_ENDIAN__
44090		__ai uint16x4_t vceqz_f16(float16x4_t __p0) {
44091		uint16x4_t __ret;
44092		__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 17);
44093		return __ret;
44094		}
	44442	#define vcopy_lane_u32(__p0_285, __p1_285, __p2_285, __p3_285) __extension__ ({ \
	44443	uint32x2_t __ret_285; \
	44444	uint32x2_t __s0_285 = __p0_285; \
	44445	uint32x2_t __s2_285 = __p2_285; \
	44446	__ret_285 = vset_lane_u32(vget_lane_u32(__s2_285, __p3_285), __s0_285, __p1_285); \
	44447	__ret_285; \
	44448	})
44095	44449	#else
44096		__ai uint16x4_t vceqz_f16(float16x4_t __p0) {
44097		uint16x4_t __ret;
44098		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44099		__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
44100		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44101		return __ret;
44102		}
	44450	#define vcopy_lane_u32(__p0_286, __p1_286, __p2_286, __p3_286) __extension__ ({ \
	44451	uint32x2_t __ret_286; \
	44452	uint32x2_t __s0_286 = __p0_286; \
	44453	uint32x2_t __s2_286 = __p2_286; \
	44454	uint32x2_t __rev0_286; __rev0_286 = __builtin_shufflevector(__s0_286, __s0_286, 1, 0); \
	44455	uint32x2_t __rev2_286; __rev2_286 = __builtin_shufflevector(__s2_286, __s2_286, 1, 0); \
	44456	__ret_286 = __noswap_vset_lane_u32(__noswap_vget_lane_u32(__rev2_286, __p3_286), __rev0_286, __p1_286); \
	44457	__ret_286 = __builtin_shufflevector(__ret_286, __ret_286, 1, 0); \
	44458	__ret_286; \
	44459	})
44103	44460	#endif
44104	44461
	44462	#define vcopy_lane_u64(__p0_287, __p1_287, __p2_287, __p3_287) __extension__ ({ \
	44463	uint64x1_t __ret_287; \
	44464	uint64x1_t __s0_287 = __p0_287; \
	44465	uint64x1_t __s2_287 = __p2_287; \
	44466	__ret_287 = vset_lane_u64(vget_lane_u64(__s2_287, __p3_287), __s0_287, __p1_287); \
	44467	__ret_287; \
	44468	})
44105	44469	#ifdef __LITTLE_ENDIAN__
44106		__ai uint16x8_t vcgeq_f16(float16x8_t __p0, float16x8_t __p1) {
44107		uint16x8_t __ret;
44108		__ret = (uint16x8_t)(__p0 >= __p1);
44109		return __ret;
44110		}
	44470	#define vcopy_lane_u16(__p0_288, __p1_288, __p2_288, __p3_288) __extension__ ({ \
	44471	uint16x4_t __ret_288; \
	44472	uint16x4_t __s0_288 = __p0_288; \
	44473	uint16x4_t __s2_288 = __p2_288; \
	44474	__ret_288 = vset_lane_u16(vget_lane_u16(__s2_288, __p3_288), __s0_288, __p1_288); \
	44475	__ret_288; \
	44476	})
44111	44477	#else
44112		__ai uint16x8_t vcgeq_f16(float16x8_t __p0, float16x8_t __p1) {
44113		uint16x8_t __ret;
44114		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44115		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44116		__ret = (uint16x8_t)(__rev0 >= __rev1);
44117		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44118		return __ret;
44119		}
	44478	#define vcopy_lane_u16(__p0_289, __p1_289, __p2_289, __p3_289) __extension__ ({ \
	44479	uint16x4_t __ret_289; \
	44480	uint16x4_t __s0_289 = __p0_289; \
	44481	uint16x4_t __s2_289 = __p2_289; \
	44482	uint16x4_t __rev0_289; __rev0_289 = __builtin_shufflevector(__s0_289, __s0_289, 3, 2, 1, 0); \
	44483	uint16x4_t __rev2_289; __rev2_289 = __builtin_shufflevector(__s2_289, __s2_289, 3, 2, 1, 0); \
	44484	__ret_289 = __noswap_vset_lane_u16(__noswap_vget_lane_u16(__rev2_289, __p3_289), __rev0_289, __p1_289); \
	44485	__ret_289 = __builtin_shufflevector(__ret_289, __ret_289, 3, 2, 1, 0); \
	44486	__ret_289; \
	44487	})
44120	44488	#endif
44121	44489
44122	44490	#ifdef __LITTLE_ENDIAN__
44123		__ai uint16x4_t vcge_f16(float16x4_t __p0, float16x4_t __p1) {
44124		uint16x4_t __ret;
44125		__ret = (uint16x4_t)(__p0 >= __p1);
44126		return __ret;
44127		}
	44491	#define vcopy_lane_s8(__p0_290, __p1_290, __p2_290, __p3_290) __extension__ ({ \
	44492	int8x8_t __ret_290; \
	44493	int8x8_t __s0_290 = __p0_290; \
	44494	int8x8_t __s2_290 = __p2_290; \
	44495	__ret_290 = vset_lane_s8(vget_lane_s8(__s2_290, __p3_290), __s0_290, __p1_290); \
	44496	__ret_290; \
	44497	})
44128	44498	#else
44129		__ai uint16x4_t vcge_f16(float16x4_t __p0, float16x4_t __p1) {
44130		uint16x4_t __ret;
44131		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44132		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44133		__ret = (uint16x4_t)(__rev0 >= __rev1);
44134		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44135		return __ret;
44136		}
	44499	#define vcopy_lane_s8(__p0_291, __p1_291, __p2_291, __p3_291) __extension__ ({ \
	44500	int8x8_t __ret_291; \
	44501	int8x8_t __s0_291 = __p0_291; \
	44502	int8x8_t __s2_291 = __p2_291; \
	44503	int8x8_t __rev0_291; __rev0_291 = __builtin_shufflevector(__s0_291, __s0_291, 7, 6, 5, 4, 3, 2, 1, 0); \
	44504	int8x8_t __rev2_291; __rev2_291 = __builtin_shufflevector(__s2_291, __s2_291, 7, 6, 5, 4, 3, 2, 1, 0); \
	44505	__ret_291 = __noswap_vset_lane_s8(__noswap_vget_lane_s8(__rev2_291, __p3_291), __rev0_291, __p1_291); \
	44506	__ret_291 = __builtin_shufflevector(__ret_291, __ret_291, 7, 6, 5, 4, 3, 2, 1, 0); \
	44507	__ret_291; \
	44508	})
44137	44509	#endif
44138	44510
44139	44511	#ifdef __LITTLE_ENDIAN__
44140		__ai uint16x8_t vcgezq_f16(float16x8_t __p0) {
44141		uint16x8_t __ret;
44142		__ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 49);
44143		return __ret;
44144		}
	44512	#define vcopy_lane_f32(__p0_292, __p1_292, __p2_292, __p3_292) __extension__ ({ \
	44513	float32x2_t __ret_292; \
	44514	float32x2_t __s0_292 = __p0_292; \
	44515	float32x2_t __s2_292 = __p2_292; \
	44516	__ret_292 = vset_lane_f32(vget_lane_f32(__s2_292, __p3_292), __s0_292, __p1_292); \
	44517	__ret_292; \
	44518	})
44145	44519	#else
44146		__ai uint16x8_t vcgezq_f16(float16x8_t __p0) {
44147		uint16x8_t __ret;
44148		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44149		__ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 49);
44150		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44151		return __ret;
44152		}
	44520	#define vcopy_lane_f32(__p0_293, __p1_293, __p2_293, __p3_293) __extension__ ({ \
	44521	float32x2_t __ret_293; \
	44522	float32x2_t __s0_293 = __p0_293; \
	44523	float32x2_t __s2_293 = __p2_293; \
	44524	float32x2_t __rev0_293; __rev0_293 = __builtin_shufflevector(__s0_293, __s0_293, 1, 0); \
	44525	float32x2_t __rev2_293; __rev2_293 = __builtin_shufflevector(__s2_293, __s2_293, 1, 0); \
	44526	__ret_293 = __noswap_vset_lane_f32(__noswap_vget_lane_f32(__rev2_293, __p3_293), __rev0_293, __p1_293); \
	44527	__ret_293 = __builtin_shufflevector(__ret_293, __ret_293, 1, 0); \
	44528	__ret_293; \
	44529	})
44153	44530	#endif
44154	44531
44155	44532	#ifdef __LITTLE_ENDIAN__
44156		__ai uint16x4_t vcgez_f16(float16x4_t __p0) {
44157		uint16x4_t __ret;
44158		__ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 17);
44159		return __ret;
44160		}
	44533	#define vcopy_lane_s32(__p0_294, __p1_294, __p2_294, __p3_294) __extension__ ({ \
	44534	int32x2_t __ret_294; \
	44535	int32x2_t __s0_294 = __p0_294; \
	44536	int32x2_t __s2_294 = __p2_294; \
	44537	__ret_294 = vset_lane_s32(vget_lane_s32(__s2_294, __p3_294), __s0_294, __p1_294); \
	44538	__ret_294; \
	44539	})
44161	44540	#else
44162		__ai uint16x4_t vcgez_f16(float16x4_t __p0) {
44163		uint16x4_t __ret;
44164		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44165		__ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 17);
44166		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44167		return __ret;
44168		}
	44541	#define vcopy_lane_s32(__p0_295, __p1_295, __p2_295, __p3_295) __extension__ ({ \
	44542	int32x2_t __ret_295; \
	44543	int32x2_t __s0_295 = __p0_295; \
	44544	int32x2_t __s2_295 = __p2_295; \
	44545	int32x2_t __rev0_295; __rev0_295 = __builtin_shufflevector(__s0_295, __s0_295, 1, 0); \
	44546	int32x2_t __rev2_295; __rev2_295 = __builtin_shufflevector(__s2_295, __s2_295, 1, 0); \
	44547	__ret_295 = __noswap_vset_lane_s32(__noswap_vget_lane_s32(__rev2_295, __p3_295), __rev0_295, __p1_295); \
	44548	__ret_295 = __builtin_shufflevector(__ret_295, __ret_295, 1, 0); \
	44549	__ret_295; \
	44550	})
44169	44551	#endif
44170	44552
	44553	#define vcopy_lane_s64(__p0_296, __p1_296, __p2_296, __p3_296) __extension__ ({ \
	44554	int64x1_t __ret_296; \
	44555	int64x1_t __s0_296 = __p0_296; \
	44556	int64x1_t __s2_296 = __p2_296; \
	44557	__ret_296 = vset_lane_s64(vget_lane_s64(__s2_296, __p3_296), __s0_296, __p1_296); \
	44558	__ret_296; \
	44559	})
44171	44560	#ifdef __LITTLE_ENDIAN__
44172		__ai uint16x8_t vcgtq_f16(float16x8_t __p0, float16x8_t __p1) {
44173		uint16x8_t __ret;
44174		__ret = (uint16x8_t)(__p0 > __p1);
44175		return __ret;
44176		}
	44561	#define vcopy_lane_s16(__p0_297, __p1_297, __p2_297, __p3_297) __extension__ ({ \
	44562	int16x4_t __ret_297; \
	44563	int16x4_t __s0_297 = __p0_297; \
	44564	int16x4_t __s2_297 = __p2_297; \
	44565	__ret_297 = vset_lane_s16(vget_lane_s16(__s2_297, __p3_297), __s0_297, __p1_297); \
	44566	__ret_297; \
	44567	})
44177	44568	#else
44178		__ai uint16x8_t vcgtq_f16(float16x8_t __p0, float16x8_t __p1) {
44179		uint16x8_t __ret;
44180		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44181		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44182		__ret = (uint16x8_t)(__rev0 > __rev1);
44183		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44184		return __ret;
44185		}
	44569	#define vcopy_lane_s16(__p0_298, __p1_298, __p2_298, __p3_298) __extension__ ({ \
	44570	int16x4_t __ret_298; \
	44571	int16x4_t __s0_298 = __p0_298; \
	44572	int16x4_t __s2_298 = __p2_298; \
	44573	int16x4_t __rev0_298; __rev0_298 = __builtin_shufflevector(__s0_298, __s0_298, 3, 2, 1, 0); \
	44574	int16x4_t __rev2_298; __rev2_298 = __builtin_shufflevector(__s2_298, __s2_298, 3, 2, 1, 0); \
	44575	__ret_298 = __noswap_vset_lane_s16(__noswap_vget_lane_s16(__rev2_298, __p3_298), __rev0_298, __p1_298); \
	44576	__ret_298 = __builtin_shufflevector(__ret_298, __ret_298, 3, 2, 1, 0); \
	44577	__ret_298; \
	44578	})
44186	44579	#endif
44187	44580
44188	44581	#ifdef __LITTLE_ENDIAN__
44189		__ai uint16x4_t vcgt_f16(float16x4_t __p0, float16x4_t __p1) {
44190		uint16x4_t __ret;
44191		__ret = (uint16x4_t)(__p0 > __p1);
44192		return __ret;
44193		}
	44582	#define vcopyq_laneq_p8(__p0_299, __p1_299, __p2_299, __p3_299) __extension__ ({ \
	44583	poly8x16_t __ret_299; \
	44584	poly8x16_t __s0_299 = __p0_299; \
	44585	poly8x16_t __s2_299 = __p2_299; \
	44586	__ret_299 = vsetq_lane_p8(vgetq_lane_p8(__s2_299, __p3_299), __s0_299, __p1_299); \
	44587	__ret_299; \
	44588	})
44194	44589	#else
44195		__ai uint16x4_t vcgt_f16(float16x4_t __p0, float16x4_t __p1) {
44196		uint16x4_t __ret;
44197		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44198		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44199		__ret = (uint16x4_t)(__rev0 > __rev1);
44200		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44201		return __ret;
44202		}
	44590	#define vcopyq_laneq_p8(__p0_300, __p1_300, __p2_300, __p3_300) __extension__ ({ \
	44591	poly8x16_t __ret_300; \
	44592	poly8x16_t __s0_300 = __p0_300; \
	44593	poly8x16_t __s2_300 = __p2_300; \
	44594	poly8x16_t __rev0_300; __rev0_300 = __builtin_shufflevector(__s0_300, __s0_300, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44595	poly8x16_t __rev2_300; __rev2_300 = __builtin_shufflevector(__s2_300, __s2_300, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44596	__ret_300 = __noswap_vsetq_lane_p8(__noswap_vgetq_lane_p8(__rev2_300, __p3_300), __rev0_300, __p1_300); \
	44597	__ret_300 = __builtin_shufflevector(__ret_300, __ret_300, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44598	__ret_300; \
	44599	})
44203	44600	#endif
44204	44601
44205	44602	#ifdef __LITTLE_ENDIAN__
44206		__ai uint16x8_t vcgtzq_f16(float16x8_t __p0) {
44207		uint16x8_t __ret;
44208		__ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 49);
44209		return __ret;
44210		}
	44603	#define vcopyq_laneq_p16(__p0_301, __p1_301, __p2_301, __p3_301) __extension__ ({ \
	44604	poly16x8_t __ret_301; \
	44605	poly16x8_t __s0_301 = __p0_301; \
	44606	poly16x8_t __s2_301 = __p2_301; \
	44607	__ret_301 = vsetq_lane_p16(vgetq_lane_p16(__s2_301, __p3_301), __s0_301, __p1_301); \
	44608	__ret_301; \
	44609	})
44211	44610	#else
44212		__ai uint16x8_t vcgtzq_f16(float16x8_t __p0) {
44213		uint16x8_t __ret;
44214		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44215		__ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 49);
44216		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44217		return __ret;
44218		}
	44611	#define vcopyq_laneq_p16(__p0_302, __p1_302, __p2_302, __p3_302) __extension__ ({ \
	44612	poly16x8_t __ret_302; \
	44613	poly16x8_t __s0_302 = __p0_302; \
	44614	poly16x8_t __s2_302 = __p2_302; \
	44615	poly16x8_t __rev0_302; __rev0_302 = __builtin_shufflevector(__s0_302, __s0_302, 7, 6, 5, 4, 3, 2, 1, 0); \
	44616	poly16x8_t __rev2_302; __rev2_302 = __builtin_shufflevector(__s2_302, __s2_302, 7, 6, 5, 4, 3, 2, 1, 0); \
	44617	__ret_302 = __noswap_vsetq_lane_p16(__noswap_vgetq_lane_p16(__rev2_302, __p3_302), __rev0_302, __p1_302); \
	44618	__ret_302 = __builtin_shufflevector(__ret_302, __ret_302, 7, 6, 5, 4, 3, 2, 1, 0); \
	44619	__ret_302; \
	44620	})
44219	44621	#endif
44220	44622
44221	44623	#ifdef __LITTLE_ENDIAN__
44222		__ai uint16x4_t vcgtz_f16(float16x4_t __p0) {
44223		uint16x4_t __ret;
44224		__ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 17);
44225		return __ret;
44226		}
	44624	#define vcopyq_laneq_u8(__p0_303, __p1_303, __p2_303, __p3_303) __extension__ ({ \
	44625	uint8x16_t __ret_303; \
	44626	uint8x16_t __s0_303 = __p0_303; \
	44627	uint8x16_t __s2_303 = __p2_303; \
	44628	__ret_303 = vsetq_lane_u8(vgetq_lane_u8(__s2_303, __p3_303), __s0_303, __p1_303); \
	44629	__ret_303; \
	44630	})
44227	44631	#else
44228		__ai uint16x4_t vcgtz_f16(float16x4_t __p0) {
44229		uint16x4_t __ret;
44230		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44231		__ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 17);
44232		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44233		return __ret;
44234		}
	44632	#define vcopyq_laneq_u8(__p0_304, __p1_304, __p2_304, __p3_304) __extension__ ({ \
	44633	uint8x16_t __ret_304; \
	44634	uint8x16_t __s0_304 = __p0_304; \
	44635	uint8x16_t __s2_304 = __p2_304; \
	44636	uint8x16_t __rev0_304; __rev0_304 = __builtin_shufflevector(__s0_304, __s0_304, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44637	uint8x16_t __rev2_304; __rev2_304 = __builtin_shufflevector(__s2_304, __s2_304, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44638	__ret_304 = __noswap_vsetq_lane_u8(__noswap_vgetq_lane_u8(__rev2_304, __p3_304), __rev0_304, __p1_304); \
	44639	__ret_304 = __builtin_shufflevector(__ret_304, __ret_304, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44640	__ret_304; \
	44641	})
44235	44642	#endif
44236	44643
44237	44644	#ifdef __LITTLE_ENDIAN__
44238		__ai uint16x8_t vcleq_f16(float16x8_t __p0, float16x8_t __p1) {
44239		uint16x8_t __ret;
44240		__ret = (uint16x8_t)(__p0 <= __p1);
44241		return __ret;
44242		}
	44645	#define vcopyq_laneq_u32(__p0_305, __p1_305, __p2_305, __p3_305) __extension__ ({ \
	44646	uint32x4_t __ret_305; \
	44647	uint32x4_t __s0_305 = __p0_305; \
	44648	uint32x4_t __s2_305 = __p2_305; \
	44649	__ret_305 = vsetq_lane_u32(vgetq_lane_u32(__s2_305, __p3_305), __s0_305, __p1_305); \
	44650	__ret_305; \
	44651	})
44243	44652	#else
44244		__ai uint16x8_t vcleq_f16(float16x8_t __p0, float16x8_t __p1) {
44245		uint16x8_t __ret;
44246		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44247		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44248		__ret = (uint16x8_t)(__rev0 <= __rev1);
44249		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44250		return __ret;
44251		}
	44653	#define vcopyq_laneq_u32(__p0_306, __p1_306, __p2_306, __p3_306) __extension__ ({ \
	44654	uint32x4_t __ret_306; \
	44655	uint32x4_t __s0_306 = __p0_306; \
	44656	uint32x4_t __s2_306 = __p2_306; \
	44657	uint32x4_t __rev0_306; __rev0_306 = __builtin_shufflevector(__s0_306, __s0_306, 3, 2, 1, 0); \
	44658	uint32x4_t __rev2_306; __rev2_306 = __builtin_shufflevector(__s2_306, __s2_306, 3, 2, 1, 0); \
	44659	__ret_306 = __noswap_vsetq_lane_u32(__noswap_vgetq_lane_u32(__rev2_306, __p3_306), __rev0_306, __p1_306); \
	44660	__ret_306 = __builtin_shufflevector(__ret_306, __ret_306, 3, 2, 1, 0); \
	44661	__ret_306; \
	44662	})
44252	44663	#endif
44253	44664
44254	44665	#ifdef __LITTLE_ENDIAN__
44255		__ai uint16x4_t vcle_f16(float16x4_t __p0, float16x4_t __p1) {
44256		uint16x4_t __ret;
44257		__ret = (uint16x4_t)(__p0 <= __p1);
44258		return __ret;
44259		}
	44666	#define vcopyq_laneq_u64(__p0_307, __p1_307, __p2_307, __p3_307) __extension__ ({ \
	44667	uint64x2_t __ret_307; \
	44668	uint64x2_t __s0_307 = __p0_307; \
	44669	uint64x2_t __s2_307 = __p2_307; \
	44670	__ret_307 = vsetq_lane_u64(vgetq_lane_u64(__s2_307, __p3_307), __s0_307, __p1_307); \
	44671	__ret_307; \
	44672	})
44260	44673	#else
44261		__ai uint16x4_t vcle_f16(float16x4_t __p0, float16x4_t __p1) {
44262		uint16x4_t __ret;
44263		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44264		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44265		__ret = (uint16x4_t)(__rev0 <= __rev1);
44266		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44267		return __ret;
44268		}
	44674	#define vcopyq_laneq_u64(__p0_308, __p1_308, __p2_308, __p3_308) __extension__ ({ \
	44675	uint64x2_t __ret_308; \
	44676	uint64x2_t __s0_308 = __p0_308; \
	44677	uint64x2_t __s2_308 = __p2_308; \
	44678	uint64x2_t __rev0_308; __rev0_308 = __builtin_shufflevector(__s0_308, __s0_308, 1, 0); \
	44679	uint64x2_t __rev2_308; __rev2_308 = __builtin_shufflevector(__s2_308, __s2_308, 1, 0); \
	44680	__ret_308 = __noswap_vsetq_lane_u64(__noswap_vgetq_lane_u64(__rev2_308, __p3_308), __rev0_308, __p1_308); \
	44681	__ret_308 = __builtin_shufflevector(__ret_308, __ret_308, 1, 0); \
	44682	__ret_308; \
	44683	})
44269	44684	#endif
44270	44685
44271	44686	#ifdef __LITTLE_ENDIAN__
44272		__ai uint16x8_t vclezq_f16(float16x8_t __p0) {
44273		uint16x8_t __ret;
44274		__ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 49);
44275		return __ret;
44276		}
	44687	#define vcopyq_laneq_u16(__p0_309, __p1_309, __p2_309, __p3_309) __extension__ ({ \
	44688	uint16x8_t __ret_309; \
	44689	uint16x8_t __s0_309 = __p0_309; \
	44690	uint16x8_t __s2_309 = __p2_309; \
	44691	__ret_309 = vsetq_lane_u16(vgetq_lane_u16(__s2_309, __p3_309), __s0_309, __p1_309); \
	44692	__ret_309; \
	44693	})
44277	44694	#else
44278		__ai uint16x8_t vclezq_f16(float16x8_t __p0) {
44279		uint16x8_t __ret;
44280		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44281		__ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 49);
44282		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44283		return __ret;
44284		}
	44695	#define vcopyq_laneq_u16(__p0_310, __p1_310, __p2_310, __p3_310) __extension__ ({ \
	44696	uint16x8_t __ret_310; \
	44697	uint16x8_t __s0_310 = __p0_310; \
	44698	uint16x8_t __s2_310 = __p2_310; \
	44699	uint16x8_t __rev0_310; __rev0_310 = __builtin_shufflevector(__s0_310, __s0_310, 7, 6, 5, 4, 3, 2, 1, 0); \
	44700	uint16x8_t __rev2_310; __rev2_310 = __builtin_shufflevector(__s2_310, __s2_310, 7, 6, 5, 4, 3, 2, 1, 0); \
	44701	__ret_310 = __noswap_vsetq_lane_u16(__noswap_vgetq_lane_u16(__rev2_310, __p3_310), __rev0_310, __p1_310); \
	44702	__ret_310 = __builtin_shufflevector(__ret_310, __ret_310, 7, 6, 5, 4, 3, 2, 1, 0); \
	44703	__ret_310; \
	44704	})
44285	44705	#endif
44286	44706
44287	44707	#ifdef __LITTLE_ENDIAN__
44288		__ai uint16x4_t vclez_f16(float16x4_t __p0) {
44289		uint16x4_t __ret;
44290		__ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__p0, 17);
44291		return __ret;
44292		}
	44708	#define vcopyq_laneq_s8(__p0_311, __p1_311, __p2_311, __p3_311) __extension__ ({ \
	44709	int8x16_t __ret_311; \
	44710	int8x16_t __s0_311 = __p0_311; \
	44711	int8x16_t __s2_311 = __p2_311; \
	44712	__ret_311 = vsetq_lane_s8(vgetq_lane_s8(__s2_311, __p3_311), __s0_311, __p1_311); \
	44713	__ret_311; \
	44714	})
44293	44715	#else
44294		__ai uint16x4_t vclez_f16(float16x4_t __p0) {
44295		uint16x4_t __ret;
44296		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44297		__ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 17);
44298		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44299		return __ret;
44300		}
	44716	#define vcopyq_laneq_s8(__p0_312, __p1_312, __p2_312, __p3_312) __extension__ ({ \
	44717	int8x16_t __ret_312; \
	44718	int8x16_t __s0_312 = __p0_312; \
	44719	int8x16_t __s2_312 = __p2_312; \
	44720	int8x16_t __rev0_312; __rev0_312 = __builtin_shufflevector(__s0_312, __s0_312, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44721	int8x16_t __rev2_312; __rev2_312 = __builtin_shufflevector(__s2_312, __s2_312, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44722	__ret_312 = __noswap_vsetq_lane_s8(__noswap_vgetq_lane_s8(__rev2_312, __p3_312), __rev0_312, __p1_312); \
	44723	__ret_312 = __builtin_shufflevector(__ret_312, __ret_312, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44724	__ret_312; \
	44725	})
44301	44726	#endif
44302	44727
44303	44728	#ifdef __LITTLE_ENDIAN__
44304		__ai uint16x8_t vcltq_f16(float16x8_t __p0, float16x8_t __p1) {
44305		uint16x8_t __ret;
44306		__ret = (uint16x8_t)(__p0 < __p1);
44307		return __ret;
44308		}
	44729	#define vcopyq_laneq_f32(__p0_313, __p1_313, __p2_313, __p3_313) __extension__ ({ \
	44730	float32x4_t __ret_313; \
	44731	float32x4_t __s0_313 = __p0_313; \
	44732	float32x4_t __s2_313 = __p2_313; \
	44733	__ret_313 = vsetq_lane_f32(vgetq_lane_f32(__s2_313, __p3_313), __s0_313, __p1_313); \
	44734	__ret_313; \
	44735	})
44309	44736	#else
44310		__ai uint16x8_t vcltq_f16(float16x8_t __p0, float16x8_t __p1) {
44311		uint16x8_t __ret;
44312		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44313		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44314		__ret = (uint16x8_t)(__rev0 < __rev1);
44315		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44316		return __ret;
44317		}
	44737	#define vcopyq_laneq_f32(__p0_314, __p1_314, __p2_314, __p3_314) __extension__ ({ \
	44738	float32x4_t __ret_314; \
	44739	float32x4_t __s0_314 = __p0_314; \
	44740	float32x4_t __s2_314 = __p2_314; \
	44741	float32x4_t __rev0_314; __rev0_314 = __builtin_shufflevector(__s0_314, __s0_314, 3, 2, 1, 0); \
	44742	float32x4_t __rev2_314; __rev2_314 = __builtin_shufflevector(__s2_314, __s2_314, 3, 2, 1, 0); \
	44743	__ret_314 = __noswap_vsetq_lane_f32(__noswap_vgetq_lane_f32(__rev2_314, __p3_314), __rev0_314, __p1_314); \
	44744	__ret_314 = __builtin_shufflevector(__ret_314, __ret_314, 3, 2, 1, 0); \
	44745	__ret_314; \
	44746	})
44318	44747	#endif
44319	44748
44320	44749	#ifdef __LITTLE_ENDIAN__
44321		__ai uint16x4_t vclt_f16(float16x4_t __p0, float16x4_t __p1) {
44322		uint16x4_t __ret;
44323		__ret = (uint16x4_t)(__p0 < __p1);
44324		return __ret;
44325		}
	44750	#define vcopyq_laneq_s32(__p0_315, __p1_315, __p2_315, __p3_315) __extension__ ({ \
	44751	int32x4_t __ret_315; \
	44752	int32x4_t __s0_315 = __p0_315; \
	44753	int32x4_t __s2_315 = __p2_315; \
	44754	__ret_315 = vsetq_lane_s32(vgetq_lane_s32(__s2_315, __p3_315), __s0_315, __p1_315); \
	44755	__ret_315; \
	44756	})
44326	44757	#else
44327		__ai uint16x4_t vclt_f16(float16x4_t __p0, float16x4_t __p1) {
44328		uint16x4_t __ret;
44329		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44330		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44331		__ret = (uint16x4_t)(__rev0 < __rev1);
44332		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44333		return __ret;
44334		}
	44758	#define vcopyq_laneq_s32(__p0_316, __p1_316, __p2_316, __p3_316) __extension__ ({ \
	44759	int32x4_t __ret_316; \
	44760	int32x4_t __s0_316 = __p0_316; \
	44761	int32x4_t __s2_316 = __p2_316; \
	44762	int32x4_t __rev0_316; __rev0_316 = __builtin_shufflevector(__s0_316, __s0_316, 3, 2, 1, 0); \
	44763	int32x4_t __rev2_316; __rev2_316 = __builtin_shufflevector(__s2_316, __s2_316, 3, 2, 1, 0); \
	44764	__ret_316 = __noswap_vsetq_lane_s32(__noswap_vgetq_lane_s32(__rev2_316, __p3_316), __rev0_316, __p1_316); \
	44765	__ret_316 = __builtin_shufflevector(__ret_316, __ret_316, 3, 2, 1, 0); \
	44766	__ret_316; \
	44767	})
44335	44768	#endif
44336	44769
44337	44770	#ifdef __LITTLE_ENDIAN__
44338		__ai uint16x8_t vcltzq_f16(float16x8_t __p0) {
44339		uint16x8_t __ret;
44340		__ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 49);
44341		return __ret;
44342		}
	44771	#define vcopyq_laneq_s64(__p0_317, __p1_317, __p2_317, __p3_317) __extension__ ({ \
	44772	int64x2_t __ret_317; \
	44773	int64x2_t __s0_317 = __p0_317; \
	44774	int64x2_t __s2_317 = __p2_317; \
	44775	__ret_317 = vsetq_lane_s64(vgetq_lane_s64(__s2_317, __p3_317), __s0_317, __p1_317); \
	44776	__ret_317; \
	44777	})
44343	44778	#else
44344		__ai uint16x8_t vcltzq_f16(float16x8_t __p0) {
44345		uint16x8_t __ret;
44346		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44347		__ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 49);
44348		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44349		return __ret;
44350		}
	44779	#define vcopyq_laneq_s64(__p0_318, __p1_318, __p2_318, __p3_318) __extension__ ({ \
	44780	int64x2_t __ret_318; \
	44781	int64x2_t __s0_318 = __p0_318; \
	44782	int64x2_t __s2_318 = __p2_318; \
	44783	int64x2_t __rev0_318; __rev0_318 = __builtin_shufflevector(__s0_318, __s0_318, 1, 0); \
	44784	int64x2_t __rev2_318; __rev2_318 = __builtin_shufflevector(__s2_318, __s2_318, 1, 0); \
	44785	__ret_318 = __noswap_vsetq_lane_s64(__noswap_vgetq_lane_s64(__rev2_318, __p3_318), __rev0_318, __p1_318); \
	44786	__ret_318 = __builtin_shufflevector(__ret_318, __ret_318, 1, 0); \
	44787	__ret_318; \
	44788	})
44351	44789	#endif
44352	44790
44353	44791	#ifdef __LITTLE_ENDIAN__
44354		__ai uint16x4_t vcltz_f16(float16x4_t __p0) {
44355		uint16x4_t __ret;
44356		__ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 17);
44357		return __ret;
44358		}
	44792	#define vcopyq_laneq_s16(__p0_319, __p1_319, __p2_319, __p3_319) __extension__ ({ \
	44793	int16x8_t __ret_319; \
	44794	int16x8_t __s0_319 = __p0_319; \
	44795	int16x8_t __s2_319 = __p2_319; \
	44796	__ret_319 = vsetq_lane_s16(vgetq_lane_s16(__s2_319, __p3_319), __s0_319, __p1_319); \
	44797	__ret_319; \
	44798	})
44359	44799	#else
44360		__ai uint16x4_t vcltz_f16(float16x4_t __p0) {
44361		uint16x4_t __ret;
44362		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44363		__ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 17);
44364		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44365		return __ret;
44366		}
	44800	#define vcopyq_laneq_s16(__p0_320, __p1_320, __p2_320, __p3_320) __extension__ ({ \
	44801	int16x8_t __ret_320; \
	44802	int16x8_t __s0_320 = __p0_320; \
	44803	int16x8_t __s2_320 = __p2_320; \
	44804	int16x8_t __rev0_320; __rev0_320 = __builtin_shufflevector(__s0_320, __s0_320, 7, 6, 5, 4, 3, 2, 1, 0); \
	44805	int16x8_t __rev2_320; __rev2_320 = __builtin_shufflevector(__s2_320, __s2_320, 7, 6, 5, 4, 3, 2, 1, 0); \
	44806	__ret_320 = __noswap_vsetq_lane_s16(__noswap_vgetq_lane_s16(__rev2_320, __p3_320), __rev0_320, __p1_320); \
	44807	__ret_320 = __builtin_shufflevector(__ret_320, __ret_320, 7, 6, 5, 4, 3, 2, 1, 0); \
	44808	__ret_320; \
	44809	})
44367	44810	#endif
44368	44811
44369	44812	#ifdef __LITTLE_ENDIAN__
44370		__ai float16x8_t vcvtq_f16_u16(uint16x8_t __p0) {
44371		float16x8_t __ret;
44372		__ret = (float16x8_t) __builtin_neon_vcvtq_f16_v((int8x16_t)__p0, 49);
44373		return __ret;
44374		}
	44813	#define vcopy_laneq_p8(__p0_321, __p1_321, __p2_321, __p3_321) __extension__ ({ \
	44814	poly8x8_t __ret_321; \
	44815	poly8x8_t __s0_321 = __p0_321; \
	44816	poly8x16_t __s2_321 = __p2_321; \
	44817	__ret_321 = vset_lane_p8(vgetq_lane_p8(__s2_321, __p3_321), __s0_321, __p1_321); \
	44818	__ret_321; \
	44819	})
44375	44820	#else
44376		__ai float16x8_t vcvtq_f16_u16(uint16x8_t __p0) {
44377		float16x8_t __ret;
44378		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44379		__ret = (float16x8_t) __builtin_neon_vcvtq_f16_v((int8x16_t)__rev0, 49);
44380		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44381		return __ret;
44382		}
	44821	#define vcopy_laneq_p8(__p0_322, __p1_322, __p2_322, __p3_322) __extension__ ({ \
	44822	poly8x8_t __ret_322; \
	44823	poly8x8_t __s0_322 = __p0_322; \
	44824	poly8x16_t __s2_322 = __p2_322; \
	44825	poly8x8_t __rev0_322; __rev0_322 = __builtin_shufflevector(__s0_322, __s0_322, 7, 6, 5, 4, 3, 2, 1, 0); \
	44826	poly8x16_t __rev2_322; __rev2_322 = __builtin_shufflevector(__s2_322, __s2_322, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44827	__ret_322 = __noswap_vset_lane_p8(__noswap_vgetq_lane_p8(__rev2_322, __p3_322), __rev0_322, __p1_322); \
	44828	__ret_322 = __builtin_shufflevector(__ret_322, __ret_322, 7, 6, 5, 4, 3, 2, 1, 0); \
	44829	__ret_322; \
	44830	})
44383	44831	#endif
44384	44832
44385	44833	#ifdef __LITTLE_ENDIAN__
44386		__ai float16x8_t vcvtq_f16_s16(int16x8_t __p0) {
44387		float16x8_t __ret;
44388		__ret = (float16x8_t) __builtin_neon_vcvtq_f16_v((int8x16_t)__p0, 33);
44389		return __ret;
44390		}
	44834	#define vcopy_laneq_p16(__p0_323, __p1_323, __p2_323, __p3_323) __extension__ ({ \
	44835	poly16x4_t __ret_323; \
	44836	poly16x4_t __s0_323 = __p0_323; \
	44837	poly16x8_t __s2_323 = __p2_323; \
	44838	__ret_323 = vset_lane_p16(vgetq_lane_p16(__s2_323, __p3_323), __s0_323, __p1_323); \
	44839	__ret_323; \
	44840	})
44391	44841	#else
44392		__ai float16x8_t vcvtq_f16_s16(int16x8_t __p0) {
44393		float16x8_t __ret;
44394		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44395		__ret = (float16x8_t) __builtin_neon_vcvtq_f16_v((int8x16_t)__rev0, 33);
44396		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44397		return __ret;
44398		}
	44842	#define vcopy_laneq_p16(__p0_324, __p1_324, __p2_324, __p3_324) __extension__ ({ \
	44843	poly16x4_t __ret_324; \
	44844	poly16x4_t __s0_324 = __p0_324; \
	44845	poly16x8_t __s2_324 = __p2_324; \
	44846	poly16x4_t __rev0_324; __rev0_324 = __builtin_shufflevector(__s0_324, __s0_324, 3, 2, 1, 0); \
	44847	poly16x8_t __rev2_324; __rev2_324 = __builtin_shufflevector(__s2_324, __s2_324, 7, 6, 5, 4, 3, 2, 1, 0); \
	44848	__ret_324 = __noswap_vset_lane_p16(__noswap_vgetq_lane_p16(__rev2_324, __p3_324), __rev0_324, __p1_324); \
	44849	__ret_324 = __builtin_shufflevector(__ret_324, __ret_324, 3, 2, 1, 0); \
	44850	__ret_324; \
	44851	})
44399	44852	#endif
44400	44853
44401	44854	#ifdef __LITTLE_ENDIAN__
44402		__ai float16x4_t vcvt_f16_u16(uint16x4_t __p0) {
44403		float16x4_t __ret;
44404		__ret = (float16x4_t) __builtin_neon_vcvt_f16_v((int8x8_t)__p0, 17);
44405		return __ret;
44406		}
	44855	#define vcopy_laneq_u8(__p0_325, __p1_325, __p2_325, __p3_325) __extension__ ({ \
	44856	uint8x8_t __ret_325; \
	44857	uint8x8_t __s0_325 = __p0_325; \
	44858	uint8x16_t __s2_325 = __p2_325; \
	44859	__ret_325 = vset_lane_u8(vgetq_lane_u8(__s2_325, __p3_325), __s0_325, __p1_325); \
	44860	__ret_325; \
	44861	})
44407	44862	#else
44408		__ai float16x4_t vcvt_f16_u16(uint16x4_t __p0) {
44409		float16x4_t __ret;
44410		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44411		__ret = (float16x4_t) __builtin_neon_vcvt_f16_v((int8x8_t)__rev0, 17);
44412		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44413		return __ret;
44414		}
	44863	#define vcopy_laneq_u8(__p0_326, __p1_326, __p2_326, __p3_326) __extension__ ({ \
	44864	uint8x8_t __ret_326; \
	44865	uint8x8_t __s0_326 = __p0_326; \
	44866	uint8x16_t __s2_326 = __p2_326; \
	44867	uint8x8_t __rev0_326; __rev0_326 = __builtin_shufflevector(__s0_326, __s0_326, 7, 6, 5, 4, 3, 2, 1, 0); \
	44868	uint8x16_t __rev2_326; __rev2_326 = __builtin_shufflevector(__s2_326, __s2_326, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44869	__ret_326 = __noswap_vset_lane_u8(__noswap_vgetq_lane_u8(__rev2_326, __p3_326), __rev0_326, __p1_326); \
	44870	__ret_326 = __builtin_shufflevector(__ret_326, __ret_326, 7, 6, 5, 4, 3, 2, 1, 0); \
	44871	__ret_326; \
	44872	})
44415	44873	#endif
44416	44874
44417	44875	#ifdef __LITTLE_ENDIAN__
44418		__ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
44419		float16x4_t __ret;
44420		__ret = (float16x4_t) __builtin_neon_vcvt_f16_v((int8x8_t)__p0, 1);
44421		return __ret;
44422		}
	44876	#define vcopy_laneq_u32(__p0_327, __p1_327, __p2_327, __p3_327) __extension__ ({ \
	44877	uint32x2_t __ret_327; \
	44878	uint32x2_t __s0_327 = __p0_327; \
	44879	uint32x4_t __s2_327 = __p2_327; \
	44880	__ret_327 = vset_lane_u32(vgetq_lane_u32(__s2_327, __p3_327), __s0_327, __p1_327); \
	44881	__ret_327; \
	44882	})
44423	44883	#else
44424		__ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
44425		float16x4_t __ret;
44426		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44427		__ret = (float16x4_t) __builtin_neon_vcvt_f16_v((int8x8_t)__rev0, 1);
44428		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44429		return __ret;
44430		}
	44884	#define vcopy_laneq_u32(__p0_328, __p1_328, __p2_328, __p3_328) __extension__ ({ \
	44885	uint32x2_t __ret_328; \
	44886	uint32x2_t __s0_328 = __p0_328; \
	44887	uint32x4_t __s2_328 = __p2_328; \
	44888	uint32x2_t __rev0_328; __rev0_328 = __builtin_shufflevector(__s0_328, __s0_328, 1, 0); \
	44889	uint32x4_t __rev2_328; __rev2_328 = __builtin_shufflevector(__s2_328, __s2_328, 3, 2, 1, 0); \
	44890	__ret_328 = __noswap_vset_lane_u32(__noswap_vgetq_lane_u32(__rev2_328, __p3_328), __rev0_328, __p1_328); \
	44891	__ret_328 = __builtin_shufflevector(__ret_328, __ret_328, 1, 0); \
	44892	__ret_328; \
	44893	})
44431	44894	#endif
44432	44895
44433	44896	#ifdef __LITTLE_ENDIAN__
44434		#define vcvtq_n_f16_u16(__p0, __p1) __extension__ ({ \
44435		float16x8_t __ret; \
44436		uint16x8_t __s0 = __p0; \
44437		__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__s0, __p1, 49); \
44438		__ret; \
	44897	#define vcopy_laneq_u64(__p0_329, __p1_329, __p2_329, __p3_329) __extension__ ({ \
	44898	uint64x1_t __ret_329; \
	44899	uint64x1_t __s0_329 = __p0_329; \
	44900	uint64x2_t __s2_329 = __p2_329; \
	44901	__ret_329 = vset_lane_u64(vgetq_lane_u64(__s2_329, __p3_329), __s0_329, __p1_329); \
	44902	__ret_329; \
44439	44903	})
44440	44904	#else
44441		#define vcvtq_n_f16_u16(__p0, __p1) __extension__ ({ \
44442		float16x8_t __ret; \
44443		uint16x8_t __s0 = __p0; \
44444		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44445		__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__rev0, __p1, 49); \
44446		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
44447		__ret; \
	44905	#define vcopy_laneq_u64(__p0_330, __p1_330, __p2_330, __p3_330) __extension__ ({ \
	44906	uint64x1_t __ret_330; \
	44907	uint64x1_t __s0_330 = __p0_330; \
	44908	uint64x2_t __s2_330 = __p2_330; \
	44909	uint64x2_t __rev2_330; __rev2_330 = __builtin_shufflevector(__s2_330, __s2_330, 1, 0); \
	44910	__ret_330 = vset_lane_u64(__noswap_vgetq_lane_u64(__rev2_330, __p3_330), __s0_330, __p1_330); \
	44911	__ret_330; \
44448	44912	})
44449	44913	#endif
44450	44914
44451	44915	#ifdef __LITTLE_ENDIAN__
44452		#define vcvtq_n_f16_s16(__p0, __p1) __extension__ ({ \
44453		float16x8_t __ret; \
44454		int16x8_t __s0 = __p0; \
44455		__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__s0, __p1, 33); \
44456		__ret; \
	44916	#define vcopy_laneq_u16(__p0_331, __p1_331, __p2_331, __p3_331) __extension__ ({ \
	44917	uint16x4_t __ret_331; \
	44918	uint16x4_t __s0_331 = __p0_331; \
	44919	uint16x8_t __s2_331 = __p2_331; \
	44920	__ret_331 = vset_lane_u16(vgetq_lane_u16(__s2_331, __p3_331), __s0_331, __p1_331); \
	44921	__ret_331; \
44457	44922	})
44458	44923	#else
44459		#define vcvtq_n_f16_s16(__p0, __p1) __extension__ ({ \
44460		float16x8_t __ret; \
44461		int16x8_t __s0 = __p0; \
44462		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44463		__ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__rev0, __p1, 33); \
44464		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
44465		__ret; \
	44924	#define vcopy_laneq_u16(__p0_332, __p1_332, __p2_332, __p3_332) __extension__ ({ \
	44925	uint16x4_t __ret_332; \
	44926	uint16x4_t __s0_332 = __p0_332; \
	44927	uint16x8_t __s2_332 = __p2_332; \
	44928	uint16x4_t __rev0_332; __rev0_332 = __builtin_shufflevector(__s0_332, __s0_332, 3, 2, 1, 0); \
	44929	uint16x8_t __rev2_332; __rev2_332 = __builtin_shufflevector(__s2_332, __s2_332, 7, 6, 5, 4, 3, 2, 1, 0); \
	44930	__ret_332 = __noswap_vset_lane_u16(__noswap_vgetq_lane_u16(__rev2_332, __p3_332), __rev0_332, __p1_332); \
	44931	__ret_332 = __builtin_shufflevector(__ret_332, __ret_332, 3, 2, 1, 0); \
	44932	__ret_332; \
44466	44933	})
44467	44934	#endif
44468	44935
44469	44936	#ifdef __LITTLE_ENDIAN__
44470		#define vcvt_n_f16_u16(__p0, __p1) __extension__ ({ \
44471		float16x4_t __ret; \
44472		uint16x4_t __s0 = __p0; \
44473		__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__s0, __p1, 17); \
44474		__ret; \
	44937	#define vcopy_laneq_s8(__p0_333, __p1_333, __p2_333, __p3_333) __extension__ ({ \
	44938	int8x8_t __ret_333; \
	44939	int8x8_t __s0_333 = __p0_333; \
	44940	int8x16_t __s2_333 = __p2_333; \
	44941	__ret_333 = vset_lane_s8(vgetq_lane_s8(__s2_333, __p3_333), __s0_333, __p1_333); \
	44942	__ret_333; \
44475	44943	})
44476	44944	#else
44477		#define vcvt_n_f16_u16(__p0, __p1) __extension__ ({ \
44478		float16x4_t __ret; \
44479		uint16x4_t __s0 = __p0; \
44480		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44481		__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__rev0, __p1, 17); \
44482		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
44483		__ret; \
	44945	#define vcopy_laneq_s8(__p0_334, __p1_334, __p2_334, __p3_334) __extension__ ({ \
	44946	int8x8_t __ret_334; \
	44947	int8x8_t __s0_334 = __p0_334; \
	44948	int8x16_t __s2_334 = __p2_334; \
	44949	int8x8_t __rev0_334; __rev0_334 = __builtin_shufflevector(__s0_334, __s0_334, 7, 6, 5, 4, 3, 2, 1, 0); \
	44950	int8x16_t __rev2_334; __rev2_334 = __builtin_shufflevector(__s2_334, __s2_334, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	44951	__ret_334 = __noswap_vset_lane_s8(__noswap_vgetq_lane_s8(__rev2_334, __p3_334), __rev0_334, __p1_334); \
	44952	__ret_334 = __builtin_shufflevector(__ret_334, __ret_334, 7, 6, 5, 4, 3, 2, 1, 0); \
	44953	__ret_334; \
44484	44954	})
44485	44955	#endif
44486	44956
44487	44957	#ifdef __LITTLE_ENDIAN__
44488		#define vcvt_n_f16_s16(__p0, __p1) __extension__ ({ \
44489		float16x4_t __ret; \
44490		int16x4_t __s0 = __p0; \
44491		__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__s0, __p1, 1); \
44492		__ret; \
	44958	#define vcopy_laneq_f32(__p0_335, __p1_335, __p2_335, __p3_335) __extension__ ({ \
	44959	float32x2_t __ret_335; \
	44960	float32x2_t __s0_335 = __p0_335; \
	44961	float32x4_t __s2_335 = __p2_335; \
	44962	__ret_335 = vset_lane_f32(vgetq_lane_f32(__s2_335, __p3_335), __s0_335, __p1_335); \
	44963	__ret_335; \
44493	44964	})
44494	44965	#else
44495		#define vcvt_n_f16_s16(__p0, __p1) __extension__ ({ \
44496		float16x4_t __ret; \
44497		int16x4_t __s0 = __p0; \
44498		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44499		__ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__rev0, __p1, 1); \
44500		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
44501		__ret; \
	44966	#define vcopy_laneq_f32(__p0_336, __p1_336, __p2_336, __p3_336) __extension__ ({ \
	44967	float32x2_t __ret_336; \
	44968	float32x2_t __s0_336 = __p0_336; \
	44969	float32x4_t __s2_336 = __p2_336; \
	44970	float32x2_t __rev0_336; __rev0_336 = __builtin_shufflevector(__s0_336, __s0_336, 1, 0); \
	44971	float32x4_t __rev2_336; __rev2_336 = __builtin_shufflevector(__s2_336, __s2_336, 3, 2, 1, 0); \
	44972	__ret_336 = __noswap_vset_lane_f32(__noswap_vgetq_lane_f32(__rev2_336, __p3_336), __rev0_336, __p1_336); \
	44973	__ret_336 = __builtin_shufflevector(__ret_336, __ret_336, 1, 0); \
	44974	__ret_336; \
44502	44975	})
44503	44976	#endif
44504	44977
44505	44978	#ifdef __LITTLE_ENDIAN__
44506		#define vcvtq_n_s16_f16(__p0, __p1) __extension__ ({ \
44507		int16x8_t __ret; \
44508		float16x8_t __s0 = __p0; \
44509		__ret = (int16x8_t) __builtin_neon_vcvtq_n_s16_v((int8x16_t)__s0, __p1, 33); \
44510		__ret; \
	44979	#define vcopy_laneq_s32(__p0_337, __p1_337, __p2_337, __p3_337) __extension__ ({ \
	44980	int32x2_t __ret_337; \
	44981	int32x2_t __s0_337 = __p0_337; \
	44982	int32x4_t __s2_337 = __p2_337; \
	44983	__ret_337 = vset_lane_s32(vgetq_lane_s32(__s2_337, __p3_337), __s0_337, __p1_337); \
	44984	__ret_337; \
44511	44985	})
44512	44986	#else
44513		#define vcvtq_n_s16_f16(__p0, __p1) __extension__ ({ \
44514		int16x8_t __ret; \
44515		float16x8_t __s0 = __p0; \
44516		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44517		__ret = (int16x8_t) __builtin_neon_vcvtq_n_s16_v((int8x16_t)__rev0, __p1, 33); \
44518		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
44519		__ret; \
	44987	#define vcopy_laneq_s32(__p0_338, __p1_338, __p2_338, __p3_338) __extension__ ({ \
	44988	int32x2_t __ret_338; \
	44989	int32x2_t __s0_338 = __p0_338; \
	44990	int32x4_t __s2_338 = __p2_338; \
	44991	int32x2_t __rev0_338; __rev0_338 = __builtin_shufflevector(__s0_338, __s0_338, 1, 0); \
	44992	int32x4_t __rev2_338; __rev2_338 = __builtin_shufflevector(__s2_338, __s2_338, 3, 2, 1, 0); \
	44993	__ret_338 = __noswap_vset_lane_s32(__noswap_vgetq_lane_s32(__rev2_338, __p3_338), __rev0_338, __p1_338); \
	44994	__ret_338 = __builtin_shufflevector(__ret_338, __ret_338, 1, 0); \
	44995	__ret_338; \
44520	44996	})
44521	44997	#endif
44522	44998
44523	44999	#ifdef __LITTLE_ENDIAN__
44524		#define vcvt_n_s16_f16(__p0, __p1) __extension__ ({ \
44525		int16x4_t __ret; \
44526		float16x4_t __s0 = __p0; \
44527		__ret = (int16x4_t) __builtin_neon_vcvt_n_s16_v((int8x8_t)__s0, __p1, 1); \
44528		__ret; \
	45000	#define vcopy_laneq_s64(__p0_339, __p1_339, __p2_339, __p3_339) __extension__ ({ \
	45001	int64x1_t __ret_339; \
	45002	int64x1_t __s0_339 = __p0_339; \
	45003	int64x2_t __s2_339 = __p2_339; \
	45004	__ret_339 = vset_lane_s64(vgetq_lane_s64(__s2_339, __p3_339), __s0_339, __p1_339); \
	45005	__ret_339; \
44529	45006	})
44530	45007	#else
44531		#define vcvt_n_s16_f16(__p0, __p1) __extension__ ({ \
44532		int16x4_t __ret; \
44533		float16x4_t __s0 = __p0; \
44534		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44535		__ret = (int16x4_t) __builtin_neon_vcvt_n_s16_v((int8x8_t)__rev0, __p1, 1); \
44536		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
44537		__ret; \
	45008	#define vcopy_laneq_s64(__p0_340, __p1_340, __p2_340, __p3_340) __extension__ ({ \
	45009	int64x1_t __ret_340; \
	45010	int64x1_t __s0_340 = __p0_340; \
	45011	int64x2_t __s2_340 = __p2_340; \
	45012	int64x2_t __rev2_340; __rev2_340 = __builtin_shufflevector(__s2_340, __s2_340, 1, 0); \
	45013	__ret_340 = vset_lane_s64(__noswap_vgetq_lane_s64(__rev2_340, __p3_340), __s0_340, __p1_340); \
	45014	__ret_340; \
44538	45015	})
44539	45016	#endif
44540	45017
44541	45018	#ifdef __LITTLE_ENDIAN__
44542		#define vcvtq_n_u16_f16(__p0, __p1) __extension__ ({ \
44543		uint16x8_t __ret; \
44544		float16x8_t __s0 = __p0; \
44545		__ret = (uint16x8_t) __builtin_neon_vcvtq_n_u16_v((int8x16_t)__s0, __p1, 49); \
44546		__ret; \
	45019	#define vcopy_laneq_s16(__p0_341, __p1_341, __p2_341, __p3_341) __extension__ ({ \
	45020	int16x4_t __ret_341; \
	45021	int16x4_t __s0_341 = __p0_341; \
	45022	int16x8_t __s2_341 = __p2_341; \
	45023	__ret_341 = vset_lane_s16(vgetq_lane_s16(__s2_341, __p3_341), __s0_341, __p1_341); \
	45024	__ret_341; \
44547	45025	})
44548	45026	#else
44549		#define vcvtq_n_u16_f16(__p0, __p1) __extension__ ({ \
44550		uint16x8_t __ret; \
44551		float16x8_t __s0 = __p0; \
44552		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44553		__ret = (uint16x8_t) __builtin_neon_vcvtq_n_u16_v((int8x16_t)__rev0, __p1, 49); \
44554		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
44555		__ret; \
	45027	#define vcopy_laneq_s16(__p0_342, __p1_342, __p2_342, __p3_342) __extension__ ({ \
	45028	int16x4_t __ret_342; \
	45029	int16x4_t __s0_342 = __p0_342; \
	45030	int16x8_t __s2_342 = __p2_342; \
	45031	int16x4_t __rev0_342; __rev0_342 = __builtin_shufflevector(__s0_342, __s0_342, 3, 2, 1, 0); \
	45032	int16x8_t __rev2_342; __rev2_342 = __builtin_shufflevector(__s2_342, __s2_342, 7, 6, 5, 4, 3, 2, 1, 0); \
	45033	__ret_342 = __noswap_vset_lane_s16(__noswap_vgetq_lane_s16(__rev2_342, __p3_342), __rev0_342, __p1_342); \
	45034	__ret_342 = __builtin_shufflevector(__ret_342, __ret_342, 3, 2, 1, 0); \
	45035	__ret_342; \
44556	45036	})
44557	45037	#endif
44558	45038
44559		#ifdef __LITTLE_ENDIAN__
44560		#define vcvt_n_u16_f16(__p0, __p1) __extension__ ({ \
44561		uint16x4_t __ret; \
44562		float16x4_t __s0 = __p0; \
44563		__ret = (uint16x4_t) __builtin_neon_vcvt_n_u16_v((int8x8_t)__s0, __p1, 17); \
	45039	#define vcreate_p64(__p0) __extension__ ({ \
	45040	poly64x1_t __ret; \
	45041	uint64_t __promote = __p0; \
	45042	__ret = (poly64x1_t)(__promote); \
44564	45043	__ret; \
44565	45044	})
44566		#else
44567		#define vcvt_n_u16_f16(__p0, __p1) __extension__ ({ \
44568		uint16x4_t __ret; \
44569		float16x4_t __s0 = __p0; \
44570		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44571		__ret = (uint16x4_t) __builtin_neon_vcvt_n_u16_v((int8x8_t)__rev0, __p1, 17); \
44572		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	45045	#define vcreate_f64(__p0) __extension__ ({ \
	45046	float64x1_t __ret; \
	45047	uint64_t __promote = __p0; \
	45048	__ret = (float64x1_t)(__promote); \
44573	45049	__ret; \
44574	45050	})
44575		#endif
44576
44577		#ifdef __LITTLE_ENDIAN__
44578		__ai int16x8_t vcvtq_s16_f16(float16x8_t __p0) {
44579		int16x8_t __ret;
44580		__ret = (int16x8_t) __builtin_neon_vcvtq_s16_v((int8x16_t)__p0, 33);
	45051	__ai float32_t vcvts_f32_s32(int32_t __p0) {
	45052	float32_t __ret;
	45053	__ret = (float32_t) __builtin_neon_vcvts_f32_s32(__p0);
44581	45054	return __ret;
44582	45055	}
44583		#else
44584		__ai int16x8_t vcvtq_s16_f16(float16x8_t __p0) {
44585		int16x8_t __ret;
44586		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44587		__ret = (int16x8_t) __builtin_neon_vcvtq_s16_v((int8x16_t)__rev0, 33);
44588		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45056	__ai float32_t vcvts_f32_u32(uint32_t __p0) {
	45057	float32_t __ret;
	45058	__ret = (float32_t) __builtin_neon_vcvts_f32_u32(__p0);
44589	45059	return __ret;
44590	45060	}
44591		#endif
44592
44593	45061	#ifdef __LITTLE_ENDIAN__
44594		__ai int16x4_t vcvt_s16_f16(float16x4_t __p0) {
44595		int16x4_t __ret;
44596		__ret = (int16x4_t) __builtin_neon_vcvt_s16_v((int8x8_t)__p0, 1);
	45062	__ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
	45063	float32x2_t __ret;
	45064	__ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__p0, 9);
44597	45065	return __ret;
44598	45066	}
44599	45067	#else
44600		__ai int16x4_t vcvt_s16_f16(float16x4_t __p0) {
44601		int16x4_t __ret;
44602		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44603		__ret = (int16x4_t) __builtin_neon_vcvt_s16_v((int8x8_t)__rev0, 1);
44604		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45068	__ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
	45069	float32x2_t __ret;
	45070	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45071	__ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__rev0, 9);
	45072	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	45073	return __ret;
	45074	}
	45075	__ai float32x2_t __noswap_vcvt_f32_f64(float64x2_t __p0) {
	45076	float32x2_t __ret;
	45077	__ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__p0, 9);
44605	45078	return __ret;
44606	45079	}
44607	45080	#endif
44608	45081
44609		#ifdef __LITTLE_ENDIAN__
44610		__ai uint16x8_t vcvtq_u16_f16(float16x8_t __p0) {
44611		uint16x8_t __ret;
44612		__ret = (uint16x8_t) __builtin_neon_vcvtq_u16_v((int8x16_t)__p0, 49);
	45082	__ai float64_t vcvtd_f64_s64(int64_t __p0) {
	45083	float64_t __ret;
	45084	__ret = (float64_t) __builtin_neon_vcvtd_f64_s64(__p0);
44613	45085	return __ret;
44614	45086	}
44615		#else
44616		__ai uint16x8_t vcvtq_u16_f16(float16x8_t __p0) {
44617		uint16x8_t __ret;
44618		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44619		__ret = (uint16x8_t) __builtin_neon_vcvtq_u16_v((int8x16_t)__rev0, 49);
44620		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45087	__ai float64_t vcvtd_f64_u64(uint64_t __p0) {
	45088	float64_t __ret;
	45089	__ret = (float64_t) __builtin_neon_vcvtd_f64_u64(__p0);
44621	45090	return __ret;
44622	45091	}
44623		#endif
44624
44625	45092	#ifdef __LITTLE_ENDIAN__
44626		__ai uint16x4_t vcvt_u16_f16(float16x4_t __p0) {
44627		uint16x4_t __ret;
44628		__ret = (uint16x4_t) __builtin_neon_vcvt_u16_v((int8x8_t)__p0, 17);
	45093	__ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
	45094	float64x2_t __ret;
	45095	__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__p0, 51);
44629	45096	return __ret;
44630	45097	}
44631	45098	#else
44632		__ai uint16x4_t vcvt_u16_f16(float16x4_t __p0) {
44633		uint16x4_t __ret;
44634		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44635		__ret = (uint16x4_t) __builtin_neon_vcvt_u16_v((int8x8_t)__rev0, 17);
44636		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45099	__ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
	45100	float64x2_t __ret;
	45101	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45102	__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__rev0, 51);
	45103	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44637	45104	return __ret;
44638	45105	}
44639	45106	#endif
44640	45107
44641	45108	#ifdef __LITTLE_ENDIAN__
44642		__ai int16x8_t vcvtaq_s16_f16(float16x8_t __p0) {
44643		int16x8_t __ret;
44644		__ret = (int16x8_t) __builtin_neon_vcvtaq_s16_v((int8x16_t)__p0, 33);
	45109	__ai float64x2_t vcvtq_f64_s64(int64x2_t __p0) {
	45110	float64x2_t __ret;
	45111	__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__p0, 35);
44645	45112	return __ret;
44646	45113	}
44647	45114	#else
44648		__ai int16x8_t vcvtaq_s16_f16(float16x8_t __p0) {
44649		int16x8_t __ret;
44650		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44651		__ret = (int16x8_t) __builtin_neon_vcvtaq_s16_v((int8x16_t)__rev0, 33);
44652		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45115	__ai float64x2_t vcvtq_f64_s64(int64x2_t __p0) {
	45116	float64x2_t __ret;
	45117	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45118	__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__rev0, 35);
	45119	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44653	45120	return __ret;
44654	45121	}
44655	45122	#endif
44656	45123
44657		#ifdef __LITTLE_ENDIAN__
44658		__ai int16x4_t vcvta_s16_f16(float16x4_t __p0) {
44659		int16x4_t __ret;
44660		__ret = (int16x4_t) __builtin_neon_vcvta_s16_v((int8x8_t)__p0, 1);
	45124	__ai float64x1_t vcvt_f64_u64(uint64x1_t __p0) {
	45125	float64x1_t __ret;
	45126	__ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 19);
44661	45127	return __ret;
44662	45128	}
44663		#else
44664		__ai int16x4_t vcvta_s16_f16(float16x4_t __p0) {
44665		int16x4_t __ret;
44666		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44667		__ret = (int16x4_t) __builtin_neon_vcvta_s16_v((int8x8_t)__rev0, 1);
44668		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45129	__ai float64x1_t vcvt_f64_s64(int64x1_t __p0) {
	45130	float64x1_t __ret;
	45131	__ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 3);
44669	45132	return __ret;
44670	45133	}
44671		#endif
44672
44673	45134	#ifdef __LITTLE_ENDIAN__
44674		__ai uint16x8_t vcvtaq_u16_f16(float16x8_t __p0) {
44675		uint16x8_t __ret;
44676		__ret = (uint16x8_t) __builtin_neon_vcvtaq_u16_v((int8x16_t)__p0, 49);
	45135	__ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
	45136	float64x2_t __ret;
	45137	__ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__p0, 42);
44677	45138	return __ret;
44678	45139	}
44679	45140	#else
44680		__ai uint16x8_t vcvtaq_u16_f16(float16x8_t __p0) {
44681		uint16x8_t __ret;
44682		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44683		__ret = (uint16x8_t) __builtin_neon_vcvtaq_u16_v((int8x16_t)__rev0, 49);
44684		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45141	__ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
	45142	float64x2_t __ret;
	45143	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45144	__ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__rev0, 42);
	45145	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	45146	return __ret;
	45147	}
	45148	__ai float64x2_t __noswap_vcvt_f64_f32(float32x2_t __p0) {
	45149	float64x2_t __ret;
	45150	__ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__p0, 42);
44685	45151	return __ret;
44686	45152	}
44687	45153	#endif
44688	45154
44689	45155	#ifdef __LITTLE_ENDIAN__
44690		__ai uint16x4_t vcvta_u16_f16(float16x4_t __p0) {
44691		uint16x4_t __ret;
44692		__ret = (uint16x4_t) __builtin_neon_vcvta_u16_v((int8x8_t)__p0, 17);
	45156	__ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
	45157	float16x8_t __ret;
	45158	__ret = vcombine_f16(__p0, vcvt_f16_f32(__p1));
44693	45159	return __ret;
44694	45160	}
44695	45161	#else
44696		__ai uint16x4_t vcvta_u16_f16(float16x4_t __p0) {
44697		uint16x4_t __ret;
	45162	__ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
	45163	float16x8_t __ret;
44698	45164	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44699		__ret = (uint16x4_t) __builtin_neon_vcvta_u16_v((int8x8_t)__rev0, 17);
44700		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45165	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	45166	__ret = __noswap_vcombine_f16(__rev0, __noswap_vcvt_f16_f32(__rev1));
	45167	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44701	45168	return __ret;
44702	45169	}
44703	45170	#endif
44704	45171
44705	45172	#ifdef __LITTLE_ENDIAN__
44706		__ai int16x8_t vcvtmq_s16_f16(float16x8_t __p0) {
44707		int16x8_t __ret;
44708		__ret = (int16x8_t) __builtin_neon_vcvtmq_s16_v((int8x16_t)__p0, 33);
	45173	__ai float32x4_t vcvt_high_f32_f16(float16x8_t __p0) {
	45174	float32x4_t __ret;
	45175	__ret = vcvt_f32_f16(vget_high_f16(__p0));
44709	45176	return __ret;
44710	45177	}
44711	45178	#else
44712		__ai int16x8_t vcvtmq_s16_f16(float16x8_t __p0) {
44713		int16x8_t __ret;
	45179	__ai float32x4_t vcvt_high_f32_f16(float16x8_t __p0) {
	45180	float32x4_t __ret;
44714	45181	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44715		__ret = (int16x8_t) __builtin_neon_vcvtmq_s16_v((int8x16_t)__rev0, 33);
44716		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45182	__ret = __noswap_vcvt_f32_f16(__noswap_vget_high_f16(__rev0));
	45183	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44717	45184	return __ret;
44718	45185	}
44719	45186	#endif
44720	45187
44721	45188	#ifdef __LITTLE_ENDIAN__
44722		__ai int16x4_t vcvtm_s16_f16(float16x4_t __p0) {
44723		int16x4_t __ret;
44724		__ret = (int16x4_t) __builtin_neon_vcvtm_s16_v((int8x8_t)__p0, 1);
	45189	__ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
	45190	float32x4_t __ret;
	45191	__ret = vcombine_f32(__p0, vcvt_f32_f64(__p1));
44725	45192	return __ret;
44726	45193	}
44727	45194	#else
44728		__ai int16x4_t vcvtm_s16_f16(float16x4_t __p0) {
44729		int16x4_t __ret;
44730		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44731		__ret = (int16x4_t) __builtin_neon_vcvtm_s16_v((int8x8_t)__rev0, 1);
	45195	__ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
	45196	float32x4_t __ret;
	45197	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45198	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	45199	__ret = __noswap_vcombine_f32(__rev0, __noswap_vcvt_f32_f64(__rev1));
44732	45200	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44733	45201	return __ret;
44734	45202	}
44735	45203	#endif
44736	45204
44737	45205	#ifdef __LITTLE_ENDIAN__
44738		__ai uint16x8_t vcvtmq_u16_f16(float16x8_t __p0) {
44739		uint16x8_t __ret;
44740		__ret = (uint16x8_t) __builtin_neon_vcvtmq_u16_v((int8x16_t)__p0, 49);
	45206	__ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
	45207	float64x2_t __ret;
	45208	__ret = vcvt_f64_f32(vget_high_f32(__p0));
44741	45209	return __ret;
44742	45210	}
44743	45211	#else
44744		__ai uint16x8_t vcvtmq_u16_f16(float16x8_t __p0) {
44745		uint16x8_t __ret;
44746		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44747		__ret = (uint16x8_t) __builtin_neon_vcvtmq_u16_v((int8x16_t)__rev0, 49);
44748		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45212	__ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
	45213	float64x2_t __ret;
	45214	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	45215	__ret = __noswap_vcvt_f64_f32(__noswap_vget_high_f32(__rev0));
	45216	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44749	45217	return __ret;
44750	45218	}
44751	45219	#endif
44752	45220
	45221	#define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
	45222	float32_t __ret; \
	45223	uint32_t __s0 = __p0; \
	45224	__ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
	45225	__ret; \
	45226	})
	45227	#define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
	45228	float32_t __ret; \
	45229	int32_t __s0 = __p0; \
	45230	__ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
	45231	__ret; \
	45232	})
44753	45233	#ifdef __LITTLE_ENDIAN__
44754		__ai uint16x4_t vcvtm_u16_f16(float16x4_t __p0) {
44755		uint16x4_t __ret;
44756		__ret = (uint16x4_t) __builtin_neon_vcvtm_u16_v((int8x8_t)__p0, 17);
44757		return __ret;
44758		}
	45234	#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
	45235	float64x2_t __ret; \
	45236	uint64x2_t __s0 = __p0; \
	45237	__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 51); \
	45238	__ret; \
	45239	})
44759	45240	#else
44760		__ai uint16x4_t vcvtm_u16_f16(float16x4_t __p0) {
44761		uint16x4_t __ret;
44762		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44763		__ret = (uint16x4_t) __builtin_neon_vcvtm_u16_v((int8x8_t)__rev0, 17);
44764		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44765		return __ret;
44766		}
	45241	#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
	45242	float64x2_t __ret; \
	45243	uint64x2_t __s0 = __p0; \
	45244	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45245	__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 51); \
	45246	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	45247	__ret; \
	45248	})
44767	45249	#endif
44768	45250
44769	45251	#ifdef __LITTLE_ENDIAN__
44770		__ai int16x8_t vcvtnq_s16_f16(float16x8_t __p0) {
44771		int16x8_t __ret;
44772		__ret = (int16x8_t) __builtin_neon_vcvtnq_s16_v((int8x16_t)__p0, 33);
44773		return __ret;
44774		}
	45252	#define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
	45253	float64x2_t __ret; \
	45254	int64x2_t __s0 = __p0; \
	45255	__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 35); \
	45256	__ret; \
	45257	})
44775	45258	#else
44776		__ai int16x8_t vcvtnq_s16_f16(float16x8_t __p0) {
44777		int16x8_t __ret;
44778		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44779		__ret = (int16x8_t) __builtin_neon_vcvtnq_s16_v((int8x16_t)__rev0, 33);
44780		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44781		return __ret;
44782		}
	45259	#define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
	45260	float64x2_t __ret; \
	45261	int64x2_t __s0 = __p0; \
	45262	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45263	__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 35); \
	45264	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	45265	__ret; \
	45266	})
	45267	#endif
	45268
	45269	#define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
	45270	float64x1_t __ret; \
	45271	uint64x1_t __s0 = __p0; \
	45272	__ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
	45273	__ret; \
	45274	})
	45275	#define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
	45276	float64x1_t __ret; \
	45277	int64x1_t __s0 = __p0; \
	45278	__ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
	45279	__ret; \
	45280	})
	45281	#define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
	45282	float64_t __ret; \
	45283	uint64_t __s0 = __p0; \
	45284	__ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
	45285	__ret; \
	45286	})
	45287	#define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
	45288	float64_t __ret; \
	45289	int64_t __s0 = __p0; \
	45290	__ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
	45291	__ret; \
	45292	})
	45293	#define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
	45294	int32_t __ret; \
	45295	float32_t __s0 = __p0; \
	45296	__ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
	45297	__ret; \
	45298	})
	45299	#ifdef __LITTLE_ENDIAN__
	45300	#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
	45301	int64x2_t __ret; \
	45302	float64x2_t __s0 = __p0; \
	45303	__ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__s0, __p1, 35); \
	45304	__ret; \
	45305	})
	45306	#else
	45307	#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
	45308	int64x2_t __ret; \
	45309	float64x2_t __s0 = __p0; \
	45310	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45311	__ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__rev0, __p1, 35); \
	45312	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	45313	__ret; \
	45314	})
44783	45315	#endif
44784	45316
	45317	#define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
	45318	int64x1_t __ret; \
	45319	float64x1_t __s0 = __p0; \
	45320	__ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
	45321	__ret; \
	45322	})
	45323	#define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
	45324	int64_t __ret; \
	45325	float64_t __s0 = __p0; \
	45326	__ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
	45327	__ret; \
	45328	})
	45329	#define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
	45330	uint32_t __ret; \
	45331	float32_t __s0 = __p0; \
	45332	__ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
	45333	__ret; \
	45334	})
44785	45335	#ifdef __LITTLE_ENDIAN__
44786		__ai int16x4_t vcvtn_s16_f16(float16x4_t __p0) {
44787		int16x4_t __ret;
44788		__ret = (int16x4_t) __builtin_neon_vcvtn_s16_v((int8x8_t)__p0, 1);
	45336	#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
	45337	uint64x2_t __ret; \
	45338	float64x2_t __s0 = __p0; \
	45339	__ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__s0, __p1, 51); \
	45340	__ret; \
	45341	})
	45342	#else
	45343	#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
	45344	uint64x2_t __ret; \
	45345	float64x2_t __s0 = __p0; \
	45346	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45347	__ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__rev0, __p1, 51); \
	45348	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	45349	__ret; \
	45350	})
	45351	#endif
	45352
	45353	#define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
	45354	uint64x1_t __ret; \
	45355	float64x1_t __s0 = __p0; \
	45356	__ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
	45357	__ret; \
	45358	})
	45359	#define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
	45360	uint64_t __ret; \
	45361	float64_t __s0 = __p0; \
	45362	__ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
	45363	__ret; \
	45364	})
	45365	__ai int32_t vcvts_s32_f32(float32_t __p0) {
	45366	int32_t __ret;
	45367	__ret = (int32_t) __builtin_neon_vcvts_s32_f32(__p0);
44789	45368	return __ret;
44790	45369	}
44791		#else
44792		__ai int16x4_t vcvtn_s16_f16(float16x4_t __p0) {
44793		int16x4_t __ret;
44794		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44795		__ret = (int16x4_t) __builtin_neon_vcvtn_s16_v((int8x8_t)__rev0, 1);
44796		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45370	__ai int64_t vcvtd_s64_f64(float64_t __p0) {
	45371	int64_t __ret;
	45372	__ret = (int64_t) __builtin_neon_vcvtd_s64_f64(__p0);
44797	45373	return __ret;
44798	45374	}
44799		#endif
44800
44801	45375	#ifdef __LITTLE_ENDIAN__
44802		__ai uint16x8_t vcvtnq_u16_f16(float16x8_t __p0) {
44803		uint16x8_t __ret;
44804		__ret = (uint16x8_t) __builtin_neon_vcvtnq_u16_v((int8x16_t)__p0, 49);
	45376	__ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
	45377	int64x2_t __ret;
	45378	__ret = (int64x2_t) __builtin_neon_vcvtq_s64_v((int8x16_t)__p0, 35);
44805	45379	return __ret;
44806	45380	}
44807	45381	#else
44808		__ai uint16x8_t vcvtnq_u16_f16(float16x8_t __p0) {
44809		uint16x8_t __ret;
44810		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44811		__ret = (uint16x8_t) __builtin_neon_vcvtnq_u16_v((int8x16_t)__rev0, 49);
44812		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45382	__ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
	45383	int64x2_t __ret;
	45384	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45385	__ret = (int64x2_t) __builtin_neon_vcvtq_s64_v((int8x16_t)__rev0, 35);
	45386	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44813	45387	return __ret;
44814	45388	}
44815	45389	#endif
44816	45390
44817		#ifdef __LITTLE_ENDIAN__
44818		__ai uint16x4_t vcvtn_u16_f16(float16x4_t __p0) {
44819		uint16x4_t __ret;
44820		__ret = (uint16x4_t) __builtin_neon_vcvtn_u16_v((int8x8_t)__p0, 17);
	45391	__ai int64x1_t vcvt_s64_f64(float64x1_t __p0) {
	45392	int64x1_t __ret;
	45393	__ret = (int64x1_t) __builtin_neon_vcvt_s64_v((int8x8_t)__p0, 3);
44821	45394	return __ret;
44822	45395	}
44823		#else
44824		__ai uint16x4_t vcvtn_u16_f16(float16x4_t __p0) {
44825		uint16x4_t __ret;
44826		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44827		__ret = (uint16x4_t) __builtin_neon_vcvtn_u16_v((int8x8_t)__rev0, 17);
44828		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45396	__ai uint32_t vcvts_u32_f32(float32_t __p0) {
	45397	uint32_t __ret;
	45398	__ret = (uint32_t) __builtin_neon_vcvts_u32_f32(__p0);
	45399	return __ret;
	45400	}
	45401	__ai uint64_t vcvtd_u64_f64(float64_t __p0) {
	45402	uint64_t __ret;
	45403	__ret = (uint64_t) __builtin_neon_vcvtd_u64_f64(__p0);
44829	45404	return __ret;
44830	45405	}
44831		#endif
44832
44833	45406	#ifdef __LITTLE_ENDIAN__
44834		__ai int16x8_t vcvtpq_s16_f16(float16x8_t __p0) {
44835		int16x8_t __ret;
44836		__ret = (int16x8_t) __builtin_neon_vcvtpq_s16_v((int8x16_t)__p0, 33);
	45407	__ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
	45408	uint64x2_t __ret;
	45409	__ret = (uint64x2_t) __builtin_neon_vcvtq_u64_v((int8x16_t)__p0, 51);
44837	45410	return __ret;
44838	45411	}
44839	45412	#else
44840		__ai int16x8_t vcvtpq_s16_f16(float16x8_t __p0) {
44841		int16x8_t __ret;
44842		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44843		__ret = (int16x8_t) __builtin_neon_vcvtpq_s16_v((int8x16_t)__rev0, 33);
44844		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45413	__ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
	45414	uint64x2_t __ret;
	45415	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45416	__ret = (uint64x2_t) __builtin_neon_vcvtq_u64_v((int8x16_t)__rev0, 51);
	45417	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44845	45418	return __ret;
44846	45419	}
44847	45420	#endif
44848	45421
44849		#ifdef __LITTLE_ENDIAN__
44850		__ai int16x4_t vcvtp_s16_f16(float16x4_t __p0) {
44851		int16x4_t __ret;
44852		__ret = (int16x4_t) __builtin_neon_vcvtp_s16_v((int8x8_t)__p0, 1);
	45422	__ai uint64x1_t vcvt_u64_f64(float64x1_t __p0) {
	45423	uint64x1_t __ret;
	45424	__ret = (uint64x1_t) __builtin_neon_vcvt_u64_v((int8x8_t)__p0, 19);
44853	45425	return __ret;
44854	45426	}
44855		#else
44856		__ai int16x4_t vcvtp_s16_f16(float16x4_t __p0) {
44857		int16x4_t __ret;
44858		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44859		__ret = (int16x4_t) __builtin_neon_vcvtp_s16_v((int8x8_t)__rev0, 1);
44860		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45427	__ai int32_t vcvtas_s32_f32(float32_t __p0) {
	45428	int32_t __ret;
	45429	__ret = (int32_t) __builtin_neon_vcvtas_s32_f32(__p0);
44861	45430	return __ret;
44862	45431	}
44863		#endif
44864
44865	45432	#ifdef __LITTLE_ENDIAN__
44866		__ai uint16x8_t vcvtpq_u16_f16(float16x8_t __p0) {
44867		uint16x8_t __ret;
44868		__ret = (uint16x8_t) __builtin_neon_vcvtpq_u16_v((int8x16_t)__p0, 49);
	45433	__ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
	45434	int64x2_t __ret;
	45435	__ret = (int64x2_t) __builtin_neon_vcvtaq_s64_v((int8x16_t)__p0, 35);
44869	45436	return __ret;
44870	45437	}
44871	45438	#else
44872		__ai uint16x8_t vcvtpq_u16_f16(float16x8_t __p0) {
44873		uint16x8_t __ret;
44874		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44875		__ret = (uint16x8_t) __builtin_neon_vcvtpq_u16_v((int8x16_t)__rev0, 49);
44876		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45439	__ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
	45440	int64x2_t __ret;
	45441	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45442	__ret = (int64x2_t) __builtin_neon_vcvtaq_s64_v((int8x16_t)__rev0, 35);
	45443	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44877	45444	return __ret;
44878	45445	}
44879	45446	#endif
44880	45447
44881		#ifdef __LITTLE_ENDIAN__
44882		__ai uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
44883		uint16x4_t __ret;
44884		__ret = (uint16x4_t) __builtin_neon_vcvtp_u16_v((int8x8_t)__p0, 17);
	45448	__ai int64x1_t vcvta_s64_f64(float64x1_t __p0) {
	45449	int64x1_t __ret;
	45450	__ret = (int64x1_t) __builtin_neon_vcvta_s64_v((int8x8_t)__p0, 3);
44885	45451	return __ret;
44886	45452	}
44887		#else
44888		__ai uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
44889		uint16x4_t __ret;
44890		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44891		__ret = (uint16x4_t) __builtin_neon_vcvtp_u16_v((int8x8_t)__rev0, 17);
44892		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45453	__ai int64_t vcvtad_s64_f64(float64_t __p0) {
	45454	int64_t __ret;
	45455	__ret = (int64_t) __builtin_neon_vcvtad_s64_f64(__p0);
	45456	return __ret;
	45457	}
	45458	__ai uint32_t vcvtas_u32_f32(float32_t __p0) {
	45459	uint32_t __ret;
	45460	__ret = (uint32_t) __builtin_neon_vcvtas_u32_f32(__p0);
44893	45461	return __ret;
44894	45462	}
44895		#endif
44896
44897		#ifdef __LITTLE_ENDIAN__
44898		#define vextq_f16(__p0, __p1, __p2) __extension__ ({ \
44899		float16x8_t __ret; \
44900		float16x8_t __s0 = __p0; \
44901		float16x8_t __s1 = __p1; \
44902		__ret = (float16x8_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 40); \
44903		__ret; \
44904		})
44905		#else
44906		#define vextq_f16(__p0, __p1, __p2) __extension__ ({ \
44907		float16x8_t __ret; \
44908		float16x8_t __s0 = __p0; \
44909		float16x8_t __s1 = __p1; \
44910		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44911		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
44912		__ret = (float16x8_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 40); \
44913		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
44914		__ret; \
44915		})
44916		#endif
44917
44918	45463	#ifdef __LITTLE_ENDIAN__
44919		#define vext_f16(__p0, __p1, __p2) __extension__ ({ \
44920		float16x4_t __ret; \
44921		float16x4_t __s0 = __p0; \
44922		float16x4_t __s1 = __p1; \
44923		__ret = (float16x4_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 8); \
44924		__ret; \
44925		})
	45464	__ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
	45465	uint64x2_t __ret;
	45466	__ret = (uint64x2_t) __builtin_neon_vcvtaq_u64_v((int8x16_t)__p0, 51);
	45467	return __ret;
	45468	}
44926	45469	#else
44927		#define vext_f16(__p0, __p1, __p2) __extension__ ({ \
44928		float16x4_t __ret; \
44929		float16x4_t __s0 = __p0; \
44930		float16x4_t __s1 = __p1; \
44931		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44932		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
44933		__ret = (float16x4_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 8); \
44934		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
44935		__ret; \
44936		})
	45470	__ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
	45471	uint64x2_t __ret;
	45472	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45473	__ret = (uint64x2_t) __builtin_neon_vcvtaq_u64_v((int8x16_t)__rev0, 51);
	45474	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	45475	return __ret;
	45476	}
44937	45477	#endif
44938	45478
44939		#ifdef __LITTLE_ENDIAN__
44940		__ai float16x8_t vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44941		float16x8_t __ret;
44942		__ret = (float16x8_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	45479	__ai uint64x1_t vcvta_u64_f64(float64x1_t __p0) {
	45480	uint64x1_t __ret;
	45481	__ret = (uint64x1_t) __builtin_neon_vcvta_u64_v((int8x8_t)__p0, 19);
44943	45482	return __ret;
44944	45483	}
44945		#else
44946		__ai float16x8_t vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44947		float16x8_t __ret;
44948		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44949		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44950		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
44951		__ret = (float16x8_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
44952		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45484	__ai uint64_t vcvtad_u64_f64(float64_t __p0) {
	45485	uint64_t __ret;
	45486	__ret = (uint64_t) __builtin_neon_vcvtad_u64_f64(__p0);
	45487	return __ret;
	45488	}
	45489	__ai int32_t vcvtms_s32_f32(float32_t __p0) {
	45490	int32_t __ret;
	45491	__ret = (int32_t) __builtin_neon_vcvtms_s32_f32(__p0);
	45492	return __ret;
	45493	}
	45494	#ifdef __LITTLE_ENDIAN__
	45495	__ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
	45496	int64x2_t __ret;
	45497	__ret = (int64x2_t) __builtin_neon_vcvtmq_s64_v((int8x16_t)__p0, 35);
44953	45498	return __ret;
44954	45499	}
44955		__ai float16x8_t __noswap_vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44956		float16x8_t __ret;
44957		__ret = (float16x8_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 40);
	45500	#else
	45501	__ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
	45502	int64x2_t __ret;
	45503	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45504	__ret = (int64x2_t) __builtin_neon_vcvtmq_s64_v((int8x16_t)__rev0, 35);
	45505	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44958	45506	return __ret;
44959	45507	}
44960	45508	#endif
44961	45509
44962		#ifdef __LITTLE_ENDIAN__
44963		__ai float16x4_t vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
44964		float16x4_t __ret;
44965		__ret = (float16x4_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	45510	__ai int64x1_t vcvtm_s64_f64(float64x1_t __p0) {
	45511	int64x1_t __ret;
	45512	__ret = (int64x1_t) __builtin_neon_vcvtm_s64_v((int8x8_t)__p0, 3);
44966	45513	return __ret;
44967	45514	}
44968		#else
44969		__ai float16x4_t vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
44970		float16x4_t __ret;
44971		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44972		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44973		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
44974		__ret = (float16x4_t) __builtin_neon_vfma_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
44975		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45515	__ai int64_t vcvtmd_s64_f64(float64_t __p0) {
	45516	int64_t __ret;
	45517	__ret = (int64_t) __builtin_neon_vcvtmd_s64_f64(__p0);
44976	45518	return __ret;
44977	45519	}
44978		__ai float16x4_t __noswap_vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
44979		float16x4_t __ret;
44980		__ret = (float16x4_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 8);
	45520	__ai uint32_t vcvtms_u32_f32(float32_t __p0) {
	45521	uint32_t __ret;
	45522	__ret = (uint32_t) __builtin_neon_vcvtms_u32_f32(__p0);
44981	45523	return __ret;
44982	45524	}
44983		#endif
44984
44985	45525	#ifdef __LITTLE_ENDIAN__
44986		__ai float16x8_t vfmsq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44987		float16x8_t __ret;
44988		__ret = vfmaq_f16(__p0, -__p1, __p2);
	45526	__ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
	45527	uint64x2_t __ret;
	45528	__ret = (uint64x2_t) __builtin_neon_vcvtmq_u64_v((int8x16_t)__p0, 51);
44989	45529	return __ret;
44990	45530	}
44991	45531	#else
44992		__ai float16x8_t vfmsq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44993		float16x8_t __ret;
44994		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44995		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44996		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
44997		__ret = __noswap_vfmaq_f16(__rev0, -__rev1, __rev2);
44998		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45532	__ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
	45533	uint64x2_t __ret;
	45534	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45535	__ret = (uint64x2_t) __builtin_neon_vcvtmq_u64_v((int8x16_t)__rev0, 51);
	45536	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
44999	45537	return __ret;
45000	45538	}
45001	45539	#endif
45002	45540
45003		#ifdef __LITTLE_ENDIAN__
45004		__ai float16x4_t vfms_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
45005		float16x4_t __ret;
45006		__ret = vfma_f16(__p0, -__p1, __p2);
	45541	__ai uint64x1_t vcvtm_u64_f64(float64x1_t __p0) {
	45542	uint64x1_t __ret;
	45543	__ret = (uint64x1_t) __builtin_neon_vcvtm_u64_v((int8x8_t)__p0, 19);
45007	45544	return __ret;
45008	45545	}
45009		#else
45010		__ai float16x4_t vfms_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
45011		float16x4_t __ret;
45012		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45013		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45014		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
45015		__ret = __noswap_vfma_f16(__rev0, -__rev1, __rev2);
45016		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45546	__ai uint64_t vcvtmd_u64_f64(float64_t __p0) {
	45547	uint64_t __ret;
	45548	__ret = (uint64_t) __builtin_neon_vcvtmd_u64_f64(__p0);
	45549	return __ret;
	45550	}
	45551	__ai int32_t vcvtns_s32_f32(float32_t __p0) {
	45552	int32_t __ret;
	45553	__ret = (int32_t) __builtin_neon_vcvtns_s32_f32(__p0);
45017	45554	return __ret;
45018	45555	}
45019		#endif
45020
45021	45556	#ifdef __LITTLE_ENDIAN__
45022		__ai float16x8_t vmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
45023		float16x8_t __ret;
45024		__ret = (float16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	45557	__ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
	45558	int64x2_t __ret;
	45559	__ret = (int64x2_t) __builtin_neon_vcvtnq_s64_v((int8x16_t)__p0, 35);
45025	45560	return __ret;
45026	45561	}
45027	45562	#else
45028		__ai float16x8_t vmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
45029		float16x8_t __ret;
45030		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45031		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45032		__ret = (float16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45033		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45563	__ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
	45564	int64x2_t __ret;
	45565	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45566	__ret = (int64x2_t) __builtin_neon_vcvtnq_s64_v((int8x16_t)__rev0, 35);
	45567	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45034	45568	return __ret;
45035	45569	}
45036	45570	#endif
45037	45571
45038		#ifdef __LITTLE_ENDIAN__
45039		__ai float16x4_t vmax_f16(float16x4_t __p0, float16x4_t __p1) {
45040		float16x4_t __ret;
45041		__ret = (float16x4_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	45572	__ai int64x1_t vcvtn_s64_f64(float64x1_t __p0) {
	45573	int64x1_t __ret;
	45574	__ret = (int64x1_t) __builtin_neon_vcvtn_s64_v((int8x8_t)__p0, 3);
45042	45575	return __ret;
45043	45576	}
45044		#else
45045		__ai float16x4_t vmax_f16(float16x4_t __p0, float16x4_t __p1) {
45046		float16x4_t __ret;
45047		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45048		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45049		__ret = (float16x4_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45050		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45577	__ai int64_t vcvtnd_s64_f64(float64_t __p0) {
	45578	int64_t __ret;
	45579	__ret = (int64_t) __builtin_neon_vcvtnd_s64_f64(__p0);
	45580	return __ret;
	45581	}
	45582	__ai uint32_t vcvtns_u32_f32(float32_t __p0) {
	45583	uint32_t __ret;
	45584	__ret = (uint32_t) __builtin_neon_vcvtns_u32_f32(__p0);
45051	45585	return __ret;
45052	45586	}
45053		#endif
45054
45055	45587	#ifdef __LITTLE_ENDIAN__
45056		__ai float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
45057		float16x8_t __ret;
45058		__ret = (float16x8_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	45588	__ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
	45589	uint64x2_t __ret;
	45590	__ret = (uint64x2_t) __builtin_neon_vcvtnq_u64_v((int8x16_t)__p0, 51);
45059	45591	return __ret;
45060	45592	}
45061	45593	#else
45062		__ai float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
45063		float16x8_t __ret;
45064		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45065		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45066		__ret = (float16x8_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45067		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45594	__ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
	45595	uint64x2_t __ret;
	45596	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45597	__ret = (uint64x2_t) __builtin_neon_vcvtnq_u64_v((int8x16_t)__rev0, 51);
	45598	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45068	45599	return __ret;
45069	45600	}
45070	45601	#endif
45071	45602
45072		#ifdef __LITTLE_ENDIAN__
45073		__ai float16x4_t vmin_f16(float16x4_t __p0, float16x4_t __p1) {
45074		float16x4_t __ret;
45075		__ret = (float16x4_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	45603	__ai uint64x1_t vcvtn_u64_f64(float64x1_t __p0) {
	45604	uint64x1_t __ret;
	45605	__ret = (uint64x1_t) __builtin_neon_vcvtn_u64_v((int8x8_t)__p0, 19);
45076	45606	return __ret;
45077	45607	}
45078		#else
45079		__ai float16x4_t vmin_f16(float16x4_t __p0, float16x4_t __p1) {
45080		float16x4_t __ret;
45081		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45082		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45083		__ret = (float16x4_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45084		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45608	__ai uint64_t vcvtnd_u64_f64(float64_t __p0) {
	45609	uint64_t __ret;
	45610	__ret = (uint64_t) __builtin_neon_vcvtnd_u64_f64(__p0);
	45611	return __ret;
	45612	}
	45613	__ai int32_t vcvtps_s32_f32(float32_t __p0) {
	45614	int32_t __ret;
	45615	__ret = (int32_t) __builtin_neon_vcvtps_s32_f32(__p0);
45085	45616	return __ret;
45086	45617	}
45087		#endif
45088
45089	45618	#ifdef __LITTLE_ENDIAN__
45090		__ai float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
45091		float16x8_t __ret;
45092		__ret = __p0 * __p1;
	45619	__ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
	45620	int64x2_t __ret;
	45621	__ret = (int64x2_t) __builtin_neon_vcvtpq_s64_v((int8x16_t)__p0, 35);
45093	45622	return __ret;
45094	45623	}
45095	45624	#else
45096		__ai float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
45097		float16x8_t __ret;
45098		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45099		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45100		__ret = __rev0 * __rev1;
45101		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45625	__ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
	45626	int64x2_t __ret;
	45627	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45628	__ret = (int64x2_t) __builtin_neon_vcvtpq_s64_v((int8x16_t)__rev0, 35);
	45629	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45102	45630	return __ret;
45103	45631	}
45104	45632	#endif
45105	45633
45106		#ifdef __LITTLE_ENDIAN__
45107		__ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
45108		float16x4_t __ret;
45109		__ret = __p0 * __p1;
	45634	__ai int64x1_t vcvtp_s64_f64(float64x1_t __p0) {
	45635	int64x1_t __ret;
	45636	__ret = (int64x1_t) __builtin_neon_vcvtp_s64_v((int8x8_t)__p0, 3);
45110	45637	return __ret;
45111	45638	}
45112		#else
45113		__ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
45114		float16x4_t __ret;
45115		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45116		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45117		__ret = __rev0 * __rev1;
45118		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45639	__ai int64_t vcvtpd_s64_f64(float64_t __p0) {
	45640	int64_t __ret;
	45641	__ret = (int64_t) __builtin_neon_vcvtpd_s64_f64(__p0);
	45642	return __ret;
	45643	}
	45644	__ai uint32_t vcvtps_u32_f32(float32_t __p0) {
	45645	uint32_t __ret;
	45646	__ret = (uint32_t) __builtin_neon_vcvtps_u32_f32(__p0);
45119	45647	return __ret;
45120	45648	}
45121		#endif
45122
45123		#ifdef __LITTLE_ENDIAN__
45124		#define vmulq_lane_f16(__p0_277, __p1_277, __p2_277) __extension__ ({ \
45125		float16x8_t __ret_277; \
45126		float16x8_t __s0_277 = __p0_277; \
45127		float16x4_t __s1_277 = __p1_277; \
45128		__ret_277 = __s0_277 * splatq_lane_f16(__s1_277, __p2_277); \
45129		__ret_277; \
45130		})
45131		#else
45132		#define vmulq_lane_f16(__p0_278, __p1_278, __p2_278) __extension__ ({ \
45133		float16x8_t __ret_278; \
45134		float16x8_t __s0_278 = __p0_278; \
45135		float16x4_t __s1_278 = __p1_278; \
45136		float16x8_t __rev0_278; __rev0_278 = __builtin_shufflevector(__s0_278, __s0_278, 7, 6, 5, 4, 3, 2, 1, 0); \
45137		float16x4_t __rev1_278; __rev1_278 = __builtin_shufflevector(__s1_278, __s1_278, 3, 2, 1, 0); \
45138		__ret_278 = __rev0_278 * __noswap_splatq_lane_f16(__rev1_278, __p2_278); \
45139		__ret_278 = __builtin_shufflevector(__ret_278, __ret_278, 7, 6, 5, 4, 3, 2, 1, 0); \
45140		__ret_278; \
45141		})
45142		#endif
45143
45144		#ifdef __LITTLE_ENDIAN__
45145		#define vmul_lane_f16(__p0_279, __p1_279, __p2_279) __extension__ ({ \
45146		float16x4_t __ret_279; \
45147		float16x4_t __s0_279 = __p0_279; \
45148		float16x4_t __s1_279 = __p1_279; \
45149		__ret_279 = __s0_279 * splat_lane_f16(__s1_279, __p2_279); \
45150		__ret_279; \
45151		})
45152		#else
45153		#define vmul_lane_f16(__p0_280, __p1_280, __p2_280) __extension__ ({ \
45154		float16x4_t __ret_280; \
45155		float16x4_t __s0_280 = __p0_280; \
45156		float16x4_t __s1_280 = __p1_280; \
45157		float16x4_t __rev0_280; __rev0_280 = __builtin_shufflevector(__s0_280, __s0_280, 3, 2, 1, 0); \
45158		float16x4_t __rev1_280; __rev1_280 = __builtin_shufflevector(__s1_280, __s1_280, 3, 2, 1, 0); \
45159		__ret_280 = __rev0_280 * __noswap_splat_lane_f16(__rev1_280, __p2_280); \
45160		__ret_280 = __builtin_shufflevector(__ret_280, __ret_280, 3, 2, 1, 0); \
45161		__ret_280; \
45162		})
45163		#endif
45164
45165		#ifdef __LITTLE_ENDIAN__
45166		#define vmulq_n_f16(__p0, __p1) __extension__ ({ \
45167		float16x8_t __ret; \
45168		float16x8_t __s0 = __p0; \
45169		float16_t __s1 = __p1; \
45170		__ret = __s0 * (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}; \
45171		__ret; \
45172		})
45173		#else
45174		#define vmulq_n_f16(__p0, __p1) __extension__ ({ \
45175		float16x8_t __ret; \
45176		float16x8_t __s0 = __p0; \
45177		float16_t __s1 = __p1; \
45178		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45179		__ret = __rev0 * (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}; \
45180		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
45181		__ret; \
45182		})
45183		#endif
45184
45185		#ifdef __LITTLE_ENDIAN__
45186		#define vmul_n_f16(__p0, __p1) __extension__ ({ \
45187		float16x4_t __ret; \
45188		float16x4_t __s0 = __p0; \
45189		float16_t __s1 = __p1; \
45190		__ret = __s0 * (float16x4_t) {__s1, __s1, __s1, __s1}; \
45191		__ret; \
45192		})
45193		#else
45194		#define vmul_n_f16(__p0, __p1) __extension__ ({ \
45195		float16x4_t __ret; \
45196		float16x4_t __s0 = __p0; \
45197		float16_t __s1 = __p1; \
45198		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45199		__ret = __rev0 * (float16x4_t) {__s1, __s1, __s1, __s1}; \
45200		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
45201		__ret; \
45202		})
45203		#endif
45204
45205	45649	#ifdef __LITTLE_ENDIAN__
45206		__ai float16x8_t vnegq_f16(float16x8_t __p0) {
45207		float16x8_t __ret;
45208		__ret = -__p0;
	45650	__ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
	45651	uint64x2_t __ret;
	45652	__ret = (uint64x2_t) __builtin_neon_vcvtpq_u64_v((int8x16_t)__p0, 51);
45209	45653	return __ret;
45210	45654	}
45211	45655	#else
45212		__ai float16x8_t vnegq_f16(float16x8_t __p0) {
45213		float16x8_t __ret;
45214		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45215		__ret = -__rev0;
45216		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45656	__ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
	45657	uint64x2_t __ret;
	45658	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45659	__ret = (uint64x2_t) __builtin_neon_vcvtpq_u64_v((int8x16_t)__rev0, 51);
	45660	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45217	45661	return __ret;
45218	45662	}
45219	45663	#endif
45220	45664
	45665	__ai uint64x1_t vcvtp_u64_f64(float64x1_t __p0) {
	45666	uint64x1_t __ret;
	45667	__ret = (uint64x1_t) __builtin_neon_vcvtp_u64_v((int8x8_t)__p0, 19);
	45668	return __ret;
	45669	}
	45670	__ai uint64_t vcvtpd_u64_f64(float64_t __p0) {
	45671	uint64_t __ret;
	45672	__ret = (uint64_t) __builtin_neon_vcvtpd_u64_f64(__p0);
	45673	return __ret;
	45674	}
	45675	__ai float32_t vcvtxd_f32_f64(float64_t __p0) {
	45676	float32_t __ret;
	45677	__ret = (float32_t) __builtin_neon_vcvtxd_f32_f64(__p0);
	45678	return __ret;
	45679	}
45221	45680	#ifdef __LITTLE_ENDIAN__
45222		__ai float16x4_t vneg_f16(float16x4_t __p0) {
45223		float16x4_t __ret;
45224		__ret = -__p0;
	45681	__ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
	45682	float32x2_t __ret;
	45683	__ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__p0, 42);
45225	45684	return __ret;
45226	45685	}
45227	45686	#else
45228		__ai float16x4_t vneg_f16(float16x4_t __p0) {
45229		float16x4_t __ret;
45230		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45231		__ret = -__rev0;
45232		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45687	__ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
	45688	float32x2_t __ret;
	45689	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45690	__ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__rev0, 42);
	45691	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	45692	return __ret;
	45693	}
	45694	__ai float32x2_t __noswap_vcvtx_f32_f64(float64x2_t __p0) {
	45695	float32x2_t __ret;
	45696	__ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__p0, 42);
45233	45697	return __ret;
45234	45698	}
45235	45699	#endif
45236	45700
45237	45701	#ifdef __LITTLE_ENDIAN__
45238		__ai float16x4_t vpadd_f16(float16x4_t __p0, float16x4_t __p1) {
45239		float16x4_t __ret;
45240		__ret = (float16x4_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	45702	__ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
	45703	float32x4_t __ret;
	45704	__ret = vcombine_f32(__p0, vcvtx_f32_f64(__p1));
45241	45705	return __ret;
45242	45706	}
45243	45707	#else
45244		__ai float16x4_t vpadd_f16(float16x4_t __p0, float16x4_t __p1) {
45245		float16x4_t __ret;
45246		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45247		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45248		__ret = (float16x4_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	45708	__ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
	45709	float32x4_t __ret;
	45710	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45711	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	45712	__ret = __noswap_vcombine_f32(__rev0, __noswap_vcvtx_f32_f64(__rev1));
45249	45713	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45250	45714	return __ret;
45251	45715	}
45252	45716	#endif
45253	45717
45254	45718	#ifdef __LITTLE_ENDIAN__
45255		__ai float16x4_t vpmax_f16(float16x4_t __p0, float16x4_t __p1) {
45256		float16x4_t __ret;
45257		__ret = (float16x4_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	45719	__ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
	45720	float64x2_t __ret;
	45721	__ret = __p0 / __p1;
45258	45722	return __ret;
45259	45723	}
45260	45724	#else
45261		__ai float16x4_t vpmax_f16(float16x4_t __p0, float16x4_t __p1) {
45262		float16x4_t __ret;
45263		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45264		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45265		__ret = (float16x4_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45266		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	45725	__ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
	45726	float64x2_t __ret;
	45727	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45728	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	45729	__ret = __rev0 / __rev1;
	45730	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45267	45731	return __ret;
45268	45732	}
45269	45733	#endif
45270	45734
45271	45735	#ifdef __LITTLE_ENDIAN__
45272		__ai float16x4_t vpmin_f16(float16x4_t __p0, float16x4_t __p1) {
45273		float16x4_t __ret;
45274		__ret = (float16x4_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	45736	__ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
	45737	float32x4_t __ret;
	45738	__ret = __p0 / __p1;
45275	45739	return __ret;
45276	45740	}
45277	45741	#else
45278		__ai float16x4_t vpmin_f16(float16x4_t __p0, float16x4_t __p1) {
45279		float16x4_t __ret;
45280		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45281		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45282		__ret = (float16x4_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	45742	__ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
	45743	float32x4_t __ret;
	45744	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	45745	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	45746	__ret = __rev0 / __rev1;
45283	45747	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45284	45748	return __ret;
45285	45749	}
45286	45750	#endif
45287	45751
	45752	__ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
	45753	float64x1_t __ret;
	45754	__ret = __p0 / __p1;
	45755	return __ret;
	45756	}
45288	45757	#ifdef __LITTLE_ENDIAN__
45289		__ai float16x8_t vrecpeq_f16(float16x8_t __p0) {
45290		float16x8_t __ret;
45291		__ret = (float16x8_t) __builtin_neon_vrecpeq_v((int8x16_t)__p0, 40);
	45758	__ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
	45759	float32x2_t __ret;
	45760	__ret = __p0 / __p1;
45292	45761	return __ret;
45293	45762	}
45294	45763	#else
45295		__ai float16x8_t vrecpeq_f16(float16x8_t __p0) {
45296		float16x8_t __ret;
45297		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45298		__ret = (float16x8_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 40);
45299		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	45764	__ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
	45765	float32x2_t __ret;
	45766	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	45767	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	45768	__ret = __rev0 / __rev1;
	45769	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45300	45770	return __ret;
45301	45771	}
45302	45772	#endif
45303	45773
45304	45774	#ifdef __LITTLE_ENDIAN__
45305		__ai float16x4_t vrecpe_f16(float16x4_t __p0) {
45306		float16x4_t __ret;
45307		__ret = (float16x4_t) __builtin_neon_vrecpe_v((int8x8_t)__p0, 8);
45308		return __ret;
45309		}
	45775	#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
	45776	poly8_t __ret; \
	45777	poly8x8_t __s0 = __p0; \
	45778	__ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__s0, __p1); \
	45779	__ret; \
	45780	})
45310	45781	#else
45311		__ai float16x4_t vrecpe_f16(float16x4_t __p0) {
45312		float16x4_t __ret;
45313		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45314		__ret = (float16x4_t) __builtin_neon_vrecpe_v((int8x8_t)__rev0, 8);
45315		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45316		return __ret;
45317		}
	45782	#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
	45783	poly8_t __ret; \
	45784	poly8x8_t __s0 = __p0; \
	45785	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	45786	__ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__rev0, __p1); \
	45787	__ret; \
	45788	})
45318	45789	#endif
45319	45790
45320	45791	#ifdef __LITTLE_ENDIAN__
45321		__ai float16x8_t vrecpsq_f16(float16x8_t __p0, float16x8_t __p1) {
45322		float16x8_t __ret;
45323		__ret = (float16x8_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
45324		return __ret;
45325		}
	45792	#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
	45793	poly16_t __ret; \
	45794	poly16x4_t __s0 = __p0; \
	45795	__ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__s0, __p1); \
	45796	__ret; \
	45797	})
45326	45798	#else
45327		__ai float16x8_t vrecpsq_f16(float16x8_t __p0, float16x8_t __p1) {
45328		float16x8_t __ret;
45329		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45330		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45331		__ret = (float16x8_t) __builtin_neon_vrecpsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45332		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45333		return __ret;
45334		}
	45799	#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
	45800	poly16_t __ret; \
	45801	poly16x4_t __s0 = __p0; \
	45802	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	45803	__ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__rev0, __p1); \
	45804	__ret; \
	45805	})
45335	45806	#endif
45336	45807
45337	45808	#ifdef __LITTLE_ENDIAN__
45338		__ai float16x4_t vrecps_f16(float16x4_t __p0, float16x4_t __p1) {
45339		float16x4_t __ret;
45340		__ret = (float16x4_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
45341		return __ret;
45342		}
	45809	#define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
	45810	uint8_t __ret; \
	45811	uint8x8_t __s0 = __p0; \
	45812	__ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
	45813	__ret; \
	45814	})
45343	45815	#else
45344		__ai float16x4_t vrecps_f16(float16x4_t __p0, float16x4_t __p1) {
45345		float16x4_t __ret;
45346		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45347		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45348		__ret = (float16x4_t) __builtin_neon_vrecps_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45349		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45350		return __ret;
45351		}
	45816	#define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
	45817	uint8_t __ret; \
	45818	uint8x8_t __s0 = __p0; \
	45819	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	45820	__ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
	45821	__ret; \
	45822	})
45352	45823	#endif
45353	45824
45354	45825	#ifdef __LITTLE_ENDIAN__
45355		__ai float16x8_t vrev64q_f16(float16x8_t __p0) {
45356		float16x8_t __ret;
45357		__ret = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0, 7, 6, 5, 4);
45358		return __ret;
45359		}
	45826	#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
	45827	uint32_t __ret; \
	45828	uint32x2_t __s0 = __p0; \
	45829	__ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
	45830	__ret; \
	45831	})
45360	45832	#else
45361		__ai float16x8_t vrev64q_f16(float16x8_t __p0) {
45362		float16x8_t __ret;
45363		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45364		__ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
45365		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45366		return __ret;
45367		}
	45833	#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
	45834	uint32_t __ret; \
	45835	uint32x2_t __s0 = __p0; \
	45836	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45837	__ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
	45838	__ret; \
	45839	})
45368	45840	#endif
45369	45841
	45842	#define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
	45843	uint64_t __ret; \
	45844	uint64x1_t __s0 = __p0; \
	45845	__ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
	45846	__ret; \
	45847	})
45370	45848	#ifdef __LITTLE_ENDIAN__
45371		__ai float16x4_t vrev64_f16(float16x4_t __p0) {
45372		float16x4_t __ret;
45373		__ret = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45374		return __ret;
45375		}
	45849	#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
	45850	uint16_t __ret; \
	45851	uint16x4_t __s0 = __p0; \
	45852	__ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
	45853	__ret; \
	45854	})
45376	45855	#else
45377		__ai float16x4_t vrev64_f16(float16x4_t __p0) {
45378		float16x4_t __ret;
45379		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45380		__ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0);
45381		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45382		return __ret;
45383		}
	45856	#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
	45857	uint16_t __ret; \
	45858	uint16x4_t __s0 = __p0; \
	45859	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	45860	__ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
	45861	__ret; \
	45862	})
45384	45863	#endif
45385	45864
45386	45865	#ifdef __LITTLE_ENDIAN__
45387		__ai float16x8_t vrsqrteq_f16(float16x8_t __p0) {
45388		float16x8_t __ret;
45389		__ret = (float16x8_t) __builtin_neon_vrsqrteq_v((int8x16_t)__p0, 40);
45390		return __ret;
45391		}
	45866	#define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
	45867	int8_t __ret; \
	45868	int8x8_t __s0 = __p0; \
	45869	__ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
	45870	__ret; \
	45871	})
45392	45872	#else
45393		__ai float16x8_t vrsqrteq_f16(float16x8_t __p0) {
45394		float16x8_t __ret;
45395		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45396		__ret = (float16x8_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 40);
45397		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45398		return __ret;
45399		}
	45873	#define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
	45874	int8_t __ret; \
	45875	int8x8_t __s0 = __p0; \
	45876	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	45877	__ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
	45878	__ret; \
	45879	})
45400	45880	#endif
45401	45881
	45882	#define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
	45883	float64_t __ret; \
	45884	float64x1_t __s0 = __p0; \
	45885	__ret = (float64_t) __builtin_neon_vdupd_lane_f64((float64x1_t)__s0, __p1); \
	45886	__ret; \
	45887	})
45402	45888	#ifdef __LITTLE_ENDIAN__
45403		__ai float16x4_t vrsqrte_f16(float16x4_t __p0) {
45404		float16x4_t __ret;
45405		__ret = (float16x4_t) __builtin_neon_vrsqrte_v((int8x8_t)__p0, 8);
45406		return __ret;
45407		}
	45889	#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
	45890	float32_t __ret; \
	45891	float32x2_t __s0 = __p0; \
	45892	__ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__s0, __p1); \
	45893	__ret; \
	45894	})
45408	45895	#else
45409		__ai float16x4_t vrsqrte_f16(float16x4_t __p0) {
45410		float16x4_t __ret;
45411		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45412		__ret = (float16x4_t) __builtin_neon_vrsqrte_v((int8x8_t)__rev0, 8);
45413		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45414		return __ret;
45415		}
	45896	#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
	45897	float32_t __ret; \
	45898	float32x2_t __s0 = __p0; \
	45899	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45900	__ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__rev0, __p1); \
	45901	__ret; \
	45902	})
	45903	#endif
	45904
	45905	#ifdef __LITTLE_ENDIAN__
	45906	#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
	45907	int32_t __ret; \
	45908	int32x2_t __s0 = __p0; \
	45909	__ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
	45910	__ret; \
	45911	})
	45912	#else
	45913	#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
	45914	int32_t __ret; \
	45915	int32x2_t __s0 = __p0; \
	45916	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	45917	__ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
	45918	__ret; \
	45919	})
45416	45920	#endif
45417	45921
	45922	#define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
	45923	int64_t __ret; \
	45924	int64x1_t __s0 = __p0; \
	45925	__ret = (int64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
	45926	__ret; \
	45927	})
45418	45928	#ifdef __LITTLE_ENDIAN__
45419		__ai float16x8_t vrsqrtsq_f16(float16x8_t __p0, float16x8_t __p1) {
45420		float16x8_t __ret;
45421		__ret = (float16x8_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
45422		return __ret;
45423		}
	45929	#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
	45930	int16_t __ret; \
	45931	int16x4_t __s0 = __p0; \
	45932	__ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
	45933	__ret; \
	45934	})
45424	45935	#else
45425		__ai float16x8_t vrsqrtsq_f16(float16x8_t __p0, float16x8_t __p1) {
45426		float16x8_t __ret;
45427		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45428		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45429		__ret = (float16x8_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45430		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45431		return __ret;
45432		}
	45936	#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
	45937	int16_t __ret; \
	45938	int16x4_t __s0 = __p0; \
	45939	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	45940	__ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
	45941	__ret; \
	45942	})
45433	45943	#endif
45434	45944
	45945	#define vdup_lane_p64(__p0_343, __p1_343) __extension__ ({ \
	45946	poly64x1_t __ret_343; \
	45947	poly64x1_t __s0_343 = __p0_343; \
	45948	__ret_343 = splat_lane_p64(__s0_343, __p1_343); \
	45949	__ret_343; \
	45950	})
45435	45951	#ifdef __LITTLE_ENDIAN__
45436		__ai float16x4_t vrsqrts_f16(float16x4_t __p0, float16x4_t __p1) {
45437		float16x4_t __ret;
45438		__ret = (float16x4_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
45439		return __ret;
45440		}
	45952	#define vdupq_lane_p64(__p0_344, __p1_344) __extension__ ({ \
	45953	poly64x2_t __ret_344; \
	45954	poly64x1_t __s0_344 = __p0_344; \
	45955	__ret_344 = splatq_lane_p64(__s0_344, __p1_344); \
	45956	__ret_344; \
	45957	})
45441	45958	#else
45442		__ai float16x4_t vrsqrts_f16(float16x4_t __p0, float16x4_t __p1) {
45443		float16x4_t __ret;
45444		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45445		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45446		__ret = (float16x4_t) __builtin_neon_vrsqrts_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45447		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45448		return __ret;
45449		}
	45959	#define vdupq_lane_p64(__p0_345, __p1_345) __extension__ ({ \
	45960	poly64x2_t __ret_345; \
	45961	poly64x1_t __s0_345 = __p0_345; \
	45962	__ret_345 = __noswap_splatq_lane_p64(__s0_345, __p1_345); \
	45963	__ret_345 = __builtin_shufflevector(__ret_345, __ret_345, 1, 0); \
	45964	__ret_345; \
	45965	})
45450	45966	#endif
45451	45967
45452	45968	#ifdef __LITTLE_ENDIAN__
45453		__ai float16x8_t vsubq_f16(float16x8_t __p0, float16x8_t __p1) {
45454		float16x8_t __ret;
45455		__ret = __p0 - __p1;
45456		return __ret;
45457		}
	45969	#define vdupq_lane_f64(__p0_346, __p1_346) __extension__ ({ \
	45970	float64x2_t __ret_346; \
	45971	float64x1_t __s0_346 = __p0_346; \
	45972	__ret_346 = splatq_lane_f64(__s0_346, __p1_346); \
	45973	__ret_346; \
	45974	})
45458	45975	#else
45459		__ai float16x8_t vsubq_f16(float16x8_t __p0, float16x8_t __p1) {
45460		float16x8_t __ret;
45461		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45462		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45463		__ret = __rev0 - __rev1;
45464		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45465		return __ret;
45466		}
	45976	#define vdupq_lane_f64(__p0_347, __p1_347) __extension__ ({ \
	45977	float64x2_t __ret_347; \
	45978	float64x1_t __s0_347 = __p0_347; \
	45979	__ret_347 = __noswap_splatq_lane_f64(__s0_347, __p1_347); \
	45980	__ret_347 = __builtin_shufflevector(__ret_347, __ret_347, 1, 0); \
	45981	__ret_347; \
	45982	})
45467	45983	#endif
45468	45984
	45985	#define vdup_lane_f64(__p0_348, __p1_348) __extension__ ({ \
	45986	float64x1_t __ret_348; \
	45987	float64x1_t __s0_348 = __p0_348; \
	45988	__ret_348 = splat_lane_f64(__s0_348, __p1_348); \
	45989	__ret_348; \
	45990	})
45469	45991	#ifdef __LITTLE_ENDIAN__
45470		__ai float16x4_t vsub_f16(float16x4_t __p0, float16x4_t __p1) {
45471		float16x4_t __ret;
45472		__ret = __p0 - __p1;
45473		return __ret;
45474		}
	45992	#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
	45993	poly8_t __ret; \
	45994	poly8x16_t __s0 = __p0; \
	45995	__ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__s0, __p1); \
	45996	__ret; \
	45997	})
45475	45998	#else
45476		__ai float16x4_t vsub_f16(float16x4_t __p0, float16x4_t __p1) {
45477		float16x4_t __ret;
45478		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45479		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45480		__ret = __rev0 - __rev1;
45481		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45482		return __ret;
45483		}
	45999	#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
	46000	poly8_t __ret; \
	46001	poly8x16_t __s0 = __p0; \
	46002	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46003	__ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__rev0, __p1); \
	46004	__ret; \
	46005	})
45484	46006	#endif
45485	46007
45486	46008	#ifdef __LITTLE_ENDIAN__
45487		__ai float16x8x2_t vtrnq_f16(float16x8_t __p0, float16x8_t __p1) {
45488		float16x8x2_t __ret;
45489		__builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 40);
45490		return __ret;
45491		}
	46009	#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
	46010	poly16_t __ret; \
	46011	poly16x8_t __s0 = __p0; \
	46012	__ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__s0, __p1); \
	46013	__ret; \
	46014	})
45492	46015	#else
45493		__ai float16x8x2_t vtrnq_f16(float16x8_t __p0, float16x8_t __p1) {
45494		float16x8x2_t __ret;
45495		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45496		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45497		__builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45498
45499		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
45500		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
45501		return __ret;
45502		}
	46016	#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
	46017	poly16_t __ret; \
	46018	poly16x8_t __s0 = __p0; \
	46019	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	46020	__ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__rev0, __p1); \
	46021	__ret; \
	46022	})
45503	46023	#endif
45504	46024
45505	46025	#ifdef __LITTLE_ENDIAN__
45506		__ai float16x4x2_t vtrn_f16(float16x4_t __p0, float16x4_t __p1) {
45507		float16x4x2_t __ret;
45508		__builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 8);
45509		return __ret;
45510		}
	46026	#define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
	46027	uint8_t __ret; \
	46028	uint8x16_t __s0 = __p0; \
	46029	__ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
	46030	__ret; \
	46031	})
45511	46032	#else
45512		__ai float16x4x2_t vtrn_f16(float16x4_t __p0, float16x4_t __p1) {
45513		float16x4x2_t __ret;
45514		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45515		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45516		__builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45517
45518		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
45519		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0);
45520		return __ret;
45521		}
	46033	#define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
	46034	uint8_t __ret; \
	46035	uint8x16_t __s0 = __p0; \
	46036	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46037	__ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
	46038	__ret; \
	46039	})
45522	46040	#endif
45523	46041
45524	46042	#ifdef __LITTLE_ENDIAN__
45525		__ai float16x8x2_t vuzpq_f16(float16x8_t __p0, float16x8_t __p1) {
45526		float16x8x2_t __ret;
45527		__builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 40);
45528		return __ret;
45529		}
	46043	#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
	46044	uint32_t __ret; \
	46045	uint32x4_t __s0 = __p0; \
	46046	__ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
	46047	__ret; \
	46048	})
45530	46049	#else
45531		__ai float16x8x2_t vuzpq_f16(float16x8_t __p0, float16x8_t __p1) {
45532		float16x8x2_t __ret;
45533		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45534		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45535		__builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45536
45537		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
45538		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
45539		return __ret;
45540		}
	46050	#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
	46051	uint32_t __ret; \
	46052	uint32x4_t __s0 = __p0; \
	46053	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	46054	__ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
	46055	__ret; \
	46056	})
45541	46057	#endif
45542	46058
45543	46059	#ifdef __LITTLE_ENDIAN__
45544		__ai float16x4x2_t vuzp_f16(float16x4_t __p0, float16x4_t __p1) {
45545		float16x4x2_t __ret;
45546		__builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 8);
45547		return __ret;
45548		}
	46060	#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
	46061	uint64_t __ret; \
	46062	uint64x2_t __s0 = __p0; \
	46063	__ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
	46064	__ret; \
	46065	})
45549	46066	#else
45550		__ai float16x4x2_t vuzp_f16(float16x4_t __p0, float16x4_t __p1) {
45551		float16x4x2_t __ret;
45552		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45553		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45554		__builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45555
45556		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
45557		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0);
45558		return __ret;
45559		}
	46067	#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
	46068	uint64_t __ret; \
	46069	uint64x2_t __s0 = __p0; \
	46070	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46071	__ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
	46072	__ret; \
	46073	})
45560	46074	#endif
45561	46075
45562	46076	#ifdef __LITTLE_ENDIAN__
45563		__ai float16x8x2_t vzipq_f16(float16x8_t __p0, float16x8_t __p1) {
45564		float16x8x2_t __ret;
45565		__builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 40);
45566		return __ret;
45567		}
	46077	#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
	46078	uint16_t __ret; \
	46079	uint16x8_t __s0 = __p0; \
	46080	__ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
	46081	__ret; \
	46082	})
45568	46083	#else
45569		__ai float16x8x2_t vzipq_f16(float16x8_t __p0, float16x8_t __p1) {
45570		float16x8x2_t __ret;
45571		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45572		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45573		__builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
45574
45575		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
45576		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
45577		return __ret;
45578		}
	46084	#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
	46085	uint16_t __ret; \
	46086	uint16x8_t __s0 = __p0; \
	46087	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	46088	__ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
	46089	__ret; \
	46090	})
45579	46091	#endif
45580	46092
45581	46093	#ifdef __LITTLE_ENDIAN__
45582		__ai float16x4x2_t vzip_f16(float16x4_t __p0, float16x4_t __p1) {
45583		float16x4x2_t __ret;
45584		__builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 8);
45585		return __ret;
45586		}
	46094	#define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
	46095	int8_t __ret; \
	46096	int8x16_t __s0 = __p0; \
	46097	__ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
	46098	__ret; \
	46099	})
45587	46100	#else
45588		__ai float16x4x2_t vzip_f16(float16x4_t __p0, float16x4_t __p1) {
45589		float16x4x2_t __ret;
45590		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45591		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45592		__builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
45593
45594		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
45595		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 3, 2, 1, 0);
45596		return __ret;
45597		}
	46101	#define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
	46102	int8_t __ret; \
	46103	int8x16_t __s0 = __p0; \
	46104	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46105	__ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
	46106	__ret; \
	46107	})
45598	46108	#endif
45599	46109
45600		#endif
45601		#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(__aarch64__)
45602	46110	#ifdef __LITTLE_ENDIAN__
45603		__ai float16x8_t vdivq_f16(float16x8_t __p0, float16x8_t __p1) {
45604		float16x8_t __ret;
45605		__ret = __p0 / __p1;
45606		return __ret;
45607		}
	46111	#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
	46112	float64_t __ret; \
	46113	float64x2_t __s0 = __p0; \
	46114	__ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__s0, __p1); \
	46115	__ret; \
	46116	})
45608	46117	#else
45609		__ai float16x8_t vdivq_f16(float16x8_t __p0, float16x8_t __p1) {
45610		float16x8_t __ret;
45611		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45612		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45613		__ret = __rev0 / __rev1;
45614		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45615		return __ret;
45616		}
	46118	#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
	46119	float64_t __ret; \
	46120	float64x2_t __s0 = __p0; \
	46121	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46122	__ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__rev0, __p1); \
	46123	__ret; \
	46124	})
45617	46125	#endif
45618	46126
45619	46127	#ifdef __LITTLE_ENDIAN__
45620		__ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
45621		float16x4_t __ret;
45622		__ret = __p0 / __p1;
45623		return __ret;
45624		}
	46128	#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
	46129	float32_t __ret; \
	46130	float32x4_t __s0 = __p0; \
	46131	__ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__s0, __p1); \
	46132	__ret; \
	46133	})
45625	46134	#else
45626		__ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
45627		float16x4_t __ret;
45628		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45629		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45630		__ret = __rev0 / __rev1;
45631		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45632		return __ret;
45633		}
	46135	#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
	46136	float32_t __ret; \
	46137	float32x4_t __s0 = __p0; \
	46138	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	46139	__ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__rev0, __p1); \
	46140	__ret; \
	46141	})
45634	46142	#endif
45635	46143
45636	46144	#ifdef __LITTLE_ENDIAN__
45637		#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
45638		float16_t __ret; \
45639		float16x4_t __s0 = __p0; \
45640		__ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__s0, __p1); \
	46145	#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
	46146	int32_t __ret; \
	46147	int32x4_t __s0 = __p0; \
	46148	__ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
45641	46149	__ret; \
45642	46150	})
45643	46151	#else
45644		#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
45645		float16_t __ret; \
45646		float16x4_t __s0 = __p0; \
45647		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45648		__ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__rev0, __p1); \
	46152	#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
	46153	int32_t __ret; \
	46154	int32x4_t __s0 = __p0; \
	46155	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	46156	__ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
45649	46157	__ret; \
45650	46158	})
45651	46159	#endif
45652	46160
45653	46161	#ifdef __LITTLE_ENDIAN__
45654		#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
45655		float16_t __ret; \
45656		float16x8_t __s0 = __p0; \
45657		__ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__s0, __p1); \
	46162	#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
	46163	int64_t __ret; \
	46164	int64x2_t __s0 = __p0; \
	46165	__ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
45658	46166	__ret; \
45659	46167	})
45660	46168	#else
45661		#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
45662		float16_t __ret; \
45663		float16x8_t __s0 = __p0; \
45664		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45665		__ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__rev0, __p1); \
	46169	#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
	46170	int64_t __ret; \
	46171	int64x2_t __s0 = __p0; \
	46172	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46173	__ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
45666	46174	__ret; \
45667	46175	})
45668	46176	#endif
45669	46177
45670	46178	#ifdef __LITTLE_ENDIAN__
45671		#define vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45672		float16_t __ret; \
45673		float16_t __s0 = __p0; \
45674		float16_t __s1 = __p1; \
45675		float16x4_t __s2 = __p2; \
45676		__ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
	46179	#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
	46180	int16_t __ret; \
	46181	int16x8_t __s0 = __p0; \
	46182	__ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
45677	46183	__ret; \
45678	46184	})
45679	46185	#else
45680		#define vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45681		float16_t __ret; \
45682		float16_t __s0 = __p0; \
45683		float16_t __s1 = __p1; \
45684		float16x4_t __s2 = __p2; \
45685		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
45686		__ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__rev2, __p3); \
45687		__ret; \
45688		})
45689		#define __noswap_vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45690		float16_t __ret; \
45691		float16_t __s0 = __p0; \
45692		float16_t __s1 = __p1; \
45693		float16x4_t __s2 = __p2; \
45694		__ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
	46186	#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
	46187	int16_t __ret; \
	46188	int16x8_t __s0 = __p0; \
	46189	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	46190	__ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
45695	46191	__ret; \
45696	46192	})
45697	46193	#endif
45698	46194
45699	46195	#ifdef __LITTLE_ENDIAN__
45700		#define vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45701		float16x8_t __ret; \
45702		float16x8_t __s0 = __p0; \
45703		float16x8_t __s1 = __p1; \
45704		float16x4_t __s2 = __p2; \
45705		__ret = (float16x8_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 40); \
45706		__ret; \
	46196	#define vdup_laneq_p8(__p0_349, __p1_349) __extension__ ({ \
	46197	poly8x8_t __ret_349; \
	46198	poly8x16_t __s0_349 = __p0_349; \
	46199	__ret_349 = splat_laneq_p8(__s0_349, __p1_349); \
	46200	__ret_349; \
45707	46201	})
45708	46202	#else
45709		#define vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45710		float16x8_t __ret; \
45711		float16x8_t __s0 = __p0; \
45712		float16x8_t __s1 = __p1; \
45713		float16x4_t __s2 = __p2; \
45714		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45715		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
45716		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
45717		__ret = (float16x8_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, __p3, 40); \
45718		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
45719		__ret; \
45720		})
45721		#define __noswap_vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45722		float16x8_t __ret; \
45723		float16x8_t __s0 = __p0; \
45724		float16x8_t __s1 = __p1; \
45725		float16x4_t __s2 = __p2; \
45726		__ret = (float16x8_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 40); \
45727		__ret; \
	46203	#define vdup_laneq_p8(__p0_350, __p1_350) __extension__ ({ \
	46204	poly8x8_t __ret_350; \
	46205	poly8x16_t __s0_350 = __p0_350; \
	46206	poly8x16_t __rev0_350; __rev0_350 = __builtin_shufflevector(__s0_350, __s0_350, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46207	__ret_350 = __noswap_splat_laneq_p8(__rev0_350, __p1_350); \
	46208	__ret_350 = __builtin_shufflevector(__ret_350, __ret_350, 7, 6, 5, 4, 3, 2, 1, 0); \
	46209	__ret_350; \
45728	46210	})
45729	46211	#endif
45730	46212
45731	46213	#ifdef __LITTLE_ENDIAN__
45732		#define vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45733		float16x4_t __ret; \
45734		float16x4_t __s0 = __p0; \
45735		float16x4_t __s1 = __p1; \
45736		float16x4_t __s2 = __p2; \
45737		__ret = (float16x4_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 8); \
45738		__ret; \
	46214	#define vdup_laneq_p64(__p0_351, __p1_351) __extension__ ({ \
	46215	poly64x1_t __ret_351; \
	46216	poly64x2_t __s0_351 = __p0_351; \
	46217	__ret_351 = splat_laneq_p64(__s0_351, __p1_351); \
	46218	__ret_351; \
45739	46219	})
45740	46220	#else
45741		#define vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45742		float16x4_t __ret; \
45743		float16x4_t __s0 = __p0; \
45744		float16x4_t __s1 = __p1; \
45745		float16x4_t __s2 = __p2; \
45746		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45747		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
45748		float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
45749		__ret = (float16x4_t) __builtin_neon_vfma_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, __p3, 8); \
45750		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
45751		__ret; \
	46221	#define vdup_laneq_p64(__p0_352, __p1_352) __extension__ ({ \
	46222	poly64x1_t __ret_352; \
	46223	poly64x2_t __s0_352 = __p0_352; \
	46224	poly64x2_t __rev0_352; __rev0_352 = __builtin_shufflevector(__s0_352, __s0_352, 1, 0); \
	46225	__ret_352 = __noswap_splat_laneq_p64(__rev0_352, __p1_352); \
	46226	__ret_352; \
45752	46227	})
45753		#define __noswap_vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45754		float16x4_t __ret; \
45755		float16x4_t __s0 = __p0; \
45756		float16x4_t __s1 = __p1; \
45757		float16x4_t __s2 = __p2; \
45758		__ret = (float16x4_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 8); \
45759		__ret; \
	46228	#endif
	46229
	46230	#ifdef __LITTLE_ENDIAN__
	46231	#define vdup_laneq_p16(__p0_353, __p1_353) __extension__ ({ \
	46232	poly16x4_t __ret_353; \
	46233	poly16x8_t __s0_353 = __p0_353; \
	46234	__ret_353 = splat_laneq_p16(__s0_353, __p1_353); \
	46235	__ret_353; \
	46236	})
	46237	#else
	46238	#define vdup_laneq_p16(__p0_354, __p1_354) __extension__ ({ \
	46239	poly16x4_t __ret_354; \
	46240	poly16x8_t __s0_354 = __p0_354; \
	46241	poly16x8_t __rev0_354; __rev0_354 = __builtin_shufflevector(__s0_354, __s0_354, 7, 6, 5, 4, 3, 2, 1, 0); \
	46242	__ret_354 = __noswap_splat_laneq_p16(__rev0_354, __p1_354); \
	46243	__ret_354 = __builtin_shufflevector(__ret_354, __ret_354, 3, 2, 1, 0); \
	46244	__ret_354; \
45760	46245	})
45761	46246	#endif
45762	46247
45763	46248	#ifdef __LITTLE_ENDIAN__
45764		#define vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45765		float16_t __ret; \
45766		float16_t __s0 = __p0; \
45767		float16_t __s1 = __p1; \
45768		float16x8_t __s2 = __p2; \
45769		__ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
45770		__ret; \
	46249	#define vdupq_laneq_p8(__p0_355, __p1_355) __extension__ ({ \
	46250	poly8x16_t __ret_355; \
	46251	poly8x16_t __s0_355 = __p0_355; \
	46252	__ret_355 = splatq_laneq_p8(__s0_355, __p1_355); \
	46253	__ret_355; \
45771	46254	})
45772	46255	#else
45773		#define vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45774		float16_t __ret; \
45775		float16_t __s0 = __p0; \
45776		float16_t __s1 = __p1; \
45777		float16x8_t __s2 = __p2; \
45778		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
45779		__ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__rev2, __p3); \
45780		__ret; \
	46256	#define vdupq_laneq_p8(__p0_356, __p1_356) __extension__ ({ \
	46257	poly8x16_t __ret_356; \
	46258	poly8x16_t __s0_356 = __p0_356; \
	46259	poly8x16_t __rev0_356; __rev0_356 = __builtin_shufflevector(__s0_356, __s0_356, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46260	__ret_356 = __noswap_splatq_laneq_p8(__rev0_356, __p1_356); \
	46261	__ret_356 = __builtin_shufflevector(__ret_356, __ret_356, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46262	__ret_356; \
45781	46263	})
45782		#define __noswap_vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45783		float16_t __ret; \
45784		float16_t __s0 = __p0; \
45785		float16_t __s1 = __p1; \
45786		float16x8_t __s2 = __p2; \
45787		__ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
45788		__ret; \
	46264	#endif
	46265
	46266	#ifdef __LITTLE_ENDIAN__
	46267	#define vdupq_laneq_p64(__p0_357, __p1_357) __extension__ ({ \
	46268	poly64x2_t __ret_357; \
	46269	poly64x2_t __s0_357 = __p0_357; \
	46270	__ret_357 = splatq_laneq_p64(__s0_357, __p1_357); \
	46271	__ret_357; \
	46272	})
	46273	#else
	46274	#define vdupq_laneq_p64(__p0_358, __p1_358) __extension__ ({ \
	46275	poly64x2_t __ret_358; \
	46276	poly64x2_t __s0_358 = __p0_358; \
	46277	poly64x2_t __rev0_358; __rev0_358 = __builtin_shufflevector(__s0_358, __s0_358, 1, 0); \
	46278	__ret_358 = __noswap_splatq_laneq_p64(__rev0_358, __p1_358); \
	46279	__ret_358 = __builtin_shufflevector(__ret_358, __ret_358, 1, 0); \
	46280	__ret_358; \
45789	46281	})
45790	46282	#endif
45791	46283
45792	46284	#ifdef __LITTLE_ENDIAN__
45793		#define vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45794		float16x8_t __ret; \
45795		float16x8_t __s0 = __p0; \
45796		float16x8_t __s1 = __p1; \
45797		float16x8_t __s2 = __p2; \
45798		__ret = (float16x8_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 40); \
45799		__ret; \
	46285	#define vdupq_laneq_p16(__p0_359, __p1_359) __extension__ ({ \
	46286	poly16x8_t __ret_359; \
	46287	poly16x8_t __s0_359 = __p0_359; \
	46288	__ret_359 = splatq_laneq_p16(__s0_359, __p1_359); \
	46289	__ret_359; \
45800	46290	})
45801	46291	#else
45802		#define vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45803		float16x8_t __ret; \
45804		float16x8_t __s0 = __p0; \
45805		float16x8_t __s1 = __p1; \
45806		float16x8_t __s2 = __p2; \
45807		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45808		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
45809		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
45810		__ret = (float16x8_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 40); \
45811		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
45812		__ret; \
	46292	#define vdupq_laneq_p16(__p0_360, __p1_360) __extension__ ({ \
	46293	poly16x8_t __ret_360; \
	46294	poly16x8_t __s0_360 = __p0_360; \
	46295	poly16x8_t __rev0_360; __rev0_360 = __builtin_shufflevector(__s0_360, __s0_360, 7, 6, 5, 4, 3, 2, 1, 0); \
	46296	__ret_360 = __noswap_splatq_laneq_p16(__rev0_360, __p1_360); \
	46297	__ret_360 = __builtin_shufflevector(__ret_360, __ret_360, 7, 6, 5, 4, 3, 2, 1, 0); \
	46298	__ret_360; \
45813	46299	})
45814		#define __noswap_vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45815		float16x8_t __ret; \
45816		float16x8_t __s0 = __p0; \
45817		float16x8_t __s1 = __p1; \
45818		float16x8_t __s2 = __p2; \
45819		__ret = (float16x8_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 40); \
45820		__ret; \
	46300	#endif
	46301
	46302	#ifdef __LITTLE_ENDIAN__
	46303	#define vdupq_laneq_u8(__p0_361, __p1_361) __extension__ ({ \
	46304	uint8x16_t __ret_361; \
	46305	uint8x16_t __s0_361 = __p0_361; \
	46306	__ret_361 = splatq_laneq_u8(__s0_361, __p1_361); \
	46307	__ret_361; \
	46308	})
	46309	#else
	46310	#define vdupq_laneq_u8(__p0_362, __p1_362) __extension__ ({ \
	46311	uint8x16_t __ret_362; \
	46312	uint8x16_t __s0_362 = __p0_362; \
	46313	uint8x16_t __rev0_362; __rev0_362 = __builtin_shufflevector(__s0_362, __s0_362, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46314	__ret_362 = __noswap_splatq_laneq_u8(__rev0_362, __p1_362); \
	46315	__ret_362 = __builtin_shufflevector(__ret_362, __ret_362, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46316	__ret_362; \
45821	46317	})
45822	46318	#endif
45823	46319
45824	46320	#ifdef __LITTLE_ENDIAN__
45825		#define vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45826		float16x4_t __ret; \
45827		float16x4_t __s0 = __p0; \
45828		float16x4_t __s1 = __p1; \
45829		float16x8_t __s2 = __p2; \
45830		__ret = (float16x4_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 8); \
45831		__ret; \
	46321	#define vdupq_laneq_u32(__p0_363, __p1_363) __extension__ ({ \
	46322	uint32x4_t __ret_363; \
	46323	uint32x4_t __s0_363 = __p0_363; \
	46324	__ret_363 = splatq_laneq_u32(__s0_363, __p1_363); \
	46325	__ret_363; \
45832	46326	})
45833	46327	#else
45834		#define vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45835		float16x4_t __ret; \
45836		float16x4_t __s0 = __p0; \
45837		float16x4_t __s1 = __p1; \
45838		float16x8_t __s2 = __p2; \
45839		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45840		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
45841		float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
45842		__ret = (float16x4_t) __builtin_neon_vfma_laneq_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x16_t)__rev2, __p3, 8); \
45843		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
45844		__ret; \
	46328	#define vdupq_laneq_u32(__p0_364, __p1_364) __extension__ ({ \
	46329	uint32x4_t __ret_364; \
	46330	uint32x4_t __s0_364 = __p0_364; \
	46331	uint32x4_t __rev0_364; __rev0_364 = __builtin_shufflevector(__s0_364, __s0_364, 3, 2, 1, 0); \
	46332	__ret_364 = __noswap_splatq_laneq_u32(__rev0_364, __p1_364); \
	46333	__ret_364 = __builtin_shufflevector(__ret_364, __ret_364, 3, 2, 1, 0); \
	46334	__ret_364; \
45845	46335	})
45846		#define __noswap_vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45847		float16x4_t __ret; \
45848		float16x4_t __s0 = __p0; \
45849		float16x4_t __s1 = __p1; \
45850		float16x8_t __s2 = __p2; \
45851		__ret = (float16x4_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 8); \
45852		__ret; \
	46336	#endif
	46337
	46338	#ifdef __LITTLE_ENDIAN__
	46339	#define vdupq_laneq_u64(__p0_365, __p1_365) __extension__ ({ \
	46340	uint64x2_t __ret_365; \
	46341	uint64x2_t __s0_365 = __p0_365; \
	46342	__ret_365 = splatq_laneq_u64(__s0_365, __p1_365); \
	46343	__ret_365; \
	46344	})
	46345	#else
	46346	#define vdupq_laneq_u64(__p0_366, __p1_366) __extension__ ({ \
	46347	uint64x2_t __ret_366; \
	46348	uint64x2_t __s0_366 = __p0_366; \
	46349	uint64x2_t __rev0_366; __rev0_366 = __builtin_shufflevector(__s0_366, __s0_366, 1, 0); \
	46350	__ret_366 = __noswap_splatq_laneq_u64(__rev0_366, __p1_366); \
	46351	__ret_366 = __builtin_shufflevector(__ret_366, __ret_366, 1, 0); \
	46352	__ret_366; \
45853	46353	})
45854	46354	#endif
45855	46355
45856	46356	#ifdef __LITTLE_ENDIAN__
45857		#define vfmaq_n_f16(__p0, __p1, __p2) __extension__ ({ \
45858		float16x8_t __ret; \
45859		float16x8_t __s0 = __p0; \
45860		float16x8_t __s1 = __p1; \
45861		float16_t __s2 = __p2; \
45862		__ret = vfmaq_f16(__s0, __s1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
45863		__ret; \
	46357	#define vdupq_laneq_u16(__p0_367, __p1_367) __extension__ ({ \
	46358	uint16x8_t __ret_367; \
	46359	uint16x8_t __s0_367 = __p0_367; \
	46360	__ret_367 = splatq_laneq_u16(__s0_367, __p1_367); \
	46361	__ret_367; \
45864	46362	})
45865	46363	#else
45866		#define vfmaq_n_f16(__p0, __p1, __p2) __extension__ ({ \
45867		float16x8_t __ret; \
45868		float16x8_t __s0 = __p0; \
45869		float16x8_t __s1 = __p1; \
45870		float16_t __s2 = __p2; \
45871		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45872		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
45873		__ret = __noswap_vfmaq_f16(__rev0, __rev1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
45874		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
45875		__ret; \
	46364	#define vdupq_laneq_u16(__p0_368, __p1_368) __extension__ ({ \
	46365	uint16x8_t __ret_368; \
	46366	uint16x8_t __s0_368 = __p0_368; \
	46367	uint16x8_t __rev0_368; __rev0_368 = __builtin_shufflevector(__s0_368, __s0_368, 7, 6, 5, 4, 3, 2, 1, 0); \
	46368	__ret_368 = __noswap_splatq_laneq_u16(__rev0_368, __p1_368); \
	46369	__ret_368 = __builtin_shufflevector(__ret_368, __ret_368, 7, 6, 5, 4, 3, 2, 1, 0); \
	46370	__ret_368; \
45876	46371	})
45877	46372	#endif
45878	46373
45879	46374	#ifdef __LITTLE_ENDIAN__
45880		#define vfma_n_f16(__p0, __p1, __p2) __extension__ ({ \
45881		float16x4_t __ret; \
45882		float16x4_t __s0 = __p0; \
45883		float16x4_t __s1 = __p1; \
45884		float16_t __s2 = __p2; \
45885		__ret = vfma_f16(__s0, __s1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
45886		__ret; \
	46375	#define vdupq_laneq_s8(__p0_369, __p1_369) __extension__ ({ \
	46376	int8x16_t __ret_369; \
	46377	int8x16_t __s0_369 = __p0_369; \
	46378	__ret_369 = splatq_laneq_s8(__s0_369, __p1_369); \
	46379	__ret_369; \
45887	46380	})
45888	46381	#else
45889		#define vfma_n_f16(__p0, __p1, __p2) __extension__ ({ \
45890		float16x4_t __ret; \
45891		float16x4_t __s0 = __p0; \
45892		float16x4_t __s1 = __p1; \
45893		float16_t __s2 = __p2; \
45894		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45895		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
45896		__ret = __noswap_vfma_f16(__rev0, __rev1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
45897		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
45898		__ret; \
	46382	#define vdupq_laneq_s8(__p0_370, __p1_370) __extension__ ({ \
	46383	int8x16_t __ret_370; \
	46384	int8x16_t __s0_370 = __p0_370; \
	46385	int8x16_t __rev0_370; __rev0_370 = __builtin_shufflevector(__s0_370, __s0_370, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46386	__ret_370 = __noswap_splatq_laneq_s8(__rev0_370, __p1_370); \
	46387	__ret_370 = __builtin_shufflevector(__ret_370, __ret_370, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46388	__ret_370; \
45899	46389	})
45900	46390	#endif
45901	46391
45902	46392	#ifdef __LITTLE_ENDIAN__
45903		#define vfmsh_lane_f16(__p0_281, __p1_281, __p2_281, __p3_281) __extension__ ({ \
45904		float16_t __ret_281; \
45905		float16_t __s0_281 = __p0_281; \
45906		float16_t __s1_281 = __p1_281; \
45907		float16x4_t __s2_281 = __p2_281; \
45908		__ret_281 = vfmah_lane_f16(__s0_281, -__s1_281, __s2_281, __p3_281); \
45909		__ret_281; \
	46393	#define vdupq_laneq_f64(__p0_371, __p1_371) __extension__ ({ \
	46394	float64x2_t __ret_371; \
	46395	float64x2_t __s0_371 = __p0_371; \
	46396	__ret_371 = splatq_laneq_f64(__s0_371, __p1_371); \
	46397	__ret_371; \
45910	46398	})
45911	46399	#else
45912		#define vfmsh_lane_f16(__p0_282, __p1_282, __p2_282, __p3_282) __extension__ ({ \
45913		float16_t __ret_282; \
45914		float16_t __s0_282 = __p0_282; \
45915		float16_t __s1_282 = __p1_282; \
45916		float16x4_t __s2_282 = __p2_282; \
45917		float16x4_t __rev2_282; __rev2_282 = __builtin_shufflevector(__s2_282, __s2_282, 3, 2, 1, 0); \
45918		__ret_282 = __noswap_vfmah_lane_f16(__s0_282, -__s1_282, __rev2_282, __p3_282); \
45919		__ret_282; \
	46400	#define vdupq_laneq_f64(__p0_372, __p1_372) __extension__ ({ \
	46401	float64x2_t __ret_372; \
	46402	float64x2_t __s0_372 = __p0_372; \
	46403	float64x2_t __rev0_372; __rev0_372 = __builtin_shufflevector(__s0_372, __s0_372, 1, 0); \
	46404	__ret_372 = __noswap_splatq_laneq_f64(__rev0_372, __p1_372); \
	46405	__ret_372 = __builtin_shufflevector(__ret_372, __ret_372, 1, 0); \
	46406	__ret_372; \
45920	46407	})
45921	46408	#endif
45922	46409
45923	46410	#ifdef __LITTLE_ENDIAN__
45924		#define vfmsq_lane_f16(__p0_283, __p1_283, __p2_283, __p3_283) __extension__ ({ \
45925		float16x8_t __ret_283; \
45926		float16x8_t __s0_283 = __p0_283; \
45927		float16x8_t __s1_283 = __p1_283; \
45928		float16x4_t __s2_283 = __p2_283; \
45929		__ret_283 = vfmaq_lane_f16(__s0_283, -__s1_283, __s2_283, __p3_283); \
45930		__ret_283; \
	46411	#define vdupq_laneq_f32(__p0_373, __p1_373) __extension__ ({ \
	46412	float32x4_t __ret_373; \
	46413	float32x4_t __s0_373 = __p0_373; \
	46414	__ret_373 = splatq_laneq_f32(__s0_373, __p1_373); \
	46415	__ret_373; \
45931	46416	})
45932	46417	#else
45933		#define vfmsq_lane_f16(__p0_284, __p1_284, __p2_284, __p3_284) __extension__ ({ \
45934		float16x8_t __ret_284; \
45935		float16x8_t __s0_284 = __p0_284; \
45936		float16x8_t __s1_284 = __p1_284; \
45937		float16x4_t __s2_284 = __p2_284; \
45938		float16x8_t __rev0_284; __rev0_284 = __builtin_shufflevector(__s0_284, __s0_284, 7, 6, 5, 4, 3, 2, 1, 0); \
45939		float16x8_t __rev1_284; __rev1_284 = __builtin_shufflevector(__s1_284, __s1_284, 7, 6, 5, 4, 3, 2, 1, 0); \
45940		float16x4_t __rev2_284; __rev2_284 = __builtin_shufflevector(__s2_284, __s2_284, 3, 2, 1, 0); \
45941		__ret_284 = __noswap_vfmaq_lane_f16(__rev0_284, -__rev1_284, __rev2_284, __p3_284); \
45942		__ret_284 = __builtin_shufflevector(__ret_284, __ret_284, 7, 6, 5, 4, 3, 2, 1, 0); \
45943		__ret_284; \
	46418	#define vdupq_laneq_f32(__p0_374, __p1_374) __extension__ ({ \
	46419	float32x4_t __ret_374; \
	46420	float32x4_t __s0_374 = __p0_374; \
	46421	float32x4_t __rev0_374; __rev0_374 = __builtin_shufflevector(__s0_374, __s0_374, 3, 2, 1, 0); \
	46422	__ret_374 = __noswap_splatq_laneq_f32(__rev0_374, __p1_374); \
	46423	__ret_374 = __builtin_shufflevector(__ret_374, __ret_374, 3, 2, 1, 0); \
	46424	__ret_374; \
45944	46425	})
45945	46426	#endif
45946	46427
45947	46428	#ifdef __LITTLE_ENDIAN__
45948		#define vfms_lane_f16(__p0_285, __p1_285, __p2_285, __p3_285) __extension__ ({ \
45949		float16x4_t __ret_285; \
45950		float16x4_t __s0_285 = __p0_285; \
45951		float16x4_t __s1_285 = __p1_285; \
45952		float16x4_t __s2_285 = __p2_285; \
45953		__ret_285 = vfma_lane_f16(__s0_285, -__s1_285, __s2_285, __p3_285); \
45954		__ret_285; \
	46429	#define vdupq_laneq_f16(__p0_375, __p1_375) __extension__ ({ \
	46430	float16x8_t __ret_375; \
	46431	float16x8_t __s0_375 = __p0_375; \
	46432	__ret_375 = splatq_laneq_f16(__s0_375, __p1_375); \
	46433	__ret_375; \
45955	46434	})
45956	46435	#else
45957		#define vfms_lane_f16(__p0_286, __p1_286, __p2_286, __p3_286) __extension__ ({ \
45958		float16x4_t __ret_286; \
45959		float16x4_t __s0_286 = __p0_286; \
45960		float16x4_t __s1_286 = __p1_286; \
45961		float16x4_t __s2_286 = __p2_286; \
45962		float16x4_t __rev0_286; __rev0_286 = __builtin_shufflevector(__s0_286, __s0_286, 3, 2, 1, 0); \
45963		float16x4_t __rev1_286; __rev1_286 = __builtin_shufflevector(__s1_286, __s1_286, 3, 2, 1, 0); \
45964		float16x4_t __rev2_286; __rev2_286 = __builtin_shufflevector(__s2_286, __s2_286, 3, 2, 1, 0); \
45965		__ret_286 = __noswap_vfma_lane_f16(__rev0_286, -__rev1_286, __rev2_286, __p3_286); \
45966		__ret_286 = __builtin_shufflevector(__ret_286, __ret_286, 3, 2, 1, 0); \
45967		__ret_286; \
	46436	#define vdupq_laneq_f16(__p0_376, __p1_376) __extension__ ({ \
	46437	float16x8_t __ret_376; \
	46438	float16x8_t __s0_376 = __p0_376; \
	46439	float16x8_t __rev0_376; __rev0_376 = __builtin_shufflevector(__s0_376, __s0_376, 7, 6, 5, 4, 3, 2, 1, 0); \
	46440	__ret_376 = __noswap_splatq_laneq_f16(__rev0_376, __p1_376); \
	46441	__ret_376 = __builtin_shufflevector(__ret_376, __ret_376, 7, 6, 5, 4, 3, 2, 1, 0); \
	46442	__ret_376; \
45968	46443	})
45969	46444	#endif
45970	46445
45971	46446	#ifdef __LITTLE_ENDIAN__
45972		#define vfmsh_laneq_f16(__p0_287, __p1_287, __p2_287, __p3_287) __extension__ ({ \
45973		float16_t __ret_287; \
45974		float16_t __s0_287 = __p0_287; \
45975		float16_t __s1_287 = __p1_287; \
45976		float16x8_t __s2_287 = __p2_287; \
45977		__ret_287 = vfmah_laneq_f16(__s0_287, -__s1_287, __s2_287, __p3_287); \
45978		__ret_287; \
	46447	#define vdupq_laneq_s32(__p0_377, __p1_377) __extension__ ({ \
	46448	int32x4_t __ret_377; \
	46449	int32x4_t __s0_377 = __p0_377; \
	46450	__ret_377 = splatq_laneq_s32(__s0_377, __p1_377); \
	46451	__ret_377; \
45979	46452	})
45980	46453	#else
45981		#define vfmsh_laneq_f16(__p0_288, __p1_288, __p2_288, __p3_288) __extension__ ({ \
45982		float16_t __ret_288; \
45983		float16_t __s0_288 = __p0_288; \
45984		float16_t __s1_288 = __p1_288; \
45985		float16x8_t __s2_288 = __p2_288; \
45986		float16x8_t __rev2_288; __rev2_288 = __builtin_shufflevector(__s2_288, __s2_288, 7, 6, 5, 4, 3, 2, 1, 0); \
45987		__ret_288 = __noswap_vfmah_laneq_f16(__s0_288, -__s1_288, __rev2_288, __p3_288); \
45988		__ret_288; \
	46454	#define vdupq_laneq_s32(__p0_378, __p1_378) __extension__ ({ \
	46455	int32x4_t __ret_378; \
	46456	int32x4_t __s0_378 = __p0_378; \
	46457	int32x4_t __rev0_378; __rev0_378 = __builtin_shufflevector(__s0_378, __s0_378, 3, 2, 1, 0); \
	46458	__ret_378 = __noswap_splatq_laneq_s32(__rev0_378, __p1_378); \
	46459	__ret_378 = __builtin_shufflevector(__ret_378, __ret_378, 3, 2, 1, 0); \
	46460	__ret_378; \
45989	46461	})
45990	46462	#endif
45991	46463
45992	46464	#ifdef __LITTLE_ENDIAN__
45993		#define vfmsq_laneq_f16(__p0_289, __p1_289, __p2_289, __p3_289) __extension__ ({ \
45994		float16x8_t __ret_289; \
45995		float16x8_t __s0_289 = __p0_289; \
45996		float16x8_t __s1_289 = __p1_289; \
45997		float16x8_t __s2_289 = __p2_289; \
45998		__ret_289 = vfmaq_laneq_f16(__s0_289, -__s1_289, __s2_289, __p3_289); \
45999		__ret_289; \
	46465	#define vdupq_laneq_s64(__p0_379, __p1_379) __extension__ ({ \
	46466	int64x2_t __ret_379; \
	46467	int64x2_t __s0_379 = __p0_379; \
	46468	__ret_379 = splatq_laneq_s64(__s0_379, __p1_379); \
	46469	__ret_379; \
46000	46470	})
46001	46471	#else
46002		#define vfmsq_laneq_f16(__p0_290, __p1_290, __p2_290, __p3_290) __extension__ ({ \
46003		float16x8_t __ret_290; \
46004		float16x8_t __s0_290 = __p0_290; \
46005		float16x8_t __s1_290 = __p1_290; \
46006		float16x8_t __s2_290 = __p2_290; \
46007		float16x8_t __rev0_290; __rev0_290 = __builtin_shufflevector(__s0_290, __s0_290, 7, 6, 5, 4, 3, 2, 1, 0); \
46008		float16x8_t __rev1_290; __rev1_290 = __builtin_shufflevector(__s1_290, __s1_290, 7, 6, 5, 4, 3, 2, 1, 0); \
46009		float16x8_t __rev2_290; __rev2_290 = __builtin_shufflevector(__s2_290, __s2_290, 7, 6, 5, 4, 3, 2, 1, 0); \
46010		__ret_290 = __noswap_vfmaq_laneq_f16(__rev0_290, -__rev1_290, __rev2_290, __p3_290); \
46011		__ret_290 = __builtin_shufflevector(__ret_290, __ret_290, 7, 6, 5, 4, 3, 2, 1, 0); \
46012		__ret_290; \
	46472	#define vdupq_laneq_s64(__p0_380, __p1_380) __extension__ ({ \
	46473	int64x2_t __ret_380; \
	46474	int64x2_t __s0_380 = __p0_380; \
	46475	int64x2_t __rev0_380; __rev0_380 = __builtin_shufflevector(__s0_380, __s0_380, 1, 0); \
	46476	__ret_380 = __noswap_splatq_laneq_s64(__rev0_380, __p1_380); \
	46477	__ret_380 = __builtin_shufflevector(__ret_380, __ret_380, 1, 0); \
	46478	__ret_380; \
46013	46479	})
46014	46480	#endif
46015	46481
46016	46482	#ifdef __LITTLE_ENDIAN__
46017		#define vfms_laneq_f16(__p0_291, __p1_291, __p2_291, __p3_291) __extension__ ({ \
46018		float16x4_t __ret_291; \
46019		float16x4_t __s0_291 = __p0_291; \
46020		float16x4_t __s1_291 = __p1_291; \
46021		float16x8_t __s2_291 = __p2_291; \
46022		__ret_291 = vfma_laneq_f16(__s0_291, -__s1_291, __s2_291, __p3_291); \
46023		__ret_291; \
	46483	#define vdupq_laneq_s16(__p0_381, __p1_381) __extension__ ({ \
	46484	int16x8_t __ret_381; \
	46485	int16x8_t __s0_381 = __p0_381; \
	46486	__ret_381 = splatq_laneq_s16(__s0_381, __p1_381); \
	46487	__ret_381; \
46024	46488	})
46025	46489	#else
46026		#define vfms_laneq_f16(__p0_292, __p1_292, __p2_292, __p3_292) __extension__ ({ \
46027		float16x4_t __ret_292; \
46028		float16x4_t __s0_292 = __p0_292; \
46029		float16x4_t __s1_292 = __p1_292; \
46030		float16x8_t __s2_292 = __p2_292; \
46031		float16x4_t __rev0_292; __rev0_292 = __builtin_shufflevector(__s0_292, __s0_292, 3, 2, 1, 0); \
46032		float16x4_t __rev1_292; __rev1_292 = __builtin_shufflevector(__s1_292, __s1_292, 3, 2, 1, 0); \
46033		float16x8_t __rev2_292; __rev2_292 = __builtin_shufflevector(__s2_292, __s2_292, 7, 6, 5, 4, 3, 2, 1, 0); \
46034		__ret_292 = __noswap_vfma_laneq_f16(__rev0_292, -__rev1_292, __rev2_292, __p3_292); \
46035		__ret_292 = __builtin_shufflevector(__ret_292, __ret_292, 3, 2, 1, 0); \
46036		__ret_292; \
	46490	#define vdupq_laneq_s16(__p0_382, __p1_382) __extension__ ({ \
	46491	int16x8_t __ret_382; \
	46492	int16x8_t __s0_382 = __p0_382; \
	46493	int16x8_t __rev0_382; __rev0_382 = __builtin_shufflevector(__s0_382, __s0_382, 7, 6, 5, 4, 3, 2, 1, 0); \
	46494	__ret_382 = __noswap_splatq_laneq_s16(__rev0_382, __p1_382); \
	46495	__ret_382 = __builtin_shufflevector(__ret_382, __ret_382, 7, 6, 5, 4, 3, 2, 1, 0); \
	46496	__ret_382; \
46037	46497	})
46038	46498	#endif
46039	46499
46040	46500	#ifdef __LITTLE_ENDIAN__
46041		#define vfmsq_n_f16(__p0, __p1, __p2) __extension__ ({ \
46042		float16x8_t __ret; \
46043		float16x8_t __s0 = __p0; \
46044		float16x8_t __s1 = __p1; \
46045		float16_t __s2 = __p2; \
46046		__ret = vfmaq_f16(__s0, -__s1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
46047		__ret; \
	46501	#define vdup_laneq_u8(__p0_383, __p1_383) __extension__ ({ \
	46502	uint8x8_t __ret_383; \
	46503	uint8x16_t __s0_383 = __p0_383; \
	46504	__ret_383 = splat_laneq_u8(__s0_383, __p1_383); \
	46505	__ret_383; \
46048	46506	})
46049	46507	#else
46050		#define vfmsq_n_f16(__p0, __p1, __p2) __extension__ ({ \
46051		float16x8_t __ret; \
46052		float16x8_t __s0 = __p0; \
46053		float16x8_t __s1 = __p1; \
46054		float16_t __s2 = __p2; \
46055		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46056		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
46057		__ret = __noswap_vfmaq_f16(__rev0, -__rev1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
46058		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
46059		__ret; \
	46508	#define vdup_laneq_u8(__p0_384, __p1_384) __extension__ ({ \
	46509	uint8x8_t __ret_384; \
	46510	uint8x16_t __s0_384 = __p0_384; \
	46511	uint8x16_t __rev0_384; __rev0_384 = __builtin_shufflevector(__s0_384, __s0_384, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46512	__ret_384 = __noswap_splat_laneq_u8(__rev0_384, __p1_384); \
	46513	__ret_384 = __builtin_shufflevector(__ret_384, __ret_384, 7, 6, 5, 4, 3, 2, 1, 0); \
	46514	__ret_384; \
46060	46515	})
46061	46516	#endif
46062	46517
46063	46518	#ifdef __LITTLE_ENDIAN__
46064		#define vfms_n_f16(__p0, __p1, __p2) __extension__ ({ \
46065		float16x4_t __ret; \
46066		float16x4_t __s0 = __p0; \
46067		float16x4_t __s1 = __p1; \
46068		float16_t __s2 = __p2; \
46069		__ret = vfma_f16(__s0, -__s1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
46070		__ret; \
	46519	#define vdup_laneq_u32(__p0_385, __p1_385) __extension__ ({ \
	46520	uint32x2_t __ret_385; \
	46521	uint32x4_t __s0_385 = __p0_385; \
	46522	__ret_385 = splat_laneq_u32(__s0_385, __p1_385); \
	46523	__ret_385; \
46071	46524	})
46072	46525	#else
46073		#define vfms_n_f16(__p0, __p1, __p2) __extension__ ({ \
46074		float16x4_t __ret; \
46075		float16x4_t __s0 = __p0; \
46076		float16x4_t __s1 = __p1; \
46077		float16_t __s2 = __p2; \
46078		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46079		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
46080		__ret = __noswap_vfma_f16(__rev0, -__rev1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
46081		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
46082		__ret; \
	46526	#define vdup_laneq_u32(__p0_386, __p1_386) __extension__ ({ \
	46527	uint32x2_t __ret_386; \
	46528	uint32x4_t __s0_386 = __p0_386; \
	46529	uint32x4_t __rev0_386; __rev0_386 = __builtin_shufflevector(__s0_386, __s0_386, 3, 2, 1, 0); \
	46530	__ret_386 = __noswap_splat_laneq_u32(__rev0_386, __p1_386); \
	46531	__ret_386 = __builtin_shufflevector(__ret_386, __ret_386, 1, 0); \
	46532	__ret_386; \
46083	46533	})
46084	46534	#endif
46085	46535
46086	46536	#ifdef __LITTLE_ENDIAN__
46087		#define vmaxnmvq_f16(__p0) __extension__ ({ \
46088		float16_t __ret; \
46089		float16x8_t __s0 = __p0; \
46090		__ret = (float16_t) __builtin_neon_vmaxnmvq_f16((int8x16_t)__s0); \
46091		__ret; \
	46537	#define vdup_laneq_u64(__p0_387, __p1_387) __extension__ ({ \
	46538	uint64x1_t __ret_387; \
	46539	uint64x2_t __s0_387 = __p0_387; \
	46540	__ret_387 = splat_laneq_u64(__s0_387, __p1_387); \
	46541	__ret_387; \
46092	46542	})
46093	46543	#else
46094		#define vmaxnmvq_f16(__p0) __extension__ ({ \
46095		float16_t __ret; \
46096		float16x8_t __s0 = __p0; \
46097		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46098		__ret = (float16_t) __builtin_neon_vmaxnmvq_f16((int8x16_t)__rev0); \
46099		__ret; \
	46544	#define vdup_laneq_u64(__p0_388, __p1_388) __extension__ ({ \
	46545	uint64x1_t __ret_388; \
	46546	uint64x2_t __s0_388 = __p0_388; \
	46547	uint64x2_t __rev0_388; __rev0_388 = __builtin_shufflevector(__s0_388, __s0_388, 1, 0); \
	46548	__ret_388 = __noswap_splat_laneq_u64(__rev0_388, __p1_388); \
	46549	__ret_388; \
46100	46550	})
46101	46551	#endif
46102	46552
46103	46553	#ifdef __LITTLE_ENDIAN__
46104		#define vmaxnmv_f16(__p0) __extension__ ({ \
46105		float16_t __ret; \
46106		float16x4_t __s0 = __p0; \
46107		__ret = (float16_t) __builtin_neon_vmaxnmv_f16((int8x8_t)__s0); \
46108		__ret; \
	46554	#define vdup_laneq_u16(__p0_389, __p1_389) __extension__ ({ \
	46555	uint16x4_t __ret_389; \
	46556	uint16x8_t __s0_389 = __p0_389; \
	46557	__ret_389 = splat_laneq_u16(__s0_389, __p1_389); \
	46558	__ret_389; \
46109	46559	})
46110	46560	#else
46111		#define vmaxnmv_f16(__p0) __extension__ ({ \
46112		float16_t __ret; \
46113		float16x4_t __s0 = __p0; \
46114		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46115		__ret = (float16_t) __builtin_neon_vmaxnmv_f16((int8x8_t)__rev0); \
46116		__ret; \
	46561	#define vdup_laneq_u16(__p0_390, __p1_390) __extension__ ({ \
	46562	uint16x4_t __ret_390; \
	46563	uint16x8_t __s0_390 = __p0_390; \
	46564	uint16x8_t __rev0_390; __rev0_390 = __builtin_shufflevector(__s0_390, __s0_390, 7, 6, 5, 4, 3, 2, 1, 0); \
	46565	__ret_390 = __noswap_splat_laneq_u16(__rev0_390, __p1_390); \
	46566	__ret_390 = __builtin_shufflevector(__ret_390, __ret_390, 3, 2, 1, 0); \
	46567	__ret_390; \
46117	46568	})
46118	46569	#endif
46119	46570
46120	46571	#ifdef __LITTLE_ENDIAN__
46121		#define vmaxvq_f16(__p0) __extension__ ({ \
46122		float16_t __ret; \
46123		float16x8_t __s0 = __p0; \
46124		__ret = (float16_t) __builtin_neon_vmaxvq_f16((int8x16_t)__s0); \
46125		__ret; \
	46572	#define vdup_laneq_s8(__p0_391, __p1_391) __extension__ ({ \
	46573	int8x8_t __ret_391; \
	46574	int8x16_t __s0_391 = __p0_391; \
	46575	__ret_391 = splat_laneq_s8(__s0_391, __p1_391); \
	46576	__ret_391; \
46126	46577	})
46127	46578	#else
46128		#define vmaxvq_f16(__p0) __extension__ ({ \
46129		float16_t __ret; \
46130		float16x8_t __s0 = __p0; \
46131		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46132		__ret = (float16_t) __builtin_neon_vmaxvq_f16((int8x16_t)__rev0); \
46133		__ret; \
	46579	#define vdup_laneq_s8(__p0_392, __p1_392) __extension__ ({ \
	46580	int8x8_t __ret_392; \
	46581	int8x16_t __s0_392 = __p0_392; \
	46582	int8x16_t __rev0_392; __rev0_392 = __builtin_shufflevector(__s0_392, __s0_392, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	46583	__ret_392 = __noswap_splat_laneq_s8(__rev0_392, __p1_392); \
	46584	__ret_392 = __builtin_shufflevector(__ret_392, __ret_392, 7, 6, 5, 4, 3, 2, 1, 0); \
	46585	__ret_392; \
46134	46586	})
46135	46587	#endif
46136	46588
46137	46589	#ifdef __LITTLE_ENDIAN__
46138		#define vmaxv_f16(__p0) __extension__ ({ \
46139		float16_t __ret; \
46140		float16x4_t __s0 = __p0; \
46141		__ret = (float16_t) __builtin_neon_vmaxv_f16((int8x8_t)__s0); \
46142		__ret; \
	46590	#define vdup_laneq_f64(__p0_393, __p1_393) __extension__ ({ \
	46591	float64x1_t __ret_393; \
	46592	float64x2_t __s0_393 = __p0_393; \
	46593	__ret_393 = splat_laneq_f64(__s0_393, __p1_393); \
	46594	__ret_393; \
46143	46595	})
46144	46596	#else
46145		#define vmaxv_f16(__p0) __extension__ ({ \
46146		float16_t __ret; \
46147		float16x4_t __s0 = __p0; \
46148		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46149		__ret = (float16_t) __builtin_neon_vmaxv_f16((int8x8_t)__rev0); \
46150		__ret; \
	46597	#define vdup_laneq_f64(__p0_394, __p1_394) __extension__ ({ \
	46598	float64x1_t __ret_394; \
	46599	float64x2_t __s0_394 = __p0_394; \
	46600	float64x2_t __rev0_394; __rev0_394 = __builtin_shufflevector(__s0_394, __s0_394, 1, 0); \
	46601	__ret_394 = __noswap_splat_laneq_f64(__rev0_394, __p1_394); \
	46602	__ret_394; \
46151	46603	})
46152	46604	#endif
46153	46605
46154	46606	#ifdef __LITTLE_ENDIAN__
46155		#define vminnmvq_f16(__p0) __extension__ ({ \
46156		float16_t __ret; \
46157		float16x8_t __s0 = __p0; \
46158		__ret = (float16_t) __builtin_neon_vminnmvq_f16((int8x16_t)__s0); \
46159		__ret; \
	46607	#define vdup_laneq_f32(__p0_395, __p1_395) __extension__ ({ \
	46608	float32x2_t __ret_395; \
	46609	float32x4_t __s0_395 = __p0_395; \
	46610	__ret_395 = splat_laneq_f32(__s0_395, __p1_395); \
	46611	__ret_395; \
46160	46612	})
46161	46613	#else
46162		#define vminnmvq_f16(__p0) __extension__ ({ \
46163		float16_t __ret; \
46164		float16x8_t __s0 = __p0; \
46165		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46166		__ret = (float16_t) __builtin_neon_vminnmvq_f16((int8x16_t)__rev0); \
46167		__ret; \
	46614	#define vdup_laneq_f32(__p0_396, __p1_396) __extension__ ({ \
	46615	float32x2_t __ret_396; \
	46616	float32x4_t __s0_396 = __p0_396; \
	46617	float32x4_t __rev0_396; __rev0_396 = __builtin_shufflevector(__s0_396, __s0_396, 3, 2, 1, 0); \
	46618	__ret_396 = __noswap_splat_laneq_f32(__rev0_396, __p1_396); \
	46619	__ret_396 = __builtin_shufflevector(__ret_396, __ret_396, 1, 0); \
	46620	__ret_396; \
46168	46621	})
46169	46622	#endif
46170	46623
46171	46624	#ifdef __LITTLE_ENDIAN__
46172		#define vminnmv_f16(__p0) __extension__ ({ \
46173		float16_t __ret; \
46174		float16x4_t __s0 = __p0; \
46175		__ret = (float16_t) __builtin_neon_vminnmv_f16((int8x8_t)__s0); \
46176		__ret; \
	46625	#define vdup_laneq_f16(__p0_397, __p1_397) __extension__ ({ \
	46626	float16x4_t __ret_397; \
	46627	float16x8_t __s0_397 = __p0_397; \
	46628	__ret_397 = splat_laneq_f16(__s0_397, __p1_397); \
	46629	__ret_397; \
46177	46630	})
46178	46631	#else
46179		#define vminnmv_f16(__p0) __extension__ ({ \
46180		float16_t __ret; \
46181		float16x4_t __s0 = __p0; \
46182		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46183		__ret = (float16_t) __builtin_neon_vminnmv_f16((int8x8_t)__rev0); \
46184		__ret; \
	46632	#define vdup_laneq_f16(__p0_398, __p1_398) __extension__ ({ \
	46633	float16x4_t __ret_398; \
	46634	float16x8_t __s0_398 = __p0_398; \
	46635	float16x8_t __rev0_398; __rev0_398 = __builtin_shufflevector(__s0_398, __s0_398, 7, 6, 5, 4, 3, 2, 1, 0); \
	46636	__ret_398 = __noswap_splat_laneq_f16(__rev0_398, __p1_398); \
	46637	__ret_398 = __builtin_shufflevector(__ret_398, __ret_398, 3, 2, 1, 0); \
	46638	__ret_398; \
46185	46639	})
46186	46640	#endif
46187	46641
46188	46642	#ifdef __LITTLE_ENDIAN__
46189		#define vminvq_f16(__p0) __extension__ ({ \
46190		float16_t __ret; \
46191		float16x8_t __s0 = __p0; \
46192		__ret = (float16_t) __builtin_neon_vminvq_f16((int8x16_t)__s0); \
46193		__ret; \
	46643	#define vdup_laneq_s32(__p0_399, __p1_399) __extension__ ({ \
	46644	int32x2_t __ret_399; \
	46645	int32x4_t __s0_399 = __p0_399; \
	46646	__ret_399 = splat_laneq_s32(__s0_399, __p1_399); \
	46647	__ret_399; \
46194	46648	})
46195	46649	#else
46196		#define vminvq_f16(__p0) __extension__ ({ \
46197		float16_t __ret; \
46198		float16x8_t __s0 = __p0; \
46199		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46200		__ret = (float16_t) __builtin_neon_vminvq_f16((int8x16_t)__rev0); \
46201		__ret; \
	46650	#define vdup_laneq_s32(__p0_400, __p1_400) __extension__ ({ \
	46651	int32x2_t __ret_400; \
	46652	int32x4_t __s0_400 = __p0_400; \
	46653	int32x4_t __rev0_400; __rev0_400 = __builtin_shufflevector(__s0_400, __s0_400, 3, 2, 1, 0); \
	46654	__ret_400 = __noswap_splat_laneq_s32(__rev0_400, __p1_400); \
	46655	__ret_400 = __builtin_shufflevector(__ret_400, __ret_400, 1, 0); \
	46656	__ret_400; \
46202	46657	})
46203	46658	#endif
46204	46659
46205	46660	#ifdef __LITTLE_ENDIAN__
46206		#define vminv_f16(__p0) __extension__ ({ \
46207		float16_t __ret; \
46208		float16x4_t __s0 = __p0; \
46209		__ret = (float16_t) __builtin_neon_vminv_f16((int8x8_t)__s0); \
46210		__ret; \
	46661	#define vdup_laneq_s64(__p0_401, __p1_401) __extension__ ({ \
	46662	int64x1_t __ret_401; \
	46663	int64x2_t __s0_401 = __p0_401; \
	46664	__ret_401 = splat_laneq_s64(__s0_401, __p1_401); \
	46665	__ret_401; \
46211	46666	})
46212	46667	#else
46213		#define vminv_f16(__p0) __extension__ ({ \
46214		float16_t __ret; \
46215		float16x4_t __s0 = __p0; \
46216		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46217		__ret = (float16_t) __builtin_neon_vminv_f16((int8x8_t)__rev0); \
46218		__ret; \
	46668	#define vdup_laneq_s64(__p0_402, __p1_402) __extension__ ({ \
	46669	int64x1_t __ret_402; \
	46670	int64x2_t __s0_402 = __p0_402; \
	46671	int64x2_t __rev0_402; __rev0_402 = __builtin_shufflevector(__s0_402, __s0_402, 1, 0); \
	46672	__ret_402 = __noswap_splat_laneq_s64(__rev0_402, __p1_402); \
	46673	__ret_402; \
46219	46674	})
46220	46675	#endif
46221	46676
46222	46677	#ifdef __LITTLE_ENDIAN__
46223		#define vmulq_laneq_f16(__p0_293, __p1_293, __p2_293) __extension__ ({ \
46224		float16x8_t __ret_293; \
46225		float16x8_t __s0_293 = __p0_293; \
46226		float16x8_t __s1_293 = __p1_293; \
46227		__ret_293 = __s0_293 * splatq_laneq_f16(__s1_293, __p2_293); \
46228		__ret_293; \
	46678	#define vdup_laneq_s16(__p0_403, __p1_403) __extension__ ({ \
	46679	int16x4_t __ret_403; \
	46680	int16x8_t __s0_403 = __p0_403; \
	46681	__ret_403 = splat_laneq_s16(__s0_403, __p1_403); \
	46682	__ret_403; \
46229	46683	})
46230	46684	#else
46231		#define vmulq_laneq_f16(__p0_294, __p1_294, __p2_294) __extension__ ({ \
46232		float16x8_t __ret_294; \
46233		float16x8_t __s0_294 = __p0_294; \
46234		float16x8_t __s1_294 = __p1_294; \
46235		float16x8_t __rev0_294; __rev0_294 = __builtin_shufflevector(__s0_294, __s0_294, 7, 6, 5, 4, 3, 2, 1, 0); \
46236		float16x8_t __rev1_294; __rev1_294 = __builtin_shufflevector(__s1_294, __s1_294, 7, 6, 5, 4, 3, 2, 1, 0); \
46237		__ret_294 = __rev0_294 * __noswap_splatq_laneq_f16(__rev1_294, __p2_294); \
46238		__ret_294 = __builtin_shufflevector(__ret_294, __ret_294, 7, 6, 5, 4, 3, 2, 1, 0); \
46239		__ret_294; \
	46685	#define vdup_laneq_s16(__p0_404, __p1_404) __extension__ ({ \
	46686	int16x4_t __ret_404; \
	46687	int16x8_t __s0_404 = __p0_404; \
	46688	int16x8_t __rev0_404; __rev0_404 = __builtin_shufflevector(__s0_404, __s0_404, 7, 6, 5, 4, 3, 2, 1, 0); \
	46689	__ret_404 = __noswap_splat_laneq_s16(__rev0_404, __p1_404); \
	46690	__ret_404 = __builtin_shufflevector(__ret_404, __ret_404, 3, 2, 1, 0); \
	46691	__ret_404; \
46240	46692	})
46241	46693	#endif
46242	46694
	46695	__ai poly64x1_t vdup_n_p64(poly64_t __p0) {
	46696	poly64x1_t __ret;
	46697	__ret = (poly64x1_t) {__p0};
	46698	return __ret;
	46699	}
46243	46700	#ifdef __LITTLE_ENDIAN__
46244		#define vmul_laneq_f16(__p0_295, __p1_295, __p2_295) __extension__ ({ \
46245		float16x4_t __ret_295; \
46246		float16x4_t __s0_295 = __p0_295; \
46247		float16x8_t __s1_295 = __p1_295; \
46248		__ret_295 = __s0_295 * splat_laneq_f16(__s1_295, __p2_295); \
46249		__ret_295; \
46250		})
	46701	__ai poly64x2_t vdupq_n_p64(poly64_t __p0) {
	46702	poly64x2_t __ret;
	46703	__ret = (poly64x2_t) {__p0, __p0};
	46704	return __ret;
	46705	}
46251	46706	#else
46252		#define vmul_laneq_f16(__p0_296, __p1_296, __p2_296) __extension__ ({ \
46253		float16x4_t __ret_296; \
46254		float16x4_t __s0_296 = __p0_296; \
46255		float16x8_t __s1_296 = __p1_296; \
46256		float16x4_t __rev0_296; __rev0_296 = __builtin_shufflevector(__s0_296, __s0_296, 3, 2, 1, 0); \
46257		float16x8_t __rev1_296; __rev1_296 = __builtin_shufflevector(__s1_296, __s1_296, 7, 6, 5, 4, 3, 2, 1, 0); \
46258		__ret_296 = __rev0_296 * __noswap_splat_laneq_f16(__rev1_296, __p2_296); \
46259		__ret_296 = __builtin_shufflevector(__ret_296, __ret_296, 3, 2, 1, 0); \
46260		__ret_296; \
46261		})
	46707	__ai poly64x2_t vdupq_n_p64(poly64_t __p0) {
	46708	poly64x2_t __ret;
	46709	__ret = (poly64x2_t) {__p0, __p0};
	46710	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	46711	return __ret;
	46712	}
46262	46713	#endif
46263	46714
46264	46715	#ifdef __LITTLE_ENDIAN__
46265		__ai float16x8_t vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
46266		float16x8_t __ret;
46267		__ret = (float16x8_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	46716	__ai float64x2_t vdupq_n_f64(float64_t __p0) {
	46717	float64x2_t __ret;
	46718	__ret = (float64x2_t) {__p0, __p0};
46268	46719	return __ret;
46269	46720	}
46270	46721	#else
46271		__ai float16x8_t vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
46272		float16x8_t __ret;
46273		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46274		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46275		__ret = (float16x8_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
46276		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	46722	__ai float64x2_t vdupq_n_f64(float64_t __p0) {
	46723	float64x2_t __ret;
	46724	__ret = (float64x2_t) {__p0, __p0};
	46725	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46277	46726	return __ret;
46278	46727	}
46279		__ai float16x8_t __noswap_vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
46280		float16x8_t __ret;
46281		__ret = (float16x8_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	46728	#endif
	46729
	46730	__ai float64x1_t vdup_n_f64(float64_t __p0) {
	46731	float64x1_t __ret;
	46732	__ret = (float64x1_t) {__p0};
46282	46733	return __ret;
46283	46734	}
	46735	#define vext_p64(__p0, __p1, __p2) __extension__ ({ \
	46736	poly64x1_t __ret; \
	46737	poly64x1_t __s0 = __p0; \
	46738	poly64x1_t __s1 = __p1; \
	46739	__ret = (poly64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
	46740	__ret; \
	46741	})
	46742	#ifdef __LITTLE_ENDIAN__
	46743	#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
	46744	poly64x2_t __ret; \
	46745	poly64x2_t __s0 = __p0; \
	46746	poly64x2_t __s1 = __p1; \
	46747	__ret = (poly64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
	46748	__ret; \
	46749	})
	46750	#else
	46751	#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
	46752	poly64x2_t __ret; \
	46753	poly64x2_t __s0 = __p0; \
	46754	poly64x2_t __s1 = __p1; \
	46755	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46756	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	46757	__ret = (poly64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
	46758	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	46759	__ret; \
	46760	})
46284	46761	#endif
46285	46762
46286	46763	#ifdef __LITTLE_ENDIAN__
46287		__ai float16x4_t vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
46288		float16x4_t __ret;
46289		__ret = (float16x4_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	46764	#define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
	46765	float64x2_t __ret; \
	46766	float64x2_t __s0 = __p0; \
	46767	float64x2_t __s1 = __p1; \
	46768	__ret = (float64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 42); \
	46769	__ret; \
	46770	})
	46771	#else
	46772	#define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
	46773	float64x2_t __ret; \
	46774	float64x2_t __s0 = __p0; \
	46775	float64x2_t __s1 = __p1; \
	46776	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46777	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	46778	__ret = (float64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 42); \
	46779	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	46780	__ret; \
	46781	})
	46782	#endif
	46783
	46784	#define vext_f64(__p0, __p1, __p2) __extension__ ({ \
	46785	float64x1_t __ret; \
	46786	float64x1_t __s0 = __p0; \
	46787	float64x1_t __s1 = __p1; \
	46788	__ret = (float64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
	46789	__ret; \
	46790	})
	46791	#ifdef __LITTLE_ENDIAN__
	46792	__ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	46793	float64x2_t __ret;
	46794	__ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
46290	46795	return __ret;
46291	46796	}
46292	46797	#else
46293		__ai float16x4_t vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
46294		float16x4_t __ret;
46295		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46296		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46297		__ret = (float16x4_t) __builtin_neon_vmulx_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
46298		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	46798	__ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	46799	float64x2_t __ret;
	46800	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	46801	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	46802	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	46803	__ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	46804	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46299	46805	return __ret;
46300	46806	}
46301		__ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
46302		float16x4_t __ret;
46303		__ret = (float16x4_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	46807	__ai float64x2_t __noswap_vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	46808	float64x2_t __ret;
	46809	__ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
46304	46810	return __ret;
46305	46811	}
46306	46812	#endif
46307	46813
	46814	__ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	46815	float64x1_t __ret;
	46816	__ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	46817	return __ret;
	46818	}
	46819	#define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46820	float64_t __ret; \
	46821	float64_t __s0 = __p0; \
	46822	float64_t __s1 = __p1; \
	46823	float64x1_t __s2 = __p2; \
	46824	__ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (float64x1_t)__s2, __p3); \
	46825	__ret; \
	46826	})
46308	46827	#ifdef __LITTLE_ENDIAN__
46309		#define vmulxh_lane_f16(__p0, __p1, __p2) __extension__ ({ \
46310		float16_t __ret; \
46311		float16_t __s0 = __p0; \
46312		float16x4_t __s1 = __p1; \
46313		__ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__s1, __p2); \
	46828	#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46829	float32_t __ret; \
	46830	float32_t __s0 = __p0; \
	46831	float32_t __s1 = __p1; \
	46832	float32x2_t __s2 = __p2; \
	46833	__ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
46314	46834	__ret; \
46315	46835	})
46316	46836	#else
46317		#define vmulxh_lane_f16(__p0, __p1, __p2) __extension__ ({ \
46318		float16_t __ret; \
46319		float16_t __s0 = __p0; \
46320		float16x4_t __s1 = __p1; \
46321		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
46322		__ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__rev1, __p2); \
	46837	#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46838	float32_t __ret; \
	46839	float32_t __s0 = __p0; \
	46840	float32_t __s1 = __p1; \
	46841	float32x2_t __s2 = __p2; \
	46842	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	46843	__ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__rev2, __p3); \
	46844	__ret; \
	46845	})
	46846	#define __noswap_vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46847	float32_t __ret; \
	46848	float32_t __s0 = __p0; \
	46849	float32_t __s1 = __p1; \
	46850	float32x2_t __s2 = __p2; \
	46851	__ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
46323	46852	__ret; \
46324	46853	})
46325	46854	#endif
46326	46855
46327	46856	#ifdef __LITTLE_ENDIAN__
46328		#define vmulxq_lane_f16(__p0_297, __p1_297, __p2_297) __extension__ ({ \
46329		float16x8_t __ret_297; \
46330		float16x8_t __s0_297 = __p0_297; \
46331		float16x4_t __s1_297 = __p1_297; \
46332		__ret_297 = vmulxq_f16(__s0_297, splatq_lane_f16(__s1_297, __p2_297)); \
46333		__ret_297; \
	46857	#define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46858	float64x2_t __ret; \
	46859	float64x2_t __s0 = __p0; \
	46860	float64x2_t __s1 = __p1; \
	46861	float64x1_t __s2 = __p2; \
	46862	__ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 42); \
	46863	__ret; \
46334	46864	})
46335	46865	#else
46336		#define vmulxq_lane_f16(__p0_298, __p1_298, __p2_298) __extension__ ({ \
46337		float16x8_t __ret_298; \
46338		float16x8_t __s0_298 = __p0_298; \
46339		float16x4_t __s1_298 = __p1_298; \
46340		float16x8_t __rev0_298; __rev0_298 = __builtin_shufflevector(__s0_298, __s0_298, 7, 6, 5, 4, 3, 2, 1, 0); \
46341		float16x4_t __rev1_298; __rev1_298 = __builtin_shufflevector(__s1_298, __s1_298, 3, 2, 1, 0); \
46342		__ret_298 = __noswap_vmulxq_f16(__rev0_298, __noswap_splatq_lane_f16(__rev1_298, __p2_298)); \
46343		__ret_298 = __builtin_shufflevector(__ret_298, __ret_298, 7, 6, 5, 4, 3, 2, 1, 0); \
46344		__ret_298; \
	46866	#define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46867	float64x2_t __ret; \
	46868	float64x2_t __s0 = __p0; \
	46869	float64x2_t __s1 = __p1; \
	46870	float64x1_t __s2 = __p2; \
	46871	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46872	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	46873	__ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__s2, __p3, 42); \
	46874	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	46875	__ret; \
	46876	})
	46877	#define __noswap_vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46878	float64x2_t __ret; \
	46879	float64x2_t __s0 = __p0; \
	46880	float64x2_t __s1 = __p1; \
	46881	float64x1_t __s2 = __p2; \
	46882	__ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 42); \
	46883	__ret; \
46345	46884	})
46346	46885	#endif
46347	46886
46348	46887	#ifdef __LITTLE_ENDIAN__
46349		#define vmulx_lane_f16(__p0_299, __p1_299, __p2_299) __extension__ ({ \
46350		float16x4_t __ret_299; \
46351		float16x4_t __s0_299 = __p0_299; \
46352		float16x4_t __s1_299 = __p1_299; \
46353		__ret_299 = vmulx_f16(__s0_299, splat_lane_f16(__s1_299, __p2_299)); \
46354		__ret_299; \
	46888	#define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46889	float32x4_t __ret; \
	46890	float32x4_t __s0 = __p0; \
	46891	float32x4_t __s1 = __p1; \
	46892	float32x2_t __s2 = __p2; \
	46893	__ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 41); \
	46894	__ret; \
46355	46895	})
46356	46896	#else
46357		#define vmulx_lane_f16(__p0_300, __p1_300, __p2_300) __extension__ ({ \
46358		float16x4_t __ret_300; \
46359		float16x4_t __s0_300 = __p0_300; \
46360		float16x4_t __s1_300 = __p1_300; \
46361		float16x4_t __rev0_300; __rev0_300 = __builtin_shufflevector(__s0_300, __s0_300, 3, 2, 1, 0); \
46362		float16x4_t __rev1_300; __rev1_300 = __builtin_shufflevector(__s1_300, __s1_300, 3, 2, 1, 0); \
46363		__ret_300 = __noswap_vmulx_f16(__rev0_300, __noswap_splat_lane_f16(__rev1_300, __p2_300)); \
46364		__ret_300 = __builtin_shufflevector(__ret_300, __ret_300, 3, 2, 1, 0); \
46365		__ret_300; \
	46897	#define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46898	float32x4_t __ret; \
	46899	float32x4_t __s0 = __p0; \
	46900	float32x4_t __s1 = __p1; \
	46901	float32x2_t __s2 = __p2; \
	46902	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	46903	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	46904	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	46905	__ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, __p3, 41); \
	46906	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	46907	__ret; \
	46908	})
	46909	#define __noswap_vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46910	float32x4_t __ret; \
	46911	float32x4_t __s0 = __p0; \
	46912	float32x4_t __s1 = __p1; \
	46913	float32x2_t __s2 = __p2; \
	46914	__ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 41); \
	46915	__ret; \
	46916	})
	46917	#endif
	46918
	46919	#define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46920	float64x1_t __ret; \
	46921	float64x1_t __s0 = __p0; \
	46922	float64x1_t __s1 = __p1; \
	46923	float64x1_t __s2 = __p2; \
	46924	__ret = (float64x1_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 10); \
	46925	__ret; \
	46926	})
	46927	#ifdef __LITTLE_ENDIAN__
	46928	#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46929	float32x2_t __ret; \
	46930	float32x2_t __s0 = __p0; \
	46931	float32x2_t __s1 = __p1; \
	46932	float32x2_t __s2 = __p2; \
	46933	__ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 9); \
	46934	__ret; \
	46935	})
	46936	#else
	46937	#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46938	float32x2_t __ret; \
	46939	float32x2_t __s0 = __p0; \
	46940	float32x2_t __s1 = __p1; \
	46941	float32x2_t __s2 = __p2; \
	46942	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	46943	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	46944	float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	46945	__ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, __p3, 9); \
	46946	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	46947	__ret; \
	46948	})
	46949	#define __noswap_vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46950	float32x2_t __ret; \
	46951	float32x2_t __s0 = __p0; \
	46952	float32x2_t __s1 = __p1; \
	46953	float32x2_t __s2 = __p2; \
	46954	__ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 9); \
	46955	__ret; \
	46956	})
	46957	#endif
	46958
	46959	#ifdef __LITTLE_ENDIAN__
	46960	#define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46961	float64_t __ret; \
	46962	float64_t __s0 = __p0; \
	46963	float64_t __s1 = __p1; \
	46964	float64x2_t __s2 = __p2; \
	46965	__ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
	46966	__ret; \
	46967	})
	46968	#else
	46969	#define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46970	float64_t __ret; \
	46971	float64_t __s0 = __p0; \
	46972	float64_t __s1 = __p1; \
	46973	float64x2_t __s2 = __p2; \
	46974	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	46975	__ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__rev2, __p3); \
	46976	__ret; \
	46977	})
	46978	#define __noswap_vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	46979	float64_t __ret; \
	46980	float64_t __s0 = __p0; \
	46981	float64_t __s1 = __p1; \
	46982	float64x2_t __s2 = __p2; \
	46983	__ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
	46984	__ret; \
46366	46985	})
46367	46986	#endif
46368	46987
46369	46988	#ifdef __LITTLE_ENDIAN__
46370		#define vmulxh_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
46371		float16_t __ret; \
46372		float16_t __s0 = __p0; \
46373		float16x8_t __s1 = __p1; \
46374		__ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__s1, __p2); \
	46989	#define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46990	float32_t __ret; \
	46991	float32_t __s0 = __p0; \
	46992	float32_t __s1 = __p1; \
	46993	float32x4_t __s2 = __p2; \
	46994	__ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
46375	46995	__ret; \
46376	46996	})
46377	46997	#else
46378		#define vmulxh_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
46379		float16_t __ret; \
46380		float16_t __s0 = __p0; \
46381		float16x8_t __s1 = __p1; \
46382		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
46383		__ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__rev1, __p2); \
	46998	#define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	46999	float32_t __ret; \
	47000	float32_t __s0 = __p0; \
	47001	float32_t __s1 = __p1; \
	47002	float32x4_t __s2 = __p2; \
	47003	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	47004	__ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__rev2, __p3); \
	47005	__ret; \
	47006	})
	47007	#define __noswap_vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47008	float32_t __ret; \
	47009	float32_t __s0 = __p0; \
	47010	float32_t __s1 = __p1; \
	47011	float32x4_t __s2 = __p2; \
	47012	__ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
46384	47013	__ret; \
46385	47014	})
46386	47015	#endif
46387	47016
46388	47017	#ifdef __LITTLE_ENDIAN__
46389		#define vmulxq_laneq_f16(__p0_301, __p1_301, __p2_301) __extension__ ({ \
46390		float16x8_t __ret_301; \
46391		float16x8_t __s0_301 = __p0_301; \
46392		float16x8_t __s1_301 = __p1_301; \
46393		__ret_301 = vmulxq_f16(__s0_301, splatq_laneq_f16(__s1_301, __p2_301)); \
46394		__ret_301; \
	47018	#define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	47019	float64x2_t __ret; \
	47020	float64x2_t __s0 = __p0; \
	47021	float64x2_t __s1 = __p1; \
	47022	float64x2_t __s2 = __p2; \
	47023	__ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 42); \
	47024	__ret; \
46395	47025	})
46396	47026	#else
46397		#define vmulxq_laneq_f16(__p0_302, __p1_302, __p2_302) __extension__ ({ \
46398		float16x8_t __ret_302; \
46399		float16x8_t __s0_302 = __p0_302; \
46400		float16x8_t __s1_302 = __p1_302; \
46401		float16x8_t __rev0_302; __rev0_302 = __builtin_shufflevector(__s0_302, __s0_302, 7, 6, 5, 4, 3, 2, 1, 0); \
46402		float16x8_t __rev1_302; __rev1_302 = __builtin_shufflevector(__s1_302, __s1_302, 7, 6, 5, 4, 3, 2, 1, 0); \
46403		__ret_302 = __noswap_vmulxq_f16(__rev0_302, __noswap_splatq_laneq_f16(__rev1_302, __p2_302)); \
46404		__ret_302 = __builtin_shufflevector(__ret_302, __ret_302, 7, 6, 5, 4, 3, 2, 1, 0); \
46405		__ret_302; \
	47027	#define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	47028	float64x2_t __ret; \
	47029	float64x2_t __s0 = __p0; \
	47030	float64x2_t __s1 = __p1; \
	47031	float64x2_t __s2 = __p2; \
	47032	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	47033	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	47034	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	47035	__ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 42); \
	47036	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47037	__ret; \
	47038	})
	47039	#define __noswap_vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	47040	float64x2_t __ret; \
	47041	float64x2_t __s0 = __p0; \
	47042	float64x2_t __s1 = __p1; \
	47043	float64x2_t __s2 = __p2; \
	47044	__ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 42); \
	47045	__ret; \
46406	47046	})
46407	47047	#endif
46408	47048
46409	47049	#ifdef __LITTLE_ENDIAN__
46410		#define vmulx_laneq_f16(__p0_303, __p1_303, __p2_303) __extension__ ({ \
46411		float16x4_t __ret_303; \
46412		float16x4_t __s0_303 = __p0_303; \
46413		float16x8_t __s1_303 = __p1_303; \
46414		__ret_303 = vmulx_f16(__s0_303, splat_laneq_f16(__s1_303, __p2_303)); \
46415		__ret_303; \
	47050	#define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47051	float32x4_t __ret; \
	47052	float32x4_t __s0 = __p0; \
	47053	float32x4_t __s1 = __p1; \
	47054	float32x4_t __s2 = __p2; \
	47055	__ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 41); \
	47056	__ret; \
46416	47057	})
46417	47058	#else
46418		#define vmulx_laneq_f16(__p0_304, __p1_304, __p2_304) __extension__ ({ \
46419		float16x4_t __ret_304; \
46420		float16x4_t __s0_304 = __p0_304; \
46421		float16x8_t __s1_304 = __p1_304; \
46422		float16x4_t __rev0_304; __rev0_304 = __builtin_shufflevector(__s0_304, __s0_304, 3, 2, 1, 0); \
46423		float16x8_t __rev1_304; __rev1_304 = __builtin_shufflevector(__s1_304, __s1_304, 7, 6, 5, 4, 3, 2, 1, 0); \
46424		__ret_304 = __noswap_vmulx_f16(__rev0_304, __noswap_splat_laneq_f16(__rev1_304, __p2_304)); \
46425		__ret_304 = __builtin_shufflevector(__ret_304, __ret_304, 3, 2, 1, 0); \
46426		__ret_304; \
	47059	#define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47060	float32x4_t __ret; \
	47061	float32x4_t __s0 = __p0; \
	47062	float32x4_t __s1 = __p1; \
	47063	float32x4_t __s2 = __p2; \
	47064	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	47065	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	47066	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	47067	__ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 41); \
	47068	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	47069	__ret; \
	47070	})
	47071	#define __noswap_vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47072	float32x4_t __ret; \
	47073	float32x4_t __s0 = __p0; \
	47074	float32x4_t __s1 = __p1; \
	47075	float32x4_t __s2 = __p2; \
	47076	__ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 41); \
	47077	__ret; \
46427	47078	})
46428	47079	#endif
46429	47080
46430	47081	#ifdef __LITTLE_ENDIAN__
46431		#define vmulxq_n_f16(__p0, __p1) __extension__ ({ \
46432		float16x8_t __ret; \
46433		float16x8_t __s0 = __p0; \
46434		float16_t __s1 = __p1; \
46435		__ret = vmulxq_f16(__s0, (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}); \
	47082	#define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	47083	float64x1_t __ret; \
	47084	float64x1_t __s0 = __p0; \
	47085	float64x1_t __s1 = __p1; \
	47086	float64x2_t __s2 = __p2; \
	47087	__ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 10); \
46436	47088	__ret; \
46437	47089	})
46438	47090	#else
46439		#define vmulxq_n_f16(__p0, __p1) __extension__ ({ \
46440		float16x8_t __ret; \
46441		float16x8_t __s0 = __p0; \
46442		float16_t __s1 = __p1; \
46443		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46444		__ret = __noswap_vmulxq_f16(__rev0, (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}); \
46445		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	47091	#define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	47092	float64x1_t __ret; \
	47093	float64x1_t __s0 = __p0; \
	47094	float64x1_t __s1 = __p1; \
	47095	float64x2_t __s2 = __p2; \
	47096	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	47097	__ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__rev2, __p3, 10); \
	47098	__ret; \
	47099	})
	47100	#define __noswap_vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
	47101	float64x1_t __ret; \
	47102	float64x1_t __s0 = __p0; \
	47103	float64x1_t __s1 = __p1; \
	47104	float64x2_t __s2 = __p2; \
	47105	__ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 10); \
46446	47106	__ret; \
46447	47107	})
46448	47108	#endif
46449	47109
46450	47110	#ifdef __LITTLE_ENDIAN__
46451		#define vmulx_n_f16(__p0, __p1) __extension__ ({ \
46452		float16x4_t __ret; \
46453		float16x4_t __s0 = __p0; \
46454		float16_t __s1 = __p1; \
46455		__ret = vmulx_f16(__s0, (float16x4_t) {__s1, __s1, __s1, __s1}); \
	47111	#define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47112	float32x2_t __ret; \
	47113	float32x2_t __s0 = __p0; \
	47114	float32x2_t __s1 = __p1; \
	47115	float32x4_t __s2 = __p2; \
	47116	__ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 9); \
46456	47117	__ret; \
46457	47118	})
46458	47119	#else
46459		#define vmulx_n_f16(__p0, __p1) __extension__ ({ \
46460		float16x4_t __ret; \
46461		float16x4_t __s0 = __p0; \
46462		float16_t __s1 = __p1; \
46463		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46464		__ret = __noswap_vmulx_f16(__rev0, (float16x4_t) {__s1, __s1, __s1, __s1}); \
46465		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	47120	#define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47121	float32x2_t __ret; \
	47122	float32x2_t __s0 = __p0; \
	47123	float32x2_t __s1 = __p1; \
	47124	float32x4_t __s2 = __p2; \
	47125	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	47126	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	47127	float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	47128	__ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x16_t)__rev2, __p3, 9); \
	47129	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47130	__ret; \
	47131	})
	47132	#define __noswap_vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
	47133	float32x2_t __ret; \
	47134	float32x2_t __s0 = __p0; \
	47135	float32x2_t __s1 = __p1; \
	47136	float32x4_t __s2 = __p2; \
	47137	__ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 9); \
46466	47138	__ret; \
46467	47139	})
46468	47140	#endif
46469	47141
46470	47142	#ifdef __LITTLE_ENDIAN__
46471		__ai float16x8_t vpaddq_f16(float16x8_t __p0, float16x8_t __p1) {
46472		float16x8_t __ret;
46473		__ret = (float16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	47143	__ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
	47144	float64x2_t __ret;
	47145	__ret = vfmaq_f64(__p0, __p1, (float64x2_t) {__p2, __p2});
46474	47146	return __ret;
46475	47147	}
46476	47148	#else
46477		__ai float16x8_t vpaddq_f16(float16x8_t __p0, float16x8_t __p1) {
46478		float16x8_t __ret;
46479		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46480		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46481		__ret = (float16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
46482		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	47149	__ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
	47150	float64x2_t __ret;
	47151	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47152	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	47153	__ret = __noswap_vfmaq_f64(__rev0, __rev1, (float64x2_t) {__p2, __p2});
	47154	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46483	47155	return __ret;
46484	47156	}
46485	47157	#endif
46486	47158
46487		#ifdef __LITTLE_ENDIAN__
46488		__ai float16x8_t vpmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
46489		float16x8_t __ret;
46490		__ret = (float16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
46491		return __ret;
46492		}
46493		#else
46494		__ai float16x8_t vpmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
46495		float16x8_t __ret;
46496		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46497		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46498		__ret = (float16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
46499		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	47159	__ai float64x1_t vfma_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
	47160	float64x1_t __ret;
	47161	__ret = vfma_f64(__p0, __p1, (float64x1_t) {__p2});
46500	47162	return __ret;
46501	47163	}
46502		#endif
46503
46504	47164	#ifdef __LITTLE_ENDIAN__
46505		__ai float16x8_t vpmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
46506		float16x8_t __ret;
46507		__ret = (float16x8_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
	47165	__ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	47166	float64x2_t __ret;
	47167	__ret = vfmaq_f64(__p0, -__p1, __p2);
46508	47168	return __ret;
46509	47169	}
46510	47170	#else
46511		__ai float16x8_t vpmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
46512		float16x8_t __ret;
46513		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46514		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46515		__ret = (float16x8_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
46516		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	47171	__ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	47172	float64x2_t __ret;
	47173	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47174	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	47175	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	47176	__ret = __noswap_vfmaq_f64(__rev0, -__rev1, __rev2);
	47177	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46517	47178	return __ret;
46518	47179	}
46519	47180	#endif
46520	47181
46521		#ifdef __LITTLE_ENDIAN__
46522		__ai float16x4_t vpmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
46523		float16x4_t __ret;
46524		__ret = (float16x4_t) __builtin_neon_vpmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
	47182	__ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	47183	float64x1_t __ret;
	47184	__ret = vfma_f64(__p0, -__p1, __p2);
46525	47185	return __ret;
46526	47186	}
	47187	#define vfmsd_lane_f64(__p0_405, __p1_405, __p2_405, __p3_405) __extension__ ({ \
	47188	float64_t __ret_405; \
	47189	float64_t __s0_405 = __p0_405; \
	47190	float64_t __s1_405 = __p1_405; \
	47191	float64x1_t __s2_405 = __p2_405; \
	47192	__ret_405 = vfmad_lane_f64(__s0_405, -__s1_405, __s2_405, __p3_405); \
	47193	__ret_405; \
	47194	})
	47195	#ifdef __LITTLE_ENDIAN__
	47196	#define vfmss_lane_f32(__p0_406, __p1_406, __p2_406, __p3_406) __extension__ ({ \
	47197	float32_t __ret_406; \
	47198	float32_t __s0_406 = __p0_406; \
	47199	float32_t __s1_406 = __p1_406; \
	47200	float32x2_t __s2_406 = __p2_406; \
	47201	__ret_406 = vfmas_lane_f32(__s0_406, -__s1_406, __s2_406, __p3_406); \
	47202	__ret_406; \
	47203	})
46527	47204	#else
46528		__ai float16x4_t vpmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
46529		float16x4_t __ret;
46530		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46531		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46532		__ret = (float16x4_t) __builtin_neon_vpmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
46533		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46534		return __ret;
46535		}
	47205	#define vfmss_lane_f32(__p0_407, __p1_407, __p2_407, __p3_407) __extension__ ({ \
	47206	float32_t __ret_407; \
	47207	float32_t __s0_407 = __p0_407; \
	47208	float32_t __s1_407 = __p1_407; \
	47209	float32x2_t __s2_407 = __p2_407; \
	47210	float32x2_t __rev2_407; __rev2_407 = __builtin_shufflevector(__s2_407, __s2_407, 1, 0); \
	47211	__ret_407 = __noswap_vfmas_lane_f32(__s0_407, -__s1_407, __rev2_407, __p3_407); \
	47212	__ret_407; \
	47213	})
46536	47214	#endif
46537	47215
46538	47216	#ifdef __LITTLE_ENDIAN__
46539		__ai float16x8_t vpminq_f16(float16x8_t __p0, float16x8_t __p1) {
46540		float16x8_t __ret;
46541		__ret = (float16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
46542		return __ret;
46543		}
	47217	#define vfmsq_lane_f64(__p0_408, __p1_408, __p2_408, __p3_408) __extension__ ({ \
	47218	float64x2_t __ret_408; \
	47219	float64x2_t __s0_408 = __p0_408; \
	47220	float64x2_t __s1_408 = __p1_408; \
	47221	float64x1_t __s2_408 = __p2_408; \
	47222	__ret_408 = vfmaq_lane_f64(__s0_408, -__s1_408, __s2_408, __p3_408); \
	47223	__ret_408; \
	47224	})
46544	47225	#else
46545		__ai float16x8_t vpminq_f16(float16x8_t __p0, float16x8_t __p1) {
46546		float16x8_t __ret;
46547		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46548		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46549		__ret = (float16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
46550		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46551		return __ret;
46552		}
	47226	#define vfmsq_lane_f64(__p0_409, __p1_409, __p2_409, __p3_409) __extension__ ({ \
	47227	float64x2_t __ret_409; \
	47228	float64x2_t __s0_409 = __p0_409; \
	47229	float64x2_t __s1_409 = __p1_409; \
	47230	float64x1_t __s2_409 = __p2_409; \
	47231	float64x2_t __rev0_409; __rev0_409 = __builtin_shufflevector(__s0_409, __s0_409, 1, 0); \
	47232	float64x2_t __rev1_409; __rev1_409 = __builtin_shufflevector(__s1_409, __s1_409, 1, 0); \
	47233	__ret_409 = __noswap_vfmaq_lane_f64(__rev0_409, -__rev1_409, __s2_409, __p3_409); \
	47234	__ret_409 = __builtin_shufflevector(__ret_409, __ret_409, 1, 0); \
	47235	__ret_409; \
	47236	})
46553	47237	#endif
46554	47238
46555	47239	#ifdef __LITTLE_ENDIAN__
46556		__ai float16x8_t vpminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
46557		float16x8_t __ret;
46558		__ret = (float16x8_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 40);
46559		return __ret;
46560		}
	47240	#define vfmsq_lane_f32(__p0_410, __p1_410, __p2_410, __p3_410) __extension__ ({ \
	47241	float32x4_t __ret_410; \
	47242	float32x4_t __s0_410 = __p0_410; \
	47243	float32x4_t __s1_410 = __p1_410; \
	47244	float32x2_t __s2_410 = __p2_410; \
	47245	__ret_410 = vfmaq_lane_f32(__s0_410, -__s1_410, __s2_410, __p3_410); \
	47246	__ret_410; \
	47247	})
46561	47248	#else
46562		__ai float16x8_t vpminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
46563		float16x8_t __ret;
46564		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46565		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46566		__ret = (float16x8_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
46567		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46568		return __ret;
46569		}
	47249	#define vfmsq_lane_f32(__p0_411, __p1_411, __p2_411, __p3_411) __extension__ ({ \
	47250	float32x4_t __ret_411; \
	47251	float32x4_t __s0_411 = __p0_411; \
	47252	float32x4_t __s1_411 = __p1_411; \
	47253	float32x2_t __s2_411 = __p2_411; \
	47254	float32x4_t __rev0_411; __rev0_411 = __builtin_shufflevector(__s0_411, __s0_411, 3, 2, 1, 0); \
	47255	float32x4_t __rev1_411; __rev1_411 = __builtin_shufflevector(__s1_411, __s1_411, 3, 2, 1, 0); \
	47256	float32x2_t __rev2_411; __rev2_411 = __builtin_shufflevector(__s2_411, __s2_411, 1, 0); \
	47257	__ret_411 = __noswap_vfmaq_lane_f32(__rev0_411, -__rev1_411, __rev2_411, __p3_411); \
	47258	__ret_411 = __builtin_shufflevector(__ret_411, __ret_411, 3, 2, 1, 0); \
	47259	__ret_411; \
	47260	})
46570	47261	#endif
46571	47262
	47263	#define vfms_lane_f64(__p0_412, __p1_412, __p2_412, __p3_412) __extension__ ({ \
	47264	float64x1_t __ret_412; \
	47265	float64x1_t __s0_412 = __p0_412; \
	47266	float64x1_t __s1_412 = __p1_412; \
	47267	float64x1_t __s2_412 = __p2_412; \
	47268	__ret_412 = vfma_lane_f64(__s0_412, -__s1_412, __s2_412, __p3_412); \
	47269	__ret_412; \
	47270	})
46572	47271	#ifdef __LITTLE_ENDIAN__
46573		__ai float16x4_t vpminnm_f16(float16x4_t __p0, float16x4_t __p1) {
46574		float16x4_t __ret;
46575		__ret = (float16x4_t) __builtin_neon_vpminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 8);
46576		return __ret;
46577		}
	47272	#define vfms_lane_f32(__p0_413, __p1_413, __p2_413, __p3_413) __extension__ ({ \
	47273	float32x2_t __ret_413; \
	47274	float32x2_t __s0_413 = __p0_413; \
	47275	float32x2_t __s1_413 = __p1_413; \
	47276	float32x2_t __s2_413 = __p2_413; \
	47277	__ret_413 = vfma_lane_f32(__s0_413, -__s1_413, __s2_413, __p3_413); \
	47278	__ret_413; \
	47279	})
46578	47280	#else
46579		__ai float16x4_t vpminnm_f16(float16x4_t __p0, float16x4_t __p1) {
46580		float16x4_t __ret;
46581		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46582		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46583		__ret = (float16x4_t) __builtin_neon_vpminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
46584		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46585		return __ret;
46586		}
	47281	#define vfms_lane_f32(__p0_414, __p1_414, __p2_414, __p3_414) __extension__ ({ \
	47282	float32x2_t __ret_414; \
	47283	float32x2_t __s0_414 = __p0_414; \
	47284	float32x2_t __s1_414 = __p1_414; \
	47285	float32x2_t __s2_414 = __p2_414; \
	47286	float32x2_t __rev0_414; __rev0_414 = __builtin_shufflevector(__s0_414, __s0_414, 1, 0); \
	47287	float32x2_t __rev1_414; __rev1_414 = __builtin_shufflevector(__s1_414, __s1_414, 1, 0); \
	47288	float32x2_t __rev2_414; __rev2_414 = __builtin_shufflevector(__s2_414, __s2_414, 1, 0); \
	47289	__ret_414 = __noswap_vfma_lane_f32(__rev0_414, -__rev1_414, __rev2_414, __p3_414); \
	47290	__ret_414 = __builtin_shufflevector(__ret_414, __ret_414, 1, 0); \
	47291	__ret_414; \
	47292	})
46587	47293	#endif
46588	47294
46589	47295	#ifdef __LITTLE_ENDIAN__
46590		__ai float16x8_t vrndiq_f16(float16x8_t __p0) {
46591		float16x8_t __ret;
46592		__ret = (float16x8_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 40);
46593		return __ret;
46594		}
	47296	#define vfmsd_laneq_f64(__p0_415, __p1_415, __p2_415, __p3_415) __extension__ ({ \
	47297	float64_t __ret_415; \
	47298	float64_t __s0_415 = __p0_415; \
	47299	float64_t __s1_415 = __p1_415; \
	47300	float64x2_t __s2_415 = __p2_415; \
	47301	__ret_415 = vfmad_laneq_f64(__s0_415, -__s1_415, __s2_415, __p3_415); \
	47302	__ret_415; \
	47303	})
46595	47304	#else
46596		__ai float16x8_t vrndiq_f16(float16x8_t __p0) {
46597		float16x8_t __ret;
46598		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46599		__ret = (float16x8_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 40);
46600		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46601		return __ret;
46602		}
	47305	#define vfmsd_laneq_f64(__p0_416, __p1_416, __p2_416, __p3_416) __extension__ ({ \
	47306	float64_t __ret_416; \
	47307	float64_t __s0_416 = __p0_416; \
	47308	float64_t __s1_416 = __p1_416; \
	47309	float64x2_t __s2_416 = __p2_416; \
	47310	float64x2_t __rev2_416; __rev2_416 = __builtin_shufflevector(__s2_416, __s2_416, 1, 0); \
	47311	__ret_416 = __noswap_vfmad_laneq_f64(__s0_416, -__s1_416, __rev2_416, __p3_416); \
	47312	__ret_416; \
	47313	})
46603	47314	#endif
46604	47315
46605	47316	#ifdef __LITTLE_ENDIAN__
46606		__ai float16x4_t vrndi_f16(float16x4_t __p0) {
46607		float16x4_t __ret;
46608		__ret = (float16x4_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 8);
46609		return __ret;
46610		}
	47317	#define vfmss_laneq_f32(__p0_417, __p1_417, __p2_417, __p3_417) __extension__ ({ \
	47318	float32_t __ret_417; \
	47319	float32_t __s0_417 = __p0_417; \
	47320	float32_t __s1_417 = __p1_417; \
	47321	float32x4_t __s2_417 = __p2_417; \
	47322	__ret_417 = vfmas_laneq_f32(__s0_417, -__s1_417, __s2_417, __p3_417); \
	47323	__ret_417; \
	47324	})
46611	47325	#else
46612		__ai float16x4_t vrndi_f16(float16x4_t __p0) {
46613		float16x4_t __ret;
46614		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46615		__ret = (float16x4_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 8);
46616		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46617		return __ret;
46618		}
	47326	#define vfmss_laneq_f32(__p0_418, __p1_418, __p2_418, __p3_418) __extension__ ({ \
	47327	float32_t __ret_418; \
	47328	float32_t __s0_418 = __p0_418; \
	47329	float32_t __s1_418 = __p1_418; \
	47330	float32x4_t __s2_418 = __p2_418; \
	47331	float32x4_t __rev2_418; __rev2_418 = __builtin_shufflevector(__s2_418, __s2_418, 3, 2, 1, 0); \
	47332	__ret_418 = __noswap_vfmas_laneq_f32(__s0_418, -__s1_418, __rev2_418, __p3_418); \
	47333	__ret_418; \
	47334	})
46619	47335	#endif
46620	47336
46621	47337	#ifdef __LITTLE_ENDIAN__
46622		__ai float16x8_t vsqrtq_f16(float16x8_t __p0) {
46623		float16x8_t __ret;
46624		__ret = (float16x8_t) __builtin_neon_vsqrtq_v((int8x16_t)__p0, 40);
46625		return __ret;
46626		}
	47338	#define vfmsq_laneq_f64(__p0_419, __p1_419, __p2_419, __p3_419) __extension__ ({ \
	47339	float64x2_t __ret_419; \
	47340	float64x2_t __s0_419 = __p0_419; \
	47341	float64x2_t __s1_419 = __p1_419; \
	47342	float64x2_t __s2_419 = __p2_419; \
	47343	__ret_419 = vfmaq_laneq_f64(__s0_419, -__s1_419, __s2_419, __p3_419); \
	47344	__ret_419; \
	47345	})
46627	47346	#else
46628		__ai float16x8_t vsqrtq_f16(float16x8_t __p0) {
46629		float16x8_t __ret;
46630		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46631		__ret = (float16x8_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 40);
46632		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46633		return __ret;
46634		}
	47347	#define vfmsq_laneq_f64(__p0_420, __p1_420, __p2_420, __p3_420) __extension__ ({ \
	47348	float64x2_t __ret_420; \
	47349	float64x2_t __s0_420 = __p0_420; \
	47350	float64x2_t __s1_420 = __p1_420; \
	47351	float64x2_t __s2_420 = __p2_420; \
	47352	float64x2_t __rev0_420; __rev0_420 = __builtin_shufflevector(__s0_420, __s0_420, 1, 0); \
	47353	float64x2_t __rev1_420; __rev1_420 = __builtin_shufflevector(__s1_420, __s1_420, 1, 0); \
	47354	float64x2_t __rev2_420; __rev2_420 = __builtin_shufflevector(__s2_420, __s2_420, 1, 0); \
	47355	__ret_420 = __noswap_vfmaq_laneq_f64(__rev0_420, -__rev1_420, __rev2_420, __p3_420); \
	47356	__ret_420 = __builtin_shufflevector(__ret_420, __ret_420, 1, 0); \
	47357	__ret_420; \
	47358	})
46635	47359	#endif
46636	47360
46637	47361	#ifdef __LITTLE_ENDIAN__
46638		__ai float16x4_t vsqrt_f16(float16x4_t __p0) {
46639		float16x4_t __ret;
46640		__ret = (float16x4_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 8);
46641		return __ret;
46642		}
	47362	#define vfmsq_laneq_f32(__p0_421, __p1_421, __p2_421, __p3_421) __extension__ ({ \
	47363	float32x4_t __ret_421; \
	47364	float32x4_t __s0_421 = __p0_421; \
	47365	float32x4_t __s1_421 = __p1_421; \
	47366	float32x4_t __s2_421 = __p2_421; \
	47367	__ret_421 = vfmaq_laneq_f32(__s0_421, -__s1_421, __s2_421, __p3_421); \
	47368	__ret_421; \
	47369	})
46643	47370	#else
46644		__ai float16x4_t vsqrt_f16(float16x4_t __p0) {
46645		float16x4_t __ret;
46646		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46647		__ret = (float16x4_t) __builtin_neon_vsqrt_v((int8x8_t)__rev0, 8);
46648		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46649		return __ret;
46650		}
	47371	#define vfmsq_laneq_f32(__p0_422, __p1_422, __p2_422, __p3_422) __extension__ ({ \
	47372	float32x4_t __ret_422; \
	47373	float32x4_t __s0_422 = __p0_422; \
	47374	float32x4_t __s1_422 = __p1_422; \
	47375	float32x4_t __s2_422 = __p2_422; \
	47376	float32x4_t __rev0_422; __rev0_422 = __builtin_shufflevector(__s0_422, __s0_422, 3, 2, 1, 0); \
	47377	float32x4_t __rev1_422; __rev1_422 = __builtin_shufflevector(__s1_422, __s1_422, 3, 2, 1, 0); \
	47378	float32x4_t __rev2_422; __rev2_422 = __builtin_shufflevector(__s2_422, __s2_422, 3, 2, 1, 0); \
	47379	__ret_422 = __noswap_vfmaq_laneq_f32(__rev0_422, -__rev1_422, __rev2_422, __p3_422); \
	47380	__ret_422 = __builtin_shufflevector(__ret_422, __ret_422, 3, 2, 1, 0); \
	47381	__ret_422; \
	47382	})
46651	47383	#endif
46652	47384
46653	47385	#ifdef __LITTLE_ENDIAN__
46654		__ai float16x8_t vtrn1q_f16(float16x8_t __p0, float16x8_t __p1) {
46655		float16x8_t __ret;
46656		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
46657		return __ret;
46658		}
	47386	#define vfms_laneq_f64(__p0_423, __p1_423, __p2_423, __p3_423) __extension__ ({ \
	47387	float64x1_t __ret_423; \
	47388	float64x1_t __s0_423 = __p0_423; \
	47389	float64x1_t __s1_423 = __p1_423; \
	47390	float64x2_t __s2_423 = __p2_423; \
	47391	__ret_423 = vfma_laneq_f64(__s0_423, -__s1_423, __s2_423, __p3_423); \
	47392	__ret_423; \
	47393	})
46659	47394	#else
46660		__ai float16x8_t vtrn1q_f16(float16x8_t __p0, float16x8_t __p1) {
46661		float16x8_t __ret;
46662		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46663		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46664		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
46665		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46666		return __ret;
46667		}
	47395	#define vfms_laneq_f64(__p0_424, __p1_424, __p2_424, __p3_424) __extension__ ({ \
	47396	float64x1_t __ret_424; \
	47397	float64x1_t __s0_424 = __p0_424; \
	47398	float64x1_t __s1_424 = __p1_424; \
	47399	float64x2_t __s2_424 = __p2_424; \
	47400	float64x2_t __rev2_424; __rev2_424 = __builtin_shufflevector(__s2_424, __s2_424, 1, 0); \
	47401	__ret_424 = __noswap_vfma_laneq_f64(__s0_424, -__s1_424, __rev2_424, __p3_424); \
	47402	__ret_424; \
	47403	})
46668	47404	#endif
46669	47405
46670	47406	#ifdef __LITTLE_ENDIAN__
46671		__ai float16x4_t vtrn1_f16(float16x4_t __p0, float16x4_t __p1) {
46672		float16x4_t __ret;
46673		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
46674		return __ret;
46675		}
	47407	#define vfms_laneq_f32(__p0_425, __p1_425, __p2_425, __p3_425) __extension__ ({ \
	47408	float32x2_t __ret_425; \
	47409	float32x2_t __s0_425 = __p0_425; \
	47410	float32x2_t __s1_425 = __p1_425; \
	47411	float32x4_t __s2_425 = __p2_425; \
	47412	__ret_425 = vfma_laneq_f32(__s0_425, -__s1_425, __s2_425, __p3_425); \
	47413	__ret_425; \
	47414	})
46676	47415	#else
46677		__ai float16x4_t vtrn1_f16(float16x4_t __p0, float16x4_t __p1) {
46678		float16x4_t __ret;
46679		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46680		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46681		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
46682		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46683		return __ret;
46684		}
	47416	#define vfms_laneq_f32(__p0_426, __p1_426, __p2_426, __p3_426) __extension__ ({ \
	47417	float32x2_t __ret_426; \
	47418	float32x2_t __s0_426 = __p0_426; \
	47419	float32x2_t __s1_426 = __p1_426; \
	47420	float32x4_t __s2_426 = __p2_426; \
	47421	float32x2_t __rev0_426; __rev0_426 = __builtin_shufflevector(__s0_426, __s0_426, 1, 0); \
	47422	float32x2_t __rev1_426; __rev1_426 = __builtin_shufflevector(__s1_426, __s1_426, 1, 0); \
	47423	float32x4_t __rev2_426; __rev2_426 = __builtin_shufflevector(__s2_426, __s2_426, 3, 2, 1, 0); \
	47424	__ret_426 = __noswap_vfma_laneq_f32(__rev0_426, -__rev1_426, __rev2_426, __p3_426); \
	47425	__ret_426 = __builtin_shufflevector(__ret_426, __ret_426, 1, 0); \
	47426	__ret_426; \
	47427	})
46685	47428	#endif
46686	47429
46687	47430	#ifdef __LITTLE_ENDIAN__
46688		__ai float16x8_t vtrn2q_f16(float16x8_t __p0, float16x8_t __p1) {
46689		float16x8_t __ret;
46690		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
	47431	__ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
	47432	float64x2_t __ret;
	47433	__ret = vfmaq_f64(__p0, -__p1, (float64x2_t) {__p2, __p2});
46691	47434	return __ret;
46692	47435	}
46693	47436	#else
46694		__ai float16x8_t vtrn2q_f16(float16x8_t __p0, float16x8_t __p1) {
46695		float16x8_t __ret;
46696		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46697		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46698		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
46699		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	47437	__ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
	47438	float64x2_t __ret;
	47439	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47440	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	47441	__ret = __noswap_vfmaq_f64(__rev0, -__rev1, (float64x2_t) {__p2, __p2});
	47442	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46700	47443	return __ret;
46701	47444	}
46702	47445	#endif
46703	47446
46704	47447	#ifdef __LITTLE_ENDIAN__
46705		__ai float16x4_t vtrn2_f16(float16x4_t __p0, float16x4_t __p1) {
46706		float16x4_t __ret;
46707		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
	47448	__ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
	47449	float32x4_t __ret;
	47450	__ret = vfmaq_f32(__p0, -__p1, (float32x4_t) {__p2, __p2, __p2, __p2});
46708	47451	return __ret;
46709	47452	}
46710	47453	#else
46711		__ai float16x4_t vtrn2_f16(float16x4_t __p0, float16x4_t __p1) {
46712		float16x4_t __ret;
46713		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46714		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46715		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	47454	__ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
	47455	float32x4_t __ret;
	47456	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	47457	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	47458	__ret = __noswap_vfmaq_f32(__rev0, -__rev1, (float32x4_t) {__p2, __p2, __p2, __p2});
46716	47459	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46717	47460	return __ret;
46718	47461	}
46719	47462	#endif
46720	47463
46721		#ifdef __LITTLE_ENDIAN__
46722		__ai float16x8_t vuzp1q_f16(float16x8_t __p0, float16x8_t __p1) {
46723		float16x8_t __ret;
46724		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
46725		return __ret;
46726		}
46727		#else
46728		__ai float16x8_t vuzp1q_f16(float16x8_t __p0, float16x8_t __p1) {
46729		float16x8_t __ret;
46730		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46731		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46732		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
46733		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	47464	__ai float64x1_t vfms_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
	47465	float64x1_t __ret;
	47466	__ret = vfma_f64(__p0, -__p1, (float64x1_t) {__p2});
46734	47467	return __ret;
46735	47468	}
46736		#endif
46737
46738	47469	#ifdef __LITTLE_ENDIAN__
46739		__ai float16x4_t vuzp1_f16(float16x4_t __p0, float16x4_t __p1) {
46740		float16x4_t __ret;
46741		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
	47470	__ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
	47471	float32x2_t __ret;
	47472	__ret = vfma_f32(__p0, -__p1, (float32x2_t) {__p2, __p2});
46742	47473	return __ret;
46743	47474	}
46744	47475	#else
46745		__ai float16x4_t vuzp1_f16(float16x4_t __p0, float16x4_t __p1) {
46746		float16x4_t __ret;
46747		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46748		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46749		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
46750		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	47476	__ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
	47477	float32x2_t __ret;
	47478	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47479	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	47480	__ret = __noswap_vfma_f32(__rev0, -__rev1, (float32x2_t) {__p2, __p2});
	47481	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46751	47482	return __ret;
46752	47483	}
46753	47484	#endif
46754	47485
46755	47486	#ifdef __LITTLE_ENDIAN__
46756		__ai float16x8_t vuzp2q_f16(float16x8_t __p0, float16x8_t __p1) {
46757		float16x8_t __ret;
46758		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
	47487	__ai poly64x1_t vget_high_p64(poly64x2_t __p0) {
	47488	poly64x1_t __ret;
	47489	__ret = __builtin_shufflevector(__p0, __p0, 1);
46759	47490	return __ret;
46760	47491	}
46761	47492	#else
46762		__ai float16x8_t vuzp2q_f16(float16x8_t __p0, float16x8_t __p1) {
46763		float16x8_t __ret;
46764		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46765		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46766		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
46767		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46768		return __ret;
46769		}
46770		#endif
46771
46772		#ifdef __LITTLE_ENDIAN__
46773		__ai float16x4_t vuzp2_f16(float16x4_t __p0, float16x4_t __p1) {
46774		float16x4_t __ret;
46775		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	47493	__ai poly64x1_t vget_high_p64(poly64x2_t __p0) {
	47494	poly64x1_t __ret;
	47495	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47496	__ret = __builtin_shufflevector(__rev0, __rev0, 1);
46776	47497	return __ret;
46777	47498	}
46778		#else
46779		__ai float16x4_t vuzp2_f16(float16x4_t __p0, float16x4_t __p1) {
46780		float16x4_t __ret;
46781		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46782		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46783		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
46784		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	47499	__ai poly64x1_t __noswap_vget_high_p64(poly64x2_t __p0) {
	47500	poly64x1_t __ret;
	47501	__ret = __builtin_shufflevector(__p0, __p0, 1);
46785	47502	return __ret;
46786	47503	}
46787	47504	#endif
46788	47505
46789	47506	#ifdef __LITTLE_ENDIAN__
46790		__ai float16x8_t vzip1q_f16(float16x8_t __p0, float16x8_t __p1) {
46791		float16x8_t __ret;
46792		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
	47507	__ai float64x1_t vget_high_f64(float64x2_t __p0) {
	47508	float64x1_t __ret;
	47509	__ret = __builtin_shufflevector(__p0, __p0, 1);
46793	47510	return __ret;
46794	47511	}
46795	47512	#else
46796		__ai float16x8_t vzip1q_f16(float16x8_t __p0, float16x8_t __p1) {
46797		float16x8_t __ret;
46798		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46799		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46800		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
46801		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	47513	__ai float64x1_t vget_high_f64(float64x2_t __p0) {
	47514	float64x1_t __ret;
	47515	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47516	__ret = __builtin_shufflevector(__rev0, __rev0, 1);
46802	47517	return __ret;
46803	47518	}
46804	47519	#endif
46805	47520
	47521	#define vget_lane_p64(__p0, __p1) __extension__ ({ \
	47522	poly64_t __ret; \
	47523	poly64x1_t __s0 = __p0; \
	47524	__ret = (poly64_t) __builtin_neon_vget_lane_i64((poly64x1_t)__s0, __p1); \
	47525	__ret; \
	47526	})
46806	47527	#ifdef __LITTLE_ENDIAN__
46807		__ai float16x4_t vzip1_f16(float16x4_t __p0, float16x4_t __p1) {
46808		float16x4_t __ret;
46809		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
46810		return __ret;
46811		}
	47528	#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
	47529	poly64_t __ret; \
	47530	poly64x2_t __s0 = __p0; \
	47531	__ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
	47532	__ret; \
	47533	})
46812	47534	#else
46813		__ai float16x4_t vzip1_f16(float16x4_t __p0, float16x4_t __p1) {
46814		float16x4_t __ret;
46815		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46816		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46817		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
46818		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46819		return __ret;
46820		}
	47535	#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
	47536	poly64_t __ret; \
	47537	poly64x2_t __s0 = __p0; \
	47538	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	47539	__ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__rev0, __p1); \
	47540	__ret; \
	47541	})
	47542	#define __noswap_vgetq_lane_p64(__p0, __p1) __extension__ ({ \
	47543	poly64_t __ret; \
	47544	poly64x2_t __s0 = __p0; \
	47545	__ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
	47546	__ret; \
	47547	})
46821	47548	#endif
46822	47549
46823	47550	#ifdef __LITTLE_ENDIAN__
46824		__ai float16x8_t vzip2q_f16(float16x8_t __p0, float16x8_t __p1) {
46825		float16x8_t __ret;
46826		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
46827		return __ret;
46828		}
	47551	#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
	47552	float64_t __ret; \
	47553	float64x2_t __s0 = __p0; \
	47554	__ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
	47555	__ret; \
	47556	})
46829	47557	#else
46830		__ai float16x8_t vzip2q_f16(float16x8_t __p0, float16x8_t __p1) {
46831		float16x8_t __ret;
46832		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46833		float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46834		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
46835		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46836		return __ret;
46837		}
	47558	#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
	47559	float64_t __ret; \
	47560	float64x2_t __s0 = __p0; \
	47561	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	47562	__ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__rev0, __p1); \
	47563	__ret; \
	47564	})
	47565	#define __noswap_vgetq_lane_f64(__p0, __p1) __extension__ ({ \
	47566	float64_t __ret; \
	47567	float64x2_t __s0 = __p0; \
	47568	__ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
	47569	__ret; \
	47570	})
46838	47571	#endif
46839	47572
	47573	#define vget_lane_f64(__p0, __p1) __extension__ ({ \
	47574	float64_t __ret; \
	47575	float64x1_t __s0 = __p0; \
	47576	__ret = (float64_t) __builtin_neon_vget_lane_f64((float64x1_t)__s0, __p1); \
	47577	__ret; \
	47578	})
46840	47579	#ifdef __LITTLE_ENDIAN__
46841		__ai float16x4_t vzip2_f16(float16x4_t __p0, float16x4_t __p1) {
46842		float16x4_t __ret;
46843		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	47580	__ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
	47581	poly64x1_t __ret;
	47582	__ret = __builtin_shufflevector(__p0, __p0, 0);
46844	47583	return __ret;
46845	47584	}
46846	47585	#else
46847		__ai float16x4_t vzip2_f16(float16x4_t __p0, float16x4_t __p1) {
46848		float16x4_t __ret;
46849		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46850		float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46851		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
46852		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	47586	__ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
	47587	poly64x1_t __ret;
	47588	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47589	__ret = __builtin_shufflevector(__rev0, __rev0, 0);
46853	47590	return __ret;
46854	47591	}
46855	47592	#endif
46856	47593
46857		#endif
46858		#if defined(__ARM_FEATURE_MATMUL_INT8)
46859	47594	#ifdef __LITTLE_ENDIAN__
46860		__ai uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
46861		uint32x4_t __ret;
46862		__ret = (uint32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
	47595	__ai float64x1_t vget_low_f64(float64x2_t __p0) {
	47596	float64x1_t __ret;
	47597	__ret = __builtin_shufflevector(__p0, __p0, 0);
46863	47598	return __ret;
46864	47599	}
46865	47600	#else
46866		__ai uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
46867		uint32x4_t __ret;
46868		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46869		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46870		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46871		__ret = (uint32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
46872		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	47601	__ai float64x1_t vget_low_f64(float64x2_t __p0) {
	47602	float64x1_t __ret;
	47603	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	47604	__ret = __builtin_shufflevector(__rev0, __rev0, 0);
46873	47605	return __ret;
46874	47606	}
46875	47607	#endif
46876	47608
	47609	#define vld1_p64(__p0) __extension__ ({ \
	47610	poly64x1_t __ret; \
	47611	__ret = (poly64x1_t) __builtin_neon_vld1_v(__p0, 6); \
	47612	__ret; \
	47613	})
46877	47614	#ifdef __LITTLE_ENDIAN__
46878		__ai int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
46879		int32x4_t __ret;
46880		__ret = (int32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
46881		return __ret;
46882		}
	47615	#define vld1q_p64(__p0) __extension__ ({ \
	47616	poly64x2_t __ret; \
	47617	__ret = (poly64x2_t) __builtin_neon_vld1q_v(__p0, 38); \
	47618	__ret; \
	47619	})
46883	47620	#else
46884		__ai int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
46885		int32x4_t __ret;
46886		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46887		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46888		int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46889		__ret = (int32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
46890		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46891		return __ret;
46892		}
	47621	#define vld1q_p64(__p0) __extension__ ({ \
	47622	poly64x2_t __ret; \
	47623	__ret = (poly64x2_t) __builtin_neon_vld1q_v(__p0, 38); \
	47624	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47625	__ret; \
	47626	})
46893	47627	#endif
46894	47628
46895	47629	#ifdef __LITTLE_ENDIAN__
46896		__ai int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
46897		int32x4_t __ret;
46898		__ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
46899		return __ret;
46900		}
	47630	#define vld1q_f64(__p0) __extension__ ({ \
	47631	float64x2_t __ret; \
	47632	__ret = (float64x2_t) __builtin_neon_vld1q_v(__p0, 42); \
	47633	__ret; \
	47634	})
46901	47635	#else
46902		__ai int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
46903		int32x4_t __ret;
46904		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46905		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46906		int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46907		__ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
46908		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46909		return __ret;
46910		}
46911		__ai int32x4_t __noswap_vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
46912		int32x4_t __ret;
46913		__ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
46914		return __ret;
46915		}
	47636	#define vld1q_f64(__p0) __extension__ ({ \
	47637	float64x2_t __ret; \
	47638	__ret = (float64x2_t) __builtin_neon_vld1q_v(__p0, 42); \
	47639	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47640	__ret; \
	47641	})
46916	47642	#endif
46917	47643
	47644	#define vld1_f64(__p0) __extension__ ({ \
	47645	float64x1_t __ret; \
	47646	__ret = (float64x1_t) __builtin_neon_vld1_v(__p0, 10); \
	47647	__ret; \
	47648	})
	47649	#define vld1_dup_p64(__p0) __extension__ ({ \
	47650	poly64x1_t __ret; \
	47651	__ret = (poly64x1_t) __builtin_neon_vld1_dup_v(__p0, 6); \
	47652	__ret; \
	47653	})
46918	47654	#ifdef __LITTLE_ENDIAN__
46919		__ai int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
46920		int32x2_t __ret;
46921		__ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
46922		return __ret;
46923		}
	47655	#define vld1q_dup_p64(__p0) __extension__ ({ \
	47656	poly64x2_t __ret; \
	47657	__ret = (poly64x2_t) __builtin_neon_vld1q_dup_v(__p0, 38); \
	47658	__ret; \
	47659	})
46924	47660	#else
46925		__ai int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
46926		int32x2_t __ret;
46927		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
46928		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46929		int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
46930		__ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
46931		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46932		return __ret;
46933		}
46934		__ai int32x2_t __noswap_vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
46935		int32x2_t __ret;
46936		__ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
46937		return __ret;
46938		}
	47661	#define vld1q_dup_p64(__p0) __extension__ ({ \
	47662	poly64x2_t __ret; \
	47663	__ret = (poly64x2_t) __builtin_neon_vld1q_dup_v(__p0, 38); \
	47664	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47665	__ret; \
	47666	})
46939	47667	#endif
46940	47668
46941	47669	#ifdef __LITTLE_ENDIAN__
46942		#define vusdotq_lane_s32(__p0_305, __p1_305, __p2_305, __p3_305) __extension__ ({ \
46943		int32x4_t __ret_305; \
46944		int32x4_t __s0_305 = __p0_305; \
46945		uint8x16_t __s1_305 = __p1_305; \
46946		int8x8_t __s2_305 = __p2_305; \
46947		int8x8_t __reint_305 = __s2_305; \
46948		__ret_305 = vusdotq_s32(__s0_305, __s1_305, (int8x16_t)(splatq_lane_s32((int32x2_t ) &__reint_305, __p3_305))); \
46949		__ret_305; \
	47670	#define vld1q_dup_f64(__p0) __extension__ ({ \
	47671	float64x2_t __ret; \
	47672	__ret = (float64x2_t) __builtin_neon_vld1q_dup_v(__p0, 42); \
	47673	__ret; \
46950	47674	})
46951	47675	#else
46952		#define vusdotq_lane_s32(__p0_306, __p1_306, __p2_306, __p3_306) __extension__ ({ \
46953		int32x4_t __ret_306; \
46954		int32x4_t __s0_306 = __p0_306; \
46955		uint8x16_t __s1_306 = __p1_306; \
46956		int8x8_t __s2_306 = __p2_306; \
46957		int32x4_t __rev0_306; __rev0_306 = __builtin_shufflevector(__s0_306, __s0_306, 3, 2, 1, 0); \
46958		uint8x16_t __rev1_306; __rev1_306 = __builtin_shufflevector(__s1_306, __s1_306, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
46959		int8x8_t __rev2_306; __rev2_306 = __builtin_shufflevector(__s2_306, __s2_306, 7, 6, 5, 4, 3, 2, 1, 0); \
46960		int8x8_t __reint_306 = __rev2_306; \
46961		__ret_306 = __noswap_vusdotq_s32(__rev0_306, __rev1_306, (int8x16_t)(__noswap_splatq_lane_s32((int32x2_t ) &__reint_306, __p3_306))); \
46962		__ret_306 = __builtin_shufflevector(__ret_306, __ret_306, 3, 2, 1, 0); \
46963		__ret_306; \
	47676	#define vld1q_dup_f64(__p0) __extension__ ({ \
	47677	float64x2_t __ret; \
	47678	__ret = (float64x2_t) __builtin_neon_vld1q_dup_v(__p0, 42); \
	47679	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47680	__ret; \
46964	47681	})
46965	47682	#endif
46966	47683
	47684	#define vld1_dup_f64(__p0) __extension__ ({ \
	47685	float64x1_t __ret; \
	47686	__ret = (float64x1_t) __builtin_neon_vld1_dup_v(__p0, 10); \
	47687	__ret; \
	47688	})
	47689	#define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	47690	poly64x1_t __ret; \
	47691	poly64x1_t __s1 = __p1; \
	47692	__ret = (poly64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
	47693	__ret; \
	47694	})
46967	47695	#ifdef __LITTLE_ENDIAN__
46968		#define vusdot_lane_s32(__p0_307, __p1_307, __p2_307, __p3_307) __extension__ ({ \
46969		int32x2_t __ret_307; \
46970		int32x2_t __s0_307 = __p0_307; \
46971		uint8x8_t __s1_307 = __p1_307; \
46972		int8x8_t __s2_307 = __p2_307; \
46973		int8x8_t __reint_307 = __s2_307; \
46974		__ret_307 = vusdot_s32(__s0_307, __s1_307, (int8x8_t)(splat_lane_s32((int32x2_t ) &__reint_307, __p3_307))); \
46975		__ret_307; \
	47696	#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	47697	poly64x2_t __ret; \
	47698	poly64x2_t __s1 = __p1; \
	47699	__ret = (poly64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 38); \
	47700	__ret; \
46976	47701	})
46977	47702	#else
46978		#define vusdot_lane_s32(__p0_308, __p1_308, __p2_308, __p3_308) __extension__ ({ \
46979		int32x2_t __ret_308; \
46980		int32x2_t __s0_308 = __p0_308; \
46981		uint8x8_t __s1_308 = __p1_308; \
46982		int8x8_t __s2_308 = __p2_308; \
46983		int32x2_t __rev0_308; __rev0_308 = __builtin_shufflevector(__s0_308, __s0_308, 1, 0); \
46984		uint8x8_t __rev1_308; __rev1_308 = __builtin_shufflevector(__s1_308, __s1_308, 7, 6, 5, 4, 3, 2, 1, 0); \
46985		int8x8_t __rev2_308; __rev2_308 = __builtin_shufflevector(__s2_308, __s2_308, 7, 6, 5, 4, 3, 2, 1, 0); \
46986		int8x8_t __reint_308 = __rev2_308; \
46987		__ret_308 = __noswap_vusdot_s32(__rev0_308, __rev1_308, (int8x8_t)(__noswap_splat_lane_s32((int32x2_t ) &__reint_308, __p3_308))); \
46988		__ret_308 = __builtin_shufflevector(__ret_308, __ret_308, 1, 0); \
46989		__ret_308; \
	47703	#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	47704	poly64x2_t __ret; \
	47705	poly64x2_t __s1 = __p1; \
	47706	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	47707	__ret = (poly64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 38); \
	47708	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47709	__ret; \
46990	47710	})
46991	47711	#endif
46992	47712
46993	47713	#ifdef __LITTLE_ENDIAN__
46994		__ai int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
46995		int32x4_t __ret;
46996		__ret = (int32x4_t) __builtin_neon_vusmmlaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
46997		return __ret;
46998		}
	47714	#define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	47715	float64x2_t __ret; \
	47716	float64x2_t __s1 = __p1; \
	47717	__ret = (float64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 42); \
	47718	__ret; \
	47719	})
46999	47720	#else
47000		__ai int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
47001		int32x4_t __ret;
47002		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47003		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47004		int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47005		__ret = (int32x4_t) __builtin_neon_vusmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
47006		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47007		return __ret;
47008		}
	47721	#define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	47722	float64x2_t __ret; \
	47723	float64x2_t __s1 = __p1; \
	47724	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	47725	__ret = (float64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 42); \
	47726	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	47727	__ret; \
	47728	})
47009	47729	#endif
47010	47730
47011		#endif
47012		#if defined(__ARM_FEATURE_QRDMX)
	47731	#define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	47732	float64x1_t __ret; \
	47733	float64x1_t __s1 = __p1; \
	47734	__ret = (float64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
	47735	__ret; \
	47736	})
	47737	#define vld1_p64_x2(__p0) __extension__ ({ \
	47738	poly64x1x2_t __ret; \
	47739	__builtin_neon_vld1_x2_v(&__ret, __p0, 6); \
	47740	__ret; \
	47741	})
47013	47742	#ifdef __LITTLE_ENDIAN__
47014		__ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47015		int32x4_t __ret;
47016		__ret = (int32x4_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
47017		return __ret;
47018		}
	47743	#define vld1q_p64_x2(__p0) __extension__ ({ \
	47744	poly64x2x2_t __ret; \
	47745	__builtin_neon_vld1q_x2_v(&__ret, __p0, 38); \
	47746	__ret; \
	47747	})
47019	47748	#else
47020		__ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47021		int32x4_t __ret;
47022		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47023		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47024		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47025		__ret = (int32x4_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
47026		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47027		return __ret;
47028		}
47029		__ai int32x4_t __noswap_vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47030		int32x4_t __ret;
47031		__ret = (int32x4_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
47032		return __ret;
47033		}
	47749	#define vld1q_p64_x2(__p0) __extension__ ({ \
	47750	poly64x2x2_t __ret; \
	47751	__builtin_neon_vld1q_x2_v(&__ret, __p0, 38); \
	47752	\
	47753	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47754	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47755	__ret; \
	47756	})
47034	47757	#endif
47035	47758
47036	47759	#ifdef __LITTLE_ENDIAN__
47037		__ai int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47038		int16x8_t __ret;
47039		__ret = (int16x8_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
47040		return __ret;
47041		}
	47760	#define vld1q_f64_x2(__p0) __extension__ ({ \
	47761	float64x2x2_t __ret; \
	47762	__builtin_neon_vld1q_x2_v(&__ret, __p0, 42); \
	47763	__ret; \
	47764	})
47042	47765	#else
47043		__ai int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47044		int16x8_t __ret;
47045		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47046		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
47047		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47048		__ret = (int16x8_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
47049		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
47050		return __ret;
47051		}
47052		__ai int16x8_t __noswap_vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47053		int16x8_t __ret;
47054		__ret = (int16x8_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
47055		return __ret;
47056		}
	47766	#define vld1q_f64_x2(__p0) __extension__ ({ \
	47767	float64x2x2_t __ret; \
	47768	__builtin_neon_vld1q_x2_v(&__ret, __p0, 42); \
	47769	\
	47770	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47771	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47772	__ret; \
	47773	})
47057	47774	#endif
47058	47775
	47776	#define vld1_f64_x2(__p0) __extension__ ({ \
	47777	float64x1x2_t __ret; \
	47778	__builtin_neon_vld1_x2_v(&__ret, __p0, 10); \
	47779	__ret; \
	47780	})
	47781	#define vld1_p64_x3(__p0) __extension__ ({ \
	47782	poly64x1x3_t __ret; \
	47783	__builtin_neon_vld1_x3_v(&__ret, __p0, 6); \
	47784	__ret; \
	47785	})
47059	47786	#ifdef __LITTLE_ENDIAN__
47060		__ai int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47061		int32x2_t __ret;
47062		__ret = (int32x2_t) __builtin_neon_vqrdmlah_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
47063		return __ret;
47064		}
	47787	#define vld1q_p64_x3(__p0) __extension__ ({ \
	47788	poly64x2x3_t __ret; \
	47789	__builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
	47790	__ret; \
	47791	})
47065	47792	#else
47066		__ai int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47067		int32x2_t __ret;
47068		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47069		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47070		int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47071		__ret = (int32x2_t) __builtin_neon_vqrdmlah_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
47072		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47073		return __ret;
47074		}
47075		__ai int32x2_t __noswap_vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47076		int32x2_t __ret;
47077		__ret = (int32x2_t) __builtin_neon_vqrdmlah_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
47078		return __ret;
47079		}
	47793	#define vld1q_p64_x3(__p0) __extension__ ({ \
	47794	poly64x2x3_t __ret; \
	47795	__builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
	47796	\
	47797	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47798	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47799	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	47800	__ret; \
	47801	})
47080	47802	#endif
47081	47803
47082	47804	#ifdef __LITTLE_ENDIAN__
47083		__ai int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47084		int16x4_t __ret;
47085		__ret = (int16x4_t) __builtin_neon_vqrdmlah_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
47086		return __ret;
47087		}
	47805	#define vld1q_f64_x3(__p0) __extension__ ({ \
	47806	float64x2x3_t __ret; \
	47807	__builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
	47808	__ret; \
	47809	})
47088	47810	#else
47089		__ai int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47090		int16x4_t __ret;
47091		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47092		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47093		int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47094		__ret = (int16x4_t) __builtin_neon_vqrdmlah_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
47095		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47096		return __ret;
47097		}
47098		__ai int16x4_t __noswap_vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47099		int16x4_t __ret;
47100		__ret = (int16x4_t) __builtin_neon_vqrdmlah_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
47101		return __ret;
47102		}
	47811	#define vld1q_f64_x3(__p0) __extension__ ({ \
	47812	float64x2x3_t __ret; \
	47813	__builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
	47814	\
	47815	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47816	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47817	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	47818	__ret; \
	47819	})
47103	47820	#endif
47104	47821
	47822	#define vld1_f64_x3(__p0) __extension__ ({ \
	47823	float64x1x3_t __ret; \
	47824	__builtin_neon_vld1_x3_v(&__ret, __p0, 10); \
	47825	__ret; \
	47826	})
	47827	#define vld1_p64_x4(__p0) __extension__ ({ \
	47828	poly64x1x4_t __ret; \
	47829	__builtin_neon_vld1_x4_v(&__ret, __p0, 6); \
	47830	__ret; \
	47831	})
47105	47832	#ifdef __LITTLE_ENDIAN__
47106		#define vqrdmlahq_lane_s32(__p0_309, __p1_309, __p2_309, __p3_309) __extension__ ({ \
47107		int32x4_t __ret_309; \
47108		int32x4_t __s0_309 = __p0_309; \
47109		int32x4_t __s1_309 = __p1_309; \
47110		int32x2_t __s2_309 = __p2_309; \
47111		__ret_309 = vqrdmlahq_s32(__s0_309, __s1_309, splatq_lane_s32(__s2_309, __p3_309)); \
47112		__ret_309; \
	47833	#define vld1q_p64_x4(__p0) __extension__ ({ \
	47834	poly64x2x4_t __ret; \
	47835	__builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
	47836	__ret; \
47113	47837	})
47114	47838	#else
47115		#define vqrdmlahq_lane_s32(__p0_310, __p1_310, __p2_310, __p3_310) __extension__ ({ \
47116		int32x4_t __ret_310; \
47117		int32x4_t __s0_310 = __p0_310; \
47118		int32x4_t __s1_310 = __p1_310; \
47119		int32x2_t __s2_310 = __p2_310; \
47120		int32x4_t __rev0_310; __rev0_310 = __builtin_shufflevector(__s0_310, __s0_310, 3, 2, 1, 0); \
47121		int32x4_t __rev1_310; __rev1_310 = __builtin_shufflevector(__s1_310, __s1_310, 3, 2, 1, 0); \
47122		int32x2_t __rev2_310; __rev2_310 = __builtin_shufflevector(__s2_310, __s2_310, 1, 0); \
47123		__ret_310 = __noswap_vqrdmlahq_s32(__rev0_310, __rev1_310, __noswap_splatq_lane_s32(__rev2_310, __p3_310)); \
47124		__ret_310 = __builtin_shufflevector(__ret_310, __ret_310, 3, 2, 1, 0); \
47125		__ret_310; \
	47839	#define vld1q_p64_x4(__p0) __extension__ ({ \
	47840	poly64x2x4_t __ret; \
	47841	__builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
	47842	\
	47843	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47844	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47845	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	47846	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	47847	__ret; \
47126	47848	})
47127	47849	#endif
47128	47850
47129	47851	#ifdef __LITTLE_ENDIAN__
47130		#define vqrdmlahq_lane_s16(__p0_311, __p1_311, __p2_311, __p3_311) __extension__ ({ \
47131		int16x8_t __ret_311; \
47132		int16x8_t __s0_311 = __p0_311; \
47133		int16x8_t __s1_311 = __p1_311; \
47134		int16x4_t __s2_311 = __p2_311; \
47135		__ret_311 = vqrdmlahq_s16(__s0_311, __s1_311, splatq_lane_s16(__s2_311, __p3_311)); \
47136		__ret_311; \
	47852	#define vld1q_f64_x4(__p0) __extension__ ({ \
	47853	float64x2x4_t __ret; \
	47854	__builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
	47855	__ret; \
47137	47856	})
47138	47857	#else
47139		#define vqrdmlahq_lane_s16(__p0_312, __p1_312, __p2_312, __p3_312) __extension__ ({ \
47140		int16x8_t __ret_312; \
47141		int16x8_t __s0_312 = __p0_312; \
47142		int16x8_t __s1_312 = __p1_312; \
47143		int16x4_t __s2_312 = __p2_312; \
47144		int16x8_t __rev0_312; __rev0_312 = __builtin_shufflevector(__s0_312, __s0_312, 7, 6, 5, 4, 3, 2, 1, 0); \
47145		int16x8_t __rev1_312; __rev1_312 = __builtin_shufflevector(__s1_312, __s1_312, 7, 6, 5, 4, 3, 2, 1, 0); \
47146		int16x4_t __rev2_312; __rev2_312 = __builtin_shufflevector(__s2_312, __s2_312, 3, 2, 1, 0); \
47147		__ret_312 = __noswap_vqrdmlahq_s16(__rev0_312, __rev1_312, __noswap_splatq_lane_s16(__rev2_312, __p3_312)); \
47148		__ret_312 = __builtin_shufflevector(__ret_312, __ret_312, 7, 6, 5, 4, 3, 2, 1, 0); \
47149		__ret_312; \
	47858	#define vld1q_f64_x4(__p0) __extension__ ({ \
	47859	float64x2x4_t __ret; \
	47860	__builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
	47861	\
	47862	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47863	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47864	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	47865	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	47866	__ret; \
47150	47867	})
47151	47868	#endif
47152	47869
	47870	#define vld1_f64_x4(__p0) __extension__ ({ \
	47871	float64x1x4_t __ret; \
	47872	__builtin_neon_vld1_x4_v(&__ret, __p0, 10); \
	47873	__ret; \
	47874	})
	47875	#define vld2_p64(__p0) __extension__ ({ \
	47876	poly64x1x2_t __ret; \
	47877	__builtin_neon_vld2_v(&__ret, __p0, 6); \
	47878	__ret; \
	47879	})
47153	47880	#ifdef __LITTLE_ENDIAN__
47154		#define vqrdmlah_lane_s32(__p0_313, __p1_313, __p2_313, __p3_313) __extension__ ({ \
47155		int32x2_t __ret_313; \
47156		int32x2_t __s0_313 = __p0_313; \
47157		int32x2_t __s1_313 = __p1_313; \
47158		int32x2_t __s2_313 = __p2_313; \
47159		__ret_313 = vqrdmlah_s32(__s0_313, __s1_313, splat_lane_s32(__s2_313, __p3_313)); \
47160		__ret_313; \
	47881	#define vld2q_p64(__p0) __extension__ ({ \
	47882	poly64x2x2_t __ret; \
	47883	__builtin_neon_vld2q_v(&__ret, __p0, 38); \
	47884	__ret; \
47161	47885	})
47162	47886	#else
47163		#define vqrdmlah_lane_s32(__p0_314, __p1_314, __p2_314, __p3_314) __extension__ ({ \
47164		int32x2_t __ret_314; \
47165		int32x2_t __s0_314 = __p0_314; \
47166		int32x2_t __s1_314 = __p1_314; \
47167		int32x2_t __s2_314 = __p2_314; \
47168		int32x2_t __rev0_314; __rev0_314 = __builtin_shufflevector(__s0_314, __s0_314, 1, 0); \
47169		int32x2_t __rev1_314; __rev1_314 = __builtin_shufflevector(__s1_314, __s1_314, 1, 0); \
47170		int32x2_t __rev2_314; __rev2_314 = __builtin_shufflevector(__s2_314, __s2_314, 1, 0); \
47171		__ret_314 = __noswap_vqrdmlah_s32(__rev0_314, __rev1_314, __noswap_splat_lane_s32(__rev2_314, __p3_314)); \
47172		__ret_314 = __builtin_shufflevector(__ret_314, __ret_314, 1, 0); \
47173		__ret_314; \
	47887	#define vld2q_p64(__p0) __extension__ ({ \
	47888	poly64x2x2_t __ret; \
	47889	__builtin_neon_vld2q_v(&__ret, __p0, 38); \
	47890	\
	47891	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47892	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47893	__ret; \
47174	47894	})
47175	47895	#endif
47176	47896
47177	47897	#ifdef __LITTLE_ENDIAN__
47178		#define vqrdmlah_lane_s16(__p0_315, __p1_315, __p2_315, __p3_315) __extension__ ({ \
47179		int16x4_t __ret_315; \
47180		int16x4_t __s0_315 = __p0_315; \
47181		int16x4_t __s1_315 = __p1_315; \
47182		int16x4_t __s2_315 = __p2_315; \
47183		__ret_315 = vqrdmlah_s16(__s0_315, __s1_315, splat_lane_s16(__s2_315, __p3_315)); \
47184		__ret_315; \
	47898	#define vld2q_u64(__p0) __extension__ ({ \
	47899	uint64x2x2_t __ret; \
	47900	__builtin_neon_vld2q_v(&__ret, __p0, 51); \
	47901	__ret; \
47185	47902	})
47186	47903	#else
47187		#define vqrdmlah_lane_s16(__p0_316, __p1_316, __p2_316, __p3_316) __extension__ ({ \
47188		int16x4_t __ret_316; \
47189		int16x4_t __s0_316 = __p0_316; \
47190		int16x4_t __s1_316 = __p1_316; \
47191		int16x4_t __s2_316 = __p2_316; \
47192		int16x4_t __rev0_316; __rev0_316 = __builtin_shufflevector(__s0_316, __s0_316, 3, 2, 1, 0); \
47193		int16x4_t __rev1_316; __rev1_316 = __builtin_shufflevector(__s1_316, __s1_316, 3, 2, 1, 0); \
47194		int16x4_t __rev2_316; __rev2_316 = __builtin_shufflevector(__s2_316, __s2_316, 3, 2, 1, 0); \
47195		__ret_316 = __noswap_vqrdmlah_s16(__rev0_316, __rev1_316, __noswap_splat_lane_s16(__rev2_316, __p3_316)); \
47196		__ret_316 = __builtin_shufflevector(__ret_316, __ret_316, 3, 2, 1, 0); \
47197		__ret_316; \
	47904	#define vld2q_u64(__p0) __extension__ ({ \
	47905	uint64x2x2_t __ret; \
	47906	__builtin_neon_vld2q_v(&__ret, __p0, 51); \
	47907	\
	47908	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47909	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47910	__ret; \
47198	47911	})
47199	47912	#endif
47200	47913
47201	47914	#ifdef __LITTLE_ENDIAN__
47202		__ai int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47203		int32x4_t __ret;
47204		__ret = (int32x4_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
47205		return __ret;
47206		}
	47915	#define vld2q_f64(__p0) __extension__ ({ \
	47916	float64x2x2_t __ret; \
	47917	__builtin_neon_vld2q_v(&__ret, __p0, 42); \
	47918	__ret; \
	47919	})
47207	47920	#else
47208		__ai int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47209		int32x4_t __ret;
47210		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47211		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47212		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47213		__ret = (int32x4_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
47214		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47215		return __ret;
47216		}
47217		__ai int32x4_t __noswap_vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47218		int32x4_t __ret;
47219		__ret = (int32x4_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
47220		return __ret;
47221		}
	47921	#define vld2q_f64(__p0) __extension__ ({ \
	47922	float64x2x2_t __ret; \
	47923	__builtin_neon_vld2q_v(&__ret, __p0, 42); \
	47924	\
	47925	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47926	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47927	__ret; \
	47928	})
47222	47929	#endif
47223	47930
47224	47931	#ifdef __LITTLE_ENDIAN__
47225		__ai int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47226		int16x8_t __ret;
47227		__ret = (int16x8_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
47228		return __ret;
47229		}
	47932	#define vld2q_s64(__p0) __extension__ ({ \
	47933	int64x2x2_t __ret; \
	47934	__builtin_neon_vld2q_v(&__ret, __p0, 35); \
	47935	__ret; \
	47936	})
47230	47937	#else
47231		__ai int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47232		int16x8_t __ret;
47233		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47234		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
47235		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47236		__ret = (int16x8_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
47237		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
47238		return __ret;
47239		}
47240		__ai int16x8_t __noswap_vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47241		int16x8_t __ret;
47242		__ret = (int16x8_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
47243		return __ret;
47244		}
	47938	#define vld2q_s64(__p0) __extension__ ({ \
	47939	int64x2x2_t __ret; \
	47940	__builtin_neon_vld2q_v(&__ret, __p0, 35); \
	47941	\
	47942	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47943	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47944	__ret; \
	47945	})
47245	47946	#endif
47246	47947
	47948	#define vld2_f64(__p0) __extension__ ({ \
	47949	float64x1x2_t __ret; \
	47950	__builtin_neon_vld2_v(&__ret, __p0, 10); \
	47951	__ret; \
	47952	})
	47953	#define vld2_dup_p64(__p0) __extension__ ({ \
	47954	poly64x1x2_t __ret; \
	47955	__builtin_neon_vld2_dup_v(&__ret, __p0, 6); \
	47956	__ret; \
	47957	})
47247	47958	#ifdef __LITTLE_ENDIAN__
47248		__ai int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47249		int32x2_t __ret;
47250		__ret = (int32x2_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
47251		return __ret;
47252		}
	47959	#define vld2q_dup_p64(__p0) __extension__ ({ \
	47960	poly64x2x2_t __ret; \
	47961	__builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
	47962	__ret; \
	47963	})
47253	47964	#else
47254		__ai int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47255		int32x2_t __ret;
47256		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47257		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47258		int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47259		__ret = (int32x2_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
47260		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47261		return __ret;
47262		}
47263		__ai int32x2_t __noswap_vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47264		int32x2_t __ret;
47265		__ret = (int32x2_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
47266		return __ret;
47267		}
	47965	#define vld2q_dup_p64(__p0) __extension__ ({ \
	47966	poly64x2x2_t __ret; \
	47967	__builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
	47968	\
	47969	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47970	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47971	__ret; \
	47972	})
47268	47973	#endif
47269	47974
47270	47975	#ifdef __LITTLE_ENDIAN__
47271		__ai int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47272		int16x4_t __ret;
47273		__ret = (int16x4_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
47274		return __ret;
47275		}
	47976	#define vld2q_dup_f64(__p0) __extension__ ({ \
	47977	float64x2x2_t __ret; \
	47978	__builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
	47979	__ret; \
	47980	})
47276	47981	#else
47277		__ai int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47278		int16x4_t __ret;
47279		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47280		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47281		int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47282		__ret = (int16x4_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
47283		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47284		return __ret;
47285		}
47286		__ai int16x4_t __noswap_vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47287		int16x4_t __ret;
47288		__ret = (int16x4_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
47289		return __ret;
47290		}
	47982	#define vld2q_dup_f64(__p0) __extension__ ({ \
	47983	float64x2x2_t __ret; \
	47984	__builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
	47985	\
	47986	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	47987	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	47988	__ret; \
	47989	})
47291	47990	#endif
47292	47991
	47992	#define vld2_dup_f64(__p0) __extension__ ({ \
	47993	float64x1x2_t __ret; \
	47994	__builtin_neon_vld2_dup_v(&__ret, __p0, 10); \
	47995	__ret; \
	47996	})
	47997	#define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	47998	poly64x1x2_t __ret; \
	47999	poly64x1x2_t __s1 = __p1; \
	48000	__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
	48001	__ret; \
	48002	})
47293	48003	#ifdef __LITTLE_ENDIAN__
47294		#define vqrdmlshq_lane_s32(__p0_317, __p1_317, __p2_317, __p3_317) __extension__ ({ \
47295		int32x4_t __ret_317; \
47296		int32x4_t __s0_317 = __p0_317; \
47297		int32x4_t __s1_317 = __p1_317; \
47298		int32x2_t __s2_317 = __p2_317; \
47299		__ret_317 = vqrdmlshq_s32(__s0_317, __s1_317, splatq_lane_s32(__s2_317, __p3_317)); \
47300		__ret_317; \
	48004	#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	48005	poly8x16x2_t __ret; \
	48006	poly8x16x2_t __s1 = __p1; \
	48007	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 36); \
	48008	__ret; \
47301	48009	})
47302	48010	#else
47303		#define vqrdmlshq_lane_s32(__p0_318, __p1_318, __p2_318, __p3_318) __extension__ ({ \
47304		int32x4_t __ret_318; \
47305		int32x4_t __s0_318 = __p0_318; \
47306		int32x4_t __s1_318 = __p1_318; \
47307		int32x2_t __s2_318 = __p2_318; \
47308		int32x4_t __rev0_318; __rev0_318 = __builtin_shufflevector(__s0_318, __s0_318, 3, 2, 1, 0); \
47309		int32x4_t __rev1_318; __rev1_318 = __builtin_shufflevector(__s1_318, __s1_318, 3, 2, 1, 0); \
47310		int32x2_t __rev2_318; __rev2_318 = __builtin_shufflevector(__s2_318, __s2_318, 1, 0); \
47311		__ret_318 = __noswap_vqrdmlshq_s32(__rev0_318, __rev1_318, __noswap_splatq_lane_s32(__rev2_318, __p3_318)); \
47312		__ret_318 = __builtin_shufflevector(__ret_318, __ret_318, 3, 2, 1, 0); \
47313		__ret_318; \
	48011	#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	48012	poly8x16x2_t __ret; \
	48013	poly8x16x2_t __s1 = __p1; \
	48014	poly8x16x2_t __rev1; \
	48015	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48016	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48017	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 36); \
	48018	\
	48019	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48020	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48021	__ret; \
47314	48022	})
47315	48023	#endif
47316	48024
47317	48025	#ifdef __LITTLE_ENDIAN__
47318		#define vqrdmlshq_lane_s16(__p0_319, __p1_319, __p2_319, __p3_319) __extension__ ({ \
47319		int16x8_t __ret_319; \
47320		int16x8_t __s0_319 = __p0_319; \
47321		int16x8_t __s1_319 = __p1_319; \
47322		int16x4_t __s2_319 = __p2_319; \
47323		__ret_319 = vqrdmlshq_s16(__s0_319, __s1_319, splatq_lane_s16(__s2_319, __p3_319)); \
47324		__ret_319; \
	48026	#define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48027	poly64x2x2_t __ret; \
	48028	poly64x2x2_t __s1 = __p1; \
	48029	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 38); \
	48030	__ret; \
47325	48031	})
47326	48032	#else
47327		#define vqrdmlshq_lane_s16(__p0_320, __p1_320, __p2_320, __p3_320) __extension__ ({ \
47328		int16x8_t __ret_320; \
47329		int16x8_t __s0_320 = __p0_320; \
47330		int16x8_t __s1_320 = __p1_320; \
47331		int16x4_t __s2_320 = __p2_320; \
47332		int16x8_t __rev0_320; __rev0_320 = __builtin_shufflevector(__s0_320, __s0_320, 7, 6, 5, 4, 3, 2, 1, 0); \
47333		int16x8_t __rev1_320; __rev1_320 = __builtin_shufflevector(__s1_320, __s1_320, 7, 6, 5, 4, 3, 2, 1, 0); \
47334		int16x4_t __rev2_320; __rev2_320 = __builtin_shufflevector(__s2_320, __s2_320, 3, 2, 1, 0); \
47335		__ret_320 = __noswap_vqrdmlshq_s16(__rev0_320, __rev1_320, __noswap_splatq_lane_s16(__rev2_320, __p3_320)); \
47336		__ret_320 = __builtin_shufflevector(__ret_320, __ret_320, 7, 6, 5, 4, 3, 2, 1, 0); \
47337		__ret_320; \
	48033	#define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48034	poly64x2x2_t __ret; \
	48035	poly64x2x2_t __s1 = __p1; \
	48036	poly64x2x2_t __rev1; \
	48037	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48038	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48039	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 38); \
	48040	\
	48041	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48042	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48043	__ret; \
47338	48044	})
47339	48045	#endif
47340	48046
47341	48047	#ifdef __LITTLE_ENDIAN__
47342		#define vqrdmlsh_lane_s32(__p0_321, __p1_321, __p2_321, __p3_321) __extension__ ({ \
47343		int32x2_t __ret_321; \
47344		int32x2_t __s0_321 = __p0_321; \
47345		int32x2_t __s1_321 = __p1_321; \
47346		int32x2_t __s2_321 = __p2_321; \
47347		__ret_321 = vqrdmlsh_s32(__s0_321, __s1_321, splat_lane_s32(__s2_321, __p3_321)); \
47348		__ret_321; \
	48048	#define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	48049	uint8x16x2_t __ret; \
	48050	uint8x16x2_t __s1 = __p1; \
	48051	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 48); \
	48052	__ret; \
47349	48053	})
47350	48054	#else
47351		#define vqrdmlsh_lane_s32(__p0_322, __p1_322, __p2_322, __p3_322) __extension__ ({ \
47352		int32x2_t __ret_322; \
47353		int32x2_t __s0_322 = __p0_322; \
47354		int32x2_t __s1_322 = __p1_322; \
47355		int32x2_t __s2_322 = __p2_322; \
47356		int32x2_t __rev0_322; __rev0_322 = __builtin_shufflevector(__s0_322, __s0_322, 1, 0); \
47357		int32x2_t __rev1_322; __rev1_322 = __builtin_shufflevector(__s1_322, __s1_322, 1, 0); \
47358		int32x2_t __rev2_322; __rev2_322 = __builtin_shufflevector(__s2_322, __s2_322, 1, 0); \
47359		__ret_322 = __noswap_vqrdmlsh_s32(__rev0_322, __rev1_322, __noswap_splat_lane_s32(__rev2_322, __p3_322)); \
47360		__ret_322 = __builtin_shufflevector(__ret_322, __ret_322, 1, 0); \
47361		__ret_322; \
	48055	#define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	48056	uint8x16x2_t __ret; \
	48057	uint8x16x2_t __s1 = __p1; \
	48058	uint8x16x2_t __rev1; \
	48059	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48060	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48061	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 48); \
	48062	\
	48063	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48064	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48065	__ret; \
47362	48066	})
47363	48067	#endif
47364	48068
47365	48069	#ifdef __LITTLE_ENDIAN__
47366		#define vqrdmlsh_lane_s16(__p0_323, __p1_323, __p2_323, __p3_323) __extension__ ({ \
47367		int16x4_t __ret_323; \
47368		int16x4_t __s0_323 = __p0_323; \
47369		int16x4_t __s1_323 = __p1_323; \
47370		int16x4_t __s2_323 = __p2_323; \
47371		__ret_323 = vqrdmlsh_s16(__s0_323, __s1_323, splat_lane_s16(__s2_323, __p3_323)); \
47372		__ret_323; \
	48070	#define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48071	uint64x2x2_t __ret; \
	48072	uint64x2x2_t __s1 = __p1; \
	48073	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 51); \
	48074	__ret; \
47373	48075	})
47374	48076	#else
47375		#define vqrdmlsh_lane_s16(__p0_324, __p1_324, __p2_324, __p3_324) __extension__ ({ \
47376		int16x4_t __ret_324; \
47377		int16x4_t __s0_324 = __p0_324; \
47378		int16x4_t __s1_324 = __p1_324; \
47379		int16x4_t __s2_324 = __p2_324; \
47380		int16x4_t __rev0_324; __rev0_324 = __builtin_shufflevector(__s0_324, __s0_324, 3, 2, 1, 0); \
47381		int16x4_t __rev1_324; __rev1_324 = __builtin_shufflevector(__s1_324, __s1_324, 3, 2, 1, 0); \
47382		int16x4_t __rev2_324; __rev2_324 = __builtin_shufflevector(__s2_324, __s2_324, 3, 2, 1, 0); \
47383		__ret_324 = __noswap_vqrdmlsh_s16(__rev0_324, __rev1_324, __noswap_splat_lane_s16(__rev2_324, __p3_324)); \
47384		__ret_324 = __builtin_shufflevector(__ret_324, __ret_324, 3, 2, 1, 0); \
47385		__ret_324; \
	48077	#define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48078	uint64x2x2_t __ret; \
	48079	uint64x2x2_t __s1 = __p1; \
	48080	uint64x2x2_t __rev1; \
	48081	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48082	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48083	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 51); \
	48084	\
	48085	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48086	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48087	__ret; \
47386	48088	})
47387	48089	#endif
47388	48090
47389		#endif
47390		#if defined(__ARM_FEATURE_QRDMX) && defined(__aarch64__)
47391		__ai int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
47392		int32_t __ret;
47393		__ret = (int32_t) __builtin_neon_vqrdmlahs_s32(__p0, __p1, __p2);
47394		return __ret;
47395		}
47396		__ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
47397		int16_t __ret;
47398		__ret = (int16_t) __builtin_neon_vqrdmlahh_s16(__p0, __p1, __p2);
47399		return __ret;
47400		}
47401	48091	#ifdef __LITTLE_ENDIAN__
47402		#define vqrdmlahs_lane_s32(__p0_325, __p1_325, __p2_325, __p3_325) __extension__ ({ \
47403		int32_t __ret_325; \
47404		int32_t __s0_325 = __p0_325; \
47405		int32_t __s1_325 = __p1_325; \
47406		int32x2_t __s2_325 = __p2_325; \
47407		__ret_325 = vqrdmlahs_s32(__s0_325, __s1_325, vget_lane_s32(__s2_325, __p3_325)); \
47408		__ret_325; \
	48092	#define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	48093	int8x16x2_t __ret; \
	48094	int8x16x2_t __s1 = __p1; \
	48095	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 32); \
	48096	__ret; \
47409	48097	})
47410	48098	#else
47411		#define vqrdmlahs_lane_s32(__p0_326, __p1_326, __p2_326, __p3_326) __extension__ ({ \
47412		int32_t __ret_326; \
47413		int32_t __s0_326 = __p0_326; \
47414		int32_t __s1_326 = __p1_326; \
47415		int32x2_t __s2_326 = __p2_326; \
47416		int32x2_t __rev2_326; __rev2_326 = __builtin_shufflevector(__s2_326, __s2_326, 1, 0); \
47417		__ret_326 = vqrdmlahs_s32(__s0_326, __s1_326, __noswap_vget_lane_s32(__rev2_326, __p3_326)); \
47418		__ret_326; \
	48099	#define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	48100	int8x16x2_t __ret; \
	48101	int8x16x2_t __s1 = __p1; \
	48102	int8x16x2_t __rev1; \
	48103	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48104	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48105	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 32); \
	48106	\
	48107	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48108	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48109	__ret; \
47419	48110	})
47420	48111	#endif
47421	48112
47422	48113	#ifdef __LITTLE_ENDIAN__
47423		#define vqrdmlahh_lane_s16(__p0_327, __p1_327, __p2_327, __p3_327) __extension__ ({ \
47424		int16_t __ret_327; \
47425		int16_t __s0_327 = __p0_327; \
47426		int16_t __s1_327 = __p1_327; \
47427		int16x4_t __s2_327 = __p2_327; \
47428		__ret_327 = vqrdmlahh_s16(__s0_327, __s1_327, vget_lane_s16(__s2_327, __p3_327)); \
47429		__ret_327; \
	48114	#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48115	float64x2x2_t __ret; \
	48116	float64x2x2_t __s1 = __p1; \
	48117	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
	48118	__ret; \
47430	48119	})
47431	48120	#else
47432		#define vqrdmlahh_lane_s16(__p0_328, __p1_328, __p2_328, __p3_328) __extension__ ({ \
47433		int16_t __ret_328; \
47434		int16_t __s0_328 = __p0_328; \
47435		int16_t __s1_328 = __p1_328; \
47436		int16x4_t __s2_328 = __p2_328; \
47437		int16x4_t __rev2_328; __rev2_328 = __builtin_shufflevector(__s2_328, __s2_328, 3, 2, 1, 0); \
47438		__ret_328 = vqrdmlahh_s16(__s0_328, __s1_328, __noswap_vget_lane_s16(__rev2_328, __p3_328)); \
47439		__ret_328; \
	48121	#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48122	float64x2x2_t __ret; \
	48123	float64x2x2_t __s1 = __p1; \
	48124	float64x2x2_t __rev1; \
	48125	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48126	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48127	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
	48128	\
	48129	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48130	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48131	__ret; \
47440	48132	})
47441	48133	#endif
47442	48134
47443	48135	#ifdef __LITTLE_ENDIAN__
47444		#define vqrdmlahs_laneq_s32(__p0_329, __p1_329, __p2_329, __p3_329) __extension__ ({ \
47445		int32_t __ret_329; \
47446		int32_t __s0_329 = __p0_329; \
47447		int32_t __s1_329 = __p1_329; \
47448		int32x4_t __s2_329 = __p2_329; \
47449		__ret_329 = vqrdmlahs_s32(__s0_329, __s1_329, vgetq_lane_s32(__s2_329, __p3_329)); \
47450		__ret_329; \
	48136	#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48137	int64x2x2_t __ret; \
	48138	int64x2x2_t __s1 = __p1; \
	48139	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
	48140	__ret; \
47451	48141	})
47452	48142	#else
47453		#define vqrdmlahs_laneq_s32(__p0_330, __p1_330, __p2_330, __p3_330) __extension__ ({ \
47454		int32_t __ret_330; \
47455		int32_t __s0_330 = __p0_330; \
47456		int32_t __s1_330 = __p1_330; \
47457		int32x4_t __s2_330 = __p2_330; \
47458		int32x4_t __rev2_330; __rev2_330 = __builtin_shufflevector(__s2_330, __s2_330, 3, 2, 1, 0); \
47459		__ret_330 = vqrdmlahs_s32(__s0_330, __s1_330, __noswap_vgetq_lane_s32(__rev2_330, __p3_330)); \
47460		__ret_330; \
	48143	#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48144	int64x2x2_t __ret; \
	48145	int64x2x2_t __s1 = __p1; \
	48146	int64x2x2_t __rev1; \
	48147	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48148	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48149	__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
	48150	\
	48151	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48152	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48153	__ret; \
47461	48154	})
47462	48155	#endif
47463	48156
47464		#ifdef __LITTLE_ENDIAN__
47465		#define vqrdmlahh_laneq_s16(__p0_331, __p1_331, __p2_331, __p3_331) __extension__ ({ \
47466		int16_t __ret_331; \
47467		int16_t __s0_331 = __p0_331; \
47468		int16_t __s1_331 = __p1_331; \
47469		int16x8_t __s2_331 = __p2_331; \
47470		__ret_331 = vqrdmlahh_s16(__s0_331, __s1_331, vgetq_lane_s16(__s2_331, __p3_331)); \
47471		__ret_331; \
	48157	#define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48158	uint64x1x2_t __ret; \
	48159	uint64x1x2_t __s1 = __p1; \
	48160	__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
	48161	__ret; \
47472	48162	})
47473		#else
47474		#define vqrdmlahh_laneq_s16(__p0_332, __p1_332, __p2_332, __p3_332) __extension__ ({ \
47475		int16_t __ret_332; \
47476		int16_t __s0_332 = __p0_332; \
47477		int16_t __s1_332 = __p1_332; \
47478		int16x8_t __s2_332 = __p2_332; \
47479		int16x8_t __rev2_332; __rev2_332 = __builtin_shufflevector(__s2_332, __s2_332, 7, 6, 5, 4, 3, 2, 1, 0); \
47480		__ret_332 = vqrdmlahh_s16(__s0_332, __s1_332, __noswap_vgetq_lane_s16(__rev2_332, __p3_332)); \
47481		__ret_332; \
	48163	#define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48164	float64x1x2_t __ret; \
	48165	float64x1x2_t __s1 = __p1; \
	48166	__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
	48167	__ret; \
47482	48168	})
47483		#endif
47484
47485		#ifdef __LITTLE_ENDIAN__
47486		#define vqrdmlahq_laneq_s32(__p0_333, __p1_333, __p2_333, __p3_333) __extension__ ({ \
47487		int32x4_t __ret_333; \
47488		int32x4_t __s0_333 = __p0_333; \
47489		int32x4_t __s1_333 = __p1_333; \
47490		int32x4_t __s2_333 = __p2_333; \
47491		__ret_333 = vqrdmlahq_s32(__s0_333, __s1_333, splatq_laneq_s32(__s2_333, __p3_333)); \
47492		__ret_333; \
	48169	#define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48170	int64x1x2_t __ret; \
	48171	int64x1x2_t __s1 = __p1; \
	48172	__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
	48173	__ret; \
47493	48174	})
47494		#else
47495		#define vqrdmlahq_laneq_s32(__p0_334, __p1_334, __p2_334, __p3_334) __extension__ ({ \
47496		int32x4_t __ret_334; \
47497		int32x4_t __s0_334 = __p0_334; \
47498		int32x4_t __s1_334 = __p1_334; \
47499		int32x4_t __s2_334 = __p2_334; \
47500		int32x4_t __rev0_334; __rev0_334 = __builtin_shufflevector(__s0_334, __s0_334, 3, 2, 1, 0); \
47501		int32x4_t __rev1_334; __rev1_334 = __builtin_shufflevector(__s1_334, __s1_334, 3, 2, 1, 0); \
47502		int32x4_t __rev2_334; __rev2_334 = __builtin_shufflevector(__s2_334, __s2_334, 3, 2, 1, 0); \
47503		__ret_334 = __noswap_vqrdmlahq_s32(__rev0_334, __rev1_334, __noswap_splatq_laneq_s32(__rev2_334, __p3_334)); \
47504		__ret_334 = __builtin_shufflevector(__ret_334, __ret_334, 3, 2, 1, 0); \
47505		__ret_334; \
	48175	#define vld3_p64(__p0) __extension__ ({ \
	48176	poly64x1x3_t __ret; \
	48177	__builtin_neon_vld3_v(&__ret, __p0, 6); \
	48178	__ret; \
47506	48179	})
47507		#endif
47508
47509	48180	#ifdef __LITTLE_ENDIAN__
47510		#define vqrdmlahq_laneq_s16(__p0_335, __p1_335, __p2_335, __p3_335) __extension__ ({ \
47511		int16x8_t __ret_335; \
47512		int16x8_t __s0_335 = __p0_335; \
47513		int16x8_t __s1_335 = __p1_335; \
47514		int16x8_t __s2_335 = __p2_335; \
47515		__ret_335 = vqrdmlahq_s16(__s0_335, __s1_335, splatq_laneq_s16(__s2_335, __p3_335)); \
47516		__ret_335; \
	48181	#define vld3q_p64(__p0) __extension__ ({ \
	48182	poly64x2x3_t __ret; \
	48183	__builtin_neon_vld3q_v(&__ret, __p0, 38); \
	48184	__ret; \
47517	48185	})
47518	48186	#else
47519		#define vqrdmlahq_laneq_s16(__p0_336, __p1_336, __p2_336, __p3_336) __extension__ ({ \
47520		int16x8_t __ret_336; \
47521		int16x8_t __s0_336 = __p0_336; \
47522		int16x8_t __s1_336 = __p1_336; \
47523		int16x8_t __s2_336 = __p2_336; \
47524		int16x8_t __rev0_336; __rev0_336 = __builtin_shufflevector(__s0_336, __s0_336, 7, 6, 5, 4, 3, 2, 1, 0); \
47525		int16x8_t __rev1_336; __rev1_336 = __builtin_shufflevector(__s1_336, __s1_336, 7, 6, 5, 4, 3, 2, 1, 0); \
47526		int16x8_t __rev2_336; __rev2_336 = __builtin_shufflevector(__s2_336, __s2_336, 7, 6, 5, 4, 3, 2, 1, 0); \
47527		__ret_336 = __noswap_vqrdmlahq_s16(__rev0_336, __rev1_336, __noswap_splatq_laneq_s16(__rev2_336, __p3_336)); \
47528		__ret_336 = __builtin_shufflevector(__ret_336, __ret_336, 7, 6, 5, 4, 3, 2, 1, 0); \
47529		__ret_336; \
	48187	#define vld3q_p64(__p0) __extension__ ({ \
	48188	poly64x2x3_t __ret; \
	48189	__builtin_neon_vld3q_v(&__ret, __p0, 38); \
	48190	\
	48191	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48192	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48193	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48194	__ret; \
47530	48195	})
47531	48196	#endif
47532	48197
47533	48198	#ifdef __LITTLE_ENDIAN__
47534		#define vqrdmlah_laneq_s32(__p0_337, __p1_337, __p2_337, __p3_337) __extension__ ({ \
47535		int32x2_t __ret_337; \
47536		int32x2_t __s0_337 = __p0_337; \
47537		int32x2_t __s1_337 = __p1_337; \
47538		int32x4_t __s2_337 = __p2_337; \
47539		__ret_337 = vqrdmlah_s32(__s0_337, __s1_337, splat_laneq_s32(__s2_337, __p3_337)); \
47540		__ret_337; \
	48199	#define vld3q_u64(__p0) __extension__ ({ \
	48200	uint64x2x3_t __ret; \
	48201	__builtin_neon_vld3q_v(&__ret, __p0, 51); \
	48202	__ret; \
47541	48203	})
47542	48204	#else
47543		#define vqrdmlah_laneq_s32(__p0_338, __p1_338, __p2_338, __p3_338) __extension__ ({ \
47544		int32x2_t __ret_338; \
47545		int32x2_t __s0_338 = __p0_338; \
47546		int32x2_t __s1_338 = __p1_338; \
47547		int32x4_t __s2_338 = __p2_338; \
47548		int32x2_t __rev0_338; __rev0_338 = __builtin_shufflevector(__s0_338, __s0_338, 1, 0); \
47549		int32x2_t __rev1_338; __rev1_338 = __builtin_shufflevector(__s1_338, __s1_338, 1, 0); \
47550		int32x4_t __rev2_338; __rev2_338 = __builtin_shufflevector(__s2_338, __s2_338, 3, 2, 1, 0); \
47551		__ret_338 = __noswap_vqrdmlah_s32(__rev0_338, __rev1_338, __noswap_splat_laneq_s32(__rev2_338, __p3_338)); \
47552		__ret_338 = __builtin_shufflevector(__ret_338, __ret_338, 1, 0); \
47553		__ret_338; \
	48205	#define vld3q_u64(__p0) __extension__ ({ \
	48206	uint64x2x3_t __ret; \
	48207	__builtin_neon_vld3q_v(&__ret, __p0, 51); \
	48208	\
	48209	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48210	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48211	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48212	__ret; \
47554	48213	})
47555	48214	#endif
47556	48215
47557	48216	#ifdef __LITTLE_ENDIAN__
47558		#define vqrdmlah_laneq_s16(__p0_339, __p1_339, __p2_339, __p3_339) __extension__ ({ \
47559		int16x4_t __ret_339; \
47560		int16x4_t __s0_339 = __p0_339; \
47561		int16x4_t __s1_339 = __p1_339; \
47562		int16x8_t __s2_339 = __p2_339; \
47563		__ret_339 = vqrdmlah_s16(__s0_339, __s1_339, splat_laneq_s16(__s2_339, __p3_339)); \
47564		__ret_339; \
	48217	#define vld3q_f64(__p0) __extension__ ({ \
	48218	float64x2x3_t __ret; \
	48219	__builtin_neon_vld3q_v(&__ret, __p0, 42); \
	48220	__ret; \
47565	48221	})
47566	48222	#else
47567		#define vqrdmlah_laneq_s16(__p0_340, __p1_340, __p2_340, __p3_340) __extension__ ({ \
47568		int16x4_t __ret_340; \
47569		int16x4_t __s0_340 = __p0_340; \
47570		int16x4_t __s1_340 = __p1_340; \
47571		int16x8_t __s2_340 = __p2_340; \
47572		int16x4_t __rev0_340; __rev0_340 = __builtin_shufflevector(__s0_340, __s0_340, 3, 2, 1, 0); \
47573		int16x4_t __rev1_340; __rev1_340 = __builtin_shufflevector(__s1_340, __s1_340, 3, 2, 1, 0); \
47574		int16x8_t __rev2_340; __rev2_340 = __builtin_shufflevector(__s2_340, __s2_340, 7, 6, 5, 4, 3, 2, 1, 0); \
47575		__ret_340 = __noswap_vqrdmlah_s16(__rev0_340, __rev1_340, __noswap_splat_laneq_s16(__rev2_340, __p3_340)); \
47576		__ret_340 = __builtin_shufflevector(__ret_340, __ret_340, 3, 2, 1, 0); \
47577		__ret_340; \
	48223	#define vld3q_f64(__p0) __extension__ ({ \
	48224	float64x2x3_t __ret; \
	48225	__builtin_neon_vld3q_v(&__ret, __p0, 42); \
	48226	\
	48227	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48228	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48229	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48230	__ret; \
47578	48231	})
47579	48232	#endif
47580	48233
47581		__ai int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
47582		int32_t __ret;
47583		__ret = (int32_t) __builtin_neon_vqrdmlshs_s32(__p0, __p1, __p2);
47584		return __ret;
47585		}
47586		__ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
47587		int16_t __ret;
47588		__ret = (int16_t) __builtin_neon_vqrdmlshh_s16(__p0, __p1, __p2);
47589		return __ret;
47590		}
47591	48234	#ifdef __LITTLE_ENDIAN__
47592		#define vqrdmlshs_lane_s32(__p0_341, __p1_341, __p2_341, __p3_341) __extension__ ({ \
47593		int32_t __ret_341; \
47594		int32_t __s0_341 = __p0_341; \
47595		int32_t __s1_341 = __p1_341; \
47596		int32x2_t __s2_341 = __p2_341; \
47597		__ret_341 = vqrdmlshs_s32(__s0_341, __s1_341, vget_lane_s32(__s2_341, __p3_341)); \
47598		__ret_341; \
	48235	#define vld3q_s64(__p0) __extension__ ({ \
	48236	int64x2x3_t __ret; \
	48237	__builtin_neon_vld3q_v(&__ret, __p0, 35); \
	48238	__ret; \
47599	48239	})
47600	48240	#else
47601		#define vqrdmlshs_lane_s32(__p0_342, __p1_342, __p2_342, __p3_342) __extension__ ({ \
47602		int32_t __ret_342; \
47603		int32_t __s0_342 = __p0_342; \
47604		int32_t __s1_342 = __p1_342; \
47605		int32x2_t __s2_342 = __p2_342; \
47606		int32x2_t __rev2_342; __rev2_342 = __builtin_shufflevector(__s2_342, __s2_342, 1, 0); \
47607		__ret_342 = vqrdmlshs_s32(__s0_342, __s1_342, __noswap_vget_lane_s32(__rev2_342, __p3_342)); \
47608		__ret_342; \
	48241	#define vld3q_s64(__p0) __extension__ ({ \
	48242	int64x2x3_t __ret; \
	48243	__builtin_neon_vld3q_v(&__ret, __p0, 35); \
	48244	\
	48245	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48246	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48247	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48248	__ret; \
47609	48249	})
47610	48250	#endif
47611	48251
47612		#ifdef __LITTLE_ENDIAN__
47613		#define vqrdmlshh_lane_s16(__p0_343, __p1_343, __p2_343, __p3_343) __extension__ ({ \
47614		int16_t __ret_343; \
47615		int16_t __s0_343 = __p0_343; \
47616		int16_t __s1_343 = __p1_343; \
47617		int16x4_t __s2_343 = __p2_343; \
47618		__ret_343 = vqrdmlshh_s16(__s0_343, __s1_343, vget_lane_s16(__s2_343, __p3_343)); \
47619		__ret_343; \
	48252	#define vld3_f64(__p0) __extension__ ({ \
	48253	float64x1x3_t __ret; \
	48254	__builtin_neon_vld3_v(&__ret, __p0, 10); \
	48255	__ret; \
47620	48256	})
47621		#else
47622		#define vqrdmlshh_lane_s16(__p0_344, __p1_344, __p2_344, __p3_344) __extension__ ({ \
47623		int16_t __ret_344; \
47624		int16_t __s0_344 = __p0_344; \
47625		int16_t __s1_344 = __p1_344; \
47626		int16x4_t __s2_344 = __p2_344; \
47627		int16x4_t __rev2_344; __rev2_344 = __builtin_shufflevector(__s2_344, __s2_344, 3, 2, 1, 0); \
47628		__ret_344 = vqrdmlshh_s16(__s0_344, __s1_344, __noswap_vget_lane_s16(__rev2_344, __p3_344)); \
47629		__ret_344; \
	48257	#define vld3_dup_p64(__p0) __extension__ ({ \
	48258	poly64x1x3_t __ret; \
	48259	__builtin_neon_vld3_dup_v(&__ret, __p0, 6); \
	48260	__ret; \
47630	48261	})
47631		#endif
47632
47633	48262	#ifdef __LITTLE_ENDIAN__
47634		#define vqrdmlshs_laneq_s32(__p0_345, __p1_345, __p2_345, __p3_345) __extension__ ({ \
47635		int32_t __ret_345; \
47636		int32_t __s0_345 = __p0_345; \
47637		int32_t __s1_345 = __p1_345; \
47638		int32x4_t __s2_345 = __p2_345; \
47639		__ret_345 = vqrdmlshs_s32(__s0_345, __s1_345, vgetq_lane_s32(__s2_345, __p3_345)); \
47640		__ret_345; \
	48263	#define vld3q_dup_p64(__p0) __extension__ ({ \
	48264	poly64x2x3_t __ret; \
	48265	__builtin_neon_vld3q_dup_v(&__ret, __p0, 38); \
	48266	__ret; \
47641	48267	})
47642	48268	#else
47643		#define vqrdmlshs_laneq_s32(__p0_346, __p1_346, __p2_346, __p3_346) __extension__ ({ \
47644		int32_t __ret_346; \
47645		int32_t __s0_346 = __p0_346; \
47646		int32_t __s1_346 = __p1_346; \
47647		int32x4_t __s2_346 = __p2_346; \
47648		int32x4_t __rev2_346; __rev2_346 = __builtin_shufflevector(__s2_346, __s2_346, 3, 2, 1, 0); \
47649		__ret_346 = vqrdmlshs_s32(__s0_346, __s1_346, __noswap_vgetq_lane_s32(__rev2_346, __p3_346)); \
47650		__ret_346; \
	48269	#define vld3q_dup_p64(__p0) __extension__ ({ \
	48270	poly64x2x3_t __ret; \
	48271	__builtin_neon_vld3q_dup_v(&__ret, __p0, 38); \
	48272	\
	48273	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48274	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48275	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48276	__ret; \
47651	48277	})
47652	48278	#endif
47653	48279
47654	48280	#ifdef __LITTLE_ENDIAN__
47655		#define vqrdmlshh_laneq_s16(__p0_347, __p1_347, __p2_347, __p3_347) __extension__ ({ \
47656		int16_t __ret_347; \
47657		int16_t __s0_347 = __p0_347; \
47658		int16_t __s1_347 = __p1_347; \
47659		int16x8_t __s2_347 = __p2_347; \
47660		__ret_347 = vqrdmlshh_s16(__s0_347, __s1_347, vgetq_lane_s16(__s2_347, __p3_347)); \
47661		__ret_347; \
	48281	#define vld3q_dup_f64(__p0) __extension__ ({ \
	48282	float64x2x3_t __ret; \
	48283	__builtin_neon_vld3q_dup_v(&__ret, __p0, 42); \
	48284	__ret; \
47662	48285	})
47663	48286	#else
47664		#define vqrdmlshh_laneq_s16(__p0_348, __p1_348, __p2_348, __p3_348) __extension__ ({ \
47665		int16_t __ret_348; \
47666		int16_t __s0_348 = __p0_348; \
47667		int16_t __s1_348 = __p1_348; \
47668		int16x8_t __s2_348 = __p2_348; \
47669		int16x8_t __rev2_348; __rev2_348 = __builtin_shufflevector(__s2_348, __s2_348, 7, 6, 5, 4, 3, 2, 1, 0); \
47670		__ret_348 = vqrdmlshh_s16(__s0_348, __s1_348, __noswap_vgetq_lane_s16(__rev2_348, __p3_348)); \
47671		__ret_348; \
	48287	#define vld3q_dup_f64(__p0) __extension__ ({ \
	48288	float64x2x3_t __ret; \
	48289	__builtin_neon_vld3q_dup_v(&__ret, __p0, 42); \
	48290	\
	48291	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48292	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48293	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48294	__ret; \
47672	48295	})
47673	48296	#endif
47674	48297
47675		#ifdef __LITTLE_ENDIAN__
47676		#define vqrdmlshq_laneq_s32(__p0_349, __p1_349, __p2_349, __p3_349) __extension__ ({ \
47677		int32x4_t __ret_349; \
47678		int32x4_t __s0_349 = __p0_349; \
47679		int32x4_t __s1_349 = __p1_349; \
47680		int32x4_t __s2_349 = __p2_349; \
47681		__ret_349 = vqrdmlshq_s32(__s0_349, __s1_349, splatq_laneq_s32(__s2_349, __p3_349)); \
47682		__ret_349; \
	48298	#define vld3_dup_f64(__p0) __extension__ ({ \
	48299	float64x1x3_t __ret; \
	48300	__builtin_neon_vld3_dup_v(&__ret, __p0, 10); \
	48301	__ret; \
47683	48302	})
47684		#else
47685		#define vqrdmlshq_laneq_s32(__p0_350, __p1_350, __p2_350, __p3_350) __extension__ ({ \
47686		int32x4_t __ret_350; \
47687		int32x4_t __s0_350 = __p0_350; \
47688		int32x4_t __s1_350 = __p1_350; \
47689		int32x4_t __s2_350 = __p2_350; \
47690		int32x4_t __rev0_350; __rev0_350 = __builtin_shufflevector(__s0_350, __s0_350, 3, 2, 1, 0); \
47691		int32x4_t __rev1_350; __rev1_350 = __builtin_shufflevector(__s1_350, __s1_350, 3, 2, 1, 0); \
47692		int32x4_t __rev2_350; __rev2_350 = __builtin_shufflevector(__s2_350, __s2_350, 3, 2, 1, 0); \
47693		__ret_350 = __noswap_vqrdmlshq_s32(__rev0_350, __rev1_350, __noswap_splatq_laneq_s32(__rev2_350, __p3_350)); \
47694		__ret_350 = __builtin_shufflevector(__ret_350, __ret_350, 3, 2, 1, 0); \
47695		__ret_350; \
	48303	#define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48304	poly64x1x3_t __ret; \
	48305	poly64x1x3_t __s1 = __p1; \
	48306	__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
	48307	__ret; \
47696	48308	})
47697		#endif
47698
47699	48309	#ifdef __LITTLE_ENDIAN__
47700		#define vqrdmlshq_laneq_s16(__p0_351, __p1_351, __p2_351, __p3_351) __extension__ ({ \
47701		int16x8_t __ret_351; \
47702		int16x8_t __s0_351 = __p0_351; \
47703		int16x8_t __s1_351 = __p1_351; \
47704		int16x8_t __s2_351 = __p2_351; \
47705		__ret_351 = vqrdmlshq_s16(__s0_351, __s1_351, splatq_laneq_s16(__s2_351, __p3_351)); \
47706		__ret_351; \
	48310	#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	48311	poly8x16x3_t __ret; \
	48312	poly8x16x3_t __s1 = __p1; \
	48313	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 36); \
	48314	__ret; \
47707	48315	})
47708	48316	#else
47709		#define vqrdmlshq_laneq_s16(__p0_352, __p1_352, __p2_352, __p3_352) __extension__ ({ \
47710		int16x8_t __ret_352; \
47711		int16x8_t __s0_352 = __p0_352; \
47712		int16x8_t __s1_352 = __p1_352; \
47713		int16x8_t __s2_352 = __p2_352; \
47714		int16x8_t __rev0_352; __rev0_352 = __builtin_shufflevector(__s0_352, __s0_352, 7, 6, 5, 4, 3, 2, 1, 0); \
47715		int16x8_t __rev1_352; __rev1_352 = __builtin_shufflevector(__s1_352, __s1_352, 7, 6, 5, 4, 3, 2, 1, 0); \
47716		int16x8_t __rev2_352; __rev2_352 = __builtin_shufflevector(__s2_352, __s2_352, 7, 6, 5, 4, 3, 2, 1, 0); \
47717		__ret_352 = __noswap_vqrdmlshq_s16(__rev0_352, __rev1_352, __noswap_splatq_laneq_s16(__rev2_352, __p3_352)); \
47718		__ret_352 = __builtin_shufflevector(__ret_352, __ret_352, 7, 6, 5, 4, 3, 2, 1, 0); \
47719		__ret_352; \
	48317	#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	48318	poly8x16x3_t __ret; \
	48319	poly8x16x3_t __s1 = __p1; \
	48320	poly8x16x3_t __rev1; \
	48321	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48322	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48323	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48324	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 36); \
	48325	\
	48326	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48327	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48328	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48329	__ret; \
47720	48330	})
47721	48331	#endif
47722	48332
47723	48333	#ifdef __LITTLE_ENDIAN__
47724		#define vqrdmlsh_laneq_s32(__p0_353, __p1_353, __p2_353, __p3_353) __extension__ ({ \
47725		int32x2_t __ret_353; \
47726		int32x2_t __s0_353 = __p0_353; \
47727		int32x2_t __s1_353 = __p1_353; \
47728		int32x4_t __s2_353 = __p2_353; \
47729		__ret_353 = vqrdmlsh_s32(__s0_353, __s1_353, splat_laneq_s32(__s2_353, __p3_353)); \
47730		__ret_353; \
	48334	#define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48335	poly64x2x3_t __ret; \
	48336	poly64x2x3_t __s1 = __p1; \
	48337	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 38); \
	48338	__ret; \
47731	48339	})
47732	48340	#else
47733		#define vqrdmlsh_laneq_s32(__p0_354, __p1_354, __p2_354, __p3_354) __extension__ ({ \
47734		int32x2_t __ret_354; \
47735		int32x2_t __s0_354 = __p0_354; \
47736		int32x2_t __s1_354 = __p1_354; \
47737		int32x4_t __s2_354 = __p2_354; \
47738		int32x2_t __rev0_354; __rev0_354 = __builtin_shufflevector(__s0_354, __s0_354, 1, 0); \
47739		int32x2_t __rev1_354; __rev1_354 = __builtin_shufflevector(__s1_354, __s1_354, 1, 0); \
47740		int32x4_t __rev2_354; __rev2_354 = __builtin_shufflevector(__s2_354, __s2_354, 3, 2, 1, 0); \
47741		__ret_354 = __noswap_vqrdmlsh_s32(__rev0_354, __rev1_354, __noswap_splat_laneq_s32(__rev2_354, __p3_354)); \
47742		__ret_354 = __builtin_shufflevector(__ret_354, __ret_354, 1, 0); \
47743		__ret_354; \
	48341	#define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48342	poly64x2x3_t __ret; \
	48343	poly64x2x3_t __s1 = __p1; \
	48344	poly64x2x3_t __rev1; \
	48345	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48346	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48347	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48348	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 38); \
	48349	\
	48350	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48351	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48352	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48353	__ret; \
47744	48354	})
47745	48355	#endif
47746	48356
47747	48357	#ifdef __LITTLE_ENDIAN__
47748		#define vqrdmlsh_laneq_s16(__p0_355, __p1_355, __p2_355, __p3_355) __extension__ ({ \
47749		int16x4_t __ret_355; \
47750		int16x4_t __s0_355 = __p0_355; \
47751		int16x4_t __s1_355 = __p1_355; \
47752		int16x8_t __s2_355 = __p2_355; \
47753		__ret_355 = vqrdmlsh_s16(__s0_355, __s1_355, splat_laneq_s16(__s2_355, __p3_355)); \
47754		__ret_355; \
	48358	#define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	48359	uint8x16x3_t __ret; \
	48360	uint8x16x3_t __s1 = __p1; \
	48361	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 48); \
	48362	__ret; \
47755	48363	})
47756	48364	#else
47757		#define vqrdmlsh_laneq_s16(__p0_356, __p1_356, __p2_356, __p3_356) __extension__ ({ \
47758		int16x4_t __ret_356; \
47759		int16x4_t __s0_356 = __p0_356; \
47760		int16x4_t __s1_356 = __p1_356; \
47761		int16x8_t __s2_356 = __p2_356; \
47762		int16x4_t __rev0_356; __rev0_356 = __builtin_shufflevector(__s0_356, __s0_356, 3, 2, 1, 0); \
47763		int16x4_t __rev1_356; __rev1_356 = __builtin_shufflevector(__s1_356, __s1_356, 3, 2, 1, 0); \
47764		int16x8_t __rev2_356; __rev2_356 = __builtin_shufflevector(__s2_356, __s2_356, 7, 6, 5, 4, 3, 2, 1, 0); \
47765		__ret_356 = __noswap_vqrdmlsh_s16(__rev0_356, __rev1_356, __noswap_splat_laneq_s16(__rev2_356, __p3_356)); \
47766		__ret_356 = __builtin_shufflevector(__ret_356, __ret_356, 3, 2, 1, 0); \
47767		__ret_356; \
	48365	#define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	48366	uint8x16x3_t __ret; \
	48367	uint8x16x3_t __s1 = __p1; \
	48368	uint8x16x3_t __rev1; \
	48369	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48370	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48371	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48372	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 48); \
	48373	\
	48374	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48375	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48376	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48377	__ret; \
47768	48378	})
47769	48379	#endif
47770	48380
47771		#endif
47772		#if defined(__aarch64__)
47773	48381	#ifdef __LITTLE_ENDIAN__
47774		__ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
47775		float64x2_t __ret;
47776		__ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
47777		return __ret;
47778		}
	48382	#define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48383	uint64x2x3_t __ret; \
	48384	uint64x2x3_t __s1 = __p1; \
	48385	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 51); \
	48386	__ret; \
	48387	})
47779	48388	#else
47780		__ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
47781		float64x2_t __ret;
47782		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47783		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47784		__ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
47785		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47786		return __ret;
47787		}
	48389	#define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48390	uint64x2x3_t __ret; \
	48391	uint64x2x3_t __s1 = __p1; \
	48392	uint64x2x3_t __rev1; \
	48393	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48394	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48395	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48396	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 51); \
	48397	\
	48398	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48399	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48400	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48401	__ret; \
	48402	})
47788	48403	#endif
47789	48404
47790		__ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
47791		float64x1_t __ret;
47792		__ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
47793		return __ret;
47794		}
47795		__ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
47796		float64_t __ret;
47797		__ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
47798		return __ret;
47799		}
47800		__ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
47801		float32_t __ret;
47802		__ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
47803		return __ret;
47804		}
47805	48405	#ifdef __LITTLE_ENDIAN__
47806		__ai float64x2_t vabsq_f64(float64x2_t __p0) {
47807		float64x2_t __ret;
47808		__ret = (float64x2_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 42);
47809		return __ret;
47810		}
	48406	#define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	48407	int8x16x3_t __ret; \
	48408	int8x16x3_t __s1 = __p1; \
	48409	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 32); \
	48410	__ret; \
	48411	})
47811	48412	#else
47812		__ai float64x2_t vabsq_f64(float64x2_t __p0) {
47813		float64x2_t __ret;
47814		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47815		__ret = (float64x2_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 42);
47816		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47817		return __ret;
47818		}
	48413	#define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	48414	int8x16x3_t __ret; \
	48415	int8x16x3_t __s1 = __p1; \
	48416	int8x16x3_t __rev1; \
	48417	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48418	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48419	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48420	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 32); \
	48421	\
	48422	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48423	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48424	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48425	__ret; \
	48426	})
47819	48427	#endif
47820	48428
47821	48429	#ifdef __LITTLE_ENDIAN__
47822		__ai int64x2_t vabsq_s64(int64x2_t __p0) {
47823		int64x2_t __ret;
47824		__ret = (int64x2_t) __builtin_neon_vabsq_v((int8x16_t)__p0, 35);
47825		return __ret;
47826		}
	48430	#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48431	float64x2x3_t __ret; \
	48432	float64x2x3_t __s1 = __p1; \
	48433	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
	48434	__ret; \
	48435	})
47827	48436	#else
47828		__ai int64x2_t vabsq_s64(int64x2_t __p0) {
47829		int64x2_t __ret;
47830		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47831		__ret = (int64x2_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 35);
47832		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47833		return __ret;
47834		}
	48437	#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48438	float64x2x3_t __ret; \
	48439	float64x2x3_t __s1 = __p1; \
	48440	float64x2x3_t __rev1; \
	48441	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48442	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48443	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48444	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
	48445	\
	48446	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48447	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48448	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48449	__ret; \
	48450	})
47835	48451	#endif
47836	48452
47837		__ai float64x1_t vabs_f64(float64x1_t __p0) {
47838		float64x1_t __ret;
47839		__ret = (float64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 10);
47840		return __ret;
47841		}
47842		__ai int64x1_t vabs_s64(int64x1_t __p0) {
47843		int64x1_t __ret;
47844		__ret = (int64x1_t) __builtin_neon_vabs_v((int8x8_t)__p0, 3);
47845		return __ret;
47846		}
47847		__ai int64_t vabsd_s64(int64_t __p0) {
47848		int64_t __ret;
47849		__ret = (int64_t) __builtin_neon_vabsd_s64(__p0);
47850		return __ret;
47851		}
47852	48453	#ifdef __LITTLE_ENDIAN__
47853		__ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
47854		float64x2_t __ret;
47855		__ret = __p0 + __p1;
47856		return __ret;
47857		}
47858		#else
47859		__ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
47860		float64x2_t __ret;
47861		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47862		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47863		__ret = __rev0 + __rev1;
47864		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47865		return __ret;
47866		}
	48454	#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48455	int64x2x3_t __ret; \
	48456	int64x2x3_t __s1 = __p1; \
	48457	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
	48458	__ret; \
	48459	})
	48460	#else
	48461	#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48462	int64x2x3_t __ret; \
	48463	int64x2x3_t __s1 = __p1; \
	48464	int64x2x3_t __rev1; \
	48465	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48466	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48467	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48468	__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
	48469	\
	48470	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48471	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48472	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48473	__ret; \
	48474	})
47867	48475	#endif
47868	48476
47869		__ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
47870		float64x1_t __ret;
47871		__ret = __p0 + __p1;
47872		return __ret;
47873		}
47874		__ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
47875		uint64_t __ret;
47876		__ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
47877		return __ret;
47878		}
47879		__ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
47880		int64_t __ret;
47881		__ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
47882		return __ret;
47883		}
47884		__ai poly128_t vaddq_p128(poly128_t __p0, poly128_t __p1) {
47885		poly128_t __ret;
47886		__ret = (poly128_t) __builtin_neon_vaddq_p128(__p0, __p1);
47887		return __ret;
47888		}
	48477	#define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48478	uint64x1x3_t __ret; \
	48479	uint64x1x3_t __s1 = __p1; \
	48480	__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
	48481	__ret; \
	48482	})
	48483	#define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48484	float64x1x3_t __ret; \
	48485	float64x1x3_t __s1 = __p1; \
	48486	__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
	48487	__ret; \
	48488	})
	48489	#define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48490	int64x1x3_t __ret; \
	48491	int64x1x3_t __s1 = __p1; \
	48492	__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
	48493	__ret; \
	48494	})
	48495	#define vld4_p64(__p0) __extension__ ({ \
	48496	poly64x1x4_t __ret; \
	48497	__builtin_neon_vld4_v(&__ret, __p0, 6); \
	48498	__ret; \
	48499	})
47889	48500	#ifdef __LITTLE_ENDIAN__
47890		__ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
47891		uint16x8_t __ret;
47892		__ret = vcombine_u16(__p0, vaddhn_u32(__p1, __p2));
47893		return __ret;
47894		}
	48501	#define vld4q_p64(__p0) __extension__ ({ \
	48502	poly64x2x4_t __ret; \
	48503	__builtin_neon_vld4q_v(&__ret, __p0, 38); \
	48504	__ret; \
	48505	})
47895	48506	#else
47896		__ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
47897		uint16x8_t __ret;
47898		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47899		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47900		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47901		__ret = __noswap_vcombine_u16(__rev0, __noswap_vaddhn_u32(__rev1, __rev2));
47902		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
47903		return __ret;
47904		}
	48507	#define vld4q_p64(__p0) __extension__ ({ \
	48508	poly64x2x4_t __ret; \
	48509	__builtin_neon_vld4q_v(&__ret, __p0, 38); \
	48510	\
	48511	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48512	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48513	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48514	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48515	__ret; \
	48516	})
47905	48517	#endif
47906	48518
47907	48519	#ifdef __LITTLE_ENDIAN__
47908		__ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
47909		uint32x4_t __ret;
47910		__ret = vcombine_u32(__p0, vaddhn_u64(__p1, __p2));
47911		return __ret;
47912		}
	48520	#define vld4q_u64(__p0) __extension__ ({ \
	48521	uint64x2x4_t __ret; \
	48522	__builtin_neon_vld4q_v(&__ret, __p0, 51); \
	48523	__ret; \
	48524	})
47913	48525	#else
47914		__ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
47915		uint32x4_t __ret;
47916		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47917		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47918		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47919		__ret = __noswap_vcombine_u32(__rev0, __noswap_vaddhn_u64(__rev1, __rev2));
47920		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47921		return __ret;
47922		}
	48526	#define vld4q_u64(__p0) __extension__ ({ \
	48527	uint64x2x4_t __ret; \
	48528	__builtin_neon_vld4q_v(&__ret, __p0, 51); \
	48529	\
	48530	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48531	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48532	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48533	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48534	__ret; \
	48535	})
47923	48536	#endif
47924	48537
47925	48538	#ifdef __LITTLE_ENDIAN__
47926		__ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
47927		uint8x16_t __ret;
47928		__ret = vcombine_u8(__p0, vaddhn_u16(__p1, __p2));
47929		return __ret;
47930		}
	48539	#define vld4q_f64(__p0) __extension__ ({ \
	48540	float64x2x4_t __ret; \
	48541	__builtin_neon_vld4q_v(&__ret, __p0, 42); \
	48542	__ret; \
	48543	})
47931	48544	#else
47932		__ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
47933		uint8x16_t __ret;
47934		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47935		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
47936		uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47937		__ret = __noswap_vcombine_u8(__rev0, __noswap_vaddhn_u16(__rev1, __rev2));
47938		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47939		return __ret;
47940		}
	48545	#define vld4q_f64(__p0) __extension__ ({ \
	48546	float64x2x4_t __ret; \
	48547	__builtin_neon_vld4q_v(&__ret, __p0, 42); \
	48548	\
	48549	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48550	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48551	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48552	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48553	__ret; \
	48554	})
47941	48555	#endif
47942	48556
47943	48557	#ifdef __LITTLE_ENDIAN__
47944		__ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47945		int16x8_t __ret;
47946		__ret = vcombine_s16(__p0, vaddhn_s32(__p1, __p2));
47947		return __ret;
47948		}
	48558	#define vld4q_s64(__p0) __extension__ ({ \
	48559	int64x2x4_t __ret; \
	48560	__builtin_neon_vld4q_v(&__ret, __p0, 35); \
	48561	__ret; \
	48562	})
47949	48563	#else
47950		__ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47951		int16x8_t __ret;
47952		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47953		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47954		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47955		__ret = __noswap_vcombine_s16(__rev0, __noswap_vaddhn_s32(__rev1, __rev2));
47956		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
47957		return __ret;
47958		}
	48564	#define vld4q_s64(__p0) __extension__ ({ \
	48565	int64x2x4_t __ret; \
	48566	__builtin_neon_vld4q_v(&__ret, __p0, 35); \
	48567	\
	48568	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48569	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48570	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48571	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48572	__ret; \
	48573	})
47959	48574	#endif
47960	48575
	48576	#define vld4_f64(__p0) __extension__ ({ \
	48577	float64x1x4_t __ret; \
	48578	__builtin_neon_vld4_v(&__ret, __p0, 10); \
	48579	__ret; \
	48580	})
	48581	#define vld4_dup_p64(__p0) __extension__ ({ \
	48582	poly64x1x4_t __ret; \
	48583	__builtin_neon_vld4_dup_v(&__ret, __p0, 6); \
	48584	__ret; \
	48585	})
47961	48586	#ifdef __LITTLE_ENDIAN__
47962		__ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
47963		int32x4_t __ret;
47964		__ret = vcombine_s32(__p0, vaddhn_s64(__p1, __p2));
47965		return __ret;
47966		}
	48587	#define vld4q_dup_p64(__p0) __extension__ ({ \
	48588	poly64x2x4_t __ret; \
	48589	__builtin_neon_vld4q_dup_v(&__ret, __p0, 38); \
	48590	__ret; \
	48591	})
47967	48592	#else
47968		__ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
47969		int32x4_t __ret;
47970		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47971		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47972		int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47973		__ret = __noswap_vcombine_s32(__rev0, __noswap_vaddhn_s64(__rev1, __rev2));
47974		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47975		return __ret;
47976		}
	48593	#define vld4q_dup_p64(__p0) __extension__ ({ \
	48594	poly64x2x4_t __ret; \
	48595	__builtin_neon_vld4q_dup_v(&__ret, __p0, 38); \
	48596	\
	48597	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48598	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48599	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48600	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48601	__ret; \
	48602	})
47977	48603	#endif
47978	48604
47979	48605	#ifdef __LITTLE_ENDIAN__
47980		__ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47981		int8x16_t __ret;
47982		__ret = vcombine_s8(__p0, vaddhn_s16(__p1, __p2));
47983		return __ret;
47984		}
	48606	#define vld4q_dup_f64(__p0) __extension__ ({ \
	48607	float64x2x4_t __ret; \
	48608	__builtin_neon_vld4q_dup_v(&__ret, __p0, 42); \
	48609	__ret; \
	48610	})
47985	48611	#else
47986		__ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47987		int8x16_t __ret;
47988		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47989		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
47990		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47991		__ret = __noswap_vcombine_s8(__rev0, __noswap_vaddhn_s16(__rev1, __rev2));
47992		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47993		return __ret;
47994		}
	48612	#define vld4q_dup_f64(__p0) __extension__ ({ \
	48613	float64x2x4_t __ret; \
	48614	__builtin_neon_vld4q_dup_v(&__ret, __p0, 42); \
	48615	\
	48616	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48617	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48618	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48619	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48620	__ret; \
	48621	})
47995	48622	#endif
47996	48623
	48624	#define vld4_dup_f64(__p0) __extension__ ({ \
	48625	float64x1x4_t __ret; \
	48626	__builtin_neon_vld4_dup_v(&__ret, __p0, 10); \
	48627	__ret; \
	48628	})
	48629	#define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48630	poly64x1x4_t __ret; \
	48631	poly64x1x4_t __s1 = __p1; \
	48632	__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
	48633	__ret; \
	48634	})
47997	48635	#ifdef __LITTLE_ENDIAN__
47998		__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
47999		uint16_t __ret;
48000		__ret = (uint16_t) __builtin_neon_vaddlvq_u8(__p0);
48001		return __ret;
48002		}
	48636	#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	48637	poly8x16x4_t __ret; \
	48638	poly8x16x4_t __s1 = __p1; \
	48639	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 36); \
	48640	__ret; \
	48641	})
48003	48642	#else
48004		__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
48005		uint16_t __ret;
48006		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48007		__ret = (uint16_t) __builtin_neon_vaddlvq_u8(__rev0);
48008		return __ret;
48009		}
	48643	#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	48644	poly8x16x4_t __ret; \
	48645	poly8x16x4_t __s1 = __p1; \
	48646	poly8x16x4_t __rev1; \
	48647	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48648	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48649	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48650	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48651	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 36); \
	48652	\
	48653	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48654	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48655	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48656	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48657	__ret; \
	48658	})
48010	48659	#endif
48011	48660
48012	48661	#ifdef __LITTLE_ENDIAN__
48013		__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
48014		uint64_t __ret;
48015		__ret = (uint64_t) __builtin_neon_vaddlvq_u32(__p0);
48016		return __ret;
48017		}
	48662	#define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48663	poly64x2x4_t __ret; \
	48664	poly64x2x4_t __s1 = __p1; \
	48665	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 38); \
	48666	__ret; \
	48667	})
48018	48668	#else
48019		__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
48020		uint64_t __ret;
48021		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48022		__ret = (uint64_t) __builtin_neon_vaddlvq_u32(__rev0);
48023		return __ret;
48024		}
	48669	#define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	48670	poly64x2x4_t __ret; \
	48671	poly64x2x4_t __s1 = __p1; \
	48672	poly64x2x4_t __rev1; \
	48673	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48674	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48675	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48676	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	48677	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 38); \
	48678	\
	48679	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48680	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48681	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48682	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48683	__ret; \
	48684	})
48025	48685	#endif
48026	48686
48027	48687	#ifdef __LITTLE_ENDIAN__
48028		__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
48029		uint32_t __ret;
48030		__ret = (uint32_t) __builtin_neon_vaddlvq_u16(__p0);
48031		return __ret;
48032		}
	48688	#define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	48689	uint8x16x4_t __ret; \
	48690	uint8x16x4_t __s1 = __p1; \
	48691	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 48); \
	48692	__ret; \
	48693	})
48033	48694	#else
48034		__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
48035		uint32_t __ret;
48036		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48037		__ret = (uint32_t) __builtin_neon_vaddlvq_u16(__rev0);
48038		return __ret;
48039		}
	48695	#define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	48696	uint8x16x4_t __ret; \
	48697	uint8x16x4_t __s1 = __p1; \
	48698	uint8x16x4_t __rev1; \
	48699	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48700	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48701	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48702	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48703	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 48); \
	48704	\
	48705	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48706	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48707	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48708	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48709	__ret; \
	48710	})
48040	48711	#endif
48041	48712
48042	48713	#ifdef __LITTLE_ENDIAN__
48043		__ai int16_t vaddlvq_s8(int8x16_t __p0) {
48044		int16_t __ret;
48045		__ret = (int16_t) __builtin_neon_vaddlvq_s8(__p0);
48046		return __ret;
48047		}
	48714	#define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48715	uint64x2x4_t __ret; \
	48716	uint64x2x4_t __s1 = __p1; \
	48717	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 51); \
	48718	__ret; \
	48719	})
48048	48720	#else
48049		__ai int16_t vaddlvq_s8(int8x16_t __p0) {
48050		int16_t __ret;
48051		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48052		__ret = (int16_t) __builtin_neon_vaddlvq_s8(__rev0);
48053		return __ret;
48054		}
	48721	#define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48722	uint64x2x4_t __ret; \
	48723	uint64x2x4_t __s1 = __p1; \
	48724	uint64x2x4_t __rev1; \
	48725	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48726	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48727	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48728	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	48729	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 51); \
	48730	\
	48731	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48732	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48733	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48734	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48735	__ret; \
	48736	})
48055	48737	#endif
48056	48738
48057	48739	#ifdef __LITTLE_ENDIAN__
48058		__ai int64_t vaddlvq_s32(int32x4_t __p0) {
48059		int64_t __ret;
48060		__ret = (int64_t) __builtin_neon_vaddlvq_s32(__p0);
48061		return __ret;
48062		}
	48740	#define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	48741	int8x16x4_t __ret; \
	48742	int8x16x4_t __s1 = __p1; \
	48743	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 32); \
	48744	__ret; \
	48745	})
48063	48746	#else
48064		__ai int64_t vaddlvq_s32(int32x4_t __p0) {
48065		int64_t __ret;
48066		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48067		__ret = (int64_t) __builtin_neon_vaddlvq_s32(__rev0);
48068		return __ret;
48069		}
	48747	#define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	48748	int8x16x4_t __ret; \
	48749	int8x16x4_t __s1 = __p1; \
	48750	int8x16x4_t __rev1; \
	48751	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48752	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48753	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48754	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48755	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 32); \
	48756	\
	48757	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48758	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48759	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48760	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	48761	__ret; \
	48762	})
48070	48763	#endif
48071	48764
48072	48765	#ifdef __LITTLE_ENDIAN__
48073		__ai int32_t vaddlvq_s16(int16x8_t __p0) {
48074		int32_t __ret;
48075		__ret = (int32_t) __builtin_neon_vaddlvq_s16(__p0);
48076		return __ret;
48077		}
	48766	#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48767	float64x2x4_t __ret; \
	48768	float64x2x4_t __s1 = __p1; \
	48769	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 42); \
	48770	__ret; \
	48771	})
48078	48772	#else
48079		__ai int32_t vaddlvq_s16(int16x8_t __p0) {
48080		int32_t __ret;
48081		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48082		__ret = (int32_t) __builtin_neon_vaddlvq_s16(__rev0);
48083		return __ret;
48084		}
	48773	#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48774	float64x2x4_t __ret; \
	48775	float64x2x4_t __s1 = __p1; \
	48776	float64x2x4_t __rev1; \
	48777	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48778	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48779	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48780	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	48781	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 42); \
	48782	\
	48783	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48784	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48785	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48786	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48787	__ret; \
	48788	})
48085	48789	#endif
48086	48790
48087	48791	#ifdef __LITTLE_ENDIAN__
48088		__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
48089		uint16_t __ret;
48090		__ret = (uint16_t) __builtin_neon_vaddlv_u8(__p0);
48091		return __ret;
48092		}
	48792	#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48793	int64x2x4_t __ret; \
	48794	int64x2x4_t __s1 = __p1; \
	48795	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 35); \
	48796	__ret; \
	48797	})
48093	48798	#else
48094		__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
48095		uint16_t __ret;
48096		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48097		__ret = (uint16_t) __builtin_neon_vaddlv_u8(__rev0);
48098		return __ret;
48099		}
	48799	#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48800	int64x2x4_t __ret; \
	48801	int64x2x4_t __s1 = __p1; \
	48802	int64x2x4_t __rev1; \
	48803	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	48804	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	48805	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	48806	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	48807	__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 35); \
	48808	\
	48809	__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
	48810	__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
	48811	__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
	48812	__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
	48813	__ret; \
	48814	})
48100	48815	#endif
48101	48816
	48817	#define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	48818	uint64x1x4_t __ret; \
	48819	uint64x1x4_t __s1 = __p1; \
	48820	__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
	48821	__ret; \
	48822	})
	48823	#define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	48824	float64x1x4_t __ret; \
	48825	float64x1x4_t __s1 = __p1; \
	48826	__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 10); \
	48827	__ret; \
	48828	})
	48829	#define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	48830	int64x1x4_t __ret; \
	48831	int64x1x4_t __s1 = __p1; \
	48832	__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 3); \
	48833	__ret; \
	48834	})
	48835	#define vldrq_p128(__p0) __extension__ ({ \
	48836	poly128_t __ret; \
	48837	__ret = (poly128_t) __builtin_neon_vldrq_p128(__p0); \
	48838	__ret; \
	48839	})
48102	48840	#ifdef __LITTLE_ENDIAN__
48103		__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
48104		uint64_t __ret;
48105		__ret = (uint64_t) __builtin_neon_vaddlv_u32(__p0);
	48841	__ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
	48842	float64x2_t __ret;
	48843	__ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
48106	48844	return __ret;
48107	48845	}
48108	48846	#else
48109		__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
48110		uint64_t __ret;
48111		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48112		__ret = (uint64_t) __builtin_neon_vaddlv_u32(__rev0);
	48847	__ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
	48848	float64x2_t __ret;
	48849	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	48850	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	48851	__ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	48852	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48113	48853	return __ret;
48114	48854	}
48115	48855	#endif
48116	48856
48117		#ifdef __LITTLE_ENDIAN__
48118		__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
48119		uint32_t __ret;
48120		__ret = (uint32_t) __builtin_neon_vaddlv_u16(__p0);
48121		return __ret;
48122		}
48123		#else
48124		__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
48125		uint32_t __ret;
48126		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48127		__ret = (uint32_t) __builtin_neon_vaddlv_u16(__rev0);
	48857	__ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
	48858	float64x1_t __ret;
	48859	__ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
48128	48860	return __ret;
48129	48861	}
48130		#endif
48131
48132	48862	#ifdef __LITTLE_ENDIAN__
48133		__ai int16_t vaddlv_s8(int8x8_t __p0) {
48134		int16_t __ret;
48135		__ret = (int16_t) __builtin_neon_vaddlv_s8(__p0);
	48863	__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
	48864	float64_t __ret;
	48865	__ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__p0);
48136	48866	return __ret;
48137	48867	}
48138	48868	#else
48139		__ai int16_t vaddlv_s8(int8x8_t __p0) {
48140		int16_t __ret;
48141		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48142		__ret = (int16_t) __builtin_neon_vaddlv_s8(__rev0);
	48869	__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
	48870	float64_t __ret;
	48871	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	48872	__ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__rev0);
48143	48873	return __ret;
48144	48874	}
48145	48875	#endif
48146	48876
48147	48877	#ifdef __LITTLE_ENDIAN__
48148		__ai int64_t vaddlv_s32(int32x2_t __p0) {
48149		int64_t __ret;
48150		__ret = (int64_t) __builtin_neon_vaddlv_s32(__p0);
	48878	__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
	48879	float32_t __ret;
	48880	__ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__p0);
48151	48881	return __ret;
48152	48882	}
48153	48883	#else
48154		__ai int64_t vaddlv_s32(int32x2_t __p0) {
48155		int64_t __ret;
48156		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48157		__ret = (int64_t) __builtin_neon_vaddlv_s32(__rev0);
	48884	__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
	48885	float32_t __ret;
	48886	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	48887	__ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__rev0);
48158	48888	return __ret;
48159	48889	}
48160	48890	#endif
48161	48891
48162	48892	#ifdef __LITTLE_ENDIAN__
48163		__ai int32_t vaddlv_s16(int16x4_t __p0) {
48164		int32_t __ret;
48165		__ret = (int32_t) __builtin_neon_vaddlv_s16(__p0);
	48893	__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
	48894	float32_t __ret;
	48895	__ret = (float32_t) __builtin_neon_vmaxnmv_f32(__p0);
48166	48896	return __ret;
48167	48897	}
48168	48898	#else
48169		__ai int32_t vaddlv_s16(int16x4_t __p0) {
48170		int32_t __ret;
48171		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48172		__ret = (int32_t) __builtin_neon_vaddlv_s16(__rev0);
	48899	__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
	48900	float32_t __ret;
	48901	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	48902	__ret = (float32_t) __builtin_neon_vmaxnmv_f32(__rev0);
48173	48903	return __ret;
48174	48904	}
48175	48905	#endif
48176	48906
48177	48907	#ifdef __LITTLE_ENDIAN__
48178		__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
	48908	__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
48179	48909	uint8_t __ret;
48180		__ret = (uint8_t) __builtin_neon_vaddvq_u8(__p0);
	48910	__ret = (uint8_t) __builtin_neon_vmaxvq_u8(__p0);
48181	48911	return __ret;
48182	48912	}
48183	48913	#else
48184		__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
	48914	__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
48185	48915	uint8_t __ret;
48186	48916	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48187		__ret = (uint8_t) __builtin_neon_vaddvq_u8(__rev0);
	48917	__ret = (uint8_t) __builtin_neon_vmaxvq_u8(__rev0);
48188	48918	return __ret;
48189	48919	}
48190	48920	#endif
48191	48921
48192	48922	#ifdef __LITTLE_ENDIAN__
48193		__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
	48923	__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
48194	48924	uint32_t __ret;
48195		__ret = (uint32_t) __builtin_neon_vaddvq_u32(__p0);
	48925	__ret = (uint32_t) __builtin_neon_vmaxvq_u32(__p0);
48196	48926	return __ret;
48197	48927	}
48198	48928	#else
48199		__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
	48929	__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
48200	48930	uint32_t __ret;
48201	48931	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48202		__ret = (uint32_t) __builtin_neon_vaddvq_u32(__rev0);
48203		return __ret;
48204		}
48205		#endif
48206
48207		#ifdef __LITTLE_ENDIAN__
48208		__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
48209		uint64_t __ret;
48210		__ret = (uint64_t) __builtin_neon_vaddvq_u64(__p0);
48211		return __ret;
48212		}
48213		#else
48214		__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
48215		uint64_t __ret;
48216		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48217		__ret = (uint64_t) __builtin_neon_vaddvq_u64(__rev0);
	48932	__ret = (uint32_t) __builtin_neon_vmaxvq_u32(__rev0);
48218	48933	return __ret;
48219	48934	}
48220	48935	#endif
48221	48936
48222	48937	#ifdef __LITTLE_ENDIAN__
48223		__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
	48938	__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
48224	48939	uint16_t __ret;
48225		__ret = (uint16_t) __builtin_neon_vaddvq_u16(__p0);
	48940	__ret = (uint16_t) __builtin_neon_vmaxvq_u16(__p0);
48226	48941	return __ret;
48227	48942	}
48228	48943	#else
48229		__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
	48944	__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
48230	48945	uint16_t __ret;
48231	48946	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48232		__ret = (uint16_t) __builtin_neon_vaddvq_u16(__rev0);
	48947	__ret = (uint16_t) __builtin_neon_vmaxvq_u16(__rev0);
48233	48948	return __ret;
48234	48949	}
48235	48950	#endif
48236	48951
48237	48952	#ifdef __LITTLE_ENDIAN__
48238		__ai int8_t vaddvq_s8(int8x16_t __p0) {
	48953	__ai int8_t vmaxvq_s8(int8x16_t __p0) {
48239	48954	int8_t __ret;
48240		__ret = (int8_t) __builtin_neon_vaddvq_s8(__p0);
	48955	__ret = (int8_t) __builtin_neon_vmaxvq_s8(__p0);
48241	48956	return __ret;
48242	48957	}
48243	48958	#else
48244		__ai int8_t vaddvq_s8(int8x16_t __p0) {
	48959	__ai int8_t vmaxvq_s8(int8x16_t __p0) {
48245	48960	int8_t __ret;
48246	48961	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48247		__ret = (int8_t) __builtin_neon_vaddvq_s8(__rev0);
	48962	__ret = (int8_t) __builtin_neon_vmaxvq_s8(__rev0);
48248	48963	return __ret;
48249	48964	}
48250	48965	#endif
48251	48966
48252	48967	#ifdef __LITTLE_ENDIAN__
48253		__ai float64_t vaddvq_f64(float64x2_t __p0) {
	48968	__ai float64_t vmaxvq_f64(float64x2_t __p0) {
48254	48969	float64_t __ret;
48255		__ret = (float64_t) __builtin_neon_vaddvq_f64(__p0);
	48970	__ret = (float64_t) __builtin_neon_vmaxvq_f64(__p0);
48256	48971	return __ret;
48257	48972	}
48258	48973	#else
48259		__ai float64_t vaddvq_f64(float64x2_t __p0) {
	48974	__ai float64_t vmaxvq_f64(float64x2_t __p0) {
48260	48975	float64_t __ret;
48261	48976	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48262		__ret = (float64_t) __builtin_neon_vaddvq_f64(__rev0);
	48977	__ret = (float64_t) __builtin_neon_vmaxvq_f64(__rev0);
48263	48978	return __ret;
48264	48979	}
48265	48980	#endif
48266	48981
48267	48982	#ifdef __LITTLE_ENDIAN__
48268		__ai float32_t vaddvq_f32(float32x4_t __p0) {
	48983	__ai float32_t vmaxvq_f32(float32x4_t __p0) {
48269	48984	float32_t __ret;
48270		__ret = (float32_t) __builtin_neon_vaddvq_f32(__p0);
	48985	__ret = (float32_t) __builtin_neon_vmaxvq_f32(__p0);
48271	48986	return __ret;
48272	48987	}
48273	48988	#else
48274		__ai float32_t vaddvq_f32(float32x4_t __p0) {
	48989	__ai float32_t vmaxvq_f32(float32x4_t __p0) {
48275	48990	float32_t __ret;
48276	48991	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48277		__ret = (float32_t) __builtin_neon_vaddvq_f32(__rev0);
	48992	__ret = (float32_t) __builtin_neon_vmaxvq_f32(__rev0);
48278	48993	return __ret;
48279	48994	}
48280	48995	#endif
48281	48996
48282	48997	#ifdef __LITTLE_ENDIAN__
48283		__ai int32_t vaddvq_s32(int32x4_t __p0) {
	48998	__ai int32_t vmaxvq_s32(int32x4_t __p0) {
48284	48999	int32_t __ret;
48285		__ret = (int32_t) __builtin_neon_vaddvq_s32(__p0);
	49000	__ret = (int32_t) __builtin_neon_vmaxvq_s32(__p0);
48286	49001	return __ret;
48287	49002	}
48288	49003	#else
48289		__ai int32_t vaddvq_s32(int32x4_t __p0) {
	49004	__ai int32_t vmaxvq_s32(int32x4_t __p0) {
48290	49005	int32_t __ret;
48291	49006	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48292		__ret = (int32_t) __builtin_neon_vaddvq_s32(__rev0);
48293		return __ret;
48294		}
48295		#endif
48296
48297		#ifdef __LITTLE_ENDIAN__
48298		__ai int64_t vaddvq_s64(int64x2_t __p0) {
48299		int64_t __ret;
48300		__ret = (int64_t) __builtin_neon_vaddvq_s64(__p0);
48301		return __ret;
48302		}
48303		#else
48304		__ai int64_t vaddvq_s64(int64x2_t __p0) {
48305		int64_t __ret;
48306		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48307		__ret = (int64_t) __builtin_neon_vaddvq_s64(__rev0);
	49007	__ret = (int32_t) __builtin_neon_vmaxvq_s32(__rev0);
48308	49008	return __ret;
48309	49009	}
48310	49010	#endif
48311	49011
48312	49012	#ifdef __LITTLE_ENDIAN__
48313		__ai int16_t vaddvq_s16(int16x8_t __p0) {
	49013	__ai int16_t vmaxvq_s16(int16x8_t __p0) {
48314	49014	int16_t __ret;
48315		__ret = (int16_t) __builtin_neon_vaddvq_s16(__p0);
	49015	__ret = (int16_t) __builtin_neon_vmaxvq_s16(__p0);
48316	49016	return __ret;
48317	49017	}
48318	49018	#else
48319		__ai int16_t vaddvq_s16(int16x8_t __p0) {
	49019	__ai int16_t vmaxvq_s16(int16x8_t __p0) {
48320	49020	int16_t __ret;
48321	49021	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48322		__ret = (int16_t) __builtin_neon_vaddvq_s16(__rev0);
	49022	__ret = (int16_t) __builtin_neon_vmaxvq_s16(__rev0);
48323	49023	return __ret;
48324	49024	}
48325	49025	#endif
48326	49026
48327	49027	#ifdef __LITTLE_ENDIAN__
48328		__ai uint8_t vaddv_u8(uint8x8_t __p0) {
	49028	__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
48329	49029	uint8_t __ret;
48330		__ret = (uint8_t) __builtin_neon_vaddv_u8(__p0);
	49030	__ret = (uint8_t) __builtin_neon_vmaxv_u8(__p0);
48331	49031	return __ret;
48332	49032	}
48333	49033	#else
48334		__ai uint8_t vaddv_u8(uint8x8_t __p0) {
	49034	__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
48335	49035	uint8_t __ret;
48336	49036	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48337		__ret = (uint8_t) __builtin_neon_vaddv_u8(__rev0);
	49037	__ret = (uint8_t) __builtin_neon_vmaxv_u8(__rev0);
48338	49038	return __ret;
48339	49039	}
48340	49040	#endif
48341	49041
48342	49042	#ifdef __LITTLE_ENDIAN__
48343		__ai uint32_t vaddv_u32(uint32x2_t __p0) {
	49043	__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
48344	49044	uint32_t __ret;
48345		__ret = (uint32_t) __builtin_neon_vaddv_u32(__p0);
	49045	__ret = (uint32_t) __builtin_neon_vmaxv_u32(__p0);
48346	49046	return __ret;
48347	49047	}
48348	49048	#else
48349		__ai uint32_t vaddv_u32(uint32x2_t __p0) {
	49049	__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
48350	49050	uint32_t __ret;
48351	49051	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48352		__ret = (uint32_t) __builtin_neon_vaddv_u32(__rev0);
	49052	__ret = (uint32_t) __builtin_neon_vmaxv_u32(__rev0);
48353	49053	return __ret;
48354	49054	}
48355	49055	#endif
48356	49056
48357	49057	#ifdef __LITTLE_ENDIAN__
48358		__ai uint16_t vaddv_u16(uint16x4_t __p0) {
	49058	__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
48359	49059	uint16_t __ret;
48360		__ret = (uint16_t) __builtin_neon_vaddv_u16(__p0);
	49060	__ret = (uint16_t) __builtin_neon_vmaxv_u16(__p0);
48361	49061	return __ret;
48362	49062	}
48363	49063	#else
48364		__ai uint16_t vaddv_u16(uint16x4_t __p0) {
	49064	__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
48365	49065	uint16_t __ret;
48366	49066	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48367		__ret = (uint16_t) __builtin_neon_vaddv_u16(__rev0);
	49067	__ret = (uint16_t) __builtin_neon_vmaxv_u16(__rev0);
48368	49068	return __ret;
48369	49069	}
48370	49070	#endif
48371	49071
48372	49072	#ifdef __LITTLE_ENDIAN__
48373		__ai int8_t vaddv_s8(int8x8_t __p0) {
	49073	__ai int8_t vmaxv_s8(int8x8_t __p0) {
48374	49074	int8_t __ret;
48375		__ret = (int8_t) __builtin_neon_vaddv_s8(__p0);
	49075	__ret = (int8_t) __builtin_neon_vmaxv_s8(__p0);
48376	49076	return __ret;
48377	49077	}
48378	49078	#else
48379		__ai int8_t vaddv_s8(int8x8_t __p0) {
	49079	__ai int8_t vmaxv_s8(int8x8_t __p0) {
48380	49080	int8_t __ret;
48381	49081	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48382		__ret = (int8_t) __builtin_neon_vaddv_s8(__rev0);
	49082	__ret = (int8_t) __builtin_neon_vmaxv_s8(__rev0);
48383	49083	return __ret;
48384	49084	}
48385	49085	#endif
48386	49086
48387	49087	#ifdef __LITTLE_ENDIAN__
48388		__ai float32_t vaddv_f32(float32x2_t __p0) {
	49088	__ai float32_t vmaxv_f32(float32x2_t __p0) {
48389	49089	float32_t __ret;
48390		__ret = (float32_t) __builtin_neon_vaddv_f32(__p0);
	49090	__ret = (float32_t) __builtin_neon_vmaxv_f32(__p0);
48391	49091	return __ret;
48392	49092	}
48393	49093	#else
48394		__ai float32_t vaddv_f32(float32x2_t __p0) {
	49094	__ai float32_t vmaxv_f32(float32x2_t __p0) {
48395	49095	float32_t __ret;
48396	49096	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48397		__ret = (float32_t) __builtin_neon_vaddv_f32(__rev0);
	49097	__ret = (float32_t) __builtin_neon_vmaxv_f32(__rev0);
48398	49098	return __ret;
48399	49099	}
48400	49100	#endif
48401	49101
48402	49102	#ifdef __LITTLE_ENDIAN__
48403		__ai int32_t vaddv_s32(int32x2_t __p0) {
	49103	__ai int32_t vmaxv_s32(int32x2_t __p0) {
48404	49104	int32_t __ret;
48405		__ret = (int32_t) __builtin_neon_vaddv_s32(__p0);
	49105	__ret = (int32_t) __builtin_neon_vmaxv_s32(__p0);
48406	49106	return __ret;
48407	49107	}
48408	49108	#else
48409		__ai int32_t vaddv_s32(int32x2_t __p0) {
	49109	__ai int32_t vmaxv_s32(int32x2_t __p0) {
48410	49110	int32_t __ret;
48411	49111	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48412		__ret = (int32_t) __builtin_neon_vaddv_s32(__rev0);
	49112	__ret = (int32_t) __builtin_neon_vmaxv_s32(__rev0);
48413	49113	return __ret;
48414	49114	}
48415	49115	#endif
48416	49116
48417	49117	#ifdef __LITTLE_ENDIAN__
48418		__ai int16_t vaddv_s16(int16x4_t __p0) {
	49118	__ai int16_t vmaxv_s16(int16x4_t __p0) {
48419	49119	int16_t __ret;
48420		__ret = (int16_t) __builtin_neon_vaddv_s16(__p0);
	49120	__ret = (int16_t) __builtin_neon_vmaxv_s16(__p0);
48421	49121	return __ret;
48422	49122	}
48423	49123	#else
48424		__ai int16_t vaddv_s16(int16x4_t __p0) {
	49124	__ai int16_t vmaxv_s16(int16x4_t __p0) {
48425	49125	int16_t __ret;
48426	49126	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48427		__ret = (int16_t) __builtin_neon_vaddv_s16(__rev0);
48428		return __ret;
48429		}
48430		#endif
48431
48432		__ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
48433		poly64x1_t __ret;
48434		__ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
48435		return __ret;
48436		}
48437		#ifdef __LITTLE_ENDIAN__
48438		__ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
48439		poly64x2_t __ret;
48440		__ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 38);
48441		return __ret;
48442		}
48443		#else
48444		__ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
48445		poly64x2_t __ret;
48446		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48447		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48448		poly64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
48449		__ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 38);
48450		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49127	__ret = (int16_t) __builtin_neon_vmaxv_s16(__rev0);
48451	49128	return __ret;
48452	49129	}
48453	49130	#endif
48454	49131
48455	49132	#ifdef __LITTLE_ENDIAN__
48456		__ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
48457		float64x2_t __ret;
48458		__ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
48459		return __ret;
48460		}
48461		#else
48462		__ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	49133	__ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
48463	49134	float64x2_t __ret;
48464		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48465		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48466		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
48467		__ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
48468		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48469		return __ret;
48470		}
48471		#endif
48472
48473		__ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
48474		float64x1_t __ret;
48475		__ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
48476		return __ret;
48477		}
48478		#ifdef __LITTLE_ENDIAN__
48479		__ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
48480		uint64x2_t __ret;
48481		__ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
48482		return __ret;
48483		}
48484		#else
48485		__ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
48486		uint64x2_t __ret;
48487		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48488		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48489		__ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
48490		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48491		return __ret;
48492		}
48493		#endif
48494
48495		__ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
48496		uint64x1_t __ret;
48497		__ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
48498		return __ret;
48499		}
48500		__ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
48501		uint64_t __ret;
48502		__ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
48503		return __ret;
48504		}
48505		__ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
48506		uint32_t __ret;
48507		__ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
48508		return __ret;
48509		}
48510		#ifdef __LITTLE_ENDIAN__
48511		__ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
48512		uint64x2_t __ret;
48513		__ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
48514		return __ret;
48515		}
48516		#else
48517		__ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
48518		uint64x2_t __ret;
48519		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48520		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48521		__ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
48522		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48523		return __ret;
48524		}
48525		#endif
48526
48527		__ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
48528		uint64x1_t __ret;
48529		__ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
48530		return __ret;
48531		}
48532		__ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
48533		uint64_t __ret;
48534		__ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
48535		return __ret;
48536		}
48537		__ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
48538		uint32_t __ret;
48539		__ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
48540		return __ret;
48541		}
48542		#ifdef __LITTLE_ENDIAN__
48543		__ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
48544		uint64x2_t __ret;
48545		__ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
48546		return __ret;
48547		}
48548		#else
48549		__ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
48550		uint64x2_t __ret;
48551		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48552		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48553		__ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
48554		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48555		return __ret;
48556		}
48557		#endif
48558
48559		__ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
48560		uint64x1_t __ret;
48561		__ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
48562		return __ret;
48563		}
48564		__ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
48565		uint64_t __ret;
48566		__ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
48567		return __ret;
48568		}
48569		__ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
48570		uint32_t __ret;
48571		__ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
48572		return __ret;
48573		}
48574		#ifdef __LITTLE_ENDIAN__
48575		__ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
48576		uint64x2_t __ret;
48577		__ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
48578		return __ret;
48579		}
48580		#else
48581		__ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
48582		uint64x2_t __ret;
48583		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48584		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48585		__ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
48586		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48587		return __ret;
48588		}
48589		#endif
48590
48591		__ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
48592		uint64x1_t __ret;
48593		__ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
48594		return __ret;
48595		}
48596		__ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
48597		uint64_t __ret;
48598		__ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
48599		return __ret;
48600		}
48601		__ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
48602		uint32_t __ret;
48603		__ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
48604		return __ret;
48605		}
48606		__ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
48607		uint64x1_t __ret;
48608		__ret = (uint64x1_t)(__p0 == __p1);
48609		return __ret;
48610		}
48611		#ifdef __LITTLE_ENDIAN__
48612		__ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
48613		uint64x2_t __ret;
48614		__ret = (uint64x2_t)(__p0 == __p1);
48615		return __ret;
48616		}
48617		#else
48618		__ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
48619		uint64x2_t __ret;
48620		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48621		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48622		__ret = (uint64x2_t)(__rev0 == __rev1);
48623		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48624		return __ret;
48625		}
48626		#endif
48627
48628		#ifdef __LITTLE_ENDIAN__
48629		__ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
48630		uint64x2_t __ret;
48631		__ret = (uint64x2_t)(__p0 == __p1);
48632		return __ret;
48633		}
48634		#else
48635		__ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
48636		uint64x2_t __ret;
48637		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48638		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48639		__ret = (uint64x2_t)(__rev0 == __rev1);
48640		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48641		return __ret;
48642		}
48643		#endif
48644
48645		#ifdef __LITTLE_ENDIAN__
48646		__ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
48647		uint64x2_t __ret;
48648		__ret = (uint64x2_t)(__p0 == __p1);
48649		return __ret;
48650		}
48651		#else
48652		__ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
48653		uint64x2_t __ret;
48654		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48655		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48656		__ret = (uint64x2_t)(__rev0 == __rev1);
48657		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48658		return __ret;
48659		}
48660		#endif
48661
48662		#ifdef __LITTLE_ENDIAN__
48663		__ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
48664		uint64x2_t __ret;
48665		__ret = (uint64x2_t)(__p0 == __p1);
48666		return __ret;
48667		}
48668		#else
48669		__ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
48670		uint64x2_t __ret;
48671		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48672		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48673		__ret = (uint64x2_t)(__rev0 == __rev1);
48674		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48675		return __ret;
48676		}
48677		#endif
48678
48679		__ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
48680		uint64x1_t __ret;
48681		__ret = (uint64x1_t)(__p0 == __p1);
48682		return __ret;
48683		}
48684		__ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
48685		uint64x1_t __ret;
48686		__ret = (uint64x1_t)(__p0 == __p1);
48687		return __ret;
48688		}
48689		__ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
48690		uint64x1_t __ret;
48691		__ret = (uint64x1_t)(__p0 == __p1);
48692		return __ret;
48693		}
48694		__ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
48695		uint64_t __ret;
48696		__ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
48697		return __ret;
48698		}
48699		__ai uint64_t vceqd_s64(int64_t __p0, int64_t __p1) {
48700		uint64_t __ret;
48701		__ret = (uint64_t) __builtin_neon_vceqd_s64(__p0, __p1);
48702		return __ret;
48703		}
48704		__ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
48705		uint64_t __ret;
48706		__ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
48707		return __ret;
48708		}
48709		__ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
48710		uint32_t __ret;
48711		__ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
48712		return __ret;
48713		}
48714		#ifdef __LITTLE_ENDIAN__
48715		__ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
48716		uint8x8_t __ret;
48717		__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 16);
	49135	__ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
48718	49136	return __ret;
48719		}
48720		#else
48721		__ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
48722		uint8x8_t __ret;
48723		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48724		__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
48725		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	49137	}
	49138	#else
	49139	__ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
	49140	float64x2_t __ret;
	49141	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	49142	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	49143	__ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	49144	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48726	49145	return __ret;
48727	49146	}
48728	49147	#endif
48729	49148
48730		__ai uint64x1_t vceqz_p64(poly64x1_t __p0) {
48731		uint64x1_t __ret;
48732		__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
	49149	__ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
	49150	float64x1_t __ret;
	49151	__ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
48733	49152	return __ret;
48734	49153	}
48735	49154	#ifdef __LITTLE_ENDIAN__
48736		__ai uint8x16_t vceqzq_p8(poly8x16_t __p0) {
48737		uint8x16_t __ret;
48738		__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 48);
	49155	__ai float64_t vminnmvq_f64(float64x2_t __p0) {
	49156	float64_t __ret;
	49157	__ret = (float64_t) __builtin_neon_vminnmvq_f64(__p0);
48739	49158	return __ret;
48740	49159	}
48741	49160	#else
48742		__ai uint8x16_t vceqzq_p8(poly8x16_t __p0) {
48743		uint8x16_t __ret;
48744		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48745		__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
48746		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	49161	__ai float64_t vminnmvq_f64(float64x2_t __p0) {
	49162	float64_t __ret;
	49163	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	49164	__ret = (float64_t) __builtin_neon_vminnmvq_f64(__rev0);
48747	49165	return __ret;
48748	49166	}
48749	49167	#endif
48750	49168
48751	49169	#ifdef __LITTLE_ENDIAN__
48752		__ai uint64x2_t vceqzq_p64(poly64x2_t __p0) {
48753		uint64x2_t __ret;
48754		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
	49170	__ai float32_t vminnmvq_f32(float32x4_t __p0) {
	49171	float32_t __ret;
	49172	__ret = (float32_t) __builtin_neon_vminnmvq_f32(__p0);
48755	49173	return __ret;
48756	49174	}
48757	49175	#else
48758		__ai uint64x2_t vceqzq_p64(poly64x2_t __p0) {
48759		uint64x2_t __ret;
48760		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48761		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
48762		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49176	__ai float32_t vminnmvq_f32(float32x4_t __p0) {
	49177	float32_t __ret;
	49178	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	49179	__ret = (float32_t) __builtin_neon_vminnmvq_f32(__rev0);
48763	49180	return __ret;
48764	49181	}
48765	49182	#endif
48766	49183
48767	49184	#ifdef __LITTLE_ENDIAN__
48768		__ai uint8x16_t vceqzq_u8(uint8x16_t __p0) {
48769		uint8x16_t __ret;
48770		__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 48);
	49185	__ai float32_t vminnmv_f32(float32x2_t __p0) {
	49186	float32_t __ret;
	49187	__ret = (float32_t) __builtin_neon_vminnmv_f32(__p0);
48771	49188	return __ret;
48772	49189	}
48773	49190	#else
48774		__ai uint8x16_t vceqzq_u8(uint8x16_t __p0) {
48775		uint8x16_t __ret;
48776		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48777		__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
48778		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	49191	__ai float32_t vminnmv_f32(float32x2_t __p0) {
	49192	float32_t __ret;
	49193	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	49194	__ret = (float32_t) __builtin_neon_vminnmv_f32(__rev0);
48779	49195	return __ret;
48780	49196	}
48781	49197	#endif
48782	49198
48783	49199	#ifdef __LITTLE_ENDIAN__
48784		__ai uint32x4_t vceqzq_u32(uint32x4_t __p0) {
48785		uint32x4_t __ret;
48786		__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 50);
	49200	__ai uint8_t vminvq_u8(uint8x16_t __p0) {
	49201	uint8_t __ret;
	49202	__ret = (uint8_t) __builtin_neon_vminvq_u8(__p0);
48787	49203	return __ret;
48788	49204	}
48789	49205	#else
48790		__ai uint32x4_t vceqzq_u32(uint32x4_t __p0) {
48791		uint32x4_t __ret;
48792		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48793		__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
48794		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	49206	__ai uint8_t vminvq_u8(uint8x16_t __p0) {
	49207	uint8_t __ret;
	49208	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	49209	__ret = (uint8_t) __builtin_neon_vminvq_u8(__rev0);
48795	49210	return __ret;
48796	49211	}
48797	49212	#endif
48798	49213
48799	49214	#ifdef __LITTLE_ENDIAN__
48800		__ai uint64x2_t vceqzq_u64(uint64x2_t __p0) {
48801		uint64x2_t __ret;
48802		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
	49215	__ai uint32_t vminvq_u32(uint32x4_t __p0) {
	49216	uint32_t __ret;
	49217	__ret = (uint32_t) __builtin_neon_vminvq_u32(__p0);
48803	49218	return __ret;
48804	49219	}
48805	49220	#else
48806		__ai uint64x2_t vceqzq_u64(uint64x2_t __p0) {
48807		uint64x2_t __ret;
48808		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48809		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
48810		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49221	__ai uint32_t vminvq_u32(uint32x4_t __p0) {
	49222	uint32_t __ret;
	49223	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	49224	__ret = (uint32_t) __builtin_neon_vminvq_u32(__rev0);
48811	49225	return __ret;
48812	49226	}
48813	49227	#endif
48814	49228
48815	49229	#ifdef __LITTLE_ENDIAN__
48816		__ai uint16x8_t vceqzq_u16(uint16x8_t __p0) {
48817		uint16x8_t __ret;
48818		__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 49);
	49230	__ai uint16_t vminvq_u16(uint16x8_t __p0) {
	49231	uint16_t __ret;
	49232	__ret = (uint16_t) __builtin_neon_vminvq_u16(__p0);
48819	49233	return __ret;
48820	49234	}
48821	49235	#else
48822		__ai uint16x8_t vceqzq_u16(uint16x8_t __p0) {
48823		uint16x8_t __ret;
	49236	__ai uint16_t vminvq_u16(uint16x8_t __p0) {
	49237	uint16_t __ret;
48824	49238	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48825		__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
48826		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	49239	__ret = (uint16_t) __builtin_neon_vminvq_u16(__rev0);
48827	49240	return __ret;
48828	49241	}
48829	49242	#endif
48830	49243
48831	49244	#ifdef __LITTLE_ENDIAN__
48832		__ai uint8x16_t vceqzq_s8(int8x16_t __p0) {
48833		uint8x16_t __ret;
48834		__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 48);
	49245	__ai int8_t vminvq_s8(int8x16_t __p0) {
	49246	int8_t __ret;
	49247	__ret = (int8_t) __builtin_neon_vminvq_s8(__p0);
48835	49248	return __ret;
48836	49249	}
48837	49250	#else
48838		__ai uint8x16_t vceqzq_s8(int8x16_t __p0) {
48839		uint8x16_t __ret;
	49251	__ai int8_t vminvq_s8(int8x16_t __p0) {
	49252	int8_t __ret;
48840	49253	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48841		__ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
48842		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	49254	__ret = (int8_t) __builtin_neon_vminvq_s8(__rev0);
48843	49255	return __ret;
48844	49256	}
48845	49257	#endif
48846	49258
48847	49259	#ifdef __LITTLE_ENDIAN__
48848		__ai uint64x2_t vceqzq_f64(float64x2_t __p0) {
48849		uint64x2_t __ret;
48850		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
	49260	__ai float64_t vminvq_f64(float64x2_t __p0) {
	49261	float64_t __ret;
	49262	__ret = (float64_t) __builtin_neon_vminvq_f64(__p0);
48851	49263	return __ret;
48852	49264	}
48853	49265	#else
48854		__ai uint64x2_t vceqzq_f64(float64x2_t __p0) {
48855		uint64x2_t __ret;
	49266	__ai float64_t vminvq_f64(float64x2_t __p0) {
	49267	float64_t __ret;
48856	49268	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48857		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
48858		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49269	__ret = (float64_t) __builtin_neon_vminvq_f64(__rev0);
48859	49270	return __ret;
48860	49271	}
48861	49272	#endif
48862	49273
48863	49274	#ifdef __LITTLE_ENDIAN__
48864		__ai uint32x4_t vceqzq_f32(float32x4_t __p0) {
48865		uint32x4_t __ret;
48866		__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 50);
	49275	__ai float32_t vminvq_f32(float32x4_t __p0) {
	49276	float32_t __ret;
	49277	__ret = (float32_t) __builtin_neon_vminvq_f32(__p0);
48867	49278	return __ret;
48868	49279	}
48869	49280	#else
48870		__ai uint32x4_t vceqzq_f32(float32x4_t __p0) {
48871		uint32x4_t __ret;
	49281	__ai float32_t vminvq_f32(float32x4_t __p0) {
	49282	float32_t __ret;
48872	49283	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48873		__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
48874		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	49284	__ret = (float32_t) __builtin_neon_vminvq_f32(__rev0);
48875	49285	return __ret;
48876	49286	}
48877	49287	#endif
48878	49288
48879	49289	#ifdef __LITTLE_ENDIAN__
48880		__ai uint32x4_t vceqzq_s32(int32x4_t __p0) {
48881		uint32x4_t __ret;
48882		__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 50);
	49290	__ai int32_t vminvq_s32(int32x4_t __p0) {
	49291	int32_t __ret;
	49292	__ret = (int32_t) __builtin_neon_vminvq_s32(__p0);
48883	49293	return __ret;
48884	49294	}
48885	49295	#else
48886		__ai uint32x4_t vceqzq_s32(int32x4_t __p0) {
48887		uint32x4_t __ret;
	49296	__ai int32_t vminvq_s32(int32x4_t __p0) {
	49297	int32_t __ret;
48888	49298	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48889		__ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
48890		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
48891		return __ret;
48892		}
48893		#endif
48894
48895		#ifdef __LITTLE_ENDIAN__
48896		__ai uint64x2_t vceqzq_s64(int64x2_t __p0) {
48897		uint64x2_t __ret;
48898		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 51);
48899		return __ret;
48900		}
48901		#else
48902		__ai uint64x2_t vceqzq_s64(int64x2_t __p0) {
48903		uint64x2_t __ret;
48904		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48905		__ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
48906		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49299	__ret = (int32_t) __builtin_neon_vminvq_s32(__rev0);
48907	49300	return __ret;
48908	49301	}
48909	49302	#endif
48910	49303
48911	49304	#ifdef __LITTLE_ENDIAN__
48912		__ai uint16x8_t vceqzq_s16(int16x8_t __p0) {
48913		uint16x8_t __ret;
48914		__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__p0, 49);
	49305	__ai int16_t vminvq_s16(int16x8_t __p0) {
	49306	int16_t __ret;
	49307	__ret = (int16_t) __builtin_neon_vminvq_s16(__p0);
48915	49308	return __ret;
48916	49309	}
48917	49310	#else
48918		__ai uint16x8_t vceqzq_s16(int16x8_t __p0) {
48919		uint16x8_t __ret;
	49311	__ai int16_t vminvq_s16(int16x8_t __p0) {
	49312	int16_t __ret;
48920	49313	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48921		__ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
48922		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	49314	__ret = (int16_t) __builtin_neon_vminvq_s16(__rev0);
48923	49315	return __ret;
48924	49316	}
48925	49317	#endif
48926	49318
48927	49319	#ifdef __LITTLE_ENDIAN__
48928		__ai uint8x8_t vceqz_u8(uint8x8_t __p0) {
48929		uint8x8_t __ret;
48930		__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 16);
	49320	__ai uint8_t vminv_u8(uint8x8_t __p0) {
	49321	uint8_t __ret;
	49322	__ret = (uint8_t) __builtin_neon_vminv_u8(__p0);
48931	49323	return __ret;
48932	49324	}
48933	49325	#else
48934		__ai uint8x8_t vceqz_u8(uint8x8_t __p0) {
48935		uint8x8_t __ret;
	49326	__ai uint8_t vminv_u8(uint8x8_t __p0) {
	49327	uint8_t __ret;
48936	49328	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48937		__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
48938		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	49329	__ret = (uint8_t) __builtin_neon_vminv_u8(__rev0);
48939	49330	return __ret;
48940	49331	}
48941	49332	#endif
48942	49333
48943	49334	#ifdef __LITTLE_ENDIAN__
48944		__ai uint32x2_t vceqz_u32(uint32x2_t __p0) {
48945		uint32x2_t __ret;
48946		__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 18);
	49335	__ai uint32_t vminv_u32(uint32x2_t __p0) {
	49336	uint32_t __ret;
	49337	__ret = (uint32_t) __builtin_neon_vminv_u32(__p0);
48947	49338	return __ret;
48948	49339	}
48949	49340	#else
48950		__ai uint32x2_t vceqz_u32(uint32x2_t __p0) {
48951		uint32x2_t __ret;
	49341	__ai uint32_t vminv_u32(uint32x2_t __p0) {
	49342	uint32_t __ret;
48952	49343	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48953		__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
48954		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49344	__ret = (uint32_t) __builtin_neon_vminv_u32(__rev0);
48955	49345	return __ret;
48956	49346	}
48957	49347	#endif
48958	49348
48959		__ai uint64x1_t vceqz_u64(uint64x1_t __p0) {
48960		uint64x1_t __ret;
48961		__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
48962		return __ret;
48963		}
48964	49349	#ifdef __LITTLE_ENDIAN__
48965		__ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
48966		uint16x4_t __ret;
48967		__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 17);
	49350	__ai uint16_t vminv_u16(uint16x4_t __p0) {
	49351	uint16_t __ret;
	49352	__ret = (uint16_t) __builtin_neon_vminv_u16(__p0);
48968	49353	return __ret;
48969	49354	}
48970	49355	#else
48971		__ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
48972		uint16x4_t __ret;
	49356	__ai uint16_t vminv_u16(uint16x4_t __p0) {
	49357	uint16_t __ret;
48973	49358	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48974		__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
48975		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	49359	__ret = (uint16_t) __builtin_neon_vminv_u16(__rev0);
48976	49360	return __ret;
48977	49361	}
48978	49362	#endif
48979	49363
48980	49364	#ifdef __LITTLE_ENDIAN__
48981		__ai uint8x8_t vceqz_s8(int8x8_t __p0) {
48982		uint8x8_t __ret;
48983		__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 16);
	49365	__ai int8_t vminv_s8(int8x8_t __p0) {
	49366	int8_t __ret;
	49367	__ret = (int8_t) __builtin_neon_vminv_s8(__p0);
48984	49368	return __ret;
48985	49369	}
48986	49370	#else
48987		__ai uint8x8_t vceqz_s8(int8x8_t __p0) {
48988		uint8x8_t __ret;
	49371	__ai int8_t vminv_s8(int8x8_t __p0) {
	49372	int8_t __ret;
48989	49373	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48990		__ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
48991		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	49374	__ret = (int8_t) __builtin_neon_vminv_s8(__rev0);
48992	49375	return __ret;
48993	49376	}
48994	49377	#endif
48995	49378
48996		__ai uint64x1_t vceqz_f64(float64x1_t __p0) {
48997		uint64x1_t __ret;
48998		__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
48999		return __ret;
49000		}
49001	49379	#ifdef __LITTLE_ENDIAN__
49002		__ai uint32x2_t vceqz_f32(float32x2_t __p0) {
49003		uint32x2_t __ret;
49004		__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 18);
	49380	__ai float32_t vminv_f32(float32x2_t __p0) {
	49381	float32_t __ret;
	49382	__ret = (float32_t) __builtin_neon_vminv_f32(__p0);
49005	49383	return __ret;
49006	49384	}
49007	49385	#else
49008		__ai uint32x2_t vceqz_f32(float32x2_t __p0) {
49009		uint32x2_t __ret;
	49386	__ai float32_t vminv_f32(float32x2_t __p0) {
	49387	float32_t __ret;
49010	49388	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49011		__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
49012		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49389	__ret = (float32_t) __builtin_neon_vminv_f32(__rev0);
49013	49390	return __ret;
49014	49391	}
49015	49392	#endif
49016	49393
49017	49394	#ifdef __LITTLE_ENDIAN__
49018		__ai uint32x2_t vceqz_s32(int32x2_t __p0) {
49019		uint32x2_t __ret;
49020		__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 18);
	49395	__ai int32_t vminv_s32(int32x2_t __p0) {
	49396	int32_t __ret;
	49397	__ret = (int32_t) __builtin_neon_vminv_s32(__p0);
49021	49398	return __ret;
49022	49399	}
49023	49400	#else
49024		__ai uint32x2_t vceqz_s32(int32x2_t __p0) {
49025		uint32x2_t __ret;
	49401	__ai int32_t vminv_s32(int32x2_t __p0) {
	49402	int32_t __ret;
49026	49403	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49027		__ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
49028		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	49404	__ret = (int32_t) __builtin_neon_vminv_s32(__rev0);
49029	49405	return __ret;
49030	49406	}
49031	49407	#endif
49032	49408
49033		__ai uint64x1_t vceqz_s64(int64x1_t __p0) {
49034		uint64x1_t __ret;
49035		__ret = (uint64x1_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 19);
49036		return __ret;
49037		}
49038	49409	#ifdef __LITTLE_ENDIAN__
49039		__ai uint16x4_t vceqz_s16(int16x4_t __p0) {
49040		uint16x4_t __ret;
49041		__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__p0, 17);
	49410	__ai int16_t vminv_s16(int16x4_t __p0) {
	49411	int16_t __ret;
	49412	__ret = (int16_t) __builtin_neon_vminv_s16(__p0);
49042	49413	return __ret;
49043	49414	}
49044	49415	#else
49045		__ai uint16x4_t vceqz_s16(int16x4_t __p0) {
49046		uint16x4_t __ret;
	49416	__ai int16_t vminv_s16(int16x4_t __p0) {
	49417	int16_t __ret;
49047	49418	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49048		__ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
49049		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	49419	__ret = (int16_t) __builtin_neon_vminv_s16(__rev0);
49050	49420	return __ret;
49051	49421	}
49052	49422	#endif
49053	49423
49054		__ai uint64_t vceqzd_u64(uint64_t __p0) {
49055		uint64_t __ret;
49056		__ret = (uint64_t) __builtin_neon_vceqzd_u64(__p0);
49057		return __ret;
49058		}
49059		__ai uint64_t vceqzd_s64(int64_t __p0) {
49060		uint64_t __ret;
49061		__ret = (uint64_t) __builtin_neon_vceqzd_s64(__p0);
49062		return __ret;
49063		}
49064		__ai uint64_t vceqzd_f64(float64_t __p0) {
49065		uint64_t __ret;
49066		__ret = (uint64_t) __builtin_neon_vceqzd_f64(__p0);
49067		return __ret;
49068		}
49069		__ai uint32_t vceqzs_f32(float32_t __p0) {
49070		uint32_t __ret;
49071		__ret = (uint32_t) __builtin_neon_vceqzs_f32(__p0);
49072		return __ret;
49073		}
49074	49424	#ifdef __LITTLE_ENDIAN__
49075		__ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49076		uint64x2_t __ret;
49077		__ret = (uint64x2_t)(__p0 >= __p1);
	49425	__ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	49426	float64x2_t __ret;
	49427	__ret = __p0 + __p1 * __p2;
49078	49428	return __ret;
49079	49429	}
49080	49430	#else
49081		__ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49082		uint64x2_t __ret;
49083		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49084		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49085		__ret = (uint64x2_t)(__rev0 >= __rev1);
	49431	__ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	49432	float64x2_t __ret;
	49433	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	49434	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	49435	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	49436	__ret = __rev0 + __rev1 * __rev2;
49086	49437	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49087	49438	return __ret;
49088	49439	}
49089	49440	#endif
49090	49441
49091		#ifdef __LITTLE_ENDIAN__
49092		__ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
49093		uint64x2_t __ret;
49094		__ret = (uint64x2_t)(__p0 >= __p1);
	49442	__ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	49443	float64x1_t __ret;
	49444	__ret = __p0 + __p1 * __p2;
49095	49445	return __ret;
49096	49446	}
	49447	#ifdef __LITTLE_ENDIAN__
	49448	#define vmlaq_laneq_u32(__p0_427, __p1_427, __p2_427, __p3_427) __extension__ ({ \
	49449	uint32x4_t __ret_427; \
	49450	uint32x4_t __s0_427 = __p0_427; \
	49451	uint32x4_t __s1_427 = __p1_427; \
	49452	uint32x4_t __s2_427 = __p2_427; \
	49453	__ret_427 = __s0_427 + __s1_427 * splatq_laneq_u32(__s2_427, __p3_427); \
	49454	__ret_427; \
	49455	})
49097	49456	#else
49098		__ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
49099		uint64x2_t __ret;
49100		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49101		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49102		__ret = (uint64x2_t)(__rev0 >= __rev1);
49103		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49104		return __ret;
49105		}
	49457	#define vmlaq_laneq_u32(__p0_428, __p1_428, __p2_428, __p3_428) __extension__ ({ \
	49458	uint32x4_t __ret_428; \
	49459	uint32x4_t __s0_428 = __p0_428; \
	49460	uint32x4_t __s1_428 = __p1_428; \
	49461	uint32x4_t __s2_428 = __p2_428; \
	49462	uint32x4_t __rev0_428; __rev0_428 = __builtin_shufflevector(__s0_428, __s0_428, 3, 2, 1, 0); \
	49463	uint32x4_t __rev1_428; __rev1_428 = __builtin_shufflevector(__s1_428, __s1_428, 3, 2, 1, 0); \
	49464	uint32x4_t __rev2_428; __rev2_428 = __builtin_shufflevector(__s2_428, __s2_428, 3, 2, 1, 0); \
	49465	__ret_428 = __rev0_428 + __rev1_428 * __noswap_splatq_laneq_u32(__rev2_428, __p3_428); \
	49466	__ret_428 = __builtin_shufflevector(__ret_428, __ret_428, 3, 2, 1, 0); \
	49467	__ret_428; \
	49468	})
49106	49469	#endif
49107	49470
49108	49471	#ifdef __LITTLE_ENDIAN__
49109		__ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
49110		uint64x2_t __ret;
49111		__ret = (uint64x2_t)(__p0 >= __p1);
49112		return __ret;
49113		}
	49472	#define vmlaq_laneq_u16(__p0_429, __p1_429, __p2_429, __p3_429) __extension__ ({ \
	49473	uint16x8_t __ret_429; \
	49474	uint16x8_t __s0_429 = __p0_429; \
	49475	uint16x8_t __s1_429 = __p1_429; \
	49476	uint16x8_t __s2_429 = __p2_429; \
	49477	__ret_429 = __s0_429 + __s1_429 * splatq_laneq_u16(__s2_429, __p3_429); \
	49478	__ret_429; \
	49479	})
49114	49480	#else
49115		__ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
49116		uint64x2_t __ret;
49117		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49118		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49119		__ret = (uint64x2_t)(__rev0 >= __rev1);
49120		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49121		return __ret;
49122		}
	49481	#define vmlaq_laneq_u16(__p0_430, __p1_430, __p2_430, __p3_430) __extension__ ({ \
	49482	uint16x8_t __ret_430; \
	49483	uint16x8_t __s0_430 = __p0_430; \
	49484	uint16x8_t __s1_430 = __p1_430; \
	49485	uint16x8_t __s2_430 = __p2_430; \
	49486	uint16x8_t __rev0_430; __rev0_430 = __builtin_shufflevector(__s0_430, __s0_430, 7, 6, 5, 4, 3, 2, 1, 0); \
	49487	uint16x8_t __rev1_430; __rev1_430 = __builtin_shufflevector(__s1_430, __s1_430, 7, 6, 5, 4, 3, 2, 1, 0); \
	49488	uint16x8_t __rev2_430; __rev2_430 = __builtin_shufflevector(__s2_430, __s2_430, 7, 6, 5, 4, 3, 2, 1, 0); \
	49489	__ret_430 = __rev0_430 + __rev1_430 * __noswap_splatq_laneq_u16(__rev2_430, __p3_430); \
	49490	__ret_430 = __builtin_shufflevector(__ret_430, __ret_430, 7, 6, 5, 4, 3, 2, 1, 0); \
	49491	__ret_430; \
	49492	})
	49493	#endif
	49494
	49495	#ifdef __LITTLE_ENDIAN__
	49496	#define vmlaq_laneq_f32(__p0_431, __p1_431, __p2_431, __p3_431) __extension__ ({ \
	49497	float32x4_t __ret_431; \
	49498	float32x4_t __s0_431 = __p0_431; \
	49499	float32x4_t __s1_431 = __p1_431; \
	49500	float32x4_t __s2_431 = __p2_431; \
	49501	__ret_431 = __s0_431 + __s1_431 * splatq_laneq_f32(__s2_431, __p3_431); \
	49502	__ret_431; \
	49503	})
	49504	#else
	49505	#define vmlaq_laneq_f32(__p0_432, __p1_432, __p2_432, __p3_432) __extension__ ({ \
	49506	float32x4_t __ret_432; \
	49507	float32x4_t __s0_432 = __p0_432; \
	49508	float32x4_t __s1_432 = __p1_432; \
	49509	float32x4_t __s2_432 = __p2_432; \
	49510	float32x4_t __rev0_432; __rev0_432 = __builtin_shufflevector(__s0_432, __s0_432, 3, 2, 1, 0); \
	49511	float32x4_t __rev1_432; __rev1_432 = __builtin_shufflevector(__s1_432, __s1_432, 3, 2, 1, 0); \
	49512	float32x4_t __rev2_432; __rev2_432 = __builtin_shufflevector(__s2_432, __s2_432, 3, 2, 1, 0); \
	49513	__ret_432 = __rev0_432 + __rev1_432 * __noswap_splatq_laneq_f32(__rev2_432, __p3_432); \
	49514	__ret_432 = __builtin_shufflevector(__ret_432, __ret_432, 3, 2, 1, 0); \
	49515	__ret_432; \
	49516	})
	49517	#endif
	49518
	49519	#ifdef __LITTLE_ENDIAN__
	49520	#define vmlaq_laneq_s32(__p0_433, __p1_433, __p2_433, __p3_433) __extension__ ({ \
	49521	int32x4_t __ret_433; \
	49522	int32x4_t __s0_433 = __p0_433; \
	49523	int32x4_t __s1_433 = __p1_433; \
	49524	int32x4_t __s2_433 = __p2_433; \
	49525	__ret_433 = __s0_433 + __s1_433 * splatq_laneq_s32(__s2_433, __p3_433); \
	49526	__ret_433; \
	49527	})
	49528	#else
	49529	#define vmlaq_laneq_s32(__p0_434, __p1_434, __p2_434, __p3_434) __extension__ ({ \
	49530	int32x4_t __ret_434; \
	49531	int32x4_t __s0_434 = __p0_434; \
	49532	int32x4_t __s1_434 = __p1_434; \
	49533	int32x4_t __s2_434 = __p2_434; \
	49534	int32x4_t __rev0_434; __rev0_434 = __builtin_shufflevector(__s0_434, __s0_434, 3, 2, 1, 0); \
	49535	int32x4_t __rev1_434; __rev1_434 = __builtin_shufflevector(__s1_434, __s1_434, 3, 2, 1, 0); \
	49536	int32x4_t __rev2_434; __rev2_434 = __builtin_shufflevector(__s2_434, __s2_434, 3, 2, 1, 0); \
	49537	__ret_434 = __rev0_434 + __rev1_434 * __noswap_splatq_laneq_s32(__rev2_434, __p3_434); \
	49538	__ret_434 = __builtin_shufflevector(__ret_434, __ret_434, 3, 2, 1, 0); \
	49539	__ret_434; \
	49540	})
	49541	#endif
	49542
	49543	#ifdef __LITTLE_ENDIAN__
	49544	#define vmlaq_laneq_s16(__p0_435, __p1_435, __p2_435, __p3_435) __extension__ ({ \
	49545	int16x8_t __ret_435; \
	49546	int16x8_t __s0_435 = __p0_435; \
	49547	int16x8_t __s1_435 = __p1_435; \
	49548	int16x8_t __s2_435 = __p2_435; \
	49549	__ret_435 = __s0_435 + __s1_435 * splatq_laneq_s16(__s2_435, __p3_435); \
	49550	__ret_435; \
	49551	})
	49552	#else
	49553	#define vmlaq_laneq_s16(__p0_436, __p1_436, __p2_436, __p3_436) __extension__ ({ \
	49554	int16x8_t __ret_436; \
	49555	int16x8_t __s0_436 = __p0_436; \
	49556	int16x8_t __s1_436 = __p1_436; \
	49557	int16x8_t __s2_436 = __p2_436; \
	49558	int16x8_t __rev0_436; __rev0_436 = __builtin_shufflevector(__s0_436, __s0_436, 7, 6, 5, 4, 3, 2, 1, 0); \
	49559	int16x8_t __rev1_436; __rev1_436 = __builtin_shufflevector(__s1_436, __s1_436, 7, 6, 5, 4, 3, 2, 1, 0); \
	49560	int16x8_t __rev2_436; __rev2_436 = __builtin_shufflevector(__s2_436, __s2_436, 7, 6, 5, 4, 3, 2, 1, 0); \
	49561	__ret_436 = __rev0_436 + __rev1_436 * __noswap_splatq_laneq_s16(__rev2_436, __p3_436); \
	49562	__ret_436 = __builtin_shufflevector(__ret_436, __ret_436, 7, 6, 5, 4, 3, 2, 1, 0); \
	49563	__ret_436; \
	49564	})
	49565	#endif
	49566
	49567	#ifdef __LITTLE_ENDIAN__
	49568	#define vmla_laneq_u32(__p0_437, __p1_437, __p2_437, __p3_437) __extension__ ({ \
	49569	uint32x2_t __ret_437; \
	49570	uint32x2_t __s0_437 = __p0_437; \
	49571	uint32x2_t __s1_437 = __p1_437; \
	49572	uint32x4_t __s2_437 = __p2_437; \
	49573	__ret_437 = __s0_437 + __s1_437 * splat_laneq_u32(__s2_437, __p3_437); \
	49574	__ret_437; \
	49575	})
	49576	#else
	49577	#define vmla_laneq_u32(__p0_438, __p1_438, __p2_438, __p3_438) __extension__ ({ \
	49578	uint32x2_t __ret_438; \
	49579	uint32x2_t __s0_438 = __p0_438; \
	49580	uint32x2_t __s1_438 = __p1_438; \
	49581	uint32x4_t __s2_438 = __p2_438; \
	49582	uint32x2_t __rev0_438; __rev0_438 = __builtin_shufflevector(__s0_438, __s0_438, 1, 0); \
	49583	uint32x2_t __rev1_438; __rev1_438 = __builtin_shufflevector(__s1_438, __s1_438, 1, 0); \
	49584	uint32x4_t __rev2_438; __rev2_438 = __builtin_shufflevector(__s2_438, __s2_438, 3, 2, 1, 0); \
	49585	__ret_438 = __rev0_438 + __rev1_438 * __noswap_splat_laneq_u32(__rev2_438, __p3_438); \
	49586	__ret_438 = __builtin_shufflevector(__ret_438, __ret_438, 1, 0); \
	49587	__ret_438; \
	49588	})
	49589	#endif
	49590
	49591	#ifdef __LITTLE_ENDIAN__
	49592	#define vmla_laneq_u16(__p0_439, __p1_439, __p2_439, __p3_439) __extension__ ({ \
	49593	uint16x4_t __ret_439; \
	49594	uint16x4_t __s0_439 = __p0_439; \
	49595	uint16x4_t __s1_439 = __p1_439; \
	49596	uint16x8_t __s2_439 = __p2_439; \
	49597	__ret_439 = __s0_439 + __s1_439 * splat_laneq_u16(__s2_439, __p3_439); \
	49598	__ret_439; \
	49599	})
	49600	#else
	49601	#define vmla_laneq_u16(__p0_440, __p1_440, __p2_440, __p3_440) __extension__ ({ \
	49602	uint16x4_t __ret_440; \
	49603	uint16x4_t __s0_440 = __p0_440; \
	49604	uint16x4_t __s1_440 = __p1_440; \
	49605	uint16x8_t __s2_440 = __p2_440; \
	49606	uint16x4_t __rev0_440; __rev0_440 = __builtin_shufflevector(__s0_440, __s0_440, 3, 2, 1, 0); \
	49607	uint16x4_t __rev1_440; __rev1_440 = __builtin_shufflevector(__s1_440, __s1_440, 3, 2, 1, 0); \
	49608	uint16x8_t __rev2_440; __rev2_440 = __builtin_shufflevector(__s2_440, __s2_440, 7, 6, 5, 4, 3, 2, 1, 0); \
	49609	__ret_440 = __rev0_440 + __rev1_440 * __noswap_splat_laneq_u16(__rev2_440, __p3_440); \
	49610	__ret_440 = __builtin_shufflevector(__ret_440, __ret_440, 3, 2, 1, 0); \
	49611	__ret_440; \
	49612	})
	49613	#endif
	49614
	49615	#ifdef __LITTLE_ENDIAN__
	49616	#define vmla_laneq_f32(__p0_441, __p1_441, __p2_441, __p3_441) __extension__ ({ \
	49617	float32x2_t __ret_441; \
	49618	float32x2_t __s0_441 = __p0_441; \
	49619	float32x2_t __s1_441 = __p1_441; \
	49620	float32x4_t __s2_441 = __p2_441; \
	49621	__ret_441 = __s0_441 + __s1_441 * splat_laneq_f32(__s2_441, __p3_441); \
	49622	__ret_441; \
	49623	})
	49624	#else
	49625	#define vmla_laneq_f32(__p0_442, __p1_442, __p2_442, __p3_442) __extension__ ({ \
	49626	float32x2_t __ret_442; \
	49627	float32x2_t __s0_442 = __p0_442; \
	49628	float32x2_t __s1_442 = __p1_442; \
	49629	float32x4_t __s2_442 = __p2_442; \
	49630	float32x2_t __rev0_442; __rev0_442 = __builtin_shufflevector(__s0_442, __s0_442, 1, 0); \
	49631	float32x2_t __rev1_442; __rev1_442 = __builtin_shufflevector(__s1_442, __s1_442, 1, 0); \
	49632	float32x4_t __rev2_442; __rev2_442 = __builtin_shufflevector(__s2_442, __s2_442, 3, 2, 1, 0); \
	49633	__ret_442 = __rev0_442 + __rev1_442 * __noswap_splat_laneq_f32(__rev2_442, __p3_442); \
	49634	__ret_442 = __builtin_shufflevector(__ret_442, __ret_442, 1, 0); \
	49635	__ret_442; \
	49636	})
	49637	#endif
	49638
	49639	#ifdef __LITTLE_ENDIAN__
	49640	#define vmla_laneq_s32(__p0_443, __p1_443, __p2_443, __p3_443) __extension__ ({ \
	49641	int32x2_t __ret_443; \
	49642	int32x2_t __s0_443 = __p0_443; \
	49643	int32x2_t __s1_443 = __p1_443; \
	49644	int32x4_t __s2_443 = __p2_443; \
	49645	__ret_443 = __s0_443 + __s1_443 * splat_laneq_s32(__s2_443, __p3_443); \
	49646	__ret_443; \
	49647	})
	49648	#else
	49649	#define vmla_laneq_s32(__p0_444, __p1_444, __p2_444, __p3_444) __extension__ ({ \
	49650	int32x2_t __ret_444; \
	49651	int32x2_t __s0_444 = __p0_444; \
	49652	int32x2_t __s1_444 = __p1_444; \
	49653	int32x4_t __s2_444 = __p2_444; \
	49654	int32x2_t __rev0_444; __rev0_444 = __builtin_shufflevector(__s0_444, __s0_444, 1, 0); \
	49655	int32x2_t __rev1_444; __rev1_444 = __builtin_shufflevector(__s1_444, __s1_444, 1, 0); \
	49656	int32x4_t __rev2_444; __rev2_444 = __builtin_shufflevector(__s2_444, __s2_444, 3, 2, 1, 0); \
	49657	__ret_444 = __rev0_444 + __rev1_444 * __noswap_splat_laneq_s32(__rev2_444, __p3_444); \
	49658	__ret_444 = __builtin_shufflevector(__ret_444, __ret_444, 1, 0); \
	49659	__ret_444; \
	49660	})
	49661	#endif
	49662
	49663	#ifdef __LITTLE_ENDIAN__
	49664	#define vmla_laneq_s16(__p0_445, __p1_445, __p2_445, __p3_445) __extension__ ({ \
	49665	int16x4_t __ret_445; \
	49666	int16x4_t __s0_445 = __p0_445; \
	49667	int16x4_t __s1_445 = __p1_445; \
	49668	int16x8_t __s2_445 = __p2_445; \
	49669	__ret_445 = __s0_445 + __s1_445 * splat_laneq_s16(__s2_445, __p3_445); \
	49670	__ret_445; \
	49671	})
	49672	#else
	49673	#define vmla_laneq_s16(__p0_446, __p1_446, __p2_446, __p3_446) __extension__ ({ \
	49674	int16x4_t __ret_446; \
	49675	int16x4_t __s0_446 = __p0_446; \
	49676	int16x4_t __s1_446 = __p1_446; \
	49677	int16x8_t __s2_446 = __p2_446; \
	49678	int16x4_t __rev0_446; __rev0_446 = __builtin_shufflevector(__s0_446, __s0_446, 3, 2, 1, 0); \
	49679	int16x4_t __rev1_446; __rev1_446 = __builtin_shufflevector(__s1_446, __s1_446, 3, 2, 1, 0); \
	49680	int16x8_t __rev2_446; __rev2_446 = __builtin_shufflevector(__s2_446, __s2_446, 7, 6, 5, 4, 3, 2, 1, 0); \
	49681	__ret_446 = __rev0_446 + __rev1_446 * __noswap_splat_laneq_s16(__rev2_446, __p3_446); \
	49682	__ret_446 = __builtin_shufflevector(__ret_446, __ret_446, 3, 2, 1, 0); \
	49683	__ret_446; \
	49684	})
	49685	#endif
	49686
	49687	#ifdef __LITTLE_ENDIAN__
	49688	#define vmlal_high_lane_u32(__p0_447, __p1_447, __p2_447, __p3_447) __extension__ ({ \
	49689	uint64x2_t __ret_447; \
	49690	uint64x2_t __s0_447 = __p0_447; \
	49691	uint32x4_t __s1_447 = __p1_447; \
	49692	uint32x2_t __s2_447 = __p2_447; \
	49693	__ret_447 = __s0_447 + vmull_u32(vget_high_u32(__s1_447), splat_lane_u32(__s2_447, __p3_447)); \
	49694	__ret_447; \
	49695	})
	49696	#else
	49697	#define vmlal_high_lane_u32(__p0_448, __p1_448, __p2_448, __p3_448) __extension__ ({ \
	49698	uint64x2_t __ret_448; \
	49699	uint64x2_t __s0_448 = __p0_448; \
	49700	uint32x4_t __s1_448 = __p1_448; \
	49701	uint32x2_t __s2_448 = __p2_448; \
	49702	uint64x2_t __rev0_448; __rev0_448 = __builtin_shufflevector(__s0_448, __s0_448, 1, 0); \
	49703	uint32x4_t __rev1_448; __rev1_448 = __builtin_shufflevector(__s1_448, __s1_448, 3, 2, 1, 0); \
	49704	uint32x2_t __rev2_448; __rev2_448 = __builtin_shufflevector(__s2_448, __s2_448, 1, 0); \
	49705	__ret_448 = __rev0_448 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_448), __noswap_splat_lane_u32(__rev2_448, __p3_448)); \
	49706	__ret_448 = __builtin_shufflevector(__ret_448, __ret_448, 1, 0); \
	49707	__ret_448; \
	49708	})
	49709	#endif
	49710
	49711	#ifdef __LITTLE_ENDIAN__
	49712	#define vmlal_high_lane_u16(__p0_449, __p1_449, __p2_449, __p3_449) __extension__ ({ \
	49713	uint32x4_t __ret_449; \
	49714	uint32x4_t __s0_449 = __p0_449; \
	49715	uint16x8_t __s1_449 = __p1_449; \
	49716	uint16x4_t __s2_449 = __p2_449; \
	49717	__ret_449 = __s0_449 + vmull_u16(vget_high_u16(__s1_449), splat_lane_u16(__s2_449, __p3_449)); \
	49718	__ret_449; \
	49719	})
	49720	#else
	49721	#define vmlal_high_lane_u16(__p0_450, __p1_450, __p2_450, __p3_450) __extension__ ({ \
	49722	uint32x4_t __ret_450; \
	49723	uint32x4_t __s0_450 = __p0_450; \
	49724	uint16x8_t __s1_450 = __p1_450; \
	49725	uint16x4_t __s2_450 = __p2_450; \
	49726	uint32x4_t __rev0_450; __rev0_450 = __builtin_shufflevector(__s0_450, __s0_450, 3, 2, 1, 0); \
	49727	uint16x8_t __rev1_450; __rev1_450 = __builtin_shufflevector(__s1_450, __s1_450, 7, 6, 5, 4, 3, 2, 1, 0); \
	49728	uint16x4_t __rev2_450; __rev2_450 = __builtin_shufflevector(__s2_450, __s2_450, 3, 2, 1, 0); \
	49729	__ret_450 = __rev0_450 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_450), __noswap_splat_lane_u16(__rev2_450, __p3_450)); \
	49730	__ret_450 = __builtin_shufflevector(__ret_450, __ret_450, 3, 2, 1, 0); \
	49731	__ret_450; \
	49732	})
	49733	#endif
	49734
	49735	#ifdef __LITTLE_ENDIAN__
	49736	#define vmlal_high_lane_s32(__p0_451, __p1_451, __p2_451, __p3_451) __extension__ ({ \
	49737	int64x2_t __ret_451; \
	49738	int64x2_t __s0_451 = __p0_451; \
	49739	int32x4_t __s1_451 = __p1_451; \
	49740	int32x2_t __s2_451 = __p2_451; \
	49741	__ret_451 = __s0_451 + vmull_s32(vget_high_s32(__s1_451), splat_lane_s32(__s2_451, __p3_451)); \
	49742	__ret_451; \
	49743	})
	49744	#else
	49745	#define vmlal_high_lane_s32(__p0_452, __p1_452, __p2_452, __p3_452) __extension__ ({ \
	49746	int64x2_t __ret_452; \
	49747	int64x2_t __s0_452 = __p0_452; \
	49748	int32x4_t __s1_452 = __p1_452; \
	49749	int32x2_t __s2_452 = __p2_452; \
	49750	int64x2_t __rev0_452; __rev0_452 = __builtin_shufflevector(__s0_452, __s0_452, 1, 0); \
	49751	int32x4_t __rev1_452; __rev1_452 = __builtin_shufflevector(__s1_452, __s1_452, 3, 2, 1, 0); \
	49752	int32x2_t __rev2_452; __rev2_452 = __builtin_shufflevector(__s2_452, __s2_452, 1, 0); \
	49753	__ret_452 = __rev0_452 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_452), __noswap_splat_lane_s32(__rev2_452, __p3_452)); \
	49754	__ret_452 = __builtin_shufflevector(__ret_452, __ret_452, 1, 0); \
	49755	__ret_452; \
	49756	})
	49757	#endif
	49758
	49759	#ifdef __LITTLE_ENDIAN__
	49760	#define vmlal_high_lane_s16(__p0_453, __p1_453, __p2_453, __p3_453) __extension__ ({ \
	49761	int32x4_t __ret_453; \
	49762	int32x4_t __s0_453 = __p0_453; \
	49763	int16x8_t __s1_453 = __p1_453; \
	49764	int16x4_t __s2_453 = __p2_453; \
	49765	__ret_453 = __s0_453 + vmull_s16(vget_high_s16(__s1_453), splat_lane_s16(__s2_453, __p3_453)); \
	49766	__ret_453; \
	49767	})
	49768	#else
	49769	#define vmlal_high_lane_s16(__p0_454, __p1_454, __p2_454, __p3_454) __extension__ ({ \
	49770	int32x4_t __ret_454; \
	49771	int32x4_t __s0_454 = __p0_454; \
	49772	int16x8_t __s1_454 = __p1_454; \
	49773	int16x4_t __s2_454 = __p2_454; \
	49774	int32x4_t __rev0_454; __rev0_454 = __builtin_shufflevector(__s0_454, __s0_454, 3, 2, 1, 0); \
	49775	int16x8_t __rev1_454; __rev1_454 = __builtin_shufflevector(__s1_454, __s1_454, 7, 6, 5, 4, 3, 2, 1, 0); \
	49776	int16x4_t __rev2_454; __rev2_454 = __builtin_shufflevector(__s2_454, __s2_454, 3, 2, 1, 0); \
	49777	__ret_454 = __rev0_454 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_454), __noswap_splat_lane_s16(__rev2_454, __p3_454)); \
	49778	__ret_454 = __builtin_shufflevector(__ret_454, __ret_454, 3, 2, 1, 0); \
	49779	__ret_454; \
	49780	})
	49781	#endif
	49782
	49783	#ifdef __LITTLE_ENDIAN__
	49784	#define vmlal_high_laneq_u32(__p0_455, __p1_455, __p2_455, __p3_455) __extension__ ({ \
	49785	uint64x2_t __ret_455; \
	49786	uint64x2_t __s0_455 = __p0_455; \
	49787	uint32x4_t __s1_455 = __p1_455; \
	49788	uint32x4_t __s2_455 = __p2_455; \
	49789	__ret_455 = __s0_455 + vmull_u32(vget_high_u32(__s1_455), splat_laneq_u32(__s2_455, __p3_455)); \
	49790	__ret_455; \
	49791	})
	49792	#else
	49793	#define vmlal_high_laneq_u32(__p0_456, __p1_456, __p2_456, __p3_456) __extension__ ({ \
	49794	uint64x2_t __ret_456; \
	49795	uint64x2_t __s0_456 = __p0_456; \
	49796	uint32x4_t __s1_456 = __p1_456; \
	49797	uint32x4_t __s2_456 = __p2_456; \
	49798	uint64x2_t __rev0_456; __rev0_456 = __builtin_shufflevector(__s0_456, __s0_456, 1, 0); \
	49799	uint32x4_t __rev1_456; __rev1_456 = __builtin_shufflevector(__s1_456, __s1_456, 3, 2, 1, 0); \
	49800	uint32x4_t __rev2_456; __rev2_456 = __builtin_shufflevector(__s2_456, __s2_456, 3, 2, 1, 0); \
	49801	__ret_456 = __rev0_456 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_456), __noswap_splat_laneq_u32(__rev2_456, __p3_456)); \
	49802	__ret_456 = __builtin_shufflevector(__ret_456, __ret_456, 1, 0); \
	49803	__ret_456; \
	49804	})
	49805	#endif
	49806
	49807	#ifdef __LITTLE_ENDIAN__
	49808	#define vmlal_high_laneq_u16(__p0_457, __p1_457, __p2_457, __p3_457) __extension__ ({ \
	49809	uint32x4_t __ret_457; \
	49810	uint32x4_t __s0_457 = __p0_457; \
	49811	uint16x8_t __s1_457 = __p1_457; \
	49812	uint16x8_t __s2_457 = __p2_457; \
	49813	__ret_457 = __s0_457 + vmull_u16(vget_high_u16(__s1_457), splat_laneq_u16(__s2_457, __p3_457)); \
	49814	__ret_457; \
	49815	})
	49816	#else
	49817	#define vmlal_high_laneq_u16(__p0_458, __p1_458, __p2_458, __p3_458) __extension__ ({ \
	49818	uint32x4_t __ret_458; \
	49819	uint32x4_t __s0_458 = __p0_458; \
	49820	uint16x8_t __s1_458 = __p1_458; \
	49821	uint16x8_t __s2_458 = __p2_458; \
	49822	uint32x4_t __rev0_458; __rev0_458 = __builtin_shufflevector(__s0_458, __s0_458, 3, 2, 1, 0); \
	49823	uint16x8_t __rev1_458; __rev1_458 = __builtin_shufflevector(__s1_458, __s1_458, 7, 6, 5, 4, 3, 2, 1, 0); \
	49824	uint16x8_t __rev2_458; __rev2_458 = __builtin_shufflevector(__s2_458, __s2_458, 7, 6, 5, 4, 3, 2, 1, 0); \
	49825	__ret_458 = __rev0_458 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_458), __noswap_splat_laneq_u16(__rev2_458, __p3_458)); \
	49826	__ret_458 = __builtin_shufflevector(__ret_458, __ret_458, 3, 2, 1, 0); \
	49827	__ret_458; \
	49828	})
	49829	#endif
	49830
	49831	#ifdef __LITTLE_ENDIAN__
	49832	#define vmlal_high_laneq_s32(__p0_459, __p1_459, __p2_459, __p3_459) __extension__ ({ \
	49833	int64x2_t __ret_459; \
	49834	int64x2_t __s0_459 = __p0_459; \
	49835	int32x4_t __s1_459 = __p1_459; \
	49836	int32x4_t __s2_459 = __p2_459; \
	49837	__ret_459 = __s0_459 + vmull_s32(vget_high_s32(__s1_459), splat_laneq_s32(__s2_459, __p3_459)); \
	49838	__ret_459; \
	49839	})
	49840	#else
	49841	#define vmlal_high_laneq_s32(__p0_460, __p1_460, __p2_460, __p3_460) __extension__ ({ \
	49842	int64x2_t __ret_460; \
	49843	int64x2_t __s0_460 = __p0_460; \
	49844	int32x4_t __s1_460 = __p1_460; \
	49845	int32x4_t __s2_460 = __p2_460; \
	49846	int64x2_t __rev0_460; __rev0_460 = __builtin_shufflevector(__s0_460, __s0_460, 1, 0); \
	49847	int32x4_t __rev1_460; __rev1_460 = __builtin_shufflevector(__s1_460, __s1_460, 3, 2, 1, 0); \
	49848	int32x4_t __rev2_460; __rev2_460 = __builtin_shufflevector(__s2_460, __s2_460, 3, 2, 1, 0); \
	49849	__ret_460 = __rev0_460 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_460), __noswap_splat_laneq_s32(__rev2_460, __p3_460)); \
	49850	__ret_460 = __builtin_shufflevector(__ret_460, __ret_460, 1, 0); \
	49851	__ret_460; \
	49852	})
	49853	#endif
	49854
	49855	#ifdef __LITTLE_ENDIAN__
	49856	#define vmlal_high_laneq_s16(__p0_461, __p1_461, __p2_461, __p3_461) __extension__ ({ \
	49857	int32x4_t __ret_461; \
	49858	int32x4_t __s0_461 = __p0_461; \
	49859	int16x8_t __s1_461 = __p1_461; \
	49860	int16x8_t __s2_461 = __p2_461; \
	49861	__ret_461 = __s0_461 + vmull_s16(vget_high_s16(__s1_461), splat_laneq_s16(__s2_461, __p3_461)); \
	49862	__ret_461; \
	49863	})
	49864	#else
	49865	#define vmlal_high_laneq_s16(__p0_462, __p1_462, __p2_462, __p3_462) __extension__ ({ \
	49866	int32x4_t __ret_462; \
	49867	int32x4_t __s0_462 = __p0_462; \
	49868	int16x8_t __s1_462 = __p1_462; \
	49869	int16x8_t __s2_462 = __p2_462; \
	49870	int32x4_t __rev0_462; __rev0_462 = __builtin_shufflevector(__s0_462, __s0_462, 3, 2, 1, 0); \
	49871	int16x8_t __rev1_462; __rev1_462 = __builtin_shufflevector(__s1_462, __s1_462, 7, 6, 5, 4, 3, 2, 1, 0); \
	49872	int16x8_t __rev2_462; __rev2_462 = __builtin_shufflevector(__s2_462, __s2_462, 7, 6, 5, 4, 3, 2, 1, 0); \
	49873	__ret_462 = __rev0_462 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_462), __noswap_splat_laneq_s16(__rev2_462, __p3_462)); \
	49874	__ret_462 = __builtin_shufflevector(__ret_462, __ret_462, 3, 2, 1, 0); \
	49875	__ret_462; \
	49876	})
	49877	#endif
	49878
	49879	#ifdef __LITTLE_ENDIAN__
	49880	#define vmlal_laneq_u32(__p0_463, __p1_463, __p2_463, __p3_463) __extension__ ({ \
	49881	uint64x2_t __ret_463; \
	49882	uint64x2_t __s0_463 = __p0_463; \
	49883	uint32x2_t __s1_463 = __p1_463; \
	49884	uint32x4_t __s2_463 = __p2_463; \
	49885	__ret_463 = __s0_463 + vmull_u32(__s1_463, splat_laneq_u32(__s2_463, __p3_463)); \
	49886	__ret_463; \
	49887	})
	49888	#else
	49889	#define vmlal_laneq_u32(__p0_464, __p1_464, __p2_464, __p3_464) __extension__ ({ \
	49890	uint64x2_t __ret_464; \
	49891	uint64x2_t __s0_464 = __p0_464; \
	49892	uint32x2_t __s1_464 = __p1_464; \
	49893	uint32x4_t __s2_464 = __p2_464; \
	49894	uint64x2_t __rev0_464; __rev0_464 = __builtin_shufflevector(__s0_464, __s0_464, 1, 0); \
	49895	uint32x2_t __rev1_464; __rev1_464 = __builtin_shufflevector(__s1_464, __s1_464, 1, 0); \
	49896	uint32x4_t __rev2_464; __rev2_464 = __builtin_shufflevector(__s2_464, __s2_464, 3, 2, 1, 0); \
	49897	__ret_464 = __rev0_464 + __noswap_vmull_u32(__rev1_464, __noswap_splat_laneq_u32(__rev2_464, __p3_464)); \
	49898	__ret_464 = __builtin_shufflevector(__ret_464, __ret_464, 1, 0); \
	49899	__ret_464; \
	49900	})
	49901	#endif
	49902
	49903	#ifdef __LITTLE_ENDIAN__
	49904	#define vmlal_laneq_u16(__p0_465, __p1_465, __p2_465, __p3_465) __extension__ ({ \
	49905	uint32x4_t __ret_465; \
	49906	uint32x4_t __s0_465 = __p0_465; \
	49907	uint16x4_t __s1_465 = __p1_465; \
	49908	uint16x8_t __s2_465 = __p2_465; \
	49909	__ret_465 = __s0_465 + vmull_u16(__s1_465, splat_laneq_u16(__s2_465, __p3_465)); \
	49910	__ret_465; \
	49911	})
	49912	#else
	49913	#define vmlal_laneq_u16(__p0_466, __p1_466, __p2_466, __p3_466) __extension__ ({ \
	49914	uint32x4_t __ret_466; \
	49915	uint32x4_t __s0_466 = __p0_466; \
	49916	uint16x4_t __s1_466 = __p1_466; \
	49917	uint16x8_t __s2_466 = __p2_466; \
	49918	uint32x4_t __rev0_466; __rev0_466 = __builtin_shufflevector(__s0_466, __s0_466, 3, 2, 1, 0); \
	49919	uint16x4_t __rev1_466; __rev1_466 = __builtin_shufflevector(__s1_466, __s1_466, 3, 2, 1, 0); \
	49920	uint16x8_t __rev2_466; __rev2_466 = __builtin_shufflevector(__s2_466, __s2_466, 7, 6, 5, 4, 3, 2, 1, 0); \
	49921	__ret_466 = __rev0_466 + __noswap_vmull_u16(__rev1_466, __noswap_splat_laneq_u16(__rev2_466, __p3_466)); \
	49922	__ret_466 = __builtin_shufflevector(__ret_466, __ret_466, 3, 2, 1, 0); \
	49923	__ret_466; \
	49924	})
49123	49925	#endif
49124	49926
49125		__ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
49126		uint64x1_t __ret;
49127		__ret = (uint64x1_t)(__p0 >= __p1);
49128		return __ret;
49129		}
49130		__ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
49131		uint64x1_t __ret;
49132		__ret = (uint64x1_t)(__p0 >= __p1);
49133		return __ret;
49134		}
49135		__ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
49136		uint64x1_t __ret;
49137		__ret = (uint64x1_t)(__p0 >= __p1);
49138		return __ret;
49139		}
49140		__ai uint64_t vcged_s64(int64_t __p0, int64_t __p1) {
49141		uint64_t __ret;
49142		__ret = (uint64_t) __builtin_neon_vcged_s64(__p0, __p1);
49143		return __ret;
49144		}
49145		__ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
49146		uint64_t __ret;
49147		__ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
49148		return __ret;
49149		}
49150		__ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
49151		uint64_t __ret;
49152		__ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
49153		return __ret;
49154		}
49155		__ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
49156		uint32_t __ret;
49157		__ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
49158		return __ret;
49159		}
49160	49927	#ifdef __LITTLE_ENDIAN__
49161		__ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
49162		uint8x16_t __ret;
49163		__ret = (uint8x16_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 48);
49164		return __ret;
49165		}
	49928	#define vmlal_laneq_s32(__p0_467, __p1_467, __p2_467, __p3_467) __extension__ ({ \
	49929	int64x2_t __ret_467; \
	49930	int64x2_t __s0_467 = __p0_467; \
	49931	int32x2_t __s1_467 = __p1_467; \
	49932	int32x4_t __s2_467 = __p2_467; \
	49933	__ret_467 = __s0_467 + vmull_s32(__s1_467, splat_laneq_s32(__s2_467, __p3_467)); \
	49934	__ret_467; \
	49935	})
49166	49936	#else
49167		__ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
49168		uint8x16_t __ret;
49169		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49170		__ret = (uint8x16_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 48);
49171		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49172		return __ret;
49173		}
	49937	#define vmlal_laneq_s32(__p0_468, __p1_468, __p2_468, __p3_468) __extension__ ({ \
	49938	int64x2_t __ret_468; \
	49939	int64x2_t __s0_468 = __p0_468; \
	49940	int32x2_t __s1_468 = __p1_468; \
	49941	int32x4_t __s2_468 = __p2_468; \
	49942	int64x2_t __rev0_468; __rev0_468 = __builtin_shufflevector(__s0_468, __s0_468, 1, 0); \
	49943	int32x2_t __rev1_468; __rev1_468 = __builtin_shufflevector(__s1_468, __s1_468, 1, 0); \
	49944	int32x4_t __rev2_468; __rev2_468 = __builtin_shufflevector(__s2_468, __s2_468, 3, 2, 1, 0); \
	49945	__ret_468 = __rev0_468 + __noswap_vmull_s32(__rev1_468, __noswap_splat_laneq_s32(__rev2_468, __p3_468)); \
	49946	__ret_468 = __builtin_shufflevector(__ret_468, __ret_468, 1, 0); \
	49947	__ret_468; \
	49948	})
49174	49949	#endif
49175	49950
49176	49951	#ifdef __LITTLE_ENDIAN__
49177		__ai uint64x2_t vcgezq_f64(float64x2_t __p0) {
49178		uint64x2_t __ret;
49179		__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 51);
49180		return __ret;
49181		}
	49952	#define vmlal_laneq_s16(__p0_469, __p1_469, __p2_469, __p3_469) __extension__ ({ \
	49953	int32x4_t __ret_469; \
	49954	int32x4_t __s0_469 = __p0_469; \
	49955	int16x4_t __s1_469 = __p1_469; \
	49956	int16x8_t __s2_469 = __p2_469; \
	49957	__ret_469 = __s0_469 + vmull_s16(__s1_469, splat_laneq_s16(__s2_469, __p3_469)); \
	49958	__ret_469; \
	49959	})
49182	49960	#else
49183		__ai uint64x2_t vcgezq_f64(float64x2_t __p0) {
49184		uint64x2_t __ret;
49185		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49186		__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 51);
49187		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49188		return __ret;
49189		}
	49961	#define vmlal_laneq_s16(__p0_470, __p1_470, __p2_470, __p3_470) __extension__ ({ \
	49962	int32x4_t __ret_470; \
	49963	int32x4_t __s0_470 = __p0_470; \
	49964	int16x4_t __s1_470 = __p1_470; \
	49965	int16x8_t __s2_470 = __p2_470; \
	49966	int32x4_t __rev0_470; __rev0_470 = __builtin_shufflevector(__s0_470, __s0_470, 3, 2, 1, 0); \
	49967	int16x4_t __rev1_470; __rev1_470 = __builtin_shufflevector(__s1_470, __s1_470, 3, 2, 1, 0); \
	49968	int16x8_t __rev2_470; __rev2_470 = __builtin_shufflevector(__s2_470, __s2_470, 7, 6, 5, 4, 3, 2, 1, 0); \
	49969	__ret_470 = __rev0_470 + __noswap_vmull_s16(__rev1_470, __noswap_splat_laneq_s16(__rev2_470, __p3_470)); \
	49970	__ret_470 = __builtin_shufflevector(__ret_470, __ret_470, 3, 2, 1, 0); \
	49971	__ret_470; \
	49972	})
49190	49973	#endif
49191	49974
49192	49975	#ifdef __LITTLE_ENDIAN__
49193		__ai uint32x4_t vcgezq_f32(float32x4_t __p0) {
49194		uint32x4_t __ret;
49195		__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 50);
	49976	__ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	49977	float64x2_t __ret;
	49978	__ret = __p0 - __p1 * __p2;
49196	49979	return __ret;
49197	49980	}
49198	49981	#else
49199		__ai uint32x4_t vcgezq_f32(float32x4_t __p0) {
49200		uint32x4_t __ret;
49201		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49202		__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 50);
49203		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	49982	__ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	49983	float64x2_t __ret;
	49984	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	49985	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	49986	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	49987	__ret = __rev0 - __rev1 * __rev2;
	49988	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49204	49989	return __ret;
49205	49990	}
49206	49991	#endif
49207	49992
49208		#ifdef __LITTLE_ENDIAN__
49209		__ai uint32x4_t vcgezq_s32(int32x4_t __p0) {
49210		uint32x4_t __ret;
49211		__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 50);
	49993	__ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	49994	float64x1_t __ret;
	49995	__ret = __p0 - __p1 * __p2;
49212	49996	return __ret;
49213	49997	}
	49998	#ifdef __LITTLE_ENDIAN__
	49999	#define vmlsq_laneq_u32(__p0_471, __p1_471, __p2_471, __p3_471) __extension__ ({ \
	50000	uint32x4_t __ret_471; \
	50001	uint32x4_t __s0_471 = __p0_471; \
	50002	uint32x4_t __s1_471 = __p1_471; \
	50003	uint32x4_t __s2_471 = __p2_471; \
	50004	__ret_471 = __s0_471 - __s1_471 * splatq_laneq_u32(__s2_471, __p3_471); \
	50005	__ret_471; \
	50006	})
49214	50007	#else
49215		__ai uint32x4_t vcgezq_s32(int32x4_t __p0) {
49216		uint32x4_t __ret;
49217		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49218		__ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 50);
49219		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49220		return __ret;
49221		}
	50008	#define vmlsq_laneq_u32(__p0_472, __p1_472, __p2_472, __p3_472) __extension__ ({ \
	50009	uint32x4_t __ret_472; \
	50010	uint32x4_t __s0_472 = __p0_472; \
	50011	uint32x4_t __s1_472 = __p1_472; \
	50012	uint32x4_t __s2_472 = __p2_472; \
	50013	uint32x4_t __rev0_472; __rev0_472 = __builtin_shufflevector(__s0_472, __s0_472, 3, 2, 1, 0); \
	50014	uint32x4_t __rev1_472; __rev1_472 = __builtin_shufflevector(__s1_472, __s1_472, 3, 2, 1, 0); \
	50015	uint32x4_t __rev2_472; __rev2_472 = __builtin_shufflevector(__s2_472, __s2_472, 3, 2, 1, 0); \
	50016	__ret_472 = __rev0_472 - __rev1_472 * __noswap_splatq_laneq_u32(__rev2_472, __p3_472); \
	50017	__ret_472 = __builtin_shufflevector(__ret_472, __ret_472, 3, 2, 1, 0); \
	50018	__ret_472; \
	50019	})
49222	50020	#endif
49223	50021
49224	50022	#ifdef __LITTLE_ENDIAN__
49225		__ai uint64x2_t vcgezq_s64(int64x2_t __p0) {
49226		uint64x2_t __ret;
49227		__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 51);
49228		return __ret;
49229		}
	50023	#define vmlsq_laneq_u16(__p0_473, __p1_473, __p2_473, __p3_473) __extension__ ({ \
	50024	uint16x8_t __ret_473; \
	50025	uint16x8_t __s0_473 = __p0_473; \
	50026	uint16x8_t __s1_473 = __p1_473; \
	50027	uint16x8_t __s2_473 = __p2_473; \
	50028	__ret_473 = __s0_473 - __s1_473 * splatq_laneq_u16(__s2_473, __p3_473); \
	50029	__ret_473; \
	50030	})
49230	50031	#else
49231		__ai uint64x2_t vcgezq_s64(int64x2_t __p0) {
49232		uint64x2_t __ret;
49233		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49234		__ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 51);
49235		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49236		return __ret;
49237		}
	50032	#define vmlsq_laneq_u16(__p0_474, __p1_474, __p2_474, __p3_474) __extension__ ({ \
	50033	uint16x8_t __ret_474; \
	50034	uint16x8_t __s0_474 = __p0_474; \
	50035	uint16x8_t __s1_474 = __p1_474; \
	50036	uint16x8_t __s2_474 = __p2_474; \
	50037	uint16x8_t __rev0_474; __rev0_474 = __builtin_shufflevector(__s0_474, __s0_474, 7, 6, 5, 4, 3, 2, 1, 0); \
	50038	uint16x8_t __rev1_474; __rev1_474 = __builtin_shufflevector(__s1_474, __s1_474, 7, 6, 5, 4, 3, 2, 1, 0); \
	50039	uint16x8_t __rev2_474; __rev2_474 = __builtin_shufflevector(__s2_474, __s2_474, 7, 6, 5, 4, 3, 2, 1, 0); \
	50040	__ret_474 = __rev0_474 - __rev1_474 * __noswap_splatq_laneq_u16(__rev2_474, __p3_474); \
	50041	__ret_474 = __builtin_shufflevector(__ret_474, __ret_474, 7, 6, 5, 4, 3, 2, 1, 0); \
	50042	__ret_474; \
	50043	})
49238	50044	#endif
49239	50045
49240	50046	#ifdef __LITTLE_ENDIAN__
49241		__ai uint16x8_t vcgezq_s16(int16x8_t __p0) {
49242		uint16x8_t __ret;
49243		__ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__p0, 49);
49244		return __ret;
49245		}
	50047	#define vmlsq_laneq_f32(__p0_475, __p1_475, __p2_475, __p3_475) __extension__ ({ \
	50048	float32x4_t __ret_475; \
	50049	float32x4_t __s0_475 = __p0_475; \
	50050	float32x4_t __s1_475 = __p1_475; \
	50051	float32x4_t __s2_475 = __p2_475; \
	50052	__ret_475 = __s0_475 - __s1_475 * splatq_laneq_f32(__s2_475, __p3_475); \
	50053	__ret_475; \
	50054	})
49246	50055	#else
49247		__ai uint16x8_t vcgezq_s16(int16x8_t __p0) {
49248		uint16x8_t __ret;
49249		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49250		__ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 49);
49251		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49252		return __ret;
49253		}
	50056	#define vmlsq_laneq_f32(__p0_476, __p1_476, __p2_476, __p3_476) __extension__ ({ \
	50057	float32x4_t __ret_476; \
	50058	float32x4_t __s0_476 = __p0_476; \
	50059	float32x4_t __s1_476 = __p1_476; \
	50060	float32x4_t __s2_476 = __p2_476; \
	50061	float32x4_t __rev0_476; __rev0_476 = __builtin_shufflevector(__s0_476, __s0_476, 3, 2, 1, 0); \
	50062	float32x4_t __rev1_476; __rev1_476 = __builtin_shufflevector(__s1_476, __s1_476, 3, 2, 1, 0); \
	50063	float32x4_t __rev2_476; __rev2_476 = __builtin_shufflevector(__s2_476, __s2_476, 3, 2, 1, 0); \
	50064	__ret_476 = __rev0_476 - __rev1_476 * __noswap_splatq_laneq_f32(__rev2_476, __p3_476); \
	50065	__ret_476 = __builtin_shufflevector(__ret_476, __ret_476, 3, 2, 1, 0); \
	50066	__ret_476; \
	50067	})
49254	50068	#endif
49255	50069
49256	50070	#ifdef __LITTLE_ENDIAN__
49257		__ai uint8x8_t vcgez_s8(int8x8_t __p0) {
49258		uint8x8_t __ret;
49259		__ret = (uint8x8_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 16);
49260		return __ret;
49261		}
	50071	#define vmlsq_laneq_s32(__p0_477, __p1_477, __p2_477, __p3_477) __extension__ ({ \
	50072	int32x4_t __ret_477; \
	50073	int32x4_t __s0_477 = __p0_477; \
	50074	int32x4_t __s1_477 = __p1_477; \
	50075	int32x4_t __s2_477 = __p2_477; \
	50076	__ret_477 = __s0_477 - __s1_477 * splatq_laneq_s32(__s2_477, __p3_477); \
	50077	__ret_477; \
	50078	})
49262	50079	#else
49263		__ai uint8x8_t vcgez_s8(int8x8_t __p0) {
49264		uint8x8_t __ret;
49265		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49266		__ret = (uint8x8_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 16);
49267		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49268		return __ret;
49269		}
	50080	#define vmlsq_laneq_s32(__p0_478, __p1_478, __p2_478, __p3_478) __extension__ ({ \
	50081	int32x4_t __ret_478; \
	50082	int32x4_t __s0_478 = __p0_478; \
	50083	int32x4_t __s1_478 = __p1_478; \
	50084	int32x4_t __s2_478 = __p2_478; \
	50085	int32x4_t __rev0_478; __rev0_478 = __builtin_shufflevector(__s0_478, __s0_478, 3, 2, 1, 0); \
	50086	int32x4_t __rev1_478; __rev1_478 = __builtin_shufflevector(__s1_478, __s1_478, 3, 2, 1, 0); \
	50087	int32x4_t __rev2_478; __rev2_478 = __builtin_shufflevector(__s2_478, __s2_478, 3, 2, 1, 0); \
	50088	__ret_478 = __rev0_478 - __rev1_478 * __noswap_splatq_laneq_s32(__rev2_478, __p3_478); \
	50089	__ret_478 = __builtin_shufflevector(__ret_478, __ret_478, 3, 2, 1, 0); \
	50090	__ret_478; \
	50091	})
49270	50092	#endif
49271	50093
49272		__ai uint64x1_t vcgez_f64(float64x1_t __p0) {
49273		uint64x1_t __ret;
49274		__ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
49275		return __ret;
49276		}
49277	50094	#ifdef __LITTLE_ENDIAN__
49278		__ai uint32x2_t vcgez_f32(float32x2_t __p0) {
49279		uint32x2_t __ret;
49280		__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 18);
49281		return __ret;
49282		}
	50095	#define vmlsq_laneq_s16(__p0_479, __p1_479, __p2_479, __p3_479) __extension__ ({ \
	50096	int16x8_t __ret_479; \
	50097	int16x8_t __s0_479 = __p0_479; \
	50098	int16x8_t __s1_479 = __p1_479; \
	50099	int16x8_t __s2_479 = __p2_479; \
	50100	__ret_479 = __s0_479 - __s1_479 * splatq_laneq_s16(__s2_479, __p3_479); \
	50101	__ret_479; \
	50102	})
49283	50103	#else
49284		__ai uint32x2_t vcgez_f32(float32x2_t __p0) {
49285		uint32x2_t __ret;
49286		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49287		__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 18);
49288		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49289		return __ret;
49290		}
	50104	#define vmlsq_laneq_s16(__p0_480, __p1_480, __p2_480, __p3_480) __extension__ ({ \
	50105	int16x8_t __ret_480; \
	50106	int16x8_t __s0_480 = __p0_480; \
	50107	int16x8_t __s1_480 = __p1_480; \
	50108	int16x8_t __s2_480 = __p2_480; \
	50109	int16x8_t __rev0_480; __rev0_480 = __builtin_shufflevector(__s0_480, __s0_480, 7, 6, 5, 4, 3, 2, 1, 0); \
	50110	int16x8_t __rev1_480; __rev1_480 = __builtin_shufflevector(__s1_480, __s1_480, 7, 6, 5, 4, 3, 2, 1, 0); \
	50111	int16x8_t __rev2_480; __rev2_480 = __builtin_shufflevector(__s2_480, __s2_480, 7, 6, 5, 4, 3, 2, 1, 0); \
	50112	__ret_480 = __rev0_480 - __rev1_480 * __noswap_splatq_laneq_s16(__rev2_480, __p3_480); \
	50113	__ret_480 = __builtin_shufflevector(__ret_480, __ret_480, 7, 6, 5, 4, 3, 2, 1, 0); \
	50114	__ret_480; \
	50115	})
49291	50116	#endif
49292	50117
49293	50118	#ifdef __LITTLE_ENDIAN__
49294		__ai uint32x2_t vcgez_s32(int32x2_t __p0) {
49295		uint32x2_t __ret;
49296		__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 18);
49297		return __ret;
49298		}
	50119	#define vmls_laneq_u32(__p0_481, __p1_481, __p2_481, __p3_481) __extension__ ({ \
	50120	uint32x2_t __ret_481; \
	50121	uint32x2_t __s0_481 = __p0_481; \
	50122	uint32x2_t __s1_481 = __p1_481; \
	50123	uint32x4_t __s2_481 = __p2_481; \
	50124	__ret_481 = __s0_481 - __s1_481 * splat_laneq_u32(__s2_481, __p3_481); \
	50125	__ret_481; \
	50126	})
49299	50127	#else
49300		__ai uint32x2_t vcgez_s32(int32x2_t __p0) {
49301		uint32x2_t __ret;
49302		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49303		__ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 18);
49304		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49305		return __ret;
49306		}
	50128	#define vmls_laneq_u32(__p0_482, __p1_482, __p2_482, __p3_482) __extension__ ({ \
	50129	uint32x2_t __ret_482; \
	50130	uint32x2_t __s0_482 = __p0_482; \
	50131	uint32x2_t __s1_482 = __p1_482; \
	50132	uint32x4_t __s2_482 = __p2_482; \
	50133	uint32x2_t __rev0_482; __rev0_482 = __builtin_shufflevector(__s0_482, __s0_482, 1, 0); \
	50134	uint32x2_t __rev1_482; __rev1_482 = __builtin_shufflevector(__s1_482, __s1_482, 1, 0); \
	50135	uint32x4_t __rev2_482; __rev2_482 = __builtin_shufflevector(__s2_482, __s2_482, 3, 2, 1, 0); \
	50136	__ret_482 = __rev0_482 - __rev1_482 * __noswap_splat_laneq_u32(__rev2_482, __p3_482); \
	50137	__ret_482 = __builtin_shufflevector(__ret_482, __ret_482, 1, 0); \
	50138	__ret_482; \
	50139	})
49307	50140	#endif
49308	50141
49309		__ai uint64x1_t vcgez_s64(int64x1_t __p0) {
49310		uint64x1_t __ret;
49311		__ret = (uint64x1_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 19);
49312		return __ret;
49313		}
49314	50142	#ifdef __LITTLE_ENDIAN__
49315		__ai uint16x4_t vcgez_s16(int16x4_t __p0) {
49316		uint16x4_t __ret;
49317		__ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__p0, 17);
49318		return __ret;
49319		}
	50143	#define vmls_laneq_u16(__p0_483, __p1_483, __p2_483, __p3_483) __extension__ ({ \
	50144	uint16x4_t __ret_483; \
	50145	uint16x4_t __s0_483 = __p0_483; \
	50146	uint16x4_t __s1_483 = __p1_483; \
	50147	uint16x8_t __s2_483 = __p2_483; \
	50148	__ret_483 = __s0_483 - __s1_483 * splat_laneq_u16(__s2_483, __p3_483); \
	50149	__ret_483; \
	50150	})
49320	50151	#else
49321		__ai uint16x4_t vcgez_s16(int16x4_t __p0) {
49322		uint16x4_t __ret;
49323		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49324		__ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 17);
49325		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49326		return __ret;
49327		}
	50152	#define vmls_laneq_u16(__p0_484, __p1_484, __p2_484, __p3_484) __extension__ ({ \
	50153	uint16x4_t __ret_484; \
	50154	uint16x4_t __s0_484 = __p0_484; \
	50155	uint16x4_t __s1_484 = __p1_484; \
	50156	uint16x8_t __s2_484 = __p2_484; \
	50157	uint16x4_t __rev0_484; __rev0_484 = __builtin_shufflevector(__s0_484, __s0_484, 3, 2, 1, 0); \
	50158	uint16x4_t __rev1_484; __rev1_484 = __builtin_shufflevector(__s1_484, __s1_484, 3, 2, 1, 0); \
	50159	uint16x8_t __rev2_484; __rev2_484 = __builtin_shufflevector(__s2_484, __s2_484, 7, 6, 5, 4, 3, 2, 1, 0); \
	50160	__ret_484 = __rev0_484 - __rev1_484 * __noswap_splat_laneq_u16(__rev2_484, __p3_484); \
	50161	__ret_484 = __builtin_shufflevector(__ret_484, __ret_484, 3, 2, 1, 0); \
	50162	__ret_484; \
	50163	})
49328	50164	#endif
49329	50165
49330		__ai uint64_t vcgezd_s64(int64_t __p0) {
49331		uint64_t __ret;
49332		__ret = (uint64_t) __builtin_neon_vcgezd_s64(__p0);
49333		return __ret;
49334		}
49335		__ai uint64_t vcgezd_f64(float64_t __p0) {
49336		uint64_t __ret;
49337		__ret = (uint64_t) __builtin_neon_vcgezd_f64(__p0);
49338		return __ret;
49339		}
49340		__ai uint32_t vcgezs_f32(float32_t __p0) {
49341		uint32_t __ret;
49342		__ret = (uint32_t) __builtin_neon_vcgezs_f32(__p0);
49343		return __ret;
49344		}
49345	50166	#ifdef __LITTLE_ENDIAN__
49346		__ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49347		uint64x2_t __ret;
49348		__ret = (uint64x2_t)(__p0 > __p1);
49349		return __ret;
49350		}
	50167	#define vmls_laneq_f32(__p0_485, __p1_485, __p2_485, __p3_485) __extension__ ({ \
	50168	float32x2_t __ret_485; \
	50169	float32x2_t __s0_485 = __p0_485; \
	50170	float32x2_t __s1_485 = __p1_485; \
	50171	float32x4_t __s2_485 = __p2_485; \
	50172	__ret_485 = __s0_485 - __s1_485 * splat_laneq_f32(__s2_485, __p3_485); \
	50173	__ret_485; \
	50174	})
49351	50175	#else
49352		__ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49353		uint64x2_t __ret;
49354		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49355		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49356		__ret = (uint64x2_t)(__rev0 > __rev1);
49357		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49358		return __ret;
49359		}
	50176	#define vmls_laneq_f32(__p0_486, __p1_486, __p2_486, __p3_486) __extension__ ({ \
	50177	float32x2_t __ret_486; \
	50178	float32x2_t __s0_486 = __p0_486; \
	50179	float32x2_t __s1_486 = __p1_486; \
	50180	float32x4_t __s2_486 = __p2_486; \
	50181	float32x2_t __rev0_486; __rev0_486 = __builtin_shufflevector(__s0_486, __s0_486, 1, 0); \
	50182	float32x2_t __rev1_486; __rev1_486 = __builtin_shufflevector(__s1_486, __s1_486, 1, 0); \
	50183	float32x4_t __rev2_486; __rev2_486 = __builtin_shufflevector(__s2_486, __s2_486, 3, 2, 1, 0); \
	50184	__ret_486 = __rev0_486 - __rev1_486 * __noswap_splat_laneq_f32(__rev2_486, __p3_486); \
	50185	__ret_486 = __builtin_shufflevector(__ret_486, __ret_486, 1, 0); \
	50186	__ret_486; \
	50187	})
49360	50188	#endif
49361	50189
49362	50190	#ifdef __LITTLE_ENDIAN__
49363		__ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
49364		uint64x2_t __ret;
49365		__ret = (uint64x2_t)(__p0 > __p1);
49366		return __ret;
49367		}
	50191	#define vmls_laneq_s32(__p0_487, __p1_487, __p2_487, __p3_487) __extension__ ({ \
	50192	int32x2_t __ret_487; \
	50193	int32x2_t __s0_487 = __p0_487; \
	50194	int32x2_t __s1_487 = __p1_487; \
	50195	int32x4_t __s2_487 = __p2_487; \
	50196	__ret_487 = __s0_487 - __s1_487 * splat_laneq_s32(__s2_487, __p3_487); \
	50197	__ret_487; \
	50198	})
49368	50199	#else
49369		__ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
49370		uint64x2_t __ret;
49371		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49372		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49373		__ret = (uint64x2_t)(__rev0 > __rev1);
49374		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49375		return __ret;
49376		}
	50200	#define vmls_laneq_s32(__p0_488, __p1_488, __p2_488, __p3_488) __extension__ ({ \
	50201	int32x2_t __ret_488; \
	50202	int32x2_t __s0_488 = __p0_488; \
	50203	int32x2_t __s1_488 = __p1_488; \
	50204	int32x4_t __s2_488 = __p2_488; \
	50205	int32x2_t __rev0_488; __rev0_488 = __builtin_shufflevector(__s0_488, __s0_488, 1, 0); \
	50206	int32x2_t __rev1_488; __rev1_488 = __builtin_shufflevector(__s1_488, __s1_488, 1, 0); \
	50207	int32x4_t __rev2_488; __rev2_488 = __builtin_shufflevector(__s2_488, __s2_488, 3, 2, 1, 0); \
	50208	__ret_488 = __rev0_488 - __rev1_488 * __noswap_splat_laneq_s32(__rev2_488, __p3_488); \
	50209	__ret_488 = __builtin_shufflevector(__ret_488, __ret_488, 1, 0); \
	50210	__ret_488; \
	50211	})
49377	50212	#endif
49378	50213
49379	50214	#ifdef __LITTLE_ENDIAN__
49380		__ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
49381		uint64x2_t __ret;
49382		__ret = (uint64x2_t)(__p0 > __p1);
49383		return __ret;
49384		}
	50215	#define vmls_laneq_s16(__p0_489, __p1_489, __p2_489, __p3_489) __extension__ ({ \
	50216	int16x4_t __ret_489; \
	50217	int16x4_t __s0_489 = __p0_489; \
	50218	int16x4_t __s1_489 = __p1_489; \
	50219	int16x8_t __s2_489 = __p2_489; \
	50220	__ret_489 = __s0_489 - __s1_489 * splat_laneq_s16(__s2_489, __p3_489); \
	50221	__ret_489; \
	50222	})
49385	50223	#else
49386		__ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
49387		uint64x2_t __ret;
49388		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49389		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49390		__ret = (uint64x2_t)(__rev0 > __rev1);
49391		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49392		return __ret;
49393		}
	50224	#define vmls_laneq_s16(__p0_490, __p1_490, __p2_490, __p3_490) __extension__ ({ \
	50225	int16x4_t __ret_490; \
	50226	int16x4_t __s0_490 = __p0_490; \
	50227	int16x4_t __s1_490 = __p1_490; \
	50228	int16x8_t __s2_490 = __p2_490; \
	50229	int16x4_t __rev0_490; __rev0_490 = __builtin_shufflevector(__s0_490, __s0_490, 3, 2, 1, 0); \
	50230	int16x4_t __rev1_490; __rev1_490 = __builtin_shufflevector(__s1_490, __s1_490, 3, 2, 1, 0); \
	50231	int16x8_t __rev2_490; __rev2_490 = __builtin_shufflevector(__s2_490, __s2_490, 7, 6, 5, 4, 3, 2, 1, 0); \
	50232	__ret_490 = __rev0_490 - __rev1_490 * __noswap_splat_laneq_s16(__rev2_490, __p3_490); \
	50233	__ret_490 = __builtin_shufflevector(__ret_490, __ret_490, 3, 2, 1, 0); \
	50234	__ret_490; \
	50235	})
49394	50236	#endif
49395	50237
49396		__ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
49397		uint64x1_t __ret;
49398		__ret = (uint64x1_t)(__p0 > __p1);
49399		return __ret;
49400		}
49401		__ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
49402		uint64x1_t __ret;
49403		__ret = (uint64x1_t)(__p0 > __p1);
49404		return __ret;
49405		}
49406		__ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
49407		uint64x1_t __ret;
49408		__ret = (uint64x1_t)(__p0 > __p1);
49409		return __ret;
49410		}
49411		__ai uint64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
49412		uint64_t __ret;
49413		__ret = (uint64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
49414		return __ret;
49415		}
49416		__ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
49417		uint64_t __ret;
49418		__ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
49419		return __ret;
49420		}
49421		__ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
49422		uint64_t __ret;
49423		__ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
49424		return __ret;
49425		}
49426		__ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
49427		uint32_t __ret;
49428		__ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
49429		return __ret;
49430		}
49431	50238	#ifdef __LITTLE_ENDIAN__
49432		__ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
49433		uint8x16_t __ret;
49434		__ret = (uint8x16_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 48);
49435		return __ret;
49436		}
	50239	#define vmlsl_high_lane_u32(__p0_491, __p1_491, __p2_491, __p3_491) __extension__ ({ \
	50240	uint64x2_t __ret_491; \
	50241	uint64x2_t __s0_491 = __p0_491; \
	50242	uint32x4_t __s1_491 = __p1_491; \
	50243	uint32x2_t __s2_491 = __p2_491; \
	50244	__ret_491 = __s0_491 - vmull_u32(vget_high_u32(__s1_491), splat_lane_u32(__s2_491, __p3_491)); \
	50245	__ret_491; \
	50246	})
49437	50247	#else
49438		__ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
49439		uint8x16_t __ret;
49440		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49441		__ret = (uint8x16_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 48);
49442		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49443		return __ret;
49444		}
	50248	#define vmlsl_high_lane_u32(__p0_492, __p1_492, __p2_492, __p3_492) __extension__ ({ \
	50249	uint64x2_t __ret_492; \
	50250	uint64x2_t __s0_492 = __p0_492; \
	50251	uint32x4_t __s1_492 = __p1_492; \
	50252	uint32x2_t __s2_492 = __p2_492; \
	50253	uint64x2_t __rev0_492; __rev0_492 = __builtin_shufflevector(__s0_492, __s0_492, 1, 0); \
	50254	uint32x4_t __rev1_492; __rev1_492 = __builtin_shufflevector(__s1_492, __s1_492, 3, 2, 1, 0); \
	50255	uint32x2_t __rev2_492; __rev2_492 = __builtin_shufflevector(__s2_492, __s2_492, 1, 0); \
	50256	__ret_492 = __rev0_492 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_492), __noswap_splat_lane_u32(__rev2_492, __p3_492)); \
	50257	__ret_492 = __builtin_shufflevector(__ret_492, __ret_492, 1, 0); \
	50258	__ret_492; \
	50259	})
49445	50260	#endif
49446	50261
49447	50262	#ifdef __LITTLE_ENDIAN__
49448		__ai uint64x2_t vcgtzq_f64(float64x2_t __p0) {
49449		uint64x2_t __ret;
49450		__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 51);
49451		return __ret;
49452		}
	50263	#define vmlsl_high_lane_u16(__p0_493, __p1_493, __p2_493, __p3_493) __extension__ ({ \
	50264	uint32x4_t __ret_493; \
	50265	uint32x4_t __s0_493 = __p0_493; \
	50266	uint16x8_t __s1_493 = __p1_493; \
	50267	uint16x4_t __s2_493 = __p2_493; \
	50268	__ret_493 = __s0_493 - vmull_u16(vget_high_u16(__s1_493), splat_lane_u16(__s2_493, __p3_493)); \
	50269	__ret_493; \
	50270	})
49453	50271	#else
49454		__ai uint64x2_t vcgtzq_f64(float64x2_t __p0) {
49455		uint64x2_t __ret;
49456		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49457		__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 51);
49458		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49459		return __ret;
49460		}
	50272	#define vmlsl_high_lane_u16(__p0_494, __p1_494, __p2_494, __p3_494) __extension__ ({ \
	50273	uint32x4_t __ret_494; \
	50274	uint32x4_t __s0_494 = __p0_494; \
	50275	uint16x8_t __s1_494 = __p1_494; \
	50276	uint16x4_t __s2_494 = __p2_494; \
	50277	uint32x4_t __rev0_494; __rev0_494 = __builtin_shufflevector(__s0_494, __s0_494, 3, 2, 1, 0); \
	50278	uint16x8_t __rev1_494; __rev1_494 = __builtin_shufflevector(__s1_494, __s1_494, 7, 6, 5, 4, 3, 2, 1, 0); \
	50279	uint16x4_t __rev2_494; __rev2_494 = __builtin_shufflevector(__s2_494, __s2_494, 3, 2, 1, 0); \
	50280	__ret_494 = __rev0_494 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_494), __noswap_splat_lane_u16(__rev2_494, __p3_494)); \
	50281	__ret_494 = __builtin_shufflevector(__ret_494, __ret_494, 3, 2, 1, 0); \
	50282	__ret_494; \
	50283	})
49461	50284	#endif
49462	50285
49463	50286	#ifdef __LITTLE_ENDIAN__
49464		__ai uint32x4_t vcgtzq_f32(float32x4_t __p0) {
49465		uint32x4_t __ret;
49466		__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 50);
49467		return __ret;
49468		}
	50287	#define vmlsl_high_lane_s32(__p0_495, __p1_495, __p2_495, __p3_495) __extension__ ({ \
	50288	int64x2_t __ret_495; \
	50289	int64x2_t __s0_495 = __p0_495; \
	50290	int32x4_t __s1_495 = __p1_495; \
	50291	int32x2_t __s2_495 = __p2_495; \
	50292	__ret_495 = __s0_495 - vmull_s32(vget_high_s32(__s1_495), splat_lane_s32(__s2_495, __p3_495)); \
	50293	__ret_495; \
	50294	})
49469	50295	#else
49470		__ai uint32x4_t vcgtzq_f32(float32x4_t __p0) {
49471		uint32x4_t __ret;
49472		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49473		__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 50);
49474		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49475		return __ret;
49476		}
	50296	#define vmlsl_high_lane_s32(__p0_496, __p1_496, __p2_496, __p3_496) __extension__ ({ \
	50297	int64x2_t __ret_496; \
	50298	int64x2_t __s0_496 = __p0_496; \
	50299	int32x4_t __s1_496 = __p1_496; \
	50300	int32x2_t __s2_496 = __p2_496; \
	50301	int64x2_t __rev0_496; __rev0_496 = __builtin_shufflevector(__s0_496, __s0_496, 1, 0); \
	50302	int32x4_t __rev1_496; __rev1_496 = __builtin_shufflevector(__s1_496, __s1_496, 3, 2, 1, 0); \
	50303	int32x2_t __rev2_496; __rev2_496 = __builtin_shufflevector(__s2_496, __s2_496, 1, 0); \
	50304	__ret_496 = __rev0_496 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_496), __noswap_splat_lane_s32(__rev2_496, __p3_496)); \
	50305	__ret_496 = __builtin_shufflevector(__ret_496, __ret_496, 1, 0); \
	50306	__ret_496; \
	50307	})
49477	50308	#endif
49478	50309
49479	50310	#ifdef __LITTLE_ENDIAN__
49480		__ai uint32x4_t vcgtzq_s32(int32x4_t __p0) {
49481		uint32x4_t __ret;
49482		__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 50);
49483		return __ret;
49484		}
	50311	#define vmlsl_high_lane_s16(__p0_497, __p1_497, __p2_497, __p3_497) __extension__ ({ \
	50312	int32x4_t __ret_497; \
	50313	int32x4_t __s0_497 = __p0_497; \
	50314	int16x8_t __s1_497 = __p1_497; \
	50315	int16x4_t __s2_497 = __p2_497; \
	50316	__ret_497 = __s0_497 - vmull_s16(vget_high_s16(__s1_497), splat_lane_s16(__s2_497, __p3_497)); \
	50317	__ret_497; \
	50318	})
49485	50319	#else
49486		__ai uint32x4_t vcgtzq_s32(int32x4_t __p0) {
49487		uint32x4_t __ret;
49488		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49489		__ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 50);
49490		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49491		return __ret;
49492		}
	50320	#define vmlsl_high_lane_s16(__p0_498, __p1_498, __p2_498, __p3_498) __extension__ ({ \
	50321	int32x4_t __ret_498; \
	50322	int32x4_t __s0_498 = __p0_498; \
	50323	int16x8_t __s1_498 = __p1_498; \
	50324	int16x4_t __s2_498 = __p2_498; \
	50325	int32x4_t __rev0_498; __rev0_498 = __builtin_shufflevector(__s0_498, __s0_498, 3, 2, 1, 0); \
	50326	int16x8_t __rev1_498; __rev1_498 = __builtin_shufflevector(__s1_498, __s1_498, 7, 6, 5, 4, 3, 2, 1, 0); \
	50327	int16x4_t __rev2_498; __rev2_498 = __builtin_shufflevector(__s2_498, __s2_498, 3, 2, 1, 0); \
	50328	__ret_498 = __rev0_498 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_498), __noswap_splat_lane_s16(__rev2_498, __p3_498)); \
	50329	__ret_498 = __builtin_shufflevector(__ret_498, __ret_498, 3, 2, 1, 0); \
	50330	__ret_498; \
	50331	})
49493	50332	#endif
49494	50333
49495	50334	#ifdef __LITTLE_ENDIAN__
49496		__ai uint64x2_t vcgtzq_s64(int64x2_t __p0) {
49497		uint64x2_t __ret;
49498		__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 51);
49499		return __ret;
49500		}
	50335	#define vmlsl_high_laneq_u32(__p0_499, __p1_499, __p2_499, __p3_499) __extension__ ({ \
	50336	uint64x2_t __ret_499; \
	50337	uint64x2_t __s0_499 = __p0_499; \
	50338	uint32x4_t __s1_499 = __p1_499; \
	50339	uint32x4_t __s2_499 = __p2_499; \
	50340	__ret_499 = __s0_499 - vmull_u32(vget_high_u32(__s1_499), splat_laneq_u32(__s2_499, __p3_499)); \
	50341	__ret_499; \
	50342	})
49501	50343	#else
49502		__ai uint64x2_t vcgtzq_s64(int64x2_t __p0) {
49503		uint64x2_t __ret;
49504		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49505		__ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 51);
49506		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49507		return __ret;
49508		}
	50344	#define vmlsl_high_laneq_u32(__p0_500, __p1_500, __p2_500, __p3_500) __extension__ ({ \
	50345	uint64x2_t __ret_500; \
	50346	uint64x2_t __s0_500 = __p0_500; \
	50347	uint32x4_t __s1_500 = __p1_500; \
	50348	uint32x4_t __s2_500 = __p2_500; \
	50349	uint64x2_t __rev0_500; __rev0_500 = __builtin_shufflevector(__s0_500, __s0_500, 1, 0); \
	50350	uint32x4_t __rev1_500; __rev1_500 = __builtin_shufflevector(__s1_500, __s1_500, 3, 2, 1, 0); \
	50351	uint32x4_t __rev2_500; __rev2_500 = __builtin_shufflevector(__s2_500, __s2_500, 3, 2, 1, 0); \
	50352	__ret_500 = __rev0_500 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_500), __noswap_splat_laneq_u32(__rev2_500, __p3_500)); \
	50353	__ret_500 = __builtin_shufflevector(__ret_500, __ret_500, 1, 0); \
	50354	__ret_500; \
	50355	})
49509	50356	#endif
49510	50357
49511	50358	#ifdef __LITTLE_ENDIAN__
49512		__ai uint16x8_t vcgtzq_s16(int16x8_t __p0) {
49513		uint16x8_t __ret;
49514		__ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__p0, 49);
49515		return __ret;
49516		}
	50359	#define vmlsl_high_laneq_u16(__p0_501, __p1_501, __p2_501, __p3_501) __extension__ ({ \
	50360	uint32x4_t __ret_501; \
	50361	uint32x4_t __s0_501 = __p0_501; \
	50362	uint16x8_t __s1_501 = __p1_501; \
	50363	uint16x8_t __s2_501 = __p2_501; \
	50364	__ret_501 = __s0_501 - vmull_u16(vget_high_u16(__s1_501), splat_laneq_u16(__s2_501, __p3_501)); \
	50365	__ret_501; \
	50366	})
49517	50367	#else
49518		__ai uint16x8_t vcgtzq_s16(int16x8_t __p0) {
49519		uint16x8_t __ret;
49520		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49521		__ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 49);
49522		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49523		return __ret;
49524		}
	50368	#define vmlsl_high_laneq_u16(__p0_502, __p1_502, __p2_502, __p3_502) __extension__ ({ \
	50369	uint32x4_t __ret_502; \
	50370	uint32x4_t __s0_502 = __p0_502; \
	50371	uint16x8_t __s1_502 = __p1_502; \
	50372	uint16x8_t __s2_502 = __p2_502; \
	50373	uint32x4_t __rev0_502; __rev0_502 = __builtin_shufflevector(__s0_502, __s0_502, 3, 2, 1, 0); \
	50374	uint16x8_t __rev1_502; __rev1_502 = __builtin_shufflevector(__s1_502, __s1_502, 7, 6, 5, 4, 3, 2, 1, 0); \
	50375	uint16x8_t __rev2_502; __rev2_502 = __builtin_shufflevector(__s2_502, __s2_502, 7, 6, 5, 4, 3, 2, 1, 0); \
	50376	__ret_502 = __rev0_502 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_502), __noswap_splat_laneq_u16(__rev2_502, __p3_502)); \
	50377	__ret_502 = __builtin_shufflevector(__ret_502, __ret_502, 3, 2, 1, 0); \
	50378	__ret_502; \
	50379	})
49525	50380	#endif
49526	50381
49527	50382	#ifdef __LITTLE_ENDIAN__
49528		__ai uint8x8_t vcgtz_s8(int8x8_t __p0) {
49529		uint8x8_t __ret;
49530		__ret = (uint8x8_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 16);
49531		return __ret;
49532		}
	50383	#define vmlsl_high_laneq_s32(__p0_503, __p1_503, __p2_503, __p3_503) __extension__ ({ \
	50384	int64x2_t __ret_503; \
	50385	int64x2_t __s0_503 = __p0_503; \
	50386	int32x4_t __s1_503 = __p1_503; \
	50387	int32x4_t __s2_503 = __p2_503; \
	50388	__ret_503 = __s0_503 - vmull_s32(vget_high_s32(__s1_503), splat_laneq_s32(__s2_503, __p3_503)); \
	50389	__ret_503; \
	50390	})
49533	50391	#else
49534		__ai uint8x8_t vcgtz_s8(int8x8_t __p0) {
49535		uint8x8_t __ret;
49536		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49537		__ret = (uint8x8_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 16);
49538		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49539		return __ret;
49540		}
	50392	#define vmlsl_high_laneq_s32(__p0_504, __p1_504, __p2_504, __p3_504) __extension__ ({ \
	50393	int64x2_t __ret_504; \
	50394	int64x2_t __s0_504 = __p0_504; \
	50395	int32x4_t __s1_504 = __p1_504; \
	50396	int32x4_t __s2_504 = __p2_504; \
	50397	int64x2_t __rev0_504; __rev0_504 = __builtin_shufflevector(__s0_504, __s0_504, 1, 0); \
	50398	int32x4_t __rev1_504; __rev1_504 = __builtin_shufflevector(__s1_504, __s1_504, 3, 2, 1, 0); \
	50399	int32x4_t __rev2_504; __rev2_504 = __builtin_shufflevector(__s2_504, __s2_504, 3, 2, 1, 0); \
	50400	__ret_504 = __rev0_504 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_504), __noswap_splat_laneq_s32(__rev2_504, __p3_504)); \
	50401	__ret_504 = __builtin_shufflevector(__ret_504, __ret_504, 1, 0); \
	50402	__ret_504; \
	50403	})
49541	50404	#endif
49542	50405
49543		__ai uint64x1_t vcgtz_f64(float64x1_t __p0) {
49544		uint64x1_t __ret;
49545		__ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
49546		return __ret;
49547		}
49548	50406	#ifdef __LITTLE_ENDIAN__
49549		__ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
49550		uint32x2_t __ret;
49551		__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 18);
49552		return __ret;
49553		}
	50407	#define vmlsl_high_laneq_s16(__p0_505, __p1_505, __p2_505, __p3_505) __extension__ ({ \
	50408	int32x4_t __ret_505; \
	50409	int32x4_t __s0_505 = __p0_505; \
	50410	int16x8_t __s1_505 = __p1_505; \
	50411	int16x8_t __s2_505 = __p2_505; \
	50412	__ret_505 = __s0_505 - vmull_s16(vget_high_s16(__s1_505), splat_laneq_s16(__s2_505, __p3_505)); \
	50413	__ret_505; \
	50414	})
49554	50415	#else
49555		__ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
49556		uint32x2_t __ret;
49557		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49558		__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 18);
49559		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49560		return __ret;
49561		}
	50416	#define vmlsl_high_laneq_s16(__p0_506, __p1_506, __p2_506, __p3_506) __extension__ ({ \
	50417	int32x4_t __ret_506; \
	50418	int32x4_t __s0_506 = __p0_506; \
	50419	int16x8_t __s1_506 = __p1_506; \
	50420	int16x8_t __s2_506 = __p2_506; \
	50421	int32x4_t __rev0_506; __rev0_506 = __builtin_shufflevector(__s0_506, __s0_506, 3, 2, 1, 0); \
	50422	int16x8_t __rev1_506; __rev1_506 = __builtin_shufflevector(__s1_506, __s1_506, 7, 6, 5, 4, 3, 2, 1, 0); \
	50423	int16x8_t __rev2_506; __rev2_506 = __builtin_shufflevector(__s2_506, __s2_506, 7, 6, 5, 4, 3, 2, 1, 0); \
	50424	__ret_506 = __rev0_506 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_506), __noswap_splat_laneq_s16(__rev2_506, __p3_506)); \
	50425	__ret_506 = __builtin_shufflevector(__ret_506, __ret_506, 3, 2, 1, 0); \
	50426	__ret_506; \
	50427	})
49562	50428	#endif
49563	50429
49564	50430	#ifdef __LITTLE_ENDIAN__
49565		__ai uint32x2_t vcgtz_s32(int32x2_t __p0) {
49566		uint32x2_t __ret;
49567		__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 18);
49568		return __ret;
49569		}
	50431	#define vmlsl_laneq_u32(__p0_507, __p1_507, __p2_507, __p3_507) __extension__ ({ \
	50432	uint64x2_t __ret_507; \
	50433	uint64x2_t __s0_507 = __p0_507; \
	50434	uint32x2_t __s1_507 = __p1_507; \
	50435	uint32x4_t __s2_507 = __p2_507; \
	50436	__ret_507 = __s0_507 - vmull_u32(__s1_507, splat_laneq_u32(__s2_507, __p3_507)); \
	50437	__ret_507; \
	50438	})
49570	50439	#else
49571		__ai uint32x2_t vcgtz_s32(int32x2_t __p0) {
49572		uint32x2_t __ret;
49573		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49574		__ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 18);
49575		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49576		return __ret;
49577		}
	50440	#define vmlsl_laneq_u32(__p0_508, __p1_508, __p2_508, __p3_508) __extension__ ({ \
	50441	uint64x2_t __ret_508; \
	50442	uint64x2_t __s0_508 = __p0_508; \
	50443	uint32x2_t __s1_508 = __p1_508; \
	50444	uint32x4_t __s2_508 = __p2_508; \
	50445	uint64x2_t __rev0_508; __rev0_508 = __builtin_shufflevector(__s0_508, __s0_508, 1, 0); \
	50446	uint32x2_t __rev1_508; __rev1_508 = __builtin_shufflevector(__s1_508, __s1_508, 1, 0); \
	50447	uint32x4_t __rev2_508; __rev2_508 = __builtin_shufflevector(__s2_508, __s2_508, 3, 2, 1, 0); \
	50448	__ret_508 = __rev0_508 - __noswap_vmull_u32(__rev1_508, __noswap_splat_laneq_u32(__rev2_508, __p3_508)); \
	50449	__ret_508 = __builtin_shufflevector(__ret_508, __ret_508, 1, 0); \
	50450	__ret_508; \
	50451	})
49578	50452	#endif
49579	50453
49580		__ai uint64x1_t vcgtz_s64(int64x1_t __p0) {
49581		uint64x1_t __ret;
49582		__ret = (uint64x1_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 19);
49583		return __ret;
49584		}
49585	50454	#ifdef __LITTLE_ENDIAN__
49586		__ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
49587		uint16x4_t __ret;
49588		__ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__p0, 17);
49589		return __ret;
49590		}
	50455	#define vmlsl_laneq_u16(__p0_509, __p1_509, __p2_509, __p3_509) __extension__ ({ \
	50456	uint32x4_t __ret_509; \
	50457	uint32x4_t __s0_509 = __p0_509; \
	50458	uint16x4_t __s1_509 = __p1_509; \
	50459	uint16x8_t __s2_509 = __p2_509; \
	50460	__ret_509 = __s0_509 - vmull_u16(__s1_509, splat_laneq_u16(__s2_509, __p3_509)); \
	50461	__ret_509; \
	50462	})
49591	50463	#else
49592		__ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
49593		uint16x4_t __ret;
49594		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49595		__ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 17);
49596		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49597		return __ret;
49598		}
	50464	#define vmlsl_laneq_u16(__p0_510, __p1_510, __p2_510, __p3_510) __extension__ ({ \
	50465	uint32x4_t __ret_510; \
	50466	uint32x4_t __s0_510 = __p0_510; \
	50467	uint16x4_t __s1_510 = __p1_510; \
	50468	uint16x8_t __s2_510 = __p2_510; \
	50469	uint32x4_t __rev0_510; __rev0_510 = __builtin_shufflevector(__s0_510, __s0_510, 3, 2, 1, 0); \
	50470	uint16x4_t __rev1_510; __rev1_510 = __builtin_shufflevector(__s1_510, __s1_510, 3, 2, 1, 0); \
	50471	uint16x8_t __rev2_510; __rev2_510 = __builtin_shufflevector(__s2_510, __s2_510, 7, 6, 5, 4, 3, 2, 1, 0); \
	50472	__ret_510 = __rev0_510 - __noswap_vmull_u16(__rev1_510, __noswap_splat_laneq_u16(__rev2_510, __p3_510)); \
	50473	__ret_510 = __builtin_shufflevector(__ret_510, __ret_510, 3, 2, 1, 0); \
	50474	__ret_510; \
	50475	})
	50476	#endif
	50477
	50478	#ifdef __LITTLE_ENDIAN__
	50479	#define vmlsl_laneq_s32(__p0_511, __p1_511, __p2_511, __p3_511) __extension__ ({ \
	50480	int64x2_t __ret_511; \
	50481	int64x2_t __s0_511 = __p0_511; \
	50482	int32x2_t __s1_511 = __p1_511; \
	50483	int32x4_t __s2_511 = __p2_511; \
	50484	__ret_511 = __s0_511 - vmull_s32(__s1_511, splat_laneq_s32(__s2_511, __p3_511)); \
	50485	__ret_511; \
	50486	})
	50487	#else
	50488	#define vmlsl_laneq_s32(__p0_512, __p1_512, __p2_512, __p3_512) __extension__ ({ \
	50489	int64x2_t __ret_512; \
	50490	int64x2_t __s0_512 = __p0_512; \
	50491	int32x2_t __s1_512 = __p1_512; \
	50492	int32x4_t __s2_512 = __p2_512; \
	50493	int64x2_t __rev0_512; __rev0_512 = __builtin_shufflevector(__s0_512, __s0_512, 1, 0); \
	50494	int32x2_t __rev1_512; __rev1_512 = __builtin_shufflevector(__s1_512, __s1_512, 1, 0); \
	50495	int32x4_t __rev2_512; __rev2_512 = __builtin_shufflevector(__s2_512, __s2_512, 3, 2, 1, 0); \
	50496	__ret_512 = __rev0_512 - __noswap_vmull_s32(__rev1_512, __noswap_splat_laneq_s32(__rev2_512, __p3_512)); \
	50497	__ret_512 = __builtin_shufflevector(__ret_512, __ret_512, 1, 0); \
	50498	__ret_512; \
	50499	})
49599	50500	#endif
49600	50501
49601		__ai uint64_t vcgtzd_s64(int64_t __p0) {
49602		uint64_t __ret;
49603		__ret = (uint64_t) __builtin_neon_vcgtzd_s64(__p0);
49604		return __ret;
49605		}
49606		__ai uint64_t vcgtzd_f64(float64_t __p0) {
49607		uint64_t __ret;
49608		__ret = (uint64_t) __builtin_neon_vcgtzd_f64(__p0);
49609		return __ret;
49610		}
49611		__ai uint32_t vcgtzs_f32(float32_t __p0) {
49612		uint32_t __ret;
49613		__ret = (uint32_t) __builtin_neon_vcgtzs_f32(__p0);
49614		return __ret;
49615		}
49616	50502	#ifdef __LITTLE_ENDIAN__
49617		__ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49618		uint64x2_t __ret;
49619		__ret = (uint64x2_t)(__p0 <= __p1);
49620		return __ret;
49621		}
	50503	#define vmlsl_laneq_s16(__p0_513, __p1_513, __p2_513, __p3_513) __extension__ ({ \
	50504	int32x4_t __ret_513; \
	50505	int32x4_t __s0_513 = __p0_513; \
	50506	int16x4_t __s1_513 = __p1_513; \
	50507	int16x8_t __s2_513 = __p2_513; \
	50508	__ret_513 = __s0_513 - vmull_s16(__s1_513, splat_laneq_s16(__s2_513, __p3_513)); \
	50509	__ret_513; \
	50510	})
49622	50511	#else
49623		__ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49624		uint64x2_t __ret;
49625		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49626		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49627		__ret = (uint64x2_t)(__rev0 <= __rev1);
49628		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49629		return __ret;
49630		}
	50512	#define vmlsl_laneq_s16(__p0_514, __p1_514, __p2_514, __p3_514) __extension__ ({ \
	50513	int32x4_t __ret_514; \
	50514	int32x4_t __s0_514 = __p0_514; \
	50515	int16x4_t __s1_514 = __p1_514; \
	50516	int16x8_t __s2_514 = __p2_514; \
	50517	int32x4_t __rev0_514; __rev0_514 = __builtin_shufflevector(__s0_514, __s0_514, 3, 2, 1, 0); \
	50518	int16x4_t __rev1_514; __rev1_514 = __builtin_shufflevector(__s1_514, __s1_514, 3, 2, 1, 0); \
	50519	int16x8_t __rev2_514; __rev2_514 = __builtin_shufflevector(__s2_514, __s2_514, 7, 6, 5, 4, 3, 2, 1, 0); \
	50520	__ret_514 = __rev0_514 - __noswap_vmull_s16(__rev1_514, __noswap_splat_laneq_s16(__rev2_514, __p3_514)); \
	50521	__ret_514 = __builtin_shufflevector(__ret_514, __ret_514, 3, 2, 1, 0); \
	50522	__ret_514; \
	50523	})
49631	50524	#endif
49632	50525
	50526	__ai poly64x1_t vmov_n_p64(poly64_t __p0) {
	50527	poly64x1_t __ret;
	50528	__ret = (poly64x1_t) {__p0};
	50529	return __ret;
	50530	}
49633	50531	#ifdef __LITTLE_ENDIAN__
49634		__ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
49635		uint64x2_t __ret;
49636		__ret = (uint64x2_t)(__p0 <= __p1);
	50532	__ai poly64x2_t vmovq_n_p64(poly64_t __p0) {
	50533	poly64x2_t __ret;
	50534	__ret = (poly64x2_t) {__p0, __p0};
49637	50535	return __ret;
49638	50536	}
49639	50537	#else
49640		__ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
49641		uint64x2_t __ret;
49642		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49643		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49644		__ret = (uint64x2_t)(__rev0 <= __rev1);
	50538	__ai poly64x2_t vmovq_n_p64(poly64_t __p0) {
	50539	poly64x2_t __ret;
	50540	__ret = (poly64x2_t) {__p0, __p0};
49645	50541	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49646	50542	return __ret;
49647	50543	}
49648	50544	#endif
49649	50545
49650	50546	#ifdef __LITTLE_ENDIAN__
49651		__ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
49652		uint64x2_t __ret;
49653		__ret = (uint64x2_t)(__p0 <= __p1);
	50547	__ai float64x2_t vmovq_n_f64(float64_t __p0) {
	50548	float64x2_t __ret;
	50549	__ret = (float64x2_t) {__p0, __p0};
49654	50550	return __ret;
49655	50551	}
49656	50552	#else
49657		__ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
49658		uint64x2_t __ret;
49659		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49660		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49661		__ret = (uint64x2_t)(__rev0 <= __rev1);
	50553	__ai float64x2_t vmovq_n_f64(float64_t __p0) {
	50554	float64x2_t __ret;
	50555	__ret = (float64x2_t) {__p0, __p0};
49662	50556	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49663	50557	return __ret;
49664	50558	}
49665	50559	#endif
49666	50560
49667		__ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
49668		uint64x1_t __ret;
49669		__ret = (uint64x1_t)(__p0 <= __p1);
49670		return __ret;
49671		}
49672		__ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
49673		uint64x1_t __ret;
49674		__ret = (uint64x1_t)(__p0 <= __p1);
49675		return __ret;
49676		}
49677		__ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
49678		uint64x1_t __ret;
49679		__ret = (uint64x1_t)(__p0 <= __p1);
49680		return __ret;
49681		}
49682		__ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
49683		uint64_t __ret;
49684		__ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
	50561	__ai float64x1_t vmov_n_f64(float64_t __p0) {
	50562	float64x1_t __ret;
	50563	__ret = (float64x1_t) {__p0};
49685	50564	return __ret;
49686	50565	}
49687		__ai uint64_t vcled_s64(int64_t __p0, int64_t __p1) {
49688		uint64_t __ret;
49689		__ret = (uint64_t) __builtin_neon_vcled_s64(__p0, __p1);
49690		return __ret;
	50566	#ifdef __LITTLE_ENDIAN__
	50567	__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_515) {
	50568	uint16x8_t __ret_515;
	50569	uint8x8_t __a1_515 = vget_high_u8(__p0_515);
	50570	__ret_515 = (uint16x8_t)(vshll_n_u8(__a1_515, 0));
	50571	return __ret_515;
49691	50572	}
49692		__ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
49693		uint64_t __ret;
49694		__ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
49695		return __ret;
	50573	#else
	50574	__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_516) {
	50575	uint16x8_t __ret_516;
	50576	uint8x16_t __rev0_516; __rev0_516 = __builtin_shufflevector(__p0_516, __p0_516, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	50577	uint8x8_t __a1_516 = __noswap_vget_high_u8(__rev0_516);
	50578	__ret_516 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_516, 0));
	50579	__ret_516 = __builtin_shufflevector(__ret_516, __ret_516, 7, 6, 5, 4, 3, 2, 1, 0);
	50580	return __ret_516;
49696	50581	}
49697		__ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
49698		uint32_t __ret;
49699		__ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
49700		return __ret;
	50582	__ai uint16x8_t __noswap_vmovl_high_u8(uint8x16_t __p0_517) {
	50583	uint16x8_t __ret_517;
	50584	uint8x8_t __a1_517 = __noswap_vget_high_u8(__p0_517);
	50585	__ret_517 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_517, 0));
	50586	return __ret_517;
49701	50587	}
	50588	#endif
	50589
49702	50590	#ifdef __LITTLE_ENDIAN__
49703		__ai uint8x16_t vclezq_s8(int8x16_t __p0) {
49704		uint8x16_t __ret;
49705		__ret = (uint8x16_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 48);
49706		return __ret;
	50591	__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_518) {
	50592	uint64x2_t __ret_518;
	50593	uint32x2_t __a1_518 = vget_high_u32(__p0_518);
	50594	__ret_518 = (uint64x2_t)(vshll_n_u32(__a1_518, 0));
	50595	return __ret_518;
49707	50596	}
49708	50597	#else
49709		__ai uint8x16_t vclezq_s8(int8x16_t __p0) {
49710		uint8x16_t __ret;
49711		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49712		__ret = (uint8x16_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 48);
49713		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49714		return __ret;
	50598	__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_519) {
	50599	uint64x2_t __ret_519;
	50600	uint32x4_t __rev0_519; __rev0_519 = __builtin_shufflevector(__p0_519, __p0_519, 3, 2, 1, 0);
	50601	uint32x2_t __a1_519 = __noswap_vget_high_u32(__rev0_519);
	50602	__ret_519 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_519, 0));
	50603	__ret_519 = __builtin_shufflevector(__ret_519, __ret_519, 1, 0);
	50604	return __ret_519;
	50605	}
	50606	__ai uint64x2_t __noswap_vmovl_high_u32(uint32x4_t __p0_520) {
	50607	uint64x2_t __ret_520;
	50608	uint32x2_t __a1_520 = __noswap_vget_high_u32(__p0_520);
	50609	__ret_520 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_520, 0));
	50610	return __ret_520;
49715	50611	}
49716	50612	#endif
49717	50613
49718	50614	#ifdef __LITTLE_ENDIAN__
49719		__ai uint64x2_t vclezq_f64(float64x2_t __p0) {
49720		uint64x2_t __ret;
49721		__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 51);
49722		return __ret;
	50615	__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_521) {
	50616	uint32x4_t __ret_521;
	50617	uint16x4_t __a1_521 = vget_high_u16(__p0_521);
	50618	__ret_521 = (uint32x4_t)(vshll_n_u16(__a1_521, 0));
	50619	return __ret_521;
49723	50620	}
49724	50621	#else
49725		__ai uint64x2_t vclezq_f64(float64x2_t __p0) {
49726		uint64x2_t __ret;
49727		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49728		__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 51);
49729		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49730		return __ret;
	50622	__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_522) {
	50623	uint32x4_t __ret_522;
	50624	uint16x8_t __rev0_522; __rev0_522 = __builtin_shufflevector(__p0_522, __p0_522, 7, 6, 5, 4, 3, 2, 1, 0);
	50625	uint16x4_t __a1_522 = __noswap_vget_high_u16(__rev0_522);
	50626	__ret_522 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_522, 0));
	50627	__ret_522 = __builtin_shufflevector(__ret_522, __ret_522, 3, 2, 1, 0);
	50628	return __ret_522;
	50629	}
	50630	__ai uint32x4_t __noswap_vmovl_high_u16(uint16x8_t __p0_523) {
	50631	uint32x4_t __ret_523;
	50632	uint16x4_t __a1_523 = __noswap_vget_high_u16(__p0_523);
	50633	__ret_523 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_523, 0));
	50634	return __ret_523;
49731	50635	}
49732	50636	#endif
49733	50637
49734	50638	#ifdef __LITTLE_ENDIAN__
49735		__ai uint32x4_t vclezq_f32(float32x4_t __p0) {
49736		uint32x4_t __ret;
49737		__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 50);
49738		return __ret;
	50639	__ai int16x8_t vmovl_high_s8(int8x16_t __p0_524) {
	50640	int16x8_t __ret_524;
	50641	int8x8_t __a1_524 = vget_high_s8(__p0_524);
	50642	__ret_524 = (int16x8_t)(vshll_n_s8(__a1_524, 0));
	50643	return __ret_524;
49739	50644	}
49740	50645	#else
49741		__ai uint32x4_t vclezq_f32(float32x4_t __p0) {
49742		uint32x4_t __ret;
49743		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49744		__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 50);
49745		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49746		return __ret;
	50646	__ai int16x8_t vmovl_high_s8(int8x16_t __p0_525) {
	50647	int16x8_t __ret_525;
	50648	int8x16_t __rev0_525; __rev0_525 = __builtin_shufflevector(__p0_525, __p0_525, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	50649	int8x8_t __a1_525 = __noswap_vget_high_s8(__rev0_525);
	50650	__ret_525 = (int16x8_t)(__noswap_vshll_n_s8(__a1_525, 0));
	50651	__ret_525 = __builtin_shufflevector(__ret_525, __ret_525, 7, 6, 5, 4, 3, 2, 1, 0);
	50652	return __ret_525;
	50653	}
	50654	__ai int16x8_t __noswap_vmovl_high_s8(int8x16_t __p0_526) {
	50655	int16x8_t __ret_526;
	50656	int8x8_t __a1_526 = __noswap_vget_high_s8(__p0_526);
	50657	__ret_526 = (int16x8_t)(__noswap_vshll_n_s8(__a1_526, 0));
	50658	return __ret_526;
49747	50659	}
49748	50660	#endif
49749	50661
49750	50662	#ifdef __LITTLE_ENDIAN__
49751		__ai uint32x4_t vclezq_s32(int32x4_t __p0) {
49752		uint32x4_t __ret;
49753		__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 50);
49754		return __ret;
	50663	__ai int64x2_t vmovl_high_s32(int32x4_t __p0_527) {
	50664	int64x2_t __ret_527;
	50665	int32x2_t __a1_527 = vget_high_s32(__p0_527);
	50666	__ret_527 = (int64x2_t)(vshll_n_s32(__a1_527, 0));
	50667	return __ret_527;
49755	50668	}
49756	50669	#else
49757		__ai uint32x4_t vclezq_s32(int32x4_t __p0) {
49758		uint32x4_t __ret;
49759		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49760		__ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 50);
49761		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49762		return __ret;
	50670	__ai int64x2_t vmovl_high_s32(int32x4_t __p0_528) {
	50671	int64x2_t __ret_528;
	50672	int32x4_t __rev0_528; __rev0_528 = __builtin_shufflevector(__p0_528, __p0_528, 3, 2, 1, 0);
	50673	int32x2_t __a1_528 = __noswap_vget_high_s32(__rev0_528);
	50674	__ret_528 = (int64x2_t)(__noswap_vshll_n_s32(__a1_528, 0));
	50675	__ret_528 = __builtin_shufflevector(__ret_528, __ret_528, 1, 0);
	50676	return __ret_528;
	50677	}
	50678	__ai int64x2_t __noswap_vmovl_high_s32(int32x4_t __p0_529) {
	50679	int64x2_t __ret_529;
	50680	int32x2_t __a1_529 = __noswap_vget_high_s32(__p0_529);
	50681	__ret_529 = (int64x2_t)(__noswap_vshll_n_s32(__a1_529, 0));
	50682	return __ret_529;
49763	50683	}
49764	50684	#endif
49765	50685
49766	50686	#ifdef __LITTLE_ENDIAN__
49767		__ai uint64x2_t vclezq_s64(int64x2_t __p0) {
49768		uint64x2_t __ret;
49769		__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 51);
49770		return __ret;
	50687	__ai int32x4_t vmovl_high_s16(int16x8_t __p0_530) {
	50688	int32x4_t __ret_530;
	50689	int16x4_t __a1_530 = vget_high_s16(__p0_530);
	50690	__ret_530 = (int32x4_t)(vshll_n_s16(__a1_530, 0));
	50691	return __ret_530;
49771	50692	}
49772	50693	#else
49773		__ai uint64x2_t vclezq_s64(int64x2_t __p0) {
49774		uint64x2_t __ret;
49775		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49776		__ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 51);
49777		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49778		return __ret;
	50694	__ai int32x4_t vmovl_high_s16(int16x8_t __p0_531) {
	50695	int32x4_t __ret_531;
	50696	int16x8_t __rev0_531; __rev0_531 = __builtin_shufflevector(__p0_531, __p0_531, 7, 6, 5, 4, 3, 2, 1, 0);
	50697	int16x4_t __a1_531 = __noswap_vget_high_s16(__rev0_531);
	50698	__ret_531 = (int32x4_t)(__noswap_vshll_n_s16(__a1_531, 0));
	50699	__ret_531 = __builtin_shufflevector(__ret_531, __ret_531, 3, 2, 1, 0);
	50700	return __ret_531;
	50701	}
	50702	__ai int32x4_t __noswap_vmovl_high_s16(int16x8_t __p0_532) {
	50703	int32x4_t __ret_532;
	50704	int16x4_t __a1_532 = __noswap_vget_high_s16(__p0_532);
	50705	__ret_532 = (int32x4_t)(__noswap_vshll_n_s16(__a1_532, 0));
	50706	return __ret_532;
49779	50707	}
49780	50708	#endif
49781	50709
49782	50710	#ifdef __LITTLE_ENDIAN__
49783		__ai uint16x8_t vclezq_s16(int16x8_t __p0) {
	50711	__ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
49784	50712	uint16x8_t __ret;
49785		__ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__p0, 49);
	50713	__ret = vcombine_u16(__p0, vmovn_u32(__p1));
49786	50714	return __ret;
49787	50715	}
49788	50716	#else
49789		__ai uint16x8_t vclezq_s16(int16x8_t __p0) {
	50717	__ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
49790	50718	uint16x8_t __ret;
49791		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49792		__ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 49);
	50719	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	50720	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	50721	__ret = __noswap_vcombine_u16(__rev0, __noswap_vmovn_u32(__rev1));
49793	50722	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49794	50723	return __ret;
49795	50724	}
49796	50725	#endif
49797	50726
49798	50727	#ifdef __LITTLE_ENDIAN__
49799		__ai uint8x8_t vclez_s8(int8x8_t __p0) {
49800		uint8x8_t __ret;
49801		__ret = (uint8x8_t) __builtin_neon_vclez_v((int8x8_t)__p0, 16);
	50728	__ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
	50729	uint32x4_t __ret;
	50730	__ret = vcombine_u32(__p0, vmovn_u64(__p1));
49802	50731	return __ret;
49803	50732	}
49804	50733	#else
49805		__ai uint8x8_t vclez_s8(int8x8_t __p0) {
49806		uint8x8_t __ret;
49807		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49808		__ret = (uint8x8_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 16);
49809		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	50734	__ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
	50735	uint32x4_t __ret;
	50736	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	50737	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	50738	__ret = __noswap_vcombine_u32(__rev0, __noswap_vmovn_u64(__rev1));
	50739	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49810	50740	return __ret;
49811	50741	}
49812	50742	#endif
49813	50743
49814		__ai uint64x1_t vclez_f64(float64x1_t __p0) {
49815		uint64x1_t __ret;
49816		__ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
49817		return __ret;
49818		}
49819	50744	#ifdef __LITTLE_ENDIAN__
49820		__ai uint32x2_t vclez_f32(float32x2_t __p0) {
49821		uint32x2_t __ret;
49822		__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__p0, 18);
	50745	__ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
	50746	uint8x16_t __ret;
	50747	__ret = vcombine_u8(__p0, vmovn_u16(__p1));
49823	50748	return __ret;
49824	50749	}
49825	50750	#else
49826		__ai uint32x2_t vclez_f32(float32x2_t __p0) {
49827		uint32x2_t __ret;
49828		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49829		__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 18);
49830		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	50751	__ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
	50752	uint8x16_t __ret;
	50753	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	50754	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	50755	__ret = __noswap_vcombine_u8(__rev0, __noswap_vmovn_u16(__rev1));
	50756	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49831	50757	return __ret;
49832	50758	}
49833	50759	#endif
49834	50760
49835	50761	#ifdef __LITTLE_ENDIAN__
49836		__ai uint32x2_t vclez_s32(int32x2_t __p0) {
49837		uint32x2_t __ret;
49838		__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__p0, 18);
	50762	__ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
	50763	int16x8_t __ret;
	50764	__ret = vcombine_s16(__p0, vmovn_s32(__p1));
49839	50765	return __ret;
49840	50766	}
49841	50767	#else
49842		__ai uint32x2_t vclez_s32(int32x2_t __p0) {
49843		uint32x2_t __ret;
49844		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49845		__ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 18);
49846		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	50768	__ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
	50769	int16x8_t __ret;
	50770	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	50771	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	50772	__ret = __noswap_vcombine_s16(__rev0, __noswap_vmovn_s32(__rev1));
	50773	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49847	50774	return __ret;
49848	50775	}
49849	50776	#endif
49850	50777
49851		__ai uint64x1_t vclez_s64(int64x1_t __p0) {
49852		uint64x1_t __ret;
49853		__ret = (uint64x1_t) __builtin_neon_vclez_v((int8x8_t)__p0, 19);
49854		return __ret;
49855		}
49856	50778	#ifdef __LITTLE_ENDIAN__
49857		__ai uint16x4_t vclez_s16(int16x4_t __p0) {
49858		uint16x4_t __ret;
49859		__ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__p0, 17);
	50779	__ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
	50780	int32x4_t __ret;
	50781	__ret = vcombine_s32(__p0, vmovn_s64(__p1));
49860	50782	return __ret;
49861	50783	}
49862	50784	#else
49863		__ai uint16x4_t vclez_s16(int16x4_t __p0) {
49864		uint16x4_t __ret;
49865		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49866		__ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 17);
	50785	__ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
	50786	int32x4_t __ret;
	50787	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	50788	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	50789	__ret = __noswap_vcombine_s32(__rev0, __noswap_vmovn_s64(__rev1));
49867	50790	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49868	50791	return __ret;
49869	50792	}
49870	50793	#endif
49871	50794
49872		__ai uint64_t vclezd_s64(int64_t __p0) {
49873		uint64_t __ret;
49874		__ret = (uint64_t) __builtin_neon_vclezd_s64(__p0);
49875		return __ret;
49876		}
49877		__ai uint64_t vclezd_f64(float64_t __p0) {
49878		uint64_t __ret;
49879		__ret = (uint64_t) __builtin_neon_vclezd_f64(__p0);
49880		return __ret;
49881		}
49882		__ai uint32_t vclezs_f32(float32_t __p0) {
49883		uint32_t __ret;
49884		__ret = (uint32_t) __builtin_neon_vclezs_f32(__p0);
49885		return __ret;
49886		}
49887	50795	#ifdef __LITTLE_ENDIAN__
49888		__ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49889		uint64x2_t __ret;
49890		__ret = (uint64x2_t)(__p0 < __p1);
	50796	__ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
	50797	int8x16_t __ret;
	50798	__ret = vcombine_s8(__p0, vmovn_s16(__p1));
49891	50799	return __ret;
49892	50800	}
49893	50801	#else
49894		__ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49895		uint64x2_t __ret;
49896		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49897		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49898		__ret = (uint64x2_t)(__rev0 < __rev1);
49899		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	50802	__ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
	50803	int8x16_t __ret;
	50804	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	50805	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	50806	__ret = __noswap_vcombine_s8(__rev0, __noswap_vmovn_s16(__rev1));
	50807	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49900	50808	return __ret;
49901	50809	}
49902	50810	#endif
49903	50811
49904	50812	#ifdef __LITTLE_ENDIAN__
49905		__ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
49906		uint64x2_t __ret;
49907		__ret = (uint64x2_t)(__p0 < __p1);
	50813	__ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
	50814	float64x2_t __ret;
	50815	__ret = __p0 * __p1;
49908	50816	return __ret;
49909	50817	}
49910	50818	#else
49911		__ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
49912		uint64x2_t __ret;
	50819	__ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
	50820	float64x2_t __ret;
49913	50821	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49914	50822	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49915		__ret = (uint64x2_t)(__rev0 < __rev1);
	50823	__ret = __rev0 * __rev1;
49916	50824	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49917	50825	return __ret;
49918	50826	}
49919	50827	#endif
49920	50828
49921		#ifdef __LITTLE_ENDIAN__
49922		__ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
49923		uint64x2_t __ret;
49924		__ret = (uint64x2_t)(__p0 < __p1);
	50829	__ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
	50830	float64x1_t __ret;
	50831	__ret = __p0 * __p1;
49925	50832	return __ret;
49926	50833	}
	50834	#define vmuld_lane_f64(__p0_533, __p1_533, __p2_533) __extension__ ({ \
	50835	float64_t __ret_533; \
	50836	float64_t __s0_533 = __p0_533; \
	50837	float64x1_t __s1_533 = __p1_533; \
	50838	__ret_533 = __s0_533 * vget_lane_f64(__s1_533, __p2_533); \
	50839	__ret_533; \
	50840	})
	50841	#ifdef __LITTLE_ENDIAN__
	50842	#define vmuls_lane_f32(__p0_534, __p1_534, __p2_534) __extension__ ({ \
	50843	float32_t __ret_534; \
	50844	float32_t __s0_534 = __p0_534; \
	50845	float32x2_t __s1_534 = __p1_534; \
	50846	__ret_534 = __s0_534 * vget_lane_f32(__s1_534, __p2_534); \
	50847	__ret_534; \
	50848	})
49927	50849	#else
49928		__ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
49929		uint64x2_t __ret;
49930		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49931		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49932		__ret = (uint64x2_t)(__rev0 < __rev1);
49933		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49934		return __ret;
49935		}
	50850	#define vmuls_lane_f32(__p0_535, __p1_535, __p2_535) __extension__ ({ \
	50851	float32_t __ret_535; \
	50852	float32_t __s0_535 = __p0_535; \
	50853	float32x2_t __s1_535 = __p1_535; \
	50854	float32x2_t __rev1_535; __rev1_535 = __builtin_shufflevector(__s1_535, __s1_535, 1, 0); \
	50855	__ret_535 = __s0_535 * __noswap_vget_lane_f32(__rev1_535, __p2_535); \
	50856	__ret_535; \
	50857	})
49936	50858	#endif
49937	50859
49938		__ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
49939		uint64x1_t __ret;
49940		__ret = (uint64x1_t)(__p0 < __p1);
49941		return __ret;
49942		}
49943		__ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
49944		uint64x1_t __ret;
49945		__ret = (uint64x1_t)(__p0 < __p1);
49946		return __ret;
49947		}
49948		__ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
49949		uint64x1_t __ret;
49950		__ret = (uint64x1_t)(__p0 < __p1);
	50860	#define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	50861	float64x1_t __ret; \
	50862	float64x1_t __s0 = __p0; \
	50863	float64x1_t __s1 = __p1; \
	50864	__ret = (float64x1_t) __builtin_neon_vmul_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
	50865	__ret; \
	50866	})
	50867	#ifdef __LITTLE_ENDIAN__
	50868	#define vmulq_lane_f64(__p0_536, __p1_536, __p2_536) __extension__ ({ \
	50869	float64x2_t __ret_536; \
	50870	float64x2_t __s0_536 = __p0_536; \
	50871	float64x1_t __s1_536 = __p1_536; \
	50872	__ret_536 = __s0_536 * splatq_lane_f64(__s1_536, __p2_536); \
	50873	__ret_536; \
	50874	})
	50875	#else
	50876	#define vmulq_lane_f64(__p0_537, __p1_537, __p2_537) __extension__ ({ \
	50877	float64x2_t __ret_537; \
	50878	float64x2_t __s0_537 = __p0_537; \
	50879	float64x1_t __s1_537 = __p1_537; \
	50880	float64x2_t __rev0_537; __rev0_537 = __builtin_shufflevector(__s0_537, __s0_537, 1, 0); \
	50881	__ret_537 = __rev0_537 * __noswap_splatq_lane_f64(__s1_537, __p2_537); \
	50882	__ret_537 = __builtin_shufflevector(__ret_537, __ret_537, 1, 0); \
	50883	__ret_537; \
	50884	})
	50885	#endif
	50886
	50887	#ifdef __LITTLE_ENDIAN__
	50888	#define vmuld_laneq_f64(__p0_538, __p1_538, __p2_538) __extension__ ({ \
	50889	float64_t __ret_538; \
	50890	float64_t __s0_538 = __p0_538; \
	50891	float64x2_t __s1_538 = __p1_538; \
	50892	__ret_538 = __s0_538 * vgetq_lane_f64(__s1_538, __p2_538); \
	50893	__ret_538; \
	50894	})
	50895	#else
	50896	#define vmuld_laneq_f64(__p0_539, __p1_539, __p2_539) __extension__ ({ \
	50897	float64_t __ret_539; \
	50898	float64_t __s0_539 = __p0_539; \
	50899	float64x2_t __s1_539 = __p1_539; \
	50900	float64x2_t __rev1_539; __rev1_539 = __builtin_shufflevector(__s1_539, __s1_539, 1, 0); \
	50901	__ret_539 = __s0_539 * __noswap_vgetq_lane_f64(__rev1_539, __p2_539); \
	50902	__ret_539; \
	50903	})
	50904	#endif
	50905
	50906	#ifdef __LITTLE_ENDIAN__
	50907	#define vmuls_laneq_f32(__p0_540, __p1_540, __p2_540) __extension__ ({ \
	50908	float32_t __ret_540; \
	50909	float32_t __s0_540 = __p0_540; \
	50910	float32x4_t __s1_540 = __p1_540; \
	50911	__ret_540 = __s0_540 * vgetq_lane_f32(__s1_540, __p2_540); \
	50912	__ret_540; \
	50913	})
	50914	#else
	50915	#define vmuls_laneq_f32(__p0_541, __p1_541, __p2_541) __extension__ ({ \
	50916	float32_t __ret_541; \
	50917	float32_t __s0_541 = __p0_541; \
	50918	float32x4_t __s1_541 = __p1_541; \
	50919	float32x4_t __rev1_541; __rev1_541 = __builtin_shufflevector(__s1_541, __s1_541, 3, 2, 1, 0); \
	50920	__ret_541 = __s0_541 * __noswap_vgetq_lane_f32(__rev1_541, __p2_541); \
	50921	__ret_541; \
	50922	})
	50923	#endif
	50924
	50925	#ifdef __LITTLE_ENDIAN__
	50926	#define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
	50927	float64x1_t __ret; \
	50928	float64x1_t __s0 = __p0; \
	50929	float64x2_t __s1 = __p1; \
	50930	__ret = (float64x1_t) __builtin_neon_vmul_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 10); \
	50931	__ret; \
	50932	})
	50933	#else
	50934	#define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
	50935	float64x1_t __ret; \
	50936	float64x1_t __s0 = __p0; \
	50937	float64x2_t __s1 = __p1; \
	50938	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	50939	__ret = (float64x1_t) __builtin_neon_vmul_laneq_v((int8x8_t)__s0, (int8x16_t)__rev1, __p2, 10); \
	50940	__ret; \
	50941	})
	50942	#endif
	50943
	50944	#ifdef __LITTLE_ENDIAN__
	50945	#define vmulq_laneq_u32(__p0_542, __p1_542, __p2_542) __extension__ ({ \
	50946	uint32x4_t __ret_542; \
	50947	uint32x4_t __s0_542 = __p0_542; \
	50948	uint32x4_t __s1_542 = __p1_542; \
	50949	__ret_542 = __s0_542 * splatq_laneq_u32(__s1_542, __p2_542); \
	50950	__ret_542; \
	50951	})
	50952	#else
	50953	#define vmulq_laneq_u32(__p0_543, __p1_543, __p2_543) __extension__ ({ \
	50954	uint32x4_t __ret_543; \
	50955	uint32x4_t __s0_543 = __p0_543; \
	50956	uint32x4_t __s1_543 = __p1_543; \
	50957	uint32x4_t __rev0_543; __rev0_543 = __builtin_shufflevector(__s0_543, __s0_543, 3, 2, 1, 0); \
	50958	uint32x4_t __rev1_543; __rev1_543 = __builtin_shufflevector(__s1_543, __s1_543, 3, 2, 1, 0); \
	50959	__ret_543 = __rev0_543 * __noswap_splatq_laneq_u32(__rev1_543, __p2_543); \
	50960	__ret_543 = __builtin_shufflevector(__ret_543, __ret_543, 3, 2, 1, 0); \
	50961	__ret_543; \
	50962	})
	50963	#endif
	50964
	50965	#ifdef __LITTLE_ENDIAN__
	50966	#define vmulq_laneq_u16(__p0_544, __p1_544, __p2_544) __extension__ ({ \
	50967	uint16x8_t __ret_544; \
	50968	uint16x8_t __s0_544 = __p0_544; \
	50969	uint16x8_t __s1_544 = __p1_544; \
	50970	__ret_544 = __s0_544 * splatq_laneq_u16(__s1_544, __p2_544); \
	50971	__ret_544; \
	50972	})
	50973	#else
	50974	#define vmulq_laneq_u16(__p0_545, __p1_545, __p2_545) __extension__ ({ \
	50975	uint16x8_t __ret_545; \
	50976	uint16x8_t __s0_545 = __p0_545; \
	50977	uint16x8_t __s1_545 = __p1_545; \
	50978	uint16x8_t __rev0_545; __rev0_545 = __builtin_shufflevector(__s0_545, __s0_545, 7, 6, 5, 4, 3, 2, 1, 0); \
	50979	uint16x8_t __rev1_545; __rev1_545 = __builtin_shufflevector(__s1_545, __s1_545, 7, 6, 5, 4, 3, 2, 1, 0); \
	50980	__ret_545 = __rev0_545 * __noswap_splatq_laneq_u16(__rev1_545, __p2_545); \
	50981	__ret_545 = __builtin_shufflevector(__ret_545, __ret_545, 7, 6, 5, 4, 3, 2, 1, 0); \
	50982	__ret_545; \
	50983	})
	50984	#endif
	50985
	50986	#ifdef __LITTLE_ENDIAN__
	50987	#define vmulq_laneq_f64(__p0_546, __p1_546, __p2_546) __extension__ ({ \
	50988	float64x2_t __ret_546; \
	50989	float64x2_t __s0_546 = __p0_546; \
	50990	float64x2_t __s1_546 = __p1_546; \
	50991	__ret_546 = __s0_546 * splatq_laneq_f64(__s1_546, __p2_546); \
	50992	__ret_546; \
	50993	})
	50994	#else
	50995	#define vmulq_laneq_f64(__p0_547, __p1_547, __p2_547) __extension__ ({ \
	50996	float64x2_t __ret_547; \
	50997	float64x2_t __s0_547 = __p0_547; \
	50998	float64x2_t __s1_547 = __p1_547; \
	50999	float64x2_t __rev0_547; __rev0_547 = __builtin_shufflevector(__s0_547, __s0_547, 1, 0); \
	51000	float64x2_t __rev1_547; __rev1_547 = __builtin_shufflevector(__s1_547, __s1_547, 1, 0); \
	51001	__ret_547 = __rev0_547 * __noswap_splatq_laneq_f64(__rev1_547, __p2_547); \
	51002	__ret_547 = __builtin_shufflevector(__ret_547, __ret_547, 1, 0); \
	51003	__ret_547; \
	51004	})
	51005	#endif
	51006
	51007	#ifdef __LITTLE_ENDIAN__
	51008	#define vmulq_laneq_f32(__p0_548, __p1_548, __p2_548) __extension__ ({ \
	51009	float32x4_t __ret_548; \
	51010	float32x4_t __s0_548 = __p0_548; \
	51011	float32x4_t __s1_548 = __p1_548; \
	51012	__ret_548 = __s0_548 * splatq_laneq_f32(__s1_548, __p2_548); \
	51013	__ret_548; \
	51014	})
	51015	#else
	51016	#define vmulq_laneq_f32(__p0_549, __p1_549, __p2_549) __extension__ ({ \
	51017	float32x4_t __ret_549; \
	51018	float32x4_t __s0_549 = __p0_549; \
	51019	float32x4_t __s1_549 = __p1_549; \
	51020	float32x4_t __rev0_549; __rev0_549 = __builtin_shufflevector(__s0_549, __s0_549, 3, 2, 1, 0); \
	51021	float32x4_t __rev1_549; __rev1_549 = __builtin_shufflevector(__s1_549, __s1_549, 3, 2, 1, 0); \
	51022	__ret_549 = __rev0_549 * __noswap_splatq_laneq_f32(__rev1_549, __p2_549); \
	51023	__ret_549 = __builtin_shufflevector(__ret_549, __ret_549, 3, 2, 1, 0); \
	51024	__ret_549; \
	51025	})
	51026	#endif
	51027
	51028	#ifdef __LITTLE_ENDIAN__
	51029	#define vmulq_laneq_s32(__p0_550, __p1_550, __p2_550) __extension__ ({ \
	51030	int32x4_t __ret_550; \
	51031	int32x4_t __s0_550 = __p0_550; \
	51032	int32x4_t __s1_550 = __p1_550; \
	51033	__ret_550 = __s0_550 * splatq_laneq_s32(__s1_550, __p2_550); \
	51034	__ret_550; \
	51035	})
	51036	#else
	51037	#define vmulq_laneq_s32(__p0_551, __p1_551, __p2_551) __extension__ ({ \
	51038	int32x4_t __ret_551; \
	51039	int32x4_t __s0_551 = __p0_551; \
	51040	int32x4_t __s1_551 = __p1_551; \
	51041	int32x4_t __rev0_551; __rev0_551 = __builtin_shufflevector(__s0_551, __s0_551, 3, 2, 1, 0); \
	51042	int32x4_t __rev1_551; __rev1_551 = __builtin_shufflevector(__s1_551, __s1_551, 3, 2, 1, 0); \
	51043	__ret_551 = __rev0_551 * __noswap_splatq_laneq_s32(__rev1_551, __p2_551); \
	51044	__ret_551 = __builtin_shufflevector(__ret_551, __ret_551, 3, 2, 1, 0); \
	51045	__ret_551; \
	51046	})
	51047	#endif
	51048
	51049	#ifdef __LITTLE_ENDIAN__
	51050	#define vmulq_laneq_s16(__p0_552, __p1_552, __p2_552) __extension__ ({ \
	51051	int16x8_t __ret_552; \
	51052	int16x8_t __s0_552 = __p0_552; \
	51053	int16x8_t __s1_552 = __p1_552; \
	51054	__ret_552 = __s0_552 * splatq_laneq_s16(__s1_552, __p2_552); \
	51055	__ret_552; \
	51056	})
	51057	#else
	51058	#define vmulq_laneq_s16(__p0_553, __p1_553, __p2_553) __extension__ ({ \
	51059	int16x8_t __ret_553; \
	51060	int16x8_t __s0_553 = __p0_553; \
	51061	int16x8_t __s1_553 = __p1_553; \
	51062	int16x8_t __rev0_553; __rev0_553 = __builtin_shufflevector(__s0_553, __s0_553, 7, 6, 5, 4, 3, 2, 1, 0); \
	51063	int16x8_t __rev1_553; __rev1_553 = __builtin_shufflevector(__s1_553, __s1_553, 7, 6, 5, 4, 3, 2, 1, 0); \
	51064	__ret_553 = __rev0_553 * __noswap_splatq_laneq_s16(__rev1_553, __p2_553); \
	51065	__ret_553 = __builtin_shufflevector(__ret_553, __ret_553, 7, 6, 5, 4, 3, 2, 1, 0); \
	51066	__ret_553; \
	51067	})
	51068	#endif
	51069
	51070	#ifdef __LITTLE_ENDIAN__
	51071	#define vmul_laneq_u32(__p0_554, __p1_554, __p2_554) __extension__ ({ \
	51072	uint32x2_t __ret_554; \
	51073	uint32x2_t __s0_554 = __p0_554; \
	51074	uint32x4_t __s1_554 = __p1_554; \
	51075	__ret_554 = __s0_554 * splat_laneq_u32(__s1_554, __p2_554); \
	51076	__ret_554; \
	51077	})
	51078	#else
	51079	#define vmul_laneq_u32(__p0_555, __p1_555, __p2_555) __extension__ ({ \
	51080	uint32x2_t __ret_555; \
	51081	uint32x2_t __s0_555 = __p0_555; \
	51082	uint32x4_t __s1_555 = __p1_555; \
	51083	uint32x2_t __rev0_555; __rev0_555 = __builtin_shufflevector(__s0_555, __s0_555, 1, 0); \
	51084	uint32x4_t __rev1_555; __rev1_555 = __builtin_shufflevector(__s1_555, __s1_555, 3, 2, 1, 0); \
	51085	__ret_555 = __rev0_555 * __noswap_splat_laneq_u32(__rev1_555, __p2_555); \
	51086	__ret_555 = __builtin_shufflevector(__ret_555, __ret_555, 1, 0); \
	51087	__ret_555; \
	51088	})
	51089	#endif
	51090
	51091	#ifdef __LITTLE_ENDIAN__
	51092	#define vmul_laneq_u16(__p0_556, __p1_556, __p2_556) __extension__ ({ \
	51093	uint16x4_t __ret_556; \
	51094	uint16x4_t __s0_556 = __p0_556; \
	51095	uint16x8_t __s1_556 = __p1_556; \
	51096	__ret_556 = __s0_556 * splat_laneq_u16(__s1_556, __p2_556); \
	51097	__ret_556; \
	51098	})
	51099	#else
	51100	#define vmul_laneq_u16(__p0_557, __p1_557, __p2_557) __extension__ ({ \
	51101	uint16x4_t __ret_557; \
	51102	uint16x4_t __s0_557 = __p0_557; \
	51103	uint16x8_t __s1_557 = __p1_557; \
	51104	uint16x4_t __rev0_557; __rev0_557 = __builtin_shufflevector(__s0_557, __s0_557, 3, 2, 1, 0); \
	51105	uint16x8_t __rev1_557; __rev1_557 = __builtin_shufflevector(__s1_557, __s1_557, 7, 6, 5, 4, 3, 2, 1, 0); \
	51106	__ret_557 = __rev0_557 * __noswap_splat_laneq_u16(__rev1_557, __p2_557); \
	51107	__ret_557 = __builtin_shufflevector(__ret_557, __ret_557, 3, 2, 1, 0); \
	51108	__ret_557; \
	51109	})
	51110	#endif
	51111
	51112	#ifdef __LITTLE_ENDIAN__
	51113	#define vmul_laneq_f32(__p0_558, __p1_558, __p2_558) __extension__ ({ \
	51114	float32x2_t __ret_558; \
	51115	float32x2_t __s0_558 = __p0_558; \
	51116	float32x4_t __s1_558 = __p1_558; \
	51117	__ret_558 = __s0_558 * splat_laneq_f32(__s1_558, __p2_558); \
	51118	__ret_558; \
	51119	})
	51120	#else
	51121	#define vmul_laneq_f32(__p0_559, __p1_559, __p2_559) __extension__ ({ \
	51122	float32x2_t __ret_559; \
	51123	float32x2_t __s0_559 = __p0_559; \
	51124	float32x4_t __s1_559 = __p1_559; \
	51125	float32x2_t __rev0_559; __rev0_559 = __builtin_shufflevector(__s0_559, __s0_559, 1, 0); \
	51126	float32x4_t __rev1_559; __rev1_559 = __builtin_shufflevector(__s1_559, __s1_559, 3, 2, 1, 0); \
	51127	__ret_559 = __rev0_559 * __noswap_splat_laneq_f32(__rev1_559, __p2_559); \
	51128	__ret_559 = __builtin_shufflevector(__ret_559, __ret_559, 1, 0); \
	51129	__ret_559; \
	51130	})
	51131	#endif
	51132
	51133	#ifdef __LITTLE_ENDIAN__
	51134	#define vmul_laneq_s32(__p0_560, __p1_560, __p2_560) __extension__ ({ \
	51135	int32x2_t __ret_560; \
	51136	int32x2_t __s0_560 = __p0_560; \
	51137	int32x4_t __s1_560 = __p1_560; \
	51138	__ret_560 = __s0_560 * splat_laneq_s32(__s1_560, __p2_560); \
	51139	__ret_560; \
	51140	})
	51141	#else
	51142	#define vmul_laneq_s32(__p0_561, __p1_561, __p2_561) __extension__ ({ \
	51143	int32x2_t __ret_561; \
	51144	int32x2_t __s0_561 = __p0_561; \
	51145	int32x4_t __s1_561 = __p1_561; \
	51146	int32x2_t __rev0_561; __rev0_561 = __builtin_shufflevector(__s0_561, __s0_561, 1, 0); \
	51147	int32x4_t __rev1_561; __rev1_561 = __builtin_shufflevector(__s1_561, __s1_561, 3, 2, 1, 0); \
	51148	__ret_561 = __rev0_561 * __noswap_splat_laneq_s32(__rev1_561, __p2_561); \
	51149	__ret_561 = __builtin_shufflevector(__ret_561, __ret_561, 1, 0); \
	51150	__ret_561; \
	51151	})
	51152	#endif
	51153
	51154	#ifdef __LITTLE_ENDIAN__
	51155	#define vmul_laneq_s16(__p0_562, __p1_562, __p2_562) __extension__ ({ \
	51156	int16x4_t __ret_562; \
	51157	int16x4_t __s0_562 = __p0_562; \
	51158	int16x8_t __s1_562 = __p1_562; \
	51159	__ret_562 = __s0_562 * splat_laneq_s16(__s1_562, __p2_562); \
	51160	__ret_562; \
	51161	})
	51162	#else
	51163	#define vmul_laneq_s16(__p0_563, __p1_563, __p2_563) __extension__ ({ \
	51164	int16x4_t __ret_563; \
	51165	int16x4_t __s0_563 = __p0_563; \
	51166	int16x8_t __s1_563 = __p1_563; \
	51167	int16x4_t __rev0_563; __rev0_563 = __builtin_shufflevector(__s0_563, __s0_563, 3, 2, 1, 0); \
	51168	int16x8_t __rev1_563; __rev1_563 = __builtin_shufflevector(__s1_563, __s1_563, 7, 6, 5, 4, 3, 2, 1, 0); \
	51169	__ret_563 = __rev0_563 * __noswap_splat_laneq_s16(__rev1_563, __p2_563); \
	51170	__ret_563 = __builtin_shufflevector(__ret_563, __ret_563, 3, 2, 1, 0); \
	51171	__ret_563; \
	51172	})
	51173	#endif
	51174
	51175	__ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
	51176	float64x1_t __ret;
	51177	__ret = (float64x1_t) __builtin_neon_vmul_n_f64((float64x1_t)__p0, __p1);
49951	51178	return __ret;
49952	51179	}
49953		__ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
49954		uint64_t __ret;
49955		__ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
	51180	#ifdef __LITTLE_ENDIAN__
	51181	__ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
	51182	float64x2_t __ret;
	51183	__ret = __p0 * (float64x2_t) {__p1, __p1};
49956	51184	return __ret;
49957	51185	}
49958		__ai uint64_t vcltd_s64(int64_t __p0, int64_t __p1) {
49959		uint64_t __ret;
49960		__ret = (uint64_t) __builtin_neon_vcltd_s64(__p0, __p1);
	51186	#else
	51187	__ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
	51188	float64x2_t __ret;
	51189	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	51190	__ret = __rev0 * (float64x2_t) {__p1, __p1};
	51191	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49961	51192	return __ret;
49962	51193	}
49963		__ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
49964		uint64_t __ret;
49965		__ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
	51194	#endif
	51195
	51196	#ifdef __LITTLE_ENDIAN__
	51197	__ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
	51198	poly16x8_t __ret;
	51199	__ret = vmull_p8(vget_high_p8(__p0), vget_high_p8(__p1));
49966	51200	return __ret;
49967	51201	}
49968		__ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
49969		uint32_t __ret;
49970		__ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
	51202	#else
	51203	__ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
	51204	poly16x8_t __ret;
	51205	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51206	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51207	__ret = __noswap_vmull_p8(__noswap_vget_high_p8(__rev0), __noswap_vget_high_p8(__rev1));
	51208	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49971	51209	return __ret;
49972	51210	}
	51211	#endif
	51212
49973	51213	#ifdef __LITTLE_ENDIAN__
49974		__ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
49975		uint8x16_t __ret;
49976		__ret = (uint8x16_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 48);
	51214	__ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
	51215	uint16x8_t __ret;
	51216	__ret = vmull_u8(vget_high_u8(__p0), vget_high_u8(__p1));
49977	51217	return __ret;
49978	51218	}
49979	51219	#else
49980		__ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
49981		uint8x16_t __ret;
49982		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49983		__ret = (uint8x16_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 48);
49984		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51220	__ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
	51221	uint16x8_t __ret;
	51222	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51223	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51224	__ret = __noswap_vmull_u8(__noswap_vget_high_u8(__rev0), __noswap_vget_high_u8(__rev1));
	51225	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49985	51226	return __ret;
49986	51227	}
49987	51228	#endif
49988	51229
49989	51230	#ifdef __LITTLE_ENDIAN__
49990		__ai uint64x2_t vcltzq_f64(float64x2_t __p0) {
	51231	__ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
49991	51232	uint64x2_t __ret;
49992		__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 51);
	51233	__ret = vmull_u32(vget_high_u32(__p0), vget_high_u32(__p1));
49993	51234	return __ret;
49994	51235	}
49995	51236	#else
49996		__ai uint64x2_t vcltzq_f64(float64x2_t __p0) {
	51237	__ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
49997	51238	uint64x2_t __ret;
49998		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49999		__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 51);
	51239	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	51240	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	51241	__ret = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0), __noswap_vget_high_u32(__rev1));
50000	51242	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50001	51243	return __ret;
50002	51244	}
50003	51245	#endif
50004	51246
50005	51247	#ifdef __LITTLE_ENDIAN__
50006		__ai uint32x4_t vcltzq_f32(float32x4_t __p0) {
	51248	__ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
50007	51249	uint32x4_t __ret;
50008		__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 50);
	51250	__ret = vmull_u16(vget_high_u16(__p0), vget_high_u16(__p1));
50009	51251	return __ret;
50010	51252	}
50011	51253	#else
50012		__ai uint32x4_t vcltzq_f32(float32x4_t __p0) {
	51254	__ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
50013	51255	uint32x4_t __ret;
50014		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
50015		__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 50);
	51256	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	51257	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	51258	__ret = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0), __noswap_vget_high_u16(__rev1));
50016	51259	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
50017	51260	return __ret;
50018	51261	}
50019	51262	#endif
50020	51263
50021	51264	#ifdef __LITTLE_ENDIAN__
50022		__ai uint32x4_t vcltzq_s32(int32x4_t __p0) {
50023		uint32x4_t __ret;
50024		__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 50);
	51265	__ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
	51266	int16x8_t __ret;
	51267	__ret = vmull_s8(vget_high_s8(__p0), vget_high_s8(__p1));
50025	51268	return __ret;
50026	51269	}
50027	51270	#else
50028		__ai uint32x4_t vcltzq_s32(int32x4_t __p0) {
50029		uint32x4_t __ret;
50030		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
50031		__ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 50);
50032		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	51271	__ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
	51272	int16x8_t __ret;
	51273	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51274	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51275	__ret = __noswap_vmull_s8(__noswap_vget_high_s8(__rev0), __noswap_vget_high_s8(__rev1));
	51276	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
50033	51277	return __ret;
50034	51278	}
50035	51279	#endif
50036	51280
50037	51281	#ifdef __LITTLE_ENDIAN__
50038		__ai uint64x2_t vcltzq_s64(int64x2_t __p0) {
50039		uint64x2_t __ret;
50040		__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 51);
	51282	__ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
	51283	int64x2_t __ret;
	51284	__ret = vmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
50041	51285	return __ret;
50042	51286	}
50043	51287	#else
50044		__ai uint64x2_t vcltzq_s64(int64x2_t __p0) {
50045		uint64x2_t __ret;
50046		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
50047		__ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 51);
	51288	__ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
	51289	int64x2_t __ret;
	51290	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	51291	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	51292	__ret = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
50048	51293	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50049	51294	return __ret;
50050	51295	}
50051	51296	#endif
50052	51297
50053	51298	#ifdef __LITTLE_ENDIAN__
50054		__ai uint16x8_t vcltzq_s16(int16x8_t __p0) {
50055		uint16x8_t __ret;
50056		__ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__p0, 49);
	51299	__ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
	51300	int32x4_t __ret;
	51301	__ret = vmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
50057	51302	return __ret;
50058	51303	}
50059	51304	#else
50060		__ai uint16x8_t vcltzq_s16(int16x8_t __p0) {
50061		uint16x8_t __ret;
	51305	__ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
	51306	int32x4_t __ret;
50062	51307	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
50063		__ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 49);
50064		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	51308	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	51309	__ret = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
	51310	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
50065	51311	return __ret;
50066	51312	}
50067	51313	#endif
50068	51314
50069	51315	#ifdef __LITTLE_ENDIAN__
50070		__ai uint8x8_t vcltz_s8(int8x8_t __p0) {
50071		uint8x8_t __ret;
50072		__ret = (uint8x8_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 16);
50073		return __ret;
50074		}
	51316	#define vmull_high_lane_u32(__p0_564, __p1_564, __p2_564) __extension__ ({ \
	51317	uint64x2_t __ret_564; \
	51318	uint32x4_t __s0_564 = __p0_564; \
	51319	uint32x2_t __s1_564 = __p1_564; \
	51320	__ret_564 = vmull_u32(vget_high_u32(__s0_564), splat_lane_u32(__s1_564, __p2_564)); \
	51321	__ret_564; \
	51322	})
50075	51323	#else
50076		__ai uint8x8_t vcltz_s8(int8x8_t __p0) {
50077		uint8x8_t __ret;
50078		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
50079		__ret = (uint8x8_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 16);
50080		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
50081		return __ret;
50082		}
	51324	#define vmull_high_lane_u32(__p0_565, __p1_565, __p2_565) __extension__ ({ \
	51325	uint64x2_t __ret_565; \
	51326	uint32x4_t __s0_565 = __p0_565; \
	51327	uint32x2_t __s1_565 = __p1_565; \
	51328	uint32x4_t __rev0_565; __rev0_565 = __builtin_shufflevector(__s0_565, __s0_565, 3, 2, 1, 0); \
	51329	uint32x2_t __rev1_565; __rev1_565 = __builtin_shufflevector(__s1_565, __s1_565, 1, 0); \
	51330	__ret_565 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_565), __noswap_splat_lane_u32(__rev1_565, __p2_565)); \
	51331	__ret_565 = __builtin_shufflevector(__ret_565, __ret_565, 1, 0); \
	51332	__ret_565; \
	51333	})
50083	51334	#endif
50084	51335
50085		__ai uint64x1_t vcltz_f64(float64x1_t __p0) {
50086		uint64x1_t __ret;
50087		__ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
50088		return __ret;
50089		}
50090	51336	#ifdef __LITTLE_ENDIAN__
50091		__ai uint32x2_t vcltz_f32(float32x2_t __p0) {
50092		uint32x2_t __ret;
50093		__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 18);
50094		return __ret;
50095		}
	51337	#define vmull_high_lane_u16(__p0_566, __p1_566, __p2_566) __extension__ ({ \
	51338	uint32x4_t __ret_566; \
	51339	uint16x8_t __s0_566 = __p0_566; \
	51340	uint16x4_t __s1_566 = __p1_566; \
	51341	__ret_566 = vmull_u16(vget_high_u16(__s0_566), splat_lane_u16(__s1_566, __p2_566)); \
	51342	__ret_566; \
	51343	})
50096	51344	#else
50097		__ai uint32x2_t vcltz_f32(float32x2_t __p0) {
50098		uint32x2_t __ret;
50099		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
50100		__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 18);
50101		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50102		return __ret;
50103		}
	51345	#define vmull_high_lane_u16(__p0_567, __p1_567, __p2_567) __extension__ ({ \
	51346	uint32x4_t __ret_567; \
	51347	uint16x8_t __s0_567 = __p0_567; \
	51348	uint16x4_t __s1_567 = __p1_567; \
	51349	uint16x8_t __rev0_567; __rev0_567 = __builtin_shufflevector(__s0_567, __s0_567, 7, 6, 5, 4, 3, 2, 1, 0); \
	51350	uint16x4_t __rev1_567; __rev1_567 = __builtin_shufflevector(__s1_567, __s1_567, 3, 2, 1, 0); \
	51351	__ret_567 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_567), __noswap_splat_lane_u16(__rev1_567, __p2_567)); \
	51352	__ret_567 = __builtin_shufflevector(__ret_567, __ret_567, 3, 2, 1, 0); \
	51353	__ret_567; \
	51354	})
50104	51355	#endif
50105	51356
50106	51357	#ifdef __LITTLE_ENDIAN__
50107		__ai uint32x2_t vcltz_s32(int32x2_t __p0) {
50108		uint32x2_t __ret;
50109		__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 18);
	51358	#define vmull_high_lane_s32(__p0_568, __p1_568, __p2_568) __extension__ ({ \
	51359	int64x2_t __ret_568; \
	51360	int32x4_t __s0_568 = __p0_568; \
	51361	int32x2_t __s1_568 = __p1_568; \
	51362	__ret_568 = vmull_s32(vget_high_s32(__s0_568), splat_lane_s32(__s1_568, __p2_568)); \
	51363	__ret_568; \
	51364	})
	51365	#else
	51366	#define vmull_high_lane_s32(__p0_569, __p1_569, __p2_569) __extension__ ({ \
	51367	int64x2_t __ret_569; \
	51368	int32x4_t __s0_569 = __p0_569; \
	51369	int32x2_t __s1_569 = __p1_569; \
	51370	int32x4_t __rev0_569; __rev0_569 = __builtin_shufflevector(__s0_569, __s0_569, 3, 2, 1, 0); \
	51371	int32x2_t __rev1_569; __rev1_569 = __builtin_shufflevector(__s1_569, __s1_569, 1, 0); \
	51372	__ret_569 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_569), __noswap_splat_lane_s32(__rev1_569, __p2_569)); \
	51373	__ret_569 = __builtin_shufflevector(__ret_569, __ret_569, 1, 0); \
	51374	__ret_569; \
	51375	})
	51376	#endif
	51377
	51378	#ifdef __LITTLE_ENDIAN__
	51379	#define vmull_high_lane_s16(__p0_570, __p1_570, __p2_570) __extension__ ({ \
	51380	int32x4_t __ret_570; \
	51381	int16x8_t __s0_570 = __p0_570; \
	51382	int16x4_t __s1_570 = __p1_570; \
	51383	__ret_570 = vmull_s16(vget_high_s16(__s0_570), splat_lane_s16(__s1_570, __p2_570)); \
	51384	__ret_570; \
	51385	})
	51386	#else
	51387	#define vmull_high_lane_s16(__p0_571, __p1_571, __p2_571) __extension__ ({ \
	51388	int32x4_t __ret_571; \
	51389	int16x8_t __s0_571 = __p0_571; \
	51390	int16x4_t __s1_571 = __p1_571; \
	51391	int16x8_t __rev0_571; __rev0_571 = __builtin_shufflevector(__s0_571, __s0_571, 7, 6, 5, 4, 3, 2, 1, 0); \
	51392	int16x4_t __rev1_571; __rev1_571 = __builtin_shufflevector(__s1_571, __s1_571, 3, 2, 1, 0); \
	51393	__ret_571 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_571), __noswap_splat_lane_s16(__rev1_571, __p2_571)); \
	51394	__ret_571 = __builtin_shufflevector(__ret_571, __ret_571, 3, 2, 1, 0); \
	51395	__ret_571; \
	51396	})
	51397	#endif
	51398
	51399	#ifdef __LITTLE_ENDIAN__
	51400	#define vmull_high_laneq_u32(__p0_572, __p1_572, __p2_572) __extension__ ({ \
	51401	uint64x2_t __ret_572; \
	51402	uint32x4_t __s0_572 = __p0_572; \
	51403	uint32x4_t __s1_572 = __p1_572; \
	51404	__ret_572 = vmull_u32(vget_high_u32(__s0_572), splat_laneq_u32(__s1_572, __p2_572)); \
	51405	__ret_572; \
	51406	})
	51407	#else
	51408	#define vmull_high_laneq_u32(__p0_573, __p1_573, __p2_573) __extension__ ({ \
	51409	uint64x2_t __ret_573; \
	51410	uint32x4_t __s0_573 = __p0_573; \
	51411	uint32x4_t __s1_573 = __p1_573; \
	51412	uint32x4_t __rev0_573; __rev0_573 = __builtin_shufflevector(__s0_573, __s0_573, 3, 2, 1, 0); \
	51413	uint32x4_t __rev1_573; __rev1_573 = __builtin_shufflevector(__s1_573, __s1_573, 3, 2, 1, 0); \
	51414	__ret_573 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_573), __noswap_splat_laneq_u32(__rev1_573, __p2_573)); \
	51415	__ret_573 = __builtin_shufflevector(__ret_573, __ret_573, 1, 0); \
	51416	__ret_573; \
	51417	})
	51418	#endif
	51419
	51420	#ifdef __LITTLE_ENDIAN__
	51421	#define vmull_high_laneq_u16(__p0_574, __p1_574, __p2_574) __extension__ ({ \
	51422	uint32x4_t __ret_574; \
	51423	uint16x8_t __s0_574 = __p0_574; \
	51424	uint16x8_t __s1_574 = __p1_574; \
	51425	__ret_574 = vmull_u16(vget_high_u16(__s0_574), splat_laneq_u16(__s1_574, __p2_574)); \
	51426	__ret_574; \
	51427	})
	51428	#else
	51429	#define vmull_high_laneq_u16(__p0_575, __p1_575, __p2_575) __extension__ ({ \
	51430	uint32x4_t __ret_575; \
	51431	uint16x8_t __s0_575 = __p0_575; \
	51432	uint16x8_t __s1_575 = __p1_575; \
	51433	uint16x8_t __rev0_575; __rev0_575 = __builtin_shufflevector(__s0_575, __s0_575, 7, 6, 5, 4, 3, 2, 1, 0); \
	51434	uint16x8_t __rev1_575; __rev1_575 = __builtin_shufflevector(__s1_575, __s1_575, 7, 6, 5, 4, 3, 2, 1, 0); \
	51435	__ret_575 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_575), __noswap_splat_laneq_u16(__rev1_575, __p2_575)); \
	51436	__ret_575 = __builtin_shufflevector(__ret_575, __ret_575, 3, 2, 1, 0); \
	51437	__ret_575; \
	51438	})
	51439	#endif
	51440
	51441	#ifdef __LITTLE_ENDIAN__
	51442	#define vmull_high_laneq_s32(__p0_576, __p1_576, __p2_576) __extension__ ({ \
	51443	int64x2_t __ret_576; \
	51444	int32x4_t __s0_576 = __p0_576; \
	51445	int32x4_t __s1_576 = __p1_576; \
	51446	__ret_576 = vmull_s32(vget_high_s32(__s0_576), splat_laneq_s32(__s1_576, __p2_576)); \
	51447	__ret_576; \
	51448	})
	51449	#else
	51450	#define vmull_high_laneq_s32(__p0_577, __p1_577, __p2_577) __extension__ ({ \
	51451	int64x2_t __ret_577; \
	51452	int32x4_t __s0_577 = __p0_577; \
	51453	int32x4_t __s1_577 = __p1_577; \
	51454	int32x4_t __rev0_577; __rev0_577 = __builtin_shufflevector(__s0_577, __s0_577, 3, 2, 1, 0); \
	51455	int32x4_t __rev1_577; __rev1_577 = __builtin_shufflevector(__s1_577, __s1_577, 3, 2, 1, 0); \
	51456	__ret_577 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_577), __noswap_splat_laneq_s32(__rev1_577, __p2_577)); \
	51457	__ret_577 = __builtin_shufflevector(__ret_577, __ret_577, 1, 0); \
	51458	__ret_577; \
	51459	})
	51460	#endif
	51461
	51462	#ifdef __LITTLE_ENDIAN__
	51463	#define vmull_high_laneq_s16(__p0_578, __p1_578, __p2_578) __extension__ ({ \
	51464	int32x4_t __ret_578; \
	51465	int16x8_t __s0_578 = __p0_578; \
	51466	int16x8_t __s1_578 = __p1_578; \
	51467	__ret_578 = vmull_s16(vget_high_s16(__s0_578), splat_laneq_s16(__s1_578, __p2_578)); \
	51468	__ret_578; \
	51469	})
	51470	#else
	51471	#define vmull_high_laneq_s16(__p0_579, __p1_579, __p2_579) __extension__ ({ \
	51472	int32x4_t __ret_579; \
	51473	int16x8_t __s0_579 = __p0_579; \
	51474	int16x8_t __s1_579 = __p1_579; \
	51475	int16x8_t __rev0_579; __rev0_579 = __builtin_shufflevector(__s0_579, __s0_579, 7, 6, 5, 4, 3, 2, 1, 0); \
	51476	int16x8_t __rev1_579; __rev1_579 = __builtin_shufflevector(__s1_579, __s1_579, 7, 6, 5, 4, 3, 2, 1, 0); \
	51477	__ret_579 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_579), __noswap_splat_laneq_s16(__rev1_579, __p2_579)); \
	51478	__ret_579 = __builtin_shufflevector(__ret_579, __ret_579, 3, 2, 1, 0); \
	51479	__ret_579; \
	51480	})
	51481	#endif
	51482
	51483	#ifdef __LITTLE_ENDIAN__
	51484	__ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
	51485	uint64x2_t __ret;
	51486	__ret = vmull_n_u32(vget_high_u32(__p0), __p1);
50110	51487	return __ret;
50111	51488	}
50112	51489	#else
50113		__ai uint32x2_t vcltz_s32(int32x2_t __p0) {
50114		uint32x2_t __ret;
50115		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
50116		__ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 18);
	51490	__ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
	51491	uint64x2_t __ret;
	51492	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	51493	__ret = __noswap_vmull_n_u32(__noswap_vget_high_u32(__rev0), __p1);
50117	51494	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50118	51495	return __ret;
50119	51496	}
50120	51497	#endif
50121	51498
50122		__ai uint64x1_t vcltz_s64(int64x1_t __p0) {
50123		uint64x1_t __ret;
50124		__ret = (uint64x1_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 19);
50125		return __ret;
50126		}
50127	51499	#ifdef __LITTLE_ENDIAN__
50128		__ai uint16x4_t vcltz_s16(int16x4_t __p0) {
50129		uint16x4_t __ret;
50130		__ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__p0, 17);
	51500	__ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
	51501	uint32x4_t __ret;
	51502	__ret = vmull_n_u16(vget_high_u16(__p0), __p1);
50131	51503	return __ret;
50132	51504	}
50133	51505	#else
50134		__ai uint16x4_t vcltz_s16(int16x4_t __p0) {
50135		uint16x4_t __ret;
50136		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
50137		__ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 17);
	51506	__ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
	51507	uint32x4_t __ret;
	51508	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	51509	__ret = __noswap_vmull_n_u16(__noswap_vget_high_u16(__rev0), __p1);
50138	51510	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
50139	51511	return __ret;
50140	51512	}
50141	51513	#endif
50142	51514
50143		__ai uint64_t vcltzd_s64(int64_t __p0) {
50144		uint64_t __ret;
50145		__ret = (uint64_t) __builtin_neon_vcltzd_s64(__p0);
50146		return __ret;
50147		}
50148		__ai uint64_t vcltzd_f64(float64_t __p0) {
50149		uint64_t __ret;
50150		__ret = (uint64_t) __builtin_neon_vcltzd_f64(__p0);
50151		return __ret;
50152		}
50153		__ai uint32_t vcltzs_f32(float32_t __p0) {
50154		uint32_t __ret;
50155		__ret = (uint32_t) __builtin_neon_vcltzs_f32(__p0);
50156		return __ret;
50157		}
50158	51515	#ifdef __LITTLE_ENDIAN__
50159		__ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
50160		poly64x2_t __ret;
50161		__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
	51516	__ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
	51517	int64x2_t __ret;
	51518	__ret = vmull_n_s32(vget_high_s32(__p0), __p1);
50162	51519	return __ret;
50163	51520	}
50164	51521	#else
50165		__ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
50166		poly64x2_t __ret;
50167		__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
	51522	__ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
	51523	int64x2_t __ret;
	51524	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	51525	__ret = __noswap_vmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
50168	51526	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50169	51527	return __ret;
50170	51528	}
50171	51529	#endif
50172	51530
50173	51531	#ifdef __LITTLE_ENDIAN__
50174		__ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
50175		float64x2_t __ret;
50176		__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
	51532	__ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
	51533	int32x4_t __ret;
	51534	__ret = vmull_n_s16(vget_high_s16(__p0), __p1);
50177	51535	return __ret;
50178	51536	}
50179	51537	#else
50180		__ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
50181		float64x2_t __ret;
50182		__ret = __builtin_shufflevector(__p0, __p1, 0, 1);
50183		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	51538	__ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
	51539	int32x4_t __ret;
	51540	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	51541	__ret = __noswap_vmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
	51542	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
50184	51543	return __ret;
50185	51544	}
50186	51545	#endif
50187	51546
50188	51547	#ifdef __LITTLE_ENDIAN__
50189		#define vcopyq_lane_p8(__p0_357, __p1_357, __p2_357, __p3_357) __extension__ ({ \
50190		poly8x16_t __ret_357; \
50191		poly8x16_t __s0_357 = __p0_357; \
50192		poly8x8_t __s2_357 = __p2_357; \
50193		__ret_357 = vsetq_lane_p8(vget_lane_p8(__s2_357, __p3_357), __s0_357, __p1_357); \
50194		__ret_357; \
	51548	#define vmull_laneq_u32(__p0_580, __p1_580, __p2_580) __extension__ ({ \
	51549	uint64x2_t __ret_580; \
	51550	uint32x2_t __s0_580 = __p0_580; \
	51551	uint32x4_t __s1_580 = __p1_580; \
	51552	__ret_580 = vmull_u32(__s0_580, splat_laneq_u32(__s1_580, __p2_580)); \
	51553	__ret_580; \
50195	51554	})
50196	51555	#else
50197		#define vcopyq_lane_p8(__p0_358, __p1_358, __p2_358, __p3_358) __extension__ ({ \
50198		poly8x16_t __ret_358; \
50199		poly8x16_t __s0_358 = __p0_358; \
50200		poly8x8_t __s2_358 = __p2_358; \
50201		poly8x16_t __rev0_358; __rev0_358 = __builtin_shufflevector(__s0_358, __s0_358, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50202		poly8x8_t __rev2_358; __rev2_358 = __builtin_shufflevector(__s2_358, __s2_358, 7, 6, 5, 4, 3, 2, 1, 0); \
50203		__ret_358 = __noswap_vsetq_lane_p8(__noswap_vget_lane_p8(__rev2_358, __p3_358), __rev0_358, __p1_358); \
50204		__ret_358 = __builtin_shufflevector(__ret_358, __ret_358, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50205		__ret_358; \
	51556	#define vmull_laneq_u32(__p0_581, __p1_581, __p2_581) __extension__ ({ \
	51557	uint64x2_t __ret_581; \
	51558	uint32x2_t __s0_581 = __p0_581; \
	51559	uint32x4_t __s1_581 = __p1_581; \
	51560	uint32x2_t __rev0_581; __rev0_581 = __builtin_shufflevector(__s0_581, __s0_581, 1, 0); \
	51561	uint32x4_t __rev1_581; __rev1_581 = __builtin_shufflevector(__s1_581, __s1_581, 3, 2, 1, 0); \
	51562	__ret_581 = __noswap_vmull_u32(__rev0_581, __noswap_splat_laneq_u32(__rev1_581, __p2_581)); \
	51563	__ret_581 = __builtin_shufflevector(__ret_581, __ret_581, 1, 0); \
	51564	__ret_581; \
50206	51565	})
50207	51566	#endif
50208	51567
50209	51568	#ifdef __LITTLE_ENDIAN__
50210		#define vcopyq_lane_p16(__p0_359, __p1_359, __p2_359, __p3_359) __extension__ ({ \
50211		poly16x8_t __ret_359; \
50212		poly16x8_t __s0_359 = __p0_359; \
50213		poly16x4_t __s2_359 = __p2_359; \
50214		__ret_359 = vsetq_lane_p16(vget_lane_p16(__s2_359, __p3_359), __s0_359, __p1_359); \
50215		__ret_359; \
	51569	#define vmull_laneq_u16(__p0_582, __p1_582, __p2_582) __extension__ ({ \
	51570	uint32x4_t __ret_582; \
	51571	uint16x4_t __s0_582 = __p0_582; \
	51572	uint16x8_t __s1_582 = __p1_582; \
	51573	__ret_582 = vmull_u16(__s0_582, splat_laneq_u16(__s1_582, __p2_582)); \
	51574	__ret_582; \
50216	51575	})
50217	51576	#else
50218		#define vcopyq_lane_p16(__p0_360, __p1_360, __p2_360, __p3_360) __extension__ ({ \
50219		poly16x8_t __ret_360; \
50220		poly16x8_t __s0_360 = __p0_360; \
50221		poly16x4_t __s2_360 = __p2_360; \
50222		poly16x8_t __rev0_360; __rev0_360 = __builtin_shufflevector(__s0_360, __s0_360, 7, 6, 5, 4, 3, 2, 1, 0); \
50223		poly16x4_t __rev2_360; __rev2_360 = __builtin_shufflevector(__s2_360, __s2_360, 3, 2, 1, 0); \
50224		__ret_360 = __noswap_vsetq_lane_p16(__noswap_vget_lane_p16(__rev2_360, __p3_360), __rev0_360, __p1_360); \
50225		__ret_360 = __builtin_shufflevector(__ret_360, __ret_360, 7, 6, 5, 4, 3, 2, 1, 0); \
50226		__ret_360; \
	51577	#define vmull_laneq_u16(__p0_583, __p1_583, __p2_583) __extension__ ({ \
	51578	uint32x4_t __ret_583; \
	51579	uint16x4_t __s0_583 = __p0_583; \
	51580	uint16x8_t __s1_583 = __p1_583; \
	51581	uint16x4_t __rev0_583; __rev0_583 = __builtin_shufflevector(__s0_583, __s0_583, 3, 2, 1, 0); \
	51582	uint16x8_t __rev1_583; __rev1_583 = __builtin_shufflevector(__s1_583, __s1_583, 7, 6, 5, 4, 3, 2, 1, 0); \
	51583	__ret_583 = __noswap_vmull_u16(__rev0_583, __noswap_splat_laneq_u16(__rev1_583, __p2_583)); \
	51584	__ret_583 = __builtin_shufflevector(__ret_583, __ret_583, 3, 2, 1, 0); \
	51585	__ret_583; \
50227	51586	})
50228	51587	#endif
50229	51588
50230	51589	#ifdef __LITTLE_ENDIAN__
50231		#define vcopyq_lane_u8(__p0_361, __p1_361, __p2_361, __p3_361) __extension__ ({ \
50232		uint8x16_t __ret_361; \
50233		uint8x16_t __s0_361 = __p0_361; \
50234		uint8x8_t __s2_361 = __p2_361; \
50235		__ret_361 = vsetq_lane_u8(vget_lane_u8(__s2_361, __p3_361), __s0_361, __p1_361); \
50236		__ret_361; \
	51590	#define vmull_laneq_s32(__p0_584, __p1_584, __p2_584) __extension__ ({ \
	51591	int64x2_t __ret_584; \
	51592	int32x2_t __s0_584 = __p0_584; \
	51593	int32x4_t __s1_584 = __p1_584; \
	51594	__ret_584 = vmull_s32(__s0_584, splat_laneq_s32(__s1_584, __p2_584)); \
	51595	__ret_584; \
50237	51596	})
50238	51597	#else
50239		#define vcopyq_lane_u8(__p0_362, __p1_362, __p2_362, __p3_362) __extension__ ({ \
50240		uint8x16_t __ret_362; \
50241		uint8x16_t __s0_362 = __p0_362; \
50242		uint8x8_t __s2_362 = __p2_362; \
50243		uint8x16_t __rev0_362; __rev0_362 = __builtin_shufflevector(__s0_362, __s0_362, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50244		uint8x8_t __rev2_362; __rev2_362 = __builtin_shufflevector(__s2_362, __s2_362, 7, 6, 5, 4, 3, 2, 1, 0); \
50245		__ret_362 = __noswap_vsetq_lane_u8(__noswap_vget_lane_u8(__rev2_362, __p3_362), __rev0_362, __p1_362); \
50246		__ret_362 = __builtin_shufflevector(__ret_362, __ret_362, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50247		__ret_362; \
	51598	#define vmull_laneq_s32(__p0_585, __p1_585, __p2_585) __extension__ ({ \
	51599	int64x2_t __ret_585; \
	51600	int32x2_t __s0_585 = __p0_585; \
	51601	int32x4_t __s1_585 = __p1_585; \
	51602	int32x2_t __rev0_585; __rev0_585 = __builtin_shufflevector(__s0_585, __s0_585, 1, 0); \
	51603	int32x4_t __rev1_585; __rev1_585 = __builtin_shufflevector(__s1_585, __s1_585, 3, 2, 1, 0); \
	51604	__ret_585 = __noswap_vmull_s32(__rev0_585, __noswap_splat_laneq_s32(__rev1_585, __p2_585)); \
	51605	__ret_585 = __builtin_shufflevector(__ret_585, __ret_585, 1, 0); \
	51606	__ret_585; \
50248	51607	})
50249	51608	#endif
50250	51609
50251	51610	#ifdef __LITTLE_ENDIAN__
50252		#define vcopyq_lane_u32(__p0_363, __p1_363, __p2_363, __p3_363) __extension__ ({ \
50253		uint32x4_t __ret_363; \
50254		uint32x4_t __s0_363 = __p0_363; \
50255		uint32x2_t __s2_363 = __p2_363; \
50256		__ret_363 = vsetq_lane_u32(vget_lane_u32(__s2_363, __p3_363), __s0_363, __p1_363); \
50257		__ret_363; \
	51611	#define vmull_laneq_s16(__p0_586, __p1_586, __p2_586) __extension__ ({ \
	51612	int32x4_t __ret_586; \
	51613	int16x4_t __s0_586 = __p0_586; \
	51614	int16x8_t __s1_586 = __p1_586; \
	51615	__ret_586 = vmull_s16(__s0_586, splat_laneq_s16(__s1_586, __p2_586)); \
	51616	__ret_586; \
50258	51617	})
50259	51618	#else
50260		#define vcopyq_lane_u32(__p0_364, __p1_364, __p2_364, __p3_364) __extension__ ({ \
50261		uint32x4_t __ret_364; \
50262		uint32x4_t __s0_364 = __p0_364; \
50263		uint32x2_t __s2_364 = __p2_364; \
50264		uint32x4_t __rev0_364; __rev0_364 = __builtin_shufflevector(__s0_364, __s0_364, 3, 2, 1, 0); \
50265		uint32x2_t __rev2_364; __rev2_364 = __builtin_shufflevector(__s2_364, __s2_364, 1, 0); \
50266		__ret_364 = __noswap_vsetq_lane_u32(__noswap_vget_lane_u32(__rev2_364, __p3_364), __rev0_364, __p1_364); \
50267		__ret_364 = __builtin_shufflevector(__ret_364, __ret_364, 3, 2, 1, 0); \
50268		__ret_364; \
	51619	#define vmull_laneq_s16(__p0_587, __p1_587, __p2_587) __extension__ ({ \
	51620	int32x4_t __ret_587; \
	51621	int16x4_t __s0_587 = __p0_587; \
	51622	int16x8_t __s1_587 = __p1_587; \
	51623	int16x4_t __rev0_587; __rev0_587 = __builtin_shufflevector(__s0_587, __s0_587, 3, 2, 1, 0); \
	51624	int16x8_t __rev1_587; __rev1_587 = __builtin_shufflevector(__s1_587, __s1_587, 7, 6, 5, 4, 3, 2, 1, 0); \
	51625	__ret_587 = __noswap_vmull_s16(__rev0_587, __noswap_splat_laneq_s16(__rev1_587, __p2_587)); \
	51626	__ret_587 = __builtin_shufflevector(__ret_587, __ret_587, 3, 2, 1, 0); \
	51627	__ret_587; \
50269	51628	})
50270	51629	#endif
50271	51630
50272	51631	#ifdef __LITTLE_ENDIAN__
50273		#define vcopyq_lane_u64(__p0_365, __p1_365, __p2_365, __p3_365) __extension__ ({ \
50274		uint64x2_t __ret_365; \
50275		uint64x2_t __s0_365 = __p0_365; \
50276		uint64x1_t __s2_365 = __p2_365; \
50277		__ret_365 = vsetq_lane_u64(vget_lane_u64(__s2_365, __p3_365), __s0_365, __p1_365); \
50278		__ret_365; \
50279		})
	51632	__ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
	51633	float64x2_t __ret;
	51634	__ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	51635	return __ret;
	51636	}
50280	51637	#else
50281		#define vcopyq_lane_u64(__p0_366, __p1_366, __p2_366, __p3_366) __extension__ ({ \
50282		uint64x2_t __ret_366; \
50283		uint64x2_t __s0_366 = __p0_366; \
50284		uint64x1_t __s2_366 = __p2_366; \
50285		uint64x2_t __rev0_366; __rev0_366 = __builtin_shufflevector(__s0_366, __s0_366, 1, 0); \
50286		__ret_366 = __noswap_vsetq_lane_u64(vget_lane_u64(__s2_366, __p3_366), __rev0_366, __p1_366); \
50287		__ret_366 = __builtin_shufflevector(__ret_366, __ret_366, 1, 0); \
50288		__ret_366; \
50289		})
	51638	__ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
	51639	float64x2_t __ret;
	51640	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	51641	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	51642	__ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	51643	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	51644	return __ret;
	51645	}
	51646	__ai float64x2_t __noswap_vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
	51647	float64x2_t __ret;
	51648	__ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	51649	return __ret;
	51650	}
50290	51651	#endif
50291	51652
50292	51653	#ifdef __LITTLE_ENDIAN__
50293		#define vcopyq_lane_u16(__p0_367, __p1_367, __p2_367, __p3_367) __extension__ ({ \
50294		uint16x8_t __ret_367; \
50295		uint16x8_t __s0_367 = __p0_367; \
50296		uint16x4_t __s2_367 = __p2_367; \
50297		__ret_367 = vsetq_lane_u16(vget_lane_u16(__s2_367, __p3_367), __s0_367, __p1_367); \
50298		__ret_367; \
50299		})
	51654	__ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
	51655	float32x4_t __ret;
	51656	__ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	51657	return __ret;
	51658	}
50300	51659	#else
50301		#define vcopyq_lane_u16(__p0_368, __p1_368, __p2_368, __p3_368) __extension__ ({ \
50302		uint16x8_t __ret_368; \
50303		uint16x8_t __s0_368 = __p0_368; \
50304		uint16x4_t __s2_368 = __p2_368; \
50305		uint16x8_t __rev0_368; __rev0_368 = __builtin_shufflevector(__s0_368, __s0_368, 7, 6, 5, 4, 3, 2, 1, 0); \
50306		uint16x4_t __rev2_368; __rev2_368 = __builtin_shufflevector(__s2_368, __s2_368, 3, 2, 1, 0); \
50307		__ret_368 = __noswap_vsetq_lane_u16(__noswap_vget_lane_u16(__rev2_368, __p3_368), __rev0_368, __p1_368); \
50308		__ret_368 = __builtin_shufflevector(__ret_368, __ret_368, 7, 6, 5, 4, 3, 2, 1, 0); \
50309		__ret_368; \
50310		})
	51660	__ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
	51661	float32x4_t __ret;
	51662	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	51663	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	51664	__ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	51665	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	51666	return __ret;
	51667	}
	51668	__ai float32x4_t __noswap_vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
	51669	float32x4_t __ret;
	51670	__ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	51671	return __ret;
	51672	}
50311	51673	#endif
50312	51674
	51675	__ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
	51676	float64x1_t __ret;
	51677	__ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	51678	return __ret;
	51679	}
50313	51680	#ifdef __LITTLE_ENDIAN__
50314		#define vcopyq_lane_s8(__p0_369, __p1_369, __p2_369, __p3_369) __extension__ ({ \
50315		int8x16_t __ret_369; \
50316		int8x16_t __s0_369 = __p0_369; \
50317		int8x8_t __s2_369 = __p2_369; \
50318		__ret_369 = vsetq_lane_s8(vget_lane_s8(__s2_369, __p3_369), __s0_369, __p1_369); \
50319		__ret_369; \
50320		})
	51681	__ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
	51682	float32x2_t __ret;
	51683	__ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	51684	return __ret;
	51685	}
50321	51686	#else
50322		#define vcopyq_lane_s8(__p0_370, __p1_370, __p2_370, __p3_370) __extension__ ({ \
50323		int8x16_t __ret_370; \
50324		int8x16_t __s0_370 = __p0_370; \
50325		int8x8_t __s2_370 = __p2_370; \
50326		int8x16_t __rev0_370; __rev0_370 = __builtin_shufflevector(__s0_370, __s0_370, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50327		int8x8_t __rev2_370; __rev2_370 = __builtin_shufflevector(__s2_370, __s2_370, 7, 6, 5, 4, 3, 2, 1, 0); \
50328		__ret_370 = __noswap_vsetq_lane_s8(__noswap_vget_lane_s8(__rev2_370, __p3_370), __rev0_370, __p1_370); \
50329		__ret_370 = __builtin_shufflevector(__ret_370, __ret_370, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50330		__ret_370; \
50331		})
	51687	__ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
	51688	float32x2_t __ret;
	51689	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	51690	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	51691	__ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
	51692	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	51693	return __ret;
	51694	}
	51695	__ai float32x2_t __noswap_vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
	51696	float32x2_t __ret;
	51697	__ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	51698	return __ret;
	51699	}
50332	51700	#endif
50333	51701
	51702	__ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
	51703	float64_t __ret;
	51704	__ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
	51705	return __ret;
	51706	}
	51707	__ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
	51708	float32_t __ret;
	51709	__ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
	51710	return __ret;
	51711	}
	51712	#define vmulxd_lane_f64(__p0_588, __p1_588, __p2_588) __extension__ ({ \
	51713	float64_t __ret_588; \
	51714	float64_t __s0_588 = __p0_588; \
	51715	float64x1_t __s1_588 = __p1_588; \
	51716	__ret_588 = vmulxd_f64(__s0_588, vget_lane_f64(__s1_588, __p2_588)); \
	51717	__ret_588; \
	51718	})
50334	51719	#ifdef __LITTLE_ENDIAN__
50335		#define vcopyq_lane_f32(__p0_371, __p1_371, __p2_371, __p3_371) __extension__ ({ \
50336		float32x4_t __ret_371; \
50337		float32x4_t __s0_371 = __p0_371; \
50338		float32x2_t __s2_371 = __p2_371; \
50339		__ret_371 = vsetq_lane_f32(vget_lane_f32(__s2_371, __p3_371), __s0_371, __p1_371); \
50340		__ret_371; \
	51720	#define vmulxs_lane_f32(__p0_589, __p1_589, __p2_589) __extension__ ({ \
	51721	float32_t __ret_589; \
	51722	float32_t __s0_589 = __p0_589; \
	51723	float32x2_t __s1_589 = __p1_589; \
	51724	__ret_589 = vmulxs_f32(__s0_589, vget_lane_f32(__s1_589, __p2_589)); \
	51725	__ret_589; \
50341	51726	})
50342	51727	#else
50343		#define vcopyq_lane_f32(__p0_372, __p1_372, __p2_372, __p3_372) __extension__ ({ \
50344		float32x4_t __ret_372; \
50345		float32x4_t __s0_372 = __p0_372; \
50346		float32x2_t __s2_372 = __p2_372; \
50347		float32x4_t __rev0_372; __rev0_372 = __builtin_shufflevector(__s0_372, __s0_372, 3, 2, 1, 0); \
50348		float32x2_t __rev2_372; __rev2_372 = __builtin_shufflevector(__s2_372, __s2_372, 1, 0); \
50349		__ret_372 = __noswap_vsetq_lane_f32(__noswap_vget_lane_f32(__rev2_372, __p3_372), __rev0_372, __p1_372); \
50350		__ret_372 = __builtin_shufflevector(__ret_372, __ret_372, 3, 2, 1, 0); \
50351		__ret_372; \
	51728	#define vmulxs_lane_f32(__p0_590, __p1_590, __p2_590) __extension__ ({ \
	51729	float32_t __ret_590; \
	51730	float32_t __s0_590 = __p0_590; \
	51731	float32x2_t __s1_590 = __p1_590; \
	51732	float32x2_t __rev1_590; __rev1_590 = __builtin_shufflevector(__s1_590, __s1_590, 1, 0); \
	51733	__ret_590 = vmulxs_f32(__s0_590, __noswap_vget_lane_f32(__rev1_590, __p2_590)); \
	51734	__ret_590; \
50352	51735	})
50353	51736	#endif
50354	51737
50355	51738	#ifdef __LITTLE_ENDIAN__
50356		#define vcopyq_lane_s32(__p0_373, __p1_373, __p2_373, __p3_373) __extension__ ({ \
50357		int32x4_t __ret_373; \
50358		int32x4_t __s0_373 = __p0_373; \
50359		int32x2_t __s2_373 = __p2_373; \
50360		__ret_373 = vsetq_lane_s32(vget_lane_s32(__s2_373, __p3_373), __s0_373, __p1_373); \
50361		__ret_373; \
	51739	#define vmulxq_lane_f64(__p0_591, __p1_591, __p2_591) __extension__ ({ \
	51740	float64x2_t __ret_591; \
	51741	float64x2_t __s0_591 = __p0_591; \
	51742	float64x1_t __s1_591 = __p1_591; \
	51743	__ret_591 = vmulxq_f64(__s0_591, splatq_lane_f64(__s1_591, __p2_591)); \
	51744	__ret_591; \
50362	51745	})
50363	51746	#else
50364		#define vcopyq_lane_s32(__p0_374, __p1_374, __p2_374, __p3_374) __extension__ ({ \
50365		int32x4_t __ret_374; \
50366		int32x4_t __s0_374 = __p0_374; \
50367		int32x2_t __s2_374 = __p2_374; \
50368		int32x4_t __rev0_374; __rev0_374 = __builtin_shufflevector(__s0_374, __s0_374, 3, 2, 1, 0); \
50369		int32x2_t __rev2_374; __rev2_374 = __builtin_shufflevector(__s2_374, __s2_374, 1, 0); \
50370		__ret_374 = __noswap_vsetq_lane_s32(__noswap_vget_lane_s32(__rev2_374, __p3_374), __rev0_374, __p1_374); \
50371		__ret_374 = __builtin_shufflevector(__ret_374, __ret_374, 3, 2, 1, 0); \
50372		__ret_374; \
	51747	#define vmulxq_lane_f64(__p0_592, __p1_592, __p2_592) __extension__ ({ \
	51748	float64x2_t __ret_592; \
	51749	float64x2_t __s0_592 = __p0_592; \
	51750	float64x1_t __s1_592 = __p1_592; \
	51751	float64x2_t __rev0_592; __rev0_592 = __builtin_shufflevector(__s0_592, __s0_592, 1, 0); \
	51752	__ret_592 = __noswap_vmulxq_f64(__rev0_592, __noswap_splatq_lane_f64(__s1_592, __p2_592)); \
	51753	__ret_592 = __builtin_shufflevector(__ret_592, __ret_592, 1, 0); \
	51754	__ret_592; \
50373	51755	})
50374	51756	#endif
50375	51757
50376	51758	#ifdef __LITTLE_ENDIAN__
50377		#define vcopyq_lane_s64(__p0_375, __p1_375, __p2_375, __p3_375) __extension__ ({ \
50378		int64x2_t __ret_375; \
50379		int64x2_t __s0_375 = __p0_375; \
50380		int64x1_t __s2_375 = __p2_375; \
50381		__ret_375 = vsetq_lane_s64(vget_lane_s64(__s2_375, __p3_375), __s0_375, __p1_375); \
50382		__ret_375; \
	51759	#define vmulxq_lane_f32(__p0_593, __p1_593, __p2_593) __extension__ ({ \
	51760	float32x4_t __ret_593; \
	51761	float32x4_t __s0_593 = __p0_593; \
	51762	float32x2_t __s1_593 = __p1_593; \
	51763	__ret_593 = vmulxq_f32(__s0_593, splatq_lane_f32(__s1_593, __p2_593)); \
	51764	__ret_593; \
50383	51765	})
50384	51766	#else
50385		#define vcopyq_lane_s64(__p0_376, __p1_376, __p2_376, __p3_376) __extension__ ({ \
50386		int64x2_t __ret_376; \
50387		int64x2_t __s0_376 = __p0_376; \
50388		int64x1_t __s2_376 = __p2_376; \
50389		int64x2_t __rev0_376; __rev0_376 = __builtin_shufflevector(__s0_376, __s0_376, 1, 0); \
50390		__ret_376 = __noswap_vsetq_lane_s64(vget_lane_s64(__s2_376, __p3_376), __rev0_376, __p1_376); \
50391		__ret_376 = __builtin_shufflevector(__ret_376, __ret_376, 1, 0); \
50392		__ret_376; \
	51767	#define vmulxq_lane_f32(__p0_594, __p1_594, __p2_594) __extension__ ({ \
	51768	float32x4_t __ret_594; \
	51769	float32x4_t __s0_594 = __p0_594; \
	51770	float32x2_t __s1_594 = __p1_594; \
	51771	float32x4_t __rev0_594; __rev0_594 = __builtin_shufflevector(__s0_594, __s0_594, 3, 2, 1, 0); \
	51772	float32x2_t __rev1_594; __rev1_594 = __builtin_shufflevector(__s1_594, __s1_594, 1, 0); \
	51773	__ret_594 = __noswap_vmulxq_f32(__rev0_594, __noswap_splatq_lane_f32(__rev1_594, __p2_594)); \
	51774	__ret_594 = __builtin_shufflevector(__ret_594, __ret_594, 3, 2, 1, 0); \
	51775	__ret_594; \
50393	51776	})
50394	51777	#endif
50395	51778
50396	51779	#ifdef __LITTLE_ENDIAN__
50397		#define vcopyq_lane_s16(__p0_377, __p1_377, __p2_377, __p3_377) __extension__ ({ \
50398		int16x8_t __ret_377; \
50399		int16x8_t __s0_377 = __p0_377; \
50400		int16x4_t __s2_377 = __p2_377; \
50401		__ret_377 = vsetq_lane_s16(vget_lane_s16(__s2_377, __p3_377), __s0_377, __p1_377); \
50402		__ret_377; \
	51780	#define vmulx_lane_f32(__p0_595, __p1_595, __p2_595) __extension__ ({ \
	51781	float32x2_t __ret_595; \
	51782	float32x2_t __s0_595 = __p0_595; \
	51783	float32x2_t __s1_595 = __p1_595; \
	51784	__ret_595 = vmulx_f32(__s0_595, splat_lane_f32(__s1_595, __p2_595)); \
	51785	__ret_595; \
50403	51786	})
50404	51787	#else
50405		#define vcopyq_lane_s16(__p0_378, __p1_378, __p2_378, __p3_378) __extension__ ({ \
50406		int16x8_t __ret_378; \
50407		int16x8_t __s0_378 = __p0_378; \
50408		int16x4_t __s2_378 = __p2_378; \
50409		int16x8_t __rev0_378; __rev0_378 = __builtin_shufflevector(__s0_378, __s0_378, 7, 6, 5, 4, 3, 2, 1, 0); \
50410		int16x4_t __rev2_378; __rev2_378 = __builtin_shufflevector(__s2_378, __s2_378, 3, 2, 1, 0); \
50411		__ret_378 = __noswap_vsetq_lane_s16(__noswap_vget_lane_s16(__rev2_378, __p3_378), __rev0_378, __p1_378); \
50412		__ret_378 = __builtin_shufflevector(__ret_378, __ret_378, 7, 6, 5, 4, 3, 2, 1, 0); \
50413		__ret_378; \
	51788	#define vmulx_lane_f32(__p0_596, __p1_596, __p2_596) __extension__ ({ \
	51789	float32x2_t __ret_596; \
	51790	float32x2_t __s0_596 = __p0_596; \
	51791	float32x2_t __s1_596 = __p1_596; \
	51792	float32x2_t __rev0_596; __rev0_596 = __builtin_shufflevector(__s0_596, __s0_596, 1, 0); \
	51793	float32x2_t __rev1_596; __rev1_596 = __builtin_shufflevector(__s1_596, __s1_596, 1, 0); \
	51794	__ret_596 = __noswap_vmulx_f32(__rev0_596, __noswap_splat_lane_f32(__rev1_596, __p2_596)); \
	51795	__ret_596 = __builtin_shufflevector(__ret_596, __ret_596, 1, 0); \
	51796	__ret_596; \
50414	51797	})
50415	51798	#endif
50416	51799
50417	51800	#ifdef __LITTLE_ENDIAN__
50418		#define vcopy_lane_p8(__p0_379, __p1_379, __p2_379, __p3_379) __extension__ ({ \
50419		poly8x8_t __ret_379; \
50420		poly8x8_t __s0_379 = __p0_379; \
50421		poly8x8_t __s2_379 = __p2_379; \
50422		__ret_379 = vset_lane_p8(vget_lane_p8(__s2_379, __p3_379), __s0_379, __p1_379); \
50423		__ret_379; \
	51801	#define vmulxd_laneq_f64(__p0_597, __p1_597, __p2_597) __extension__ ({ \
	51802	float64_t __ret_597; \
	51803	float64_t __s0_597 = __p0_597; \
	51804	float64x2_t __s1_597 = __p1_597; \
	51805	__ret_597 = vmulxd_f64(__s0_597, vgetq_lane_f64(__s1_597, __p2_597)); \
	51806	__ret_597; \
50424	51807	})
50425	51808	#else
50426		#define vcopy_lane_p8(__p0_380, __p1_380, __p2_380, __p3_380) __extension__ ({ \
50427		poly8x8_t __ret_380; \
50428		poly8x8_t __s0_380 = __p0_380; \
50429		poly8x8_t __s2_380 = __p2_380; \
50430		poly8x8_t __rev0_380; __rev0_380 = __builtin_shufflevector(__s0_380, __s0_380, 7, 6, 5, 4, 3, 2, 1, 0); \
50431		poly8x8_t __rev2_380; __rev2_380 = __builtin_shufflevector(__s2_380, __s2_380, 7, 6, 5, 4, 3, 2, 1, 0); \
50432		__ret_380 = __noswap_vset_lane_p8(__noswap_vget_lane_p8(__rev2_380, __p3_380), __rev0_380, __p1_380); \
50433		__ret_380 = __builtin_shufflevector(__ret_380, __ret_380, 7, 6, 5, 4, 3, 2, 1, 0); \
50434		__ret_380; \
	51809	#define vmulxd_laneq_f64(__p0_598, __p1_598, __p2_598) __extension__ ({ \
	51810	float64_t __ret_598; \
	51811	float64_t __s0_598 = __p0_598; \
	51812	float64x2_t __s1_598 = __p1_598; \
	51813	float64x2_t __rev1_598; __rev1_598 = __builtin_shufflevector(__s1_598, __s1_598, 1, 0); \
	51814	__ret_598 = vmulxd_f64(__s0_598, __noswap_vgetq_lane_f64(__rev1_598, __p2_598)); \
	51815	__ret_598; \
50435	51816	})
50436	51817	#endif
50437	51818
50438	51819	#ifdef __LITTLE_ENDIAN__
50439		#define vcopy_lane_p16(__p0_381, __p1_381, __p2_381, __p3_381) __extension__ ({ \
50440		poly16x4_t __ret_381; \
50441		poly16x4_t __s0_381 = __p0_381; \
50442		poly16x4_t __s2_381 = __p2_381; \
50443		__ret_381 = vset_lane_p16(vget_lane_p16(__s2_381, __p3_381), __s0_381, __p1_381); \
50444		__ret_381; \
	51820	#define vmulxs_laneq_f32(__p0_599, __p1_599, __p2_599) __extension__ ({ \
	51821	float32_t __ret_599; \
	51822	float32_t __s0_599 = __p0_599; \
	51823	float32x4_t __s1_599 = __p1_599; \
	51824	__ret_599 = vmulxs_f32(__s0_599, vgetq_lane_f32(__s1_599, __p2_599)); \
	51825	__ret_599; \
50445	51826	})
50446	51827	#else
50447		#define vcopy_lane_p16(__p0_382, __p1_382, __p2_382, __p3_382) __extension__ ({ \
50448		poly16x4_t __ret_382; \
50449		poly16x4_t __s0_382 = __p0_382; \
50450		poly16x4_t __s2_382 = __p2_382; \
50451		poly16x4_t __rev0_382; __rev0_382 = __builtin_shufflevector(__s0_382, __s0_382, 3, 2, 1, 0); \
50452		poly16x4_t __rev2_382; __rev2_382 = __builtin_shufflevector(__s2_382, __s2_382, 3, 2, 1, 0); \
50453		__ret_382 = __noswap_vset_lane_p16(__noswap_vget_lane_p16(__rev2_382, __p3_382), __rev0_382, __p1_382); \
50454		__ret_382 = __builtin_shufflevector(__ret_382, __ret_382, 3, 2, 1, 0); \
50455		__ret_382; \
	51828	#define vmulxs_laneq_f32(__p0_600, __p1_600, __p2_600) __extension__ ({ \
	51829	float32_t __ret_600; \
	51830	float32_t __s0_600 = __p0_600; \
	51831	float32x4_t __s1_600 = __p1_600; \
	51832	float32x4_t __rev1_600; __rev1_600 = __builtin_shufflevector(__s1_600, __s1_600, 3, 2, 1, 0); \
	51833	__ret_600 = vmulxs_f32(__s0_600, __noswap_vgetq_lane_f32(__rev1_600, __p2_600)); \
	51834	__ret_600; \
50456	51835	})
50457	51836	#endif
50458	51837
50459	51838	#ifdef __LITTLE_ENDIAN__
50460		#define vcopy_lane_u8(__p0_383, __p1_383, __p2_383, __p3_383) __extension__ ({ \
50461		uint8x8_t __ret_383; \
50462		uint8x8_t __s0_383 = __p0_383; \
50463		uint8x8_t __s2_383 = __p2_383; \
50464		__ret_383 = vset_lane_u8(vget_lane_u8(__s2_383, __p3_383), __s0_383, __p1_383); \
50465		__ret_383; \
	51839	#define vmulxq_laneq_f64(__p0_601, __p1_601, __p2_601) __extension__ ({ \
	51840	float64x2_t __ret_601; \
	51841	float64x2_t __s0_601 = __p0_601; \
	51842	float64x2_t __s1_601 = __p1_601; \
	51843	__ret_601 = vmulxq_f64(__s0_601, splatq_laneq_f64(__s1_601, __p2_601)); \
	51844	__ret_601; \
50466	51845	})
50467	51846	#else
50468		#define vcopy_lane_u8(__p0_384, __p1_384, __p2_384, __p3_384) __extension__ ({ \
50469		uint8x8_t __ret_384; \
50470		uint8x8_t __s0_384 = __p0_384; \
50471		uint8x8_t __s2_384 = __p2_384; \
50472		uint8x8_t __rev0_384; __rev0_384 = __builtin_shufflevector(__s0_384, __s0_384, 7, 6, 5, 4, 3, 2, 1, 0); \
50473		uint8x8_t __rev2_384; __rev2_384 = __builtin_shufflevector(__s2_384, __s2_384, 7, 6, 5, 4, 3, 2, 1, 0); \
50474		__ret_384 = __noswap_vset_lane_u8(__noswap_vget_lane_u8(__rev2_384, __p3_384), __rev0_384, __p1_384); \
50475		__ret_384 = __builtin_shufflevector(__ret_384, __ret_384, 7, 6, 5, 4, 3, 2, 1, 0); \
50476		__ret_384; \
	51847	#define vmulxq_laneq_f64(__p0_602, __p1_602, __p2_602) __extension__ ({ \
	51848	float64x2_t __ret_602; \
	51849	float64x2_t __s0_602 = __p0_602; \
	51850	float64x2_t __s1_602 = __p1_602; \
	51851	float64x2_t __rev0_602; __rev0_602 = __builtin_shufflevector(__s0_602, __s0_602, 1, 0); \
	51852	float64x2_t __rev1_602; __rev1_602 = __builtin_shufflevector(__s1_602, __s1_602, 1, 0); \
	51853	__ret_602 = __noswap_vmulxq_f64(__rev0_602, __noswap_splatq_laneq_f64(__rev1_602, __p2_602)); \
	51854	__ret_602 = __builtin_shufflevector(__ret_602, __ret_602, 1, 0); \
	51855	__ret_602; \
50477	51856	})
50478	51857	#endif
50479	51858
50480	51859	#ifdef __LITTLE_ENDIAN__
50481		#define vcopy_lane_u32(__p0_385, __p1_385, __p2_385, __p3_385) __extension__ ({ \
50482		uint32x2_t __ret_385; \
50483		uint32x2_t __s0_385 = __p0_385; \
50484		uint32x2_t __s2_385 = __p2_385; \
50485		__ret_385 = vset_lane_u32(vget_lane_u32(__s2_385, __p3_385), __s0_385, __p1_385); \
50486		__ret_385; \
	51860	#define vmulxq_laneq_f32(__p0_603, __p1_603, __p2_603) __extension__ ({ \
	51861	float32x4_t __ret_603; \
	51862	float32x4_t __s0_603 = __p0_603; \
	51863	float32x4_t __s1_603 = __p1_603; \
	51864	__ret_603 = vmulxq_f32(__s0_603, splatq_laneq_f32(__s1_603, __p2_603)); \
	51865	__ret_603; \
50487	51866	})
50488	51867	#else
50489		#define vcopy_lane_u32(__p0_386, __p1_386, __p2_386, __p3_386) __extension__ ({ \
50490		uint32x2_t __ret_386; \
50491		uint32x2_t __s0_386 = __p0_386; \
50492		uint32x2_t __s2_386 = __p2_386; \
50493		uint32x2_t __rev0_386; __rev0_386 = __builtin_shufflevector(__s0_386, __s0_386, 1, 0); \
50494		uint32x2_t __rev2_386; __rev2_386 = __builtin_shufflevector(__s2_386, __s2_386, 1, 0); \
50495		__ret_386 = __noswap_vset_lane_u32(__noswap_vget_lane_u32(__rev2_386, __p3_386), __rev0_386, __p1_386); \
50496		__ret_386 = __builtin_shufflevector(__ret_386, __ret_386, 1, 0); \
50497		__ret_386; \
	51868	#define vmulxq_laneq_f32(__p0_604, __p1_604, __p2_604) __extension__ ({ \
	51869	float32x4_t __ret_604; \
	51870	float32x4_t __s0_604 = __p0_604; \
	51871	float32x4_t __s1_604 = __p1_604; \
	51872	float32x4_t __rev0_604; __rev0_604 = __builtin_shufflevector(__s0_604, __s0_604, 3, 2, 1, 0); \
	51873	float32x4_t __rev1_604; __rev1_604 = __builtin_shufflevector(__s1_604, __s1_604, 3, 2, 1, 0); \
	51874	__ret_604 = __noswap_vmulxq_f32(__rev0_604, __noswap_splatq_laneq_f32(__rev1_604, __p2_604)); \
	51875	__ret_604 = __builtin_shufflevector(__ret_604, __ret_604, 3, 2, 1, 0); \
	51876	__ret_604; \
50498	51877	})
50499	51878	#endif
50500	51879
50501		#define vcopy_lane_u64(__p0_387, __p1_387, __p2_387, __p3_387) __extension__ ({ \
50502		uint64x1_t __ret_387; \
50503		uint64x1_t __s0_387 = __p0_387; \
50504		uint64x1_t __s2_387 = __p2_387; \
50505		__ret_387 = vset_lane_u64(vget_lane_u64(__s2_387, __p3_387), __s0_387, __p1_387); \
50506		__ret_387; \
50507		})
50508	51880	#ifdef __LITTLE_ENDIAN__
50509		#define vcopy_lane_u16(__p0_388, __p1_388, __p2_388, __p3_388) __extension__ ({ \
50510		uint16x4_t __ret_388; \
50511		uint16x4_t __s0_388 = __p0_388; \
50512		uint16x4_t __s2_388 = __p2_388; \
50513		__ret_388 = vset_lane_u16(vget_lane_u16(__s2_388, __p3_388), __s0_388, __p1_388); \
50514		__ret_388; \
	51881	#define vmulx_laneq_f32(__p0_605, __p1_605, __p2_605) __extension__ ({ \
	51882	float32x2_t __ret_605; \
	51883	float32x2_t __s0_605 = __p0_605; \
	51884	float32x4_t __s1_605 = __p1_605; \
	51885	__ret_605 = vmulx_f32(__s0_605, splat_laneq_f32(__s1_605, __p2_605)); \
	51886	__ret_605; \
50515	51887	})
50516	51888	#else
50517		#define vcopy_lane_u16(__p0_389, __p1_389, __p2_389, __p3_389) __extension__ ({ \
50518		uint16x4_t __ret_389; \
50519		uint16x4_t __s0_389 = __p0_389; \
50520		uint16x4_t __s2_389 = __p2_389; \
50521		uint16x4_t __rev0_389; __rev0_389 = __builtin_shufflevector(__s0_389, __s0_389, 3, 2, 1, 0); \
50522		uint16x4_t __rev2_389; __rev2_389 = __builtin_shufflevector(__s2_389, __s2_389, 3, 2, 1, 0); \
50523		__ret_389 = __noswap_vset_lane_u16(__noswap_vget_lane_u16(__rev2_389, __p3_389), __rev0_389, __p1_389); \
50524		__ret_389 = __builtin_shufflevector(__ret_389, __ret_389, 3, 2, 1, 0); \
50525		__ret_389; \
	51889	#define vmulx_laneq_f32(__p0_606, __p1_606, __p2_606) __extension__ ({ \
	51890	float32x2_t __ret_606; \
	51891	float32x2_t __s0_606 = __p0_606; \
	51892	float32x4_t __s1_606 = __p1_606; \
	51893	float32x2_t __rev0_606; __rev0_606 = __builtin_shufflevector(__s0_606, __s0_606, 1, 0); \
	51894	float32x4_t __rev1_606; __rev1_606 = __builtin_shufflevector(__s1_606, __s1_606, 3, 2, 1, 0); \
	51895	__ret_606 = __noswap_vmulx_f32(__rev0_606, __noswap_splat_laneq_f32(__rev1_606, __p2_606)); \
	51896	__ret_606 = __builtin_shufflevector(__ret_606, __ret_606, 1, 0); \
	51897	__ret_606; \
50526	51898	})
50527	51899	#endif
50528	51900
50529	51901	#ifdef __LITTLE_ENDIAN__
50530		#define vcopy_lane_s8(__p0_390, __p1_390, __p2_390, __p3_390) __extension__ ({ \
50531		int8x8_t __ret_390; \
50532		int8x8_t __s0_390 = __p0_390; \
50533		int8x8_t __s2_390 = __p2_390; \
50534		__ret_390 = vset_lane_s8(vget_lane_s8(__s2_390, __p3_390), __s0_390, __p1_390); \
50535		__ret_390; \
50536		})
	51902	__ai float64x2_t vnegq_f64(float64x2_t __p0) {
	51903	float64x2_t __ret;
	51904	__ret = -__p0;
	51905	return __ret;
	51906	}
50537	51907	#else
50538		#define vcopy_lane_s8(__p0_391, __p1_391, __p2_391, __p3_391) __extension__ ({ \
50539		int8x8_t __ret_391; \
50540		int8x8_t __s0_391 = __p0_391; \
50541		int8x8_t __s2_391 = __p2_391; \
50542		int8x8_t __rev0_391; __rev0_391 = __builtin_shufflevector(__s0_391, __s0_391, 7, 6, 5, 4, 3, 2, 1, 0); \
50543		int8x8_t __rev2_391; __rev2_391 = __builtin_shufflevector(__s2_391, __s2_391, 7, 6, 5, 4, 3, 2, 1, 0); \
50544		__ret_391 = __noswap_vset_lane_s8(__noswap_vget_lane_s8(__rev2_391, __p3_391), __rev0_391, __p1_391); \
50545		__ret_391 = __builtin_shufflevector(__ret_391, __ret_391, 7, 6, 5, 4, 3, 2, 1, 0); \
50546		__ret_391; \
50547		})
	51908	__ai float64x2_t vnegq_f64(float64x2_t __p0) {
	51909	float64x2_t __ret;
	51910	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	51911	__ret = -__rev0;
	51912	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	51913	return __ret;
	51914	}
50548	51915	#endif
50549	51916
50550	51917	#ifdef __LITTLE_ENDIAN__
50551		#define vcopy_lane_f32(__p0_392, __p1_392, __p2_392, __p3_392) __extension__ ({ \
50552		float32x2_t __ret_392; \
50553		float32x2_t __s0_392 = __p0_392; \
50554		float32x2_t __s2_392 = __p2_392; \
50555		__ret_392 = vset_lane_f32(vget_lane_f32(__s2_392, __p3_392), __s0_392, __p1_392); \
50556		__ret_392; \
50557		})
	51918	__ai int64x2_t vnegq_s64(int64x2_t __p0) {
	51919	int64x2_t __ret;
	51920	__ret = -__p0;
	51921	return __ret;
	51922	}
50558	51923	#else
50559		#define vcopy_lane_f32(__p0_393, __p1_393, __p2_393, __p3_393) __extension__ ({ \
50560		float32x2_t __ret_393; \
50561		float32x2_t __s0_393 = __p0_393; \
50562		float32x2_t __s2_393 = __p2_393; \
50563		float32x2_t __rev0_393; __rev0_393 = __builtin_shufflevector(__s0_393, __s0_393, 1, 0); \
50564		float32x2_t __rev2_393; __rev2_393 = __builtin_shufflevector(__s2_393, __s2_393, 1, 0); \
50565		__ret_393 = __noswap_vset_lane_f32(__noswap_vget_lane_f32(__rev2_393, __p3_393), __rev0_393, __p1_393); \
50566		__ret_393 = __builtin_shufflevector(__ret_393, __ret_393, 1, 0); \
50567		__ret_393; \
50568		})
	51924	__ai int64x2_t vnegq_s64(int64x2_t __p0) {
	51925	int64x2_t __ret;
	51926	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	51927	__ret = -__rev0;
	51928	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	51929	return __ret;
	51930	}
	51931	#endif
	51932
	51933	__ai float64x1_t vneg_f64(float64x1_t __p0) {
	51934	float64x1_t __ret;
	51935	__ret = -__p0;
	51936	return __ret;
	51937	}
	51938	__ai int64x1_t vneg_s64(int64x1_t __p0) {
	51939	int64x1_t __ret;
	51940	__ret = -__p0;
	51941	return __ret;
	51942	}
	51943	__ai int64_t vnegd_s64(int64_t __p0) {
	51944	int64_t __ret;
	51945	__ret = (int64_t) __builtin_neon_vnegd_s64(__p0);
	51946	return __ret;
	51947	}
	51948	#ifdef __LITTLE_ENDIAN__
	51949	__ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	51950	uint8x16_t __ret;
	51951	__ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	51952	return __ret;
	51953	}
	51954	#else
	51955	__ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	51956	uint8x16_t __ret;
	51957	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51958	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51959	__ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	51960	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	51961	return __ret;
	51962	}
50569	51963	#endif
50570	51964
50571	51965	#ifdef __LITTLE_ENDIAN__
50572		#define vcopy_lane_s32(__p0_394, __p1_394, __p2_394, __p3_394) __extension__ ({ \
50573		int32x2_t __ret_394; \
50574		int32x2_t __s0_394 = __p0_394; \
50575		int32x2_t __s2_394 = __p2_394; \
50576		__ret_394 = vset_lane_s32(vget_lane_s32(__s2_394, __p3_394), __s0_394, __p1_394); \
50577		__ret_394; \
50578		})
	51966	__ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	51967	uint32x4_t __ret;
	51968	__ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	51969	return __ret;
	51970	}
50579	51971	#else
50580		#define vcopy_lane_s32(__p0_395, __p1_395, __p2_395, __p3_395) __extension__ ({ \
50581		int32x2_t __ret_395; \
50582		int32x2_t __s0_395 = __p0_395; \
50583		int32x2_t __s2_395 = __p2_395; \
50584		int32x2_t __rev0_395; __rev0_395 = __builtin_shufflevector(__s0_395, __s0_395, 1, 0); \
50585		int32x2_t __rev2_395; __rev2_395 = __builtin_shufflevector(__s2_395, __s2_395, 1, 0); \
50586		__ret_395 = __noswap_vset_lane_s32(__noswap_vget_lane_s32(__rev2_395, __p3_395), __rev0_395, __p1_395); \
50587		__ret_395 = __builtin_shufflevector(__ret_395, __ret_395, 1, 0); \
50588		__ret_395; \
50589		})
	51972	__ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	51973	uint32x4_t __ret;
	51974	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	51975	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	51976	__ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	51977	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	51978	return __ret;
	51979	}
50590	51980	#endif
50591	51981
50592		#define vcopy_lane_s64(__p0_396, __p1_396, __p2_396, __p3_396) __extension__ ({ \
50593		int64x1_t __ret_396; \
50594		int64x1_t __s0_396 = __p0_396; \
50595		int64x1_t __s2_396 = __p2_396; \
50596		__ret_396 = vset_lane_s64(vget_lane_s64(__s2_396, __p3_396), __s0_396, __p1_396); \
50597		__ret_396; \
50598		})
50599	51982	#ifdef __LITTLE_ENDIAN__
50600		#define vcopy_lane_s16(__p0_397, __p1_397, __p2_397, __p3_397) __extension__ ({ \
50601		int16x4_t __ret_397; \
50602		int16x4_t __s0_397 = __p0_397; \
50603		int16x4_t __s2_397 = __p2_397; \
50604		__ret_397 = vset_lane_s16(vget_lane_s16(__s2_397, __p3_397), __s0_397, __p1_397); \
50605		__ret_397; \
50606		})
	51983	__ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	51984	uint64x2_t __ret;
	51985	__ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	51986	return __ret;
	51987	}
50607	51988	#else
50608		#define vcopy_lane_s16(__p0_398, __p1_398, __p2_398, __p3_398) __extension__ ({ \
50609		int16x4_t __ret_398; \
50610		int16x4_t __s0_398 = __p0_398; \
50611		int16x4_t __s2_398 = __p2_398; \
50612		int16x4_t __rev0_398; __rev0_398 = __builtin_shufflevector(__s0_398, __s0_398, 3, 2, 1, 0); \
50613		int16x4_t __rev2_398; __rev2_398 = __builtin_shufflevector(__s2_398, __s2_398, 3, 2, 1, 0); \
50614		__ret_398 = __noswap_vset_lane_s16(__noswap_vget_lane_s16(__rev2_398, __p3_398), __rev0_398, __p1_398); \
50615		__ret_398 = __builtin_shufflevector(__ret_398, __ret_398, 3, 2, 1, 0); \
50616		__ret_398; \
50617		})
	51989	__ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	51990	uint64x2_t __ret;
	51991	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	51992	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	51993	__ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	51994	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	51995	return __ret;
	51996	}
50618	51997	#endif
50619	51998
50620	51999	#ifdef __LITTLE_ENDIAN__
50621		#define vcopyq_laneq_p8(__p0_399, __p1_399, __p2_399, __p3_399) __extension__ ({ \
50622		poly8x16_t __ret_399; \
50623		poly8x16_t __s0_399 = __p0_399; \
50624		poly8x16_t __s2_399 = __p2_399; \
50625		__ret_399 = vsetq_lane_p8(vgetq_lane_p8(__s2_399, __p3_399), __s0_399, __p1_399); \
50626		__ret_399; \
50627		})
	52000	__ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
	52001	uint16x8_t __ret;
	52002	__ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	52003	return __ret;
	52004	}
50628	52005	#else
50629		#define vcopyq_laneq_p8(__p0_400, __p1_400, __p2_400, __p3_400) __extension__ ({ \
50630		poly8x16_t __ret_400; \
50631		poly8x16_t __s0_400 = __p0_400; \
50632		poly8x16_t __s2_400 = __p2_400; \
50633		poly8x16_t __rev0_400; __rev0_400 = __builtin_shufflevector(__s0_400, __s0_400, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50634		poly8x16_t __rev2_400; __rev2_400 = __builtin_shufflevector(__s2_400, __s2_400, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50635		__ret_400 = __noswap_vsetq_lane_p8(__noswap_vgetq_lane_p8(__rev2_400, __p3_400), __rev0_400, __p1_400); \
50636		__ret_400 = __builtin_shufflevector(__ret_400, __ret_400, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50637		__ret_400; \
50638		})
	52006	__ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
	52007	uint16x8_t __ret;
	52008	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	52009	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52010	__ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	52011	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	52012	return __ret;
	52013	}
50639	52014	#endif
50640	52015
50641	52016	#ifdef __LITTLE_ENDIAN__
50642		#define vcopyq_laneq_p16(__p0_401, __p1_401, __p2_401, __p3_401) __extension__ ({ \
50643		poly16x8_t __ret_401; \
50644		poly16x8_t __s0_401 = __p0_401; \
50645		poly16x8_t __s2_401 = __p2_401; \
50646		__ret_401 = vsetq_lane_p16(vgetq_lane_p16(__s2_401, __p3_401), __s0_401, __p1_401); \
50647		__ret_401; \
50648		})
	52017	__ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
	52018	int8x16_t __ret;
	52019	__ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	52020	return __ret;
	52021	}
50649	52022	#else
50650		#define vcopyq_laneq_p16(__p0_402, __p1_402, __p2_402, __p3_402) __extension__ ({ \
50651		poly16x8_t __ret_402; \
50652		poly16x8_t __s0_402 = __p0_402; \
50653		poly16x8_t __s2_402 = __p2_402; \
50654		poly16x8_t __rev0_402; __rev0_402 = __builtin_shufflevector(__s0_402, __s0_402, 7, 6, 5, 4, 3, 2, 1, 0); \
50655		poly16x8_t __rev2_402; __rev2_402 = __builtin_shufflevector(__s2_402, __s2_402, 7, 6, 5, 4, 3, 2, 1, 0); \
50656		__ret_402 = __noswap_vsetq_lane_p16(__noswap_vgetq_lane_p16(__rev2_402, __p3_402), __rev0_402, __p1_402); \
50657		__ret_402 = __builtin_shufflevector(__ret_402, __ret_402, 7, 6, 5, 4, 3, 2, 1, 0); \
50658		__ret_402; \
50659		})
	52023	__ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
	52024	int8x16_t __ret;
	52025	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52026	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52027	__ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
	52028	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52029	return __ret;
	52030	}
50660	52031	#endif
50661	52032
50662	52033	#ifdef __LITTLE_ENDIAN__
50663		#define vcopyq_laneq_u8(__p0_403, __p1_403, __p2_403, __p3_403) __extension__ ({ \
50664		uint8x16_t __ret_403; \
50665		uint8x16_t __s0_403 = __p0_403; \
50666		uint8x16_t __s2_403 = __p2_403; \
50667		__ret_403 = vsetq_lane_u8(vgetq_lane_u8(__s2_403, __p3_403), __s0_403, __p1_403); \
50668		__ret_403; \
50669		})
	52034	__ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
	52035	float64x2_t __ret;
	52036	__ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	52037	return __ret;
	52038	}
50670	52039	#else
50671		#define vcopyq_laneq_u8(__p0_404, __p1_404, __p2_404, __p3_404) __extension__ ({ \
50672		uint8x16_t __ret_404; \
50673		uint8x16_t __s0_404 = __p0_404; \
50674		uint8x16_t __s2_404 = __p2_404; \
50675		uint8x16_t __rev0_404; __rev0_404 = __builtin_shufflevector(__s0_404, __s0_404, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50676		uint8x16_t __rev2_404; __rev2_404 = __builtin_shufflevector(__s2_404, __s2_404, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50677		__ret_404 = __noswap_vsetq_lane_u8(__noswap_vgetq_lane_u8(__rev2_404, __p3_404), __rev0_404, __p1_404); \
50678		__ret_404 = __builtin_shufflevector(__ret_404, __ret_404, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50679		__ret_404; \
50680		})
	52040	__ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
	52041	float64x2_t __ret;
	52042	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52043	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52044	__ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	52045	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52046	return __ret;
	52047	}
50681	52048	#endif
50682	52049
50683	52050	#ifdef __LITTLE_ENDIAN__
50684		#define vcopyq_laneq_u32(__p0_405, __p1_405, __p2_405, __p3_405) __extension__ ({ \
50685		uint32x4_t __ret_405; \
50686		uint32x4_t __s0_405 = __p0_405; \
50687		uint32x4_t __s2_405 = __p2_405; \
50688		__ret_405 = vsetq_lane_u32(vgetq_lane_u32(__s2_405, __p3_405), __s0_405, __p1_405); \
50689		__ret_405; \
50690		})
	52051	__ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
	52052	float32x4_t __ret;
	52053	__ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	52054	return __ret;
	52055	}
50691	52056	#else
50692		#define vcopyq_laneq_u32(__p0_406, __p1_406, __p2_406, __p3_406) __extension__ ({ \
50693		uint32x4_t __ret_406; \
50694		uint32x4_t __s0_406 = __p0_406; \
50695		uint32x4_t __s2_406 = __p2_406; \
50696		uint32x4_t __rev0_406; __rev0_406 = __builtin_shufflevector(__s0_406, __s0_406, 3, 2, 1, 0); \
50697		uint32x4_t __rev2_406; __rev2_406 = __builtin_shufflevector(__s2_406, __s2_406, 3, 2, 1, 0); \
50698		__ret_406 = __noswap_vsetq_lane_u32(__noswap_vgetq_lane_u32(__rev2_406, __p3_406), __rev0_406, __p1_406); \
50699		__ret_406 = __builtin_shufflevector(__ret_406, __ret_406, 3, 2, 1, 0); \
50700		__ret_406; \
50701		})
	52057	__ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
	52058	float32x4_t __ret;
	52059	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52060	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52061	__ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	52062	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	52063	return __ret;
	52064	}
50702	52065	#endif
50703	52066
50704	52067	#ifdef __LITTLE_ENDIAN__
50705		#define vcopyq_laneq_u64(__p0_407, __p1_407, __p2_407, __p3_407) __extension__ ({ \
50706		uint64x2_t __ret_407; \
50707		uint64x2_t __s0_407 = __p0_407; \
50708		uint64x2_t __s2_407 = __p2_407; \
50709		__ret_407 = vsetq_lane_u64(vgetq_lane_u64(__s2_407, __p3_407), __s0_407, __p1_407); \
50710		__ret_407; \
50711		})
	52068	__ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
	52069	int32x4_t __ret;
	52070	__ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
	52071	return __ret;
	52072	}
50712	52073	#else
50713		#define vcopyq_laneq_u64(__p0_408, __p1_408, __p2_408, __p3_408) __extension__ ({ \
50714		uint64x2_t __ret_408; \
50715		uint64x2_t __s0_408 = __p0_408; \
50716		uint64x2_t __s2_408 = __p2_408; \
50717		uint64x2_t __rev0_408; __rev0_408 = __builtin_shufflevector(__s0_408, __s0_408, 1, 0); \
50718		uint64x2_t __rev2_408; __rev2_408 = __builtin_shufflevector(__s2_408, __s2_408, 1, 0); \
50719		__ret_408 = __noswap_vsetq_lane_u64(__noswap_vgetq_lane_u64(__rev2_408, __p3_408), __rev0_408, __p1_408); \
50720		__ret_408 = __builtin_shufflevector(__ret_408, __ret_408, 1, 0); \
50721		__ret_408; \
50722		})
	52074	__ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
	52075	int32x4_t __ret;
	52076	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52077	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52078	__ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
	52079	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	52080	return __ret;
	52081	}
50723	52082	#endif
50724	52083
50725	52084	#ifdef __LITTLE_ENDIAN__
50726		#define vcopyq_laneq_u16(__p0_409, __p1_409, __p2_409, __p3_409) __extension__ ({ \
50727		uint16x8_t __ret_409; \
50728		uint16x8_t __s0_409 = __p0_409; \
50729		uint16x8_t __s2_409 = __p2_409; \
50730		__ret_409 = vsetq_lane_u16(vgetq_lane_u16(__s2_409, __p3_409), __s0_409, __p1_409); \
50731		__ret_409; \
50732		})
	52085	__ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
	52086	int64x2_t __ret;
	52087	__ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
	52088	return __ret;
	52089	}
50733	52090	#else
50734		#define vcopyq_laneq_u16(__p0_410, __p1_410, __p2_410, __p3_410) __extension__ ({ \
50735		uint16x8_t __ret_410; \
50736		uint16x8_t __s0_410 = __p0_410; \
50737		uint16x8_t __s2_410 = __p2_410; \
50738		uint16x8_t __rev0_410; __rev0_410 = __builtin_shufflevector(__s0_410, __s0_410, 7, 6, 5, 4, 3, 2, 1, 0); \
50739		uint16x8_t __rev2_410; __rev2_410 = __builtin_shufflevector(__s2_410, __s2_410, 7, 6, 5, 4, 3, 2, 1, 0); \
50740		__ret_410 = __noswap_vsetq_lane_u16(__noswap_vgetq_lane_u16(__rev2_410, __p3_410), __rev0_410, __p1_410); \
50741		__ret_410 = __builtin_shufflevector(__ret_410, __ret_410, 7, 6, 5, 4, 3, 2, 1, 0); \
50742		__ret_410; \
50743		})
	52091	__ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
	52092	int64x2_t __ret;
	52093	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52094	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52095	__ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
	52096	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52097	return __ret;
	52098	}
50744	52099	#endif
50745	52100
50746	52101	#ifdef __LITTLE_ENDIAN__
50747		#define vcopyq_laneq_s8(__p0_411, __p1_411, __p2_411, __p3_411) __extension__ ({ \
50748		int8x16_t __ret_411; \
50749		int8x16_t __s0_411 = __p0_411; \
50750		int8x16_t __s2_411 = __p2_411; \
50751		__ret_411 = vsetq_lane_s8(vgetq_lane_s8(__s2_411, __p3_411), __s0_411, __p1_411); \
50752		__ret_411; \
50753		})
	52102	__ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
	52103	int16x8_t __ret;
	52104	__ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
	52105	return __ret;
	52106	}
50754	52107	#else
50755		#define vcopyq_laneq_s8(__p0_412, __p1_412, __p2_412, __p3_412) __extension__ ({ \
50756		int8x16_t __ret_412; \
50757		int8x16_t __s0_412 = __p0_412; \
50758		int8x16_t __s2_412 = __p2_412; \
50759		int8x16_t __rev0_412; __rev0_412 = __builtin_shufflevector(__s0_412, __s0_412, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50760		int8x16_t __rev2_412; __rev2_412 = __builtin_shufflevector(__s2_412, __s2_412, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50761		__ret_412 = __noswap_vsetq_lane_s8(__noswap_vgetq_lane_s8(__rev2_412, __p3_412), __rev0_412, __p1_412); \
50762		__ret_412 = __builtin_shufflevector(__ret_412, __ret_412, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50763		__ret_412; \
50764		})
	52108	__ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
	52109	int16x8_t __ret;
	52110	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	52111	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52112	__ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
	52113	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	52114	return __ret;
	52115	}
50765	52116	#endif
50766	52117
50767	52118	#ifdef __LITTLE_ENDIAN__
50768		#define vcopyq_laneq_f32(__p0_413, __p1_413, __p2_413, __p3_413) __extension__ ({ \
50769		float32x4_t __ret_413; \
50770		float32x4_t __s0_413 = __p0_413; \
50771		float32x4_t __s2_413 = __p2_413; \
50772		__ret_413 = vsetq_lane_f32(vgetq_lane_f32(__s2_413, __p3_413), __s0_413, __p1_413); \
50773		__ret_413; \
50774		})
	52119	__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
	52120	uint64_t __ret;
	52121	__ret = (uint64_t) __builtin_neon_vpaddd_u64(__p0);
	52122	return __ret;
	52123	}
50775	52124	#else
50776		#define vcopyq_laneq_f32(__p0_414, __p1_414, __p2_414, __p3_414) __extension__ ({ \
50777		float32x4_t __ret_414; \
50778		float32x4_t __s0_414 = __p0_414; \
50779		float32x4_t __s2_414 = __p2_414; \
50780		float32x4_t __rev0_414; __rev0_414 = __builtin_shufflevector(__s0_414, __s0_414, 3, 2, 1, 0); \
50781		float32x4_t __rev2_414; __rev2_414 = __builtin_shufflevector(__s2_414, __s2_414, 3, 2, 1, 0); \
50782		__ret_414 = __noswap_vsetq_lane_f32(__noswap_vgetq_lane_f32(__rev2_414, __p3_414), __rev0_414, __p1_414); \
50783		__ret_414 = __builtin_shufflevector(__ret_414, __ret_414, 3, 2, 1, 0); \
50784		__ret_414; \
50785		})
	52125	__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
	52126	uint64_t __ret;
	52127	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52128	__ret = (uint64_t) __builtin_neon_vpaddd_u64(__rev0);
	52129	return __ret;
	52130	}
50786	52131	#endif
50787	52132
50788	52133	#ifdef __LITTLE_ENDIAN__
50789		#define vcopyq_laneq_s32(__p0_415, __p1_415, __p2_415, __p3_415) __extension__ ({ \
50790		int32x4_t __ret_415; \
50791		int32x4_t __s0_415 = __p0_415; \
50792		int32x4_t __s2_415 = __p2_415; \
50793		__ret_415 = vsetq_lane_s32(vgetq_lane_s32(__s2_415, __p3_415), __s0_415, __p1_415); \
50794		__ret_415; \
50795		})
	52134	__ai float64_t vpaddd_f64(float64x2_t __p0) {
	52135	float64_t __ret;
	52136	__ret = (float64_t) __builtin_neon_vpaddd_f64(__p0);
	52137	return __ret;
	52138	}
50796	52139	#else
50797		#define vcopyq_laneq_s32(__p0_416, __p1_416, __p2_416, __p3_416) __extension__ ({ \
50798		int32x4_t __ret_416; \
50799		int32x4_t __s0_416 = __p0_416; \
50800		int32x4_t __s2_416 = __p2_416; \
50801		int32x4_t __rev0_416; __rev0_416 = __builtin_shufflevector(__s0_416, __s0_416, 3, 2, 1, 0); \
50802		int32x4_t __rev2_416; __rev2_416 = __builtin_shufflevector(__s2_416, __s2_416, 3, 2, 1, 0); \
50803		__ret_416 = __noswap_vsetq_lane_s32(__noswap_vgetq_lane_s32(__rev2_416, __p3_416), __rev0_416, __p1_416); \
50804		__ret_416 = __builtin_shufflevector(__ret_416, __ret_416, 3, 2, 1, 0); \
50805		__ret_416; \
50806		})
	52140	__ai float64_t vpaddd_f64(float64x2_t __p0) {
	52141	float64_t __ret;
	52142	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52143	__ret = (float64_t) __builtin_neon_vpaddd_f64(__rev0);
	52144	return __ret;
	52145	}
50807	52146	#endif
50808	52147
50809	52148	#ifdef __LITTLE_ENDIAN__
50810		#define vcopyq_laneq_s64(__p0_417, __p1_417, __p2_417, __p3_417) __extension__ ({ \
50811		int64x2_t __ret_417; \
50812		int64x2_t __s0_417 = __p0_417; \
50813		int64x2_t __s2_417 = __p2_417; \
50814		__ret_417 = vsetq_lane_s64(vgetq_lane_s64(__s2_417, __p3_417), __s0_417, __p1_417); \
50815		__ret_417; \
50816		})
	52149	__ai int64_t vpaddd_s64(int64x2_t __p0) {
	52150	int64_t __ret;
	52151	__ret = (int64_t) __builtin_neon_vpaddd_s64(__p0);
	52152	return __ret;
	52153	}
50817	52154	#else
50818		#define vcopyq_laneq_s64(__p0_418, __p1_418, __p2_418, __p3_418) __extension__ ({ \
50819		int64x2_t __ret_418; \
50820		int64x2_t __s0_418 = __p0_418; \
50821		int64x2_t __s2_418 = __p2_418; \
50822		int64x2_t __rev0_418; __rev0_418 = __builtin_shufflevector(__s0_418, __s0_418, 1, 0); \
50823		int64x2_t __rev2_418; __rev2_418 = __builtin_shufflevector(__s2_418, __s2_418, 1, 0); \
50824		__ret_418 = __noswap_vsetq_lane_s64(__noswap_vgetq_lane_s64(__rev2_418, __p3_418), __rev0_418, __p1_418); \
50825		__ret_418 = __builtin_shufflevector(__ret_418, __ret_418, 1, 0); \
50826		__ret_418; \
50827		})
	52155	__ai int64_t vpaddd_s64(int64x2_t __p0) {
	52156	int64_t __ret;
	52157	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52158	__ret = (int64_t) __builtin_neon_vpaddd_s64(__rev0);
	52159	return __ret;
	52160	}
50828	52161	#endif
50829	52162
50830	52163	#ifdef __LITTLE_ENDIAN__
50831		#define vcopyq_laneq_s16(__p0_419, __p1_419, __p2_419, __p3_419) __extension__ ({ \
50832		int16x8_t __ret_419; \
50833		int16x8_t __s0_419 = __p0_419; \
50834		int16x8_t __s2_419 = __p2_419; \
50835		__ret_419 = vsetq_lane_s16(vgetq_lane_s16(__s2_419, __p3_419), __s0_419, __p1_419); \
50836		__ret_419; \
50837		})
	52164	__ai float32_t vpadds_f32(float32x2_t __p0) {
	52165	float32_t __ret;
	52166	__ret = (float32_t) __builtin_neon_vpadds_f32(__p0);
	52167	return __ret;
	52168	}
50838	52169	#else
50839		#define vcopyq_laneq_s16(__p0_420, __p1_420, __p2_420, __p3_420) __extension__ ({ \
50840		int16x8_t __ret_420; \
50841		int16x8_t __s0_420 = __p0_420; \
50842		int16x8_t __s2_420 = __p2_420; \
50843		int16x8_t __rev0_420; __rev0_420 = __builtin_shufflevector(__s0_420, __s0_420, 7, 6, 5, 4, 3, 2, 1, 0); \
50844		int16x8_t __rev2_420; __rev2_420 = __builtin_shufflevector(__s2_420, __s2_420, 7, 6, 5, 4, 3, 2, 1, 0); \
50845		__ret_420 = __noswap_vsetq_lane_s16(__noswap_vgetq_lane_s16(__rev2_420, __p3_420), __rev0_420, __p1_420); \
50846		__ret_420 = __builtin_shufflevector(__ret_420, __ret_420, 7, 6, 5, 4, 3, 2, 1, 0); \
50847		__ret_420; \
50848		})
	52170	__ai float32_t vpadds_f32(float32x2_t __p0) {
	52171	float32_t __ret;
	52172	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52173	__ret = (float32_t) __builtin_neon_vpadds_f32(__rev0);
	52174	return __ret;
	52175	}
50849	52176	#endif
50850	52177
50851	52178	#ifdef __LITTLE_ENDIAN__
50852		#define vcopy_laneq_p8(__p0_421, __p1_421, __p2_421, __p3_421) __extension__ ({ \
50853		poly8x8_t __ret_421; \
50854		poly8x8_t __s0_421 = __p0_421; \
50855		poly8x16_t __s2_421 = __p2_421; \
50856		__ret_421 = vset_lane_p8(vgetq_lane_p8(__s2_421, __p3_421), __s0_421, __p1_421); \
50857		__ret_421; \
50858		})
	52179	__ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	52180	uint8x16_t __ret;
	52181	__ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	52182	return __ret;
	52183	}
50859	52184	#else
50860		#define vcopy_laneq_p8(__p0_422, __p1_422, __p2_422, __p3_422) __extension__ ({ \
50861		poly8x8_t __ret_422; \
50862		poly8x8_t __s0_422 = __p0_422; \
50863		poly8x16_t __s2_422 = __p2_422; \
50864		poly8x8_t __rev0_422; __rev0_422 = __builtin_shufflevector(__s0_422, __s0_422, 7, 6, 5, 4, 3, 2, 1, 0); \
50865		poly8x16_t __rev2_422; __rev2_422 = __builtin_shufflevector(__s2_422, __s2_422, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50866		__ret_422 = __noswap_vset_lane_p8(__noswap_vgetq_lane_p8(__rev2_422, __p3_422), __rev0_422, __p1_422); \
50867		__ret_422 = __builtin_shufflevector(__ret_422, __ret_422, 7, 6, 5, 4, 3, 2, 1, 0); \
50868		__ret_422; \
50869		})
	52185	__ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	52186	uint8x16_t __ret;
	52187	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52188	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52189	__ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	52190	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52191	return __ret;
	52192	}
50870	52193	#endif
50871	52194
50872	52195	#ifdef __LITTLE_ENDIAN__
50873		#define vcopy_laneq_p16(__p0_423, __p1_423, __p2_423, __p3_423) __extension__ ({ \
50874		poly16x4_t __ret_423; \
50875		poly16x4_t __s0_423 = __p0_423; \
50876		poly16x8_t __s2_423 = __p2_423; \
50877		__ret_423 = vset_lane_p16(vgetq_lane_p16(__s2_423, __p3_423), __s0_423, __p1_423); \
50878		__ret_423; \
50879		})
	52196	__ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	52197	uint32x4_t __ret;
	52198	__ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	52199	return __ret;
	52200	}
50880	52201	#else
50881		#define vcopy_laneq_p16(__p0_424, __p1_424, __p2_424, __p3_424) __extension__ ({ \
50882		poly16x4_t __ret_424; \
50883		poly16x4_t __s0_424 = __p0_424; \
50884		poly16x8_t __s2_424 = __p2_424; \
50885		poly16x4_t __rev0_424; __rev0_424 = __builtin_shufflevector(__s0_424, __s0_424, 3, 2, 1, 0); \
50886		poly16x8_t __rev2_424; __rev2_424 = __builtin_shufflevector(__s2_424, __s2_424, 7, 6, 5, 4, 3, 2, 1, 0); \
50887		__ret_424 = __noswap_vset_lane_p16(__noswap_vgetq_lane_p16(__rev2_424, __p3_424), __rev0_424, __p1_424); \
50888		__ret_424 = __builtin_shufflevector(__ret_424, __ret_424, 3, 2, 1, 0); \
50889		__ret_424; \
50890		})
	52202	__ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	52203	uint32x4_t __ret;
	52204	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52205	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52206	__ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	52207	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	52208	return __ret;
	52209	}
50891	52210	#endif
50892	52211
50893	52212	#ifdef __LITTLE_ENDIAN__
50894		#define vcopy_laneq_u8(__p0_425, __p1_425, __p2_425, __p3_425) __extension__ ({ \
50895		uint8x8_t __ret_425; \
50896		uint8x8_t __s0_425 = __p0_425; \
50897		uint8x16_t __s2_425 = __p2_425; \
50898		__ret_425 = vset_lane_u8(vgetq_lane_u8(__s2_425, __p3_425), __s0_425, __p1_425); \
50899		__ret_425; \
50900		})
	52213	__ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
	52214	uint16x8_t __ret;
	52215	__ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	52216	return __ret;
	52217	}
50901	52218	#else
50902		#define vcopy_laneq_u8(__p0_426, __p1_426, __p2_426, __p3_426) __extension__ ({ \
50903		uint8x8_t __ret_426; \
50904		uint8x8_t __s0_426 = __p0_426; \
50905		uint8x16_t __s2_426 = __p2_426; \
50906		uint8x8_t __rev0_426; __rev0_426 = __builtin_shufflevector(__s0_426, __s0_426, 7, 6, 5, 4, 3, 2, 1, 0); \
50907		uint8x16_t __rev2_426; __rev2_426 = __builtin_shufflevector(__s2_426, __s2_426, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50908		__ret_426 = __noswap_vset_lane_u8(__noswap_vgetq_lane_u8(__rev2_426, __p3_426), __rev0_426, __p1_426); \
50909		__ret_426 = __builtin_shufflevector(__ret_426, __ret_426, 7, 6, 5, 4, 3, 2, 1, 0); \
50910		__ret_426; \
50911		})
	52219	__ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
	52220	uint16x8_t __ret;
	52221	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	52222	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52223	__ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	52224	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	52225	return __ret;
	52226	}
50912	52227	#endif
50913	52228
50914	52229	#ifdef __LITTLE_ENDIAN__
50915		#define vcopy_laneq_u32(__p0_427, __p1_427, __p2_427, __p3_427) __extension__ ({ \
50916		uint32x2_t __ret_427; \
50917		uint32x2_t __s0_427 = __p0_427; \
50918		uint32x4_t __s2_427 = __p2_427; \
50919		__ret_427 = vset_lane_u32(vgetq_lane_u32(__s2_427, __p3_427), __s0_427, __p1_427); \
50920		__ret_427; \
50921		})
	52230	__ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
	52231	int8x16_t __ret;
	52232	__ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	52233	return __ret;
	52234	}
50922	52235	#else
50923		#define vcopy_laneq_u32(__p0_428, __p1_428, __p2_428, __p3_428) __extension__ ({ \
50924		uint32x2_t __ret_428; \
50925		uint32x2_t __s0_428 = __p0_428; \
50926		uint32x4_t __s2_428 = __p2_428; \
50927		uint32x2_t __rev0_428; __rev0_428 = __builtin_shufflevector(__s0_428, __s0_428, 1, 0); \
50928		uint32x4_t __rev2_428; __rev2_428 = __builtin_shufflevector(__s2_428, __s2_428, 3, 2, 1, 0); \
50929		__ret_428 = __noswap_vset_lane_u32(__noswap_vgetq_lane_u32(__rev2_428, __p3_428), __rev0_428, __p1_428); \
50930		__ret_428 = __builtin_shufflevector(__ret_428, __ret_428, 1, 0); \
50931		__ret_428; \
50932		})
	52236	__ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
	52237	int8x16_t __ret;
	52238	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52239	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52240	__ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
	52241	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52242	return __ret;
	52243	}
50933	52244	#endif
50934	52245
50935	52246	#ifdef __LITTLE_ENDIAN__
50936		#define vcopy_laneq_u64(__p0_429, __p1_429, __p2_429, __p3_429) __extension__ ({ \
50937		uint64x1_t __ret_429; \
50938		uint64x1_t __s0_429 = __p0_429; \
50939		uint64x2_t __s2_429 = __p2_429; \
50940		__ret_429 = vset_lane_u64(vgetq_lane_u64(__s2_429, __p3_429), __s0_429, __p1_429); \
50941		__ret_429; \
50942		})
	52247	__ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
	52248	float64x2_t __ret;
	52249	__ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	52250	return __ret;
	52251	}
50943	52252	#else
50944		#define vcopy_laneq_u64(__p0_430, __p1_430, __p2_430, __p3_430) __extension__ ({ \
50945		uint64x1_t __ret_430; \
50946		uint64x1_t __s0_430 = __p0_430; \
50947		uint64x2_t __s2_430 = __p2_430; \
50948		uint64x2_t __rev2_430; __rev2_430 = __builtin_shufflevector(__s2_430, __s2_430, 1, 0); \
50949		__ret_430 = vset_lane_u64(__noswap_vgetq_lane_u64(__rev2_430, __p3_430), __s0_430, __p1_430); \
50950		__ret_430; \
50951		})
	52253	__ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
	52254	float64x2_t __ret;
	52255	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52256	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52257	__ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	52258	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52259	return __ret;
	52260	}
50952	52261	#endif
50953	52262
50954	52263	#ifdef __LITTLE_ENDIAN__
50955		#define vcopy_laneq_u16(__p0_431, __p1_431, __p2_431, __p3_431) __extension__ ({ \
50956		uint16x4_t __ret_431; \
50957		uint16x4_t __s0_431 = __p0_431; \
50958		uint16x8_t __s2_431 = __p2_431; \
50959		__ret_431 = vset_lane_u16(vgetq_lane_u16(__s2_431, __p3_431), __s0_431, __p1_431); \
50960		__ret_431; \
50961		})
	52264	__ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
	52265	float32x4_t __ret;
	52266	__ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	52267	return __ret;
	52268	}
50962	52269	#else
50963		#define vcopy_laneq_u16(__p0_432, __p1_432, __p2_432, __p3_432) __extension__ ({ \
50964		uint16x4_t __ret_432; \
50965		uint16x4_t __s0_432 = __p0_432; \
50966		uint16x8_t __s2_432 = __p2_432; \
50967		uint16x4_t __rev0_432; __rev0_432 = __builtin_shufflevector(__s0_432, __s0_432, 3, 2, 1, 0); \
50968		uint16x8_t __rev2_432; __rev2_432 = __builtin_shufflevector(__s2_432, __s2_432, 7, 6, 5, 4, 3, 2, 1, 0); \
50969		__ret_432 = __noswap_vset_lane_u16(__noswap_vgetq_lane_u16(__rev2_432, __p3_432), __rev0_432, __p1_432); \
50970		__ret_432 = __builtin_shufflevector(__ret_432, __ret_432, 3, 2, 1, 0); \
50971		__ret_432; \
50972		})
	52270	__ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
	52271	float32x4_t __ret;
	52272	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52273	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52274	__ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	52275	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	52276	return __ret;
	52277	}
50973	52278	#endif
50974	52279
50975	52280	#ifdef __LITTLE_ENDIAN__
50976		#define vcopy_laneq_s8(__p0_433, __p1_433, __p2_433, __p3_433) __extension__ ({ \
50977		int8x8_t __ret_433; \
50978		int8x8_t __s0_433 = __p0_433; \
50979		int8x16_t __s2_433 = __p2_433; \
50980		__ret_433 = vset_lane_s8(vgetq_lane_s8(__s2_433, __p3_433), __s0_433, __p1_433); \
50981		__ret_433; \
50982		})
	52281	__ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
	52282	int32x4_t __ret;
	52283	__ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
	52284	return __ret;
	52285	}
50983	52286	#else
50984		#define vcopy_laneq_s8(__p0_434, __p1_434, __p2_434, __p3_434) __extension__ ({ \
50985		int8x8_t __ret_434; \
50986		int8x8_t __s0_434 = __p0_434; \
50987		int8x16_t __s2_434 = __p2_434; \
50988		int8x8_t __rev0_434; __rev0_434 = __builtin_shufflevector(__s0_434, __s0_434, 7, 6, 5, 4, 3, 2, 1, 0); \
50989		int8x16_t __rev2_434; __rev2_434 = __builtin_shufflevector(__s2_434, __s2_434, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
50990		__ret_434 = __noswap_vset_lane_s8(__noswap_vgetq_lane_s8(__rev2_434, __p3_434), __rev0_434, __p1_434); \
50991		__ret_434 = __builtin_shufflevector(__ret_434, __ret_434, 7, 6, 5, 4, 3, 2, 1, 0); \
50992		__ret_434; \
50993		})
	52287	__ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
	52288	int32x4_t __ret;
	52289	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52290	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52291	__ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
	52292	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	52293	return __ret;
	52294	}
50994	52295	#endif
50995	52296
50996	52297	#ifdef __LITTLE_ENDIAN__
50997		#define vcopy_laneq_f32(__p0_435, __p1_435, __p2_435, __p3_435) __extension__ ({ \
50998		float32x2_t __ret_435; \
50999		float32x2_t __s0_435 = __p0_435; \
51000		float32x4_t __s2_435 = __p2_435; \
51001		__ret_435 = vset_lane_f32(vgetq_lane_f32(__s2_435, __p3_435), __s0_435, __p1_435); \
51002		__ret_435; \
51003		})
	52298	__ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
	52299	int16x8_t __ret;
	52300	__ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
	52301	return __ret;
	52302	}
51004	52303	#else
51005		#define vcopy_laneq_f32(__p0_436, __p1_436, __p2_436, __p3_436) __extension__ ({ \
51006		float32x2_t __ret_436; \
51007		float32x2_t __s0_436 = __p0_436; \
51008		float32x4_t __s2_436 = __p2_436; \
51009		float32x2_t __rev0_436; __rev0_436 = __builtin_shufflevector(__s0_436, __s0_436, 1, 0); \
51010		float32x4_t __rev2_436; __rev2_436 = __builtin_shufflevector(__s2_436, __s2_436, 3, 2, 1, 0); \
51011		__ret_436 = __noswap_vset_lane_f32(__noswap_vgetq_lane_f32(__rev2_436, __p3_436), __rev0_436, __p1_436); \
51012		__ret_436 = __builtin_shufflevector(__ret_436, __ret_436, 1, 0); \
51013		__ret_436; \
51014		})
	52304	__ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
	52305	int16x8_t __ret;
	52306	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	52307	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52308	__ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
	52309	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	52310	return __ret;
	52311	}
51015	52312	#endif
51016	52313
51017	52314	#ifdef __LITTLE_ENDIAN__
51018		#define vcopy_laneq_s32(__p0_437, __p1_437, __p2_437, __p3_437) __extension__ ({ \
51019		int32x2_t __ret_437; \
51020		int32x2_t __s0_437 = __p0_437; \
51021		int32x4_t __s2_437 = __p2_437; \
51022		__ret_437 = vset_lane_s32(vgetq_lane_s32(__s2_437, __p3_437), __s0_437, __p1_437); \
51023		__ret_437; \
51024		})
	52315	__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
	52316	float64_t __ret;
	52317	__ret = (float64_t) __builtin_neon_vpmaxqd_f64(__p0);
	52318	return __ret;
	52319	}
51025	52320	#else
51026		#define vcopy_laneq_s32(__p0_438, __p1_438, __p2_438, __p3_438) __extension__ ({ \
51027		int32x2_t __ret_438; \
51028		int32x2_t __s0_438 = __p0_438; \
51029		int32x4_t __s2_438 = __p2_438; \
51030		int32x2_t __rev0_438; __rev0_438 = __builtin_shufflevector(__s0_438, __s0_438, 1, 0); \
51031		int32x4_t __rev2_438; __rev2_438 = __builtin_shufflevector(__s2_438, __s2_438, 3, 2, 1, 0); \
51032		__ret_438 = __noswap_vset_lane_s32(__noswap_vgetq_lane_s32(__rev2_438, __p3_438), __rev0_438, __p1_438); \
51033		__ret_438 = __builtin_shufflevector(__ret_438, __ret_438, 1, 0); \
51034		__ret_438; \
51035		})
	52321	__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
	52322	float64_t __ret;
	52323	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52324	__ret = (float64_t) __builtin_neon_vpmaxqd_f64(__rev0);
	52325	return __ret;
	52326	}
51036	52327	#endif
51037	52328
51038	52329	#ifdef __LITTLE_ENDIAN__
51039		#define vcopy_laneq_s64(__p0_439, __p1_439, __p2_439, __p3_439) __extension__ ({ \
51040		int64x1_t __ret_439; \
51041		int64x1_t __s0_439 = __p0_439; \
51042		int64x2_t __s2_439 = __p2_439; \
51043		__ret_439 = vset_lane_s64(vgetq_lane_s64(__s2_439, __p3_439), __s0_439, __p1_439); \
51044		__ret_439; \
51045		})
	52330	__ai float32_t vpmaxs_f32(float32x2_t __p0) {
	52331	float32_t __ret;
	52332	__ret = (float32_t) __builtin_neon_vpmaxs_f32(__p0);
	52333	return __ret;
	52334	}
51046	52335	#else
51047		#define vcopy_laneq_s64(__p0_440, __p1_440, __p2_440, __p3_440) __extension__ ({ \
51048		int64x1_t __ret_440; \
51049		int64x1_t __s0_440 = __p0_440; \
51050		int64x2_t __s2_440 = __p2_440; \
51051		int64x2_t __rev2_440; __rev2_440 = __builtin_shufflevector(__s2_440, __s2_440, 1, 0); \
51052		__ret_440 = vset_lane_s64(__noswap_vgetq_lane_s64(__rev2_440, __p3_440), __s0_440, __p1_440); \
51053		__ret_440; \
51054		})
	52336	__ai float32_t vpmaxs_f32(float32x2_t __p0) {
	52337	float32_t __ret;
	52338	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52339	__ret = (float32_t) __builtin_neon_vpmaxs_f32(__rev0);
	52340	return __ret;
	52341	}
51055	52342	#endif
51056	52343
51057	52344	#ifdef __LITTLE_ENDIAN__
51058		#define vcopy_laneq_s16(__p0_441, __p1_441, __p2_441, __p3_441) __extension__ ({ \
51059		int16x4_t __ret_441; \
51060		int16x4_t __s0_441 = __p0_441; \
51061		int16x8_t __s2_441 = __p2_441; \
51062		__ret_441 = vset_lane_s16(vgetq_lane_s16(__s2_441, __p3_441), __s0_441, __p1_441); \
51063		__ret_441; \
51064		})
	52345	__ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	52346	float64x2_t __ret;
	52347	__ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	52348	return __ret;
	52349	}
51065	52350	#else
51066		#define vcopy_laneq_s16(__p0_442, __p1_442, __p2_442, __p3_442) __extension__ ({ \
51067		int16x4_t __ret_442; \
51068		int16x4_t __s0_442 = __p0_442; \
51069		int16x8_t __s2_442 = __p2_442; \
51070		int16x4_t __rev0_442; __rev0_442 = __builtin_shufflevector(__s0_442, __s0_442, 3, 2, 1, 0); \
51071		int16x8_t __rev2_442; __rev2_442 = __builtin_shufflevector(__s2_442, __s2_442, 7, 6, 5, 4, 3, 2, 1, 0); \
51072		__ret_442 = __noswap_vset_lane_s16(__noswap_vgetq_lane_s16(__rev2_442, __p3_442), __rev0_442, __p1_442); \
51073		__ret_442 = __builtin_shufflevector(__ret_442, __ret_442, 3, 2, 1, 0); \
51074		__ret_442; \
51075		})
	52351	__ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	52352	float64x2_t __ret;
	52353	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52354	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52355	__ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	52356	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52357	return __ret;
	52358	}
51076	52359	#endif
51077	52360
51078		#define vcreate_p64(__p0) __extension__ ({ \
51079		poly64x1_t __ret; \
51080		uint64_t __promote = __p0; \
51081		__ret = (poly64x1_t)(__promote); \
51082		__ret; \
51083		})
51084		#define vcreate_f64(__p0) __extension__ ({ \
51085		float64x1_t __ret; \
51086		uint64_t __promote = __p0; \
51087		__ret = (float64x1_t)(__promote); \
51088		__ret; \
51089		})
51090		__ai float32_t vcvts_f32_s32(int32_t __p0) {
51091		float32_t __ret;
51092		__ret = (float32_t) __builtin_neon_vcvts_f32_s32(__p0);
	52361	#ifdef __LITTLE_ENDIAN__
	52362	__ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	52363	float32x4_t __ret;
	52364	__ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
51093	52365	return __ret;
51094	52366	}
51095		__ai float32_t vcvts_f32_u32(uint32_t __p0) {
51096		float32_t __ret;
51097		__ret = (float32_t) __builtin_neon_vcvts_f32_u32(__p0);
	52367	#else
	52368	__ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	52369	float32x4_t __ret;
	52370	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52371	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52372	__ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	52373	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51098	52374	return __ret;
51099	52375	}
	52376	#endif
	52377
51100	52378	#ifdef __LITTLE_ENDIAN__
51101		__ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
	52379	__ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
51102	52380	float32x2_t __ret;
51103		__ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__p0, 9);
	52381	__ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
51104	52382	return __ret;
51105	52383	}
51106	52384	#else
51107		__ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
	52385	__ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
51108	52386	float32x2_t __ret;
51109		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51110		__ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__rev0, 9);
	52387	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52388	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52389	__ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
51111	52390	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51112	52391	return __ret;
51113	52392	}
51114		__ai float32x2_t __noswap_vcvt_f32_f64(float64x2_t __p0) {
51115		float32x2_t __ret;
51116		__ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__p0, 9);
51117		return __ret;
51118		}
51119	52393	#endif
51120	52394
51121		__ai float64_t vcvtd_f64_s64(int64_t __p0) {
	52395	#ifdef __LITTLE_ENDIAN__
	52396	__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
51122	52397	float64_t __ret;
51123		__ret = (float64_t) __builtin_neon_vcvtd_f64_s64(__p0);
	52398	__ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__p0);
51124	52399	return __ret;
51125	52400	}
51126		__ai float64_t vcvtd_f64_u64(uint64_t __p0) {
	52401	#else
	52402	__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
51127	52403	float64_t __ret;
51128		__ret = (float64_t) __builtin_neon_vcvtd_f64_u64(__p0);
	52404	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52405	__ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__rev0);
51129	52406	return __ret;
51130	52407	}
	52408	#endif
	52409
51131	52410	#ifdef __LITTLE_ENDIAN__
51132		__ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
51133		float64x2_t __ret;
51134		__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__p0, 51);
	52411	__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
	52412	float32_t __ret;
	52413	__ret = (float32_t) __builtin_neon_vpmaxnms_f32(__p0);
51135	52414	return __ret;
51136	52415	}
51137	52416	#else
51138		__ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
51139		float64x2_t __ret;
51140		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51141		__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__rev0, 51);
51142		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52417	__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
	52418	float32_t __ret;
	52419	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52420	__ret = (float32_t) __builtin_neon_vpmaxnms_f32(__rev0);
51143	52421	return __ret;
51144	52422	}
51145	52423	#endif
51146	52424
51147	52425	#ifdef __LITTLE_ENDIAN__
51148		__ai float64x2_t vcvtq_f64_s64(int64x2_t __p0) {
51149		float64x2_t __ret;
51150		__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__p0, 35);
	52426	__ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	52427	uint8x16_t __ret;
	52428	__ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
51151	52429	return __ret;
51152	52430	}
51153	52431	#else
51154		__ai float64x2_t vcvtq_f64_s64(int64x2_t __p0) {
51155		float64x2_t __ret;
51156		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51157		__ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__rev0, 35);
51158		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52432	__ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
	52433	uint8x16_t __ret;
	52434	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52435	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52436	__ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	52437	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
51159	52438	return __ret;
51160	52439	}
51161	52440	#endif
51162	52441
51163		__ai float64x1_t vcvt_f64_u64(uint64x1_t __p0) {
51164		float64x1_t __ret;
51165		__ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 19);
	52442	#ifdef __LITTLE_ENDIAN__
	52443	__ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	52444	uint32x4_t __ret;
	52445	__ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
51166	52446	return __ret;
51167	52447	}
51168		__ai float64x1_t vcvt_f64_s64(int64x1_t __p0) {
51169		float64x1_t __ret;
51170		__ret = (float64x1_t) __builtin_neon_vcvt_f64_v((int8x8_t)__p0, 3);
	52448	#else
	52449	__ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	52450	uint32x4_t __ret;
	52451	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52452	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52453	__ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	52454	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51171	52455	return __ret;
51172	52456	}
	52457	#endif
	52458
51173	52459	#ifdef __LITTLE_ENDIAN__
51174		__ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
51175		float64x2_t __ret;
51176		__ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__p0, 42);
	52460	__ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
	52461	uint16x8_t __ret;
	52462	__ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
51177	52463	return __ret;
51178	52464	}
51179	52465	#else
51180		__ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
51181		float64x2_t __ret;
51182		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51183		__ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__rev0, 42);
51184		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52466	__ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
	52467	uint16x8_t __ret;
	52468	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	52469	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52470	__ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	52471	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
51185	52472	return __ret;
51186	52473	}
51187		__ai float64x2_t __noswap_vcvt_f64_f32(float32x2_t __p0) {
51188		float64x2_t __ret;
51189		__ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__p0, 42);
	52474	#endif
	52475
	52476	#ifdef __LITTLE_ENDIAN__
	52477	__ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
	52478	int8x16_t __ret;
	52479	__ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	52480	return __ret;
	52481	}
	52482	#else
	52483	__ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
	52484	int8x16_t __ret;
	52485	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52486	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	52487	__ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
	52488	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
51190	52489	return __ret;
51191	52490	}
51192	52491	#endif
51193	52492
51194	52493	#ifdef __LITTLE_ENDIAN__
51195		__ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
51196		float16x8_t __ret;
51197		__ret = vcombine_f16(__p0, vcvt_f16_f32(__p1));
	52494	__ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
	52495	float64x2_t __ret;
	52496	__ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
51198	52497	return __ret;
51199	52498	}
51200	52499	#else
51201		__ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
51202		float16x8_t __ret;
51203		float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51204		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
51205		__ret = __noswap_vcombine_f16(__rev0, __noswap_vcvt_f16_f32(__rev1));
51206		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	52500	__ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
	52501	float64x2_t __ret;
	52502	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52503	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52504	__ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	52505	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51207	52506	return __ret;
51208	52507	}
51209	52508	#endif
51210	52509
51211	52510	#ifdef __LITTLE_ENDIAN__
51212		__ai float32x4_t vcvt_high_f32_f16(float16x8_t __p0) {
	52511	__ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
51213	52512	float32x4_t __ret;
51214		__ret = vcvt_f32_f16(vget_high_f16(__p0));
	52513	__ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
51215	52514	return __ret;
51216	52515	}
51217	52516	#else
51218		__ai float32x4_t vcvt_high_f32_f16(float16x8_t __p0) {
	52517	__ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
51219	52518	float32x4_t __ret;
51220		float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51221		__ret = __noswap_vcvt_f32_f16(__noswap_vget_high_f16(__rev0));
	52519	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52520	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52521	__ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
51222	52522	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51223	52523	return __ret;
51224	52524	}
51225	52525	#endif
51226	52526
51227	52527	#ifdef __LITTLE_ENDIAN__
51228		__ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
51229		float32x4_t __ret;
51230		__ret = vcombine_f32(__p0, vcvt_f32_f64(__p1));
	52528	__ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
	52529	int32x4_t __ret;
	52530	__ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
51231	52531	return __ret;
51232	52532	}
51233	52533	#else
51234		__ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
51235		float32x4_t __ret;
51236		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51237		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51238		__ret = __noswap_vcombine_f32(__rev0, __noswap_vcvt_f32_f64(__rev1));
	52534	__ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
	52535	int32x4_t __ret;
	52536	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52537	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52538	__ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
51239	52539	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51240	52540	return __ret;
51241	52541	}
51242	52542	#endif
51243	52543
51244	52544	#ifdef __LITTLE_ENDIAN__
51245		__ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
51246		float64x2_t __ret;
51247		__ret = vcvt_f64_f32(vget_high_f32(__p0));
	52545	__ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
	52546	int16x8_t __ret;
	52547	__ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
51248	52548	return __ret;
51249	52549	}
51250	52550	#else
51251		__ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
51252		float64x2_t __ret;
51253		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51254		__ret = __noswap_vcvt_f64_f32(__noswap_vget_high_f32(__rev0));
51255		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52551	__ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
	52552	int16x8_t __ret;
	52553	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	52554	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52555	__ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
	52556	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
51256	52557	return __ret;
51257	52558	}
51258	52559	#endif
51259	52560
51260		#define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
51261		float32_t __ret; \
51262		uint32_t __s0 = __p0; \
51263		__ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
51264		__ret; \
51265		})
51266		#define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
51267		float32_t __ret; \
51268		int32_t __s0 = __p0; \
51269		__ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
51270		__ret; \
51271		})
51272	52561	#ifdef __LITTLE_ENDIAN__
51273		#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
51274		float64x2_t __ret; \
51275		uint64x2_t __s0 = __p0; \
51276		__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 51); \
51277		__ret; \
51278		})
	52562	__ai float64_t vpminqd_f64(float64x2_t __p0) {
	52563	float64_t __ret;
	52564	__ret = (float64_t) __builtin_neon_vpminqd_f64(__p0);
	52565	return __ret;
	52566	}
51279	52567	#else
51280		#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
51281		float64x2_t __ret; \
51282		uint64x2_t __s0 = __p0; \
51283		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51284		__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 51); \
51285		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
51286		__ret; \
51287		})
	52568	__ai float64_t vpminqd_f64(float64x2_t __p0) {
	52569	float64_t __ret;
	52570	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52571	__ret = (float64_t) __builtin_neon_vpminqd_f64(__rev0);
	52572	return __ret;
	52573	}
51288	52574	#endif
51289	52575
51290	52576	#ifdef __LITTLE_ENDIAN__
51291		#define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
51292		float64x2_t __ret; \
51293		int64x2_t __s0 = __p0; \
51294		__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 35); \
51295		__ret; \
51296		})
	52577	__ai float32_t vpmins_f32(float32x2_t __p0) {
	52578	float32_t __ret;
	52579	__ret = (float32_t) __builtin_neon_vpmins_f32(__p0);
	52580	return __ret;
	52581	}
51297	52582	#else
51298		#define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
51299		float64x2_t __ret; \
51300		int64x2_t __s0 = __p0; \
51301		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51302		__ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 35); \
51303		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
51304		__ret; \
51305		})
	52583	__ai float32_t vpmins_f32(float32x2_t __p0) {
	52584	float32_t __ret;
	52585	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52586	__ret = (float32_t) __builtin_neon_vpmins_f32(__rev0);
	52587	return __ret;
	52588	}
51306	52589	#endif
51307	52590
51308		#define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
51309		float64x1_t __ret; \
51310		uint64x1_t __s0 = __p0; \
51311		__ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
51312		__ret; \
51313		})
51314		#define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
51315		float64x1_t __ret; \
51316		int64x1_t __s0 = __p0; \
51317		__ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
51318		__ret; \
51319		})
51320		#define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
51321		float64_t __ret; \
51322		uint64_t __s0 = __p0; \
51323		__ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
51324		__ret; \
51325		})
51326		#define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
51327		float64_t __ret; \
51328		int64_t __s0 = __p0; \
51329		__ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
51330		__ret; \
51331		})
51332		#define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
51333		int32_t __ret; \
51334		float32_t __s0 = __p0; \
51335		__ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
51336		__ret; \
51337		})
51338	52591	#ifdef __LITTLE_ENDIAN__
51339		#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
51340		int64x2_t __ret; \
51341		float64x2_t __s0 = __p0; \
51342		__ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__s0, __p1, 35); \
51343		__ret; \
51344		})
	52592	__ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	52593	float64x2_t __ret;
	52594	__ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	52595	return __ret;
	52596	}
51345	52597	#else
51346		#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
51347		int64x2_t __ret; \
51348		float64x2_t __s0 = __p0; \
51349		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51350		__ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__rev0, __p1, 35); \
51351		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
51352		__ret; \
51353		})
	52598	__ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	52599	float64x2_t __ret;
	52600	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52601	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52602	__ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	52603	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52604	return __ret;
	52605	}
51354	52606	#endif
51355	52607
51356		#define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
51357		int64x1_t __ret; \
51358		float64x1_t __s0 = __p0; \
51359		__ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
51360		__ret; \
51361		})
51362		#define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
51363		int64_t __ret; \
51364		float64_t __s0 = __p0; \
51365		__ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
51366		__ret; \
51367		})
51368		#define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
51369		uint32_t __ret; \
51370		float32_t __s0 = __p0; \
51371		__ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
51372		__ret; \
51373		})
51374	52608	#ifdef __LITTLE_ENDIAN__
51375		#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
51376		uint64x2_t __ret; \
51377		float64x2_t __s0 = __p0; \
51378		__ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__s0, __p1, 51); \
51379		__ret; \
51380		})
51381		#else
51382		#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
51383		uint64x2_t __ret; \
51384		float64x2_t __s0 = __p0; \
51385		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51386		__ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__rev0, __p1, 51); \
51387		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
51388		__ret; \
51389		})
51390		#endif
51391
51392		#define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
51393		uint64x1_t __ret; \
51394		float64x1_t __s0 = __p0; \
51395		__ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
51396		__ret; \
51397		})
51398		#define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
51399		uint64_t __ret; \
51400		float64_t __s0 = __p0; \
51401		__ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
51402		__ret; \
51403		})
51404		__ai int32_t vcvts_s32_f32(float32_t __p0) {
51405		int32_t __ret;
51406		__ret = (int32_t) __builtin_neon_vcvts_s32_f32(__p0);
	52609	__ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	52610	float32x4_t __ret;
	52611	__ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
51407	52612	return __ret;
51408	52613	}
51409		__ai int64_t vcvtd_s64_f64(float64_t __p0) {
51410		int64_t __ret;
51411		__ret = (int64_t) __builtin_neon_vcvtd_s64_f64(__p0);
	52614	#else
	52615	__ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
	52616	float32x4_t __ret;
	52617	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52618	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52619	__ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
	52620	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51412	52621	return __ret;
51413	52622	}
	52623	#endif
	52624
51414	52625	#ifdef __LITTLE_ENDIAN__
51415		__ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
51416		int64x2_t __ret;
51417		__ret = (int64x2_t) __builtin_neon_vcvtq_s64_v((int8x16_t)__p0, 35);
	52626	__ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
	52627	float32x2_t __ret;
	52628	__ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
51418	52629	return __ret;
51419	52630	}
51420	52631	#else
51421		__ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
51422		int64x2_t __ret;
51423		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51424		__ret = (int64x2_t) __builtin_neon_vcvtq_s64_v((int8x16_t)__rev0, 35);
	52632	__ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
	52633	float32x2_t __ret;
	52634	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52635	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	52636	__ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
51425	52637	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51426	52638	return __ret;
51427	52639	}
51428	52640	#endif
51429	52641
51430		__ai int64x1_t vcvt_s64_f64(float64x1_t __p0) {
51431		int64x1_t __ret;
51432		__ret = (int64x1_t) __builtin_neon_vcvt_s64_v((int8x8_t)__p0, 3);
51433		return __ret;
51434		}
51435		__ai uint32_t vcvts_u32_f32(float32_t __p0) {
51436		uint32_t __ret;
51437		__ret = (uint32_t) __builtin_neon_vcvts_u32_f32(__p0);
	52642	#ifdef __LITTLE_ENDIAN__
	52643	__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
	52644	float64_t __ret;
	52645	__ret = (float64_t) __builtin_neon_vpminnmqd_f64(__p0);
51438	52646	return __ret;
51439	52647	}
51440		__ai uint64_t vcvtd_u64_f64(float64_t __p0) {
51441		uint64_t __ret;
51442		__ret = (uint64_t) __builtin_neon_vcvtd_u64_f64(__p0);
	52648	#else
	52649	__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
	52650	float64_t __ret;
	52651	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52652	__ret = (float64_t) __builtin_neon_vpminnmqd_f64(__rev0);
51443	52653	return __ret;
51444	52654	}
	52655	#endif
	52656
51445	52657	#ifdef __LITTLE_ENDIAN__
51446		__ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
51447		uint64x2_t __ret;
51448		__ret = (uint64x2_t) __builtin_neon_vcvtq_u64_v((int8x16_t)__p0, 51);
	52658	__ai float32_t vpminnms_f32(float32x2_t __p0) {
	52659	float32_t __ret;
	52660	__ret = (float32_t) __builtin_neon_vpminnms_f32(__p0);
51449	52661	return __ret;
51450	52662	}
51451	52663	#else
51452		__ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
51453		uint64x2_t __ret;
51454		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51455		__ret = (uint64x2_t) __builtin_neon_vcvtq_u64_v((int8x16_t)__rev0, 51);
51456		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52664	__ai float32_t vpminnms_f32(float32x2_t __p0) {
	52665	float32_t __ret;
	52666	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52667	__ret = (float32_t) __builtin_neon_vpminnms_f32(__rev0);
51457	52668	return __ret;
51458	52669	}
51459	52670	#endif
51460	52671
51461		__ai uint64x1_t vcvt_u64_f64(float64x1_t __p0) {
51462		uint64x1_t __ret;
51463		__ret = (uint64x1_t) __builtin_neon_vcvt_u64_v((int8x8_t)__p0, 19);
51464		return __ret;
51465		}
51466		__ai int32_t vcvtas_s32_f32(float32_t __p0) {
51467		int32_t __ret;
51468		__ret = (int32_t) __builtin_neon_vcvtas_s32_f32(__p0);
	52672	#ifdef __LITTLE_ENDIAN__
	52673	__ai int64x2_t vqabsq_s64(int64x2_t __p0) {
	52674	int64x2_t __ret;
	52675	__ret = (int64x2_t) __builtin_neon_vqabsq_v((int8x16_t)__p0, 35);
51469	52676	return __ret;
51470	52677	}
51471		__ai int64_t vcvtad_s64_f64(float64_t __p0) {
51472		int64_t __ret;
51473		__ret = (int64_t) __builtin_neon_vcvtad_s64_f64(__p0);
	52678	#else
	52679	__ai int64x2_t vqabsq_s64(int64x2_t __p0) {
	52680	int64x2_t __ret;
	52681	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52682	__ret = (int64x2_t) __builtin_neon_vqabsq_v((int8x16_t)__rev0, 35);
	52683	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51474	52684	return __ret;
51475	52685	}
51476		__ai uint32_t vcvtas_u32_f32(float32_t __p0) {
51477		uint32_t __ret;
51478		__ret = (uint32_t) __builtin_neon_vcvtas_u32_f32(__p0);
	52686	#endif
	52687
	52688	__ai int64x1_t vqabs_s64(int64x1_t __p0) {
	52689	int64x1_t __ret;
	52690	__ret = (int64x1_t) __builtin_neon_vqabs_v((int8x8_t)__p0, 3);
51479	52691	return __ret;
51480	52692	}
51481		__ai uint64_t vcvtad_u64_f64(float64_t __p0) {
51482		uint64_t __ret;
51483		__ret = (uint64_t) __builtin_neon_vcvtad_u64_f64(__p0);
	52693	__ai int8_t vqabsb_s8(int8_t __p0) {
	52694	int8_t __ret;
	52695	__ret = (int8_t) __builtin_neon_vqabsb_s8(__p0);
51484	52696	return __ret;
51485	52697	}
51486		__ai int32_t vcvtms_s32_f32(float32_t __p0) {
	52698	__ai int32_t vqabss_s32(int32_t __p0) {
51487	52699	int32_t __ret;
51488		__ret = (int32_t) __builtin_neon_vcvtms_s32_f32(__p0);
	52700	__ret = (int32_t) __builtin_neon_vqabss_s32(__p0);
51489	52701	return __ret;
51490	52702	}
51491		__ai int64_t vcvtmd_s64_f64(float64_t __p0) {
	52703	__ai int64_t vqabsd_s64(int64_t __p0) {
51492	52704	int64_t __ret;
51493		__ret = (int64_t) __builtin_neon_vcvtmd_s64_f64(__p0);
	52705	__ret = (int64_t) __builtin_neon_vqabsd_s64(__p0);
51494	52706	return __ret;
51495	52707	}
51496		__ai uint32_t vcvtms_u32_f32(float32_t __p0) {
51497		uint32_t __ret;
51498		__ret = (uint32_t) __builtin_neon_vcvtms_u32_f32(__p0);
	52708	__ai int16_t vqabsh_s16(int16_t __p0) {
	52709	int16_t __ret;
	52710	__ret = (int16_t) __builtin_neon_vqabsh_s16(__p0);
51499	52711	return __ret;
51500	52712	}
51501		__ai uint64_t vcvtmd_u64_f64(float64_t __p0) {
51502		uint64_t __ret;
51503		__ret = (uint64_t) __builtin_neon_vcvtmd_u64_f64(__p0);
	52713	__ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
	52714	uint8_t __ret;
	52715	__ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
51504	52716	return __ret;
51505	52717	}
51506		__ai int32_t vcvtns_s32_f32(float32_t __p0) {
51507		int32_t __ret;
51508		__ret = (int32_t) __builtin_neon_vcvtns_s32_f32(__p0);
	52718	__ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
	52719	uint32_t __ret;
	52720	__ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
51509	52721	return __ret;
51510	52722	}
51511		__ai int64_t vcvtnd_s64_f64(float64_t __p0) {
51512		int64_t __ret;
51513		__ret = (int64_t) __builtin_neon_vcvtnd_s64_f64(__p0);
	52723	__ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
	52724	uint64_t __ret;
	52725	__ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
51514	52726	return __ret;
51515	52727	}
51516		__ai uint32_t vcvtns_u32_f32(float32_t __p0) {
51517		uint32_t __ret;
51518		__ret = (uint32_t) __builtin_neon_vcvtns_u32_f32(__p0);
	52728	__ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
	52729	uint16_t __ret;
	52730	__ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
51519	52731	return __ret;
51520	52732	}
51521		__ai uint64_t vcvtnd_u64_f64(float64_t __p0) {
51522		uint64_t __ret;
51523		__ret = (uint64_t) __builtin_neon_vcvtnd_u64_f64(__p0);
	52733	__ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
	52734	int8_t __ret;
	52735	__ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
51524	52736	return __ret;
51525	52737	}
51526		__ai int32_t vcvtps_s32_f32(float32_t __p0) {
	52738	__ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
51527	52739	int32_t __ret;
51528		__ret = (int32_t) __builtin_neon_vcvtps_s32_f32(__p0);
	52740	__ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
51529	52741	return __ret;
51530	52742	}
51531		__ai int64_t vcvtpd_s64_f64(float64_t __p0) {
	52743	__ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
51532	52744	int64_t __ret;
51533		__ret = (int64_t) __builtin_neon_vcvtpd_s64_f64(__p0);
	52745	__ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
51534	52746	return __ret;
51535	52747	}
51536		__ai uint32_t vcvtps_u32_f32(float32_t __p0) {
51537		uint32_t __ret;
51538		__ret = (uint32_t) __builtin_neon_vcvtps_u32_f32(__p0);
	52748	__ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
	52749	int16_t __ret;
	52750	__ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
51539	52751	return __ret;
51540	52752	}
51541		__ai uint64_t vcvtpd_u64_f64(float64_t __p0) {
51542		uint64_t __ret;
51543		__ret = (uint64_t) __builtin_neon_vcvtpd_u64_f64(__p0);
	52753	__ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
	52754	int64_t __ret;
	52755	__ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
51544	52756	return __ret;
51545	52757	}
51546		__ai float32_t vcvtxd_f32_f64(float64_t __p0) {
51547		float32_t __ret;
51548		__ret = (float32_t) __builtin_neon_vcvtxd_f32_f64(__p0);
	52758	__ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
	52759	int32_t __ret;
	52760	__ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
51549	52761	return __ret;
51550	52762	}
51551	52763	#ifdef __LITTLE_ENDIAN__
51552		__ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
51553		float32x2_t __ret;
51554		__ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__p0, 42);
	52764	__ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
	52765	int64x2_t __ret;
	52766	__ret = vqdmlal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
51555	52767	return __ret;
51556	52768	}
51557	52769	#else
51558		__ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
51559		float32x2_t __ret;
51560		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51561		__ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__rev0, 42);
	52770	__ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
	52771	int64x2_t __ret;
	52772	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52773	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52774	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	52775	__ret = __noswap_vqdmlal_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
51562	52776	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51563	52777	return __ret;
51564	52778	}
51565		__ai float32x2_t __noswap_vcvtx_f32_f64(float64x2_t __p0) {
51566		float32x2_t __ret;
51567		__ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__p0, 42);
51568		return __ret;
51569		}
51570	52779	#endif
51571	52780
51572	52781	#ifdef __LITTLE_ENDIAN__
51573		__ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
51574		float32x4_t __ret;
51575		__ret = vcombine_f32(__p0, vcvtx_f32_f64(__p1));
	52782	__ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
	52783	int32x4_t __ret;
	52784	__ret = vqdmlal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
51576	52785	return __ret;
51577	52786	}
51578	52787	#else
51579		__ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
51580		float32x4_t __ret;
51581		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51582		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51583		__ret = __noswap_vcombine_f32(__rev0, __noswap_vcvtx_f32_f64(__rev1));
	52788	__ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
	52789	int32x4_t __ret;
	52790	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52791	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52792	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	52793	__ret = __noswap_vqdmlal_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
51584	52794	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51585	52795	return __ret;
51586	52796	}
51587	52797	#endif
51588	52798
51589	52799	#ifdef __LITTLE_ENDIAN__
51590		__ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
51591		float64x2_t __ret;
51592		__ret = __p0 / __p1;
51593		return __ret;
51594		}
	52800	#define vqdmlal_high_lane_s32(__p0_607, __p1_607, __p2_607, __p3_607) __extension__ ({ \
	52801	int64x2_t __ret_607; \
	52802	int64x2_t __s0_607 = __p0_607; \
	52803	int32x4_t __s1_607 = __p1_607; \
	52804	int32x2_t __s2_607 = __p2_607; \
	52805	__ret_607 = vqdmlal_s32(__s0_607, vget_high_s32(__s1_607), splat_lane_s32(__s2_607, __p3_607)); \
	52806	__ret_607; \
	52807	})
51595	52808	#else
51596		__ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
51597		float64x2_t __ret;
51598		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51599		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51600		__ret = __rev0 / __rev1;
51601		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51602		return __ret;
51603		}
	52809	#define vqdmlal_high_lane_s32(__p0_608, __p1_608, __p2_608, __p3_608) __extension__ ({ \
	52810	int64x2_t __ret_608; \
	52811	int64x2_t __s0_608 = __p0_608; \
	52812	int32x4_t __s1_608 = __p1_608; \
	52813	int32x2_t __s2_608 = __p2_608; \
	52814	int64x2_t __rev0_608; __rev0_608 = __builtin_shufflevector(__s0_608, __s0_608, 1, 0); \
	52815	int32x4_t __rev1_608; __rev1_608 = __builtin_shufflevector(__s1_608, __s1_608, 3, 2, 1, 0); \
	52816	int32x2_t __rev2_608; __rev2_608 = __builtin_shufflevector(__s2_608, __s2_608, 1, 0); \
	52817	__ret_608 = __noswap_vqdmlal_s32(__rev0_608, __noswap_vget_high_s32(__rev1_608), __noswap_splat_lane_s32(__rev2_608, __p3_608)); \
	52818	__ret_608 = __builtin_shufflevector(__ret_608, __ret_608, 1, 0); \
	52819	__ret_608; \
	52820	})
51604	52821	#endif
51605	52822
51606	52823	#ifdef __LITTLE_ENDIAN__
51607		__ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
51608		float32x4_t __ret;
51609		__ret = __p0 / __p1;
	52824	#define vqdmlal_high_lane_s16(__p0_609, __p1_609, __p2_609, __p3_609) __extension__ ({ \
	52825	int32x4_t __ret_609; \
	52826	int32x4_t __s0_609 = __p0_609; \
	52827	int16x8_t __s1_609 = __p1_609; \
	52828	int16x4_t __s2_609 = __p2_609; \
	52829	__ret_609 = vqdmlal_s16(__s0_609, vget_high_s16(__s1_609), splat_lane_s16(__s2_609, __p3_609)); \
	52830	__ret_609; \
	52831	})
	52832	#else
	52833	#define vqdmlal_high_lane_s16(__p0_610, __p1_610, __p2_610, __p3_610) __extension__ ({ \
	52834	int32x4_t __ret_610; \
	52835	int32x4_t __s0_610 = __p0_610; \
	52836	int16x8_t __s1_610 = __p1_610; \
	52837	int16x4_t __s2_610 = __p2_610; \
	52838	int32x4_t __rev0_610; __rev0_610 = __builtin_shufflevector(__s0_610, __s0_610, 3, 2, 1, 0); \
	52839	int16x8_t __rev1_610; __rev1_610 = __builtin_shufflevector(__s1_610, __s1_610, 7, 6, 5, 4, 3, 2, 1, 0); \
	52840	int16x4_t __rev2_610; __rev2_610 = __builtin_shufflevector(__s2_610, __s2_610, 3, 2, 1, 0); \
	52841	__ret_610 = __noswap_vqdmlal_s16(__rev0_610, __noswap_vget_high_s16(__rev1_610), __noswap_splat_lane_s16(__rev2_610, __p3_610)); \
	52842	__ret_610 = __builtin_shufflevector(__ret_610, __ret_610, 3, 2, 1, 0); \
	52843	__ret_610; \
	52844	})
	52845	#endif
	52846
	52847	#ifdef __LITTLE_ENDIAN__
	52848	#define vqdmlal_high_laneq_s32(__p0_611, __p1_611, __p2_611, __p3_611) __extension__ ({ \
	52849	int64x2_t __ret_611; \
	52850	int64x2_t __s0_611 = __p0_611; \
	52851	int32x4_t __s1_611 = __p1_611; \
	52852	int32x4_t __s2_611 = __p2_611; \
	52853	__ret_611 = vqdmlal_s32(__s0_611, vget_high_s32(__s1_611), splat_laneq_s32(__s2_611, __p3_611)); \
	52854	__ret_611; \
	52855	})
	52856	#else
	52857	#define vqdmlal_high_laneq_s32(__p0_612, __p1_612, __p2_612, __p3_612) __extension__ ({ \
	52858	int64x2_t __ret_612; \
	52859	int64x2_t __s0_612 = __p0_612; \
	52860	int32x4_t __s1_612 = __p1_612; \
	52861	int32x4_t __s2_612 = __p2_612; \
	52862	int64x2_t __rev0_612; __rev0_612 = __builtin_shufflevector(__s0_612, __s0_612, 1, 0); \
	52863	int32x4_t __rev1_612; __rev1_612 = __builtin_shufflevector(__s1_612, __s1_612, 3, 2, 1, 0); \
	52864	int32x4_t __rev2_612; __rev2_612 = __builtin_shufflevector(__s2_612, __s2_612, 3, 2, 1, 0); \
	52865	__ret_612 = __noswap_vqdmlal_s32(__rev0_612, __noswap_vget_high_s32(__rev1_612), __noswap_splat_laneq_s32(__rev2_612, __p3_612)); \
	52866	__ret_612 = __builtin_shufflevector(__ret_612, __ret_612, 1, 0); \
	52867	__ret_612; \
	52868	})
	52869	#endif
	52870
	52871	#ifdef __LITTLE_ENDIAN__
	52872	#define vqdmlal_high_laneq_s16(__p0_613, __p1_613, __p2_613, __p3_613) __extension__ ({ \
	52873	int32x4_t __ret_613; \
	52874	int32x4_t __s0_613 = __p0_613; \
	52875	int16x8_t __s1_613 = __p1_613; \
	52876	int16x8_t __s2_613 = __p2_613; \
	52877	__ret_613 = vqdmlal_s16(__s0_613, vget_high_s16(__s1_613), splat_laneq_s16(__s2_613, __p3_613)); \
	52878	__ret_613; \
	52879	})
	52880	#else
	52881	#define vqdmlal_high_laneq_s16(__p0_614, __p1_614, __p2_614, __p3_614) __extension__ ({ \
	52882	int32x4_t __ret_614; \
	52883	int32x4_t __s0_614 = __p0_614; \
	52884	int16x8_t __s1_614 = __p1_614; \
	52885	int16x8_t __s2_614 = __p2_614; \
	52886	int32x4_t __rev0_614; __rev0_614 = __builtin_shufflevector(__s0_614, __s0_614, 3, 2, 1, 0); \
	52887	int16x8_t __rev1_614; __rev1_614 = __builtin_shufflevector(__s1_614, __s1_614, 7, 6, 5, 4, 3, 2, 1, 0); \
	52888	int16x8_t __rev2_614; __rev2_614 = __builtin_shufflevector(__s2_614, __s2_614, 7, 6, 5, 4, 3, 2, 1, 0); \
	52889	__ret_614 = __noswap_vqdmlal_s16(__rev0_614, __noswap_vget_high_s16(__rev1_614), __noswap_splat_laneq_s16(__rev2_614, __p3_614)); \
	52890	__ret_614 = __builtin_shufflevector(__ret_614, __ret_614, 3, 2, 1, 0); \
	52891	__ret_614; \
	52892	})
	52893	#endif
	52894
	52895	#ifdef __LITTLE_ENDIAN__
	52896	__ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
	52897	int64x2_t __ret;
	52898	__ret = vqdmlal_n_s32(__p0, vget_high_s32(__p1), __p2);
51610	52899	return __ret;
51611	52900	}
51612	52901	#else
51613		__ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
51614		float32x4_t __ret;
51615		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51616		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
51617		__ret = __rev0 / __rev1;
51618		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	52902	__ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
	52903	int64x2_t __ret;
	52904	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	52905	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	52906	__ret = __noswap_vqdmlal_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
	52907	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51619	52908	return __ret;
51620	52909	}
51621	52910	#endif
51622	52911
51623		__ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
51624		float64x1_t __ret;
51625		__ret = __p0 / __p1;
51626		return __ret;
51627		}
51628	52912	#ifdef __LITTLE_ENDIAN__
51629		__ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
51630		float32x2_t __ret;
51631		__ret = __p0 / __p1;
	52913	__ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
	52914	int32x4_t __ret;
	52915	__ret = vqdmlal_n_s16(__p0, vget_high_s16(__p1), __p2);
51632	52916	return __ret;
51633	52917	}
51634	52918	#else
51635		__ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
51636		float32x2_t __ret;
51637		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51638		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51639		__ret = __rev0 / __rev1;
51640		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	52919	__ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
	52920	int32x4_t __ret;
	52921	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	52922	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	52923	__ret = __noswap_vqdmlal_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
	52924	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51641	52925	return __ret;
51642	52926	}
51643	52927	#endif
51644	52928
51645	52929	#ifdef __LITTLE_ENDIAN__
51646		#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
51647		poly8_t __ret; \
51648		poly8x8_t __s0 = __p0; \
51649		__ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__s0, __p1); \
	52930	#define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	52931	int64_t __ret; \
	52932	int64_t __s0 = __p0; \
	52933	int32_t __s1 = __p1; \
	52934	int32x2_t __s2 = __p2; \
	52935	__ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __s2, __p3); \
51650	52936	__ret; \
51651	52937	})
51652	52938	#else
51653		#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
51654		poly8_t __ret; \
51655		poly8x8_t __s0 = __p0; \
51656		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51657		__ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__rev0, __p1); \
	52939	#define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	52940	int64_t __ret; \
	52941	int64_t __s0 = __p0; \
	52942	int32_t __s1 = __p1; \
	52943	int32x2_t __s2 = __p2; \
	52944	int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	52945	__ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __rev2, __p3); \
51658	52946	__ret; \
51659	52947	})
51660	52948	#endif
51661	52949
51662	52950	#ifdef __LITTLE_ENDIAN__
51663		#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
51664		poly16_t __ret; \
51665		poly16x4_t __s0 = __p0; \
51666		__ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__s0, __p1); \
	52951	#define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	52952	int32_t __ret; \
	52953	int32_t __s0 = __p0; \
	52954	int16_t __s1 = __p1; \
	52955	int16x4_t __s2 = __p2; \
	52956	__ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __s2, __p3); \
51667	52957	__ret; \
51668	52958	})
51669	52959	#else
51670		#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
51671		poly16_t __ret; \
51672		poly16x4_t __s0 = __p0; \
51673		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51674		__ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__rev0, __p1); \
	52960	#define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	52961	int32_t __ret; \
	52962	int32_t __s0 = __p0; \
	52963	int16_t __s1 = __p1; \
	52964	int16x4_t __s2 = __p2; \
	52965	int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	52966	__ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __rev2, __p3); \
51675	52967	__ret; \
51676	52968	})
51677	52969	#endif
51678	52970
51679	52971	#ifdef __LITTLE_ENDIAN__
51680		#define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
51681		uint8_t __ret; \
51682		uint8x8_t __s0 = __p0; \
51683		__ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
	52972	#define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	52973	int64_t __ret; \
	52974	int64_t __s0 = __p0; \
	52975	int32_t __s1 = __p1; \
	52976	int32x4_t __s2 = __p2; \
	52977	__ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __s2, __p3); \
51684	52978	__ret; \
51685	52979	})
51686	52980	#else
51687		#define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
51688		uint8_t __ret; \
51689		uint8x8_t __s0 = __p0; \
51690		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51691		__ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
	52981	#define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	52982	int64_t __ret; \
	52983	int64_t __s0 = __p0; \
	52984	int32_t __s1 = __p1; \
	52985	int32x4_t __s2 = __p2; \
	52986	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	52987	__ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __rev2, __p3); \
51692	52988	__ret; \
51693	52989	})
51694	52990	#endif
51695	52991
51696	52992	#ifdef __LITTLE_ENDIAN__
51697		#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
51698		uint32_t __ret; \
51699		uint32x2_t __s0 = __p0; \
51700		__ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
	52993	#define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	52994	int32_t __ret; \
	52995	int32_t __s0 = __p0; \
	52996	int16_t __s1 = __p1; \
	52997	int16x8_t __s2 = __p2; \
	52998	__ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __s2, __p3); \
51701	52999	__ret; \
51702	53000	})
51703	53001	#else
51704		#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
51705		uint32_t __ret; \
51706		uint32x2_t __s0 = __p0; \
51707		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51708		__ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
	53002	#define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	53003	int32_t __ret; \
	53004	int32_t __s0 = __p0; \
	53005	int16_t __s1 = __p1; \
	53006	int16x8_t __s2 = __p2; \
	53007	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
	53008	__ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __rev2, __p3); \
51709	53009	__ret; \
51710	53010	})
51711	53011	#endif
51712	53012
51713		#define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
51714		uint64_t __ret; \
51715		uint64x1_t __s0 = __p0; \
51716		__ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
51717		__ret; \
51718		})
51719	53013	#ifdef __LITTLE_ENDIAN__
51720		#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
51721		uint16_t __ret; \
51722		uint16x4_t __s0 = __p0; \
51723		__ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
51724		__ret; \
	53014	#define vqdmlal_laneq_s32(__p0_615, __p1_615, __p2_615, __p3_615) __extension__ ({ \
	53015	int64x2_t __ret_615; \
	53016	int64x2_t __s0_615 = __p0_615; \
	53017	int32x2_t __s1_615 = __p1_615; \
	53018	int32x4_t __s2_615 = __p2_615; \
	53019	__ret_615 = vqdmlal_s32(__s0_615, __s1_615, splat_laneq_s32(__s2_615, __p3_615)); \
	53020	__ret_615; \
51725	53021	})
51726	53022	#else
51727		#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
51728		uint16_t __ret; \
51729		uint16x4_t __s0 = __p0; \
51730		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51731		__ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
51732		__ret; \
	53023	#define vqdmlal_laneq_s32(__p0_616, __p1_616, __p2_616, __p3_616) __extension__ ({ \
	53024	int64x2_t __ret_616; \
	53025	int64x2_t __s0_616 = __p0_616; \
	53026	int32x2_t __s1_616 = __p1_616; \
	53027	int32x4_t __s2_616 = __p2_616; \
	53028	int64x2_t __rev0_616; __rev0_616 = __builtin_shufflevector(__s0_616, __s0_616, 1, 0); \
	53029	int32x2_t __rev1_616; __rev1_616 = __builtin_shufflevector(__s1_616, __s1_616, 1, 0); \
	53030	int32x4_t __rev2_616; __rev2_616 = __builtin_shufflevector(__s2_616, __s2_616, 3, 2, 1, 0); \
	53031	__ret_616 = __noswap_vqdmlal_s32(__rev0_616, __rev1_616, __noswap_splat_laneq_s32(__rev2_616, __p3_616)); \
	53032	__ret_616 = __builtin_shufflevector(__ret_616, __ret_616, 1, 0); \
	53033	__ret_616; \
51733	53034	})
51734	53035	#endif
51735	53036
51736	53037	#ifdef __LITTLE_ENDIAN__
51737		#define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
51738		int8_t __ret; \
51739		int8x8_t __s0 = __p0; \
51740		__ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
51741		__ret; \
	53038	#define vqdmlal_laneq_s16(__p0_617, __p1_617, __p2_617, __p3_617) __extension__ ({ \
	53039	int32x4_t __ret_617; \
	53040	int32x4_t __s0_617 = __p0_617; \
	53041	int16x4_t __s1_617 = __p1_617; \
	53042	int16x8_t __s2_617 = __p2_617; \
	53043	__ret_617 = vqdmlal_s16(__s0_617, __s1_617, splat_laneq_s16(__s2_617, __p3_617)); \
	53044	__ret_617; \
51742	53045	})
51743	53046	#else
51744		#define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
51745		int8_t __ret; \
51746		int8x8_t __s0 = __p0; \
51747		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51748		__ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
51749		__ret; \
	53047	#define vqdmlal_laneq_s16(__p0_618, __p1_618, __p2_618, __p3_618) __extension__ ({ \
	53048	int32x4_t __ret_618; \
	53049	int32x4_t __s0_618 = __p0_618; \
	53050	int16x4_t __s1_618 = __p1_618; \
	53051	int16x8_t __s2_618 = __p2_618; \
	53052	int32x4_t __rev0_618; __rev0_618 = __builtin_shufflevector(__s0_618, __s0_618, 3, 2, 1, 0); \
	53053	int16x4_t __rev1_618; __rev1_618 = __builtin_shufflevector(__s1_618, __s1_618, 3, 2, 1, 0); \
	53054	int16x8_t __rev2_618; __rev2_618 = __builtin_shufflevector(__s2_618, __s2_618, 7, 6, 5, 4, 3, 2, 1, 0); \
	53055	__ret_618 = __noswap_vqdmlal_s16(__rev0_618, __rev1_618, __noswap_splat_laneq_s16(__rev2_618, __p3_618)); \
	53056	__ret_618 = __builtin_shufflevector(__ret_618, __ret_618, 3, 2, 1, 0); \
	53057	__ret_618; \
51750	53058	})
51751	53059	#endif
51752	53060
51753		#define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
51754		float64_t __ret; \
51755		float64x1_t __s0 = __p0; \
51756		__ret = (float64_t) __builtin_neon_vdupd_lane_f64((float64x1_t)__s0, __p1); \
51757		__ret; \
51758		})
	53061	__ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
	53062	int64_t __ret;
	53063	__ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
	53064	return __ret;
	53065	}
	53066	__ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
	53067	int32_t __ret;
	53068	__ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
	53069	return __ret;
	53070	}
51759	53071	#ifdef __LITTLE_ENDIAN__
51760		#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
51761		float32_t __ret; \
51762		float32x2_t __s0 = __p0; \
51763		__ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__s0, __p1); \
51764		__ret; \
51765		})
	53072	__ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
	53073	int64x2_t __ret;
	53074	__ret = vqdmlsl_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
	53075	return __ret;
	53076	}
51766	53077	#else
51767		#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
51768		float32_t __ret; \
51769		float32x2_t __s0 = __p0; \
51770		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51771		__ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__rev0, __p1); \
51772		__ret; \
51773		})
	53078	__ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
	53079	int64x2_t __ret;
	53080	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	53081	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	53082	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	53083	__ret = __noswap_vqdmlsl_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
	53084	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	53085	return __ret;
	53086	}
51774	53087	#endif
51775	53088
51776	53089	#ifdef __LITTLE_ENDIAN__
51777		#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
51778		int32_t __ret; \
51779		int32x2_t __s0 = __p0; \
51780		__ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
51781		__ret; \
51782		})
	53090	__ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
	53091	int32x4_t __ret;
	53092	__ret = vqdmlsl_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
	53093	return __ret;
	53094	}
51783	53095	#else
51784		#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
51785		int32_t __ret; \
51786		int32x2_t __s0 = __p0; \
51787		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51788		__ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
51789		__ret; \
51790		})
	53096	__ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
	53097	int32x4_t __ret;
	53098	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	53099	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	53100	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	53101	__ret = __noswap_vqdmlsl_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
	53102	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	53103	return __ret;
	53104	}
51791	53105	#endif
51792	53106
51793		#define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
51794		int64_t __ret; \
51795		int64x1_t __s0 = __p0; \
51796		__ret = (int64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
51797		__ret; \
51798		})
51799	53107	#ifdef __LITTLE_ENDIAN__
51800		#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
51801		int16_t __ret; \
51802		int16x4_t __s0 = __p0; \
51803		__ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
51804		__ret; \
	53108	#define vqdmlsl_high_lane_s32(__p0_619, __p1_619, __p2_619, __p3_619) __extension__ ({ \
	53109	int64x2_t __ret_619; \
	53110	int64x2_t __s0_619 = __p0_619; \
	53111	int32x4_t __s1_619 = __p1_619; \
	53112	int32x2_t __s2_619 = __p2_619; \
	53113	__ret_619 = vqdmlsl_s32(__s0_619, vget_high_s32(__s1_619), splat_lane_s32(__s2_619, __p3_619)); \
	53114	__ret_619; \
51805	53115	})
51806	53116	#else
51807		#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
51808		int16_t __ret; \
51809		int16x4_t __s0 = __p0; \
51810		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51811		__ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
51812		__ret; \
	53117	#define vqdmlsl_high_lane_s32(__p0_620, __p1_620, __p2_620, __p3_620) __extension__ ({ \
	53118	int64x2_t __ret_620; \
	53119	int64x2_t __s0_620 = __p0_620; \
	53120	int32x4_t __s1_620 = __p1_620; \
	53121	int32x2_t __s2_620 = __p2_620; \
	53122	int64x2_t __rev0_620; __rev0_620 = __builtin_shufflevector(__s0_620, __s0_620, 1, 0); \
	53123	int32x4_t __rev1_620; __rev1_620 = __builtin_shufflevector(__s1_620, __s1_620, 3, 2, 1, 0); \
	53124	int32x2_t __rev2_620; __rev2_620 = __builtin_shufflevector(__s2_620, __s2_620, 1, 0); \
	53125	__ret_620 = __noswap_vqdmlsl_s32(__rev0_620, __noswap_vget_high_s32(__rev1_620), __noswap_splat_lane_s32(__rev2_620, __p3_620)); \
	53126	__ret_620 = __builtin_shufflevector(__ret_620, __ret_620, 1, 0); \
	53127	__ret_620; \
51813	53128	})
51814	53129	#endif
51815	53130
51816		#define vdup_lane_p64(__p0_443, __p1_443) __extension__ ({ \
51817		poly64x1_t __ret_443; \
51818		poly64x1_t __s0_443 = __p0_443; \
51819		__ret_443 = splat_lane_p64(__s0_443, __p1_443); \
51820		__ret_443; \
51821		})
51822	53131	#ifdef __LITTLE_ENDIAN__
51823		#define vdupq_lane_p64(__p0_444, __p1_444) __extension__ ({ \
51824		poly64x2_t __ret_444; \
51825		poly64x1_t __s0_444 = __p0_444; \
51826		__ret_444 = splatq_lane_p64(__s0_444, __p1_444); \
51827		__ret_444; \
	53132	#define vqdmlsl_high_lane_s16(__p0_621, __p1_621, __p2_621, __p3_621) __extension__ ({ \
	53133	int32x4_t __ret_621; \
	53134	int32x4_t __s0_621 = __p0_621; \
	53135	int16x8_t __s1_621 = __p1_621; \
	53136	int16x4_t __s2_621 = __p2_621; \
	53137	__ret_621 = vqdmlsl_s16(__s0_621, vget_high_s16(__s1_621), splat_lane_s16(__s2_621, __p3_621)); \
	53138	__ret_621; \
51828	53139	})
51829	53140	#else
51830		#define vdupq_lane_p64(__p0_445, __p1_445) __extension__ ({ \
51831		poly64x2_t __ret_445; \
51832		poly64x1_t __s0_445 = __p0_445; \
51833		__ret_445 = __noswap_splatq_lane_p64(__s0_445, __p1_445); \
51834		__ret_445 = __builtin_shufflevector(__ret_445, __ret_445, 1, 0); \
51835		__ret_445; \
	53141	#define vqdmlsl_high_lane_s16(__p0_622, __p1_622, __p2_622, __p3_622) __extension__ ({ \
	53142	int32x4_t __ret_622; \
	53143	int32x4_t __s0_622 = __p0_622; \
	53144	int16x8_t __s1_622 = __p1_622; \
	53145	int16x4_t __s2_622 = __p2_622; \
	53146	int32x4_t __rev0_622; __rev0_622 = __builtin_shufflevector(__s0_622, __s0_622, 3, 2, 1, 0); \
	53147	int16x8_t __rev1_622; __rev1_622 = __builtin_shufflevector(__s1_622, __s1_622, 7, 6, 5, 4, 3, 2, 1, 0); \
	53148	int16x4_t __rev2_622; __rev2_622 = __builtin_shufflevector(__s2_622, __s2_622, 3, 2, 1, 0); \
	53149	__ret_622 = __noswap_vqdmlsl_s16(__rev0_622, __noswap_vget_high_s16(__rev1_622), __noswap_splat_lane_s16(__rev2_622, __p3_622)); \
	53150	__ret_622 = __builtin_shufflevector(__ret_622, __ret_622, 3, 2, 1, 0); \
	53151	__ret_622; \
51836	53152	})
51837	53153	#endif
51838	53154
51839	53155	#ifdef __LITTLE_ENDIAN__
51840		#define vdupq_lane_f64(__p0_446, __p1_446) __extension__ ({ \
51841		float64x2_t __ret_446; \
51842		float64x1_t __s0_446 = __p0_446; \
51843		__ret_446 = splatq_lane_f64(__s0_446, __p1_446); \
51844		__ret_446; \
	53156	#define vqdmlsl_high_laneq_s32(__p0_623, __p1_623, __p2_623, __p3_623) __extension__ ({ \
	53157	int64x2_t __ret_623; \
	53158	int64x2_t __s0_623 = __p0_623; \
	53159	int32x4_t __s1_623 = __p1_623; \
	53160	int32x4_t __s2_623 = __p2_623; \
	53161	__ret_623 = vqdmlsl_s32(__s0_623, vget_high_s32(__s1_623), splat_laneq_s32(__s2_623, __p3_623)); \
	53162	__ret_623; \
51845	53163	})
51846	53164	#else
51847		#define vdupq_lane_f64(__p0_447, __p1_447) __extension__ ({ \
51848		float64x2_t __ret_447; \
51849		float64x1_t __s0_447 = __p0_447; \
51850		__ret_447 = __noswap_splatq_lane_f64(__s0_447, __p1_447); \
51851		__ret_447 = __builtin_shufflevector(__ret_447, __ret_447, 1, 0); \
51852		__ret_447; \
	53165	#define vqdmlsl_high_laneq_s32(__p0_624, __p1_624, __p2_624, __p3_624) __extension__ ({ \
	53166	int64x2_t __ret_624; \
	53167	int64x2_t __s0_624 = __p0_624; \
	53168	int32x4_t __s1_624 = __p1_624; \
	53169	int32x4_t __s2_624 = __p2_624; \
	53170	int64x2_t __rev0_624; __rev0_624 = __builtin_shufflevector(__s0_624, __s0_624, 1, 0); \
	53171	int32x4_t __rev1_624; __rev1_624 = __builtin_shufflevector(__s1_624, __s1_624, 3, 2, 1, 0); \
	53172	int32x4_t __rev2_624; __rev2_624 = __builtin_shufflevector(__s2_624, __s2_624, 3, 2, 1, 0); \
	53173	__ret_624 = __noswap_vqdmlsl_s32(__rev0_624, __noswap_vget_high_s32(__rev1_624), __noswap_splat_laneq_s32(__rev2_624, __p3_624)); \
	53174	__ret_624 = __builtin_shufflevector(__ret_624, __ret_624, 1, 0); \
	53175	__ret_624; \
51853	53176	})
51854	53177	#endif
51855	53178
51856	53179	#ifdef __LITTLE_ENDIAN__
51857		#define vdupq_lane_f16(__p0_448, __p1_448) __extension__ ({ \
51858		float16x8_t __ret_448; \
51859		float16x4_t __s0_448 = __p0_448; \
51860		__ret_448 = splatq_lane_f16(__s0_448, __p1_448); \
51861		__ret_448; \
	53180	#define vqdmlsl_high_laneq_s16(__p0_625, __p1_625, __p2_625, __p3_625) __extension__ ({ \
	53181	int32x4_t __ret_625; \
	53182	int32x4_t __s0_625 = __p0_625; \
	53183	int16x8_t __s1_625 = __p1_625; \
	53184	int16x8_t __s2_625 = __p2_625; \
	53185	__ret_625 = vqdmlsl_s16(__s0_625, vget_high_s16(__s1_625), splat_laneq_s16(__s2_625, __p3_625)); \
	53186	__ret_625; \
51862	53187	})
51863	53188	#else
51864		#define vdupq_lane_f16(__p0_449, __p1_449) __extension__ ({ \
51865		float16x8_t __ret_449; \
51866		float16x4_t __s0_449 = __p0_449; \
51867		float16x4_t __rev0_449; __rev0_449 = __builtin_shufflevector(__s0_449, __s0_449, 3, 2, 1, 0); \
51868		__ret_449 = __noswap_splatq_lane_f16(__rev0_449, __p1_449); \
51869		__ret_449 = __builtin_shufflevector(__ret_449, __ret_449, 7, 6, 5, 4, 3, 2, 1, 0); \
51870		__ret_449; \
	53189	#define vqdmlsl_high_laneq_s16(__p0_626, __p1_626, __p2_626, __p3_626) __extension__ ({ \
	53190	int32x4_t __ret_626; \
	53191	int32x4_t __s0_626 = __p0_626; \
	53192	int16x8_t __s1_626 = __p1_626; \
	53193	int16x8_t __s2_626 = __p2_626; \
	53194	int32x4_t __rev0_626; __rev0_626 = __builtin_shufflevector(__s0_626, __s0_626, 3, 2, 1, 0); \
	53195	int16x8_t __rev1_626; __rev1_626 = __builtin_shufflevector(__s1_626, __s1_626, 7, 6, 5, 4, 3, 2, 1, 0); \
	53196	int16x8_t __rev2_626; __rev2_626 = __builtin_shufflevector(__s2_626, __s2_626, 7, 6, 5, 4, 3, 2, 1, 0); \
	53197	__ret_626 = __noswap_vqdmlsl_s16(__rev0_626, __noswap_vget_high_s16(__rev1_626), __noswap_splat_laneq_s16(__rev2_626, __p3_626)); \
	53198	__ret_626 = __builtin_shufflevector(__ret_626, __ret_626, 3, 2, 1, 0); \
	53199	__ret_626; \
51871	53200	})
51872	53201	#endif
51873	53202
51874		#define vdup_lane_f64(__p0_450, __p1_450) __extension__ ({ \
51875		float64x1_t __ret_450; \
51876		float64x1_t __s0_450 = __p0_450; \
51877		__ret_450 = splat_lane_f64(__s0_450, __p1_450); \
51878		__ret_450; \
51879		})
51880	53203	#ifdef __LITTLE_ENDIAN__
51881		#define vdup_lane_f16(__p0_451, __p1_451) __extension__ ({ \
51882		float16x4_t __ret_451; \
51883		float16x4_t __s0_451 = __p0_451; \
51884		__ret_451 = splat_lane_f16(__s0_451, __p1_451); \
51885		__ret_451; \
	53204	__ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
	53205	int64x2_t __ret;
	53206	__ret = vqdmlsl_n_s32(__p0, vget_high_s32(__p1), __p2);
	53207	return __ret;
	53208	}
	53209	#else
	53210	__ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
	53211	int64x2_t __ret;
	53212	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	53213	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	53214	__ret = __noswap_vqdmlsl_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
	53215	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	53216	return __ret;
	53217	}
	53218	#endif
	53219
	53220	#ifdef __LITTLE_ENDIAN__
	53221	__ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
	53222	int32x4_t __ret;
	53223	__ret = vqdmlsl_n_s16(__p0, vget_high_s16(__p1), __p2);
	53224	return __ret;
	53225	}
	53226	#else
	53227	__ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
	53228	int32x4_t __ret;
	53229	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	53230	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	53231	__ret = __noswap_vqdmlsl_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
	53232	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	53233	return __ret;
	53234	}
	53235	#endif
	53236
	53237	#ifdef __LITTLE_ENDIAN__
	53238	#define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	53239	int64_t __ret; \
	53240	int64_t __s0 = __p0; \
	53241	int32_t __s1 = __p1; \
	53242	int32x2_t __s2 = __p2; \
	53243	__ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __s2, __p3); \
	53244	__ret; \
51886	53245	})
51887	53246	#else
51888		#define vdup_lane_f16(__p0_452, __p1_452) __extension__ ({ \
51889		float16x4_t __ret_452; \
51890		float16x4_t __s0_452 = __p0_452; \
51891		float16x4_t __rev0_452; __rev0_452 = __builtin_shufflevector(__s0_452, __s0_452, 3, 2, 1, 0); \
51892		__ret_452 = __noswap_splat_lane_f16(__rev0_452, __p1_452); \
51893		__ret_452 = __builtin_shufflevector(__ret_452, __ret_452, 3, 2, 1, 0); \
51894		__ret_452; \
	53247	#define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	53248	int64_t __ret; \
	53249	int64_t __s0 = __p0; \
	53250	int32_t __s1 = __p1; \
	53251	int32x2_t __s2 = __p2; \
	53252	int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
	53253	__ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __rev2, __p3); \
	53254	__ret; \
51895	53255	})
51896	53256	#endif
51897	53257
51898	53258	#ifdef __LITTLE_ENDIAN__
51899		#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
51900		poly8_t __ret; \
51901		poly8x16_t __s0 = __p0; \
51902		__ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__s0, __p1); \
	53259	#define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	53260	int32_t __ret; \
	53261	int32_t __s0 = __p0; \
	53262	int16_t __s1 = __p1; \
	53263	int16x4_t __s2 = __p2; \
	53264	__ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __s2, __p3); \
51903	53265	__ret; \
51904	53266	})
51905	53267	#else
51906		#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
51907		poly8_t __ret; \
51908		poly8x16_t __s0 = __p0; \
51909		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
51910		__ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__rev0, __p1); \
	53268	#define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	53269	int32_t __ret; \
	53270	int32_t __s0 = __p0; \
	53271	int16_t __s1 = __p1; \
	53272	int16x4_t __s2 = __p2; \
	53273	int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	53274	__ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __rev2, __p3); \
51911	53275	__ret; \
51912	53276	})
51913	53277	#endif
51914	53278
51915	53279	#ifdef __LITTLE_ENDIAN__
51916		#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
51917		poly16_t __ret; \
51918		poly16x8_t __s0 = __p0; \
51919		__ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__s0, __p1); \
	53280	#define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	53281	int64_t __ret; \
	53282	int64_t __s0 = __p0; \
	53283	int32_t __s1 = __p1; \
	53284	int32x4_t __s2 = __p2; \
	53285	__ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __s2, __p3); \
51920	53286	__ret; \
51921	53287	})
51922	53288	#else
51923		#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
51924		poly16_t __ret; \
51925		poly16x8_t __s0 = __p0; \
51926		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51927		__ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__rev0, __p1); \
	53289	#define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	53290	int64_t __ret; \
	53291	int64_t __s0 = __p0; \
	53292	int32_t __s1 = __p1; \
	53293	int32x4_t __s2 = __p2; \
	53294	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	53295	__ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __rev2, __p3); \
51928	53296	__ret; \
51929	53297	})
51930	53298	#endif
51931	53299
51932	53300	#ifdef __LITTLE_ENDIAN__
51933		#define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
51934		uint8_t __ret; \
51935		uint8x16_t __s0 = __p0; \
51936		__ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
	53301	#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	53302	int32_t __ret; \
	53303	int32_t __s0 = __p0; \
	53304	int16_t __s1 = __p1; \
	53305	int16x8_t __s2 = __p2; \
	53306	__ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __s2, __p3); \
51937	53307	__ret; \
51938	53308	})
51939	53309	#else
51940		#define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
51941		uint8_t __ret; \
51942		uint8x16_t __s0 = __p0; \
51943		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
51944		__ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
	53310	#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
	53311	int32_t __ret; \
	53312	int32_t __s0 = __p0; \
	53313	int16_t __s1 = __p1; \
	53314	int16x8_t __s2 = __p2; \
	53315	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
	53316	__ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __rev2, __p3); \
51945	53317	__ret; \
51946	53318	})
51947	53319	#endif
51948	53320
51949	53321	#ifdef __LITTLE_ENDIAN__
51950		#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
51951		uint32_t __ret; \
51952		uint32x4_t __s0 = __p0; \
51953		__ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
	53322	#define vqdmlsl_laneq_s32(__p0_627, __p1_627, __p2_627, __p3_627) __extension__ ({ \
	53323	int64x2_t __ret_627; \
	53324	int64x2_t __s0_627 = __p0_627; \
	53325	int32x2_t __s1_627 = __p1_627; \
	53326	int32x4_t __s2_627 = __p2_627; \
	53327	__ret_627 = vqdmlsl_s32(__s0_627, __s1_627, splat_laneq_s32(__s2_627, __p3_627)); \
	53328	__ret_627; \
	53329	})
	53330	#else
	53331	#define vqdmlsl_laneq_s32(__p0_628, __p1_628, __p2_628, __p3_628) __extension__ ({ \
	53332	int64x2_t __ret_628; \
	53333	int64x2_t __s0_628 = __p0_628; \
	53334	int32x2_t __s1_628 = __p1_628; \
	53335	int32x4_t __s2_628 = __p2_628; \
	53336	int64x2_t __rev0_628; __rev0_628 = __builtin_shufflevector(__s0_628, __s0_628, 1, 0); \
	53337	int32x2_t __rev1_628; __rev1_628 = __builtin_shufflevector(__s1_628, __s1_628, 1, 0); \
	53338	int32x4_t __rev2_628; __rev2_628 = __builtin_shufflevector(__s2_628, __s2_628, 3, 2, 1, 0); \
	53339	__ret_628 = __noswap_vqdmlsl_s32(__rev0_628, __rev1_628, __noswap_splat_laneq_s32(__rev2_628, __p3_628)); \
	53340	__ret_628 = __builtin_shufflevector(__ret_628, __ret_628, 1, 0); \
	53341	__ret_628; \
	53342	})
	53343	#endif
	53344
	53345	#ifdef __LITTLE_ENDIAN__
	53346	#define vqdmlsl_laneq_s16(__p0_629, __p1_629, __p2_629, __p3_629) __extension__ ({ \
	53347	int32x4_t __ret_629; \
	53348	int32x4_t __s0_629 = __p0_629; \
	53349	int16x4_t __s1_629 = __p1_629; \
	53350	int16x8_t __s2_629 = __p2_629; \
	53351	__ret_629 = vqdmlsl_s16(__s0_629, __s1_629, splat_laneq_s16(__s2_629, __p3_629)); \
	53352	__ret_629; \
	53353	})
	53354	#else
	53355	#define vqdmlsl_laneq_s16(__p0_630, __p1_630, __p2_630, __p3_630) __extension__ ({ \
	53356	int32x4_t __ret_630; \
	53357	int32x4_t __s0_630 = __p0_630; \
	53358	int16x4_t __s1_630 = __p1_630; \
	53359	int16x8_t __s2_630 = __p2_630; \
	53360	int32x4_t __rev0_630; __rev0_630 = __builtin_shufflevector(__s0_630, __s0_630, 3, 2, 1, 0); \
	53361	int16x4_t __rev1_630; __rev1_630 = __builtin_shufflevector(__s1_630, __s1_630, 3, 2, 1, 0); \
	53362	int16x8_t __rev2_630; __rev2_630 = __builtin_shufflevector(__s2_630, __s2_630, 7, 6, 5, 4, 3, 2, 1, 0); \
	53363	__ret_630 = __noswap_vqdmlsl_s16(__rev0_630, __rev1_630, __noswap_splat_laneq_s16(__rev2_630, __p3_630)); \
	53364	__ret_630 = __builtin_shufflevector(__ret_630, __ret_630, 3, 2, 1, 0); \
	53365	__ret_630; \
	53366	})
	53367	#endif
	53368
	53369	__ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
	53370	int32_t __ret;
	53371	__ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
	53372	return __ret;
	53373	}
	53374	__ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
	53375	int16_t __ret;
	53376	__ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
	53377	return __ret;
	53378	}
	53379	#ifdef __LITTLE_ENDIAN__
	53380	#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	53381	int32x4_t __ret; \
	53382	int32x4_t __s0 = __p0; \
	53383	int32x2_t __s1 = __p1; \
	53384	__ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
51954	53385	__ret; \
51955	53386	})
51956	53387	#else
51957		#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
51958		uint32_t __ret; \
51959		uint32x4_t __s0 = __p0; \
51960		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51961		__ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
	53388	#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	53389	int32x4_t __ret; \
	53390	int32x4_t __s0 = __p0; \
	53391	int32x2_t __s1 = __p1; \
	53392	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	53393	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	53394	__ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
	53395	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
51962	53396	__ret; \
51963	53397	})
51964	53398	#endif
51965	53399
51966	53400	#ifdef __LITTLE_ENDIAN__
51967		#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
51968		uint64_t __ret; \
51969		uint64x2_t __s0 = __p0; \
51970		__ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
	53401	#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	53402	int16x8_t __ret; \
	53403	int16x8_t __s0 = __p0; \
	53404	int16x4_t __s1 = __p1; \
	53405	__ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
51971	53406	__ret; \
51972	53407	})
51973	53408	#else
51974		#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
51975		uint64_t __ret; \
51976		uint64x2_t __s0 = __p0; \
51977		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51978		__ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
	53409	#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	53410	int16x8_t __ret; \
	53411	int16x8_t __s0 = __p0; \
	53412	int16x4_t __s1 = __p1; \
	53413	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	53414	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	53415	__ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
	53416	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
51979	53417	__ret; \
51980	53418	})
51981	53419	#endif
51982	53420
51983	53421	#ifdef __LITTLE_ENDIAN__
51984		#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
51985		uint16_t __ret; \
51986		uint16x8_t __s0 = __p0; \
51987		__ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
	53422	#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	53423	int32x2_t __ret; \
	53424	int32x2_t __s0 = __p0; \
	53425	int32x2_t __s1 = __p1; \
	53426	__ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
51988	53427	__ret; \
51989	53428	})
51990	53429	#else
51991		#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
51992		uint16_t __ret; \
51993		uint16x8_t __s0 = __p0; \
51994		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51995		__ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
	53430	#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	53431	int32x2_t __ret; \
	53432	int32x2_t __s0 = __p0; \
	53433	int32x2_t __s1 = __p1; \
	53434	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	53435	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	53436	__ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
	53437	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
51996	53438	__ret; \
51997	53439	})
51998	53440	#endif
51999	53441
52000	53442	#ifdef __LITTLE_ENDIAN__
52001		#define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
52002		int8_t __ret; \
52003		int8x16_t __s0 = __p0; \
52004		__ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
	53443	#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	53444	int16x4_t __ret; \
	53445	int16x4_t __s0 = __p0; \
	53446	int16x4_t __s1 = __p1; \
	53447	__ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
52005	53448	__ret; \
52006	53449	})
52007	53450	#else
52008		#define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
52009		int8_t __ret; \
52010		int8x16_t __s0 = __p0; \
52011		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52012		__ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
	53451	#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	53452	int16x4_t __ret; \
	53453	int16x4_t __s0 = __p0; \
	53454	int16x4_t __s1 = __p1; \
	53455	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	53456	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	53457	__ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
	53458	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
52013	53459	__ret; \
52014	53460	})
52015	53461	#endif
52016	53462
52017	53463	#ifdef __LITTLE_ENDIAN__
52018		#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
52019		float64_t __ret; \
52020		float64x2_t __s0 = __p0; \
52021		__ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__s0, __p1); \
52022		__ret; \
	53464	#define vqdmulhs_lane_s32(__p0_631, __p1_631, __p2_631) __extension__ ({ \
	53465	int32_t __ret_631; \
	53466	int32_t __s0_631 = __p0_631; \
	53467	int32x2_t __s1_631 = __p1_631; \
	53468	__ret_631 = vqdmulhs_s32(__s0_631, vget_lane_s32(__s1_631, __p2_631)); \
	53469	__ret_631; \
	53470	})
	53471	#else
	53472	#define vqdmulhs_lane_s32(__p0_632, __p1_632, __p2_632) __extension__ ({ \
	53473	int32_t __ret_632; \
	53474	int32_t __s0_632 = __p0_632; \
	53475	int32x2_t __s1_632 = __p1_632; \
	53476	int32x2_t __rev1_632; __rev1_632 = __builtin_shufflevector(__s1_632, __s1_632, 1, 0); \
	53477	__ret_632 = vqdmulhs_s32(__s0_632, __noswap_vget_lane_s32(__rev1_632, __p2_632)); \
	53478	__ret_632; \
	53479	})
	53480	#endif
	53481
	53482	#ifdef __LITTLE_ENDIAN__
	53483	#define vqdmulhh_lane_s16(__p0_633, __p1_633, __p2_633) __extension__ ({ \
	53484	int16_t __ret_633; \
	53485	int16_t __s0_633 = __p0_633; \
	53486	int16x4_t __s1_633 = __p1_633; \
	53487	__ret_633 = vqdmulhh_s16(__s0_633, vget_lane_s16(__s1_633, __p2_633)); \
	53488	__ret_633; \
	53489	})
	53490	#else
	53491	#define vqdmulhh_lane_s16(__p0_634, __p1_634, __p2_634) __extension__ ({ \
	53492	int16_t __ret_634; \
	53493	int16_t __s0_634 = __p0_634; \
	53494	int16x4_t __s1_634 = __p1_634; \
	53495	int16x4_t __rev1_634; __rev1_634 = __builtin_shufflevector(__s1_634, __s1_634, 3, 2, 1, 0); \
	53496	__ret_634 = vqdmulhh_s16(__s0_634, __noswap_vget_lane_s16(__rev1_634, __p2_634)); \
	53497	__ret_634; \
	53498	})
	53499	#endif
	53500
	53501	#ifdef __LITTLE_ENDIAN__
	53502	#define vqdmulhs_laneq_s32(__p0_635, __p1_635, __p2_635) __extension__ ({ \
	53503	int32_t __ret_635; \
	53504	int32_t __s0_635 = __p0_635; \
	53505	int32x4_t __s1_635 = __p1_635; \
	53506	__ret_635 = vqdmulhs_s32(__s0_635, vgetq_lane_s32(__s1_635, __p2_635)); \
	53507	__ret_635; \
52023	53508	})
52024	53509	#else
52025		#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
52026		float64_t __ret; \
52027		float64x2_t __s0 = __p0; \
52028		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52029		__ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__rev0, __p1); \
52030		__ret; \
	53510	#define vqdmulhs_laneq_s32(__p0_636, __p1_636, __p2_636) __extension__ ({ \
	53511	int32_t __ret_636; \
	53512	int32_t __s0_636 = __p0_636; \
	53513	int32x4_t __s1_636 = __p1_636; \
	53514	int32x4_t __rev1_636; __rev1_636 = __builtin_shufflevector(__s1_636, __s1_636, 3, 2, 1, 0); \
	53515	__ret_636 = vqdmulhs_s32(__s0_636, __noswap_vgetq_lane_s32(__rev1_636, __p2_636)); \
	53516	__ret_636; \
52031	53517	})
52032	53518	#endif
52033	53519
52034	53520	#ifdef __LITTLE_ENDIAN__
52035		#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
52036		float32_t __ret; \
52037		float32x4_t __s0 = __p0; \
52038		__ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__s0, __p1); \
52039		__ret; \
	53521	#define vqdmulhh_laneq_s16(__p0_637, __p1_637, __p2_637) __extension__ ({ \
	53522	int16_t __ret_637; \
	53523	int16_t __s0_637 = __p0_637; \
	53524	int16x8_t __s1_637 = __p1_637; \
	53525	__ret_637 = vqdmulhh_s16(__s0_637, vgetq_lane_s16(__s1_637, __p2_637)); \
	53526	__ret_637; \
52040	53527	})
52041	53528	#else
52042		#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
52043		float32_t __ret; \
52044		float32x4_t __s0 = __p0; \
52045		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
52046		__ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__rev0, __p1); \
52047		__ret; \
	53529	#define vqdmulhh_laneq_s16(__p0_638, __p1_638, __p2_638) __extension__ ({ \
	53530	int16_t __ret_638; \
	53531	int16_t __s0_638 = __p0_638; \
	53532	int16x8_t __s1_638 = __p1_638; \
	53533	int16x8_t __rev1_638; __rev1_638 = __builtin_shufflevector(__s1_638, __s1_638, 7, 6, 5, 4, 3, 2, 1, 0); \
	53534	__ret_638 = vqdmulhh_s16(__s0_638, __noswap_vgetq_lane_s16(__rev1_638, __p2_638)); \
	53535	__ret_638; \
52048	53536	})
52049	53537	#endif
52050	53538
52051	53539	#ifdef __LITTLE_ENDIAN__
52052		#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
52053		int32_t __ret; \
	53540	#define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	53541	int32x4_t __ret; \
52054	53542	int32x4_t __s0 = __p0; \
52055		__ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
	53543	int32x4_t __s1 = __p1; \
	53544	__ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
52056	53545	__ret; \
52057	53546	})
52058	53547	#else
52059		#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
52060		int32_t __ret; \
	53548	#define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	53549	int32x4_t __ret; \
52061	53550	int32x4_t __s0 = __p0; \
	53551	int32x4_t __s1 = __p1; \
52062	53552	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
52063		__ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
	53553	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	53554	__ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
	53555	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
52064	53556	__ret; \
52065	53557	})
52066	53558	#endif
52067	53559
52068	53560	#ifdef __LITTLE_ENDIAN__
52069		#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
52070		int64_t __ret; \
52071		int64x2_t __s0 = __p0; \
52072		__ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
	53561	#define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	53562	int16x8_t __ret; \
	53563	int16x8_t __s0 = __p0; \
	53564	int16x8_t __s1 = __p1; \
	53565	__ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
52073	53566	__ret; \
52074	53567	})
52075	53568	#else
52076		#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
52077		int64_t __ret; \
52078		int64x2_t __s0 = __p0; \
52079		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52080		__ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
	53569	#define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	53570	int16x8_t __ret; \
	53571	int16x8_t __s0 = __p0; \
	53572	int16x8_t __s1 = __p1; \
	53573	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	53574	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	53575	__ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
	53576	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
52081	53577	__ret; \
52082	53578	})
52083	53579	#endif
52084	53580
52085	53581	#ifdef __LITTLE_ENDIAN__
52086		#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
52087		int16_t __ret; \
52088		int16x8_t __s0 = __p0; \
52089		__ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
	53582	#define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	53583	int32x2_t __ret; \
	53584	int32x2_t __s0 = __p0; \
	53585	int32x4_t __s1 = __p1; \
	53586	__ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
52090	53587	__ret; \
52091	53588	})
52092	53589	#else
52093		#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
52094		int16_t __ret; \
52095		int16x8_t __s0 = __p0; \
52096		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
52097		__ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
	53590	#define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	53591	int32x2_t __ret; \
	53592	int32x2_t __s0 = __p0; \
	53593	int32x4_t __s1 = __p1; \
	53594	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	53595	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	53596	__ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
	53597	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
52098	53598	__ret; \
52099	53599	})
52100	53600	#endif
52101	53601
52102	53602	#ifdef __LITTLE_ENDIAN__
52103		#define vdup_laneq_p8(__p0_453, __p1_453) __extension__ ({ \
52104		poly8x8_t __ret_453; \
52105		poly8x16_t __s0_453 = __p0_453; \
52106		__ret_453 = splat_laneq_p8(__s0_453, __p1_453); \
52107		__ret_453; \
	53603	#define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	53604	int16x4_t __ret; \
	53605	int16x4_t __s0 = __p0; \
	53606	int16x8_t __s1 = __p1; \
	53607	__ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
	53608	__ret; \
52108	53609	})
52109	53610	#else
52110		#define vdup_laneq_p8(__p0_454, __p1_454) __extension__ ({ \
52111		poly8x8_t __ret_454; \
52112		poly8x16_t __s0_454 = __p0_454; \
52113		poly8x16_t __rev0_454; __rev0_454 = __builtin_shufflevector(__s0_454, __s0_454, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52114		__ret_454 = __noswap_splat_laneq_p8(__rev0_454, __p1_454); \
52115		__ret_454 = __builtin_shufflevector(__ret_454, __ret_454, 7, 6, 5, 4, 3, 2, 1, 0); \
52116		__ret_454; \
	53611	#define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	53612	int16x4_t __ret; \
	53613	int16x4_t __s0 = __p0; \
	53614	int16x8_t __s1 = __p1; \
	53615	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	53616	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	53617	__ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
	53618	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	53619	__ret; \
52117	53620	})
52118	53621	#endif
52119	53622
	53623	__ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
	53624	int64_t __ret;
	53625	__ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
	53626	return __ret;
	53627	}
	53628	__ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
	53629	int32_t __ret;
	53630	__ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
	53631	return __ret;
	53632	}
52120	53633	#ifdef __LITTLE_ENDIAN__
52121		#define vdup_laneq_p64(__p0_455, __p1_455) __extension__ ({ \
52122		poly64x1_t __ret_455; \
52123		poly64x2_t __s0_455 = __p0_455; \
52124		__ret_455 = splat_laneq_p64(__s0_455, __p1_455); \
52125		__ret_455; \
52126		})
	53634	__ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
	53635	int64x2_t __ret;
	53636	__ret = vqdmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
	53637	return __ret;
	53638	}
52127	53639	#else
52128		#define vdup_laneq_p64(__p0_456, __p1_456) __extension__ ({ \
52129		poly64x1_t __ret_456; \
52130		poly64x2_t __s0_456 = __p0_456; \
52131		poly64x2_t __rev0_456; __rev0_456 = __builtin_shufflevector(__s0_456, __s0_456, 1, 0); \
52132		__ret_456 = __noswap_splat_laneq_p64(__rev0_456, __p1_456); \
52133		__ret_456; \
52134		})
	53640	__ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
	53641	int64x2_t __ret;
	53642	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	53643	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	53644	__ret = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
	53645	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	53646	return __ret;
	53647	}
52135	53648	#endif
52136	53649
52137	53650	#ifdef __LITTLE_ENDIAN__
52138		#define vdup_laneq_p16(__p0_457, __p1_457) __extension__ ({ \
52139		poly16x4_t __ret_457; \
52140		poly16x8_t __s0_457 = __p0_457; \
52141		__ret_457 = splat_laneq_p16(__s0_457, __p1_457); \
52142		__ret_457; \
52143		})
	53651	__ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
	53652	int32x4_t __ret;
	53653	__ret = vqdmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
	53654	return __ret;
	53655	}
52144	53656	#else
52145		#define vdup_laneq_p16(__p0_458, __p1_458) __extension__ ({ \
52146		poly16x4_t __ret_458; \
52147		poly16x8_t __s0_458 = __p0_458; \
52148		poly16x8_t __rev0_458; __rev0_458 = __builtin_shufflevector(__s0_458, __s0_458, 7, 6, 5, 4, 3, 2, 1, 0); \
52149		__ret_458 = __noswap_splat_laneq_p16(__rev0_458, __p1_458); \
52150		__ret_458 = __builtin_shufflevector(__ret_458, __ret_458, 3, 2, 1, 0); \
52151		__ret_458; \
52152		})
	53657	__ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
	53658	int32x4_t __ret;
	53659	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	53660	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	53661	__ret = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
	53662	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	53663	return __ret;
	53664	}
52153	53665	#endif
52154	53666
52155	53667	#ifdef __LITTLE_ENDIAN__
52156		#define vdupq_laneq_p8(__p0_459, __p1_459) __extension__ ({ \
52157		poly8x16_t __ret_459; \
52158		poly8x16_t __s0_459 = __p0_459; \
52159		__ret_459 = splatq_laneq_p8(__s0_459, __p1_459); \
52160		__ret_459; \
	53668	#define vqdmull_high_lane_s32(__p0_639, __p1_639, __p2_639) __extension__ ({ \
	53669	int64x2_t __ret_639; \
	53670	int32x4_t __s0_639 = __p0_639; \
	53671	int32x2_t __s1_639 = __p1_639; \
	53672	__ret_639 = vqdmull_s32(vget_high_s32(__s0_639), splat_lane_s32(__s1_639, __p2_639)); \
	53673	__ret_639; \
52161	53674	})
52162	53675	#else
52163		#define vdupq_laneq_p8(__p0_460, __p1_460) __extension__ ({ \
52164		poly8x16_t __ret_460; \
52165		poly8x16_t __s0_460 = __p0_460; \
52166		poly8x16_t __rev0_460; __rev0_460 = __builtin_shufflevector(__s0_460, __s0_460, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52167		__ret_460 = __noswap_splatq_laneq_p8(__rev0_460, __p1_460); \
52168		__ret_460 = __builtin_shufflevector(__ret_460, __ret_460, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52169		__ret_460; \
	53676	#define vqdmull_high_lane_s32(__p0_640, __p1_640, __p2_640) __extension__ ({ \
	53677	int64x2_t __ret_640; \
	53678	int32x4_t __s0_640 = __p0_640; \
	53679	int32x2_t __s1_640 = __p1_640; \
	53680	int32x4_t __rev0_640; __rev0_640 = __builtin_shufflevector(__s0_640, __s0_640, 3, 2, 1, 0); \
	53681	int32x2_t __rev1_640; __rev1_640 = __builtin_shufflevector(__s1_640, __s1_640, 1, 0); \
	53682	__ret_640 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_640), __noswap_splat_lane_s32(__rev1_640, __p2_640)); \
	53683	__ret_640 = __builtin_shufflevector(__ret_640, __ret_640, 1, 0); \
	53684	__ret_640; \
52170	53685	})
52171	53686	#endif
52172	53687
52173	53688	#ifdef __LITTLE_ENDIAN__
52174		#define vdupq_laneq_p64(__p0_461, __p1_461) __extension__ ({ \
52175		poly64x2_t __ret_461; \
52176		poly64x2_t __s0_461 = __p0_461; \
52177		__ret_461 = splatq_laneq_p64(__s0_461, __p1_461); \
52178		__ret_461; \
	53689	#define vqdmull_high_lane_s16(__p0_641, __p1_641, __p2_641) __extension__ ({ \
	53690	int32x4_t __ret_641; \
	53691	int16x8_t __s0_641 = __p0_641; \
	53692	int16x4_t __s1_641 = __p1_641; \
	53693	__ret_641 = vqdmull_s16(vget_high_s16(__s0_641), splat_lane_s16(__s1_641, __p2_641)); \
	53694	__ret_641; \
52179	53695	})
52180	53696	#else
52181		#define vdupq_laneq_p64(__p0_462, __p1_462) __extension__ ({ \
52182		poly64x2_t __ret_462; \
52183		poly64x2_t __s0_462 = __p0_462; \
52184		poly64x2_t __rev0_462; __rev0_462 = __builtin_shufflevector(__s0_462, __s0_462, 1, 0); \
52185		__ret_462 = __noswap_splatq_laneq_p64(__rev0_462, __p1_462); \
52186		__ret_462 = __builtin_shufflevector(__ret_462, __ret_462, 1, 0); \
52187		__ret_462; \
	53697	#define vqdmull_high_lane_s16(__p0_642, __p1_642, __p2_642) __extension__ ({ \
	53698	int32x4_t __ret_642; \
	53699	int16x8_t __s0_642 = __p0_642; \
	53700	int16x4_t __s1_642 = __p1_642; \
	53701	int16x8_t __rev0_642; __rev0_642 = __builtin_shufflevector(__s0_642, __s0_642, 7, 6, 5, 4, 3, 2, 1, 0); \
	53702	int16x4_t __rev1_642; __rev1_642 = __builtin_shufflevector(__s1_642, __s1_642, 3, 2, 1, 0); \
	53703	__ret_642 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_642), __noswap_splat_lane_s16(__rev1_642, __p2_642)); \
	53704	__ret_642 = __builtin_shufflevector(__ret_642, __ret_642, 3, 2, 1, 0); \
	53705	__ret_642; \
52188	53706	})
52189	53707	#endif
52190	53708
52191	53709	#ifdef __LITTLE_ENDIAN__
52192		#define vdupq_laneq_p16(__p0_463, __p1_463) __extension__ ({ \
52193		poly16x8_t __ret_463; \
52194		poly16x8_t __s0_463 = __p0_463; \
52195		__ret_463 = splatq_laneq_p16(__s0_463, __p1_463); \
52196		__ret_463; \
	53710	#define vqdmull_high_laneq_s32(__p0_643, __p1_643, __p2_643) __extension__ ({ \
	53711	int64x2_t __ret_643; \
	53712	int32x4_t __s0_643 = __p0_643; \
	53713	int32x4_t __s1_643 = __p1_643; \
	53714	__ret_643 = vqdmull_s32(vget_high_s32(__s0_643), splat_laneq_s32(__s1_643, __p2_643)); \
	53715	__ret_643; \
52197	53716	})
52198	53717	#else
52199		#define vdupq_laneq_p16(__p0_464, __p1_464) __extension__ ({ \
52200		poly16x8_t __ret_464; \
52201		poly16x8_t __s0_464 = __p0_464; \
52202		poly16x8_t __rev0_464; __rev0_464 = __builtin_shufflevector(__s0_464, __s0_464, 7, 6, 5, 4, 3, 2, 1, 0); \
52203		__ret_464 = __noswap_splatq_laneq_p16(__rev0_464, __p1_464); \
52204		__ret_464 = __builtin_shufflevector(__ret_464, __ret_464, 7, 6, 5, 4, 3, 2, 1, 0); \
52205		__ret_464; \
	53718	#define vqdmull_high_laneq_s32(__p0_644, __p1_644, __p2_644) __extension__ ({ \
	53719	int64x2_t __ret_644; \
	53720	int32x4_t __s0_644 = __p0_644; \
	53721	int32x4_t __s1_644 = __p1_644; \
	53722	int32x4_t __rev0_644; __rev0_644 = __builtin_shufflevector(__s0_644, __s0_644, 3, 2, 1, 0); \
	53723	int32x4_t __rev1_644; __rev1_644 = __builtin_shufflevector(__s1_644, __s1_644, 3, 2, 1, 0); \
	53724	__ret_644 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_644), __noswap_splat_laneq_s32(__rev1_644, __p2_644)); \
	53725	__ret_644 = __builtin_shufflevector(__ret_644, __ret_644, 1, 0); \
	53726	__ret_644; \
52206	53727	})
52207	53728	#endif
52208	53729
52209	53730	#ifdef __LITTLE_ENDIAN__
52210		#define vdupq_laneq_u8(__p0_465, __p1_465) __extension__ ({ \
52211		uint8x16_t __ret_465; \
52212		uint8x16_t __s0_465 = __p0_465; \
52213		__ret_465 = splatq_laneq_u8(__s0_465, __p1_465); \
52214		__ret_465; \
	53731	#define vqdmull_high_laneq_s16(__p0_645, __p1_645, __p2_645) __extension__ ({ \
	53732	int32x4_t __ret_645; \
	53733	int16x8_t __s0_645 = __p0_645; \
	53734	int16x8_t __s1_645 = __p1_645; \
	53735	__ret_645 = vqdmull_s16(vget_high_s16(__s0_645), splat_laneq_s16(__s1_645, __p2_645)); \
	53736	__ret_645; \
52215	53737	})
52216	53738	#else
52217		#define vdupq_laneq_u8(__p0_466, __p1_466) __extension__ ({ \
52218		uint8x16_t __ret_466; \
52219		uint8x16_t __s0_466 = __p0_466; \
52220		uint8x16_t __rev0_466; __rev0_466 = __builtin_shufflevector(__s0_466, __s0_466, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52221		__ret_466 = __noswap_splatq_laneq_u8(__rev0_466, __p1_466); \
52222		__ret_466 = __builtin_shufflevector(__ret_466, __ret_466, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52223		__ret_466; \
	53739	#define vqdmull_high_laneq_s16(__p0_646, __p1_646, __p2_646) __extension__ ({ \
	53740	int32x4_t __ret_646; \
	53741	int16x8_t __s0_646 = __p0_646; \
	53742	int16x8_t __s1_646 = __p1_646; \
	53743	int16x8_t __rev0_646; __rev0_646 = __builtin_shufflevector(__s0_646, __s0_646, 7, 6, 5, 4, 3, 2, 1, 0); \
	53744	int16x8_t __rev1_646; __rev1_646 = __builtin_shufflevector(__s1_646, __s1_646, 7, 6, 5, 4, 3, 2, 1, 0); \
	53745	__ret_646 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_646), __noswap_splat_laneq_s16(__rev1_646, __p2_646)); \
	53746	__ret_646 = __builtin_shufflevector(__ret_646, __ret_646, 3, 2, 1, 0); \
	53747	__ret_646; \
52224	53748	})
52225	53749	#endif
52226	53750
52227	53751	#ifdef __LITTLE_ENDIAN__
52228		#define vdupq_laneq_u32(__p0_467, __p1_467) __extension__ ({ \
52229		uint32x4_t __ret_467; \
52230		uint32x4_t __s0_467 = __p0_467; \
52231		__ret_467 = splatq_laneq_u32(__s0_467, __p1_467); \
52232		__ret_467; \
52233		})
	53752	__ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
	53753	int64x2_t __ret;
	53754	__ret = vqdmull_n_s32(vget_high_s32(__p0), __p1);
	53755	return __ret;
	53756	}
52234	53757	#else
52235		#define vdupq_laneq_u32(__p0_468, __p1_468) __extension__ ({ \
52236		uint32x4_t __ret_468; \
52237		uint32x4_t __s0_468 = __p0_468; \
52238		uint32x4_t __rev0_468; __rev0_468 = __builtin_shufflevector(__s0_468, __s0_468, 3, 2, 1, 0); \
52239		__ret_468 = __noswap_splatq_laneq_u32(__rev0_468, __p1_468); \
52240		__ret_468 = __builtin_shufflevector(__ret_468, __ret_468, 3, 2, 1, 0); \
52241		__ret_468; \
52242		})
	53758	__ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
	53759	int64x2_t __ret;
	53760	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	53761	__ret = __noswap_vqdmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
	53762	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	53763	return __ret;
	53764	}
52243	53765	#endif
52244	53766
52245	53767	#ifdef __LITTLE_ENDIAN__
52246		#define vdupq_laneq_u64(__p0_469, __p1_469) __extension__ ({ \
52247		uint64x2_t __ret_469; \
52248		uint64x2_t __s0_469 = __p0_469; \
52249		__ret_469 = splatq_laneq_u64(__s0_469, __p1_469); \
52250		__ret_469; \
52251		})
	53768	__ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
	53769	int32x4_t __ret;
	53770	__ret = vqdmull_n_s16(vget_high_s16(__p0), __p1);
	53771	return __ret;
	53772	}
52252	53773	#else
52253		#define vdupq_laneq_u64(__p0_470, __p1_470) __extension__ ({ \
52254		uint64x2_t __ret_470; \
52255		uint64x2_t __s0_470 = __p0_470; \
52256		uint64x2_t __rev0_470; __rev0_470 = __builtin_shufflevector(__s0_470, __s0_470, 1, 0); \
52257		__ret_470 = __noswap_splatq_laneq_u64(__rev0_470, __p1_470); \
52258		__ret_470 = __builtin_shufflevector(__ret_470, __ret_470, 1, 0); \
52259		__ret_470; \
52260		})
	53774	__ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
	53775	int32x4_t __ret;
	53776	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	53777	__ret = __noswap_vqdmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
	53778	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	53779	return __ret;
	53780	}
52261	53781	#endif
52262	53782
52263	53783	#ifdef __LITTLE_ENDIAN__
52264		#define vdupq_laneq_u16(__p0_471, __p1_471) __extension__ ({ \
52265		uint16x8_t __ret_471; \
52266		uint16x8_t __s0_471 = __p0_471; \
52267		__ret_471 = splatq_laneq_u16(__s0_471, __p1_471); \
52268		__ret_471; \
	53784	#define vqdmulls_lane_s32(__p0_647, __p1_647, __p2_647) __extension__ ({ \
	53785	int64_t __ret_647; \
	53786	int32_t __s0_647 = __p0_647; \
	53787	int32x2_t __s1_647 = __p1_647; \
	53788	__ret_647 = vqdmulls_s32(__s0_647, vget_lane_s32(__s1_647, __p2_647)); \
	53789	__ret_647; \
52269	53790	})
52270	53791	#else
52271		#define vdupq_laneq_u16(__p0_472, __p1_472) __extension__ ({ \
52272		uint16x8_t __ret_472; \
52273		uint16x8_t __s0_472 = __p0_472; \
52274		uint16x8_t __rev0_472; __rev0_472 = __builtin_shufflevector(__s0_472, __s0_472, 7, 6, 5, 4, 3, 2, 1, 0); \
52275		__ret_472 = __noswap_splatq_laneq_u16(__rev0_472, __p1_472); \
52276		__ret_472 = __builtin_shufflevector(__ret_472, __ret_472, 7, 6, 5, 4, 3, 2, 1, 0); \
52277		__ret_472; \
	53792	#define vqdmulls_lane_s32(__p0_648, __p1_648, __p2_648) __extension__ ({ \
	53793	int64_t __ret_648; \
	53794	int32_t __s0_648 = __p0_648; \
	53795	int32x2_t __s1_648 = __p1_648; \
	53796	int32x2_t __rev1_648; __rev1_648 = __builtin_shufflevector(__s1_648, __s1_648, 1, 0); \
	53797	__ret_648 = vqdmulls_s32(__s0_648, __noswap_vget_lane_s32(__rev1_648, __p2_648)); \
	53798	__ret_648; \
52278	53799	})
52279	53800	#endif
52280	53801
52281	53802	#ifdef __LITTLE_ENDIAN__
52282		#define vdupq_laneq_s8(__p0_473, __p1_473) __extension__ ({ \
52283		int8x16_t __ret_473; \
52284		int8x16_t __s0_473 = __p0_473; \
52285		__ret_473 = splatq_laneq_s8(__s0_473, __p1_473); \
52286		__ret_473; \
	53803	#define vqdmullh_lane_s16(__p0_649, __p1_649, __p2_649) __extension__ ({ \
	53804	int32_t __ret_649; \
	53805	int16_t __s0_649 = __p0_649; \
	53806	int16x4_t __s1_649 = __p1_649; \
	53807	__ret_649 = vqdmullh_s16(__s0_649, vget_lane_s16(__s1_649, __p2_649)); \
	53808	__ret_649; \
52287	53809	})
52288	53810	#else
52289		#define vdupq_laneq_s8(__p0_474, __p1_474) __extension__ ({ \
52290		int8x16_t __ret_474; \
52291		int8x16_t __s0_474 = __p0_474; \
52292		int8x16_t __rev0_474; __rev0_474 = __builtin_shufflevector(__s0_474, __s0_474, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52293		__ret_474 = __noswap_splatq_laneq_s8(__rev0_474, __p1_474); \
52294		__ret_474 = __builtin_shufflevector(__ret_474, __ret_474, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52295		__ret_474; \
	53811	#define vqdmullh_lane_s16(__p0_650, __p1_650, __p2_650) __extension__ ({ \
	53812	int32_t __ret_650; \
	53813	int16_t __s0_650 = __p0_650; \
	53814	int16x4_t __s1_650 = __p1_650; \
	53815	int16x4_t __rev1_650; __rev1_650 = __builtin_shufflevector(__s1_650, __s1_650, 3, 2, 1, 0); \
	53816	__ret_650 = vqdmullh_s16(__s0_650, __noswap_vget_lane_s16(__rev1_650, __p2_650)); \
	53817	__ret_650; \
52296	53818	})
52297	53819	#endif
52298	53820
52299	53821	#ifdef __LITTLE_ENDIAN__
52300		#define vdupq_laneq_f64(__p0_475, __p1_475) __extension__ ({ \
52301		float64x2_t __ret_475; \
52302		float64x2_t __s0_475 = __p0_475; \
52303		__ret_475 = splatq_laneq_f64(__s0_475, __p1_475); \
52304		__ret_475; \
	53822	#define vqdmulls_laneq_s32(__p0_651, __p1_651, __p2_651) __extension__ ({ \
	53823	int64_t __ret_651; \
	53824	int32_t __s0_651 = __p0_651; \
	53825	int32x4_t __s1_651 = __p1_651; \
	53826	__ret_651 = vqdmulls_s32(__s0_651, vgetq_lane_s32(__s1_651, __p2_651)); \
	53827	__ret_651; \
52305	53828	})
52306	53829	#else
52307		#define vdupq_laneq_f64(__p0_476, __p1_476) __extension__ ({ \
52308		float64x2_t __ret_476; \
52309		float64x2_t __s0_476 = __p0_476; \
52310		float64x2_t __rev0_476; __rev0_476 = __builtin_shufflevector(__s0_476, __s0_476, 1, 0); \
52311		__ret_476 = __noswap_splatq_laneq_f64(__rev0_476, __p1_476); \
52312		__ret_476 = __builtin_shufflevector(__ret_476, __ret_476, 1, 0); \
52313		__ret_476; \
	53830	#define vqdmulls_laneq_s32(__p0_652, __p1_652, __p2_652) __extension__ ({ \
	53831	int64_t __ret_652; \
	53832	int32_t __s0_652 = __p0_652; \
	53833	int32x4_t __s1_652 = __p1_652; \
	53834	int32x4_t __rev1_652; __rev1_652 = __builtin_shufflevector(__s1_652, __s1_652, 3, 2, 1, 0); \
	53835	__ret_652 = vqdmulls_s32(__s0_652, __noswap_vgetq_lane_s32(__rev1_652, __p2_652)); \
	53836	__ret_652; \
52314	53837	})
52315	53838	#endif
52316	53839
52317	53840	#ifdef __LITTLE_ENDIAN__
52318		#define vdupq_laneq_f32(__p0_477, __p1_477) __extension__ ({ \
52319		float32x4_t __ret_477; \
52320		float32x4_t __s0_477 = __p0_477; \
52321		__ret_477 = splatq_laneq_f32(__s0_477, __p1_477); \
52322		__ret_477; \
	53841	#define vqdmullh_laneq_s16(__p0_653, __p1_653, __p2_653) __extension__ ({ \
	53842	int32_t __ret_653; \
	53843	int16_t __s0_653 = __p0_653; \
	53844	int16x8_t __s1_653 = __p1_653; \
	53845	__ret_653 = vqdmullh_s16(__s0_653, vgetq_lane_s16(__s1_653, __p2_653)); \
	53846	__ret_653; \
52323	53847	})
52324	53848	#else
52325		#define vdupq_laneq_f32(__p0_478, __p1_478) __extension__ ({ \
52326		float32x4_t __ret_478; \
52327		float32x4_t __s0_478 = __p0_478; \
52328		float32x4_t __rev0_478; __rev0_478 = __builtin_shufflevector(__s0_478, __s0_478, 3, 2, 1, 0); \
52329		__ret_478 = __noswap_splatq_laneq_f32(__rev0_478, __p1_478); \
52330		__ret_478 = __builtin_shufflevector(__ret_478, __ret_478, 3, 2, 1, 0); \
52331		__ret_478; \
	53849	#define vqdmullh_laneq_s16(__p0_654, __p1_654, __p2_654) __extension__ ({ \
	53850	int32_t __ret_654; \
	53851	int16_t __s0_654 = __p0_654; \
	53852	int16x8_t __s1_654 = __p1_654; \
	53853	int16x8_t __rev1_654; __rev1_654 = __builtin_shufflevector(__s1_654, __s1_654, 7, 6, 5, 4, 3, 2, 1, 0); \
	53854	__ret_654 = vqdmullh_s16(__s0_654, __noswap_vgetq_lane_s16(__rev1_654, __p2_654)); \
	53855	__ret_654; \
52332	53856	})
52333	53857	#endif
52334	53858
52335	53859	#ifdef __LITTLE_ENDIAN__
52336		#define vdupq_laneq_f16(__p0_479, __p1_479) __extension__ ({ \
52337		float16x8_t __ret_479; \
52338		float16x8_t __s0_479 = __p0_479; \
52339		__ret_479 = splatq_laneq_f16(__s0_479, __p1_479); \
52340		__ret_479; \
	53860	#define vqdmull_laneq_s32(__p0_655, __p1_655, __p2_655) __extension__ ({ \
	53861	int64x2_t __ret_655; \
	53862	int32x2_t __s0_655 = __p0_655; \
	53863	int32x4_t __s1_655 = __p1_655; \
	53864	__ret_655 = vqdmull_s32(__s0_655, splat_laneq_s32(__s1_655, __p2_655)); \
	53865	__ret_655; \
52341	53866	})
52342	53867	#else
52343		#define vdupq_laneq_f16(__p0_480, __p1_480) __extension__ ({ \
52344		float16x8_t __ret_480; \
52345		float16x8_t __s0_480 = __p0_480; \
52346		float16x8_t __rev0_480; __rev0_480 = __builtin_shufflevector(__s0_480, __s0_480, 7, 6, 5, 4, 3, 2, 1, 0); \
52347		__ret_480 = __noswap_splatq_laneq_f16(__rev0_480, __p1_480); \
52348		__ret_480 = __builtin_shufflevector(__ret_480, __ret_480, 7, 6, 5, 4, 3, 2, 1, 0); \
52349		__ret_480; \
	53868	#define vqdmull_laneq_s32(__p0_656, __p1_656, __p2_656) __extension__ ({ \
	53869	int64x2_t __ret_656; \
	53870	int32x2_t __s0_656 = __p0_656; \
	53871	int32x4_t __s1_656 = __p1_656; \
	53872	int32x2_t __rev0_656; __rev0_656 = __builtin_shufflevector(__s0_656, __s0_656, 1, 0); \
	53873	int32x4_t __rev1_656; __rev1_656 = __builtin_shufflevector(__s1_656, __s1_656, 3, 2, 1, 0); \
	53874	__ret_656 = __noswap_vqdmull_s32(__rev0_656, __noswap_splat_laneq_s32(__rev1_656, __p2_656)); \
	53875	__ret_656 = __builtin_shufflevector(__ret_656, __ret_656, 1, 0); \
	53876	__ret_656; \
52350	53877	})
52351	53878	#endif
52352	53879
52353	53880	#ifdef __LITTLE_ENDIAN__
52354		#define vdupq_laneq_s32(__p0_481, __p1_481) __extension__ ({ \
52355		int32x4_t __ret_481; \
52356		int32x4_t __s0_481 = __p0_481; \
52357		__ret_481 = splatq_laneq_s32(__s0_481, __p1_481); \
52358		__ret_481; \
	53881	#define vqdmull_laneq_s16(__p0_657, __p1_657, __p2_657) __extension__ ({ \
	53882	int32x4_t __ret_657; \
	53883	int16x4_t __s0_657 = __p0_657; \
	53884	int16x8_t __s1_657 = __p1_657; \
	53885	__ret_657 = vqdmull_s16(__s0_657, splat_laneq_s16(__s1_657, __p2_657)); \
	53886	__ret_657; \
52359	53887	})
52360	53888	#else
52361		#define vdupq_laneq_s32(__p0_482, __p1_482) __extension__ ({ \
52362		int32x4_t __ret_482; \
52363		int32x4_t __s0_482 = __p0_482; \
52364		int32x4_t __rev0_482; __rev0_482 = __builtin_shufflevector(__s0_482, __s0_482, 3, 2, 1, 0); \
52365		__ret_482 = __noswap_splatq_laneq_s32(__rev0_482, __p1_482); \
52366		__ret_482 = __builtin_shufflevector(__ret_482, __ret_482, 3, 2, 1, 0); \
52367		__ret_482; \
	53889	#define vqdmull_laneq_s16(__p0_658, __p1_658, __p2_658) __extension__ ({ \
	53890	int32x4_t __ret_658; \
	53891	int16x4_t __s0_658 = __p0_658; \
	53892	int16x8_t __s1_658 = __p1_658; \
	53893	int16x4_t __rev0_658; __rev0_658 = __builtin_shufflevector(__s0_658, __s0_658, 3, 2, 1, 0); \
	53894	int16x8_t __rev1_658; __rev1_658 = __builtin_shufflevector(__s1_658, __s1_658, 7, 6, 5, 4, 3, 2, 1, 0); \
	53895	__ret_658 = __noswap_vqdmull_s16(__rev0_658, __noswap_splat_laneq_s16(__rev1_658, __p2_658)); \
	53896	__ret_658 = __builtin_shufflevector(__ret_658, __ret_658, 3, 2, 1, 0); \
	53897	__ret_658; \
52368	53898	})
52369	53899	#endif
52370	53900
	53901	__ai int16_t vqmovns_s32(int32_t __p0) {
	53902	int16_t __ret;
	53903	__ret = (int16_t) __builtin_neon_vqmovns_s32(__p0);
	53904	return __ret;
	53905	}
	53906	__ai int32_t vqmovnd_s64(int64_t __p0) {
	53907	int32_t __ret;
	53908	__ret = (int32_t) __builtin_neon_vqmovnd_s64(__p0);
	53909	return __ret;
	53910	}
	53911	__ai int8_t vqmovnh_s16(int16_t __p0) {
	53912	int8_t __ret;
	53913	__ret = (int8_t) __builtin_neon_vqmovnh_s16(__p0);
	53914	return __ret;
	53915	}
	53916	__ai uint16_t vqmovns_u32(uint32_t __p0) {
	53917	uint16_t __ret;
	53918	__ret = (uint16_t) __builtin_neon_vqmovns_u32(__p0);
	53919	return __ret;
	53920	}
	53921	__ai uint32_t vqmovnd_u64(uint64_t __p0) {
	53922	uint32_t __ret;
	53923	__ret = (uint32_t) __builtin_neon_vqmovnd_u64(__p0);
	53924	return __ret;
	53925	}
	53926	__ai uint8_t vqmovnh_u16(uint16_t __p0) {
	53927	uint8_t __ret;
	53928	__ret = (uint8_t) __builtin_neon_vqmovnh_u16(__p0);
	53929	return __ret;
	53930	}
52371	53931	#ifdef __LITTLE_ENDIAN__
52372		#define vdupq_laneq_s64(__p0_483, __p1_483) __extension__ ({ \
52373		int64x2_t __ret_483; \
52374		int64x2_t __s0_483 = __p0_483; \
52375		__ret_483 = splatq_laneq_s64(__s0_483, __p1_483); \
52376		__ret_483; \
52377		})
	53932	__ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
	53933	uint16x8_t __ret;
	53934	__ret = vcombine_u16(__p0, vqmovn_u32(__p1));
	53935	return __ret;
	53936	}
52378	53937	#else
52379		#define vdupq_laneq_s64(__p0_484, __p1_484) __extension__ ({ \
52380		int64x2_t __ret_484; \
52381		int64x2_t __s0_484 = __p0_484; \
52382		int64x2_t __rev0_484; __rev0_484 = __builtin_shufflevector(__s0_484, __s0_484, 1, 0); \
52383		__ret_484 = __noswap_splatq_laneq_s64(__rev0_484, __p1_484); \
52384		__ret_484 = __builtin_shufflevector(__ret_484, __ret_484, 1, 0); \
52385		__ret_484; \
52386		})
	53938	__ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
	53939	uint16x8_t __ret;
	53940	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	53941	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	53942	__ret = __noswap_vcombine_u16(__rev0, __noswap_vqmovn_u32(__rev1));
	53943	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	53944	return __ret;
	53945	}
52387	53946	#endif
52388	53947
52389	53948	#ifdef __LITTLE_ENDIAN__
52390		#define vdupq_laneq_s16(__p0_485, __p1_485) __extension__ ({ \
52391		int16x8_t __ret_485; \
52392		int16x8_t __s0_485 = __p0_485; \
52393		__ret_485 = splatq_laneq_s16(__s0_485, __p1_485); \
52394		__ret_485; \
52395		})
	53949	__ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
	53950	uint32x4_t __ret;
	53951	__ret = vcombine_u32(__p0, vqmovn_u64(__p1));
	53952	return __ret;
	53953	}
52396	53954	#else
52397		#define vdupq_laneq_s16(__p0_486, __p1_486) __extension__ ({ \
52398		int16x8_t __ret_486; \
52399		int16x8_t __s0_486 = __p0_486; \
52400		int16x8_t __rev0_486; __rev0_486 = __builtin_shufflevector(__s0_486, __s0_486, 7, 6, 5, 4, 3, 2, 1, 0); \
52401		__ret_486 = __noswap_splatq_laneq_s16(__rev0_486, __p1_486); \
52402		__ret_486 = __builtin_shufflevector(__ret_486, __ret_486, 7, 6, 5, 4, 3, 2, 1, 0); \
52403		__ret_486; \
52404		})
	53955	__ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
	53956	uint32x4_t __ret;
	53957	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	53958	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	53959	__ret = __noswap_vcombine_u32(__rev0, __noswap_vqmovn_u64(__rev1));
	53960	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	53961	return __ret;
	53962	}
52405	53963	#endif
52406	53964
52407	53965	#ifdef __LITTLE_ENDIAN__
52408		#define vdup_laneq_u8(__p0_487, __p1_487) __extension__ ({ \
52409		uint8x8_t __ret_487; \
52410		uint8x16_t __s0_487 = __p0_487; \
52411		__ret_487 = splat_laneq_u8(__s0_487, __p1_487); \
52412		__ret_487; \
52413		})
	53966	__ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
	53967	uint8x16_t __ret;
	53968	__ret = vcombine_u8(__p0, vqmovn_u16(__p1));
	53969	return __ret;
	53970	}
52414	53971	#else
52415		#define vdup_laneq_u8(__p0_488, __p1_488) __extension__ ({ \
52416		uint8x8_t __ret_488; \
52417		uint8x16_t __s0_488 = __p0_488; \
52418		uint8x16_t __rev0_488; __rev0_488 = __builtin_shufflevector(__s0_488, __s0_488, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52419		__ret_488 = __noswap_splat_laneq_u8(__rev0_488, __p1_488); \
52420		__ret_488 = __builtin_shufflevector(__ret_488, __ret_488, 7, 6, 5, 4, 3, 2, 1, 0); \
52421		__ret_488; \
52422		})
	53972	__ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
	53973	uint8x16_t __ret;
	53974	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	53975	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	53976	__ret = __noswap_vcombine_u8(__rev0, __noswap_vqmovn_u16(__rev1));
	53977	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	53978	return __ret;
	53979	}
52423	53980	#endif
52424	53981
52425	53982	#ifdef __LITTLE_ENDIAN__
52426		#define vdup_laneq_u32(__p0_489, __p1_489) __extension__ ({ \
52427		uint32x2_t __ret_489; \
52428		uint32x4_t __s0_489 = __p0_489; \
52429		__ret_489 = splat_laneq_u32(__s0_489, __p1_489); \
52430		__ret_489; \
52431		})
	53983	__ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
	53984	int16x8_t __ret;
	53985	__ret = vcombine_s16(__p0, vqmovn_s32(__p1));
	53986	return __ret;
	53987	}
52432	53988	#else
52433		#define vdup_laneq_u32(__p0_490, __p1_490) __extension__ ({ \
52434		uint32x2_t __ret_490; \
52435		uint32x4_t __s0_490 = __p0_490; \
52436		uint32x4_t __rev0_490; __rev0_490 = __builtin_shufflevector(__s0_490, __s0_490, 3, 2, 1, 0); \
52437		__ret_490 = __noswap_splat_laneq_u32(__rev0_490, __p1_490); \
52438		__ret_490 = __builtin_shufflevector(__ret_490, __ret_490, 1, 0); \
52439		__ret_490; \
52440		})
	53989	__ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
	53990	int16x8_t __ret;
	53991	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	53992	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	53993	__ret = __noswap_vcombine_s16(__rev0, __noswap_vqmovn_s32(__rev1));
	53994	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	53995	return __ret;
	53996	}
52441	53997	#endif
52442	53998
52443	53999	#ifdef __LITTLE_ENDIAN__
52444		#define vdup_laneq_u64(__p0_491, __p1_491) __extension__ ({ \
52445		uint64x1_t __ret_491; \
52446		uint64x2_t __s0_491 = __p0_491; \
52447		__ret_491 = splat_laneq_u64(__s0_491, __p1_491); \
52448		__ret_491; \
52449		})
	54000	__ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
	54001	int32x4_t __ret;
	54002	__ret = vcombine_s32(__p0, vqmovn_s64(__p1));
	54003	return __ret;
	54004	}
52450	54005	#else
52451		#define vdup_laneq_u64(__p0_492, __p1_492) __extension__ ({ \
52452		uint64x1_t __ret_492; \
52453		uint64x2_t __s0_492 = __p0_492; \
52454		uint64x2_t __rev0_492; __rev0_492 = __builtin_shufflevector(__s0_492, __s0_492, 1, 0); \
52455		__ret_492 = __noswap_splat_laneq_u64(__rev0_492, __p1_492); \
52456		__ret_492; \
52457		})
	54006	__ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
	54007	int32x4_t __ret;
	54008	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	54009	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	54010	__ret = __noswap_vcombine_s32(__rev0, __noswap_vqmovn_s64(__rev1));
	54011	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	54012	return __ret;
	54013	}
52458	54014	#endif
52459	54015
52460	54016	#ifdef __LITTLE_ENDIAN__
52461		#define vdup_laneq_u16(__p0_493, __p1_493) __extension__ ({ \
52462		uint16x4_t __ret_493; \
52463		uint16x8_t __s0_493 = __p0_493; \
52464		__ret_493 = splat_laneq_u16(__s0_493, __p1_493); \
52465		__ret_493; \
	54017	__ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
	54018	int8x16_t __ret;
	54019	__ret = vcombine_s8(__p0, vqmovn_s16(__p1));
	54020	return __ret;
	54021	}
	54022	#else
	54023	__ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
	54024	int8x16_t __ret;
	54025	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	54026	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	54027	__ret = __noswap_vcombine_s8(__rev0, __noswap_vqmovn_s16(__rev1));
	54028	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	54029	return __ret;
	54030	}
	54031	#endif
	54032
	54033	__ai uint16_t vqmovuns_s32(int32_t __p0) {
	54034	uint16_t __ret;
	54035	__ret = (uint16_t) __builtin_neon_vqmovuns_s32(__p0);
	54036	return __ret;
	54037	}
	54038	__ai uint32_t vqmovund_s64(int64_t __p0) {
	54039	uint32_t __ret;
	54040	__ret = (uint32_t) __builtin_neon_vqmovund_s64(__p0);
	54041	return __ret;
	54042	}
	54043	__ai uint8_t vqmovunh_s16(int16_t __p0) {
	54044	uint8_t __ret;
	54045	__ret = (uint8_t) __builtin_neon_vqmovunh_s16(__p0);
	54046	return __ret;
	54047	}
	54048	#ifdef __LITTLE_ENDIAN__
	54049	__ai uint16x8_t vqmovun_high_s32(uint16x4_t __p0, int32x4_t __p1) {
	54050	uint16x8_t __ret;
	54051	__ret = vcombine_u16((uint16x4_t)(__p0), vqmovun_s32(__p1));
	54052	return __ret;
	54053	}
	54054	#else
	54055	__ai uint16x8_t vqmovun_high_s32(uint16x4_t __p0, int32x4_t __p1) {
	54056	uint16x8_t __ret;
	54057	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	54058	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	54059	__ret = __noswap_vcombine_u16((uint16x4_t)(__rev0), __noswap_vqmovun_s32(__rev1));
	54060	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	54061	return __ret;
	54062	}
	54063	#endif
	54064
	54065	#ifdef __LITTLE_ENDIAN__
	54066	__ai uint32x4_t vqmovun_high_s64(uint32x2_t __p0, int64x2_t __p1) {
	54067	uint32x4_t __ret;
	54068	__ret = vcombine_u32((uint32x2_t)(__p0), vqmovun_s64(__p1));
	54069	return __ret;
	54070	}
	54071	#else
	54072	__ai uint32x4_t vqmovun_high_s64(uint32x2_t __p0, int64x2_t __p1) {
	54073	uint32x4_t __ret;
	54074	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	54075	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	54076	__ret = __noswap_vcombine_u32((uint32x2_t)(__rev0), __noswap_vqmovun_s64(__rev1));
	54077	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	54078	return __ret;
	54079	}
	54080	#endif
	54081
	54082	#ifdef __LITTLE_ENDIAN__
	54083	__ai uint8x16_t vqmovun_high_s16(uint8x8_t __p0, int16x8_t __p1) {
	54084	uint8x16_t __ret;
	54085	__ret = vcombine_u8((uint8x8_t)(__p0), vqmovun_s16(__p1));
	54086	return __ret;
	54087	}
	54088	#else
	54089	__ai uint8x16_t vqmovun_high_s16(uint8x8_t __p0, int16x8_t __p1) {
	54090	uint8x16_t __ret;
	54091	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	54092	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	54093	__ret = __noswap_vcombine_u8((uint8x8_t)(__rev0), __noswap_vqmovun_s16(__rev1));
	54094	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	54095	return __ret;
	54096	}
	54097	#endif
	54098
	54099	#ifdef __LITTLE_ENDIAN__
	54100	__ai int64x2_t vqnegq_s64(int64x2_t __p0) {
	54101	int64x2_t __ret;
	54102	__ret = (int64x2_t) __builtin_neon_vqnegq_v((int8x16_t)__p0, 35);
	54103	return __ret;
	54104	}
	54105	#else
	54106	__ai int64x2_t vqnegq_s64(int64x2_t __p0) {
	54107	int64x2_t __ret;
	54108	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	54109	__ret = (int64x2_t) __builtin_neon_vqnegq_v((int8x16_t)__rev0, 35);
	54110	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	54111	return __ret;
	54112	}
	54113	#endif
	54114
	54115	__ai int64x1_t vqneg_s64(int64x1_t __p0) {
	54116	int64x1_t __ret;
	54117	__ret = (int64x1_t) __builtin_neon_vqneg_v((int8x8_t)__p0, 3);
	54118	return __ret;
	54119	}
	54120	__ai int8_t vqnegb_s8(int8_t __p0) {
	54121	int8_t __ret;
	54122	__ret = (int8_t) __builtin_neon_vqnegb_s8(__p0);
	54123	return __ret;
	54124	}
	54125	__ai int32_t vqnegs_s32(int32_t __p0) {
	54126	int32_t __ret;
	54127	__ret = (int32_t) __builtin_neon_vqnegs_s32(__p0);
	54128	return __ret;
	54129	}
	54130	__ai int64_t vqnegd_s64(int64_t __p0) {
	54131	int64_t __ret;
	54132	__ret = (int64_t) __builtin_neon_vqnegd_s64(__p0);
	54133	return __ret;
	54134	}
	54135	__ai int16_t vqnegh_s16(int16_t __p0) {
	54136	int16_t __ret;
	54137	__ret = (int16_t) __builtin_neon_vqnegh_s16(__p0);
	54138	return __ret;
	54139	}
	54140	__ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
	54141	int32_t __ret;
	54142	__ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
	54143	return __ret;
	54144	}
	54145	__ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
	54146	int16_t __ret;
	54147	__ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
	54148	return __ret;
	54149	}
	54150	#ifdef __LITTLE_ENDIAN__
	54151	#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	54152	int32x4_t __ret; \
	54153	int32x4_t __s0 = __p0; \
	54154	int32x2_t __s1 = __p1; \
	54155	__ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
	54156	__ret; \
52466	54157	})
52467	54158	#else
52468		#define vdup_laneq_u16(__p0_494, __p1_494) __extension__ ({ \
52469		uint16x4_t __ret_494; \
52470		uint16x8_t __s0_494 = __p0_494; \
52471		uint16x8_t __rev0_494; __rev0_494 = __builtin_shufflevector(__s0_494, __s0_494, 7, 6, 5, 4, 3, 2, 1, 0); \
52472		__ret_494 = __noswap_splat_laneq_u16(__rev0_494, __p1_494); \
52473		__ret_494 = __builtin_shufflevector(__ret_494, __ret_494, 3, 2, 1, 0); \
52474		__ret_494; \
	54159	#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	54160	int32x4_t __ret; \
	54161	int32x4_t __s0 = __p0; \
	54162	int32x2_t __s1 = __p1; \
	54163	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	54164	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	54165	__ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
	54166	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	54167	__ret; \
52475	54168	})
52476	54169	#endif
52477	54170
52478	54171	#ifdef __LITTLE_ENDIAN__
52479		#define vdup_laneq_s8(__p0_495, __p1_495) __extension__ ({ \
52480		int8x8_t __ret_495; \
52481		int8x16_t __s0_495 = __p0_495; \
52482		__ret_495 = splat_laneq_s8(__s0_495, __p1_495); \
52483		__ret_495; \
	54172	#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	54173	int16x8_t __ret; \
	54174	int16x8_t __s0 = __p0; \
	54175	int16x4_t __s1 = __p1; \
	54176	__ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
	54177	__ret; \
52484	54178	})
52485	54179	#else
52486		#define vdup_laneq_s8(__p0_496, __p1_496) __extension__ ({ \
52487		int8x8_t __ret_496; \
52488		int8x16_t __s0_496 = __p0_496; \
52489		int8x16_t __rev0_496; __rev0_496 = __builtin_shufflevector(__s0_496, __s0_496, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52490		__ret_496 = __noswap_splat_laneq_s8(__rev0_496, __p1_496); \
52491		__ret_496 = __builtin_shufflevector(__ret_496, __ret_496, 7, 6, 5, 4, 3, 2, 1, 0); \
52492		__ret_496; \
	54180	#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	54181	int16x8_t __ret; \
	54182	int16x8_t __s0 = __p0; \
	54183	int16x4_t __s1 = __p1; \
	54184	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	54185	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	54186	__ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
	54187	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	54188	__ret; \
52493	54189	})
52494	54190	#endif
52495	54191
52496	54192	#ifdef __LITTLE_ENDIAN__
52497		#define vdup_laneq_f64(__p0_497, __p1_497) __extension__ ({ \
52498		float64x1_t __ret_497; \
52499		float64x2_t __s0_497 = __p0_497; \
52500		__ret_497 = splat_laneq_f64(__s0_497, __p1_497); \
52501		__ret_497; \
	54193	#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	54194	int32x2_t __ret; \
	54195	int32x2_t __s0 = __p0; \
	54196	int32x2_t __s1 = __p1; \
	54197	__ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
	54198	__ret; \
52502	54199	})
52503	54200	#else
52504		#define vdup_laneq_f64(__p0_498, __p1_498) __extension__ ({ \
52505		float64x1_t __ret_498; \
52506		float64x2_t __s0_498 = __p0_498; \
52507		float64x2_t __rev0_498; __rev0_498 = __builtin_shufflevector(__s0_498, __s0_498, 1, 0); \
52508		__ret_498 = __noswap_splat_laneq_f64(__rev0_498, __p1_498); \
52509		__ret_498; \
	54201	#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
	54202	int32x2_t __ret; \
	54203	int32x2_t __s0 = __p0; \
	54204	int32x2_t __s1 = __p1; \
	54205	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	54206	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	54207	__ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
	54208	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	54209	__ret; \
52510	54210	})
52511	54211	#endif
52512	54212
52513	54213	#ifdef __LITTLE_ENDIAN__
52514		#define vdup_laneq_f32(__p0_499, __p1_499) __extension__ ({ \
52515		float32x2_t __ret_499; \
52516		float32x4_t __s0_499 = __p0_499; \
52517		__ret_499 = splat_laneq_f32(__s0_499, __p1_499); \
52518		__ret_499; \
	54214	#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	54215	int16x4_t __ret; \
	54216	int16x4_t __s0 = __p0; \
	54217	int16x4_t __s1 = __p1; \
	54218	__ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
	54219	__ret; \
52519	54220	})
52520	54221	#else
52521		#define vdup_laneq_f32(__p0_500, __p1_500) __extension__ ({ \
52522		float32x2_t __ret_500; \
52523		float32x4_t __s0_500 = __p0_500; \
52524		float32x4_t __rev0_500; __rev0_500 = __builtin_shufflevector(__s0_500, __s0_500, 3, 2, 1, 0); \
52525		__ret_500 = __noswap_splat_laneq_f32(__rev0_500, __p1_500); \
52526		__ret_500 = __builtin_shufflevector(__ret_500, __ret_500, 1, 0); \
52527		__ret_500; \
	54222	#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
	54223	int16x4_t __ret; \
	54224	int16x4_t __s0 = __p0; \
	54225	int16x4_t __s1 = __p1; \
	54226	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	54227	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	54228	__ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
	54229	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	54230	__ret; \
52528	54231	})
52529	54232	#endif
52530	54233
52531	54234	#ifdef __LITTLE_ENDIAN__
52532		#define vdup_laneq_f16(__p0_501, __p1_501) __extension__ ({ \
52533		float16x4_t __ret_501; \
52534		float16x8_t __s0_501 = __p0_501; \
52535		__ret_501 = splat_laneq_f16(__s0_501, __p1_501); \
52536		__ret_501; \
	54235	#define vqrdmulhs_lane_s32(__p0_659, __p1_659, __p2_659) __extension__ ({ \
	54236	int32_t __ret_659; \
	54237	int32_t __s0_659 = __p0_659; \
	54238	int32x2_t __s1_659 = __p1_659; \
	54239	__ret_659 = vqrdmulhs_s32(__s0_659, vget_lane_s32(__s1_659, __p2_659)); \
	54240	__ret_659; \
52537	54241	})
52538	54242	#else
52539		#define vdup_laneq_f16(__p0_502, __p1_502) __extension__ ({ \
52540		float16x4_t __ret_502; \
52541		float16x8_t __s0_502 = __p0_502; \
52542		float16x8_t __rev0_502; __rev0_502 = __builtin_shufflevector(__s0_502, __s0_502, 7, 6, 5, 4, 3, 2, 1, 0); \
52543		__ret_502 = __noswap_splat_laneq_f16(__rev0_502, __p1_502); \
52544		__ret_502 = __builtin_shufflevector(__ret_502, __ret_502, 3, 2, 1, 0); \
52545		__ret_502; \
	54243	#define vqrdmulhs_lane_s32(__p0_660, __p1_660, __p2_660) __extension__ ({ \
	54244	int32_t __ret_660; \
	54245	int32_t __s0_660 = __p0_660; \
	54246	int32x2_t __s1_660 = __p1_660; \
	54247	int32x2_t __rev1_660; __rev1_660 = __builtin_shufflevector(__s1_660, __s1_660, 1, 0); \
	54248	__ret_660 = vqrdmulhs_s32(__s0_660, __noswap_vget_lane_s32(__rev1_660, __p2_660)); \
	54249	__ret_660; \
52546	54250	})
52547	54251	#endif
52548	54252
52549	54253	#ifdef __LITTLE_ENDIAN__
52550		#define vdup_laneq_s32(__p0_503, __p1_503) __extension__ ({ \
52551		int32x2_t __ret_503; \
52552		int32x4_t __s0_503 = __p0_503; \
52553		__ret_503 = splat_laneq_s32(__s0_503, __p1_503); \
52554		__ret_503; \
	54254	#define vqrdmulhh_lane_s16(__p0_661, __p1_661, __p2_661) __extension__ ({ \
	54255	int16_t __ret_661; \
	54256	int16_t __s0_661 = __p0_661; \
	54257	int16x4_t __s1_661 = __p1_661; \
	54258	__ret_661 = vqrdmulhh_s16(__s0_661, vget_lane_s16(__s1_661, __p2_661)); \
	54259	__ret_661; \
52555	54260	})
52556	54261	#else
52557		#define vdup_laneq_s32(__p0_504, __p1_504) __extension__ ({ \
52558		int32x2_t __ret_504; \
52559		int32x4_t __s0_504 = __p0_504; \
52560		int32x4_t __rev0_504; __rev0_504 = __builtin_shufflevector(__s0_504, __s0_504, 3, 2, 1, 0); \
52561		__ret_504 = __noswap_splat_laneq_s32(__rev0_504, __p1_504); \
52562		__ret_504 = __builtin_shufflevector(__ret_504, __ret_504, 1, 0); \
52563		__ret_504; \
	54262	#define vqrdmulhh_lane_s16(__p0_662, __p1_662, __p2_662) __extension__ ({ \
	54263	int16_t __ret_662; \
	54264	int16_t __s0_662 = __p0_662; \
	54265	int16x4_t __s1_662 = __p1_662; \
	54266	int16x4_t __rev1_662; __rev1_662 = __builtin_shufflevector(__s1_662, __s1_662, 3, 2, 1, 0); \
	54267	__ret_662 = vqrdmulhh_s16(__s0_662, __noswap_vget_lane_s16(__rev1_662, __p2_662)); \
	54268	__ret_662; \
52564	54269	})
52565	54270	#endif
52566	54271
52567	54272	#ifdef __LITTLE_ENDIAN__
52568		#define vdup_laneq_s64(__p0_505, __p1_505) __extension__ ({ \
52569		int64x1_t __ret_505; \
52570		int64x2_t __s0_505 = __p0_505; \
52571		__ret_505 = splat_laneq_s64(__s0_505, __p1_505); \
52572		__ret_505; \
	54273	#define vqrdmulhs_laneq_s32(__p0_663, __p1_663, __p2_663) __extension__ ({ \
	54274	int32_t __ret_663; \
	54275	int32_t __s0_663 = __p0_663; \
	54276	int32x4_t __s1_663 = __p1_663; \
	54277	__ret_663 = vqrdmulhs_s32(__s0_663, vgetq_lane_s32(__s1_663, __p2_663)); \
	54278	__ret_663; \
52573	54279	})
52574	54280	#else
52575		#define vdup_laneq_s64(__p0_506, __p1_506) __extension__ ({ \
52576		int64x1_t __ret_506; \
52577		int64x2_t __s0_506 = __p0_506; \
52578		int64x2_t __rev0_506; __rev0_506 = __builtin_shufflevector(__s0_506, __s0_506, 1, 0); \
52579		__ret_506 = __noswap_splat_laneq_s64(__rev0_506, __p1_506); \
52580		__ret_506; \
	54281	#define vqrdmulhs_laneq_s32(__p0_664, __p1_664, __p2_664) __extension__ ({ \
	54282	int32_t __ret_664; \
	54283	int32_t __s0_664 = __p0_664; \
	54284	int32x4_t __s1_664 = __p1_664; \
	54285	int32x4_t __rev1_664; __rev1_664 = __builtin_shufflevector(__s1_664, __s1_664, 3, 2, 1, 0); \
	54286	__ret_664 = vqrdmulhs_s32(__s0_664, __noswap_vgetq_lane_s32(__rev1_664, __p2_664)); \
	54287	__ret_664; \
52581	54288	})
52582	54289	#endif
52583	54290
52584	54291	#ifdef __LITTLE_ENDIAN__
52585		#define vdup_laneq_s16(__p0_507, __p1_507) __extension__ ({ \
52586		int16x4_t __ret_507; \
52587		int16x8_t __s0_507 = __p0_507; \
52588		__ret_507 = splat_laneq_s16(__s0_507, __p1_507); \
52589		__ret_507; \
	54292	#define vqrdmulhh_laneq_s16(__p0_665, __p1_665, __p2_665) __extension__ ({ \
	54293	int16_t __ret_665; \
	54294	int16_t __s0_665 = __p0_665; \
	54295	int16x8_t __s1_665 = __p1_665; \
	54296	__ret_665 = vqrdmulhh_s16(__s0_665, vgetq_lane_s16(__s1_665, __p2_665)); \
	54297	__ret_665; \
52590	54298	})
52591	54299	#else
52592		#define vdup_laneq_s16(__p0_508, __p1_508) __extension__ ({ \
52593		int16x4_t __ret_508; \
52594		int16x8_t __s0_508 = __p0_508; \
52595		int16x8_t __rev0_508; __rev0_508 = __builtin_shufflevector(__s0_508, __s0_508, 7, 6, 5, 4, 3, 2, 1, 0); \
52596		__ret_508 = __noswap_splat_laneq_s16(__rev0_508, __p1_508); \
52597		__ret_508 = __builtin_shufflevector(__ret_508, __ret_508, 3, 2, 1, 0); \
52598		__ret_508; \
	54300	#define vqrdmulhh_laneq_s16(__p0_666, __p1_666, __p2_666) __extension__ ({ \
	54301	int16_t __ret_666; \
	54302	int16_t __s0_666 = __p0_666; \
	54303	int16x8_t __s1_666 = __p1_666; \
	54304	int16x8_t __rev1_666; __rev1_666 = __builtin_shufflevector(__s1_666, __s1_666, 7, 6, 5, 4, 3, 2, 1, 0); \
	54305	__ret_666 = vqrdmulhh_s16(__s0_666, __noswap_vgetq_lane_s16(__rev1_666, __p2_666)); \
	54306	__ret_666; \
52599	54307	})
52600	54308	#endif
52601	54309
52602		__ai poly64x1_t vdup_n_p64(poly64_t __p0) {
52603		poly64x1_t __ret;
52604		__ret = (poly64x1_t) {__p0};
52605		return __ret;
52606		}
52607	54310	#ifdef __LITTLE_ENDIAN__
52608		__ai poly64x2_t vdupq_n_p64(poly64_t __p0) {
52609		poly64x2_t __ret;
52610		__ret = (poly64x2_t) {__p0, __p0};
52611		return __ret;
52612		}
	54311	#define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	54312	int32x4_t __ret; \
	54313	int32x4_t __s0 = __p0; \
	54314	int32x4_t __s1 = __p1; \
	54315	__ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
	54316	__ret; \
	54317	})
52613	54318	#else
52614		__ai poly64x2_t vdupq_n_p64(poly64_t __p0) {
52615		poly64x2_t __ret;
52616		__ret = (poly64x2_t) {__p0, __p0};
52617		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
52618		return __ret;
52619		}
	54319	#define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	54320	int32x4_t __ret; \
	54321	int32x4_t __s0 = __p0; \
	54322	int32x4_t __s1 = __p1; \
	54323	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	54324	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	54325	__ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
	54326	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	54327	__ret; \
	54328	})
52620	54329	#endif
52621	54330
52622	54331	#ifdef __LITTLE_ENDIAN__
52623		__ai float64x2_t vdupq_n_f64(float64_t __p0) {
52624		float64x2_t __ret;
52625		__ret = (float64x2_t) {__p0, __p0};
52626		return __ret;
52627		}
	54332	#define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	54333	int16x8_t __ret; \
	54334	int16x8_t __s0 = __p0; \
	54335	int16x8_t __s1 = __p1; \
	54336	__ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
	54337	__ret; \
	54338	})
52628	54339	#else
52629		__ai float64x2_t vdupq_n_f64(float64_t __p0) {
52630		float64x2_t __ret;
52631		__ret = (float64x2_t) {__p0, __p0};
52632		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
52633		return __ret;
52634		}
52635		#endif
52636
52637		__ai float64x1_t vdup_n_f64(float64_t __p0) {
52638		float64x1_t __ret;
52639		__ret = (float64x1_t) {__p0};
52640		return __ret;
52641		}
52642		#define vext_p64(__p0, __p1, __p2) __extension__ ({ \
52643		poly64x1_t __ret; \
52644		poly64x1_t __s0 = __p0; \
52645		poly64x1_t __s1 = __p1; \
52646		__ret = (poly64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
	54340	#define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	54341	int16x8_t __ret; \
	54342	int16x8_t __s0 = __p0; \
	54343	int16x8_t __s1 = __p1; \
	54344	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	54345	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	54346	__ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
	54347	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
52647	54348	__ret; \
52648	54349	})
	54350	#endif
	54351
52649	54352	#ifdef __LITTLE_ENDIAN__
52650		#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
52651		poly64x2_t __ret; \
52652		poly64x2_t __s0 = __p0; \
52653		poly64x2_t __s1 = __p1; \
52654		__ret = (poly64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
	54353	#define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	54354	int32x2_t __ret; \
	54355	int32x2_t __s0 = __p0; \
	54356	int32x4_t __s1 = __p1; \
	54357	__ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
52655	54358	__ret; \
52656	54359	})
52657	54360	#else
52658		#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
52659		poly64x2_t __ret; \
52660		poly64x2_t __s0 = __p0; \
52661		poly64x2_t __s1 = __p1; \
52662		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52663		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52664		__ret = (poly64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
	54361	#define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
	54362	int32x2_t __ret; \
	54363	int32x2_t __s0 = __p0; \
	54364	int32x4_t __s1 = __p1; \
	54365	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	54366	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	54367	__ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
52665	54368	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
52666	54369	__ret; \
52667	54370	})
52668	54371	#endif
52669	54372
52670	54373	#ifdef __LITTLE_ENDIAN__
52671		#define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
52672		float64x2_t __ret; \
52673		float64x2_t __s0 = __p0; \
52674		float64x2_t __s1 = __p1; \
52675		__ret = (float64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 42); \
	54374	#define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	54375	int16x4_t __ret; \
	54376	int16x4_t __s0 = __p0; \
	54377	int16x8_t __s1 = __p1; \
	54378	__ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
52676	54379	__ret; \
52677	54380	})
52678	54381	#else
52679		#define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
52680		float64x2_t __ret; \
52681		float64x2_t __s0 = __p0; \
52682		float64x2_t __s1 = __p1; \
52683		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52684		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52685		__ret = (float64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 42); \
52686		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	54382	#define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
	54383	int16x4_t __ret; \
	54384	int16x4_t __s0 = __p0; \
	54385	int16x8_t __s1 = __p1; \
	54386	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	54387	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	54388	__ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
	54389	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
52687	54390	__ret; \
52688	54391	})
52689	54392	#endif
52690	54393
52691		#define vext_f64(__p0, __p1, __p2) __extension__ ({ \
52692		float64x1_t __ret; \
52693		float64x1_t __s0 = __p0; \
52694		float64x1_t __s1 = __p1; \
52695		__ret = (float64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
52696		__ret; \
52697		})
52698		#ifdef __LITTLE_ENDIAN__
52699		__ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52700		float64x2_t __ret;
52701		__ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	54394	__ai uint8_t vqrshlb_u8(uint8_t __p0, int8_t __p1) {
	54395	uint8_t __ret;
	54396	__ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
52702	54397	return __ret;
52703	54398	}
52704		#else
52705		__ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52706		float64x2_t __ret;
52707		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
52708		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52709		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
52710		__ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
52711		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	54399	__ai uint32_t vqrshls_u32(uint32_t __p0, int32_t __p1) {
	54400	uint32_t __ret;
	54401	__ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
52712	54402	return __ret;
52713	54403	}
52714		__ai float64x2_t __noswap_vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52715		float64x2_t __ret;
52716		__ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	54404	__ai uint64_t vqrshld_u64(uint64_t __p0, int64_t __p1) {
	54405	uint64_t __ret;
	54406	__ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
52717	54407	return __ret;
52718	54408	}
52719		#endif
52720
52721		__ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
52722		float64x1_t __ret;
52723		__ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
	54409	__ai uint16_t vqrshlh_u16(uint16_t __p0, int16_t __p1) {
	54410	uint16_t __ret;
	54411	__ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
	54412	return __ret;
	54413	}
	54414	__ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
	54415	int8_t __ret;
	54416	__ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
	54417	return __ret;
	54418	}
	54419	__ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
	54420	int32_t __ret;
	54421	__ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
	54422	return __ret;
	54423	}
	54424	__ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
	54425	int64_t __ret;
	54426	__ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
	54427	return __ret;
	54428	}
	54429	__ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
	54430	int16_t __ret;
	54431	__ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
52724	54432	return __ret;
52725	54433	}
52726		#define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52727		float64_t __ret; \
52728		float64_t __s0 = __p0; \
52729		float64_t __s1 = __p1; \
52730		float64x1_t __s2 = __p2; \
52731		__ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (float64x1_t)__s2, __p3); \
52732		__ret; \
52733		})
52734	54434	#ifdef __LITTLE_ENDIAN__
52735		#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52736		float32_t __ret; \
52737		float32_t __s0 = __p0; \
52738		float32_t __s1 = __p1; \
52739		float32x2_t __s2 = __p2; \
52740		__ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
52741		__ret; \
	54435	#define vqrshrn_high_n_u32(__p0_667, __p1_667, __p2_667) __extension__ ({ \
	54436	uint16x8_t __ret_667; \
	54437	uint16x4_t __s0_667 = __p0_667; \
	54438	uint32x4_t __s1_667 = __p1_667; \
	54439	__ret_667 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_667), (uint16x4_t)(vqrshrn_n_u32(__s1_667, __p2_667)))); \
	54440	__ret_667; \
52742	54441	})
52743	54442	#else
52744		#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52745		float32_t __ret; \
52746		float32_t __s0 = __p0; \
52747		float32_t __s1 = __p1; \
52748		float32x2_t __s2 = __p2; \
52749		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52750		__ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__rev2, __p3); \
52751		__ret; \
52752		})
52753		#define __noswap_vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52754		float32_t __ret; \
52755		float32_t __s0 = __p0; \
52756		float32_t __s1 = __p1; \
52757		float32x2_t __s2 = __p2; \
52758		__ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
52759		__ret; \
	54443	#define vqrshrn_high_n_u32(__p0_668, __p1_668, __p2_668) __extension__ ({ \
	54444	uint16x8_t __ret_668; \
	54445	uint16x4_t __s0_668 = __p0_668; \
	54446	uint32x4_t __s1_668 = __p1_668; \
	54447	uint16x4_t __rev0_668; __rev0_668 = __builtin_shufflevector(__s0_668, __s0_668, 3, 2, 1, 0); \
	54448	uint32x4_t __rev1_668; __rev1_668 = __builtin_shufflevector(__s1_668, __s1_668, 3, 2, 1, 0); \
	54449	__ret_668 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_668), (uint16x4_t)(__noswap_vqrshrn_n_u32(__rev1_668, __p2_668)))); \
	54450	__ret_668 = __builtin_shufflevector(__ret_668, __ret_668, 7, 6, 5, 4, 3, 2, 1, 0); \
	54451	__ret_668; \
52760	54452	})
52761	54453	#endif
52762	54454
52763	54455	#ifdef __LITTLE_ENDIAN__
52764		#define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52765		float64x2_t __ret; \
52766		float64x2_t __s0 = __p0; \
52767		float64x2_t __s1 = __p1; \
52768		float64x1_t __s2 = __p2; \
52769		__ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 42); \
52770		__ret; \
	54456	#define vqrshrn_high_n_u64(__p0_669, __p1_669, __p2_669) __extension__ ({ \
	54457	uint32x4_t __ret_669; \
	54458	uint32x2_t __s0_669 = __p0_669; \
	54459	uint64x2_t __s1_669 = __p1_669; \
	54460	__ret_669 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_669), (uint32x2_t)(vqrshrn_n_u64(__s1_669, __p2_669)))); \
	54461	__ret_669; \
52771	54462	})
52772	54463	#else
52773		#define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52774		float64x2_t __ret; \
52775		float64x2_t __s0 = __p0; \
52776		float64x2_t __s1 = __p1; \
52777		float64x1_t __s2 = __p2; \
52778		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52779		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52780		__ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__s2, __p3, 42); \
52781		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
52782		__ret; \
52783		})
52784		#define __noswap_vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52785		float64x2_t __ret; \
52786		float64x2_t __s0 = __p0; \
52787		float64x2_t __s1 = __p1; \
52788		float64x1_t __s2 = __p2; \
52789		__ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 42); \
52790		__ret; \
	54464	#define vqrshrn_high_n_u64(__p0_670, __p1_670, __p2_670) __extension__ ({ \
	54465	uint32x4_t __ret_670; \
	54466	uint32x2_t __s0_670 = __p0_670; \
	54467	uint64x2_t __s1_670 = __p1_670; \
	54468	uint32x2_t __rev0_670; __rev0_670 = __builtin_shufflevector(__s0_670, __s0_670, 1, 0); \
	54469	uint64x2_t __rev1_670; __rev1_670 = __builtin_shufflevector(__s1_670, __s1_670, 1, 0); \
	54470	__ret_670 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_670), (uint32x2_t)(__noswap_vqrshrn_n_u64(__rev1_670, __p2_670)))); \
	54471	__ret_670 = __builtin_shufflevector(__ret_670, __ret_670, 3, 2, 1, 0); \
	54472	__ret_670; \
52791	54473	})
52792	54474	#endif
52793	54475
52794	54476	#ifdef __LITTLE_ENDIAN__
52795		#define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52796		float32x4_t __ret; \
52797		float32x4_t __s0 = __p0; \
52798		float32x4_t __s1 = __p1; \
52799		float32x2_t __s2 = __p2; \
52800		__ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 41); \
52801		__ret; \
	54477	#define vqrshrn_high_n_u16(__p0_671, __p1_671, __p2_671) __extension__ ({ \
	54478	uint8x16_t __ret_671; \
	54479	uint8x8_t __s0_671 = __p0_671; \
	54480	uint16x8_t __s1_671 = __p1_671; \
	54481	__ret_671 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_671), (uint8x8_t)(vqrshrn_n_u16(__s1_671, __p2_671)))); \
	54482	__ret_671; \
52802	54483	})
52803	54484	#else
52804		#define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52805		float32x4_t __ret; \
52806		float32x4_t __s0 = __p0; \
52807		float32x4_t __s1 = __p1; \
52808		float32x2_t __s2 = __p2; \
52809		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
52810		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
52811		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52812		__ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, __p3, 41); \
52813		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
52814		__ret; \
52815		})
52816		#define __noswap_vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52817		float32x4_t __ret; \
52818		float32x4_t __s0 = __p0; \
52819		float32x4_t __s1 = __p1; \
52820		float32x2_t __s2 = __p2; \
52821		__ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 41); \
52822		__ret; \
	54485	#define vqrshrn_high_n_u16(__p0_672, __p1_672, __p2_672) __extension__ ({ \
	54486	uint8x16_t __ret_672; \
	54487	uint8x8_t __s0_672 = __p0_672; \
	54488	uint16x8_t __s1_672 = __p1_672; \
	54489	uint8x8_t __rev0_672; __rev0_672 = __builtin_shufflevector(__s0_672, __s0_672, 7, 6, 5, 4, 3, 2, 1, 0); \
	54490	uint16x8_t __rev1_672; __rev1_672 = __builtin_shufflevector(__s1_672, __s1_672, 7, 6, 5, 4, 3, 2, 1, 0); \
	54491	__ret_672 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_672), (uint8x8_t)(__noswap_vqrshrn_n_u16(__rev1_672, __p2_672)))); \
	54492	__ret_672 = __builtin_shufflevector(__ret_672, __ret_672, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	54493	__ret_672; \
52823	54494	})
52824	54495	#endif
52825	54496
52826		#define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52827		float64x1_t __ret; \
52828		float64x1_t __s0 = __p0; \
52829		float64x1_t __s1 = __p1; \
52830		float64x1_t __s2 = __p2; \
52831		__ret = (float64x1_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 10); \
52832		__ret; \
52833		})
52834	54497	#ifdef __LITTLE_ENDIAN__
52835		#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52836		float32x2_t __ret; \
52837		float32x2_t __s0 = __p0; \
52838		float32x2_t __s1 = __p1; \
52839		float32x2_t __s2 = __p2; \
52840		__ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 9); \
52841		__ret; \
	54498	#define vqrshrn_high_n_s32(__p0_673, __p1_673, __p2_673) __extension__ ({ \
	54499	int16x8_t __ret_673; \
	54500	int16x4_t __s0_673 = __p0_673; \
	54501	int32x4_t __s1_673 = __p1_673; \
	54502	__ret_673 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_673), (int16x4_t)(vqrshrn_n_s32(__s1_673, __p2_673)))); \
	54503	__ret_673; \
52842	54504	})
52843	54505	#else
52844		#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52845		float32x2_t __ret; \
52846		float32x2_t __s0 = __p0; \
52847		float32x2_t __s1 = __p1; \
52848		float32x2_t __s2 = __p2; \
52849		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52850		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52851		float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52852		__ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, __p3, 9); \
52853		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
52854		__ret; \
52855		})
52856		#define __noswap_vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52857		float32x2_t __ret; \
52858		float32x2_t __s0 = __p0; \
52859		float32x2_t __s1 = __p1; \
52860		float32x2_t __s2 = __p2; \
52861		__ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 9); \
52862		__ret; \
	54506	#define vqrshrn_high_n_s32(__p0_674, __p1_674, __p2_674) __extension__ ({ \
	54507	int16x8_t __ret_674; \
	54508	int16x4_t __s0_674 = __p0_674; \
	54509	int32x4_t __s1_674 = __p1_674; \
	54510	int16x4_t __rev0_674; __rev0_674 = __builtin_shufflevector(__s0_674, __s0_674, 3, 2, 1, 0); \
	54511	int32x4_t __rev1_674; __rev1_674 = __builtin_shufflevector(__s1_674, __s1_674, 3, 2, 1, 0); \
	54512	__ret_674 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_674), (int16x4_t)(__noswap_vqrshrn_n_s32(__rev1_674, __p2_674)))); \
	54513	__ret_674 = __builtin_shufflevector(__ret_674, __ret_674, 7, 6, 5, 4, 3, 2, 1, 0); \
	54514	__ret_674; \
52863	54515	})
52864	54516	#endif
52865	54517
52866	54518	#ifdef __LITTLE_ENDIAN__
52867		#define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52868		float64_t __ret; \
52869		float64_t __s0 = __p0; \
52870		float64_t __s1 = __p1; \
52871		float64x2_t __s2 = __p2; \
52872		__ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
52873		__ret; \
	54519	#define vqrshrn_high_n_s64(__p0_675, __p1_675, __p2_675) __extension__ ({ \
	54520	int32x4_t __ret_675; \
	54521	int32x2_t __s0_675 = __p0_675; \
	54522	int64x2_t __s1_675 = __p1_675; \
	54523	__ret_675 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_675), (int32x2_t)(vqrshrn_n_s64(__s1_675, __p2_675)))); \
	54524	__ret_675; \
52874	54525	})
52875	54526	#else
52876		#define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52877		float64_t __ret; \
52878		float64_t __s0 = __p0; \
52879		float64_t __s1 = __p1; \
52880		float64x2_t __s2 = __p2; \
52881		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52882		__ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__rev2, __p3); \
52883		__ret; \
52884		})
52885		#define __noswap_vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52886		float64_t __ret; \
52887		float64_t __s0 = __p0; \
52888		float64_t __s1 = __p1; \
52889		float64x2_t __s2 = __p2; \
52890		__ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
52891		__ret; \
	54527	#define vqrshrn_high_n_s64(__p0_676, __p1_676, __p2_676) __extension__ ({ \
	54528	int32x4_t __ret_676; \
	54529	int32x2_t __s0_676 = __p0_676; \
	54530	int64x2_t __s1_676 = __p1_676; \
	54531	int32x2_t __rev0_676; __rev0_676 = __builtin_shufflevector(__s0_676, __s0_676, 1, 0); \
	54532	int64x2_t __rev1_676; __rev1_676 = __builtin_shufflevector(__s1_676, __s1_676, 1, 0); \
	54533	__ret_676 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_676), (int32x2_t)(__noswap_vqrshrn_n_s64(__rev1_676, __p2_676)))); \
	54534	__ret_676 = __builtin_shufflevector(__ret_676, __ret_676, 3, 2, 1, 0); \
	54535	__ret_676; \
52892	54536	})
52893	54537	#endif
52894	54538
52895	54539	#ifdef __LITTLE_ENDIAN__
52896		#define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52897		float32_t __ret; \
52898		float32_t __s0 = __p0; \
52899		float32_t __s1 = __p1; \
52900		float32x4_t __s2 = __p2; \
52901		__ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
52902		__ret; \
	54540	#define vqrshrn_high_n_s16(__p0_677, __p1_677, __p2_677) __extension__ ({ \
	54541	int8x16_t __ret_677; \
	54542	int8x8_t __s0_677 = __p0_677; \
	54543	int16x8_t __s1_677 = __p1_677; \
	54544	__ret_677 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_677), (int8x8_t)(vqrshrn_n_s16(__s1_677, __p2_677)))); \
	54545	__ret_677; \
52903	54546	})
52904	54547	#else
52905		#define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52906		float32_t __ret; \
52907		float32_t __s0 = __p0; \
52908		float32_t __s1 = __p1; \
52909		float32x4_t __s2 = __p2; \
52910		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
52911		__ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__rev2, __p3); \
52912		__ret; \
52913		})
52914		#define __noswap_vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52915		float32_t __ret; \
52916		float32_t __s0 = __p0; \
52917		float32_t __s1 = __p1; \
52918		float32x4_t __s2 = __p2; \
52919		__ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
52920		__ret; \
	54548	#define vqrshrn_high_n_s16(__p0_678, __p1_678, __p2_678) __extension__ ({ \
	54549	int8x16_t __ret_678; \
	54550	int8x8_t __s0_678 = __p0_678; \
	54551	int16x8_t __s1_678 = __p1_678; \
	54552	int8x8_t __rev0_678; __rev0_678 = __builtin_shufflevector(__s0_678, __s0_678, 7, 6, 5, 4, 3, 2, 1, 0); \
	54553	int16x8_t __rev1_678; __rev1_678 = __builtin_shufflevector(__s1_678, __s1_678, 7, 6, 5, 4, 3, 2, 1, 0); \
	54554	__ret_678 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_678), (int8x8_t)(__noswap_vqrshrn_n_s16(__rev1_678, __p2_678)))); \
	54555	__ret_678 = __builtin_shufflevector(__ret_678, __ret_678, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	54556	__ret_678; \
52921	54557	})
52922	54558	#endif
52923	54559
52924		#ifdef __LITTLE_ENDIAN__
52925		#define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52926		float64x2_t __ret; \
52927		float64x2_t __s0 = __p0; \
52928		float64x2_t __s1 = __p1; \
52929		float64x2_t __s2 = __p2; \
52930		__ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 42); \
	54560	#define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
	54561	uint16_t __ret; \
	54562	uint32_t __s0 = __p0; \
	54563	__ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
52931	54564	__ret; \
52932	54565	})
52933		#else
52934		#define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52935		float64x2_t __ret; \
52936		float64x2_t __s0 = __p0; \
52937		float64x2_t __s1 = __p1; \
52938		float64x2_t __s2 = __p2; \
52939		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52940		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52941		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52942		__ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 42); \
52943		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	54566	#define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
	54567	uint32_t __ret; \
	54568	uint64_t __s0 = __p0; \
	54569	__ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
52944	54570	__ret; \
52945	54571	})
52946		#define __noswap_vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52947		float64x2_t __ret; \
52948		float64x2_t __s0 = __p0; \
52949		float64x2_t __s1 = __p1; \
52950		float64x2_t __s2 = __p2; \
52951		__ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 42); \
	54572	#define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
	54573	uint8_t __ret; \
	54574	uint16_t __s0 = __p0; \
	54575	__ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
52952	54576	__ret; \
52953	54577	})
52954		#endif
52955
52956		#ifdef __LITTLE_ENDIAN__
52957		#define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52958		float32x4_t __ret; \
52959		float32x4_t __s0 = __p0; \
52960		float32x4_t __s1 = __p1; \
52961		float32x4_t __s2 = __p2; \
52962		__ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 41); \
	54578	#define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
	54579	int16_t __ret; \
	54580	int32_t __s0 = __p0; \
	54581	__ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
52963	54582	__ret; \
52964	54583	})
52965		#else
52966		#define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52967		float32x4_t __ret; \
52968		float32x4_t __s0 = __p0; \
52969		float32x4_t __s1 = __p1; \
52970		float32x4_t __s2 = __p2; \
52971		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
52972		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
52973		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
52974		__ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 41); \
52975		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	54584	#define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
	54585	int32_t __ret; \
	54586	int64_t __s0 = __p0; \
	54587	__ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
52976	54588	__ret; \
52977	54589	})
52978		#define __noswap_vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52979		float32x4_t __ret; \
52980		float32x4_t __s0 = __p0; \
52981		float32x4_t __s1 = __p1; \
52982		float32x4_t __s2 = __p2; \
52983		__ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 41); \
	54590	#define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
	54591	int8_t __ret; \
	54592	int16_t __s0 = __p0; \
	54593	__ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
52984	54594	__ret; \
52985	54595	})
52986		#endif
52987
52988	54596	#ifdef __LITTLE_ENDIAN__
52989		#define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52990		float64x1_t __ret; \
52991		float64x1_t __s0 = __p0; \
52992		float64x1_t __s1 = __p1; \
52993		float64x2_t __s2 = __p2; \
52994		__ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 10); \
52995		__ret; \
	54597	#define vqrshrun_high_n_s32(__p0_679, __p1_679, __p2_679) __extension__ ({ \
	54598	int16x8_t __ret_679; \
	54599	int16x4_t __s0_679 = __p0_679; \
	54600	int32x4_t __s1_679 = __p1_679; \
	54601	__ret_679 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_679), (int16x4_t)(vqrshrun_n_s32(__s1_679, __p2_679)))); \
	54602	__ret_679; \
52996	54603	})
52997	54604	#else
52998		#define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52999		float64x1_t __ret; \
53000		float64x1_t __s0 = __p0; \
53001		float64x1_t __s1 = __p1; \
53002		float64x2_t __s2 = __p2; \
53003		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
53004		__ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__rev2, __p3, 10); \
53005		__ret; \
	54605	#define vqrshrun_high_n_s32(__p0_680, __p1_680, __p2_680) __extension__ ({ \
	54606	int16x8_t __ret_680; \
	54607	int16x4_t __s0_680 = __p0_680; \
	54608	int32x4_t __s1_680 = __p1_680; \
	54609	int16x4_t __rev0_680; __rev0_680 = __builtin_shufflevector(__s0_680, __s0_680, 3, 2, 1, 0); \
	54610	int32x4_t __rev1_680; __rev1_680 = __builtin_shufflevector(__s1_680, __s1_680, 3, 2, 1, 0); \
	54611	__ret_680 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_680), (int16x4_t)(__noswap_vqrshrun_n_s32(__rev1_680, __p2_680)))); \
	54612	__ret_680 = __builtin_shufflevector(__ret_680, __ret_680, 7, 6, 5, 4, 3, 2, 1, 0); \
	54613	__ret_680; \
	54614	})
	54615	#endif
	54616
	54617	#ifdef __LITTLE_ENDIAN__
	54618	#define vqrshrun_high_n_s64(__p0_681, __p1_681, __p2_681) __extension__ ({ \
	54619	int32x4_t __ret_681; \
	54620	int32x2_t __s0_681 = __p0_681; \
	54621	int64x2_t __s1_681 = __p1_681; \
	54622	__ret_681 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_681), (int32x2_t)(vqrshrun_n_s64(__s1_681, __p2_681)))); \
	54623	__ret_681; \
53006	54624	})
53007		#define __noswap_vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
53008		float64x1_t __ret; \
53009		float64x1_t __s0 = __p0; \
53010		float64x1_t __s1 = __p1; \
53011		float64x2_t __s2 = __p2; \
53012		__ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 10); \
53013		__ret; \
	54625	#else
	54626	#define vqrshrun_high_n_s64(__p0_682, __p1_682, __p2_682) __extension__ ({ \
	54627	int32x4_t __ret_682; \
	54628	int32x2_t __s0_682 = __p0_682; \
	54629	int64x2_t __s1_682 = __p1_682; \
	54630	int32x2_t __rev0_682; __rev0_682 = __builtin_shufflevector(__s0_682, __s0_682, 1, 0); \
	54631	int64x2_t __rev1_682; __rev1_682 = __builtin_shufflevector(__s1_682, __s1_682, 1, 0); \
	54632	__ret_682 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_682), (int32x2_t)(__noswap_vqrshrun_n_s64(__rev1_682, __p2_682)))); \
	54633	__ret_682 = __builtin_shufflevector(__ret_682, __ret_682, 3, 2, 1, 0); \
	54634	__ret_682; \
53014	54635	})
53015	54636	#endif
53016	54637
53017	54638	#ifdef __LITTLE_ENDIAN__
53018		#define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53019		float32x2_t __ret; \
53020		float32x2_t __s0 = __p0; \
53021		float32x2_t __s1 = __p1; \
53022		float32x4_t __s2 = __p2; \
53023		__ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 9); \
53024		__ret; \
	54639	#define vqrshrun_high_n_s16(__p0_683, __p1_683, __p2_683) __extension__ ({ \
	54640	int8x16_t __ret_683; \
	54641	int8x8_t __s0_683 = __p0_683; \
	54642	int16x8_t __s1_683 = __p1_683; \
	54643	__ret_683 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_683), (int8x8_t)(vqrshrun_n_s16(__s1_683, __p2_683)))); \
	54644	__ret_683; \
53025	54645	})
53026	54646	#else
53027		#define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53028		float32x2_t __ret; \
53029		float32x2_t __s0 = __p0; \
53030		float32x2_t __s1 = __p1; \
53031		float32x4_t __s2 = __p2; \
53032		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
53033		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
53034		float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
53035		__ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x16_t)__rev2, __p3, 9); \
53036		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	54647	#define vqrshrun_high_n_s16(__p0_684, __p1_684, __p2_684) __extension__ ({ \
	54648	int8x16_t __ret_684; \
	54649	int8x8_t __s0_684 = __p0_684; \
	54650	int16x8_t __s1_684 = __p1_684; \
	54651	int8x8_t __rev0_684; __rev0_684 = __builtin_shufflevector(__s0_684, __s0_684, 7, 6, 5, 4, 3, 2, 1, 0); \
	54652	int16x8_t __rev1_684; __rev1_684 = __builtin_shufflevector(__s1_684, __s1_684, 7, 6, 5, 4, 3, 2, 1, 0); \
	54653	__ret_684 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_684), (int8x8_t)(__noswap_vqrshrun_n_s16(__rev1_684, __p2_684)))); \
	54654	__ret_684 = __builtin_shufflevector(__ret_684, __ret_684, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	54655	__ret_684; \
	54656	})
	54657	#endif
	54658
	54659	#define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
	54660	int16_t __ret; \
	54661	int32_t __s0 = __p0; \
	54662	__ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
53037	54663	__ret; \
53038	54664	})
53039		#define __noswap_vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53040		float32x2_t __ret; \
53041		float32x2_t __s0 = __p0; \
53042		float32x2_t __s1 = __p1; \
53043		float32x4_t __s2 = __p2; \
53044		__ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 9); \
	54665	#define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
	54666	int32_t __ret; \
	54667	int64_t __s0 = __p0; \
	54668	__ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
53045	54669	__ret; \
53046	54670	})
53047		#endif
53048
53049		#ifdef __LITTLE_ENDIAN__
53050		__ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53051		float64x2_t __ret;
53052		__ret = vfmaq_f64(__p0, __p1, (float64x2_t) {__p2, __p2});
	54671	#define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
	54672	int8_t __ret; \
	54673	int16_t __s0 = __p0; \
	54674	__ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
	54675	__ret; \
	54676	})
	54677	__ai uint8_t vqshlb_u8(uint8_t __p0, int8_t __p1) {
	54678	uint8_t __ret;
	54679	__ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
53053	54680	return __ret;
53054	54681	}
53055		#else
53056		__ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53057		float64x2_t __ret;
53058		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53059		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53060		__ret = __noswap_vfmaq_f64(__rev0, __rev1, (float64x2_t) {__p2, __p2});
53061		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	54682	__ai uint32_t vqshls_u32(uint32_t __p0, int32_t __p1) {
	54683	uint32_t __ret;
	54684	__ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
53062	54685	return __ret;
53063	54686	}
53064		#endif
53065
53066		__ai float64x1_t vfma_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
53067		float64x1_t __ret;
53068		__ret = vfma_f64(__p0, __p1, (float64x1_t) {__p2});
	54687	__ai uint64_t vqshld_u64(uint64_t __p0, int64_t __p1) {
	54688	uint64_t __ret;
	54689	__ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
53069	54690	return __ret;
53070	54691	}
53071		#ifdef __LITTLE_ENDIAN__
53072		__ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
53073		float64x2_t __ret;
53074		__ret = vfmaq_f64(__p0, -__p1, __p2);
	54692	__ai uint16_t vqshlh_u16(uint16_t __p0, int16_t __p1) {
	54693	uint16_t __ret;
	54694	__ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
53075	54695	return __ret;
53076	54696	}
53077		#else
53078		__ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
53079		float64x2_t __ret;
53080		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53081		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53082		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
53083		__ret = __noswap_vfmaq_f64(__rev0, -__rev1, __rev2);
53084		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	54697	__ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
	54698	int8_t __ret;
	54699	__ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
53085	54700	return __ret;
53086	54701	}
53087		#endif
53088
53089		__ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
53090		float64x1_t __ret;
53091		__ret = vfma_f64(__p0, -__p1, __p2);
	54702	__ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
	54703	int32_t __ret;
	54704	__ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
53092	54705	return __ret;
53093	54706	}
53094		#define vfmsd_lane_f64(__p0_509, __p1_509, __p2_509, __p3_509) __extension__ ({ \
53095		float64_t __ret_509; \
53096		float64_t __s0_509 = __p0_509; \
53097		float64_t __s1_509 = __p1_509; \
53098		float64x1_t __s2_509 = __p2_509; \
53099		__ret_509 = vfmad_lane_f64(__s0_509, -__s1_509, __s2_509, __p3_509); \
53100		__ret_509; \
	54707	__ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
	54708	int64_t __ret;
	54709	__ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
	54710	return __ret;
	54711	}
	54712	__ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
	54713	int16_t __ret;
	54714	__ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
	54715	return __ret;
	54716	}
	54717	#define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
	54718	uint8_t __ret; \
	54719	uint8_t __s0 = __p0; \
	54720	__ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
	54721	__ret; \
53101	54722	})
53102		#ifdef __LITTLE_ENDIAN__
53103		#define vfmss_lane_f32(__p0_510, __p1_510, __p2_510, __p3_510) __extension__ ({ \
53104		float32_t __ret_510; \
53105		float32_t __s0_510 = __p0_510; \
53106		float32_t __s1_510 = __p1_510; \
53107		float32x2_t __s2_510 = __p2_510; \
53108		__ret_510 = vfmas_lane_f32(__s0_510, -__s1_510, __s2_510, __p3_510); \
53109		__ret_510; \
	54723	#define vqshls_n_u32(__p0, __p1) __extension__ ({ \
	54724	uint32_t __ret; \
	54725	uint32_t __s0 = __p0; \
	54726	__ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
	54727	__ret; \
53110	54728	})
53111		#else
53112		#define vfmss_lane_f32(__p0_511, __p1_511, __p2_511, __p3_511) __extension__ ({ \
53113		float32_t __ret_511; \
53114		float32_t __s0_511 = __p0_511; \
53115		float32_t __s1_511 = __p1_511; \
53116		float32x2_t __s2_511 = __p2_511; \
53117		float32x2_t __rev2_511; __rev2_511 = __builtin_shufflevector(__s2_511, __s2_511, 1, 0); \
53118		__ret_511 = __noswap_vfmas_lane_f32(__s0_511, -__s1_511, __rev2_511, __p3_511); \
53119		__ret_511; \
	54729	#define vqshld_n_u64(__p0, __p1) __extension__ ({ \
	54730	uint64_t __ret; \
	54731	uint64_t __s0 = __p0; \
	54732	__ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
	54733	__ret; \
	54734	})
	54735	#define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
	54736	uint16_t __ret; \
	54737	uint16_t __s0 = __p0; \
	54738	__ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
	54739	__ret; \
	54740	})
	54741	#define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
	54742	int8_t __ret; \
	54743	int8_t __s0 = __p0; \
	54744	__ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
	54745	__ret; \
	54746	})
	54747	#define vqshls_n_s32(__p0, __p1) __extension__ ({ \
	54748	int32_t __ret; \
	54749	int32_t __s0 = __p0; \
	54750	__ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
	54751	__ret; \
	54752	})
	54753	#define vqshld_n_s64(__p0, __p1) __extension__ ({ \
	54754	int64_t __ret; \
	54755	int64_t __s0 = __p0; \
	54756	__ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
	54757	__ret; \
	54758	})
	54759	#define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
	54760	int16_t __ret; \
	54761	int16_t __s0 = __p0; \
	54762	__ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
	54763	__ret; \
	54764	})
	54765	#define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
	54766	int8_t __ret; \
	54767	int8_t __s0 = __p0; \
	54768	__ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
	54769	__ret; \
	54770	})
	54771	#define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
	54772	int32_t __ret; \
	54773	int32_t __s0 = __p0; \
	54774	__ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
	54775	__ret; \
	54776	})
	54777	#define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
	54778	int64_t __ret; \
	54779	int64_t __s0 = __p0; \
	54780	__ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
	54781	__ret; \
	54782	})
	54783	#define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
	54784	int16_t __ret; \
	54785	int16_t __s0 = __p0; \
	54786	__ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
	54787	__ret; \
53120	54788	})
53121		#endif
53122
53123	54789	#ifdef __LITTLE_ENDIAN__
53124		#define vfmsq_lane_f64(__p0_512, __p1_512, __p2_512, __p3_512) __extension__ ({ \
53125		float64x2_t __ret_512; \
53126		float64x2_t __s0_512 = __p0_512; \
53127		float64x2_t __s1_512 = __p1_512; \
53128		float64x1_t __s2_512 = __p2_512; \
53129		__ret_512 = vfmaq_lane_f64(__s0_512, -__s1_512, __s2_512, __p3_512); \
53130		__ret_512; \
	54790	#define vqshrn_high_n_u32(__p0_685, __p1_685, __p2_685) __extension__ ({ \
	54791	uint16x8_t __ret_685; \
	54792	uint16x4_t __s0_685 = __p0_685; \
	54793	uint32x4_t __s1_685 = __p1_685; \
	54794	__ret_685 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_685), (uint16x4_t)(vqshrn_n_u32(__s1_685, __p2_685)))); \
	54795	__ret_685; \
53131	54796	})
53132	54797	#else
53133		#define vfmsq_lane_f64(__p0_513, __p1_513, __p2_513, __p3_513) __extension__ ({ \
53134		float64x2_t __ret_513; \
53135		float64x2_t __s0_513 = __p0_513; \
53136		float64x2_t __s1_513 = __p1_513; \
53137		float64x1_t __s2_513 = __p2_513; \
53138		float64x2_t __rev0_513; __rev0_513 = __builtin_shufflevector(__s0_513, __s0_513, 1, 0); \
53139		float64x2_t __rev1_513; __rev1_513 = __builtin_shufflevector(__s1_513, __s1_513, 1, 0); \
53140		__ret_513 = __noswap_vfmaq_lane_f64(__rev0_513, -__rev1_513, __s2_513, __p3_513); \
53141		__ret_513 = __builtin_shufflevector(__ret_513, __ret_513, 1, 0); \
53142		__ret_513; \
	54798	#define vqshrn_high_n_u32(__p0_686, __p1_686, __p2_686) __extension__ ({ \
	54799	uint16x8_t __ret_686; \
	54800	uint16x4_t __s0_686 = __p0_686; \
	54801	uint32x4_t __s1_686 = __p1_686; \
	54802	uint16x4_t __rev0_686; __rev0_686 = __builtin_shufflevector(__s0_686, __s0_686, 3, 2, 1, 0); \
	54803	uint32x4_t __rev1_686; __rev1_686 = __builtin_shufflevector(__s1_686, __s1_686, 3, 2, 1, 0); \
	54804	__ret_686 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_686), (uint16x4_t)(__noswap_vqshrn_n_u32(__rev1_686, __p2_686)))); \
	54805	__ret_686 = __builtin_shufflevector(__ret_686, __ret_686, 7, 6, 5, 4, 3, 2, 1, 0); \
	54806	__ret_686; \
53143	54807	})
53144	54808	#endif
53145	54809
53146	54810	#ifdef __LITTLE_ENDIAN__
53147		#define vfmsq_lane_f32(__p0_514, __p1_514, __p2_514, __p3_514) __extension__ ({ \
53148		float32x4_t __ret_514; \
53149		float32x4_t __s0_514 = __p0_514; \
53150		float32x4_t __s1_514 = __p1_514; \
53151		float32x2_t __s2_514 = __p2_514; \
53152		__ret_514 = vfmaq_lane_f32(__s0_514, -__s1_514, __s2_514, __p3_514); \
53153		__ret_514; \
	54811	#define vqshrn_high_n_u64(__p0_687, __p1_687, __p2_687) __extension__ ({ \
	54812	uint32x4_t __ret_687; \
	54813	uint32x2_t __s0_687 = __p0_687; \
	54814	uint64x2_t __s1_687 = __p1_687; \
	54815	__ret_687 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_687), (uint32x2_t)(vqshrn_n_u64(__s1_687, __p2_687)))); \
	54816	__ret_687; \
53154	54817	})
53155	54818	#else
53156		#define vfmsq_lane_f32(__p0_515, __p1_515, __p2_515, __p3_515) __extension__ ({ \
53157		float32x4_t __ret_515; \
53158		float32x4_t __s0_515 = __p0_515; \
53159		float32x4_t __s1_515 = __p1_515; \
53160		float32x2_t __s2_515 = __p2_515; \
53161		float32x4_t __rev0_515; __rev0_515 = __builtin_shufflevector(__s0_515, __s0_515, 3, 2, 1, 0); \
53162		float32x4_t __rev1_515; __rev1_515 = __builtin_shufflevector(__s1_515, __s1_515, 3, 2, 1, 0); \
53163		float32x2_t __rev2_515; __rev2_515 = __builtin_shufflevector(__s2_515, __s2_515, 1, 0); \
53164		__ret_515 = __noswap_vfmaq_lane_f32(__rev0_515, -__rev1_515, __rev2_515, __p3_515); \
53165		__ret_515 = __builtin_shufflevector(__ret_515, __ret_515, 3, 2, 1, 0); \
53166		__ret_515; \
	54819	#define vqshrn_high_n_u64(__p0_688, __p1_688, __p2_688) __extension__ ({ \
	54820	uint32x4_t __ret_688; \
	54821	uint32x2_t __s0_688 = __p0_688; \
	54822	uint64x2_t __s1_688 = __p1_688; \
	54823	uint32x2_t __rev0_688; __rev0_688 = __builtin_shufflevector(__s0_688, __s0_688, 1, 0); \
	54824	uint64x2_t __rev1_688; __rev1_688 = __builtin_shufflevector(__s1_688, __s1_688, 1, 0); \
	54825	__ret_688 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_688), (uint32x2_t)(__noswap_vqshrn_n_u64(__rev1_688, __p2_688)))); \
	54826	__ret_688 = __builtin_shufflevector(__ret_688, __ret_688, 3, 2, 1, 0); \
	54827	__ret_688; \
53167	54828	})
53168	54829	#endif
53169	54830
53170		#define vfms_lane_f64(__p0_516, __p1_516, __p2_516, __p3_516) __extension__ ({ \
53171		float64x1_t __ret_516; \
53172		float64x1_t __s0_516 = __p0_516; \
53173		float64x1_t __s1_516 = __p1_516; \
53174		float64x1_t __s2_516 = __p2_516; \
53175		__ret_516 = vfma_lane_f64(__s0_516, -__s1_516, __s2_516, __p3_516); \
53176		__ret_516; \
53177		})
53178	54831	#ifdef __LITTLE_ENDIAN__
53179		#define vfms_lane_f32(__p0_517, __p1_517, __p2_517, __p3_517) __extension__ ({ \
53180		float32x2_t __ret_517; \
53181		float32x2_t __s0_517 = __p0_517; \
53182		float32x2_t __s1_517 = __p1_517; \
53183		float32x2_t __s2_517 = __p2_517; \
53184		__ret_517 = vfma_lane_f32(__s0_517, -__s1_517, __s2_517, __p3_517); \
53185		__ret_517; \
	54832	#define vqshrn_high_n_u16(__p0_689, __p1_689, __p2_689) __extension__ ({ \
	54833	uint8x16_t __ret_689; \
	54834	uint8x8_t __s0_689 = __p0_689; \
	54835	uint16x8_t __s1_689 = __p1_689; \
	54836	__ret_689 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_689), (uint8x8_t)(vqshrn_n_u16(__s1_689, __p2_689)))); \
	54837	__ret_689; \
53186	54838	})
53187	54839	#else
53188		#define vfms_lane_f32(__p0_518, __p1_518, __p2_518, __p3_518) __extension__ ({ \
53189		float32x2_t __ret_518; \
53190		float32x2_t __s0_518 = __p0_518; \
53191		float32x2_t __s1_518 = __p1_518; \
53192		float32x2_t __s2_518 = __p2_518; \
53193		float32x2_t __rev0_518; __rev0_518 = __builtin_shufflevector(__s0_518, __s0_518, 1, 0); \
53194		float32x2_t __rev1_518; __rev1_518 = __builtin_shufflevector(__s1_518, __s1_518, 1, 0); \
53195		float32x2_t __rev2_518; __rev2_518 = __builtin_shufflevector(__s2_518, __s2_518, 1, 0); \
53196		__ret_518 = __noswap_vfma_lane_f32(__rev0_518, -__rev1_518, __rev2_518, __p3_518); \
53197		__ret_518 = __builtin_shufflevector(__ret_518, __ret_518, 1, 0); \
53198		__ret_518; \
	54840	#define vqshrn_high_n_u16(__p0_690, __p1_690, __p2_690) __extension__ ({ \
	54841	uint8x16_t __ret_690; \
	54842	uint8x8_t __s0_690 = __p0_690; \
	54843	uint16x8_t __s1_690 = __p1_690; \
	54844	uint8x8_t __rev0_690; __rev0_690 = __builtin_shufflevector(__s0_690, __s0_690, 7, 6, 5, 4, 3, 2, 1, 0); \
	54845	uint16x8_t __rev1_690; __rev1_690 = __builtin_shufflevector(__s1_690, __s1_690, 7, 6, 5, 4, 3, 2, 1, 0); \
	54846	__ret_690 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_690), (uint8x8_t)(__noswap_vqshrn_n_u16(__rev1_690, __p2_690)))); \
	54847	__ret_690 = __builtin_shufflevector(__ret_690, __ret_690, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	54848	__ret_690; \
53199	54849	})
53200	54850	#endif
53201	54851
53202	54852	#ifdef __LITTLE_ENDIAN__
53203		#define vfmsd_laneq_f64(__p0_519, __p1_519, __p2_519, __p3_519) __extension__ ({ \
53204		float64_t __ret_519; \
53205		float64_t __s0_519 = __p0_519; \
53206		float64_t __s1_519 = __p1_519; \
53207		float64x2_t __s2_519 = __p2_519; \
53208		__ret_519 = vfmad_laneq_f64(__s0_519, -__s1_519, __s2_519, __p3_519); \
53209		__ret_519; \
	54853	#define vqshrn_high_n_s32(__p0_691, __p1_691, __p2_691) __extension__ ({ \
	54854	int16x8_t __ret_691; \
	54855	int16x4_t __s0_691 = __p0_691; \
	54856	int32x4_t __s1_691 = __p1_691; \
	54857	__ret_691 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_691), (int16x4_t)(vqshrn_n_s32(__s1_691, __p2_691)))); \
	54858	__ret_691; \
53210	54859	})
53211	54860	#else
53212		#define vfmsd_laneq_f64(__p0_520, __p1_520, __p2_520, __p3_520) __extension__ ({ \
53213		float64_t __ret_520; \
53214		float64_t __s0_520 = __p0_520; \
53215		float64_t __s1_520 = __p1_520; \
53216		float64x2_t __s2_520 = __p2_520; \
53217		float64x2_t __rev2_520; __rev2_520 = __builtin_shufflevector(__s2_520, __s2_520, 1, 0); \
53218		__ret_520 = __noswap_vfmad_laneq_f64(__s0_520, -__s1_520, __rev2_520, __p3_520); \
53219		__ret_520; \
	54861	#define vqshrn_high_n_s32(__p0_692, __p1_692, __p2_692) __extension__ ({ \
	54862	int16x8_t __ret_692; \
	54863	int16x4_t __s0_692 = __p0_692; \
	54864	int32x4_t __s1_692 = __p1_692; \
	54865	int16x4_t __rev0_692; __rev0_692 = __builtin_shufflevector(__s0_692, __s0_692, 3, 2, 1, 0); \
	54866	int32x4_t __rev1_692; __rev1_692 = __builtin_shufflevector(__s1_692, __s1_692, 3, 2, 1, 0); \
	54867	__ret_692 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_692), (int16x4_t)(__noswap_vqshrn_n_s32(__rev1_692, __p2_692)))); \
	54868	__ret_692 = __builtin_shufflevector(__ret_692, __ret_692, 7, 6, 5, 4, 3, 2, 1, 0); \
	54869	__ret_692; \
53220	54870	})
53221	54871	#endif
53222	54872
53223	54873	#ifdef __LITTLE_ENDIAN__
53224		#define vfmss_laneq_f32(__p0_521, __p1_521, __p2_521, __p3_521) __extension__ ({ \
53225		float32_t __ret_521; \
53226		float32_t __s0_521 = __p0_521; \
53227		float32_t __s1_521 = __p1_521; \
53228		float32x4_t __s2_521 = __p2_521; \
53229		__ret_521 = vfmas_laneq_f32(__s0_521, -__s1_521, __s2_521, __p3_521); \
53230		__ret_521; \
	54874	#define vqshrn_high_n_s64(__p0_693, __p1_693, __p2_693) __extension__ ({ \
	54875	int32x4_t __ret_693; \
	54876	int32x2_t __s0_693 = __p0_693; \
	54877	int64x2_t __s1_693 = __p1_693; \
	54878	__ret_693 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_693), (int32x2_t)(vqshrn_n_s64(__s1_693, __p2_693)))); \
	54879	__ret_693; \
53231	54880	})
53232	54881	#else
53233		#define vfmss_laneq_f32(__p0_522, __p1_522, __p2_522, __p3_522) __extension__ ({ \
53234		float32_t __ret_522; \
53235		float32_t __s0_522 = __p0_522; \
53236		float32_t __s1_522 = __p1_522; \
53237		float32x4_t __s2_522 = __p2_522; \
53238		float32x4_t __rev2_522; __rev2_522 = __builtin_shufflevector(__s2_522, __s2_522, 3, 2, 1, 0); \
53239		__ret_522 = __noswap_vfmas_laneq_f32(__s0_522, -__s1_522, __rev2_522, __p3_522); \
53240		__ret_522; \
	54882	#define vqshrn_high_n_s64(__p0_694, __p1_694, __p2_694) __extension__ ({ \
	54883	int32x4_t __ret_694; \
	54884	int32x2_t __s0_694 = __p0_694; \
	54885	int64x2_t __s1_694 = __p1_694; \
	54886	int32x2_t __rev0_694; __rev0_694 = __builtin_shufflevector(__s0_694, __s0_694, 1, 0); \
	54887	int64x2_t __rev1_694; __rev1_694 = __builtin_shufflevector(__s1_694, __s1_694, 1, 0); \
	54888	__ret_694 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_694), (int32x2_t)(__noswap_vqshrn_n_s64(__rev1_694, __p2_694)))); \
	54889	__ret_694 = __builtin_shufflevector(__ret_694, __ret_694, 3, 2, 1, 0); \
	54890	__ret_694; \
53241	54891	})
53242	54892	#endif
53243	54893
53244	54894	#ifdef __LITTLE_ENDIAN__
53245		#define vfmsq_laneq_f64(__p0_523, __p1_523, __p2_523, __p3_523) __extension__ ({ \
53246		float64x2_t __ret_523; \
53247		float64x2_t __s0_523 = __p0_523; \
53248		float64x2_t __s1_523 = __p1_523; \
53249		float64x2_t __s2_523 = __p2_523; \
53250		__ret_523 = vfmaq_laneq_f64(__s0_523, -__s1_523, __s2_523, __p3_523); \
53251		__ret_523; \
	54895	#define vqshrn_high_n_s16(__p0_695, __p1_695, __p2_695) __extension__ ({ \
	54896	int8x16_t __ret_695; \
	54897	int8x8_t __s0_695 = __p0_695; \
	54898	int16x8_t __s1_695 = __p1_695; \
	54899	__ret_695 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_695), (int8x8_t)(vqshrn_n_s16(__s1_695, __p2_695)))); \
	54900	__ret_695; \
53252	54901	})
53253	54902	#else
53254		#define vfmsq_laneq_f64(__p0_524, __p1_524, __p2_524, __p3_524) __extension__ ({ \
53255		float64x2_t __ret_524; \
53256		float64x2_t __s0_524 = __p0_524; \
53257		float64x2_t __s1_524 = __p1_524; \
53258		float64x2_t __s2_524 = __p2_524; \
53259		float64x2_t __rev0_524; __rev0_524 = __builtin_shufflevector(__s0_524, __s0_524, 1, 0); \
53260		float64x2_t __rev1_524; __rev1_524 = __builtin_shufflevector(__s1_524, __s1_524, 1, 0); \
53261		float64x2_t __rev2_524; __rev2_524 = __builtin_shufflevector(__s2_524, __s2_524, 1, 0); \
53262		__ret_524 = __noswap_vfmaq_laneq_f64(__rev0_524, -__rev1_524, __rev2_524, __p3_524); \
53263		__ret_524 = __builtin_shufflevector(__ret_524, __ret_524, 1, 0); \
53264		__ret_524; \
	54903	#define vqshrn_high_n_s16(__p0_696, __p1_696, __p2_696) __extension__ ({ \
	54904	int8x16_t __ret_696; \
	54905	int8x8_t __s0_696 = __p0_696; \
	54906	int16x8_t __s1_696 = __p1_696; \
	54907	int8x8_t __rev0_696; __rev0_696 = __builtin_shufflevector(__s0_696, __s0_696, 7, 6, 5, 4, 3, 2, 1, 0); \
	54908	int16x8_t __rev1_696; __rev1_696 = __builtin_shufflevector(__s1_696, __s1_696, 7, 6, 5, 4, 3, 2, 1, 0); \
	54909	__ret_696 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_696), (int8x8_t)(__noswap_vqshrn_n_s16(__rev1_696, __p2_696)))); \
	54910	__ret_696 = __builtin_shufflevector(__ret_696, __ret_696, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	54911	__ret_696; \
53265	54912	})
53266	54913	#endif
53267	54914
	54915	#define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
	54916	uint16_t __ret; \
	54917	uint32_t __s0 = __p0; \
	54918	__ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
	54919	__ret; \
	54920	})
	54921	#define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
	54922	uint32_t __ret; \
	54923	uint64_t __s0 = __p0; \
	54924	__ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
	54925	__ret; \
	54926	})
	54927	#define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
	54928	uint8_t __ret; \
	54929	uint16_t __s0 = __p0; \
	54930	__ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
	54931	__ret; \
	54932	})
	54933	#define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
	54934	int16_t __ret; \
	54935	int32_t __s0 = __p0; \
	54936	__ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
	54937	__ret; \
	54938	})
	54939	#define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
	54940	int32_t __ret; \
	54941	int64_t __s0 = __p0; \
	54942	__ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
	54943	__ret; \
	54944	})
	54945	#define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
	54946	int8_t __ret; \
	54947	int16_t __s0 = __p0; \
	54948	__ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
	54949	__ret; \
	54950	})
53268	54951	#ifdef __LITTLE_ENDIAN__
53269		#define vfmsq_laneq_f32(__p0_525, __p1_525, __p2_525, __p3_525) __extension__ ({ \
53270		float32x4_t __ret_525; \
53271		float32x4_t __s0_525 = __p0_525; \
53272		float32x4_t __s1_525 = __p1_525; \
53273		float32x4_t __s2_525 = __p2_525; \
53274		__ret_525 = vfmaq_laneq_f32(__s0_525, -__s1_525, __s2_525, __p3_525); \
53275		__ret_525; \
	54952	#define vqshrun_high_n_s32(__p0_697, __p1_697, __p2_697) __extension__ ({ \
	54953	int16x8_t __ret_697; \
	54954	int16x4_t __s0_697 = __p0_697; \
	54955	int32x4_t __s1_697 = __p1_697; \
	54956	__ret_697 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_697), (int16x4_t)(vqshrun_n_s32(__s1_697, __p2_697)))); \
	54957	__ret_697; \
53276	54958	})
53277	54959	#else
53278		#define vfmsq_laneq_f32(__p0_526, __p1_526, __p2_526, __p3_526) __extension__ ({ \
53279		float32x4_t __ret_526; \
53280		float32x4_t __s0_526 = __p0_526; \
53281		float32x4_t __s1_526 = __p1_526; \
53282		float32x4_t __s2_526 = __p2_526; \
53283		float32x4_t __rev0_526; __rev0_526 = __builtin_shufflevector(__s0_526, __s0_526, 3, 2, 1, 0); \
53284		float32x4_t __rev1_526; __rev1_526 = __builtin_shufflevector(__s1_526, __s1_526, 3, 2, 1, 0); \
53285		float32x4_t __rev2_526; __rev2_526 = __builtin_shufflevector(__s2_526, __s2_526, 3, 2, 1, 0); \
53286		__ret_526 = __noswap_vfmaq_laneq_f32(__rev0_526, -__rev1_526, __rev2_526, __p3_526); \
53287		__ret_526 = __builtin_shufflevector(__ret_526, __ret_526, 3, 2, 1, 0); \
53288		__ret_526; \
	54960	#define vqshrun_high_n_s32(__p0_698, __p1_698, __p2_698) __extension__ ({ \
	54961	int16x8_t __ret_698; \
	54962	int16x4_t __s0_698 = __p0_698; \
	54963	int32x4_t __s1_698 = __p1_698; \
	54964	int16x4_t __rev0_698; __rev0_698 = __builtin_shufflevector(__s0_698, __s0_698, 3, 2, 1, 0); \
	54965	int32x4_t __rev1_698; __rev1_698 = __builtin_shufflevector(__s1_698, __s1_698, 3, 2, 1, 0); \
	54966	__ret_698 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_698), (int16x4_t)(__noswap_vqshrun_n_s32(__rev1_698, __p2_698)))); \
	54967	__ret_698 = __builtin_shufflevector(__ret_698, __ret_698, 7, 6, 5, 4, 3, 2, 1, 0); \
	54968	__ret_698; \
53289	54969	})
53290	54970	#endif
53291	54971
53292	54972	#ifdef __LITTLE_ENDIAN__
53293		#define vfms_laneq_f64(__p0_527, __p1_527, __p2_527, __p3_527) __extension__ ({ \
53294		float64x1_t __ret_527; \
53295		float64x1_t __s0_527 = __p0_527; \
53296		float64x1_t __s1_527 = __p1_527; \
53297		float64x2_t __s2_527 = __p2_527; \
53298		__ret_527 = vfma_laneq_f64(__s0_527, -__s1_527, __s2_527, __p3_527); \
53299		__ret_527; \
	54973	#define vqshrun_high_n_s64(__p0_699, __p1_699, __p2_699) __extension__ ({ \
	54974	int32x4_t __ret_699; \
	54975	int32x2_t __s0_699 = __p0_699; \
	54976	int64x2_t __s1_699 = __p1_699; \
	54977	__ret_699 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_699), (int32x2_t)(vqshrun_n_s64(__s1_699, __p2_699)))); \
	54978	__ret_699; \
53300	54979	})
53301	54980	#else
53302		#define vfms_laneq_f64(__p0_528, __p1_528, __p2_528, __p3_528) __extension__ ({ \
53303		float64x1_t __ret_528; \
53304		float64x1_t __s0_528 = __p0_528; \
53305		float64x1_t __s1_528 = __p1_528; \
53306		float64x2_t __s2_528 = __p2_528; \
53307		float64x2_t __rev2_528; __rev2_528 = __builtin_shufflevector(__s2_528, __s2_528, 1, 0); \
53308		__ret_528 = __noswap_vfma_laneq_f64(__s0_528, -__s1_528, __rev2_528, __p3_528); \
53309		__ret_528; \
	54981	#define vqshrun_high_n_s64(__p0_700, __p1_700, __p2_700) __extension__ ({ \
	54982	int32x4_t __ret_700; \
	54983	int32x2_t __s0_700 = __p0_700; \
	54984	int64x2_t __s1_700 = __p1_700; \
	54985	int32x2_t __rev0_700; __rev0_700 = __builtin_shufflevector(__s0_700, __s0_700, 1, 0); \
	54986	int64x2_t __rev1_700; __rev1_700 = __builtin_shufflevector(__s1_700, __s1_700, 1, 0); \
	54987	__ret_700 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_700), (int32x2_t)(__noswap_vqshrun_n_s64(__rev1_700, __p2_700)))); \
	54988	__ret_700 = __builtin_shufflevector(__ret_700, __ret_700, 3, 2, 1, 0); \
	54989	__ret_700; \
53310	54990	})
53311	54991	#endif
53312	54992
53313	54993	#ifdef __LITTLE_ENDIAN__
53314		#define vfms_laneq_f32(__p0_529, __p1_529, __p2_529, __p3_529) __extension__ ({ \
53315		float32x2_t __ret_529; \
53316		float32x2_t __s0_529 = __p0_529; \
53317		float32x2_t __s1_529 = __p1_529; \
53318		float32x4_t __s2_529 = __p2_529; \
53319		__ret_529 = vfma_laneq_f32(__s0_529, -__s1_529, __s2_529, __p3_529); \
53320		__ret_529; \
	54994	#define vqshrun_high_n_s16(__p0_701, __p1_701, __p2_701) __extension__ ({ \
	54995	int8x16_t __ret_701; \
	54996	int8x8_t __s0_701 = __p0_701; \
	54997	int16x8_t __s1_701 = __p1_701; \
	54998	__ret_701 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_701), (int8x8_t)(vqshrun_n_s16(__s1_701, __p2_701)))); \
	54999	__ret_701; \
53321	55000	})
53322	55001	#else
53323		#define vfms_laneq_f32(__p0_530, __p1_530, __p2_530, __p3_530) __extension__ ({ \
53324		float32x2_t __ret_530; \
53325		float32x2_t __s0_530 = __p0_530; \
53326		float32x2_t __s1_530 = __p1_530; \
53327		float32x4_t __s2_530 = __p2_530; \
53328		float32x2_t __rev0_530; __rev0_530 = __builtin_shufflevector(__s0_530, __s0_530, 1, 0); \
53329		float32x2_t __rev1_530; __rev1_530 = __builtin_shufflevector(__s1_530, __s1_530, 1, 0); \
53330		float32x4_t __rev2_530; __rev2_530 = __builtin_shufflevector(__s2_530, __s2_530, 3, 2, 1, 0); \
53331		__ret_530 = __noswap_vfma_laneq_f32(__rev0_530, -__rev1_530, __rev2_530, __p3_530); \
53332		__ret_530 = __builtin_shufflevector(__ret_530, __ret_530, 1, 0); \
53333		__ret_530; \
	55002	#define vqshrun_high_n_s16(__p0_702, __p1_702, __p2_702) __extension__ ({ \
	55003	int8x16_t __ret_702; \
	55004	int8x8_t __s0_702 = __p0_702; \
	55005	int16x8_t __s1_702 = __p1_702; \
	55006	int8x8_t __rev0_702; __rev0_702 = __builtin_shufflevector(__s0_702, __s0_702, 7, 6, 5, 4, 3, 2, 1, 0); \
	55007	int16x8_t __rev1_702; __rev1_702 = __builtin_shufflevector(__s1_702, __s1_702, 7, 6, 5, 4, 3, 2, 1, 0); \
	55008	__ret_702 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_702), (int8x8_t)(__noswap_vqshrun_n_s16(__rev1_702, __p2_702)))); \
	55009	__ret_702 = __builtin_shufflevector(__ret_702, __ret_702, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	55010	__ret_702; \
53334	55011	})
53335	55012	#endif
53336	55013
	55014	#define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
	55015	int16_t __ret; \
	55016	int32_t __s0 = __p0; \
	55017	__ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
	55018	__ret; \
	55019	})
	55020	#define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
	55021	int32_t __ret; \
	55022	int64_t __s0 = __p0; \
	55023	__ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
	55024	__ret; \
	55025	})
	55026	#define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
	55027	int8_t __ret; \
	55028	int16_t __s0 = __p0; \
	55029	__ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
	55030	__ret; \
	55031	})
	55032	__ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
	55033	uint8_t __ret;
	55034	__ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
	55035	return __ret;
	55036	}
	55037	__ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
	55038	uint32_t __ret;
	55039	__ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
	55040	return __ret;
	55041	}
	55042	__ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
	55043	uint64_t __ret;
	55044	__ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
	55045	return __ret;
	55046	}
	55047	__ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
	55048	uint16_t __ret;
	55049	__ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
	55050	return __ret;
	55051	}
	55052	__ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
	55053	int8_t __ret;
	55054	__ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
	55055	return __ret;
	55056	}
	55057	__ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
	55058	int32_t __ret;
	55059	__ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
	55060	return __ret;
	55061	}
	55062	__ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
	55063	int64_t __ret;
	55064	__ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
	55065	return __ret;
	55066	}
	55067	__ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
	55068	int16_t __ret;
	55069	__ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
	55070	return __ret;
	55071	}
53337	55072	#ifdef __LITTLE_ENDIAN__
53338		__ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53339		float64x2_t __ret;
53340		__ret = vfmaq_f64(__p0, -__p1, (float64x2_t) {__p2, __p2});
	55073	__ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
	55074	poly8x8_t __ret;
	55075	__ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 4);
53341	55076	return __ret;
53342	55077	}
53343	55078	#else
53344		__ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53345		float64x2_t __ret;
53346		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53347		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53348		__ret = __noswap_vfmaq_f64(__rev0, -__rev1, (float64x2_t) {__p2, __p2});
53349		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	55079	__ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
	55080	poly8x8_t __ret;
	55081	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55082	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55083	__ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 4);
	55084	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
53350	55085	return __ret;
53351	55086	}
53352	55087	#endif
53353	55088
53354	55089	#ifdef __LITTLE_ENDIAN__
53355		__ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
53356		float32x4_t __ret;
53357		__ret = vfmaq_f32(__p0, -__p1, (float32x4_t) {__p2, __p2, __p2, __p2});
	55090	__ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
	55091	poly8x16_t __ret;
	55092	__ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 36);
53358	55093	return __ret;
53359	55094	}
53360	55095	#else
53361		__ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
53362		float32x4_t __ret;
53363		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
53364		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53365		__ret = __noswap_vfmaq_f32(__rev0, -__rev1, (float32x4_t) {__p2, __p2, __p2, __p2});
53366		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	55096	__ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
	55097	poly8x16_t __ret;
	55098	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55099	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55100	__ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 36);
	55101	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53367	55102	return __ret;
53368	55103	}
53369	55104	#endif
53370	55105
53371		__ai float64x1_t vfms_n_f64(float64x1_t __p0, float64x1_t __p1, float64_t __p2) {
53372		float64x1_t __ret;
53373		__ret = vfma_f64(__p0, -__p1, (float64x1_t) {__p2});
53374		return __ret;
53375		}
53376	55106	#ifdef __LITTLE_ENDIAN__
53377		__ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
53378		float32x2_t __ret;
53379		__ret = vfma_f32(__p0, -__p1, (float32x2_t) {__p2, __p2});
	55107	__ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	55108	uint8x16_t __ret;
	55109	__ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
53380	55110	return __ret;
53381	55111	}
53382	55112	#else
53383		__ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
53384		float32x2_t __ret;
53385		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53386		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53387		__ret = __noswap_vfma_f32(__rev0, -__rev1, (float32x2_t) {__p2, __p2});
53388		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	55113	__ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	55114	uint8x16_t __ret;
	55115	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55116	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55117	__ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	55118	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53389	55119	return __ret;
53390	55120	}
53391	55121	#endif
53392	55122
53393	55123	#ifdef __LITTLE_ENDIAN__
53394		__ai poly64x1_t vget_high_p64(poly64x2_t __p0) {
53395		poly64x1_t __ret;
53396		__ret = __builtin_shufflevector(__p0, __p0, 1);
	55124	__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
	55125	int8x16_t __ret;
	55126	__ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
53397	55127	return __ret;
53398	55128	}
53399	55129	#else
53400		__ai poly64x1_t vget_high_p64(poly64x2_t __p0) {
53401		poly64x1_t __ret;
53402		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53403		__ret = __builtin_shufflevector(__rev0, __rev0, 1);
	55130	__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
	55131	int8x16_t __ret;
	55132	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55133	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55134	__ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
	55135	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53404	55136	return __ret;
53405	55137	}
53406		__ai poly64x1_t __noswap_vget_high_p64(poly64x2_t __p0) {
53407		poly64x1_t __ret;
53408		__ret = __builtin_shufflevector(__p0, __p0, 1);
	55138	#endif
	55139
	55140	#ifdef __LITTLE_ENDIAN__
	55141	__ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
	55142	uint8x8_t __ret;
	55143	__ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 16);
	55144	return __ret;
	55145	}
	55146	#else
	55147	__ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
	55148	uint8x8_t __ret;
	55149	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55150	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55151	__ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 16);
	55152	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
53409	55153	return __ret;
53410	55154	}
53411	55155	#endif
53412	55156
53413	55157	#ifdef __LITTLE_ENDIAN__
53414		__ai float64x1_t vget_high_f64(float64x2_t __p0) {
53415		float64x1_t __ret;
53416		__ret = __builtin_shufflevector(__p0, __p0, 1);
	55158	__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
	55159	int8x8_t __ret;
	55160	__ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 0);
53417	55161	return __ret;
53418	55162	}
53419	55163	#else
53420		__ai float64x1_t vget_high_f64(float64x2_t __p0) {
53421		float64x1_t __ret;
53422		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53423		__ret = __builtin_shufflevector(__rev0, __rev0, 1);
	55164	__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
	55165	int8x8_t __ret;
	55166	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55167	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55168	__ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 0);
	55169	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
53424	55170	return __ret;
53425	55171	}
53426	55172	#endif
53427	55173
53428		#define vget_lane_p64(__p0, __p1) __extension__ ({ \
53429		poly64_t __ret; \
53430		poly64x1_t __s0 = __p0; \
53431		__ret = (poly64_t) __builtin_neon_vget_lane_i64((poly64x1_t)__s0, __p1); \
53432		__ret; \
53433		})
53434	55174	#ifdef __LITTLE_ENDIAN__
53435		#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
53436		poly64_t __ret; \
53437		poly64x2_t __s0 = __p0; \
53438		__ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
53439		__ret; \
53440		})
	55175	__ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
	55176	poly8x8_t __ret;
	55177	__ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 4);
	55178	return __ret;
	55179	}
53441	55180	#else
53442		#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
53443		poly64_t __ret; \
53444		poly64x2_t __s0 = __p0; \
53445		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
53446		__ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__rev0, __p1); \
53447		__ret; \
53448		})
53449		#define __noswap_vgetq_lane_p64(__p0, __p1) __extension__ ({ \
53450		poly64_t __ret; \
53451		poly64x2_t __s0 = __p0; \
53452		__ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
53453		__ret; \
53454		})
	55181	__ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
	55182	poly8x8_t __ret;
	55183	poly8x16x2_t __rev0;
	55184	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55185	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55186	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55187	__ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 4);
	55188	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55189	return __ret;
	55190	}
53455	55191	#endif
53456	55192
53457	55193	#ifdef __LITTLE_ENDIAN__
53458		#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
53459		float64_t __ret; \
53460		float64x2_t __s0 = __p0; \
53461		__ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
53462		__ret; \
53463		})
	55194	__ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
	55195	poly8x16_t __ret;
	55196	__ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 36);
	55197	return __ret;
	55198	}
53464	55199	#else
53465		#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
53466		float64_t __ret; \
53467		float64x2_t __s0 = __p0; \
53468		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
53469		__ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__rev0, __p1); \
53470		__ret; \
53471		})
53472		#define __noswap_vgetq_lane_f64(__p0, __p1) __extension__ ({ \
53473		float64_t __ret; \
53474		float64x2_t __s0 = __p0; \
53475		__ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
53476		__ret; \
53477		})
	55200	__ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
	55201	poly8x16_t __ret;
	55202	poly8x16x2_t __rev0;
	55203	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55204	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55205	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55206	__ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 36);
	55207	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55208	return __ret;
	55209	}
53478	55210	#endif
53479	55211
53480		#define vget_lane_f64(__p0, __p1) __extension__ ({ \
53481		float64_t __ret; \
53482		float64x1_t __s0 = __p0; \
53483		__ret = (float64_t) __builtin_neon_vget_lane_f64((float64x1_t)__s0, __p1); \
53484		__ret; \
53485		})
53486	55212	#ifdef __LITTLE_ENDIAN__
53487		__ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
53488		poly64x1_t __ret;
53489		__ret = __builtin_shufflevector(__p0, __p0, 0);
	55213	__ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
	55214	uint8x16_t __ret;
	55215	__ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 48);
53490	55216	return __ret;
53491	55217	}
53492	55218	#else
53493		__ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
53494		poly64x1_t __ret;
53495		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53496		__ret = __builtin_shufflevector(__rev0, __rev0, 0);
	55219	__ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
	55220	uint8x16_t __ret;
	55221	uint8x16x2_t __rev0;
	55222	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55223	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55224	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55225	__ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 48);
	55226	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53497	55227	return __ret;
53498	55228	}
53499	55229	#endif
53500	55230
53501	55231	#ifdef __LITTLE_ENDIAN__
53502		__ai float64x1_t vget_low_f64(float64x2_t __p0) {
53503		float64x1_t __ret;
53504		__ret = __builtin_shufflevector(__p0, __p0, 0);
	55232	__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
	55233	int8x16_t __ret;
	55234	__ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 32);
53505	55235	return __ret;
53506	55236	}
53507	55237	#else
53508		__ai float64x1_t vget_low_f64(float64x2_t __p0) {
53509		float64x1_t __ret;
53510		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
53511		__ret = __builtin_shufflevector(__rev0, __rev0, 0);
	55238	__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
	55239	int8x16_t __ret;
	55240	int8x16x2_t __rev0;
	55241	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55242	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55243	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55244	__ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 32);
	55245	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53512	55246	return __ret;
53513	55247	}
53514	55248	#endif
53515	55249
53516		#define vld1_p64(__p0) __extension__ ({ \
53517		poly64x1_t __ret; \
53518		__ret = (poly64x1_t) __builtin_neon_vld1_v(__p0, 6); \
53519		__ret; \
53520		})
53521		#ifdef __LITTLE_ENDIAN__
53522		#define vld1q_p64(__p0) __extension__ ({ \
53523		poly64x2_t __ret; \
53524		__ret = (poly64x2_t) __builtin_neon_vld1q_v(__p0, 38); \
53525		__ret; \
53526		})
53527		#else
53528		#define vld1q_p64(__p0) __extension__ ({ \
53529		poly64x2_t __ret; \
53530		__ret = (poly64x2_t) __builtin_neon_vld1q_v(__p0, 38); \
53531		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
53532		__ret; \
53533		})
53534		#endif
53535
53536	55250	#ifdef __LITTLE_ENDIAN__
53537		#define vld1q_f64(__p0) __extension__ ({ \
53538		float64x2_t __ret; \
53539		__ret = (float64x2_t) __builtin_neon_vld1q_v(__p0, 42); \
53540		__ret; \
53541		})
	55251	__ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
	55252	uint8x8_t __ret;
	55253	__ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 16);
	55254	return __ret;
	55255	}
53542	55256	#else
53543		#define vld1q_f64(__p0) __extension__ ({ \
53544		float64x2_t __ret; \
53545		__ret = (float64x2_t) __builtin_neon_vld1q_v(__p0, 42); \
53546		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
53547		__ret; \
53548		})
	55257	__ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
	55258	uint8x8_t __ret;
	55259	uint8x16x2_t __rev0;
	55260	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55261	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55262	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55263	__ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 16);
	55264	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55265	return __ret;
	55266	}
53549	55267	#endif
53550	55268
53551		#define vld1_f64(__p0) __extension__ ({ \
53552		float64x1_t __ret; \
53553		__ret = (float64x1_t) __builtin_neon_vld1_v(__p0, 10); \
53554		__ret; \
53555		})
53556		#define vld1_dup_p64(__p0) __extension__ ({ \
53557		poly64x1_t __ret; \
53558		__ret = (poly64x1_t) __builtin_neon_vld1_dup_v(__p0, 6); \
53559		__ret; \
53560		})
53561	55269	#ifdef __LITTLE_ENDIAN__
53562		#define vld1q_dup_p64(__p0) __extension__ ({ \
53563		poly64x2_t __ret; \
53564		__ret = (poly64x2_t) __builtin_neon_vld1q_dup_v(__p0, 38); \
53565		__ret; \
53566		})
	55270	__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
	55271	int8x8_t __ret;
	55272	__ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 0);
	55273	return __ret;
	55274	}
53567	55275	#else
53568		#define vld1q_dup_p64(__p0) __extension__ ({ \
53569		poly64x2_t __ret; \
53570		__ret = (poly64x2_t) __builtin_neon_vld1q_dup_v(__p0, 38); \
53571		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
53572		__ret; \
53573		})
	55276	__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
	55277	int8x8_t __ret;
	55278	int8x16x2_t __rev0;
	55279	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55280	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55281	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55282	__ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 0);
	55283	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55284	return __ret;
	55285	}
53574	55286	#endif
53575	55287
53576	55288	#ifdef __LITTLE_ENDIAN__
53577		#define vld1q_dup_f64(__p0) __extension__ ({ \
53578		float64x2_t __ret; \
53579		__ret = (float64x2_t) __builtin_neon_vld1q_dup_v(__p0, 42); \
53580		__ret; \
53581		})
	55289	__ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
	55290	poly8x8_t __ret;
	55291	__ret = (poly8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 4);
	55292	return __ret;
	55293	}
53582	55294	#else
53583		#define vld1q_dup_f64(__p0) __extension__ ({ \
53584		float64x2_t __ret; \
53585		__ret = (float64x2_t) __builtin_neon_vld1q_dup_v(__p0, 42); \
53586		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
53587		__ret; \
53588		})
	55295	__ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
	55296	poly8x8_t __ret;
	55297	poly8x16x3_t __rev0;
	55298	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55299	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55300	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55301	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55302	__ret = (poly8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 4);
	55303	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55304	return __ret;
	55305	}
53589	55306	#endif
53590	55307
53591		#define vld1_dup_f64(__p0) __extension__ ({ \
53592		float64x1_t __ret; \
53593		__ret = (float64x1_t) __builtin_neon_vld1_dup_v(__p0, 10); \
53594		__ret; \
53595		})
53596		#define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53597		poly64x1_t __ret; \
53598		poly64x1_t __s1 = __p1; \
53599		__ret = (poly64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
53600		__ret; \
53601		})
53602	55308	#ifdef __LITTLE_ENDIAN__
53603		#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53604		poly64x2_t __ret; \
53605		poly64x2_t __s1 = __p1; \
53606		__ret = (poly64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 38); \
53607		__ret; \
53608		})
	55309	__ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
	55310	poly8x16_t __ret;
	55311	__ret = (poly8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 36);
	55312	return __ret;
	55313	}
53609	55314	#else
53610		#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53611		poly64x2_t __ret; \
53612		poly64x2_t __s1 = __p1; \
53613		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
53614		__ret = (poly64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 38); \
53615		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
53616		__ret; \
53617		})
	55315	__ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
	55316	poly8x16_t __ret;
	55317	poly8x16x3_t __rev0;
	55318	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55319	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55320	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55321	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55322	__ret = (poly8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 36);
	55323	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55324	return __ret;
	55325	}
53618	55326	#endif
53619	55327
53620	55328	#ifdef __LITTLE_ENDIAN__
53621		#define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53622		float64x2_t __ret; \
53623		float64x2_t __s1 = __p1; \
53624		__ret = (float64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 42); \
53625		__ret; \
53626		})
	55329	__ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
	55330	uint8x16_t __ret;
	55331	__ret = (uint8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 48);
	55332	return __ret;
	55333	}
53627	55334	#else
53628		#define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53629		float64x2_t __ret; \
53630		float64x2_t __s1 = __p1; \
53631		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
53632		__ret = (float64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 42); \
53633		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
53634		__ret; \
53635		})
	55335	__ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
	55336	uint8x16_t __ret;
	55337	uint8x16x3_t __rev0;
	55338	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55339	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55340	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55341	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55342	__ret = (uint8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 48);
	55343	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55344	return __ret;
	55345	}
53636	55346	#endif
53637	55347
53638		#define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53639		float64x1_t __ret; \
53640		float64x1_t __s1 = __p1; \
53641		__ret = (float64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
53642		__ret; \
53643		})
53644		#define vld1_p64_x2(__p0) __extension__ ({ \
53645		poly64x1x2_t __ret; \
53646		__builtin_neon_vld1_x2_v(&__ret, __p0, 6); \
53647		__ret; \
53648		})
53649	55348	#ifdef __LITTLE_ENDIAN__
53650		#define vld1q_p64_x2(__p0) __extension__ ({ \
53651		poly64x2x2_t __ret; \
53652		__builtin_neon_vld1q_x2_v(&__ret, __p0, 38); \
53653		__ret; \
53654		})
	55349	__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
	55350	int8x16_t __ret;
	55351	__ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 32);
	55352	return __ret;
	55353	}
53655	55354	#else
53656		#define vld1q_p64_x2(__p0) __extension__ ({ \
53657		poly64x2x2_t __ret; \
53658		__builtin_neon_vld1q_x2_v(&__ret, __p0, 38); \
53659		\
53660		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53661		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53662		__ret; \
53663		})
	55355	__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
	55356	int8x16_t __ret;
	55357	int8x16x3_t __rev0;
	55358	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55359	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55360	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55361	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55362	__ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 32);
	55363	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55364	return __ret;
	55365	}
53664	55366	#endif
53665	55367
53666	55368	#ifdef __LITTLE_ENDIAN__
53667		#define vld1q_f64_x2(__p0) __extension__ ({ \
53668		float64x2x2_t __ret; \
53669		__builtin_neon_vld1q_x2_v(&__ret, __p0, 42); \
53670		__ret; \
53671		})
	55369	__ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
	55370	uint8x8_t __ret;
	55371	__ret = (uint8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 16);
	55372	return __ret;
	55373	}
53672	55374	#else
53673		#define vld1q_f64_x2(__p0) __extension__ ({ \
53674		float64x2x2_t __ret; \
53675		__builtin_neon_vld1q_x2_v(&__ret, __p0, 42); \
53676		\
53677		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53678		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53679		__ret; \
53680		})
	55375	__ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
	55376	uint8x8_t __ret;
	55377	uint8x16x3_t __rev0;
	55378	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55379	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55380	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55381	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55382	__ret = (uint8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 16);
	55383	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55384	return __ret;
	55385	}
53681	55386	#endif
53682	55387
53683		#define vld1_f64_x2(__p0) __extension__ ({ \
53684		float64x1x2_t __ret; \
53685		__builtin_neon_vld1_x2_v(&__ret, __p0, 10); \
53686		__ret; \
53687		})
53688		#define vld1_p64_x3(__p0) __extension__ ({ \
53689		poly64x1x3_t __ret; \
53690		__builtin_neon_vld1_x3_v(&__ret, __p0, 6); \
53691		__ret; \
53692		})
53693	55388	#ifdef __LITTLE_ENDIAN__
53694		#define vld1q_p64_x3(__p0) __extension__ ({ \
53695		poly64x2x3_t __ret; \
53696		__builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
53697		__ret; \
53698		})
	55389	__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
	55390	int8x8_t __ret;
	55391	__ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 0);
	55392	return __ret;
	55393	}
53699	55394	#else
53700		#define vld1q_p64_x3(__p0) __extension__ ({ \
53701		poly64x2x3_t __ret; \
53702		__builtin_neon_vld1q_x3_v(&__ret, __p0, 38); \
53703		\
53704		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53705		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53706		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
53707		__ret; \
53708		})
	55395	__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
	55396	int8x8_t __ret;
	55397	int8x16x3_t __rev0;
	55398	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55399	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55400	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55401	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55402	__ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 0);
	55403	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55404	return __ret;
	55405	}
53709	55406	#endif
53710	55407
53711	55408	#ifdef __LITTLE_ENDIAN__
53712		#define vld1q_f64_x3(__p0) __extension__ ({ \
53713		float64x2x3_t __ret; \
53714		__builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
53715		__ret; \
53716		})
	55409	__ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
	55410	poly8x8_t __ret;
	55411	__ret = (poly8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 4);
	55412	return __ret;
	55413	}
53717	55414	#else
53718		#define vld1q_f64_x3(__p0) __extension__ ({ \
53719		float64x2x3_t __ret; \
53720		__builtin_neon_vld1q_x3_v(&__ret, __p0, 42); \
53721		\
53722		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53723		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53724		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
53725		__ret; \
53726		})
	55415	__ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
	55416	poly8x8_t __ret;
	55417	poly8x16x4_t __rev0;
	55418	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55419	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55420	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55421	__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55422	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55423	__ret = (poly8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 4);
	55424	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55425	return __ret;
	55426	}
53727	55427	#endif
53728	55428
53729		#define vld1_f64_x3(__p0) __extension__ ({ \
53730		float64x1x3_t __ret; \
53731		__builtin_neon_vld1_x3_v(&__ret, __p0, 10); \
53732		__ret; \
53733		})
53734		#define vld1_p64_x4(__p0) __extension__ ({ \
53735		poly64x1x4_t __ret; \
53736		__builtin_neon_vld1_x4_v(&__ret, __p0, 6); \
53737		__ret; \
53738		})
53739	55429	#ifdef __LITTLE_ENDIAN__
53740		#define vld1q_p64_x4(__p0) __extension__ ({ \
53741		poly64x2x4_t __ret; \
53742		__builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
53743		__ret; \
53744		})
	55430	__ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
	55431	poly8x16_t __ret;
	55432	__ret = (poly8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 36);
	55433	return __ret;
	55434	}
53745	55435	#else
53746		#define vld1q_p64_x4(__p0) __extension__ ({ \
53747		poly64x2x4_t __ret; \
53748		__builtin_neon_vld1q_x4_v(&__ret, __p0, 38); \
53749		\
53750		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53751		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53752		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
53753		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
53754		__ret; \
53755		})
	55436	__ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
	55437	poly8x16_t __ret;
	55438	poly8x16x4_t __rev0;
	55439	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55440	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55441	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55442	__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55443	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55444	__ret = (poly8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 36);
	55445	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55446	return __ret;
	55447	}
53756	55448	#endif
53757	55449
53758	55450	#ifdef __LITTLE_ENDIAN__
53759		#define vld1q_f64_x4(__p0) __extension__ ({ \
53760		float64x2x4_t __ret; \
53761		__builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
53762		__ret; \
53763		})
	55451	__ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
	55452	uint8x16_t __ret;
	55453	__ret = (uint8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 48);
	55454	return __ret;
	55455	}
53764	55456	#else
53765		#define vld1q_f64_x4(__p0) __extension__ ({ \
53766		float64x2x4_t __ret; \
53767		__builtin_neon_vld1q_x4_v(&__ret, __p0, 42); \
53768		\
53769		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53770		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53771		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
53772		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
53773		__ret; \
53774		})
	55457	__ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
	55458	uint8x16_t __ret;
	55459	uint8x16x4_t __rev0;
	55460	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55461	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55462	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55463	__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55464	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55465	__ret = (uint8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 48);
	55466	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55467	return __ret;
	55468	}
53775	55469	#endif
53776	55470
53777		#define vld1_f64_x4(__p0) __extension__ ({ \
53778		float64x1x4_t __ret; \
53779		__builtin_neon_vld1_x4_v(&__ret, __p0, 10); \
53780		__ret; \
53781		})
53782		#define vld2_p64(__p0) __extension__ ({ \
53783		poly64x1x2_t __ret; \
53784		__builtin_neon_vld2_v(&__ret, __p0, 6); \
53785		__ret; \
53786		})
53787	55471	#ifdef __LITTLE_ENDIAN__
53788		#define vld2q_p64(__p0) __extension__ ({ \
53789		poly64x2x2_t __ret; \
53790		__builtin_neon_vld2q_v(&__ret, __p0, 38); \
53791		__ret; \
53792		})
	55472	__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
	55473	int8x16_t __ret;
	55474	__ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 32);
	55475	return __ret;
	55476	}
53793	55477	#else
53794		#define vld2q_p64(__p0) __extension__ ({ \
53795		poly64x2x2_t __ret; \
53796		__builtin_neon_vld2q_v(&__ret, __p0, 38); \
53797		\
53798		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53799		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53800		__ret; \
53801		})
	55478	__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
	55479	int8x16_t __ret;
	55480	int8x16x4_t __rev0;
	55481	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55482	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55483	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55484	__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55485	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55486	__ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 32);
	55487	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55488	return __ret;
	55489	}
53802	55490	#endif
53803	55491
53804	55492	#ifdef __LITTLE_ENDIAN__
53805		#define vld2q_u64(__p0) __extension__ ({ \
53806		uint64x2x2_t __ret; \
53807		__builtin_neon_vld2q_v(&__ret, __p0, 51); \
53808		__ret; \
53809		})
	55493	__ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
	55494	uint8x8_t __ret;
	55495	__ret = (uint8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 16);
	55496	return __ret;
	55497	}
53810	55498	#else
53811		#define vld2q_u64(__p0) __extension__ ({ \
53812		uint64x2x2_t __ret; \
53813		__builtin_neon_vld2q_v(&__ret, __p0, 51); \
53814		\
53815		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53816		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53817		__ret; \
53818		})
	55499	__ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
	55500	uint8x8_t __ret;
	55501	uint8x16x4_t __rev0;
	55502	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55503	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55504	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55505	__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55506	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55507	__ret = (uint8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 16);
	55508	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55509	return __ret;
	55510	}
53819	55511	#endif
53820	55512
53821	55513	#ifdef __LITTLE_ENDIAN__
53822		#define vld2q_f64(__p0) __extension__ ({ \
53823		float64x2x2_t __ret; \
53824		__builtin_neon_vld2q_v(&__ret, __p0, 42); \
53825		__ret; \
53826		})
	55514	__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
	55515	int8x8_t __ret;
	55516	__ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 0);
	55517	return __ret;
	55518	}
53827	55519	#else
53828		#define vld2q_f64(__p0) __extension__ ({ \
53829		float64x2x2_t __ret; \
53830		__builtin_neon_vld2q_v(&__ret, __p0, 42); \
53831		\
53832		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53833		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53834		__ret; \
53835		})
	55520	__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
	55521	int8x8_t __ret;
	55522	int8x16x4_t __rev0;
	55523	__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55524	__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55525	__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55526	__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55527	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	55528	__ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 0);
	55529	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55530	return __ret;
	55531	}
53836	55532	#endif
53837	55533
53838	55534	#ifdef __LITTLE_ENDIAN__
53839		#define vld2q_s64(__p0) __extension__ ({ \
53840		int64x2x2_t __ret; \
53841		__builtin_neon_vld2q_v(&__ret, __p0, 35); \
53842		__ret; \
53843		})
	55535	__ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
	55536	poly8x8_t __ret;
	55537	__ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 4);
	55538	return __ret;
	55539	}
53844	55540	#else
53845		#define vld2q_s64(__p0) __extension__ ({ \
53846		int64x2x2_t __ret; \
53847		__builtin_neon_vld2q_v(&__ret, __p0, 35); \
53848		\
53849		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53850		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53851		__ret; \
53852		})
	55541	__ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
	55542	poly8x8_t __ret;
	55543	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55544	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55545	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55546	__ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 4);
	55547	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55548	return __ret;
	55549	}
53853	55550	#endif
53854	55551
53855		#define vld2_f64(__p0) __extension__ ({ \
53856		float64x1x2_t __ret; \
53857		__builtin_neon_vld2_v(&__ret, __p0, 10); \
53858		__ret; \
53859		})
53860		#define vld2_dup_p64(__p0) __extension__ ({ \
53861		poly64x1x2_t __ret; \
53862		__builtin_neon_vld2_dup_v(&__ret, __p0, 6); \
53863		__ret; \
53864		})
53865	55552	#ifdef __LITTLE_ENDIAN__
53866		#define vld2q_dup_p64(__p0) __extension__ ({ \
53867		poly64x2x2_t __ret; \
53868		__builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
53869		__ret; \
53870		})
	55553	__ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
	55554	poly8x16_t __ret;
	55555	__ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 36);
	55556	return __ret;
	55557	}
53871	55558	#else
53872		#define vld2q_dup_p64(__p0) __extension__ ({ \
53873		poly64x2x2_t __ret; \
53874		__builtin_neon_vld2q_dup_v(&__ret, __p0, 38); \
53875		\
53876		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53877		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53878		__ret; \
53879		})
	55559	__ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
	55560	poly8x16_t __ret;
	55561	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55562	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55563	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55564	__ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 36);
	55565	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55566	return __ret;
	55567	}
53880	55568	#endif
53881	55569
53882	55570	#ifdef __LITTLE_ENDIAN__
53883		#define vld2q_dup_f64(__p0) __extension__ ({ \
53884		float64x2x2_t __ret; \
53885		__builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
53886		__ret; \
53887		})
	55571	__ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	55572	uint8x16_t __ret;
	55573	__ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
	55574	return __ret;
	55575	}
53888	55576	#else
53889		#define vld2q_dup_f64(__p0) __extension__ ({ \
53890		float64x2x2_t __ret; \
53891		__builtin_neon_vld2q_dup_v(&__ret, __p0, 42); \
53892		\
53893		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53894		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53895		__ret; \
53896		})
	55577	__ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	55578	uint8x16_t __ret;
	55579	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55580	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55581	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55582	__ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
	55583	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55584	return __ret;
	55585	}
53897	55586	#endif
53898	55587
53899		#define vld2_dup_f64(__p0) __extension__ ({ \
53900		float64x1x2_t __ret; \
53901		__builtin_neon_vld2_dup_v(&__ret, __p0, 10); \
53902		__ret; \
53903		})
53904		#define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53905		poly64x1x2_t __ret; \
53906		poly64x1x2_t __s1 = __p1; \
53907		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
53908		__ret; \
53909		})
53910	55588	#ifdef __LITTLE_ENDIAN__
53911		#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
53912		poly8x16x2_t __ret; \
53913		poly8x16x2_t __s1 = __p1; \
53914		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 36); \
53915		__ret; \
53916		})
	55589	__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
	55590	int8x16_t __ret;
	55591	__ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
	55592	return __ret;
	55593	}
53917	55594	#else
53918		#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
53919		poly8x16x2_t __ret; \
53920		poly8x16x2_t __s1 = __p1; \
53921		poly8x16x2_t __rev1; \
53922		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53923		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53924		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 36); \
53925		\
53926		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53927		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53928		__ret; \
53929		})
	55595	__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
	55596	int8x16_t __ret;
	55597	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55598	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55599	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55600	__ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
	55601	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55602	return __ret;
	55603	}
53930	55604	#endif
53931	55605
53932	55606	#ifdef __LITTLE_ENDIAN__
53933		#define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53934		poly64x2x2_t __ret; \
53935		poly64x2x2_t __s1 = __p1; \
53936		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 38); \
53937		__ret; \
53938		})
	55607	__ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
	55608	uint8x8_t __ret;
	55609	__ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 16);
	55610	return __ret;
	55611	}
53939	55612	#else
53940		#define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53941		poly64x2x2_t __ret; \
53942		poly64x2x2_t __s1 = __p1; \
53943		poly64x2x2_t __rev1; \
53944		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
53945		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
53946		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 38); \
53947		\
53948		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53949		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53950		__ret; \
53951		})
	55613	__ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
	55614	uint8x8_t __ret;
	55615	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55616	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55617	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55618	__ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 16);
	55619	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55620	return __ret;
	55621	}
53952	55622	#endif
53953	55623
53954	55624	#ifdef __LITTLE_ENDIAN__
53955		#define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
53956		uint8x16x2_t __ret; \
53957		uint8x16x2_t __s1 = __p1; \
53958		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 48); \
53959		__ret; \
53960		})
	55625	__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
	55626	int8x8_t __ret;
	55627	__ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 0);
	55628	return __ret;
	55629	}
53961	55630	#else
53962		#define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
53963		uint8x16x2_t __ret; \
53964		uint8x16x2_t __s1 = __p1; \
53965		uint8x16x2_t __rev1; \
53966		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53967		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53968		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 48); \
53969		\
53970		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53971		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
53972		__ret; \
53973		})
	55631	__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
	55632	int8x8_t __ret;
	55633	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55634	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55635	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55636	__ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 0);
	55637	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55638	return __ret;
	55639	}
53974	55640	#endif
53975	55641
53976	55642	#ifdef __LITTLE_ENDIAN__
53977		#define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
53978		uint64x2x2_t __ret; \
53979		uint64x2x2_t __s1 = __p1; \
53980		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 51); \
53981		__ret; \
53982		})
	55643	__ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
	55644	poly8x8_t __ret;
	55645	__ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 4);
	55646	return __ret;
	55647	}
53983	55648	#else
53984		#define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
53985		uint64x2x2_t __ret; \
53986		uint64x2x2_t __s1 = __p1; \
53987		uint64x2x2_t __rev1; \
53988		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
53989		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
53990		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 51); \
53991		\
53992		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
53993		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
53994		__ret; \
53995		})
	55649	__ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
	55650	poly8x8_t __ret;
	55651	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55652	poly8x16x2_t __rev1;
	55653	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55654	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55655	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55656	__ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 4);
	55657	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55658	return __ret;
	55659	}
53996	55660	#endif
53997	55661
53998	55662	#ifdef __LITTLE_ENDIAN__
53999		#define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54000		int8x16x2_t __ret; \
54001		int8x16x2_t __s1 = __p1; \
54002		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 32); \
54003		__ret; \
54004		})
	55663	__ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
	55664	poly8x16_t __ret;
	55665	__ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 36);
	55666	return __ret;
	55667	}
54005	55668	#else
54006		#define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54007		int8x16x2_t __ret; \
54008		int8x16x2_t __s1 = __p1; \
54009		int8x16x2_t __rev1; \
54010		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54011		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54012		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 32); \
54013		\
54014		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54015		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54016		__ret; \
54017		})
	55669	__ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
	55670	poly8x16_t __ret;
	55671	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55672	poly8x16x2_t __rev1;
	55673	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55674	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55675	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55676	__ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 36);
	55677	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55678	return __ret;
	55679	}
54018	55680	#endif
54019	55681
54020	55682	#ifdef __LITTLE_ENDIAN__
54021		#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54022		float64x2x2_t __ret; \
54023		float64x2x2_t __s1 = __p1; \
54024		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
54025		__ret; \
54026		})
	55683	__ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
	55684	uint8x16_t __ret;
	55685	__ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 48);
	55686	return __ret;
	55687	}
54027	55688	#else
54028		#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54029		float64x2x2_t __ret; \
54030		float64x2x2_t __s1 = __p1; \
54031		float64x2x2_t __rev1; \
54032		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54033		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54034		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
54035		\
54036		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54037		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54038		__ret; \
54039		})
	55689	__ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
	55690	uint8x16_t __ret;
	55691	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55692	uint8x16x2_t __rev1;
	55693	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55694	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55695	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55696	__ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 48);
	55697	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55698	return __ret;
	55699	}
54040	55700	#endif
54041	55701
54042	55702	#ifdef __LITTLE_ENDIAN__
54043		#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54044		int64x2x2_t __ret; \
54045		int64x2x2_t __s1 = __p1; \
54046		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
54047		__ret; \
54048		})
	55703	__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
	55704	int8x16_t __ret;
	55705	__ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 32);
	55706	return __ret;
	55707	}
54049	55708	#else
54050		#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54051		int64x2x2_t __ret; \
54052		int64x2x2_t __s1 = __p1; \
54053		int64x2x2_t __rev1; \
54054		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54055		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54056		__builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
54057		\
54058		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54059		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54060		__ret; \
54061		})
	55709	__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
	55710	int8x16_t __ret;
	55711	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55712	int8x16x2_t __rev1;
	55713	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55714	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55715	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55716	__ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 32);
	55717	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55718	return __ret;
	55719	}
54062	55720	#endif
54063	55721
54064		#define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54065		uint64x1x2_t __ret; \
54066		uint64x1x2_t __s1 = __p1; \
54067		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
54068		__ret; \
54069		})
54070		#define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54071		float64x1x2_t __ret; \
54072		float64x1x2_t __s1 = __p1; \
54073		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
54074		__ret; \
54075		})
54076		#define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54077		int64x1x2_t __ret; \
54078		int64x1x2_t __s1 = __p1; \
54079		__builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
54080		__ret; \
54081		})
54082		#define vld3_p64(__p0) __extension__ ({ \
54083		poly64x1x3_t __ret; \
54084		__builtin_neon_vld3_v(&__ret, __p0, 6); \
54085		__ret; \
54086		})
54087	55722	#ifdef __LITTLE_ENDIAN__
54088		#define vld3q_p64(__p0) __extension__ ({ \
54089		poly64x2x3_t __ret; \
54090		__builtin_neon_vld3q_v(&__ret, __p0, 38); \
54091		__ret; \
54092		})
	55723	__ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
	55724	uint8x8_t __ret;
	55725	__ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 16);
	55726	return __ret;
	55727	}
54093	55728	#else
54094		#define vld3q_p64(__p0) __extension__ ({ \
54095		poly64x2x3_t __ret; \
54096		__builtin_neon_vld3q_v(&__ret, __p0, 38); \
54097		\
54098		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54099		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54100		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54101		__ret; \
54102		})
	55729	__ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
	55730	uint8x8_t __ret;
	55731	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55732	uint8x16x2_t __rev1;
	55733	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55734	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55735	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55736	__ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 16);
	55737	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55738	return __ret;
	55739	}
54103	55740	#endif
54104	55741
54105	55742	#ifdef __LITTLE_ENDIAN__
54106		#define vld3q_u64(__p0) __extension__ ({ \
54107		uint64x2x3_t __ret; \
54108		__builtin_neon_vld3q_v(&__ret, __p0, 51); \
54109		__ret; \
54110		})
	55743	__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
	55744	int8x8_t __ret;
	55745	__ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 0);
	55746	return __ret;
	55747	}
54111	55748	#else
54112		#define vld3q_u64(__p0) __extension__ ({ \
54113		uint64x2x3_t __ret; \
54114		__builtin_neon_vld3q_v(&__ret, __p0, 51); \
54115		\
54116		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54117		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54118		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54119		__ret; \
54120		})
	55749	__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
	55750	int8x8_t __ret;
	55751	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55752	int8x16x2_t __rev1;
	55753	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55754	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55755	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55756	__ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 0);
	55757	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55758	return __ret;
	55759	}
54121	55760	#endif
54122	55761
54123	55762	#ifdef __LITTLE_ENDIAN__
54124		#define vld3q_f64(__p0) __extension__ ({ \
54125		float64x2x3_t __ret; \
54126		__builtin_neon_vld3q_v(&__ret, __p0, 42); \
54127		__ret; \
54128		})
	55763	__ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
	55764	poly8x8_t __ret;
	55765	__ret = (poly8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 4);
	55766	return __ret;
	55767	}
54129	55768	#else
54130		#define vld3q_f64(__p0) __extension__ ({ \
54131		float64x2x3_t __ret; \
54132		__builtin_neon_vld3q_v(&__ret, __p0, 42); \
54133		\
54134		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54135		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54136		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54137		__ret; \
54138		})
	55769	__ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
	55770	poly8x8_t __ret;
	55771	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55772	poly8x16x3_t __rev1;
	55773	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55774	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55775	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55776	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55777	__ret = (poly8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 4);
	55778	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55779	return __ret;
	55780	}
54139	55781	#endif
54140	55782
54141	55783	#ifdef __LITTLE_ENDIAN__
54142		#define vld3q_s64(__p0) __extension__ ({ \
54143		int64x2x3_t __ret; \
54144		__builtin_neon_vld3q_v(&__ret, __p0, 35); \
54145		__ret; \
54146		})
	55784	__ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
	55785	poly8x16_t __ret;
	55786	__ret = (poly8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 36);
	55787	return __ret;
	55788	}
54147	55789	#else
54148		#define vld3q_s64(__p0) __extension__ ({ \
54149		int64x2x3_t __ret; \
54150		__builtin_neon_vld3q_v(&__ret, __p0, 35); \
54151		\
54152		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54153		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54154		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54155		__ret; \
54156		})
	55790	__ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
	55791	poly8x16_t __ret;
	55792	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55793	poly8x16x3_t __rev1;
	55794	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55795	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55796	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55797	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55798	__ret = (poly8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 36);
	55799	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55800	return __ret;
	55801	}
	55802	#endif
	55803
	55804	#ifdef __LITTLE_ENDIAN__
	55805	__ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
	55806	uint8x16_t __ret;
	55807	__ret = (uint8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 48);
	55808	return __ret;
	55809	}
	55810	#else
	55811	__ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
	55812	uint8x16_t __ret;
	55813	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55814	uint8x16x3_t __rev1;
	55815	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55816	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55817	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55818	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55819	__ret = (uint8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 48);
	55820	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55821	return __ret;
	55822	}
54157	55823	#endif
54158	55824
54159		#define vld3_f64(__p0) __extension__ ({ \
54160		float64x1x3_t __ret; \
54161		__builtin_neon_vld3_v(&__ret, __p0, 10); \
54162		__ret; \
54163		})
54164		#define vld3_dup_p64(__p0) __extension__ ({ \
54165		poly64x1x3_t __ret; \
54166		__builtin_neon_vld3_dup_v(&__ret, __p0, 6); \
54167		__ret; \
54168		})
54169	55825	#ifdef __LITTLE_ENDIAN__
54170		#define vld3q_dup_p64(__p0) __extension__ ({ \
54171		poly64x2x3_t __ret; \
54172		__builtin_neon_vld3q_dup_v(&__ret, __p0, 38); \
54173		__ret; \
54174		})
	55826	__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
	55827	int8x16_t __ret;
	55828	__ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 32);
	55829	return __ret;
	55830	}
54175	55831	#else
54176		#define vld3q_dup_p64(__p0) __extension__ ({ \
54177		poly64x2x3_t __ret; \
54178		__builtin_neon_vld3q_dup_v(&__ret, __p0, 38); \
54179		\
54180		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54181		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54182		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54183		__ret; \
54184		})
	55832	__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
	55833	int8x16_t __ret;
	55834	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55835	int8x16x3_t __rev1;
	55836	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55837	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55838	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55839	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55840	__ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 32);
	55841	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55842	return __ret;
	55843	}
54185	55844	#endif
54186	55845
54187	55846	#ifdef __LITTLE_ENDIAN__
54188		#define vld3q_dup_f64(__p0) __extension__ ({ \
54189		float64x2x3_t __ret; \
54190		__builtin_neon_vld3q_dup_v(&__ret, __p0, 42); \
54191		__ret; \
54192		})
	55847	__ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
	55848	uint8x8_t __ret;
	55849	__ret = (uint8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 16);
	55850	return __ret;
	55851	}
54193	55852	#else
54194		#define vld3q_dup_f64(__p0) __extension__ ({ \
54195		float64x2x3_t __ret; \
54196		__builtin_neon_vld3q_dup_v(&__ret, __p0, 42); \
54197		\
54198		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54199		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54200		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54201		__ret; \
54202		})
	55853	__ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
	55854	uint8x8_t __ret;
	55855	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55856	uint8x16x3_t __rev1;
	55857	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55858	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55859	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55860	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55861	__ret = (uint8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 16);
	55862	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55863	return __ret;
	55864	}
54203	55865	#endif
54204	55866
54205		#define vld3_dup_f64(__p0) __extension__ ({ \
54206		float64x1x3_t __ret; \
54207		__builtin_neon_vld3_dup_v(&__ret, __p0, 10); \
54208		__ret; \
54209		})
54210		#define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54211		poly64x1x3_t __ret; \
54212		poly64x1x3_t __s1 = __p1; \
54213		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
54214		__ret; \
54215		})
54216	55867	#ifdef __LITTLE_ENDIAN__
54217		#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54218		poly8x16x3_t __ret; \
54219		poly8x16x3_t __s1 = __p1; \
54220		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 36); \
54221		__ret; \
54222		})
	55868	__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
	55869	int8x8_t __ret;
	55870	__ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 0);
	55871	return __ret;
	55872	}
54223	55873	#else
54224		#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54225		poly8x16x3_t __ret; \
54226		poly8x16x3_t __s1 = __p1; \
54227		poly8x16x3_t __rev1; \
54228		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54229		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54230		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54231		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 36); \
54232		\
54233		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54234		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54235		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54236		__ret; \
54237		})
	55874	__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
	55875	int8x8_t __ret;
	55876	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55877	int8x16x3_t __rev1;
	55878	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55879	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55880	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55881	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55882	__ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 0);
	55883	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55884	return __ret;
	55885	}
54238	55886	#endif
54239	55887
54240	55888	#ifdef __LITTLE_ENDIAN__
54241		#define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54242		poly64x2x3_t __ret; \
54243		poly64x2x3_t __s1 = __p1; \
54244		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 38); \
54245		__ret; \
54246		})
	55889	__ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
	55890	poly8x8_t __ret;
	55891	__ret = (poly8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 4);
	55892	return __ret;
	55893	}
54247	55894	#else
54248		#define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54249		poly64x2x3_t __ret; \
54250		poly64x2x3_t __s1 = __p1; \
54251		poly64x2x3_t __rev1; \
54252		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54253		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54254		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54255		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 38); \
54256		\
54257		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54258		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54259		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54260		__ret; \
54261		})
	55895	__ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
	55896	poly8x8_t __ret;
	55897	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55898	poly8x16x4_t __rev1;
	55899	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55900	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55901	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55902	__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55903	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55904	__ret = (poly8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 4);
	55905	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55906	return __ret;
	55907	}
54262	55908	#endif
54263	55909
54264	55910	#ifdef __LITTLE_ENDIAN__
54265		#define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54266		uint8x16x3_t __ret; \
54267		uint8x16x3_t __s1 = __p1; \
54268		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 48); \
54269		__ret; \
54270		})
	55911	__ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
	55912	poly8x16_t __ret;
	55913	__ret = (poly8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 36);
	55914	return __ret;
	55915	}
54271	55916	#else
54272		#define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54273		uint8x16x3_t __ret; \
54274		uint8x16x3_t __s1 = __p1; \
54275		uint8x16x3_t __rev1; \
54276		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54277		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54278		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54279		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 48); \
54280		\
54281		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54282		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54283		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54284		__ret; \
54285		})
	55917	__ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
	55918	poly8x16_t __ret;
	55919	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55920	poly8x16x4_t __rev1;
	55921	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55922	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55923	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55924	__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55925	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55926	__ret = (poly8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 36);
	55927	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55928	return __ret;
	55929	}
54286	55930	#endif
54287	55931
54288	55932	#ifdef __LITTLE_ENDIAN__
54289		#define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54290		uint64x2x3_t __ret; \
54291		uint64x2x3_t __s1 = __p1; \
54292		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 51); \
54293		__ret; \
54294		})
	55933	__ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
	55934	uint8x16_t __ret;
	55935	__ret = (uint8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 48);
	55936	return __ret;
	55937	}
54295	55938	#else
54296		#define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54297		uint64x2x3_t __ret; \
54298		uint64x2x3_t __s1 = __p1; \
54299		uint64x2x3_t __rev1; \
54300		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54301		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54302		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54303		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 51); \
54304		\
54305		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54306		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54307		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54308		__ret; \
54309		})
	55939	__ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
	55940	uint8x16_t __ret;
	55941	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55942	uint8x16x4_t __rev1;
	55943	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55944	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55945	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55946	__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55947	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55948	__ret = (uint8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 48);
	55949	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55950	return __ret;
	55951	}
54310	55952	#endif
54311	55953
54312	55954	#ifdef __LITTLE_ENDIAN__
54313		#define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54314		int8x16x3_t __ret; \
54315		int8x16x3_t __s1 = __p1; \
54316		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 32); \
54317		__ret; \
54318		})
	55955	__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
	55956	int8x16_t __ret;
	55957	__ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 32);
	55958	return __ret;
	55959	}
54319	55960	#else
54320		#define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54321		int8x16x3_t __ret; \
54322		int8x16x3_t __s1 = __p1; \
54323		int8x16x3_t __rev1; \
54324		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54325		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54326		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54327		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 32); \
54328		\
54329		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54330		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54331		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54332		__ret; \
54333		})
	55961	__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
	55962	int8x16_t __ret;
	55963	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55964	int8x16x4_t __rev1;
	55965	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55966	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55967	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55968	__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55969	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55970	__ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 32);
	55971	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55972	return __ret;
	55973	}
54334	55974	#endif
54335	55975
54336	55976	#ifdef __LITTLE_ENDIAN__
54337		#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54338		float64x2x3_t __ret; \
54339		float64x2x3_t __s1 = __p1; \
54340		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
54341		__ret; \
54342		})
	55977	__ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
	55978	uint8x8_t __ret;
	55979	__ret = (uint8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 16);
	55980	return __ret;
	55981	}
54343	55982	#else
54344		#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54345		float64x2x3_t __ret; \
54346		float64x2x3_t __s1 = __p1; \
54347		float64x2x3_t __rev1; \
54348		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54349		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54350		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54351		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
54352		\
54353		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54354		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54355		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54356		__ret; \
54357		})
	55983	__ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
	55984	uint8x8_t __ret;
	55985	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	55986	uint8x16x4_t __rev1;
	55987	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55988	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55989	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55990	__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	55991	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	55992	__ret = (uint8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 16);
	55993	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	55994	return __ret;
	55995	}
54358	55996	#endif
54359	55997
54360	55998	#ifdef __LITTLE_ENDIAN__
54361		#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54362		int64x2x3_t __ret; \
54363		int64x2x3_t __s1 = __p1; \
54364		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
54365		__ret; \
54366		})
	55999	__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
	56000	int8x8_t __ret;
	56001	__ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 0);
	56002	return __ret;
	56003	}
54367	56004	#else
54368		#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54369		int64x2x3_t __ret; \
54370		int64x2x3_t __s1 = __p1; \
54371		int64x2x3_t __rev1; \
54372		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54373		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54374		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54375		__builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
54376		\
54377		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54378		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54379		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54380		__ret; \
54381		})
	56005	__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
	56006	int8x8_t __ret;
	56007	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	56008	int8x16x4_t __rev1;
	56009	__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56010	__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56011	__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56012	__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56013	uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	56014	__ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 0);
	56015	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	56016	return __ret;
	56017	}
54382	56018	#endif
54383	56019
54384		#define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54385		uint64x1x3_t __ret; \
54386		uint64x1x3_t __s1 = __p1; \
54387		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
54388		__ret; \
54389		})
54390		#define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54391		float64x1x3_t __ret; \
54392		float64x1x3_t __s1 = __p1; \
54393		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
54394		__ret; \
54395		})
54396		#define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54397		int64x1x3_t __ret; \
54398		int64x1x3_t __s1 = __p1; \
54399		__builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
54400		__ret; \
54401		})
54402		#define vld4_p64(__p0) __extension__ ({ \
54403		poly64x1x4_t __ret; \
54404		__builtin_neon_vld4_v(&__ret, __p0, 6); \
54405		__ret; \
54406		})
54407	56020	#ifdef __LITTLE_ENDIAN__
54408		#define vld4q_p64(__p0) __extension__ ({ \
54409		poly64x2x4_t __ret; \
54410		__builtin_neon_vld4q_v(&__ret, __p0, 38); \
54411		__ret; \
54412		})
	56021	__ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	56022	uint16x8_t __ret;
	56023	__ret = vcombine_u16(__p0, vraddhn_u32(__p1, __p2));
	56024	return __ret;
	56025	}
54413	56026	#else
54414		#define vld4q_p64(__p0) __extension__ ({ \
54415		poly64x2x4_t __ret; \
54416		__builtin_neon_vld4q_v(&__ret, __p0, 38); \
54417		\
54418		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54419		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54420		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54421		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54422		__ret; \
54423		})
	56027	__ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	56028	uint16x8_t __ret;
	56029	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	56030	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	56031	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	56032	__ret = __noswap_vcombine_u16(__rev0, __noswap_vraddhn_u32(__rev1, __rev2));
	56033	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	56034	return __ret;
	56035	}
54424	56036	#endif
54425	56037
54426	56038	#ifdef __LITTLE_ENDIAN__
54427		#define vld4q_u64(__p0) __extension__ ({ \
54428		uint64x2x4_t __ret; \
54429		__builtin_neon_vld4q_v(&__ret, __p0, 51); \
54430		__ret; \
54431		})
	56039	__ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	56040	uint32x4_t __ret;
	56041	__ret = vcombine_u32(__p0, vraddhn_u64(__p1, __p2));
	56042	return __ret;
	56043	}
54432	56044	#else
54433		#define vld4q_u64(__p0) __extension__ ({ \
54434		uint64x2x4_t __ret; \
54435		__builtin_neon_vld4q_v(&__ret, __p0, 51); \
54436		\
54437		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54438		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54439		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54440		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54441		__ret; \
54442		})
	56045	__ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	56046	uint32x4_t __ret;
	56047	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	56048	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	56049	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	56050	__ret = __noswap_vcombine_u32(__rev0, __noswap_vraddhn_u64(__rev1, __rev2));
	56051	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	56052	return __ret;
	56053	}
54443	56054	#endif
54444	56055
54445	56056	#ifdef __LITTLE_ENDIAN__
54446		#define vld4q_f64(__p0) __extension__ ({ \
54447		float64x2x4_t __ret; \
54448		__builtin_neon_vld4q_v(&__ret, __p0, 42); \
54449		__ret; \
54450		})
	56057	__ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	56058	uint8x16_t __ret;
	56059	__ret = vcombine_u8(__p0, vraddhn_u16(__p1, __p2));
	56060	return __ret;
	56061	}
54451	56062	#else
54452		#define vld4q_f64(__p0) __extension__ ({ \
54453		float64x2x4_t __ret; \
54454		__builtin_neon_vld4q_v(&__ret, __p0, 42); \
54455		\
54456		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54457		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54458		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54459		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54460		__ret; \
54461		})
	56063	__ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	56064	uint8x16_t __ret;
	56065	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	56066	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	56067	uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	56068	__ret = __noswap_vcombine_u8(__rev0, __noswap_vraddhn_u16(__rev1, __rev2));
	56069	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56070	return __ret;
	56071	}
54462	56072	#endif
54463	56073
54464	56074	#ifdef __LITTLE_ENDIAN__
54465		#define vld4q_s64(__p0) __extension__ ({ \
54466		int64x2x4_t __ret; \
54467		__builtin_neon_vld4q_v(&__ret, __p0, 35); \
54468		__ret; \
54469		})
	56075	__ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	56076	int16x8_t __ret;
	56077	__ret = vcombine_s16(__p0, vraddhn_s32(__p1, __p2));
	56078	return __ret;
	56079	}
54470	56080	#else
54471		#define vld4q_s64(__p0) __extension__ ({ \
54472		int64x2x4_t __ret; \
54473		__builtin_neon_vld4q_v(&__ret, __p0, 35); \
54474		\
54475		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54476		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54477		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54478		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54479		__ret; \
54480		})
	56081	__ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	56082	int16x8_t __ret;
	56083	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	56084	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	56085	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	56086	__ret = __noswap_vcombine_s16(__rev0, __noswap_vraddhn_s32(__rev1, __rev2));
	56087	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	56088	return __ret;
	56089	}
54481	56090	#endif
54482	56091
54483		#define vld4_f64(__p0) __extension__ ({ \
54484		float64x1x4_t __ret; \
54485		__builtin_neon_vld4_v(&__ret, __p0, 10); \
54486		__ret; \
54487		})
54488		#define vld4_dup_p64(__p0) __extension__ ({ \
54489		poly64x1x4_t __ret; \
54490		__builtin_neon_vld4_dup_v(&__ret, __p0, 6); \
54491		__ret; \
54492		})
54493	56092	#ifdef __LITTLE_ENDIAN__
54494		#define vld4q_dup_p64(__p0) __extension__ ({ \
54495		poly64x2x4_t __ret; \
54496		__builtin_neon_vld4q_dup_v(&__ret, __p0, 38); \
54497		__ret; \
54498		})
54499		#else
54500		#define vld4q_dup_p64(__p0) __extension__ ({ \
54501		poly64x2x4_t __ret; \
54502		__builtin_neon_vld4q_dup_v(&__ret, __p0, 38); \
54503		\
54504		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54505		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54506		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54507		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54508		__ret; \
54509		})
	56093	__ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	56094	int32x4_t __ret;
	56095	__ret = vcombine_s32(__p0, vraddhn_s64(__p1, __p2));
	56096	return __ret;
	56097	}
	56098	#else
	56099	__ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	56100	int32x4_t __ret;
	56101	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	56102	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	56103	int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	56104	__ret = __noswap_vcombine_s32(__rev0, __noswap_vraddhn_s64(__rev1, __rev2));
	56105	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	56106	return __ret;
	56107	}
54510	56108	#endif
54511	56109
54512	56110	#ifdef __LITTLE_ENDIAN__
54513		#define vld4q_dup_f64(__p0) __extension__ ({ \
54514		float64x2x4_t __ret; \
54515		__builtin_neon_vld4q_dup_v(&__ret, __p0, 42); \
54516		__ret; \
54517		})
	56111	__ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	56112	int8x16_t __ret;
	56113	__ret = vcombine_s8(__p0, vraddhn_s16(__p1, __p2));
	56114	return __ret;
	56115	}
54518	56116	#else
54519		#define vld4q_dup_f64(__p0) __extension__ ({ \
54520		float64x2x4_t __ret; \
54521		__builtin_neon_vld4q_dup_v(&__ret, __p0, 42); \
54522		\
54523		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54524		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54525		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54526		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54527		__ret; \
54528		})
	56117	__ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	56118	int8x16_t __ret;
	56119	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	56120	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	56121	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	56122	__ret = __noswap_vcombine_s8(__rev0, __noswap_vraddhn_s16(__rev1, __rev2));
	56123	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56124	return __ret;
	56125	}
54529	56126	#endif
54530	56127
54531		#define vld4_dup_f64(__p0) __extension__ ({ \
54532		float64x1x4_t __ret; \
54533		__builtin_neon_vld4_dup_v(&__ret, __p0, 10); \
54534		__ret; \
54535		})
54536		#define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54537		poly64x1x4_t __ret; \
54538		poly64x1x4_t __s1 = __p1; \
54539		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
54540		__ret; \
54541		})
54542	56128	#ifdef __LITTLE_ENDIAN__
54543		#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54544		poly8x16x4_t __ret; \
54545		poly8x16x4_t __s1 = __p1; \
54546		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 36); \
54547		__ret; \
54548		})
	56129	__ai poly8x8_t vrbit_p8(poly8x8_t __p0) {
	56130	poly8x8_t __ret;
	56131	__ret = (poly8x8_t) __builtin_neon_vrbit_v((int8x8_t)__p0, 4);
	56132	return __ret;
	56133	}
54549	56134	#else
54550		#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54551		poly8x16x4_t __ret; \
54552		poly8x16x4_t __s1 = __p1; \
54553		poly8x16x4_t __rev1; \
54554		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54555		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54556		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54557		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54558		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 36); \
54559		\
54560		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54561		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54562		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54563		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54564		__ret; \
54565		})
	56135	__ai poly8x8_t vrbit_p8(poly8x8_t __p0) {
	56136	poly8x8_t __ret;
	56137	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	56138	__ret = (poly8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 4);
	56139	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	56140	return __ret;
	56141	}
54566	56142	#endif
54567	56143
54568	56144	#ifdef __LITTLE_ENDIAN__
54569		#define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54570		poly64x2x4_t __ret; \
54571		poly64x2x4_t __s1 = __p1; \
54572		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 38); \
54573		__ret; \
54574		})
	56145	__ai poly8x16_t vrbitq_p8(poly8x16_t __p0) {
	56146	poly8x16_t __ret;
	56147	__ret = (poly8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__p0, 36);
	56148	return __ret;
	56149	}
54575	56150	#else
54576		#define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54577		poly64x2x4_t __ret; \
54578		poly64x2x4_t __s1 = __p1; \
54579		poly64x2x4_t __rev1; \
54580		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54581		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54582		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54583		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54584		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 38); \
54585		\
54586		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54587		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54588		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54589		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54590		__ret; \
54591		})
	56151	__ai poly8x16_t vrbitq_p8(poly8x16_t __p0) {
	56152	poly8x16_t __ret;
	56153	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56154	__ret = (poly8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 36);
	56155	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56156	return __ret;
	56157	}
54592	56158	#endif
54593	56159
54594	56160	#ifdef __LITTLE_ENDIAN__
54595		#define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54596		uint8x16x4_t __ret; \
54597		uint8x16x4_t __s1 = __p1; \
54598		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 48); \
54599		__ret; \
54600		})
	56161	__ai uint8x16_t vrbitq_u8(uint8x16_t __p0) {
	56162	uint8x16_t __ret;
	56163	__ret = (uint8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__p0, 48);
	56164	return __ret;
	56165	}
54601	56166	#else
54602		#define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54603		uint8x16x4_t __ret; \
54604		uint8x16x4_t __s1 = __p1; \
54605		uint8x16x4_t __rev1; \
54606		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54607		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54608		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54609		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54610		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 48); \
54611		\
54612		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54613		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54614		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54615		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54616		__ret; \
54617		})
	56167	__ai uint8x16_t vrbitq_u8(uint8x16_t __p0) {
	56168	uint8x16_t __ret;
	56169	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56170	__ret = (uint8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 48);
	56171	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56172	return __ret;
	56173	}
54618	56174	#endif
54619	56175
54620	56176	#ifdef __LITTLE_ENDIAN__
54621		#define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54622		uint64x2x4_t __ret; \
54623		uint64x2x4_t __s1 = __p1; \
54624		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 51); \
54625		__ret; \
54626		})
	56177	__ai int8x16_t vrbitq_s8(int8x16_t __p0) {
	56178	int8x16_t __ret;
	56179	__ret = (int8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__p0, 32);
	56180	return __ret;
	56181	}
54627	56182	#else
54628		#define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54629		uint64x2x4_t __ret; \
54630		uint64x2x4_t __s1 = __p1; \
54631		uint64x2x4_t __rev1; \
54632		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54633		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54634		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54635		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54636		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 51); \
54637		\
54638		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54639		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54640		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54641		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54642		__ret; \
54643		})
	56183	__ai int8x16_t vrbitq_s8(int8x16_t __p0) {
	56184	int8x16_t __ret;
	56185	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56186	__ret = (int8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 32);
	56187	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	56188	return __ret;
	56189	}
54644	56190	#endif
54645	56191
54646	56192	#ifdef __LITTLE_ENDIAN__
54647		#define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54648		int8x16x4_t __ret; \
54649		int8x16x4_t __s1 = __p1; \
54650		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 32); \
54651		__ret; \
54652		})
	56193	__ai uint8x8_t vrbit_u8(uint8x8_t __p0) {
	56194	uint8x8_t __ret;
	56195	__ret = (uint8x8_t) __builtin_neon_vrbit_v((int8x8_t)__p0, 16);
	56196	return __ret;
	56197	}
54653	56198	#else
54654		#define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54655		int8x16x4_t __ret; \
54656		int8x16x4_t __s1 = __p1; \
54657		int8x16x4_t __rev1; \
54658		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54659		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54660		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54661		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54662		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 32); \
54663		\
54664		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54665		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54666		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54667		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
54668		__ret; \
54669		})
	56199	__ai uint8x8_t vrbit_u8(uint8x8_t __p0) {
	56200	uint8x8_t __ret;
	56201	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	56202	__ret = (uint8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 16);
	56203	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	56204	return __ret;
	56205	}
54670	56206	#endif
54671	56207
54672	56208	#ifdef __LITTLE_ENDIAN__
54673		#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54674		float64x2x4_t __ret; \
54675		float64x2x4_t __s1 = __p1; \
54676		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 42); \
54677		__ret; \
54678		})
	56209	__ai int8x8_t vrbit_s8(int8x8_t __p0) {
	56210	int8x8_t __ret;
	56211	__ret = (int8x8_t) __builtin_neon_vrbit_v((int8x8_t)__p0, 0);
	56212	return __ret;
	56213	}
54679	56214	#else
54680		#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54681		float64x2x4_t __ret; \
54682		float64x2x4_t __s1 = __p1; \
54683		float64x2x4_t __rev1; \
54684		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54685		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54686		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54687		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54688		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 42); \
54689		\
54690		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54691		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54692		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54693		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54694		__ret; \
54695		})
	56215	__ai int8x8_t vrbit_s8(int8x8_t __p0) {
	56216	int8x8_t __ret;
	56217	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	56218	__ret = (int8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 0);
	56219	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	56220	return __ret;
	56221	}
54696	56222	#endif
54697	56223
54698	56224	#ifdef __LITTLE_ENDIAN__
54699		#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54700		int64x2x4_t __ret; \
54701		int64x2x4_t __s1 = __p1; \
54702		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 35); \
54703		__ret; \
54704		})
	56225	__ai float64x2_t vrecpeq_f64(float64x2_t __p0) {
	56226	float64x2_t __ret;
	56227	__ret = (float64x2_t) __builtin_neon_vrecpeq_v((int8x16_t)__p0, 42);
	56228	return __ret;
	56229	}
54705	56230	#else
54706		#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54707		int64x2x4_t __ret; \
54708		int64x2x4_t __s1 = __p1; \
54709		int64x2x4_t __rev1; \
54710		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
54711		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54712		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54713		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54714		__builtin_neon_vld4q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 35); \
54715		\
54716		__ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
54717		__ret.val[1] = __builtin_shufflevector(__ret.val[1], __ret.val[1], 1, 0); \
54718		__ret.val[2] = __builtin_shufflevector(__ret.val[2], __ret.val[2], 1, 0); \
54719		__ret.val[3] = __builtin_shufflevector(__ret.val[3], __ret.val[3], 1, 0); \
54720		__ret; \
54721		})
	56231	__ai float64x2_t vrecpeq_f64(float64x2_t __p0) {
	56232	float64x2_t __ret;
	56233	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	56234	__ret = (float64x2_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 42);
	56235	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	56236	return __ret;
	56237	}
54722	56238	#endif
54723	56239
54724		#define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54725		uint64x1x4_t __ret; \
54726		uint64x1x4_t __s1 = __p1; \
54727		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
54728		__ret; \
54729		})
54730		#define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54731		float64x1x4_t __ret; \
54732		float64x1x4_t __s1 = __p1; \
54733		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 10); \
54734		__ret; \
54735		})
54736		#define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54737		int64x1x4_t __ret; \
54738		int64x1x4_t __s1 = __p1; \
54739		__builtin_neon_vld4_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 3); \
54740		__ret; \
54741		})
54742		#define vldrq_p128(__p0) __extension__ ({ \
54743		poly128_t __ret; \
54744		__ret = (poly128_t) __builtin_neon_vldrq_p128(__p0); \
54745		__ret; \
54746		})
	56240	__ai float64x1_t vrecpe_f64(float64x1_t __p0) {
	56241	float64x1_t __ret;
	56242	__ret = (float64x1_t) __builtin_neon_vrecpe_v((int8x8_t)__p0, 10);
	56243	return __ret;
	56244	}
	56245	__ai float64_t vrecped_f64(float64_t __p0) {
	56246	float64_t __ret;
	56247	__ret = (float64_t) __builtin_neon_vrecped_f64(__p0);
	56248	return __ret;
	56249	}
	56250	__ai float32_t vrecpes_f32(float32_t __p0) {
	56251	float32_t __ret;
	56252	__ret = (float32_t) __builtin_neon_vrecpes_f32(__p0);
	56253	return __ret;
	56254	}
54747	56255	#ifdef __LITTLE_ENDIAN__
54748		__ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
	56256	__ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
54749	56257	float64x2_t __ret;
54750		__ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	56258	__ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	56259	return __ret;
	56260	}
	56261	#else
	56262	__ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
	56263	float64x2_t __ret;
	56264	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	56265	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	56266	__ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	56267	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	56268	return __ret;
	56269	}
	56270	#endif
	56271
	56272	__ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
	56273	float64x1_t __ret;
	56274	__ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	56275	return __ret;
	56276	}
	56277	__ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
	56278	float64_t __ret;
	56279	__ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
	56280	return __ret;
	56281	}
	56282	__ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
	56283	float32_t __ret;
	56284	__ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
	56285	return __ret;
	56286	}
	56287	__ai float64_t vrecpxd_f64(float64_t __p0) {
	56288	float64_t __ret;
	56289	__ret = (float64_t) __builtin_neon_vrecpxd_f64(__p0);
	56290	return __ret;
	56291	}
	56292	__ai float32_t vrecpxs_f32(float32_t __p0) {
	56293	float32_t __ret;
	56294	__ret = (float32_t) __builtin_neon_vrecpxs_f32(__p0);
	56295	return __ret;
	56296	}
	56297	__ai poly8x8_t vreinterpret_p8_p64(poly64x1_t __p0) {
	56298	poly8x8_t __ret;
	56299	__ret = (poly8x8_t)(__p0);
	56300	return __ret;
	56301	}
	56302	__ai poly8x8_t vreinterpret_p8_p16(poly16x4_t __p0) {
	56303	poly8x8_t __ret;
	56304	__ret = (poly8x8_t)(__p0);
	56305	return __ret;
	56306	}
	56307	__ai poly8x8_t vreinterpret_p8_u8(uint8x8_t __p0) {
	56308	poly8x8_t __ret;
	56309	__ret = (poly8x8_t)(__p0);
	56310	return __ret;
	56311	}
	56312	__ai poly8x8_t vreinterpret_p8_u32(uint32x2_t __p0) {
	56313	poly8x8_t __ret;
	56314	__ret = (poly8x8_t)(__p0);
	56315	return __ret;
	56316	}
	56317	__ai poly8x8_t vreinterpret_p8_u64(uint64x1_t __p0) {
	56318	poly8x8_t __ret;
	56319	__ret = (poly8x8_t)(__p0);
	56320	return __ret;
	56321	}
	56322	__ai poly8x8_t vreinterpret_p8_u16(uint16x4_t __p0) {
	56323	poly8x8_t __ret;
	56324	__ret = (poly8x8_t)(__p0);
	56325	return __ret;
	56326	}
	56327	__ai poly8x8_t vreinterpret_p8_s8(int8x8_t __p0) {
	56328	poly8x8_t __ret;
	56329	__ret = (poly8x8_t)(__p0);
	56330	return __ret;
	56331	}
	56332	__ai poly8x8_t vreinterpret_p8_f64(float64x1_t __p0) {
	56333	poly8x8_t __ret;
	56334	__ret = (poly8x8_t)(__p0);
	56335	return __ret;
	56336	}
	56337	__ai poly8x8_t vreinterpret_p8_f32(float32x2_t __p0) {
	56338	poly8x8_t __ret;
	56339	__ret = (poly8x8_t)(__p0);
	56340	return __ret;
	56341	}
	56342	__ai poly8x8_t vreinterpret_p8_f16(float16x4_t __p0) {
	56343	poly8x8_t __ret;
	56344	__ret = (poly8x8_t)(__p0);
	56345	return __ret;
	56346	}
	56347	__ai poly8x8_t vreinterpret_p8_s32(int32x2_t __p0) {
	56348	poly8x8_t __ret;
	56349	__ret = (poly8x8_t)(__p0);
	56350	return __ret;
	56351	}
	56352	__ai poly8x8_t vreinterpret_p8_s64(int64x1_t __p0) {
	56353	poly8x8_t __ret;
	56354	__ret = (poly8x8_t)(__p0);
	56355	return __ret;
	56356	}
	56357	__ai poly8x8_t vreinterpret_p8_s16(int16x4_t __p0) {
	56358	poly8x8_t __ret;
	56359	__ret = (poly8x8_t)(__p0);
	56360	return __ret;
	56361	}
	56362	__ai poly64x1_t vreinterpret_p64_p8(poly8x8_t __p0) {
	56363	poly64x1_t __ret;
	56364	__ret = (poly64x1_t)(__p0);
	56365	return __ret;
	56366	}
	56367	__ai poly64x1_t vreinterpret_p64_p16(poly16x4_t __p0) {
	56368	poly64x1_t __ret;
	56369	__ret = (poly64x1_t)(__p0);
	56370	return __ret;
	56371	}
	56372	__ai poly64x1_t vreinterpret_p64_u8(uint8x8_t __p0) {
	56373	poly64x1_t __ret;
	56374	__ret = (poly64x1_t)(__p0);
	56375	return __ret;
	56376	}
	56377	__ai poly64x1_t vreinterpret_p64_u32(uint32x2_t __p0) {
	56378	poly64x1_t __ret;
	56379	__ret = (poly64x1_t)(__p0);
	56380	return __ret;
	56381	}
	56382	__ai poly64x1_t vreinterpret_p64_u64(uint64x1_t __p0) {
	56383	poly64x1_t __ret;
	56384	__ret = (poly64x1_t)(__p0);
	56385	return __ret;
	56386	}
	56387	__ai poly64x1_t vreinterpret_p64_u16(uint16x4_t __p0) {
	56388	poly64x1_t __ret;
	56389	__ret = (poly64x1_t)(__p0);
	56390	return __ret;
	56391	}
	56392	__ai poly64x1_t vreinterpret_p64_s8(int8x8_t __p0) {
	56393	poly64x1_t __ret;
	56394	__ret = (poly64x1_t)(__p0);
	56395	return __ret;
	56396	}
	56397	__ai poly64x1_t vreinterpret_p64_f64(float64x1_t __p0) {
	56398	poly64x1_t __ret;
	56399	__ret = (poly64x1_t)(__p0);
	56400	return __ret;
	56401	}
	56402	__ai poly64x1_t vreinterpret_p64_f32(float32x2_t __p0) {
	56403	poly64x1_t __ret;
	56404	__ret = (poly64x1_t)(__p0);
	56405	return __ret;
	56406	}
	56407	__ai poly64x1_t vreinterpret_p64_f16(float16x4_t __p0) {
	56408	poly64x1_t __ret;
	56409	__ret = (poly64x1_t)(__p0);
	56410	return __ret;
	56411	}
	56412	__ai poly64x1_t vreinterpret_p64_s32(int32x2_t __p0) {
	56413	poly64x1_t __ret;
	56414	__ret = (poly64x1_t)(__p0);
	56415	return __ret;
	56416	}
	56417	__ai poly64x1_t vreinterpret_p64_s64(int64x1_t __p0) {
	56418	poly64x1_t __ret;
	56419	__ret = (poly64x1_t)(__p0);
	56420	return __ret;
	56421	}
	56422	__ai poly64x1_t vreinterpret_p64_s16(int16x4_t __p0) {
	56423	poly64x1_t __ret;
	56424	__ret = (poly64x1_t)(__p0);
	56425	return __ret;
	56426	}
	56427	__ai poly16x4_t vreinterpret_p16_p8(poly8x8_t __p0) {
	56428	poly16x4_t __ret;
	56429	__ret = (poly16x4_t)(__p0);
	56430	return __ret;
	56431	}
	56432	__ai poly16x4_t vreinterpret_p16_p64(poly64x1_t __p0) {
	56433	poly16x4_t __ret;
	56434	__ret = (poly16x4_t)(__p0);
	56435	return __ret;
	56436	}
	56437	__ai poly16x4_t vreinterpret_p16_u8(uint8x8_t __p0) {
	56438	poly16x4_t __ret;
	56439	__ret = (poly16x4_t)(__p0);
54751	56440	return __ret;
54752	56441	}
54753		#else
54754		__ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
54755		float64x2_t __ret;
54756		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
54757		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54758		__ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
54759		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	56442	__ai poly16x4_t vreinterpret_p16_u32(uint32x2_t __p0) {
	56443	poly16x4_t __ret;
	56444	__ret = (poly16x4_t)(__p0);
54760	56445	return __ret;
54761	56446	}
54762		#endif
54763
54764		__ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
54765		float64x1_t __ret;
54766		__ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	56447	__ai poly16x4_t vreinterpret_p16_u64(uint64x1_t __p0) {
	56448	poly16x4_t __ret;
	56449	__ret = (poly16x4_t)(__p0);
54767	56450	return __ret;
54768	56451	}
54769		#ifdef __LITTLE_ENDIAN__
54770		__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
54771		float64_t __ret;
54772		__ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__p0);
	56452	__ai poly16x4_t vreinterpret_p16_u16(uint16x4_t __p0) {
	56453	poly16x4_t __ret;
	56454	__ret = (poly16x4_t)(__p0);
54773	56455	return __ret;
54774	56456	}
54775		#else
54776		__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
54777		float64_t __ret;
54778		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
54779		__ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__rev0);
	56457	__ai poly16x4_t vreinterpret_p16_s8(int8x8_t __p0) {
	56458	poly16x4_t __ret;
	56459	__ret = (poly16x4_t)(__p0);
54780	56460	return __ret;
54781	56461	}
54782		#endif
54783
54784		#ifdef __LITTLE_ENDIAN__
54785		__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
54786		float32_t __ret;
54787		__ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__p0);
	56462	__ai poly16x4_t vreinterpret_p16_f64(float64x1_t __p0) {
	56463	poly16x4_t __ret;
	56464	__ret = (poly16x4_t)(__p0);
54788	56465	return __ret;
54789	56466	}
54790		#else
54791		__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
54792		float32_t __ret;
54793		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
54794		__ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__rev0);
	56467	__ai poly16x4_t vreinterpret_p16_f32(float32x2_t __p0) {
	56468	poly16x4_t __ret;
	56469	__ret = (poly16x4_t)(__p0);
54795	56470	return __ret;
54796	56471	}
54797		#endif
54798
54799		#ifdef __LITTLE_ENDIAN__
54800		__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
54801		float32_t __ret;
54802		__ret = (float32_t) __builtin_neon_vmaxnmv_f32(__p0);
	56472	__ai poly16x4_t vreinterpret_p16_f16(float16x4_t __p0) {
	56473	poly16x4_t __ret;
	56474	__ret = (poly16x4_t)(__p0);
54803	56475	return __ret;
54804	56476	}
54805		#else
54806		__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
54807		float32_t __ret;
54808		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
54809		__ret = (float32_t) __builtin_neon_vmaxnmv_f32(__rev0);
	56477	__ai poly16x4_t vreinterpret_p16_s32(int32x2_t __p0) {
	56478	poly16x4_t __ret;
	56479	__ret = (poly16x4_t)(__p0);
54810	56480	return __ret;
54811	56481	}
54812		#endif
54813
54814		#ifdef __LITTLE_ENDIAN__
54815		__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
54816		uint8_t __ret;
54817		__ret = (uint8_t) __builtin_neon_vmaxvq_u8(__p0);
	56482	__ai poly16x4_t vreinterpret_p16_s64(int64x1_t __p0) {
	56483	poly16x4_t __ret;
	56484	__ret = (poly16x4_t)(__p0);
54818	56485	return __ret;
54819	56486	}
54820		#else
54821		__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
54822		uint8_t __ret;
54823		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54824		__ret = (uint8_t) __builtin_neon_vmaxvq_u8(__rev0);
	56487	__ai poly16x4_t vreinterpret_p16_s16(int16x4_t __p0) {
	56488	poly16x4_t __ret;
	56489	__ret = (poly16x4_t)(__p0);
54825	56490	return __ret;
54826	56491	}
54827		#endif
54828
54829		#ifdef __LITTLE_ENDIAN__
54830		__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
54831		uint32_t __ret;
54832		__ret = (uint32_t) __builtin_neon_vmaxvq_u32(__p0);
	56492	__ai poly8x16_t vreinterpretq_p8_p128(poly128_t __p0) {
	56493	poly8x16_t __ret;
	56494	__ret = (poly8x16_t)(__p0);
	56495	return __ret;
	56496	}
	56497	__ai poly8x16_t vreinterpretq_p8_p64(poly64x2_t __p0) {
	56498	poly8x16_t __ret;
	56499	__ret = (poly8x16_t)(__p0);
	56500	return __ret;
	56501	}
	56502	__ai poly8x16_t vreinterpretq_p8_p16(poly16x8_t __p0) {
	56503	poly8x16_t __ret;
	56504	__ret = (poly8x16_t)(__p0);
	56505	return __ret;
	56506	}
	56507	__ai poly8x16_t vreinterpretq_p8_u8(uint8x16_t __p0) {
	56508	poly8x16_t __ret;
	56509	__ret = (poly8x16_t)(__p0);
	56510	return __ret;
	56511	}
	56512	__ai poly8x16_t vreinterpretq_p8_u32(uint32x4_t __p0) {
	56513	poly8x16_t __ret;
	56514	__ret = (poly8x16_t)(__p0);
	56515	return __ret;
	56516	}
	56517	__ai poly8x16_t vreinterpretq_p8_u64(uint64x2_t __p0) {
	56518	poly8x16_t __ret;
	56519	__ret = (poly8x16_t)(__p0);
	56520	return __ret;
	56521	}
	56522	__ai poly8x16_t vreinterpretq_p8_u16(uint16x8_t __p0) {
	56523	poly8x16_t __ret;
	56524	__ret = (poly8x16_t)(__p0);
	56525	return __ret;
	56526	}
	56527	__ai poly8x16_t vreinterpretq_p8_s8(int8x16_t __p0) {
	56528	poly8x16_t __ret;
	56529	__ret = (poly8x16_t)(__p0);
	56530	return __ret;
	56531	}
	56532	__ai poly8x16_t vreinterpretq_p8_f64(float64x2_t __p0) {
	56533	poly8x16_t __ret;
	56534	__ret = (poly8x16_t)(__p0);
	56535	return __ret;
	56536	}
	56537	__ai poly8x16_t vreinterpretq_p8_f32(float32x4_t __p0) {
	56538	poly8x16_t __ret;
	56539	__ret = (poly8x16_t)(__p0);
	56540	return __ret;
	56541	}
	56542	__ai poly8x16_t vreinterpretq_p8_f16(float16x8_t __p0) {
	56543	poly8x16_t __ret;
	56544	__ret = (poly8x16_t)(__p0);
	56545	return __ret;
	56546	}
	56547	__ai poly8x16_t vreinterpretq_p8_s32(int32x4_t __p0) {
	56548	poly8x16_t __ret;
	56549	__ret = (poly8x16_t)(__p0);
	56550	return __ret;
	56551	}
	56552	__ai poly8x16_t vreinterpretq_p8_s64(int64x2_t __p0) {
	56553	poly8x16_t __ret;
	56554	__ret = (poly8x16_t)(__p0);
	56555	return __ret;
	56556	}
	56557	__ai poly8x16_t vreinterpretq_p8_s16(int16x8_t __p0) {
	56558	poly8x16_t __ret;
	56559	__ret = (poly8x16_t)(__p0);
	56560	return __ret;
	56561	}
	56562	__ai poly128_t vreinterpretq_p128_p8(poly8x16_t __p0) {
	56563	poly128_t __ret;
	56564	__ret = (poly128_t)(__p0);
	56565	return __ret;
	56566	}
	56567	__ai poly128_t vreinterpretq_p128_p64(poly64x2_t __p0) {
	56568	poly128_t __ret;
	56569	__ret = (poly128_t)(__p0);
	56570	return __ret;
	56571	}
	56572	__ai poly128_t vreinterpretq_p128_p16(poly16x8_t __p0) {
	56573	poly128_t __ret;
	56574	__ret = (poly128_t)(__p0);
	56575	return __ret;
	56576	}
	56577	__ai poly128_t vreinterpretq_p128_u8(uint8x16_t __p0) {
	56578	poly128_t __ret;
	56579	__ret = (poly128_t)(__p0);
	56580	return __ret;
	56581	}
	56582	__ai poly128_t vreinterpretq_p128_u32(uint32x4_t __p0) {
	56583	poly128_t __ret;
	56584	__ret = (poly128_t)(__p0);
	56585	return __ret;
	56586	}
	56587	__ai poly128_t vreinterpretq_p128_u64(uint64x2_t __p0) {
	56588	poly128_t __ret;
	56589	__ret = (poly128_t)(__p0);
	56590	return __ret;
	56591	}
	56592	__ai poly128_t vreinterpretq_p128_u16(uint16x8_t __p0) {
	56593	poly128_t __ret;
	56594	__ret = (poly128_t)(__p0);
	56595	return __ret;
	56596	}
	56597	__ai poly128_t vreinterpretq_p128_s8(int8x16_t __p0) {
	56598	poly128_t __ret;
	56599	__ret = (poly128_t)(__p0);
	56600	return __ret;
	56601	}
	56602	__ai poly128_t vreinterpretq_p128_f64(float64x2_t __p0) {
	56603	poly128_t __ret;
	56604	__ret = (poly128_t)(__p0);
	56605	return __ret;
	56606	}
	56607	__ai poly128_t vreinterpretq_p128_f32(float32x4_t __p0) {
	56608	poly128_t __ret;
	56609	__ret = (poly128_t)(__p0);
	56610	return __ret;
	56611	}
	56612	__ai poly128_t vreinterpretq_p128_f16(float16x8_t __p0) {
	56613	poly128_t __ret;
	56614	__ret = (poly128_t)(__p0);
	56615	return __ret;
	56616	}
	56617	__ai poly128_t vreinterpretq_p128_s32(int32x4_t __p0) {
	56618	poly128_t __ret;
	56619	__ret = (poly128_t)(__p0);
	56620	return __ret;
	56621	}
	56622	__ai poly128_t vreinterpretq_p128_s64(int64x2_t __p0) {
	56623	poly128_t __ret;
	56624	__ret = (poly128_t)(__p0);
	56625	return __ret;
	56626	}
	56627	__ai poly128_t vreinterpretq_p128_s16(int16x8_t __p0) {
	56628	poly128_t __ret;
	56629	__ret = (poly128_t)(__p0);
	56630	return __ret;
	56631	}
	56632	__ai poly64x2_t vreinterpretq_p64_p8(poly8x16_t __p0) {
	56633	poly64x2_t __ret;
	56634	__ret = (poly64x2_t)(__p0);
	56635	return __ret;
	56636	}
	56637	__ai poly64x2_t vreinterpretq_p64_p128(poly128_t __p0) {
	56638	poly64x2_t __ret;
	56639	__ret = (poly64x2_t)(__p0);
	56640	return __ret;
	56641	}
	56642	__ai poly64x2_t vreinterpretq_p64_p16(poly16x8_t __p0) {
	56643	poly64x2_t __ret;
	56644	__ret = (poly64x2_t)(__p0);
	56645	return __ret;
	56646	}
	56647	__ai poly64x2_t vreinterpretq_p64_u8(uint8x16_t __p0) {
	56648	poly64x2_t __ret;
	56649	__ret = (poly64x2_t)(__p0);
	56650	return __ret;
	56651	}
	56652	__ai poly64x2_t vreinterpretq_p64_u32(uint32x4_t __p0) {
	56653	poly64x2_t __ret;
	56654	__ret = (poly64x2_t)(__p0);
	56655	return __ret;
	56656	}
	56657	__ai poly64x2_t vreinterpretq_p64_u64(uint64x2_t __p0) {
	56658	poly64x2_t __ret;
	56659	__ret = (poly64x2_t)(__p0);
	56660	return __ret;
	56661	}
	56662	__ai poly64x2_t vreinterpretq_p64_u16(uint16x8_t __p0) {
	56663	poly64x2_t __ret;
	56664	__ret = (poly64x2_t)(__p0);
	56665	return __ret;
	56666	}
	56667	__ai poly64x2_t vreinterpretq_p64_s8(int8x16_t __p0) {
	56668	poly64x2_t __ret;
	56669	__ret = (poly64x2_t)(__p0);
	56670	return __ret;
	56671	}
	56672	__ai poly64x2_t vreinterpretq_p64_f64(float64x2_t __p0) {
	56673	poly64x2_t __ret;
	56674	__ret = (poly64x2_t)(__p0);
	56675	return __ret;
	56676	}
	56677	__ai poly64x2_t vreinterpretq_p64_f32(float32x4_t __p0) {
	56678	poly64x2_t __ret;
	56679	__ret = (poly64x2_t)(__p0);
	56680	return __ret;
	56681	}
	56682	__ai poly64x2_t vreinterpretq_p64_f16(float16x8_t __p0) {
	56683	poly64x2_t __ret;
	56684	__ret = (poly64x2_t)(__p0);
	56685	return __ret;
	56686	}
	56687	__ai poly64x2_t vreinterpretq_p64_s32(int32x4_t __p0) {
	56688	poly64x2_t __ret;
	56689	__ret = (poly64x2_t)(__p0);
	56690	return __ret;
	56691	}
	56692	__ai poly64x2_t vreinterpretq_p64_s64(int64x2_t __p0) {
	56693	poly64x2_t __ret;
	56694	__ret = (poly64x2_t)(__p0);
	56695	return __ret;
	56696	}
	56697	__ai poly64x2_t vreinterpretq_p64_s16(int16x8_t __p0) {
	56698	poly64x2_t __ret;
	56699	__ret = (poly64x2_t)(__p0);
	56700	return __ret;
	56701	}
	56702	__ai poly16x8_t vreinterpretq_p16_p8(poly8x16_t __p0) {
	56703	poly16x8_t __ret;
	56704	__ret = (poly16x8_t)(__p0);
	56705	return __ret;
	56706	}
	56707	__ai poly16x8_t vreinterpretq_p16_p128(poly128_t __p0) {
	56708	poly16x8_t __ret;
	56709	__ret = (poly16x8_t)(__p0);
	56710	return __ret;
	56711	}
	56712	__ai poly16x8_t vreinterpretq_p16_p64(poly64x2_t __p0) {
	56713	poly16x8_t __ret;
	56714	__ret = (poly16x8_t)(__p0);
	56715	return __ret;
	56716	}
	56717	__ai poly16x8_t vreinterpretq_p16_u8(uint8x16_t __p0) {
	56718	poly16x8_t __ret;
	56719	__ret = (poly16x8_t)(__p0);
	56720	return __ret;
	56721	}
	56722	__ai poly16x8_t vreinterpretq_p16_u32(uint32x4_t __p0) {
	56723	poly16x8_t __ret;
	56724	__ret = (poly16x8_t)(__p0);
	56725	return __ret;
	56726	}
	56727	__ai poly16x8_t vreinterpretq_p16_u64(uint64x2_t __p0) {
	56728	poly16x8_t __ret;
	56729	__ret = (poly16x8_t)(__p0);
	56730	return __ret;
	56731	}
	56732	__ai poly16x8_t vreinterpretq_p16_u16(uint16x8_t __p0) {
	56733	poly16x8_t __ret;
	56734	__ret = (poly16x8_t)(__p0);
	56735	return __ret;
	56736	}
	56737	__ai poly16x8_t vreinterpretq_p16_s8(int8x16_t __p0) {
	56738	poly16x8_t __ret;
	56739	__ret = (poly16x8_t)(__p0);
	56740	return __ret;
	56741	}
	56742	__ai poly16x8_t vreinterpretq_p16_f64(float64x2_t __p0) {
	56743	poly16x8_t __ret;
	56744	__ret = (poly16x8_t)(__p0);
	56745	return __ret;
	56746	}
	56747	__ai poly16x8_t vreinterpretq_p16_f32(float32x4_t __p0) {
	56748	poly16x8_t __ret;
	56749	__ret = (poly16x8_t)(__p0);
	56750	return __ret;
	56751	}
	56752	__ai poly16x8_t vreinterpretq_p16_f16(float16x8_t __p0) {
	56753	poly16x8_t __ret;
	56754	__ret = (poly16x8_t)(__p0);
	56755	return __ret;
	56756	}
	56757	__ai poly16x8_t vreinterpretq_p16_s32(int32x4_t __p0) {
	56758	poly16x8_t __ret;
	56759	__ret = (poly16x8_t)(__p0);
	56760	return __ret;
	56761	}
	56762	__ai poly16x8_t vreinterpretq_p16_s64(int64x2_t __p0) {
	56763	poly16x8_t __ret;
	56764	__ret = (poly16x8_t)(__p0);
	56765	return __ret;
	56766	}
	56767	__ai poly16x8_t vreinterpretq_p16_s16(int16x8_t __p0) {
	56768	poly16x8_t __ret;
	56769	__ret = (poly16x8_t)(__p0);
	56770	return __ret;
	56771	}
	56772	__ai uint8x16_t vreinterpretq_u8_p8(poly8x16_t __p0) {
	56773	uint8x16_t __ret;
	56774	__ret = (uint8x16_t)(__p0);
	56775	return __ret;
	56776	}
	56777	__ai uint8x16_t vreinterpretq_u8_p128(poly128_t __p0) {
	56778	uint8x16_t __ret;
	56779	__ret = (uint8x16_t)(__p0);
	56780	return __ret;
	56781	}
	56782	__ai uint8x16_t vreinterpretq_u8_p64(poly64x2_t __p0) {
	56783	uint8x16_t __ret;
	56784	__ret = (uint8x16_t)(__p0);
	56785	return __ret;
	56786	}
	56787	__ai uint8x16_t vreinterpretq_u8_p16(poly16x8_t __p0) {
	56788	uint8x16_t __ret;
	56789	__ret = (uint8x16_t)(__p0);
	56790	return __ret;
	56791	}
	56792	__ai uint8x16_t vreinterpretq_u8_u32(uint32x4_t __p0) {
	56793	uint8x16_t __ret;
	56794	__ret = (uint8x16_t)(__p0);
	56795	return __ret;
	56796	}
	56797	__ai uint8x16_t vreinterpretq_u8_u64(uint64x2_t __p0) {
	56798	uint8x16_t __ret;
	56799	__ret = (uint8x16_t)(__p0);
	56800	return __ret;
	56801	}
	56802	__ai uint8x16_t vreinterpretq_u8_u16(uint16x8_t __p0) {
	56803	uint8x16_t __ret;
	56804	__ret = (uint8x16_t)(__p0);
	56805	return __ret;
	56806	}
	56807	__ai uint8x16_t vreinterpretq_u8_s8(int8x16_t __p0) {
	56808	uint8x16_t __ret;
	56809	__ret = (uint8x16_t)(__p0);
	56810	return __ret;
	56811	}
	56812	__ai uint8x16_t vreinterpretq_u8_f64(float64x2_t __p0) {
	56813	uint8x16_t __ret;
	56814	__ret = (uint8x16_t)(__p0);
	56815	return __ret;
	56816	}
	56817	__ai uint8x16_t vreinterpretq_u8_f32(float32x4_t __p0) {
	56818	uint8x16_t __ret;
	56819	__ret = (uint8x16_t)(__p0);
	56820	return __ret;
	56821	}
	56822	__ai uint8x16_t vreinterpretq_u8_f16(float16x8_t __p0) {
	56823	uint8x16_t __ret;
	56824	__ret = (uint8x16_t)(__p0);
	56825	return __ret;
	56826	}
	56827	__ai uint8x16_t vreinterpretq_u8_s32(int32x4_t __p0) {
	56828	uint8x16_t __ret;
	56829	__ret = (uint8x16_t)(__p0);
	56830	return __ret;
	56831	}
	56832	__ai uint8x16_t vreinterpretq_u8_s64(int64x2_t __p0) {
	56833	uint8x16_t __ret;
	56834	__ret = (uint8x16_t)(__p0);
	56835	return __ret;
	56836	}
	56837	__ai uint8x16_t vreinterpretq_u8_s16(int16x8_t __p0) {
	56838	uint8x16_t __ret;
	56839	__ret = (uint8x16_t)(__p0);
	56840	return __ret;
	56841	}
	56842	__ai uint32x4_t vreinterpretq_u32_p8(poly8x16_t __p0) {
	56843	uint32x4_t __ret;
	56844	__ret = (uint32x4_t)(__p0);
	56845	return __ret;
	56846	}
	56847	__ai uint32x4_t vreinterpretq_u32_p128(poly128_t __p0) {
	56848	uint32x4_t __ret;
	56849	__ret = (uint32x4_t)(__p0);
	56850	return __ret;
	56851	}
	56852	__ai uint32x4_t vreinterpretq_u32_p64(poly64x2_t __p0) {
	56853	uint32x4_t __ret;
	56854	__ret = (uint32x4_t)(__p0);
	56855	return __ret;
	56856	}
	56857	__ai uint32x4_t vreinterpretq_u32_p16(poly16x8_t __p0) {
	56858	uint32x4_t __ret;
	56859	__ret = (uint32x4_t)(__p0);
	56860	return __ret;
	56861	}
	56862	__ai uint32x4_t vreinterpretq_u32_u8(uint8x16_t __p0) {
	56863	uint32x4_t __ret;
	56864	__ret = (uint32x4_t)(__p0);
	56865	return __ret;
	56866	}
	56867	__ai uint32x4_t vreinterpretq_u32_u64(uint64x2_t __p0) {
	56868	uint32x4_t __ret;
	56869	__ret = (uint32x4_t)(__p0);
	56870	return __ret;
	56871	}
	56872	__ai uint32x4_t vreinterpretq_u32_u16(uint16x8_t __p0) {
	56873	uint32x4_t __ret;
	56874	__ret = (uint32x4_t)(__p0);
	56875	return __ret;
	56876	}
	56877	__ai uint32x4_t vreinterpretq_u32_s8(int8x16_t __p0) {
	56878	uint32x4_t __ret;
	56879	__ret = (uint32x4_t)(__p0);
	56880	return __ret;
	56881	}
	56882	__ai uint32x4_t vreinterpretq_u32_f64(float64x2_t __p0) {
	56883	uint32x4_t __ret;
	56884	__ret = (uint32x4_t)(__p0);
	56885	return __ret;
	56886	}
	56887	__ai uint32x4_t vreinterpretq_u32_f32(float32x4_t __p0) {
	56888	uint32x4_t __ret;
	56889	__ret = (uint32x4_t)(__p0);
	56890	return __ret;
	56891	}
	56892	__ai uint32x4_t vreinterpretq_u32_f16(float16x8_t __p0) {
	56893	uint32x4_t __ret;
	56894	__ret = (uint32x4_t)(__p0);
	56895	return __ret;
	56896	}
	56897	__ai uint32x4_t vreinterpretq_u32_s32(int32x4_t __p0) {
	56898	uint32x4_t __ret;
	56899	__ret = (uint32x4_t)(__p0);
	56900	return __ret;
	56901	}
	56902	__ai uint32x4_t vreinterpretq_u32_s64(int64x2_t __p0) {
	56903	uint32x4_t __ret;
	56904	__ret = (uint32x4_t)(__p0);
	56905	return __ret;
	56906	}
	56907	__ai uint32x4_t vreinterpretq_u32_s16(int16x8_t __p0) {
	56908	uint32x4_t __ret;
	56909	__ret = (uint32x4_t)(__p0);
	56910	return __ret;
	56911	}
	56912	__ai uint64x2_t vreinterpretq_u64_p8(poly8x16_t __p0) {
	56913	uint64x2_t __ret;
	56914	__ret = (uint64x2_t)(__p0);
54833	56915	return __ret;
54834	56916	}
54835		#else
54836		__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
54837		uint32_t __ret;
54838		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
54839		__ret = (uint32_t) __builtin_neon_vmaxvq_u32(__rev0);
	56917	__ai uint64x2_t vreinterpretq_u64_p128(poly128_t __p0) {
	56918	uint64x2_t __ret;
	56919	__ret = (uint64x2_t)(__p0);
54840	56920	return __ret;
54841	56921	}
54842		#endif
54843
54844		#ifdef __LITTLE_ENDIAN__
54845		__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
54846		uint16_t __ret;
54847		__ret = (uint16_t) __builtin_neon_vmaxvq_u16(__p0);
	56922	__ai uint64x2_t vreinterpretq_u64_p64(poly64x2_t __p0) {
	56923	uint64x2_t __ret;
	56924	__ret = (uint64x2_t)(__p0);
54848	56925	return __ret;
54849	56926	}
54850		#else
54851		__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
54852		uint16_t __ret;
54853		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
54854		__ret = (uint16_t) __builtin_neon_vmaxvq_u16(__rev0);
	56927	__ai uint64x2_t vreinterpretq_u64_p16(poly16x8_t __p0) {
	56928	uint64x2_t __ret;
	56929	__ret = (uint64x2_t)(__p0);
54855	56930	return __ret;
54856	56931	}
54857		#endif
54858
54859		#ifdef __LITTLE_ENDIAN__
54860		__ai int8_t vmaxvq_s8(int8x16_t __p0) {
54861		int8_t __ret;
54862		__ret = (int8_t) __builtin_neon_vmaxvq_s8(__p0);
	56932	__ai uint64x2_t vreinterpretq_u64_u8(uint8x16_t __p0) {
	56933	uint64x2_t __ret;
	56934	__ret = (uint64x2_t)(__p0);
54863	56935	return __ret;
54864	56936	}
54865		#else
54866		__ai int8_t vmaxvq_s8(int8x16_t __p0) {
54867		int8_t __ret;
54868		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54869		__ret = (int8_t) __builtin_neon_vmaxvq_s8(__rev0);
	56937	__ai uint64x2_t vreinterpretq_u64_u32(uint32x4_t __p0) {
	56938	uint64x2_t __ret;
	56939	__ret = (uint64x2_t)(__p0);
54870	56940	return __ret;
54871	56941	}
54872		#endif
54873
54874		#ifdef __LITTLE_ENDIAN__
54875		__ai float64_t vmaxvq_f64(float64x2_t __p0) {
54876		float64_t __ret;
54877		__ret = (float64_t) __builtin_neon_vmaxvq_f64(__p0);
	56942	__ai uint64x2_t vreinterpretq_u64_u16(uint16x8_t __p0) {
	56943	uint64x2_t __ret;
	56944	__ret = (uint64x2_t)(__p0);
54878	56945	return __ret;
54879	56946	}
54880		#else
54881		__ai float64_t vmaxvq_f64(float64x2_t __p0) {
54882		float64_t __ret;
54883		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
54884		__ret = (float64_t) __builtin_neon_vmaxvq_f64(__rev0);
	56947	__ai uint64x2_t vreinterpretq_u64_s8(int8x16_t __p0) {
	56948	uint64x2_t __ret;
	56949	__ret = (uint64x2_t)(__p0);
54885	56950	return __ret;
54886	56951	}
54887		#endif
54888
54889		#ifdef __LITTLE_ENDIAN__
54890		__ai float32_t vmaxvq_f32(float32x4_t __p0) {
54891		float32_t __ret;
54892		__ret = (float32_t) __builtin_neon_vmaxvq_f32(__p0);
	56952	__ai uint64x2_t vreinterpretq_u64_f64(float64x2_t __p0) {
	56953	uint64x2_t __ret;
	56954	__ret = (uint64x2_t)(__p0);
54893	56955	return __ret;
54894	56956	}
54895		#else
54896		__ai float32_t vmaxvq_f32(float32x4_t __p0) {
54897		float32_t __ret;
54898		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
54899		__ret = (float32_t) __builtin_neon_vmaxvq_f32(__rev0);
	56957	__ai uint64x2_t vreinterpretq_u64_f32(float32x4_t __p0) {
	56958	uint64x2_t __ret;
	56959	__ret = (uint64x2_t)(__p0);
54900	56960	return __ret;
54901	56961	}
54902		#endif
54903
54904		#ifdef __LITTLE_ENDIAN__
54905		__ai int32_t vmaxvq_s32(int32x4_t __p0) {
54906		int32_t __ret;
54907		__ret = (int32_t) __builtin_neon_vmaxvq_s32(__p0);
	56962	__ai uint64x2_t vreinterpretq_u64_f16(float16x8_t __p0) {
	56963	uint64x2_t __ret;
	56964	__ret = (uint64x2_t)(__p0);
54908	56965	return __ret;
54909	56966	}
54910		#else
54911		__ai int32_t vmaxvq_s32(int32x4_t __p0) {
54912		int32_t __ret;
54913		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
54914		__ret = (int32_t) __builtin_neon_vmaxvq_s32(__rev0);
	56967	__ai uint64x2_t vreinterpretq_u64_s32(int32x4_t __p0) {
	56968	uint64x2_t __ret;
	56969	__ret = (uint64x2_t)(__p0);
54915	56970	return __ret;
54916	56971	}
54917		#endif
54918
54919		#ifdef __LITTLE_ENDIAN__
54920		__ai int16_t vmaxvq_s16(int16x8_t __p0) {
54921		int16_t __ret;
54922		__ret = (int16_t) __builtin_neon_vmaxvq_s16(__p0);
	56972	__ai uint64x2_t vreinterpretq_u64_s64(int64x2_t __p0) {
	56973	uint64x2_t __ret;
	56974	__ret = (uint64x2_t)(__p0);
54923	56975	return __ret;
54924	56976	}
54925		#else
54926		__ai int16_t vmaxvq_s16(int16x8_t __p0) {
54927		int16_t __ret;
54928		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
54929		__ret = (int16_t) __builtin_neon_vmaxvq_s16(__rev0);
	56977	__ai uint64x2_t vreinterpretq_u64_s16(int16x8_t __p0) {
	56978	uint64x2_t __ret;
	56979	__ret = (uint64x2_t)(__p0);
54930	56980	return __ret;
54931	56981	}
54932		#endif
54933
54934		#ifdef __LITTLE_ENDIAN__
54935		__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
54936		uint8_t __ret;
54937		__ret = (uint8_t) __builtin_neon_vmaxv_u8(__p0);
	56982	__ai uint16x8_t vreinterpretq_u16_p8(poly8x16_t __p0) {
	56983	uint16x8_t __ret;
	56984	__ret = (uint16x8_t)(__p0);
54938	56985	return __ret;
54939	56986	}
54940		#else
54941		__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
54942		uint8_t __ret;
54943		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
54944		__ret = (uint8_t) __builtin_neon_vmaxv_u8(__rev0);
	56987	__ai uint16x8_t vreinterpretq_u16_p128(poly128_t __p0) {
	56988	uint16x8_t __ret;
	56989	__ret = (uint16x8_t)(__p0);
54945	56990	return __ret;
54946	56991	}
54947		#endif
54948
54949		#ifdef __LITTLE_ENDIAN__
54950		__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
54951		uint32_t __ret;
54952		__ret = (uint32_t) __builtin_neon_vmaxv_u32(__p0);
	56992	__ai uint16x8_t vreinterpretq_u16_p64(poly64x2_t __p0) {
	56993	uint16x8_t __ret;
	56994	__ret = (uint16x8_t)(__p0);
54953	56995	return __ret;
54954	56996	}
54955		#else
54956		__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
54957		uint32_t __ret;
54958		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
54959		__ret = (uint32_t) __builtin_neon_vmaxv_u32(__rev0);
	56997	__ai uint16x8_t vreinterpretq_u16_p16(poly16x8_t __p0) {
	56998	uint16x8_t __ret;
	56999	__ret = (uint16x8_t)(__p0);
54960	57000	return __ret;
54961	57001	}
54962		#endif
54963
54964		#ifdef __LITTLE_ENDIAN__
54965		__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
54966		uint16_t __ret;
54967		__ret = (uint16_t) __builtin_neon_vmaxv_u16(__p0);
	57002	__ai uint16x8_t vreinterpretq_u16_u8(uint8x16_t __p0) {
	57003	uint16x8_t __ret;
	57004	__ret = (uint16x8_t)(__p0);
54968	57005	return __ret;
54969	57006	}
54970		#else
54971		__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
54972		uint16_t __ret;
54973		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
54974		__ret = (uint16_t) __builtin_neon_vmaxv_u16(__rev0);
	57007	__ai uint16x8_t vreinterpretq_u16_u32(uint32x4_t __p0) {
	57008	uint16x8_t __ret;
	57009	__ret = (uint16x8_t)(__p0);
54975	57010	return __ret;
54976	57011	}
54977		#endif
54978
54979		#ifdef __LITTLE_ENDIAN__
54980		__ai int8_t vmaxv_s8(int8x8_t __p0) {
54981		int8_t __ret;
54982		__ret = (int8_t) __builtin_neon_vmaxv_s8(__p0);
	57012	__ai uint16x8_t vreinterpretq_u16_u64(uint64x2_t __p0) {
	57013	uint16x8_t __ret;
	57014	__ret = (uint16x8_t)(__p0);
54983	57015	return __ret;
54984	57016	}
54985		#else
54986		__ai int8_t vmaxv_s8(int8x8_t __p0) {
54987		int8_t __ret;
54988		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
54989		__ret = (int8_t) __builtin_neon_vmaxv_s8(__rev0);
	57017	__ai uint16x8_t vreinterpretq_u16_s8(int8x16_t __p0) {
	57018	uint16x8_t __ret;
	57019	__ret = (uint16x8_t)(__p0);
54990	57020	return __ret;
54991	57021	}
54992		#endif
54993
54994		#ifdef __LITTLE_ENDIAN__
54995		__ai float32_t vmaxv_f32(float32x2_t __p0) {
54996		float32_t __ret;
54997		__ret = (float32_t) __builtin_neon_vmaxv_f32(__p0);
	57022	__ai uint16x8_t vreinterpretq_u16_f64(float64x2_t __p0) {
	57023	uint16x8_t __ret;
	57024	__ret = (uint16x8_t)(__p0);
54998	57025	return __ret;
54999	57026	}
55000		#else
55001		__ai float32_t vmaxv_f32(float32x2_t __p0) {
55002		float32_t __ret;
55003		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55004		__ret = (float32_t) __builtin_neon_vmaxv_f32(__rev0);
	57027	__ai uint16x8_t vreinterpretq_u16_f32(float32x4_t __p0) {
	57028	uint16x8_t __ret;
	57029	__ret = (uint16x8_t)(__p0);
55005	57030	return __ret;
55006	57031	}
55007		#endif
55008
55009		#ifdef __LITTLE_ENDIAN__
55010		__ai int32_t vmaxv_s32(int32x2_t __p0) {
55011		int32_t __ret;
55012		__ret = (int32_t) __builtin_neon_vmaxv_s32(__p0);
	57032	__ai uint16x8_t vreinterpretq_u16_f16(float16x8_t __p0) {
	57033	uint16x8_t __ret;
	57034	__ret = (uint16x8_t)(__p0);
55013	57035	return __ret;
55014	57036	}
55015		#else
55016		__ai int32_t vmaxv_s32(int32x2_t __p0) {
55017		int32_t __ret;
55018		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55019		__ret = (int32_t) __builtin_neon_vmaxv_s32(__rev0);
	57037	__ai uint16x8_t vreinterpretq_u16_s32(int32x4_t __p0) {
	57038	uint16x8_t __ret;
	57039	__ret = (uint16x8_t)(__p0);
55020	57040	return __ret;
55021	57041	}
55022		#endif
55023
55024		#ifdef __LITTLE_ENDIAN__
55025		__ai int16_t vmaxv_s16(int16x4_t __p0) {
55026		int16_t __ret;
55027		__ret = (int16_t) __builtin_neon_vmaxv_s16(__p0);
	57042	__ai uint16x8_t vreinterpretq_u16_s64(int64x2_t __p0) {
	57043	uint16x8_t __ret;
	57044	__ret = (uint16x8_t)(__p0);
55028	57045	return __ret;
55029	57046	}
55030		#else
55031		__ai int16_t vmaxv_s16(int16x4_t __p0) {
55032		int16_t __ret;
55033		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55034		__ret = (int16_t) __builtin_neon_vmaxv_s16(__rev0);
	57047	__ai uint16x8_t vreinterpretq_u16_s16(int16x8_t __p0) {
	57048	uint16x8_t __ret;
	57049	__ret = (uint16x8_t)(__p0);
55035	57050	return __ret;
55036	57051	}
55037		#endif
55038
55039		#ifdef __LITTLE_ENDIAN__
55040		__ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
55041		float64x2_t __ret;
55042		__ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	57052	__ai int8x16_t vreinterpretq_s8_p8(poly8x16_t __p0) {
	57053	int8x16_t __ret;
	57054	__ret = (int8x16_t)(__p0);
55043	57055	return __ret;
55044	57056	}
55045		#else
55046		__ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
55047		float64x2_t __ret;
55048		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55049		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55050		__ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
55051		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57057	__ai int8x16_t vreinterpretq_s8_p128(poly128_t __p0) {
	57058	int8x16_t __ret;
	57059	__ret = (int8x16_t)(__p0);
55052	57060	return __ret;
55053	57061	}
55054		#endif
55055
55056		__ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
55057		float64x1_t __ret;
55058		__ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	57062	__ai int8x16_t vreinterpretq_s8_p64(poly64x2_t __p0) {
	57063	int8x16_t __ret;
	57064	__ret = (int8x16_t)(__p0);
55059	57065	return __ret;
55060	57066	}
55061		#ifdef __LITTLE_ENDIAN__
55062		__ai float64_t vminnmvq_f64(float64x2_t __p0) {
55063		float64_t __ret;
55064		__ret = (float64_t) __builtin_neon_vminnmvq_f64(__p0);
	57067	__ai int8x16_t vreinterpretq_s8_p16(poly16x8_t __p0) {
	57068	int8x16_t __ret;
	57069	__ret = (int8x16_t)(__p0);
55065	57070	return __ret;
55066	57071	}
55067		#else
55068		__ai float64_t vminnmvq_f64(float64x2_t __p0) {
55069		float64_t __ret;
55070		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55071		__ret = (float64_t) __builtin_neon_vminnmvq_f64(__rev0);
	57072	__ai int8x16_t vreinterpretq_s8_u8(uint8x16_t __p0) {
	57073	int8x16_t __ret;
	57074	__ret = (int8x16_t)(__p0);
55072	57075	return __ret;
55073	57076	}
55074		#endif
55075
55076		#ifdef __LITTLE_ENDIAN__
55077		__ai float32_t vminnmvq_f32(float32x4_t __p0) {
55078		float32_t __ret;
55079		__ret = (float32_t) __builtin_neon_vminnmvq_f32(__p0);
	57077	__ai int8x16_t vreinterpretq_s8_u32(uint32x4_t __p0) {
	57078	int8x16_t __ret;
	57079	__ret = (int8x16_t)(__p0);
55080	57080	return __ret;
55081	57081	}
55082		#else
55083		__ai float32_t vminnmvq_f32(float32x4_t __p0) {
55084		float32_t __ret;
55085		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55086		__ret = (float32_t) __builtin_neon_vminnmvq_f32(__rev0);
	57082	__ai int8x16_t vreinterpretq_s8_u64(uint64x2_t __p0) {
	57083	int8x16_t __ret;
	57084	__ret = (int8x16_t)(__p0);
55087	57085	return __ret;
55088	57086	}
55089		#endif
55090
55091		#ifdef __LITTLE_ENDIAN__
55092		__ai float32_t vminnmv_f32(float32x2_t __p0) {
55093		float32_t __ret;
55094		__ret = (float32_t) __builtin_neon_vminnmv_f32(__p0);
	57087	__ai int8x16_t vreinterpretq_s8_u16(uint16x8_t __p0) {
	57088	int8x16_t __ret;
	57089	__ret = (int8x16_t)(__p0);
55095	57090	return __ret;
55096	57091	}
55097		#else
55098		__ai float32_t vminnmv_f32(float32x2_t __p0) {
55099		float32_t __ret;
55100		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55101		__ret = (float32_t) __builtin_neon_vminnmv_f32(__rev0);
	57092	__ai int8x16_t vreinterpretq_s8_f64(float64x2_t __p0) {
	57093	int8x16_t __ret;
	57094	__ret = (int8x16_t)(__p0);
55102	57095	return __ret;
55103	57096	}
55104		#endif
55105
55106		#ifdef __LITTLE_ENDIAN__
55107		__ai uint8_t vminvq_u8(uint8x16_t __p0) {
55108		uint8_t __ret;
55109		__ret = (uint8_t) __builtin_neon_vminvq_u8(__p0);
	57097	__ai int8x16_t vreinterpretq_s8_f32(float32x4_t __p0) {
	57098	int8x16_t __ret;
	57099	__ret = (int8x16_t)(__p0);
55110	57100	return __ret;
55111	57101	}
55112		#else
55113		__ai uint8_t vminvq_u8(uint8x16_t __p0) {
55114		uint8_t __ret;
55115		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55116		__ret = (uint8_t) __builtin_neon_vminvq_u8(__rev0);
	57102	__ai int8x16_t vreinterpretq_s8_f16(float16x8_t __p0) {
	57103	int8x16_t __ret;
	57104	__ret = (int8x16_t)(__p0);
55117	57105	return __ret;
55118	57106	}
55119		#endif
55120
55121		#ifdef __LITTLE_ENDIAN__
55122		__ai uint32_t vminvq_u32(uint32x4_t __p0) {
55123		uint32_t __ret;
55124		__ret = (uint32_t) __builtin_neon_vminvq_u32(__p0);
	57107	__ai int8x16_t vreinterpretq_s8_s32(int32x4_t __p0) {
	57108	int8x16_t __ret;
	57109	__ret = (int8x16_t)(__p0);
55125	57110	return __ret;
55126	57111	}
55127		#else
55128		__ai uint32_t vminvq_u32(uint32x4_t __p0) {
55129		uint32_t __ret;
55130		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55131		__ret = (uint32_t) __builtin_neon_vminvq_u32(__rev0);
	57112	__ai int8x16_t vreinterpretq_s8_s64(int64x2_t __p0) {
	57113	int8x16_t __ret;
	57114	__ret = (int8x16_t)(__p0);
55132	57115	return __ret;
55133	57116	}
55134		#endif
55135
55136		#ifdef __LITTLE_ENDIAN__
55137		__ai uint16_t vminvq_u16(uint16x8_t __p0) {
55138		uint16_t __ret;
55139		__ret = (uint16_t) __builtin_neon_vminvq_u16(__p0);
	57117	__ai int8x16_t vreinterpretq_s8_s16(int16x8_t __p0) {
	57118	int8x16_t __ret;
	57119	__ret = (int8x16_t)(__p0);
55140	57120	return __ret;
55141	57121	}
55142		#else
55143		__ai uint16_t vminvq_u16(uint16x8_t __p0) {
55144		uint16_t __ret;
55145		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
55146		__ret = (uint16_t) __builtin_neon_vminvq_u16(__rev0);
	57122	__ai float64x2_t vreinterpretq_f64_p8(poly8x16_t __p0) {
	57123	float64x2_t __ret;
	57124	__ret = (float64x2_t)(__p0);
55147	57125	return __ret;
55148	57126	}
55149		#endif
55150
55151		#ifdef __LITTLE_ENDIAN__
55152		__ai int8_t vminvq_s8(int8x16_t __p0) {
55153		int8_t __ret;
55154		__ret = (int8_t) __builtin_neon_vminvq_s8(__p0);
	57127	__ai float64x2_t vreinterpretq_f64_p128(poly128_t __p0) {
	57128	float64x2_t __ret;
	57129	__ret = (float64x2_t)(__p0);
55155	57130	return __ret;
55156	57131	}
55157		#else
55158		__ai int8_t vminvq_s8(int8x16_t __p0) {
55159		int8_t __ret;
55160		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55161		__ret = (int8_t) __builtin_neon_vminvq_s8(__rev0);
	57132	__ai float64x2_t vreinterpretq_f64_p64(poly64x2_t __p0) {
	57133	float64x2_t __ret;
	57134	__ret = (float64x2_t)(__p0);
55162	57135	return __ret;
55163	57136	}
55164		#endif
55165
55166		#ifdef __LITTLE_ENDIAN__
55167		__ai float64_t vminvq_f64(float64x2_t __p0) {
55168		float64_t __ret;
55169		__ret = (float64_t) __builtin_neon_vminvq_f64(__p0);
	57137	__ai float64x2_t vreinterpretq_f64_p16(poly16x8_t __p0) {
	57138	float64x2_t __ret;
	57139	__ret = (float64x2_t)(__p0);
55170	57140	return __ret;
55171	57141	}
55172		#else
55173		__ai float64_t vminvq_f64(float64x2_t __p0) {
55174		float64_t __ret;
55175		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55176		__ret = (float64_t) __builtin_neon_vminvq_f64(__rev0);
	57142	__ai float64x2_t vreinterpretq_f64_u8(uint8x16_t __p0) {
	57143	float64x2_t __ret;
	57144	__ret = (float64x2_t)(__p0);
55177	57145	return __ret;
55178	57146	}
55179		#endif
55180
55181		#ifdef __LITTLE_ENDIAN__
55182		__ai float32_t vminvq_f32(float32x4_t __p0) {
55183		float32_t __ret;
55184		__ret = (float32_t) __builtin_neon_vminvq_f32(__p0);
	57147	__ai float64x2_t vreinterpretq_f64_u32(uint32x4_t __p0) {
	57148	float64x2_t __ret;
	57149	__ret = (float64x2_t)(__p0);
55185	57150	return __ret;
55186	57151	}
55187		#else
55188		__ai float32_t vminvq_f32(float32x4_t __p0) {
55189		float32_t __ret;
55190		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55191		__ret = (float32_t) __builtin_neon_vminvq_f32(__rev0);
	57152	__ai float64x2_t vreinterpretq_f64_u64(uint64x2_t __p0) {
	57153	float64x2_t __ret;
	57154	__ret = (float64x2_t)(__p0);
55192	57155	return __ret;
55193	57156	}
55194		#endif
55195
55196		#ifdef __LITTLE_ENDIAN__
55197		__ai int32_t vminvq_s32(int32x4_t __p0) {
55198		int32_t __ret;
55199		__ret = (int32_t) __builtin_neon_vminvq_s32(__p0);
	57157	__ai float64x2_t vreinterpretq_f64_u16(uint16x8_t __p0) {
	57158	float64x2_t __ret;
	57159	__ret = (float64x2_t)(__p0);
55200	57160	return __ret;
55201	57161	}
55202		#else
55203		__ai int32_t vminvq_s32(int32x4_t __p0) {
55204		int32_t __ret;
55205		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55206		__ret = (int32_t) __builtin_neon_vminvq_s32(__rev0);
	57162	__ai float64x2_t vreinterpretq_f64_s8(int8x16_t __p0) {
	57163	float64x2_t __ret;
	57164	__ret = (float64x2_t)(__p0);
55207	57165	return __ret;
55208	57166	}
55209		#endif
55210
55211		#ifdef __LITTLE_ENDIAN__
55212		__ai int16_t vminvq_s16(int16x8_t __p0) {
55213		int16_t __ret;
55214		__ret = (int16_t) __builtin_neon_vminvq_s16(__p0);
	57167	__ai float64x2_t vreinterpretq_f64_f32(float32x4_t __p0) {
	57168	float64x2_t __ret;
	57169	__ret = (float64x2_t)(__p0);
55215	57170	return __ret;
55216	57171	}
55217		#else
55218		__ai int16_t vminvq_s16(int16x8_t __p0) {
55219		int16_t __ret;
55220		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
55221		__ret = (int16_t) __builtin_neon_vminvq_s16(__rev0);
	57172	__ai float64x2_t vreinterpretq_f64_f16(float16x8_t __p0) {
	57173	float64x2_t __ret;
	57174	__ret = (float64x2_t)(__p0);
55222	57175	return __ret;
55223	57176	}
55224		#endif
55225
55226		#ifdef __LITTLE_ENDIAN__
55227		__ai uint8_t vminv_u8(uint8x8_t __p0) {
55228		uint8_t __ret;
55229		__ret = (uint8_t) __builtin_neon_vminv_u8(__p0);
	57177	__ai float64x2_t vreinterpretq_f64_s32(int32x4_t __p0) {
	57178	float64x2_t __ret;
	57179	__ret = (float64x2_t)(__p0);
55230	57180	return __ret;
55231	57181	}
55232		#else
55233		__ai uint8_t vminv_u8(uint8x8_t __p0) {
55234		uint8_t __ret;
55235		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
55236		__ret = (uint8_t) __builtin_neon_vminv_u8(__rev0);
	57182	__ai float64x2_t vreinterpretq_f64_s64(int64x2_t __p0) {
	57183	float64x2_t __ret;
	57184	__ret = (float64x2_t)(__p0);
55237	57185	return __ret;
55238	57186	}
55239		#endif
55240
55241		#ifdef __LITTLE_ENDIAN__
55242		__ai uint32_t vminv_u32(uint32x2_t __p0) {
55243		uint32_t __ret;
55244		__ret = (uint32_t) __builtin_neon_vminv_u32(__p0);
	57187	__ai float64x2_t vreinterpretq_f64_s16(int16x8_t __p0) {
	57188	float64x2_t __ret;
	57189	__ret = (float64x2_t)(__p0);
55245	57190	return __ret;
55246	57191	}
55247		#else
55248		__ai uint32_t vminv_u32(uint32x2_t __p0) {
55249		uint32_t __ret;
55250		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55251		__ret = (uint32_t) __builtin_neon_vminv_u32(__rev0);
	57192	__ai float32x4_t vreinterpretq_f32_p8(poly8x16_t __p0) {
	57193	float32x4_t __ret;
	57194	__ret = (float32x4_t)(__p0);
55252	57195	return __ret;
55253	57196	}
55254		#endif
55255
55256		#ifdef __LITTLE_ENDIAN__
55257		__ai uint16_t vminv_u16(uint16x4_t __p0) {
55258		uint16_t __ret;
55259		__ret = (uint16_t) __builtin_neon_vminv_u16(__p0);
	57197	__ai float32x4_t vreinterpretq_f32_p128(poly128_t __p0) {
	57198	float32x4_t __ret;
	57199	__ret = (float32x4_t)(__p0);
55260	57200	return __ret;
55261	57201	}
55262		#else
55263		__ai uint16_t vminv_u16(uint16x4_t __p0) {
55264		uint16_t __ret;
55265		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55266		__ret = (uint16_t) __builtin_neon_vminv_u16(__rev0);
	57202	__ai float32x4_t vreinterpretq_f32_p64(poly64x2_t __p0) {
	57203	float32x4_t __ret;
	57204	__ret = (float32x4_t)(__p0);
55267	57205	return __ret;
55268	57206	}
55269		#endif
55270
55271		#ifdef __LITTLE_ENDIAN__
55272		__ai int8_t vminv_s8(int8x8_t __p0) {
55273		int8_t __ret;
55274		__ret = (int8_t) __builtin_neon_vminv_s8(__p0);
	57207	__ai float32x4_t vreinterpretq_f32_p16(poly16x8_t __p0) {
	57208	float32x4_t __ret;
	57209	__ret = (float32x4_t)(__p0);
55275	57210	return __ret;
55276	57211	}
55277		#else
55278		__ai int8_t vminv_s8(int8x8_t __p0) {
55279		int8_t __ret;
55280		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
55281		__ret = (int8_t) __builtin_neon_vminv_s8(__rev0);
	57212	__ai float32x4_t vreinterpretq_f32_u8(uint8x16_t __p0) {
	57213	float32x4_t __ret;
	57214	__ret = (float32x4_t)(__p0);
	57215	return __ret;
	57216	}
	57217	__ai float32x4_t vreinterpretq_f32_u32(uint32x4_t __p0) {
	57218	float32x4_t __ret;
	57219	__ret = (float32x4_t)(__p0);
55282	57220	return __ret;
55283	57221	}
55284		#endif
55285
55286		#ifdef __LITTLE_ENDIAN__
55287		__ai float32_t vminv_f32(float32x2_t __p0) {
55288		float32_t __ret;
55289		__ret = (float32_t) __builtin_neon_vminv_f32(__p0);
	57222	__ai float32x4_t vreinterpretq_f32_u64(uint64x2_t __p0) {
	57223	float32x4_t __ret;
	57224	__ret = (float32x4_t)(__p0);
55290	57225	return __ret;
55291	57226	}
55292		#else
55293		__ai float32_t vminv_f32(float32x2_t __p0) {
55294		float32_t __ret;
55295		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55296		__ret = (float32_t) __builtin_neon_vminv_f32(__rev0);
	57227	__ai float32x4_t vreinterpretq_f32_u16(uint16x8_t __p0) {
	57228	float32x4_t __ret;
	57229	__ret = (float32x4_t)(__p0);
55297	57230	return __ret;
55298	57231	}
55299		#endif
55300
55301		#ifdef __LITTLE_ENDIAN__
55302		__ai int32_t vminv_s32(int32x2_t __p0) {
55303		int32_t __ret;
55304		__ret = (int32_t) __builtin_neon_vminv_s32(__p0);
	57232	__ai float32x4_t vreinterpretq_f32_s8(int8x16_t __p0) {
	57233	float32x4_t __ret;
	57234	__ret = (float32x4_t)(__p0);
55305	57235	return __ret;
55306	57236	}
55307		#else
55308		__ai int32_t vminv_s32(int32x2_t __p0) {
55309		int32_t __ret;
55310		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55311		__ret = (int32_t) __builtin_neon_vminv_s32(__rev0);
	57237	__ai float32x4_t vreinterpretq_f32_f64(float64x2_t __p0) {
	57238	float32x4_t __ret;
	57239	__ret = (float32x4_t)(__p0);
55312	57240	return __ret;
55313	57241	}
55314		#endif
55315
55316		#ifdef __LITTLE_ENDIAN__
55317		__ai int16_t vminv_s16(int16x4_t __p0) {
55318		int16_t __ret;
55319		__ret = (int16_t) __builtin_neon_vminv_s16(__p0);
	57242	__ai float32x4_t vreinterpretq_f32_f16(float16x8_t __p0) {
	57243	float32x4_t __ret;
	57244	__ret = (float32x4_t)(__p0);
55320	57245	return __ret;
55321	57246	}
55322		#else
55323		__ai int16_t vminv_s16(int16x4_t __p0) {
55324		int16_t __ret;
55325		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55326		__ret = (int16_t) __builtin_neon_vminv_s16(__rev0);
	57247	__ai float32x4_t vreinterpretq_f32_s32(int32x4_t __p0) {
	57248	float32x4_t __ret;
	57249	__ret = (float32x4_t)(__p0);
55327	57250	return __ret;
55328	57251	}
55329		#endif
55330
55331		#ifdef __LITTLE_ENDIAN__
55332		__ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
55333		float64x2_t __ret;
55334		__ret = __p0 + __p1 * __p2;
	57252	__ai float32x4_t vreinterpretq_f32_s64(int64x2_t __p0) {
	57253	float32x4_t __ret;
	57254	__ret = (float32x4_t)(__p0);
55335	57255	return __ret;
55336	57256	}
55337		#else
55338		__ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
55339		float64x2_t __ret;
55340		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55341		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55342		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
55343		__ret = __rev0 + __rev1 * __rev2;
55344		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57257	__ai float32x4_t vreinterpretq_f32_s16(int16x8_t __p0) {
	57258	float32x4_t __ret;
	57259	__ret = (float32x4_t)(__p0);
55345	57260	return __ret;
55346	57261	}
55347		#endif
55348
55349		__ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
55350		float64x1_t __ret;
55351		__ret = __p0 + __p1 * __p2;
	57262	__ai float16x8_t vreinterpretq_f16_p8(poly8x16_t __p0) {
	57263	float16x8_t __ret;
	57264	__ret = (float16x8_t)(__p0);
55352	57265	return __ret;
55353	57266	}
55354		#ifdef __LITTLE_ENDIAN__
55355		#define vmlaq_laneq_u32(__p0_531, __p1_531, __p2_531, __p3_531) __extension__ ({ \
55356		uint32x4_t __ret_531; \
55357		uint32x4_t __s0_531 = __p0_531; \
55358		uint32x4_t __s1_531 = __p1_531; \
55359		uint32x4_t __s2_531 = __p2_531; \
55360		__ret_531 = __s0_531 + __s1_531 * splatq_laneq_u32(__s2_531, __p3_531); \
55361		__ret_531; \
55362		})
55363		#else
55364		#define vmlaq_laneq_u32(__p0_532, __p1_532, __p2_532, __p3_532) __extension__ ({ \
55365		uint32x4_t __ret_532; \
55366		uint32x4_t __s0_532 = __p0_532; \
55367		uint32x4_t __s1_532 = __p1_532; \
55368		uint32x4_t __s2_532 = __p2_532; \
55369		uint32x4_t __rev0_532; __rev0_532 = __builtin_shufflevector(__s0_532, __s0_532, 3, 2, 1, 0); \
55370		uint32x4_t __rev1_532; __rev1_532 = __builtin_shufflevector(__s1_532, __s1_532, 3, 2, 1, 0); \
55371		uint32x4_t __rev2_532; __rev2_532 = __builtin_shufflevector(__s2_532, __s2_532, 3, 2, 1, 0); \
55372		__ret_532 = __rev0_532 + __rev1_532 * __noswap_splatq_laneq_u32(__rev2_532, __p3_532); \
55373		__ret_532 = __builtin_shufflevector(__ret_532, __ret_532, 3, 2, 1, 0); \
55374		__ret_532; \
55375		})
55376		#endif
55377
55378		#ifdef __LITTLE_ENDIAN__
55379		#define vmlaq_laneq_u16(__p0_533, __p1_533, __p2_533, __p3_533) __extension__ ({ \
55380		uint16x8_t __ret_533; \
55381		uint16x8_t __s0_533 = __p0_533; \
55382		uint16x8_t __s1_533 = __p1_533; \
55383		uint16x8_t __s2_533 = __p2_533; \
55384		__ret_533 = __s0_533 + __s1_533 * splatq_laneq_u16(__s2_533, __p3_533); \
55385		__ret_533; \
55386		})
55387		#else
55388		#define vmlaq_laneq_u16(__p0_534, __p1_534, __p2_534, __p3_534) __extension__ ({ \
55389		uint16x8_t __ret_534; \
55390		uint16x8_t __s0_534 = __p0_534; \
55391		uint16x8_t __s1_534 = __p1_534; \
55392		uint16x8_t __s2_534 = __p2_534; \
55393		uint16x8_t __rev0_534; __rev0_534 = __builtin_shufflevector(__s0_534, __s0_534, 7, 6, 5, 4, 3, 2, 1, 0); \
55394		uint16x8_t __rev1_534; __rev1_534 = __builtin_shufflevector(__s1_534, __s1_534, 7, 6, 5, 4, 3, 2, 1, 0); \
55395		uint16x8_t __rev2_534; __rev2_534 = __builtin_shufflevector(__s2_534, __s2_534, 7, 6, 5, 4, 3, 2, 1, 0); \
55396		__ret_534 = __rev0_534 + __rev1_534 * __noswap_splatq_laneq_u16(__rev2_534, __p3_534); \
55397		__ret_534 = __builtin_shufflevector(__ret_534, __ret_534, 7, 6, 5, 4, 3, 2, 1, 0); \
55398		__ret_534; \
55399		})
55400		#endif
55401
55402		#ifdef __LITTLE_ENDIAN__
55403		#define vmlaq_laneq_f32(__p0_535, __p1_535, __p2_535, __p3_535) __extension__ ({ \
55404		float32x4_t __ret_535; \
55405		float32x4_t __s0_535 = __p0_535; \
55406		float32x4_t __s1_535 = __p1_535; \
55407		float32x4_t __s2_535 = __p2_535; \
55408		__ret_535 = __s0_535 + __s1_535 * splatq_laneq_f32(__s2_535, __p3_535); \
55409		__ret_535; \
55410		})
55411		#else
55412		#define vmlaq_laneq_f32(__p0_536, __p1_536, __p2_536, __p3_536) __extension__ ({ \
55413		float32x4_t __ret_536; \
55414		float32x4_t __s0_536 = __p0_536; \
55415		float32x4_t __s1_536 = __p1_536; \
55416		float32x4_t __s2_536 = __p2_536; \
55417		float32x4_t __rev0_536; __rev0_536 = __builtin_shufflevector(__s0_536, __s0_536, 3, 2, 1, 0); \
55418		float32x4_t __rev1_536; __rev1_536 = __builtin_shufflevector(__s1_536, __s1_536, 3, 2, 1, 0); \
55419		float32x4_t __rev2_536; __rev2_536 = __builtin_shufflevector(__s2_536, __s2_536, 3, 2, 1, 0); \
55420		__ret_536 = __rev0_536 + __rev1_536 * __noswap_splatq_laneq_f32(__rev2_536, __p3_536); \
55421		__ret_536 = __builtin_shufflevector(__ret_536, __ret_536, 3, 2, 1, 0); \
55422		__ret_536; \
55423		})
55424		#endif
55425
55426		#ifdef __LITTLE_ENDIAN__
55427		#define vmlaq_laneq_s32(__p0_537, __p1_537, __p2_537, __p3_537) __extension__ ({ \
55428		int32x4_t __ret_537; \
55429		int32x4_t __s0_537 = __p0_537; \
55430		int32x4_t __s1_537 = __p1_537; \
55431		int32x4_t __s2_537 = __p2_537; \
55432		__ret_537 = __s0_537 + __s1_537 * splatq_laneq_s32(__s2_537, __p3_537); \
55433		__ret_537; \
55434		})
55435		#else
55436		#define vmlaq_laneq_s32(__p0_538, __p1_538, __p2_538, __p3_538) __extension__ ({ \
55437		int32x4_t __ret_538; \
55438		int32x4_t __s0_538 = __p0_538; \
55439		int32x4_t __s1_538 = __p1_538; \
55440		int32x4_t __s2_538 = __p2_538; \
55441		int32x4_t __rev0_538; __rev0_538 = __builtin_shufflevector(__s0_538, __s0_538, 3, 2, 1, 0); \
55442		int32x4_t __rev1_538; __rev1_538 = __builtin_shufflevector(__s1_538, __s1_538, 3, 2, 1, 0); \
55443		int32x4_t __rev2_538; __rev2_538 = __builtin_shufflevector(__s2_538, __s2_538, 3, 2, 1, 0); \
55444		__ret_538 = __rev0_538 + __rev1_538 * __noswap_splatq_laneq_s32(__rev2_538, __p3_538); \
55445		__ret_538 = __builtin_shufflevector(__ret_538, __ret_538, 3, 2, 1, 0); \
55446		__ret_538; \
55447		})
55448		#endif
55449
55450		#ifdef __LITTLE_ENDIAN__
55451		#define vmlaq_laneq_s16(__p0_539, __p1_539, __p2_539, __p3_539) __extension__ ({ \
55452		int16x8_t __ret_539; \
55453		int16x8_t __s0_539 = __p0_539; \
55454		int16x8_t __s1_539 = __p1_539; \
55455		int16x8_t __s2_539 = __p2_539; \
55456		__ret_539 = __s0_539 + __s1_539 * splatq_laneq_s16(__s2_539, __p3_539); \
55457		__ret_539; \
55458		})
55459		#else
55460		#define vmlaq_laneq_s16(__p0_540, __p1_540, __p2_540, __p3_540) __extension__ ({ \
55461		int16x8_t __ret_540; \
55462		int16x8_t __s0_540 = __p0_540; \
55463		int16x8_t __s1_540 = __p1_540; \
55464		int16x8_t __s2_540 = __p2_540; \
55465		int16x8_t __rev0_540; __rev0_540 = __builtin_shufflevector(__s0_540, __s0_540, 7, 6, 5, 4, 3, 2, 1, 0); \
55466		int16x8_t __rev1_540; __rev1_540 = __builtin_shufflevector(__s1_540, __s1_540, 7, 6, 5, 4, 3, 2, 1, 0); \
55467		int16x8_t __rev2_540; __rev2_540 = __builtin_shufflevector(__s2_540, __s2_540, 7, 6, 5, 4, 3, 2, 1, 0); \
55468		__ret_540 = __rev0_540 + __rev1_540 * __noswap_splatq_laneq_s16(__rev2_540, __p3_540); \
55469		__ret_540 = __builtin_shufflevector(__ret_540, __ret_540, 7, 6, 5, 4, 3, 2, 1, 0); \
55470		__ret_540; \
55471		})
55472		#endif
55473
55474		#ifdef __LITTLE_ENDIAN__
55475		#define vmla_laneq_u32(__p0_541, __p1_541, __p2_541, __p3_541) __extension__ ({ \
55476		uint32x2_t __ret_541; \
55477		uint32x2_t __s0_541 = __p0_541; \
55478		uint32x2_t __s1_541 = __p1_541; \
55479		uint32x4_t __s2_541 = __p2_541; \
55480		__ret_541 = __s0_541 + __s1_541 * splat_laneq_u32(__s2_541, __p3_541); \
55481		__ret_541; \
55482		})
55483		#else
55484		#define vmla_laneq_u32(__p0_542, __p1_542, __p2_542, __p3_542) __extension__ ({ \
55485		uint32x2_t __ret_542; \
55486		uint32x2_t __s0_542 = __p0_542; \
55487		uint32x2_t __s1_542 = __p1_542; \
55488		uint32x4_t __s2_542 = __p2_542; \
55489		uint32x2_t __rev0_542; __rev0_542 = __builtin_shufflevector(__s0_542, __s0_542, 1, 0); \
55490		uint32x2_t __rev1_542; __rev1_542 = __builtin_shufflevector(__s1_542, __s1_542, 1, 0); \
55491		uint32x4_t __rev2_542; __rev2_542 = __builtin_shufflevector(__s2_542, __s2_542, 3, 2, 1, 0); \
55492		__ret_542 = __rev0_542 + __rev1_542 * __noswap_splat_laneq_u32(__rev2_542, __p3_542); \
55493		__ret_542 = __builtin_shufflevector(__ret_542, __ret_542, 1, 0); \
55494		__ret_542; \
55495		})
55496		#endif
55497
55498		#ifdef __LITTLE_ENDIAN__
55499		#define vmla_laneq_u16(__p0_543, __p1_543, __p2_543, __p3_543) __extension__ ({ \
55500		uint16x4_t __ret_543; \
55501		uint16x4_t __s0_543 = __p0_543; \
55502		uint16x4_t __s1_543 = __p1_543; \
55503		uint16x8_t __s2_543 = __p2_543; \
55504		__ret_543 = __s0_543 + __s1_543 * splat_laneq_u16(__s2_543, __p3_543); \
55505		__ret_543; \
55506		})
55507		#else
55508		#define vmla_laneq_u16(__p0_544, __p1_544, __p2_544, __p3_544) __extension__ ({ \
55509		uint16x4_t __ret_544; \
55510		uint16x4_t __s0_544 = __p0_544; \
55511		uint16x4_t __s1_544 = __p1_544; \
55512		uint16x8_t __s2_544 = __p2_544; \
55513		uint16x4_t __rev0_544; __rev0_544 = __builtin_shufflevector(__s0_544, __s0_544, 3, 2, 1, 0); \
55514		uint16x4_t __rev1_544; __rev1_544 = __builtin_shufflevector(__s1_544, __s1_544, 3, 2, 1, 0); \
55515		uint16x8_t __rev2_544; __rev2_544 = __builtin_shufflevector(__s2_544, __s2_544, 7, 6, 5, 4, 3, 2, 1, 0); \
55516		__ret_544 = __rev0_544 + __rev1_544 * __noswap_splat_laneq_u16(__rev2_544, __p3_544); \
55517		__ret_544 = __builtin_shufflevector(__ret_544, __ret_544, 3, 2, 1, 0); \
55518		__ret_544; \
55519		})
55520		#endif
55521
55522		#ifdef __LITTLE_ENDIAN__
55523		#define vmla_laneq_f32(__p0_545, __p1_545, __p2_545, __p3_545) __extension__ ({ \
55524		float32x2_t __ret_545; \
55525		float32x2_t __s0_545 = __p0_545; \
55526		float32x2_t __s1_545 = __p1_545; \
55527		float32x4_t __s2_545 = __p2_545; \
55528		__ret_545 = __s0_545 + __s1_545 * splat_laneq_f32(__s2_545, __p3_545); \
55529		__ret_545; \
55530		})
55531		#else
55532		#define vmla_laneq_f32(__p0_546, __p1_546, __p2_546, __p3_546) __extension__ ({ \
55533		float32x2_t __ret_546; \
55534		float32x2_t __s0_546 = __p0_546; \
55535		float32x2_t __s1_546 = __p1_546; \
55536		float32x4_t __s2_546 = __p2_546; \
55537		float32x2_t __rev0_546; __rev0_546 = __builtin_shufflevector(__s0_546, __s0_546, 1, 0); \
55538		float32x2_t __rev1_546; __rev1_546 = __builtin_shufflevector(__s1_546, __s1_546, 1, 0); \
55539		float32x4_t __rev2_546; __rev2_546 = __builtin_shufflevector(__s2_546, __s2_546, 3, 2, 1, 0); \
55540		__ret_546 = __rev0_546 + __rev1_546 * __noswap_splat_laneq_f32(__rev2_546, __p3_546); \
55541		__ret_546 = __builtin_shufflevector(__ret_546, __ret_546, 1, 0); \
55542		__ret_546; \
55543		})
55544		#endif
55545
55546		#ifdef __LITTLE_ENDIAN__
55547		#define vmla_laneq_s32(__p0_547, __p1_547, __p2_547, __p3_547) __extension__ ({ \
55548		int32x2_t __ret_547; \
55549		int32x2_t __s0_547 = __p0_547; \
55550		int32x2_t __s1_547 = __p1_547; \
55551		int32x4_t __s2_547 = __p2_547; \
55552		__ret_547 = __s0_547 + __s1_547 * splat_laneq_s32(__s2_547, __p3_547); \
55553		__ret_547; \
55554		})
55555		#else
55556		#define vmla_laneq_s32(__p0_548, __p1_548, __p2_548, __p3_548) __extension__ ({ \
55557		int32x2_t __ret_548; \
55558		int32x2_t __s0_548 = __p0_548; \
55559		int32x2_t __s1_548 = __p1_548; \
55560		int32x4_t __s2_548 = __p2_548; \
55561		int32x2_t __rev0_548; __rev0_548 = __builtin_shufflevector(__s0_548, __s0_548, 1, 0); \
55562		int32x2_t __rev1_548; __rev1_548 = __builtin_shufflevector(__s1_548, __s1_548, 1, 0); \
55563		int32x4_t __rev2_548; __rev2_548 = __builtin_shufflevector(__s2_548, __s2_548, 3, 2, 1, 0); \
55564		__ret_548 = __rev0_548 + __rev1_548 * __noswap_splat_laneq_s32(__rev2_548, __p3_548); \
55565		__ret_548 = __builtin_shufflevector(__ret_548, __ret_548, 1, 0); \
55566		__ret_548; \
55567		})
55568		#endif
55569
55570		#ifdef __LITTLE_ENDIAN__
55571		#define vmla_laneq_s16(__p0_549, __p1_549, __p2_549, __p3_549) __extension__ ({ \
55572		int16x4_t __ret_549; \
55573		int16x4_t __s0_549 = __p0_549; \
55574		int16x4_t __s1_549 = __p1_549; \
55575		int16x8_t __s2_549 = __p2_549; \
55576		__ret_549 = __s0_549 + __s1_549 * splat_laneq_s16(__s2_549, __p3_549); \
55577		__ret_549; \
55578		})
55579		#else
55580		#define vmla_laneq_s16(__p0_550, __p1_550, __p2_550, __p3_550) __extension__ ({ \
55581		int16x4_t __ret_550; \
55582		int16x4_t __s0_550 = __p0_550; \
55583		int16x4_t __s1_550 = __p1_550; \
55584		int16x8_t __s2_550 = __p2_550; \
55585		int16x4_t __rev0_550; __rev0_550 = __builtin_shufflevector(__s0_550, __s0_550, 3, 2, 1, 0); \
55586		int16x4_t __rev1_550; __rev1_550 = __builtin_shufflevector(__s1_550, __s1_550, 3, 2, 1, 0); \
55587		int16x8_t __rev2_550; __rev2_550 = __builtin_shufflevector(__s2_550, __s2_550, 7, 6, 5, 4, 3, 2, 1, 0); \
55588		__ret_550 = __rev0_550 + __rev1_550 * __noswap_splat_laneq_s16(__rev2_550, __p3_550); \
55589		__ret_550 = __builtin_shufflevector(__ret_550, __ret_550, 3, 2, 1, 0); \
55590		__ret_550; \
55591		})
55592		#endif
55593
55594		#ifdef __LITTLE_ENDIAN__
55595		#define vmlal_high_lane_u32(__p0_551, __p1_551, __p2_551, __p3_551) __extension__ ({ \
55596		uint64x2_t __ret_551; \
55597		uint64x2_t __s0_551 = __p0_551; \
55598		uint32x4_t __s1_551 = __p1_551; \
55599		uint32x2_t __s2_551 = __p2_551; \
55600		__ret_551 = __s0_551 + vmull_u32(vget_high_u32(__s1_551), splat_lane_u32(__s2_551, __p3_551)); \
55601		__ret_551; \
55602		})
55603		#else
55604		#define vmlal_high_lane_u32(__p0_552, __p1_552, __p2_552, __p3_552) __extension__ ({ \
55605		uint64x2_t __ret_552; \
55606		uint64x2_t __s0_552 = __p0_552; \
55607		uint32x4_t __s1_552 = __p1_552; \
55608		uint32x2_t __s2_552 = __p2_552; \
55609		uint64x2_t __rev0_552; __rev0_552 = __builtin_shufflevector(__s0_552, __s0_552, 1, 0); \
55610		uint32x4_t __rev1_552; __rev1_552 = __builtin_shufflevector(__s1_552, __s1_552, 3, 2, 1, 0); \
55611		uint32x2_t __rev2_552; __rev2_552 = __builtin_shufflevector(__s2_552, __s2_552, 1, 0); \
55612		__ret_552 = __rev0_552 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_552), __noswap_splat_lane_u32(__rev2_552, __p3_552)); \
55613		__ret_552 = __builtin_shufflevector(__ret_552, __ret_552, 1, 0); \
55614		__ret_552; \
55615		})
55616		#endif
55617
55618		#ifdef __LITTLE_ENDIAN__
55619		#define vmlal_high_lane_u16(__p0_553, __p1_553, __p2_553, __p3_553) __extension__ ({ \
55620		uint32x4_t __ret_553; \
55621		uint32x4_t __s0_553 = __p0_553; \
55622		uint16x8_t __s1_553 = __p1_553; \
55623		uint16x4_t __s2_553 = __p2_553; \
55624		__ret_553 = __s0_553 + vmull_u16(vget_high_u16(__s1_553), splat_lane_u16(__s2_553, __p3_553)); \
55625		__ret_553; \
55626		})
55627		#else
55628		#define vmlal_high_lane_u16(__p0_554, __p1_554, __p2_554, __p3_554) __extension__ ({ \
55629		uint32x4_t __ret_554; \
55630		uint32x4_t __s0_554 = __p0_554; \
55631		uint16x8_t __s1_554 = __p1_554; \
55632		uint16x4_t __s2_554 = __p2_554; \
55633		uint32x4_t __rev0_554; __rev0_554 = __builtin_shufflevector(__s0_554, __s0_554, 3, 2, 1, 0); \
55634		uint16x8_t __rev1_554; __rev1_554 = __builtin_shufflevector(__s1_554, __s1_554, 7, 6, 5, 4, 3, 2, 1, 0); \
55635		uint16x4_t __rev2_554; __rev2_554 = __builtin_shufflevector(__s2_554, __s2_554, 3, 2, 1, 0); \
55636		__ret_554 = __rev0_554 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_554), __noswap_splat_lane_u16(__rev2_554, __p3_554)); \
55637		__ret_554 = __builtin_shufflevector(__ret_554, __ret_554, 3, 2, 1, 0); \
55638		__ret_554; \
55639		})
55640		#endif
55641
55642		#ifdef __LITTLE_ENDIAN__
55643		#define vmlal_high_lane_s32(__p0_555, __p1_555, __p2_555, __p3_555) __extension__ ({ \
55644		int64x2_t __ret_555; \
55645		int64x2_t __s0_555 = __p0_555; \
55646		int32x4_t __s1_555 = __p1_555; \
55647		int32x2_t __s2_555 = __p2_555; \
55648		__ret_555 = __s0_555 + vmull_s32(vget_high_s32(__s1_555), splat_lane_s32(__s2_555, __p3_555)); \
55649		__ret_555; \
55650		})
55651		#else
55652		#define vmlal_high_lane_s32(__p0_556, __p1_556, __p2_556, __p3_556) __extension__ ({ \
55653		int64x2_t __ret_556; \
55654		int64x2_t __s0_556 = __p0_556; \
55655		int32x4_t __s1_556 = __p1_556; \
55656		int32x2_t __s2_556 = __p2_556; \
55657		int64x2_t __rev0_556; __rev0_556 = __builtin_shufflevector(__s0_556, __s0_556, 1, 0); \
55658		int32x4_t __rev1_556; __rev1_556 = __builtin_shufflevector(__s1_556, __s1_556, 3, 2, 1, 0); \
55659		int32x2_t __rev2_556; __rev2_556 = __builtin_shufflevector(__s2_556, __s2_556, 1, 0); \
55660		__ret_556 = __rev0_556 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_556), __noswap_splat_lane_s32(__rev2_556, __p3_556)); \
55661		__ret_556 = __builtin_shufflevector(__ret_556, __ret_556, 1, 0); \
55662		__ret_556; \
55663		})
55664		#endif
55665
55666		#ifdef __LITTLE_ENDIAN__
55667		#define vmlal_high_lane_s16(__p0_557, __p1_557, __p2_557, __p3_557) __extension__ ({ \
55668		int32x4_t __ret_557; \
55669		int32x4_t __s0_557 = __p0_557; \
55670		int16x8_t __s1_557 = __p1_557; \
55671		int16x4_t __s2_557 = __p2_557; \
55672		__ret_557 = __s0_557 + vmull_s16(vget_high_s16(__s1_557), splat_lane_s16(__s2_557, __p3_557)); \
55673		__ret_557; \
55674		})
55675		#else
55676		#define vmlal_high_lane_s16(__p0_558, __p1_558, __p2_558, __p3_558) __extension__ ({ \
55677		int32x4_t __ret_558; \
55678		int32x4_t __s0_558 = __p0_558; \
55679		int16x8_t __s1_558 = __p1_558; \
55680		int16x4_t __s2_558 = __p2_558; \
55681		int32x4_t __rev0_558; __rev0_558 = __builtin_shufflevector(__s0_558, __s0_558, 3, 2, 1, 0); \
55682		int16x8_t __rev1_558; __rev1_558 = __builtin_shufflevector(__s1_558, __s1_558, 7, 6, 5, 4, 3, 2, 1, 0); \
55683		int16x4_t __rev2_558; __rev2_558 = __builtin_shufflevector(__s2_558, __s2_558, 3, 2, 1, 0); \
55684		__ret_558 = __rev0_558 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_558), __noswap_splat_lane_s16(__rev2_558, __p3_558)); \
55685		__ret_558 = __builtin_shufflevector(__ret_558, __ret_558, 3, 2, 1, 0); \
55686		__ret_558; \
55687		})
55688		#endif
55689
55690		#ifdef __LITTLE_ENDIAN__
55691		#define vmlal_high_laneq_u32(__p0_559, __p1_559, __p2_559, __p3_559) __extension__ ({ \
55692		uint64x2_t __ret_559; \
55693		uint64x2_t __s0_559 = __p0_559; \
55694		uint32x4_t __s1_559 = __p1_559; \
55695		uint32x4_t __s2_559 = __p2_559; \
55696		__ret_559 = __s0_559 + vmull_u32(vget_high_u32(__s1_559), splat_laneq_u32(__s2_559, __p3_559)); \
55697		__ret_559; \
55698		})
55699		#else
55700		#define vmlal_high_laneq_u32(__p0_560, __p1_560, __p2_560, __p3_560) __extension__ ({ \
55701		uint64x2_t __ret_560; \
55702		uint64x2_t __s0_560 = __p0_560; \
55703		uint32x4_t __s1_560 = __p1_560; \
55704		uint32x4_t __s2_560 = __p2_560; \
55705		uint64x2_t __rev0_560; __rev0_560 = __builtin_shufflevector(__s0_560, __s0_560, 1, 0); \
55706		uint32x4_t __rev1_560; __rev1_560 = __builtin_shufflevector(__s1_560, __s1_560, 3, 2, 1, 0); \
55707		uint32x4_t __rev2_560; __rev2_560 = __builtin_shufflevector(__s2_560, __s2_560, 3, 2, 1, 0); \
55708		__ret_560 = __rev0_560 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_560), __noswap_splat_laneq_u32(__rev2_560, __p3_560)); \
55709		__ret_560 = __builtin_shufflevector(__ret_560, __ret_560, 1, 0); \
55710		__ret_560; \
55711		})
55712		#endif
55713
55714		#ifdef __LITTLE_ENDIAN__
55715		#define vmlal_high_laneq_u16(__p0_561, __p1_561, __p2_561, __p3_561) __extension__ ({ \
55716		uint32x4_t __ret_561; \
55717		uint32x4_t __s0_561 = __p0_561; \
55718		uint16x8_t __s1_561 = __p1_561; \
55719		uint16x8_t __s2_561 = __p2_561; \
55720		__ret_561 = __s0_561 + vmull_u16(vget_high_u16(__s1_561), splat_laneq_u16(__s2_561, __p3_561)); \
55721		__ret_561; \
55722		})
55723		#else
55724		#define vmlal_high_laneq_u16(__p0_562, __p1_562, __p2_562, __p3_562) __extension__ ({ \
55725		uint32x4_t __ret_562; \
55726		uint32x4_t __s0_562 = __p0_562; \
55727		uint16x8_t __s1_562 = __p1_562; \
55728		uint16x8_t __s2_562 = __p2_562; \
55729		uint32x4_t __rev0_562; __rev0_562 = __builtin_shufflevector(__s0_562, __s0_562, 3, 2, 1, 0); \
55730		uint16x8_t __rev1_562; __rev1_562 = __builtin_shufflevector(__s1_562, __s1_562, 7, 6, 5, 4, 3, 2, 1, 0); \
55731		uint16x8_t __rev2_562; __rev2_562 = __builtin_shufflevector(__s2_562, __s2_562, 7, 6, 5, 4, 3, 2, 1, 0); \
55732		__ret_562 = __rev0_562 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_562), __noswap_splat_laneq_u16(__rev2_562, __p3_562)); \
55733		__ret_562 = __builtin_shufflevector(__ret_562, __ret_562, 3, 2, 1, 0); \
55734		__ret_562; \
55735		})
55736		#endif
55737
55738		#ifdef __LITTLE_ENDIAN__
55739		#define vmlal_high_laneq_s32(__p0_563, __p1_563, __p2_563, __p3_563) __extension__ ({ \
55740		int64x2_t __ret_563; \
55741		int64x2_t __s0_563 = __p0_563; \
55742		int32x4_t __s1_563 = __p1_563; \
55743		int32x4_t __s2_563 = __p2_563; \
55744		__ret_563 = __s0_563 + vmull_s32(vget_high_s32(__s1_563), splat_laneq_s32(__s2_563, __p3_563)); \
55745		__ret_563; \
55746		})
55747		#else
55748		#define vmlal_high_laneq_s32(__p0_564, __p1_564, __p2_564, __p3_564) __extension__ ({ \
55749		int64x2_t __ret_564; \
55750		int64x2_t __s0_564 = __p0_564; \
55751		int32x4_t __s1_564 = __p1_564; \
55752		int32x4_t __s2_564 = __p2_564; \
55753		int64x2_t __rev0_564; __rev0_564 = __builtin_shufflevector(__s0_564, __s0_564, 1, 0); \
55754		int32x4_t __rev1_564; __rev1_564 = __builtin_shufflevector(__s1_564, __s1_564, 3, 2, 1, 0); \
55755		int32x4_t __rev2_564; __rev2_564 = __builtin_shufflevector(__s2_564, __s2_564, 3, 2, 1, 0); \
55756		__ret_564 = __rev0_564 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_564), __noswap_splat_laneq_s32(__rev2_564, __p3_564)); \
55757		__ret_564 = __builtin_shufflevector(__ret_564, __ret_564, 1, 0); \
55758		__ret_564; \
55759		})
55760		#endif
55761
55762		#ifdef __LITTLE_ENDIAN__
55763		#define vmlal_high_laneq_s16(__p0_565, __p1_565, __p2_565, __p3_565) __extension__ ({ \
55764		int32x4_t __ret_565; \
55765		int32x4_t __s0_565 = __p0_565; \
55766		int16x8_t __s1_565 = __p1_565; \
55767		int16x8_t __s2_565 = __p2_565; \
55768		__ret_565 = __s0_565 + vmull_s16(vget_high_s16(__s1_565), splat_laneq_s16(__s2_565, __p3_565)); \
55769		__ret_565; \
55770		})
55771		#else
55772		#define vmlal_high_laneq_s16(__p0_566, __p1_566, __p2_566, __p3_566) __extension__ ({ \
55773		int32x4_t __ret_566; \
55774		int32x4_t __s0_566 = __p0_566; \
55775		int16x8_t __s1_566 = __p1_566; \
55776		int16x8_t __s2_566 = __p2_566; \
55777		int32x4_t __rev0_566; __rev0_566 = __builtin_shufflevector(__s0_566, __s0_566, 3, 2, 1, 0); \
55778		int16x8_t __rev1_566; __rev1_566 = __builtin_shufflevector(__s1_566, __s1_566, 7, 6, 5, 4, 3, 2, 1, 0); \
55779		int16x8_t __rev2_566; __rev2_566 = __builtin_shufflevector(__s2_566, __s2_566, 7, 6, 5, 4, 3, 2, 1, 0); \
55780		__ret_566 = __rev0_566 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_566), __noswap_splat_laneq_s16(__rev2_566, __p3_566)); \
55781		__ret_566 = __builtin_shufflevector(__ret_566, __ret_566, 3, 2, 1, 0); \
55782		__ret_566; \
55783		})
55784		#endif
55785
55786		#ifdef __LITTLE_ENDIAN__
55787		#define vmlal_laneq_u32(__p0_567, __p1_567, __p2_567, __p3_567) __extension__ ({ \
55788		uint64x2_t __ret_567; \
55789		uint64x2_t __s0_567 = __p0_567; \
55790		uint32x2_t __s1_567 = __p1_567; \
55791		uint32x4_t __s2_567 = __p2_567; \
55792		__ret_567 = __s0_567 + vmull_u32(__s1_567, splat_laneq_u32(__s2_567, __p3_567)); \
55793		__ret_567; \
55794		})
55795		#else
55796		#define vmlal_laneq_u32(__p0_568, __p1_568, __p2_568, __p3_568) __extension__ ({ \
55797		uint64x2_t __ret_568; \
55798		uint64x2_t __s0_568 = __p0_568; \
55799		uint32x2_t __s1_568 = __p1_568; \
55800		uint32x4_t __s2_568 = __p2_568; \
55801		uint64x2_t __rev0_568; __rev0_568 = __builtin_shufflevector(__s0_568, __s0_568, 1, 0); \
55802		uint32x2_t __rev1_568; __rev1_568 = __builtin_shufflevector(__s1_568, __s1_568, 1, 0); \
55803		uint32x4_t __rev2_568; __rev2_568 = __builtin_shufflevector(__s2_568, __s2_568, 3, 2, 1, 0); \
55804		__ret_568 = __rev0_568 + __noswap_vmull_u32(__rev1_568, __noswap_splat_laneq_u32(__rev2_568, __p3_568)); \
55805		__ret_568 = __builtin_shufflevector(__ret_568, __ret_568, 1, 0); \
55806		__ret_568; \
55807		})
55808		#endif
55809
55810		#ifdef __LITTLE_ENDIAN__
55811		#define vmlal_laneq_u16(__p0_569, __p1_569, __p2_569, __p3_569) __extension__ ({ \
55812		uint32x4_t __ret_569; \
55813		uint32x4_t __s0_569 = __p0_569; \
55814		uint16x4_t __s1_569 = __p1_569; \
55815		uint16x8_t __s2_569 = __p2_569; \
55816		__ret_569 = __s0_569 + vmull_u16(__s1_569, splat_laneq_u16(__s2_569, __p3_569)); \
55817		__ret_569; \
55818		})
55819		#else
55820		#define vmlal_laneq_u16(__p0_570, __p1_570, __p2_570, __p3_570) __extension__ ({ \
55821		uint32x4_t __ret_570; \
55822		uint32x4_t __s0_570 = __p0_570; \
55823		uint16x4_t __s1_570 = __p1_570; \
55824		uint16x8_t __s2_570 = __p2_570; \
55825		uint32x4_t __rev0_570; __rev0_570 = __builtin_shufflevector(__s0_570, __s0_570, 3, 2, 1, 0); \
55826		uint16x4_t __rev1_570; __rev1_570 = __builtin_shufflevector(__s1_570, __s1_570, 3, 2, 1, 0); \
55827		uint16x8_t __rev2_570; __rev2_570 = __builtin_shufflevector(__s2_570, __s2_570, 7, 6, 5, 4, 3, 2, 1, 0); \
55828		__ret_570 = __rev0_570 + __noswap_vmull_u16(__rev1_570, __noswap_splat_laneq_u16(__rev2_570, __p3_570)); \
55829		__ret_570 = __builtin_shufflevector(__ret_570, __ret_570, 3, 2, 1, 0); \
55830		__ret_570; \
55831		})
55832		#endif
55833
55834		#ifdef __LITTLE_ENDIAN__
55835		#define vmlal_laneq_s32(__p0_571, __p1_571, __p2_571, __p3_571) __extension__ ({ \
55836		int64x2_t __ret_571; \
55837		int64x2_t __s0_571 = __p0_571; \
55838		int32x2_t __s1_571 = __p1_571; \
55839		int32x4_t __s2_571 = __p2_571; \
55840		__ret_571 = __s0_571 + vmull_s32(__s1_571, splat_laneq_s32(__s2_571, __p3_571)); \
55841		__ret_571; \
55842		})
55843		#else
55844		#define vmlal_laneq_s32(__p0_572, __p1_572, __p2_572, __p3_572) __extension__ ({ \
55845		int64x2_t __ret_572; \
55846		int64x2_t __s0_572 = __p0_572; \
55847		int32x2_t __s1_572 = __p1_572; \
55848		int32x4_t __s2_572 = __p2_572; \
55849		int64x2_t __rev0_572; __rev0_572 = __builtin_shufflevector(__s0_572, __s0_572, 1, 0); \
55850		int32x2_t __rev1_572; __rev1_572 = __builtin_shufflevector(__s1_572, __s1_572, 1, 0); \
55851		int32x4_t __rev2_572; __rev2_572 = __builtin_shufflevector(__s2_572, __s2_572, 3, 2, 1, 0); \
55852		__ret_572 = __rev0_572 + __noswap_vmull_s32(__rev1_572, __noswap_splat_laneq_s32(__rev2_572, __p3_572)); \
55853		__ret_572 = __builtin_shufflevector(__ret_572, __ret_572, 1, 0); \
55854		__ret_572; \
55855		})
55856		#endif
55857
55858		#ifdef __LITTLE_ENDIAN__
55859		#define vmlal_laneq_s16(__p0_573, __p1_573, __p2_573, __p3_573) __extension__ ({ \
55860		int32x4_t __ret_573; \
55861		int32x4_t __s0_573 = __p0_573; \
55862		int16x4_t __s1_573 = __p1_573; \
55863		int16x8_t __s2_573 = __p2_573; \
55864		__ret_573 = __s0_573 + vmull_s16(__s1_573, splat_laneq_s16(__s2_573, __p3_573)); \
55865		__ret_573; \
55866		})
55867		#else
55868		#define vmlal_laneq_s16(__p0_574, __p1_574, __p2_574, __p3_574) __extension__ ({ \
55869		int32x4_t __ret_574; \
55870		int32x4_t __s0_574 = __p0_574; \
55871		int16x4_t __s1_574 = __p1_574; \
55872		int16x8_t __s2_574 = __p2_574; \
55873		int32x4_t __rev0_574; __rev0_574 = __builtin_shufflevector(__s0_574, __s0_574, 3, 2, 1, 0); \
55874		int16x4_t __rev1_574; __rev1_574 = __builtin_shufflevector(__s1_574, __s1_574, 3, 2, 1, 0); \
55875		int16x8_t __rev2_574; __rev2_574 = __builtin_shufflevector(__s2_574, __s2_574, 7, 6, 5, 4, 3, 2, 1, 0); \
55876		__ret_574 = __rev0_574 + __noswap_vmull_s16(__rev1_574, __noswap_splat_laneq_s16(__rev2_574, __p3_574)); \
55877		__ret_574 = __builtin_shufflevector(__ret_574, __ret_574, 3, 2, 1, 0); \
55878		__ret_574; \
55879		})
55880		#endif
55881
55882		#ifdef __LITTLE_ENDIAN__
55883		__ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
55884		float64x2_t __ret;
55885		__ret = __p0 - __p1 * __p2;
	57267	__ai float16x8_t vreinterpretq_f16_p128(poly128_t __p0) {
	57268	float16x8_t __ret;
	57269	__ret = (float16x8_t)(__p0);
55886	57270	return __ret;
55887	57271	}
55888		#else
55889		__ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
55890		float64x2_t __ret;
55891		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55892		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55893		float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
55894		__ret = __rev0 - __rev1 * __rev2;
55895		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57272	__ai float16x8_t vreinterpretq_f16_p64(poly64x2_t __p0) {
	57273	float16x8_t __ret;
	57274	__ret = (float16x8_t)(__p0);
55896	57275	return __ret;
55897	57276	}
55898		#endif
55899
55900		__ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
55901		float64x1_t __ret;
55902		__ret = __p0 - __p1 * __p2;
	57277	__ai float16x8_t vreinterpretq_f16_p16(poly16x8_t __p0) {
	57278	float16x8_t __ret;
	57279	__ret = (float16x8_t)(__p0);
55903	57280	return __ret;
55904	57281	}
55905		#ifdef __LITTLE_ENDIAN__
55906		#define vmlsq_laneq_u32(__p0_575, __p1_575, __p2_575, __p3_575) __extension__ ({ \
55907		uint32x4_t __ret_575; \
55908		uint32x4_t __s0_575 = __p0_575; \
55909		uint32x4_t __s1_575 = __p1_575; \
55910		uint32x4_t __s2_575 = __p2_575; \
55911		__ret_575 = __s0_575 - __s1_575 * splatq_laneq_u32(__s2_575, __p3_575); \
55912		__ret_575; \
55913		})
55914		#else
55915		#define vmlsq_laneq_u32(__p0_576, __p1_576, __p2_576, __p3_576) __extension__ ({ \
55916		uint32x4_t __ret_576; \
55917		uint32x4_t __s0_576 = __p0_576; \
55918		uint32x4_t __s1_576 = __p1_576; \
55919		uint32x4_t __s2_576 = __p2_576; \
55920		uint32x4_t __rev0_576; __rev0_576 = __builtin_shufflevector(__s0_576, __s0_576, 3, 2, 1, 0); \
55921		uint32x4_t __rev1_576; __rev1_576 = __builtin_shufflevector(__s1_576, __s1_576, 3, 2, 1, 0); \
55922		uint32x4_t __rev2_576; __rev2_576 = __builtin_shufflevector(__s2_576, __s2_576, 3, 2, 1, 0); \
55923		__ret_576 = __rev0_576 - __rev1_576 * __noswap_splatq_laneq_u32(__rev2_576, __p3_576); \
55924		__ret_576 = __builtin_shufflevector(__ret_576, __ret_576, 3, 2, 1, 0); \
55925		__ret_576; \
55926		})
55927		#endif
55928
55929		#ifdef __LITTLE_ENDIAN__
55930		#define vmlsq_laneq_u16(__p0_577, __p1_577, __p2_577, __p3_577) __extension__ ({ \
55931		uint16x8_t __ret_577; \
55932		uint16x8_t __s0_577 = __p0_577; \
55933		uint16x8_t __s1_577 = __p1_577; \
55934		uint16x8_t __s2_577 = __p2_577; \
55935		__ret_577 = __s0_577 - __s1_577 * splatq_laneq_u16(__s2_577, __p3_577); \
55936		__ret_577; \
55937		})
55938		#else
55939		#define vmlsq_laneq_u16(__p0_578, __p1_578, __p2_578, __p3_578) __extension__ ({ \
55940		uint16x8_t __ret_578; \
55941		uint16x8_t __s0_578 = __p0_578; \
55942		uint16x8_t __s1_578 = __p1_578; \
55943		uint16x8_t __s2_578 = __p2_578; \
55944		uint16x8_t __rev0_578; __rev0_578 = __builtin_shufflevector(__s0_578, __s0_578, 7, 6, 5, 4, 3, 2, 1, 0); \
55945		uint16x8_t __rev1_578; __rev1_578 = __builtin_shufflevector(__s1_578, __s1_578, 7, 6, 5, 4, 3, 2, 1, 0); \
55946		uint16x8_t __rev2_578; __rev2_578 = __builtin_shufflevector(__s2_578, __s2_578, 7, 6, 5, 4, 3, 2, 1, 0); \
55947		__ret_578 = __rev0_578 - __rev1_578 * __noswap_splatq_laneq_u16(__rev2_578, __p3_578); \
55948		__ret_578 = __builtin_shufflevector(__ret_578, __ret_578, 7, 6, 5, 4, 3, 2, 1, 0); \
55949		__ret_578; \
55950		})
55951		#endif
55952
55953		#ifdef __LITTLE_ENDIAN__
55954		#define vmlsq_laneq_f32(__p0_579, __p1_579, __p2_579, __p3_579) __extension__ ({ \
55955		float32x4_t __ret_579; \
55956		float32x4_t __s0_579 = __p0_579; \
55957		float32x4_t __s1_579 = __p1_579; \
55958		float32x4_t __s2_579 = __p2_579; \
55959		__ret_579 = __s0_579 - __s1_579 * splatq_laneq_f32(__s2_579, __p3_579); \
55960		__ret_579; \
55961		})
55962		#else
55963		#define vmlsq_laneq_f32(__p0_580, __p1_580, __p2_580, __p3_580) __extension__ ({ \
55964		float32x4_t __ret_580; \
55965		float32x4_t __s0_580 = __p0_580; \
55966		float32x4_t __s1_580 = __p1_580; \
55967		float32x4_t __s2_580 = __p2_580; \
55968		float32x4_t __rev0_580; __rev0_580 = __builtin_shufflevector(__s0_580, __s0_580, 3, 2, 1, 0); \
55969		float32x4_t __rev1_580; __rev1_580 = __builtin_shufflevector(__s1_580, __s1_580, 3, 2, 1, 0); \
55970		float32x4_t __rev2_580; __rev2_580 = __builtin_shufflevector(__s2_580, __s2_580, 3, 2, 1, 0); \
55971		__ret_580 = __rev0_580 - __rev1_580 * __noswap_splatq_laneq_f32(__rev2_580, __p3_580); \
55972		__ret_580 = __builtin_shufflevector(__ret_580, __ret_580, 3, 2, 1, 0); \
55973		__ret_580; \
55974		})
55975		#endif
55976
55977		#ifdef __LITTLE_ENDIAN__
55978		#define vmlsq_laneq_s32(__p0_581, __p1_581, __p2_581, __p3_581) __extension__ ({ \
55979		int32x4_t __ret_581; \
55980		int32x4_t __s0_581 = __p0_581; \
55981		int32x4_t __s1_581 = __p1_581; \
55982		int32x4_t __s2_581 = __p2_581; \
55983		__ret_581 = __s0_581 - __s1_581 * splatq_laneq_s32(__s2_581, __p3_581); \
55984		__ret_581; \
55985		})
55986		#else
55987		#define vmlsq_laneq_s32(__p0_582, __p1_582, __p2_582, __p3_582) __extension__ ({ \
55988		int32x4_t __ret_582; \
55989		int32x4_t __s0_582 = __p0_582; \
55990		int32x4_t __s1_582 = __p1_582; \
55991		int32x4_t __s2_582 = __p2_582; \
55992		int32x4_t __rev0_582; __rev0_582 = __builtin_shufflevector(__s0_582, __s0_582, 3, 2, 1, 0); \
55993		int32x4_t __rev1_582; __rev1_582 = __builtin_shufflevector(__s1_582, __s1_582, 3, 2, 1, 0); \
55994		int32x4_t __rev2_582; __rev2_582 = __builtin_shufflevector(__s2_582, __s2_582, 3, 2, 1, 0); \
55995		__ret_582 = __rev0_582 - __rev1_582 * __noswap_splatq_laneq_s32(__rev2_582, __p3_582); \
55996		__ret_582 = __builtin_shufflevector(__ret_582, __ret_582, 3, 2, 1, 0); \
55997		__ret_582; \
55998		})
55999		#endif
56000
56001		#ifdef __LITTLE_ENDIAN__
56002		#define vmlsq_laneq_s16(__p0_583, __p1_583, __p2_583, __p3_583) __extension__ ({ \
56003		int16x8_t __ret_583; \
56004		int16x8_t __s0_583 = __p0_583; \
56005		int16x8_t __s1_583 = __p1_583; \
56006		int16x8_t __s2_583 = __p2_583; \
56007		__ret_583 = __s0_583 - __s1_583 * splatq_laneq_s16(__s2_583, __p3_583); \
56008		__ret_583; \
56009		})
56010		#else
56011		#define vmlsq_laneq_s16(__p0_584, __p1_584, __p2_584, __p3_584) __extension__ ({ \
56012		int16x8_t __ret_584; \
56013		int16x8_t __s0_584 = __p0_584; \
56014		int16x8_t __s1_584 = __p1_584; \
56015		int16x8_t __s2_584 = __p2_584; \
56016		int16x8_t __rev0_584; __rev0_584 = __builtin_shufflevector(__s0_584, __s0_584, 7, 6, 5, 4, 3, 2, 1, 0); \
56017		int16x8_t __rev1_584; __rev1_584 = __builtin_shufflevector(__s1_584, __s1_584, 7, 6, 5, 4, 3, 2, 1, 0); \
56018		int16x8_t __rev2_584; __rev2_584 = __builtin_shufflevector(__s2_584, __s2_584, 7, 6, 5, 4, 3, 2, 1, 0); \
56019		__ret_584 = __rev0_584 - __rev1_584 * __noswap_splatq_laneq_s16(__rev2_584, __p3_584); \
56020		__ret_584 = __builtin_shufflevector(__ret_584, __ret_584, 7, 6, 5, 4, 3, 2, 1, 0); \
56021		__ret_584; \
56022		})
56023		#endif
56024
56025		#ifdef __LITTLE_ENDIAN__
56026		#define vmls_laneq_u32(__p0_585, __p1_585, __p2_585, __p3_585) __extension__ ({ \
56027		uint32x2_t __ret_585; \
56028		uint32x2_t __s0_585 = __p0_585; \
56029		uint32x2_t __s1_585 = __p1_585; \
56030		uint32x4_t __s2_585 = __p2_585; \
56031		__ret_585 = __s0_585 - __s1_585 * splat_laneq_u32(__s2_585, __p3_585); \
56032		__ret_585; \
56033		})
56034		#else
56035		#define vmls_laneq_u32(__p0_586, __p1_586, __p2_586, __p3_586) __extension__ ({ \
56036		uint32x2_t __ret_586; \
56037		uint32x2_t __s0_586 = __p0_586; \
56038		uint32x2_t __s1_586 = __p1_586; \
56039		uint32x4_t __s2_586 = __p2_586; \
56040		uint32x2_t __rev0_586; __rev0_586 = __builtin_shufflevector(__s0_586, __s0_586, 1, 0); \
56041		uint32x2_t __rev1_586; __rev1_586 = __builtin_shufflevector(__s1_586, __s1_586, 1, 0); \
56042		uint32x4_t __rev2_586; __rev2_586 = __builtin_shufflevector(__s2_586, __s2_586, 3, 2, 1, 0); \
56043		__ret_586 = __rev0_586 - __rev1_586 * __noswap_splat_laneq_u32(__rev2_586, __p3_586); \
56044		__ret_586 = __builtin_shufflevector(__ret_586, __ret_586, 1, 0); \
56045		__ret_586; \
56046		})
56047		#endif
56048
56049		#ifdef __LITTLE_ENDIAN__
56050		#define vmls_laneq_u16(__p0_587, __p1_587, __p2_587, __p3_587) __extension__ ({ \
56051		uint16x4_t __ret_587; \
56052		uint16x4_t __s0_587 = __p0_587; \
56053		uint16x4_t __s1_587 = __p1_587; \
56054		uint16x8_t __s2_587 = __p2_587; \
56055		__ret_587 = __s0_587 - __s1_587 * splat_laneq_u16(__s2_587, __p3_587); \
56056		__ret_587; \
56057		})
56058		#else
56059		#define vmls_laneq_u16(__p0_588, __p1_588, __p2_588, __p3_588) __extension__ ({ \
56060		uint16x4_t __ret_588; \
56061		uint16x4_t __s0_588 = __p0_588; \
56062		uint16x4_t __s1_588 = __p1_588; \
56063		uint16x8_t __s2_588 = __p2_588; \
56064		uint16x4_t __rev0_588; __rev0_588 = __builtin_shufflevector(__s0_588, __s0_588, 3, 2, 1, 0); \
56065		uint16x4_t __rev1_588; __rev1_588 = __builtin_shufflevector(__s1_588, __s1_588, 3, 2, 1, 0); \
56066		uint16x8_t __rev2_588; __rev2_588 = __builtin_shufflevector(__s2_588, __s2_588, 7, 6, 5, 4, 3, 2, 1, 0); \
56067		__ret_588 = __rev0_588 - __rev1_588 * __noswap_splat_laneq_u16(__rev2_588, __p3_588); \
56068		__ret_588 = __builtin_shufflevector(__ret_588, __ret_588, 3, 2, 1, 0); \
56069		__ret_588; \
56070		})
56071		#endif
56072
56073		#ifdef __LITTLE_ENDIAN__
56074		#define vmls_laneq_f32(__p0_589, __p1_589, __p2_589, __p3_589) __extension__ ({ \
56075		float32x2_t __ret_589; \
56076		float32x2_t __s0_589 = __p0_589; \
56077		float32x2_t __s1_589 = __p1_589; \
56078		float32x4_t __s2_589 = __p2_589; \
56079		__ret_589 = __s0_589 - __s1_589 * splat_laneq_f32(__s2_589, __p3_589); \
56080		__ret_589; \
56081		})
56082		#else
56083		#define vmls_laneq_f32(__p0_590, __p1_590, __p2_590, __p3_590) __extension__ ({ \
56084		float32x2_t __ret_590; \
56085		float32x2_t __s0_590 = __p0_590; \
56086		float32x2_t __s1_590 = __p1_590; \
56087		float32x4_t __s2_590 = __p2_590; \
56088		float32x2_t __rev0_590; __rev0_590 = __builtin_shufflevector(__s0_590, __s0_590, 1, 0); \
56089		float32x2_t __rev1_590; __rev1_590 = __builtin_shufflevector(__s1_590, __s1_590, 1, 0); \
56090		float32x4_t __rev2_590; __rev2_590 = __builtin_shufflevector(__s2_590, __s2_590, 3, 2, 1, 0); \
56091		__ret_590 = __rev0_590 - __rev1_590 * __noswap_splat_laneq_f32(__rev2_590, __p3_590); \
56092		__ret_590 = __builtin_shufflevector(__ret_590, __ret_590, 1, 0); \
56093		__ret_590; \
56094		})
56095		#endif
56096
56097		#ifdef __LITTLE_ENDIAN__
56098		#define vmls_laneq_s32(__p0_591, __p1_591, __p2_591, __p3_591) __extension__ ({ \
56099		int32x2_t __ret_591; \
56100		int32x2_t __s0_591 = __p0_591; \
56101		int32x2_t __s1_591 = __p1_591; \
56102		int32x4_t __s2_591 = __p2_591; \
56103		__ret_591 = __s0_591 - __s1_591 * splat_laneq_s32(__s2_591, __p3_591); \
56104		__ret_591; \
56105		})
56106		#else
56107		#define vmls_laneq_s32(__p0_592, __p1_592, __p2_592, __p3_592) __extension__ ({ \
56108		int32x2_t __ret_592; \
56109		int32x2_t __s0_592 = __p0_592; \
56110		int32x2_t __s1_592 = __p1_592; \
56111		int32x4_t __s2_592 = __p2_592; \
56112		int32x2_t __rev0_592; __rev0_592 = __builtin_shufflevector(__s0_592, __s0_592, 1, 0); \
56113		int32x2_t __rev1_592; __rev1_592 = __builtin_shufflevector(__s1_592, __s1_592, 1, 0); \
56114		int32x4_t __rev2_592; __rev2_592 = __builtin_shufflevector(__s2_592, __s2_592, 3, 2, 1, 0); \
56115		__ret_592 = __rev0_592 - __rev1_592 * __noswap_splat_laneq_s32(__rev2_592, __p3_592); \
56116		__ret_592 = __builtin_shufflevector(__ret_592, __ret_592, 1, 0); \
56117		__ret_592; \
56118		})
56119		#endif
56120
56121		#ifdef __LITTLE_ENDIAN__
56122		#define vmls_laneq_s16(__p0_593, __p1_593, __p2_593, __p3_593) __extension__ ({ \
56123		int16x4_t __ret_593; \
56124		int16x4_t __s0_593 = __p0_593; \
56125		int16x4_t __s1_593 = __p1_593; \
56126		int16x8_t __s2_593 = __p2_593; \
56127		__ret_593 = __s0_593 - __s1_593 * splat_laneq_s16(__s2_593, __p3_593); \
56128		__ret_593; \
56129		})
56130		#else
56131		#define vmls_laneq_s16(__p0_594, __p1_594, __p2_594, __p3_594) __extension__ ({ \
56132		int16x4_t __ret_594; \
56133		int16x4_t __s0_594 = __p0_594; \
56134		int16x4_t __s1_594 = __p1_594; \
56135		int16x8_t __s2_594 = __p2_594; \
56136		int16x4_t __rev0_594; __rev0_594 = __builtin_shufflevector(__s0_594, __s0_594, 3, 2, 1, 0); \
56137		int16x4_t __rev1_594; __rev1_594 = __builtin_shufflevector(__s1_594, __s1_594, 3, 2, 1, 0); \
56138		int16x8_t __rev2_594; __rev2_594 = __builtin_shufflevector(__s2_594, __s2_594, 7, 6, 5, 4, 3, 2, 1, 0); \
56139		__ret_594 = __rev0_594 - __rev1_594 * __noswap_splat_laneq_s16(__rev2_594, __p3_594); \
56140		__ret_594 = __builtin_shufflevector(__ret_594, __ret_594, 3, 2, 1, 0); \
56141		__ret_594; \
56142		})
56143		#endif
56144
56145		#ifdef __LITTLE_ENDIAN__
56146		#define vmlsl_high_lane_u32(__p0_595, __p1_595, __p2_595, __p3_595) __extension__ ({ \
56147		uint64x2_t __ret_595; \
56148		uint64x2_t __s0_595 = __p0_595; \
56149		uint32x4_t __s1_595 = __p1_595; \
56150		uint32x2_t __s2_595 = __p2_595; \
56151		__ret_595 = __s0_595 - vmull_u32(vget_high_u32(__s1_595), splat_lane_u32(__s2_595, __p3_595)); \
56152		__ret_595; \
56153		})
56154		#else
56155		#define vmlsl_high_lane_u32(__p0_596, __p1_596, __p2_596, __p3_596) __extension__ ({ \
56156		uint64x2_t __ret_596; \
56157		uint64x2_t __s0_596 = __p0_596; \
56158		uint32x4_t __s1_596 = __p1_596; \
56159		uint32x2_t __s2_596 = __p2_596; \
56160		uint64x2_t __rev0_596; __rev0_596 = __builtin_shufflevector(__s0_596, __s0_596, 1, 0); \
56161		uint32x4_t __rev1_596; __rev1_596 = __builtin_shufflevector(__s1_596, __s1_596, 3, 2, 1, 0); \
56162		uint32x2_t __rev2_596; __rev2_596 = __builtin_shufflevector(__s2_596, __s2_596, 1, 0); \
56163		__ret_596 = __rev0_596 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_596), __noswap_splat_lane_u32(__rev2_596, __p3_596)); \
56164		__ret_596 = __builtin_shufflevector(__ret_596, __ret_596, 1, 0); \
56165		__ret_596; \
56166		})
56167		#endif
56168
56169		#ifdef __LITTLE_ENDIAN__
56170		#define vmlsl_high_lane_u16(__p0_597, __p1_597, __p2_597, __p3_597) __extension__ ({ \
56171		uint32x4_t __ret_597; \
56172		uint32x4_t __s0_597 = __p0_597; \
56173		uint16x8_t __s1_597 = __p1_597; \
56174		uint16x4_t __s2_597 = __p2_597; \
56175		__ret_597 = __s0_597 - vmull_u16(vget_high_u16(__s1_597), splat_lane_u16(__s2_597, __p3_597)); \
56176		__ret_597; \
56177		})
56178		#else
56179		#define vmlsl_high_lane_u16(__p0_598, __p1_598, __p2_598, __p3_598) __extension__ ({ \
56180		uint32x4_t __ret_598; \
56181		uint32x4_t __s0_598 = __p0_598; \
56182		uint16x8_t __s1_598 = __p1_598; \
56183		uint16x4_t __s2_598 = __p2_598; \
56184		uint32x4_t __rev0_598; __rev0_598 = __builtin_shufflevector(__s0_598, __s0_598, 3, 2, 1, 0); \
56185		uint16x8_t __rev1_598; __rev1_598 = __builtin_shufflevector(__s1_598, __s1_598, 7, 6, 5, 4, 3, 2, 1, 0); \
56186		uint16x4_t __rev2_598; __rev2_598 = __builtin_shufflevector(__s2_598, __s2_598, 3, 2, 1, 0); \
56187		__ret_598 = __rev0_598 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_598), __noswap_splat_lane_u16(__rev2_598, __p3_598)); \
56188		__ret_598 = __builtin_shufflevector(__ret_598, __ret_598, 3, 2, 1, 0); \
56189		__ret_598; \
56190		})
56191		#endif
56192
56193		#ifdef __LITTLE_ENDIAN__
56194		#define vmlsl_high_lane_s32(__p0_599, __p1_599, __p2_599, __p3_599) __extension__ ({ \
56195		int64x2_t __ret_599; \
56196		int64x2_t __s0_599 = __p0_599; \
56197		int32x4_t __s1_599 = __p1_599; \
56198		int32x2_t __s2_599 = __p2_599; \
56199		__ret_599 = __s0_599 - vmull_s32(vget_high_s32(__s1_599), splat_lane_s32(__s2_599, __p3_599)); \
56200		__ret_599; \
56201		})
56202		#else
56203		#define vmlsl_high_lane_s32(__p0_600, __p1_600, __p2_600, __p3_600) __extension__ ({ \
56204		int64x2_t __ret_600; \
56205		int64x2_t __s0_600 = __p0_600; \
56206		int32x4_t __s1_600 = __p1_600; \
56207		int32x2_t __s2_600 = __p2_600; \
56208		int64x2_t __rev0_600; __rev0_600 = __builtin_shufflevector(__s0_600, __s0_600, 1, 0); \
56209		int32x4_t __rev1_600; __rev1_600 = __builtin_shufflevector(__s1_600, __s1_600, 3, 2, 1, 0); \
56210		int32x2_t __rev2_600; __rev2_600 = __builtin_shufflevector(__s2_600, __s2_600, 1, 0); \
56211		__ret_600 = __rev0_600 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_600), __noswap_splat_lane_s32(__rev2_600, __p3_600)); \
56212		__ret_600 = __builtin_shufflevector(__ret_600, __ret_600, 1, 0); \
56213		__ret_600; \
56214		})
56215		#endif
56216
56217		#ifdef __LITTLE_ENDIAN__
56218		#define vmlsl_high_lane_s16(__p0_601, __p1_601, __p2_601, __p3_601) __extension__ ({ \
56219		int32x4_t __ret_601; \
56220		int32x4_t __s0_601 = __p0_601; \
56221		int16x8_t __s1_601 = __p1_601; \
56222		int16x4_t __s2_601 = __p2_601; \
56223		__ret_601 = __s0_601 - vmull_s16(vget_high_s16(__s1_601), splat_lane_s16(__s2_601, __p3_601)); \
56224		__ret_601; \
56225		})
56226		#else
56227		#define vmlsl_high_lane_s16(__p0_602, __p1_602, __p2_602, __p3_602) __extension__ ({ \
56228		int32x4_t __ret_602; \
56229		int32x4_t __s0_602 = __p0_602; \
56230		int16x8_t __s1_602 = __p1_602; \
56231		int16x4_t __s2_602 = __p2_602; \
56232		int32x4_t __rev0_602; __rev0_602 = __builtin_shufflevector(__s0_602, __s0_602, 3, 2, 1, 0); \
56233		int16x8_t __rev1_602; __rev1_602 = __builtin_shufflevector(__s1_602, __s1_602, 7, 6, 5, 4, 3, 2, 1, 0); \
56234		int16x4_t __rev2_602; __rev2_602 = __builtin_shufflevector(__s2_602, __s2_602, 3, 2, 1, 0); \
56235		__ret_602 = __rev0_602 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_602), __noswap_splat_lane_s16(__rev2_602, __p3_602)); \
56236		__ret_602 = __builtin_shufflevector(__ret_602, __ret_602, 3, 2, 1, 0); \
56237		__ret_602; \
56238		})
56239		#endif
56240
56241		#ifdef __LITTLE_ENDIAN__
56242		#define vmlsl_high_laneq_u32(__p0_603, __p1_603, __p2_603, __p3_603) __extension__ ({ \
56243		uint64x2_t __ret_603; \
56244		uint64x2_t __s0_603 = __p0_603; \
56245		uint32x4_t __s1_603 = __p1_603; \
56246		uint32x4_t __s2_603 = __p2_603; \
56247		__ret_603 = __s0_603 - vmull_u32(vget_high_u32(__s1_603), splat_laneq_u32(__s2_603, __p3_603)); \
56248		__ret_603; \
56249		})
56250		#else
56251		#define vmlsl_high_laneq_u32(__p0_604, __p1_604, __p2_604, __p3_604) __extension__ ({ \
56252		uint64x2_t __ret_604; \
56253		uint64x2_t __s0_604 = __p0_604; \
56254		uint32x4_t __s1_604 = __p1_604; \
56255		uint32x4_t __s2_604 = __p2_604; \
56256		uint64x2_t __rev0_604; __rev0_604 = __builtin_shufflevector(__s0_604, __s0_604, 1, 0); \
56257		uint32x4_t __rev1_604; __rev1_604 = __builtin_shufflevector(__s1_604, __s1_604, 3, 2, 1, 0); \
56258		uint32x4_t __rev2_604; __rev2_604 = __builtin_shufflevector(__s2_604, __s2_604, 3, 2, 1, 0); \
56259		__ret_604 = __rev0_604 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_604), __noswap_splat_laneq_u32(__rev2_604, __p3_604)); \
56260		__ret_604 = __builtin_shufflevector(__ret_604, __ret_604, 1, 0); \
56261		__ret_604; \
56262		})
56263		#endif
56264
56265		#ifdef __LITTLE_ENDIAN__
56266		#define vmlsl_high_laneq_u16(__p0_605, __p1_605, __p2_605, __p3_605) __extension__ ({ \
56267		uint32x4_t __ret_605; \
56268		uint32x4_t __s0_605 = __p0_605; \
56269		uint16x8_t __s1_605 = __p1_605; \
56270		uint16x8_t __s2_605 = __p2_605; \
56271		__ret_605 = __s0_605 - vmull_u16(vget_high_u16(__s1_605), splat_laneq_u16(__s2_605, __p3_605)); \
56272		__ret_605; \
56273		})
56274		#else
56275		#define vmlsl_high_laneq_u16(__p0_606, __p1_606, __p2_606, __p3_606) __extension__ ({ \
56276		uint32x4_t __ret_606; \
56277		uint32x4_t __s0_606 = __p0_606; \
56278		uint16x8_t __s1_606 = __p1_606; \
56279		uint16x8_t __s2_606 = __p2_606; \
56280		uint32x4_t __rev0_606; __rev0_606 = __builtin_shufflevector(__s0_606, __s0_606, 3, 2, 1, 0); \
56281		uint16x8_t __rev1_606; __rev1_606 = __builtin_shufflevector(__s1_606, __s1_606, 7, 6, 5, 4, 3, 2, 1, 0); \
56282		uint16x8_t __rev2_606; __rev2_606 = __builtin_shufflevector(__s2_606, __s2_606, 7, 6, 5, 4, 3, 2, 1, 0); \
56283		__ret_606 = __rev0_606 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_606), __noswap_splat_laneq_u16(__rev2_606, __p3_606)); \
56284		__ret_606 = __builtin_shufflevector(__ret_606, __ret_606, 3, 2, 1, 0); \
56285		__ret_606; \
56286		})
56287		#endif
56288
56289		#ifdef __LITTLE_ENDIAN__
56290		#define vmlsl_high_laneq_s32(__p0_607, __p1_607, __p2_607, __p3_607) __extension__ ({ \
56291		int64x2_t __ret_607; \
56292		int64x2_t __s0_607 = __p0_607; \
56293		int32x4_t __s1_607 = __p1_607; \
56294		int32x4_t __s2_607 = __p2_607; \
56295		__ret_607 = __s0_607 - vmull_s32(vget_high_s32(__s1_607), splat_laneq_s32(__s2_607, __p3_607)); \
56296		__ret_607; \
56297		})
56298		#else
56299		#define vmlsl_high_laneq_s32(__p0_608, __p1_608, __p2_608, __p3_608) __extension__ ({ \
56300		int64x2_t __ret_608; \
56301		int64x2_t __s0_608 = __p0_608; \
56302		int32x4_t __s1_608 = __p1_608; \
56303		int32x4_t __s2_608 = __p2_608; \
56304		int64x2_t __rev0_608; __rev0_608 = __builtin_shufflevector(__s0_608, __s0_608, 1, 0); \
56305		int32x4_t __rev1_608; __rev1_608 = __builtin_shufflevector(__s1_608, __s1_608, 3, 2, 1, 0); \
56306		int32x4_t __rev2_608; __rev2_608 = __builtin_shufflevector(__s2_608, __s2_608, 3, 2, 1, 0); \
56307		__ret_608 = __rev0_608 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_608), __noswap_splat_laneq_s32(__rev2_608, __p3_608)); \
56308		__ret_608 = __builtin_shufflevector(__ret_608, __ret_608, 1, 0); \
56309		__ret_608; \
56310		})
56311		#endif
56312
56313		#ifdef __LITTLE_ENDIAN__
56314		#define vmlsl_high_laneq_s16(__p0_609, __p1_609, __p2_609, __p3_609) __extension__ ({ \
56315		int32x4_t __ret_609; \
56316		int32x4_t __s0_609 = __p0_609; \
56317		int16x8_t __s1_609 = __p1_609; \
56318		int16x8_t __s2_609 = __p2_609; \
56319		__ret_609 = __s0_609 - vmull_s16(vget_high_s16(__s1_609), splat_laneq_s16(__s2_609, __p3_609)); \
56320		__ret_609; \
56321		})
56322		#else
56323		#define vmlsl_high_laneq_s16(__p0_610, __p1_610, __p2_610, __p3_610) __extension__ ({ \
56324		int32x4_t __ret_610; \
56325		int32x4_t __s0_610 = __p0_610; \
56326		int16x8_t __s1_610 = __p1_610; \
56327		int16x8_t __s2_610 = __p2_610; \
56328		int32x4_t __rev0_610; __rev0_610 = __builtin_shufflevector(__s0_610, __s0_610, 3, 2, 1, 0); \
56329		int16x8_t __rev1_610; __rev1_610 = __builtin_shufflevector(__s1_610, __s1_610, 7, 6, 5, 4, 3, 2, 1, 0); \
56330		int16x8_t __rev2_610; __rev2_610 = __builtin_shufflevector(__s2_610, __s2_610, 7, 6, 5, 4, 3, 2, 1, 0); \
56331		__ret_610 = __rev0_610 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_610), __noswap_splat_laneq_s16(__rev2_610, __p3_610)); \
56332		__ret_610 = __builtin_shufflevector(__ret_610, __ret_610, 3, 2, 1, 0); \
56333		__ret_610; \
56334		})
56335		#endif
56336
56337		#ifdef __LITTLE_ENDIAN__
56338		#define vmlsl_laneq_u32(__p0_611, __p1_611, __p2_611, __p3_611) __extension__ ({ \
56339		uint64x2_t __ret_611; \
56340		uint64x2_t __s0_611 = __p0_611; \
56341		uint32x2_t __s1_611 = __p1_611; \
56342		uint32x4_t __s2_611 = __p2_611; \
56343		__ret_611 = __s0_611 - vmull_u32(__s1_611, splat_laneq_u32(__s2_611, __p3_611)); \
56344		__ret_611; \
56345		})
56346		#else
56347		#define vmlsl_laneq_u32(__p0_612, __p1_612, __p2_612, __p3_612) __extension__ ({ \
56348		uint64x2_t __ret_612; \
56349		uint64x2_t __s0_612 = __p0_612; \
56350		uint32x2_t __s1_612 = __p1_612; \
56351		uint32x4_t __s2_612 = __p2_612; \
56352		uint64x2_t __rev0_612; __rev0_612 = __builtin_shufflevector(__s0_612, __s0_612, 1, 0); \
56353		uint32x2_t __rev1_612; __rev1_612 = __builtin_shufflevector(__s1_612, __s1_612, 1, 0); \
56354		uint32x4_t __rev2_612; __rev2_612 = __builtin_shufflevector(__s2_612, __s2_612, 3, 2, 1, 0); \
56355		__ret_612 = __rev0_612 - __noswap_vmull_u32(__rev1_612, __noswap_splat_laneq_u32(__rev2_612, __p3_612)); \
56356		__ret_612 = __builtin_shufflevector(__ret_612, __ret_612, 1, 0); \
56357		__ret_612; \
56358		})
56359		#endif
56360
56361		#ifdef __LITTLE_ENDIAN__
56362		#define vmlsl_laneq_u16(__p0_613, __p1_613, __p2_613, __p3_613) __extension__ ({ \
56363		uint32x4_t __ret_613; \
56364		uint32x4_t __s0_613 = __p0_613; \
56365		uint16x4_t __s1_613 = __p1_613; \
56366		uint16x8_t __s2_613 = __p2_613; \
56367		__ret_613 = __s0_613 - vmull_u16(__s1_613, splat_laneq_u16(__s2_613, __p3_613)); \
56368		__ret_613; \
56369		})
56370		#else
56371		#define vmlsl_laneq_u16(__p0_614, __p1_614, __p2_614, __p3_614) __extension__ ({ \
56372		uint32x4_t __ret_614; \
56373		uint32x4_t __s0_614 = __p0_614; \
56374		uint16x4_t __s1_614 = __p1_614; \
56375		uint16x8_t __s2_614 = __p2_614; \
56376		uint32x4_t __rev0_614; __rev0_614 = __builtin_shufflevector(__s0_614, __s0_614, 3, 2, 1, 0); \
56377		uint16x4_t __rev1_614; __rev1_614 = __builtin_shufflevector(__s1_614, __s1_614, 3, 2, 1, 0); \
56378		uint16x8_t __rev2_614; __rev2_614 = __builtin_shufflevector(__s2_614, __s2_614, 7, 6, 5, 4, 3, 2, 1, 0); \
56379		__ret_614 = __rev0_614 - __noswap_vmull_u16(__rev1_614, __noswap_splat_laneq_u16(__rev2_614, __p3_614)); \
56380		__ret_614 = __builtin_shufflevector(__ret_614, __ret_614, 3, 2, 1, 0); \
56381		__ret_614; \
56382		})
56383		#endif
56384
56385		#ifdef __LITTLE_ENDIAN__
56386		#define vmlsl_laneq_s32(__p0_615, __p1_615, __p2_615, __p3_615) __extension__ ({ \
56387		int64x2_t __ret_615; \
56388		int64x2_t __s0_615 = __p0_615; \
56389		int32x2_t __s1_615 = __p1_615; \
56390		int32x4_t __s2_615 = __p2_615; \
56391		__ret_615 = __s0_615 - vmull_s32(__s1_615, splat_laneq_s32(__s2_615, __p3_615)); \
56392		__ret_615; \
56393		})
56394		#else
56395		#define vmlsl_laneq_s32(__p0_616, __p1_616, __p2_616, __p3_616) __extension__ ({ \
56396		int64x2_t __ret_616; \
56397		int64x2_t __s0_616 = __p0_616; \
56398		int32x2_t __s1_616 = __p1_616; \
56399		int32x4_t __s2_616 = __p2_616; \
56400		int64x2_t __rev0_616; __rev0_616 = __builtin_shufflevector(__s0_616, __s0_616, 1, 0); \
56401		int32x2_t __rev1_616; __rev1_616 = __builtin_shufflevector(__s1_616, __s1_616, 1, 0); \
56402		int32x4_t __rev2_616; __rev2_616 = __builtin_shufflevector(__s2_616, __s2_616, 3, 2, 1, 0); \
56403		__ret_616 = __rev0_616 - __noswap_vmull_s32(__rev1_616, __noswap_splat_laneq_s32(__rev2_616, __p3_616)); \
56404		__ret_616 = __builtin_shufflevector(__ret_616, __ret_616, 1, 0); \
56405		__ret_616; \
56406		})
56407		#endif
56408
56409		#ifdef __LITTLE_ENDIAN__
56410		#define vmlsl_laneq_s16(__p0_617, __p1_617, __p2_617, __p3_617) __extension__ ({ \
56411		int32x4_t __ret_617; \
56412		int32x4_t __s0_617 = __p0_617; \
56413		int16x4_t __s1_617 = __p1_617; \
56414		int16x8_t __s2_617 = __p2_617; \
56415		__ret_617 = __s0_617 - vmull_s16(__s1_617, splat_laneq_s16(__s2_617, __p3_617)); \
56416		__ret_617; \
56417		})
56418		#else
56419		#define vmlsl_laneq_s16(__p0_618, __p1_618, __p2_618, __p3_618) __extension__ ({ \
56420		int32x4_t __ret_618; \
56421		int32x4_t __s0_618 = __p0_618; \
56422		int16x4_t __s1_618 = __p1_618; \
56423		int16x8_t __s2_618 = __p2_618; \
56424		int32x4_t __rev0_618; __rev0_618 = __builtin_shufflevector(__s0_618, __s0_618, 3, 2, 1, 0); \
56425		int16x4_t __rev1_618; __rev1_618 = __builtin_shufflevector(__s1_618, __s1_618, 3, 2, 1, 0); \
56426		int16x8_t __rev2_618; __rev2_618 = __builtin_shufflevector(__s2_618, __s2_618, 7, 6, 5, 4, 3, 2, 1, 0); \
56427		__ret_618 = __rev0_618 - __noswap_vmull_s16(__rev1_618, __noswap_splat_laneq_s16(__rev2_618, __p3_618)); \
56428		__ret_618 = __builtin_shufflevector(__ret_618, __ret_618, 3, 2, 1, 0); \
56429		__ret_618; \
56430		})
56431		#endif
56432
56433		__ai poly64x1_t vmov_n_p64(poly64_t __p0) {
56434		poly64x1_t __ret;
56435		__ret = (poly64x1_t) {__p0};
	57282	__ai float16x8_t vreinterpretq_f16_u8(uint8x16_t __p0) {
	57283	float16x8_t __ret;
	57284	__ret = (float16x8_t)(__p0);
56436	57285	return __ret;
56437	57286	}
56438		#ifdef __LITTLE_ENDIAN__
56439		__ai poly64x2_t vmovq_n_p64(poly64_t __p0) {
56440		poly64x2_t __ret;
56441		__ret = (poly64x2_t) {__p0, __p0};
	57287	__ai float16x8_t vreinterpretq_f16_u32(uint32x4_t __p0) {
	57288	float16x8_t __ret;
	57289	__ret = (float16x8_t)(__p0);
56442	57290	return __ret;
56443	57291	}
56444		#else
56445		__ai poly64x2_t vmovq_n_p64(poly64_t __p0) {
56446		poly64x2_t __ret;
56447		__ret = (poly64x2_t) {__p0, __p0};
56448		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57292	__ai float16x8_t vreinterpretq_f16_u64(uint64x2_t __p0) {
	57293	float16x8_t __ret;
	57294	__ret = (float16x8_t)(__p0);
56449	57295	return __ret;
56450	57296	}
56451		#endif
56452
56453		#ifdef __LITTLE_ENDIAN__
56454		__ai float64x2_t vmovq_n_f64(float64_t __p0) {
56455		float64x2_t __ret;
56456		__ret = (float64x2_t) {__p0, __p0};
	57297	__ai float16x8_t vreinterpretq_f16_u16(uint16x8_t __p0) {
	57298	float16x8_t __ret;
	57299	__ret = (float16x8_t)(__p0);
56457	57300	return __ret;
56458	57301	}
56459		#else
56460		__ai float64x2_t vmovq_n_f64(float64_t __p0) {
56461		float64x2_t __ret;
56462		__ret = (float64x2_t) {__p0, __p0};
56463		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57302	__ai float16x8_t vreinterpretq_f16_s8(int8x16_t __p0) {
	57303	float16x8_t __ret;
	57304	__ret = (float16x8_t)(__p0);
56464	57305	return __ret;
56465	57306	}
56466		#endif
56467
56468		__ai float64x1_t vmov_n_f64(float64_t __p0) {
56469		float64x1_t __ret;
56470		__ret = (float64x1_t) {__p0};
	57307	__ai float16x8_t vreinterpretq_f16_f64(float64x2_t __p0) {
	57308	float16x8_t __ret;
	57309	__ret = (float16x8_t)(__p0);
56471	57310	return __ret;
56472	57311	}
56473		#ifdef __LITTLE_ENDIAN__
56474		__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_619) {
56475		uint16x8_t __ret_619;
56476		uint8x8_t __a1_619 = vget_high_u8(__p0_619);
56477		__ret_619 = (uint16x8_t)(vshll_n_u8(__a1_619, 0));
56478		return __ret_619;
	57312	__ai float16x8_t vreinterpretq_f16_f32(float32x4_t __p0) {
	57313	float16x8_t __ret;
	57314	__ret = (float16x8_t)(__p0);
	57315	return __ret;
56479	57316	}
56480		#else
56481		__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_620) {
56482		uint16x8_t __ret_620;
56483		uint8x16_t __rev0_620; __rev0_620 = __builtin_shufflevector(__p0_620, __p0_620, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
56484		uint8x8_t __a1_620 = __noswap_vget_high_u8(__rev0_620);
56485		__ret_620 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_620, 0));
56486		__ret_620 = __builtin_shufflevector(__ret_620, __ret_620, 7, 6, 5, 4, 3, 2, 1, 0);
56487		return __ret_620;
	57317	__ai float16x8_t vreinterpretq_f16_s32(int32x4_t __p0) {
	57318	float16x8_t __ret;
	57319	__ret = (float16x8_t)(__p0);
	57320	return __ret;
56488	57321	}
56489		__ai uint16x8_t __noswap_vmovl_high_u8(uint8x16_t __p0_621) {
56490		uint16x8_t __ret_621;
56491		uint8x8_t __a1_621 = __noswap_vget_high_u8(__p0_621);
56492		__ret_621 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_621, 0));
56493		return __ret_621;
	57322	__ai float16x8_t vreinterpretq_f16_s64(int64x2_t __p0) {
	57323	float16x8_t __ret;
	57324	__ret = (float16x8_t)(__p0);
	57325	return __ret;
56494	57326	}
56495		#endif
56496
56497		#ifdef __LITTLE_ENDIAN__
56498		__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_622) {
56499		uint64x2_t __ret_622;
56500		uint32x2_t __a1_622 = vget_high_u32(__p0_622);
56501		__ret_622 = (uint64x2_t)(vshll_n_u32(__a1_622, 0));
56502		return __ret_622;
	57327	__ai float16x8_t vreinterpretq_f16_s16(int16x8_t __p0) {
	57328	float16x8_t __ret;
	57329	__ret = (float16x8_t)(__p0);
	57330	return __ret;
56503	57331	}
56504		#else
56505		__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_623) {
56506		uint64x2_t __ret_623;
56507		uint32x4_t __rev0_623; __rev0_623 = __builtin_shufflevector(__p0_623, __p0_623, 3, 2, 1, 0);
56508		uint32x2_t __a1_623 = __noswap_vget_high_u32(__rev0_623);
56509		__ret_623 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_623, 0));
56510		__ret_623 = __builtin_shufflevector(__ret_623, __ret_623, 1, 0);
56511		return __ret_623;
	57332	__ai int32x4_t vreinterpretq_s32_p8(poly8x16_t __p0) {
	57333	int32x4_t __ret;
	57334	__ret = (int32x4_t)(__p0);
	57335	return __ret;
56512	57336	}
56513		__ai uint64x2_t __noswap_vmovl_high_u32(uint32x4_t __p0_624) {
56514		uint64x2_t __ret_624;
56515		uint32x2_t __a1_624 = __noswap_vget_high_u32(__p0_624);
56516		__ret_624 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_624, 0));
56517		return __ret_624;
	57337	__ai int32x4_t vreinterpretq_s32_p128(poly128_t __p0) {
	57338	int32x4_t __ret;
	57339	__ret = (int32x4_t)(__p0);
	57340	return __ret;
56518	57341	}
56519		#endif
56520
56521		#ifdef __LITTLE_ENDIAN__
56522		__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_625) {
56523		uint32x4_t __ret_625;
56524		uint16x4_t __a1_625 = vget_high_u16(__p0_625);
56525		__ret_625 = (uint32x4_t)(vshll_n_u16(__a1_625, 0));
56526		return __ret_625;
	57342	__ai int32x4_t vreinterpretq_s32_p64(poly64x2_t __p0) {
	57343	int32x4_t __ret;
	57344	__ret = (int32x4_t)(__p0);
	57345	return __ret;
56527	57346	}
56528		#else
56529		__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_626) {
56530		uint32x4_t __ret_626;
56531		uint16x8_t __rev0_626; __rev0_626 = __builtin_shufflevector(__p0_626, __p0_626, 7, 6, 5, 4, 3, 2, 1, 0);
56532		uint16x4_t __a1_626 = __noswap_vget_high_u16(__rev0_626);
56533		__ret_626 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_626, 0));
56534		__ret_626 = __builtin_shufflevector(__ret_626, __ret_626, 3, 2, 1, 0);
56535		return __ret_626;
	57347	__ai int32x4_t vreinterpretq_s32_p16(poly16x8_t __p0) {
	57348	int32x4_t __ret;
	57349	__ret = (int32x4_t)(__p0);
	57350	return __ret;
56536	57351	}
56537		__ai uint32x4_t __noswap_vmovl_high_u16(uint16x8_t __p0_627) {
56538		uint32x4_t __ret_627;
56539		uint16x4_t __a1_627 = __noswap_vget_high_u16(__p0_627);
56540		__ret_627 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_627, 0));
56541		return __ret_627;
	57352	__ai int32x4_t vreinterpretq_s32_u8(uint8x16_t __p0) {
	57353	int32x4_t __ret;
	57354	__ret = (int32x4_t)(__p0);
	57355	return __ret;
56542	57356	}
56543		#endif
56544
56545		#ifdef __LITTLE_ENDIAN__
56546		__ai int16x8_t vmovl_high_s8(int8x16_t __p0_628) {
56547		int16x8_t __ret_628;
56548		int8x8_t __a1_628 = vget_high_s8(__p0_628);
56549		__ret_628 = (int16x8_t)(vshll_n_s8(__a1_628, 0));
56550		return __ret_628;
	57357	__ai int32x4_t vreinterpretq_s32_u32(uint32x4_t __p0) {
	57358	int32x4_t __ret;
	57359	__ret = (int32x4_t)(__p0);
	57360	return __ret;
56551	57361	}
56552		#else
56553		__ai int16x8_t vmovl_high_s8(int8x16_t __p0_629) {
56554		int16x8_t __ret_629;
56555		int8x16_t __rev0_629; __rev0_629 = __builtin_shufflevector(__p0_629, __p0_629, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
56556		int8x8_t __a1_629 = __noswap_vget_high_s8(__rev0_629);
56557		__ret_629 = (int16x8_t)(__noswap_vshll_n_s8(__a1_629, 0));
56558		__ret_629 = __builtin_shufflevector(__ret_629, __ret_629, 7, 6, 5, 4, 3, 2, 1, 0);
56559		return __ret_629;
	57362	__ai int32x4_t vreinterpretq_s32_u64(uint64x2_t __p0) {
	57363	int32x4_t __ret;
	57364	__ret = (int32x4_t)(__p0);
	57365	return __ret;
56560	57366	}
56561		__ai int16x8_t __noswap_vmovl_high_s8(int8x16_t __p0_630) {
56562		int16x8_t __ret_630;
56563		int8x8_t __a1_630 = __noswap_vget_high_s8(__p0_630);
56564		__ret_630 = (int16x8_t)(__noswap_vshll_n_s8(__a1_630, 0));
56565		return __ret_630;
	57367	__ai int32x4_t vreinterpretq_s32_u16(uint16x8_t __p0) {
	57368	int32x4_t __ret;
	57369	__ret = (int32x4_t)(__p0);
	57370	return __ret;
56566	57371	}
56567		#endif
56568
56569		#ifdef __LITTLE_ENDIAN__
56570		__ai int64x2_t vmovl_high_s32(int32x4_t __p0_631) {
56571		int64x2_t __ret_631;
56572		int32x2_t __a1_631 = vget_high_s32(__p0_631);
56573		__ret_631 = (int64x2_t)(vshll_n_s32(__a1_631, 0));
56574		return __ret_631;
	57372	__ai int32x4_t vreinterpretq_s32_s8(int8x16_t __p0) {
	57373	int32x4_t __ret;
	57374	__ret = (int32x4_t)(__p0);
	57375	return __ret;
56575	57376	}
56576		#else
56577		__ai int64x2_t vmovl_high_s32(int32x4_t __p0_632) {
56578		int64x2_t __ret_632;
56579		int32x4_t __rev0_632; __rev0_632 = __builtin_shufflevector(__p0_632, __p0_632, 3, 2, 1, 0);
56580		int32x2_t __a1_632 = __noswap_vget_high_s32(__rev0_632);
56581		__ret_632 = (int64x2_t)(__noswap_vshll_n_s32(__a1_632, 0));
56582		__ret_632 = __builtin_shufflevector(__ret_632, __ret_632, 1, 0);
56583		return __ret_632;
	57377	__ai int32x4_t vreinterpretq_s32_f64(float64x2_t __p0) {
	57378	int32x4_t __ret;
	57379	__ret = (int32x4_t)(__p0);
	57380	return __ret;
56584	57381	}
56585		__ai int64x2_t __noswap_vmovl_high_s32(int32x4_t __p0_633) {
56586		int64x2_t __ret_633;
56587		int32x2_t __a1_633 = __noswap_vget_high_s32(__p0_633);
56588		__ret_633 = (int64x2_t)(__noswap_vshll_n_s32(__a1_633, 0));
56589		return __ret_633;
	57382	__ai int32x4_t vreinterpretq_s32_f32(float32x4_t __p0) {
	57383	int32x4_t __ret;
	57384	__ret = (int32x4_t)(__p0);
	57385	return __ret;
56590	57386	}
56591		#endif
56592
56593		#ifdef __LITTLE_ENDIAN__
56594		__ai int32x4_t vmovl_high_s16(int16x8_t __p0_634) {
56595		int32x4_t __ret_634;
56596		int16x4_t __a1_634 = vget_high_s16(__p0_634);
56597		__ret_634 = (int32x4_t)(vshll_n_s16(__a1_634, 0));
56598		return __ret_634;
	57387	__ai int32x4_t vreinterpretq_s32_f16(float16x8_t __p0) {
	57388	int32x4_t __ret;
	57389	__ret = (int32x4_t)(__p0);
	57390	return __ret;
56599	57391	}
56600		#else
56601		__ai int32x4_t vmovl_high_s16(int16x8_t __p0_635) {
56602		int32x4_t __ret_635;
56603		int16x8_t __rev0_635; __rev0_635 = __builtin_shufflevector(__p0_635, __p0_635, 7, 6, 5, 4, 3, 2, 1, 0);
56604		int16x4_t __a1_635 = __noswap_vget_high_s16(__rev0_635);
56605		__ret_635 = (int32x4_t)(__noswap_vshll_n_s16(__a1_635, 0));
56606		__ret_635 = __builtin_shufflevector(__ret_635, __ret_635, 3, 2, 1, 0);
56607		return __ret_635;
	57392	__ai int32x4_t vreinterpretq_s32_s64(int64x2_t __p0) {
	57393	int32x4_t __ret;
	57394	__ret = (int32x4_t)(__p0);
	57395	return __ret;
56608	57396	}
56609		__ai int32x4_t __noswap_vmovl_high_s16(int16x8_t __p0_636) {
56610		int32x4_t __ret_636;
56611		int16x4_t __a1_636 = __noswap_vget_high_s16(__p0_636);
56612		__ret_636 = (int32x4_t)(__noswap_vshll_n_s16(__a1_636, 0));
56613		return __ret_636;
	57397	__ai int32x4_t vreinterpretq_s32_s16(int16x8_t __p0) {
	57398	int32x4_t __ret;
	57399	__ret = (int32x4_t)(__p0);
	57400	return __ret;
56614	57401	}
56615		#endif
56616
56617		#ifdef __LITTLE_ENDIAN__
56618		__ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
56619		uint16x8_t __ret;
56620		__ret = vcombine_u16(__p0, vmovn_u32(__p1));
	57402	__ai int64x2_t vreinterpretq_s64_p8(poly8x16_t __p0) {
	57403	int64x2_t __ret;
	57404	__ret = (int64x2_t)(__p0);
56621	57405	return __ret;
56622	57406	}
56623		#else
56624		__ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
56625		uint16x8_t __ret;
56626		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
56627		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56628		__ret = __noswap_vcombine_u16(__rev0, __noswap_vmovn_u32(__rev1));
56629		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57407	__ai int64x2_t vreinterpretq_s64_p128(poly128_t __p0) {
	57408	int64x2_t __ret;
	57409	__ret = (int64x2_t)(__p0);
56630	57410	return __ret;
56631	57411	}
56632		#endif
56633
56634		#ifdef __LITTLE_ENDIAN__
56635		__ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
56636		uint32x4_t __ret;
56637		__ret = vcombine_u32(__p0, vmovn_u64(__p1));
	57412	__ai int64x2_t vreinterpretq_s64_p64(poly64x2_t __p0) {
	57413	int64x2_t __ret;
	57414	__ret = (int64x2_t)(__p0);
56638	57415	return __ret;
56639	57416	}
56640		#else
56641		__ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
56642		uint32x4_t __ret;
56643		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
56644		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56645		__ret = __noswap_vcombine_u32(__rev0, __noswap_vmovn_u64(__rev1));
56646		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57417	__ai int64x2_t vreinterpretq_s64_p16(poly16x8_t __p0) {
	57418	int64x2_t __ret;
	57419	__ret = (int64x2_t)(__p0);
56647	57420	return __ret;
56648	57421	}
56649		#endif
56650
56651		#ifdef __LITTLE_ENDIAN__
56652		__ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
56653		uint8x16_t __ret;
56654		__ret = vcombine_u8(__p0, vmovn_u16(__p1));
	57422	__ai int64x2_t vreinterpretq_s64_u8(uint8x16_t __p0) {
	57423	int64x2_t __ret;
	57424	__ret = (int64x2_t)(__p0);
56655	57425	return __ret;
56656	57426	}
56657		#else
56658		__ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
56659		uint8x16_t __ret;
56660		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
56661		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56662		__ret = __noswap_vcombine_u8(__rev0, __noswap_vmovn_u16(__rev1));
56663		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	57427	__ai int64x2_t vreinterpretq_s64_u32(uint32x4_t __p0) {
	57428	int64x2_t __ret;
	57429	__ret = (int64x2_t)(__p0);
56664	57430	return __ret;
56665	57431	}
56666		#endif
56667
56668		#ifdef __LITTLE_ENDIAN__
56669		__ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
	57432	__ai int64x2_t vreinterpretq_s64_u64(uint64x2_t __p0) {
	57433	int64x2_t __ret;
	57434	__ret = (int64x2_t)(__p0);
	57435	return __ret;
	57436	}
	57437	__ai int64x2_t vreinterpretq_s64_u16(uint16x8_t __p0) {
	57438	int64x2_t __ret;
	57439	__ret = (int64x2_t)(__p0);
	57440	return __ret;
	57441	}
	57442	__ai int64x2_t vreinterpretq_s64_s8(int8x16_t __p0) {
	57443	int64x2_t __ret;
	57444	__ret = (int64x2_t)(__p0);
	57445	return __ret;
	57446	}
	57447	__ai int64x2_t vreinterpretq_s64_f64(float64x2_t __p0) {
	57448	int64x2_t __ret;
	57449	__ret = (int64x2_t)(__p0);
	57450	return __ret;
	57451	}
	57452	__ai int64x2_t vreinterpretq_s64_f32(float32x4_t __p0) {
	57453	int64x2_t __ret;
	57454	__ret = (int64x2_t)(__p0);
	57455	return __ret;
	57456	}
	57457	__ai int64x2_t vreinterpretq_s64_f16(float16x8_t __p0) {
	57458	int64x2_t __ret;
	57459	__ret = (int64x2_t)(__p0);
	57460	return __ret;
	57461	}
	57462	__ai int64x2_t vreinterpretq_s64_s32(int32x4_t __p0) {
	57463	int64x2_t __ret;
	57464	__ret = (int64x2_t)(__p0);
	57465	return __ret;
	57466	}
	57467	__ai int64x2_t vreinterpretq_s64_s16(int16x8_t __p0) {
	57468	int64x2_t __ret;
	57469	__ret = (int64x2_t)(__p0);
	57470	return __ret;
	57471	}
	57472	__ai int16x8_t vreinterpretq_s16_p8(poly8x16_t __p0) {
56670	57473	int16x8_t __ret;
56671		__ret = vcombine_s16(__p0, vmovn_s32(__p1));
	57474	__ret = (int16x8_t)(__p0);
56672	57475	return __ret;
56673	57476	}
56674		#else
56675		__ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
	57477	__ai int16x8_t vreinterpretq_s16_p128(poly128_t __p0) {
56676	57478	int16x8_t __ret;
56677		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
56678		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56679		__ret = __noswap_vcombine_s16(__rev0, __noswap_vmovn_s32(__rev1));
56680		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57479	__ret = (int16x8_t)(__p0);
56681	57480	return __ret;
56682	57481	}
56683		#endif
56684
56685		#ifdef __LITTLE_ENDIAN__
56686		__ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
56687		int32x4_t __ret;
56688		__ret = vcombine_s32(__p0, vmovn_s64(__p1));
	57482	__ai int16x8_t vreinterpretq_s16_p64(poly64x2_t __p0) {
	57483	int16x8_t __ret;
	57484	__ret = (int16x8_t)(__p0);
56689	57485	return __ret;
56690	57486	}
56691		#else
56692		__ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
56693		int32x4_t __ret;
56694		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
56695		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56696		__ret = __noswap_vcombine_s32(__rev0, __noswap_vmovn_s64(__rev1));
56697		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57487	__ai int16x8_t vreinterpretq_s16_p16(poly16x8_t __p0) {
	57488	int16x8_t __ret;
	57489	__ret = (int16x8_t)(__p0);
56698	57490	return __ret;
56699	57491	}
56700		#endif
56701
56702		#ifdef __LITTLE_ENDIAN__
56703		__ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
56704		int8x16_t __ret;
56705		__ret = vcombine_s8(__p0, vmovn_s16(__p1));
	57492	__ai int16x8_t vreinterpretq_s16_u8(uint8x16_t __p0) {
	57493	int16x8_t __ret;
	57494	__ret = (int16x8_t)(__p0);
56706	57495	return __ret;
56707	57496	}
56708		#else
56709		__ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
56710		int8x16_t __ret;
56711		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
56712		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56713		__ret = __noswap_vcombine_s8(__rev0, __noswap_vmovn_s16(__rev1));
56714		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	57497	__ai int16x8_t vreinterpretq_s16_u32(uint32x4_t __p0) {
	57498	int16x8_t __ret;
	57499	__ret = (int16x8_t)(__p0);
56715	57500	return __ret;
56716	57501	}
56717		#endif
56718
56719		#ifdef __LITTLE_ENDIAN__
56720		__ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
56721		float64x2_t __ret;
56722		__ret = __p0 * __p1;
	57502	__ai int16x8_t vreinterpretq_s16_u64(uint64x2_t __p0) {
	57503	int16x8_t __ret;
	57504	__ret = (int16x8_t)(__p0);
56723	57505	return __ret;
56724	57506	}
56725		#else
56726		__ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
56727		float64x2_t __ret;
56728		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
56729		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56730		__ret = __rev0 * __rev1;
56731		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57507	__ai int16x8_t vreinterpretq_s16_u16(uint16x8_t __p0) {
	57508	int16x8_t __ret;
	57509	__ret = (int16x8_t)(__p0);
56732	57510	return __ret;
56733	57511	}
56734		#endif
56735
56736		__ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
56737		float64x1_t __ret;
56738		__ret = __p0 * __p1;
	57512	__ai int16x8_t vreinterpretq_s16_s8(int8x16_t __p0) {
	57513	int16x8_t __ret;
	57514	__ret = (int16x8_t)(__p0);
56739	57515	return __ret;
56740	57516	}
56741		#define vmuld_lane_f64(__p0_637, __p1_637, __p2_637) __extension__ ({ \
56742		float64_t __ret_637; \
56743		float64_t __s0_637 = __p0_637; \
56744		float64x1_t __s1_637 = __p1_637; \
56745		__ret_637 = __s0_637 * vget_lane_f64(__s1_637, __p2_637); \
56746		__ret_637; \
56747		})
56748		#ifdef __LITTLE_ENDIAN__
56749		#define vmuls_lane_f32(__p0_638, __p1_638, __p2_638) __extension__ ({ \
56750		float32_t __ret_638; \
56751		float32_t __s0_638 = __p0_638; \
56752		float32x2_t __s1_638 = __p1_638; \
56753		__ret_638 = __s0_638 * vget_lane_f32(__s1_638, __p2_638); \
56754		__ret_638; \
56755		})
56756		#else
56757		#define vmuls_lane_f32(__p0_639, __p1_639, __p2_639) __extension__ ({ \
56758		float32_t __ret_639; \
56759		float32_t __s0_639 = __p0_639; \
56760		float32x2_t __s1_639 = __p1_639; \
56761		float32x2_t __rev1_639; __rev1_639 = __builtin_shufflevector(__s1_639, __s1_639, 1, 0); \
56762		__ret_639 = __s0_639 * __noswap_vget_lane_f32(__rev1_639, __p2_639); \
56763		__ret_639; \
56764		})
56765		#endif
56766
56767		#define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
56768		float64x1_t __ret; \
56769		float64x1_t __s0 = __p0; \
56770		float64x1_t __s1 = __p1; \
56771		__ret = (float64x1_t) __builtin_neon_vmul_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
56772		__ret; \
56773		})
56774		#ifdef __LITTLE_ENDIAN__
56775		#define vmulq_lane_f64(__p0_640, __p1_640, __p2_640) __extension__ ({ \
56776		float64x2_t __ret_640; \
56777		float64x2_t __s0_640 = __p0_640; \
56778		float64x1_t __s1_640 = __p1_640; \
56779		__ret_640 = __s0_640 * splatq_lane_f64(__s1_640, __p2_640); \
56780		__ret_640; \
56781		})
56782		#else
56783		#define vmulq_lane_f64(__p0_641, __p1_641, __p2_641) __extension__ ({ \
56784		float64x2_t __ret_641; \
56785		float64x2_t __s0_641 = __p0_641; \
56786		float64x1_t __s1_641 = __p1_641; \
56787		float64x2_t __rev0_641; __rev0_641 = __builtin_shufflevector(__s0_641, __s0_641, 1, 0); \
56788		__ret_641 = __rev0_641 * __noswap_splatq_lane_f64(__s1_641, __p2_641); \
56789		__ret_641 = __builtin_shufflevector(__ret_641, __ret_641, 1, 0); \
56790		__ret_641; \
56791		})
56792		#endif
56793
56794		#ifdef __LITTLE_ENDIAN__
56795		#define vmuld_laneq_f64(__p0_642, __p1_642, __p2_642) __extension__ ({ \
56796		float64_t __ret_642; \
56797		float64_t __s0_642 = __p0_642; \
56798		float64x2_t __s1_642 = __p1_642; \
56799		__ret_642 = __s0_642 * vgetq_lane_f64(__s1_642, __p2_642); \
56800		__ret_642; \
56801		})
56802		#else
56803		#define vmuld_laneq_f64(__p0_643, __p1_643, __p2_643) __extension__ ({ \
56804		float64_t __ret_643; \
56805		float64_t __s0_643 = __p0_643; \
56806		float64x2_t __s1_643 = __p1_643; \
56807		float64x2_t __rev1_643; __rev1_643 = __builtin_shufflevector(__s1_643, __s1_643, 1, 0); \
56808		__ret_643 = __s0_643 * __noswap_vgetq_lane_f64(__rev1_643, __p2_643); \
56809		__ret_643; \
56810		})
56811		#endif
56812
56813		#ifdef __LITTLE_ENDIAN__
56814		#define vmuls_laneq_f32(__p0_644, __p1_644, __p2_644) __extension__ ({ \
56815		float32_t __ret_644; \
56816		float32_t __s0_644 = __p0_644; \
56817		float32x4_t __s1_644 = __p1_644; \
56818		__ret_644 = __s0_644 * vgetq_lane_f32(__s1_644, __p2_644); \
56819		__ret_644; \
56820		})
56821		#else
56822		#define vmuls_laneq_f32(__p0_645, __p1_645, __p2_645) __extension__ ({ \
56823		float32_t __ret_645; \
56824		float32_t __s0_645 = __p0_645; \
56825		float32x4_t __s1_645 = __p1_645; \
56826		float32x4_t __rev1_645; __rev1_645 = __builtin_shufflevector(__s1_645, __s1_645, 3, 2, 1, 0); \
56827		__ret_645 = __s0_645 * __noswap_vgetq_lane_f32(__rev1_645, __p2_645); \
56828		__ret_645; \
56829		})
56830		#endif
56831
56832		#ifdef __LITTLE_ENDIAN__
56833		#define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
56834		float64x1_t __ret; \
56835		float64x1_t __s0 = __p0; \
56836		float64x2_t __s1 = __p1; \
56837		__ret = (float64x1_t) __builtin_neon_vmul_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 10); \
56838		__ret; \
56839		})
56840		#else
56841		#define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
56842		float64x1_t __ret; \
56843		float64x1_t __s0 = __p0; \
56844		float64x2_t __s1 = __p1; \
56845		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
56846		__ret = (float64x1_t) __builtin_neon_vmul_laneq_v((int8x8_t)__s0, (int8x16_t)__rev1, __p2, 10); \
56847		__ret; \
56848		})
56849		#endif
56850
56851		#ifdef __LITTLE_ENDIAN__
56852		#define vmulq_laneq_u32(__p0_646, __p1_646, __p2_646) __extension__ ({ \
56853		uint32x4_t __ret_646; \
56854		uint32x4_t __s0_646 = __p0_646; \
56855		uint32x4_t __s1_646 = __p1_646; \
56856		__ret_646 = __s0_646 * splatq_laneq_u32(__s1_646, __p2_646); \
56857		__ret_646; \
56858		})
56859		#else
56860		#define vmulq_laneq_u32(__p0_647, __p1_647, __p2_647) __extension__ ({ \
56861		uint32x4_t __ret_647; \
56862		uint32x4_t __s0_647 = __p0_647; \
56863		uint32x4_t __s1_647 = __p1_647; \
56864		uint32x4_t __rev0_647; __rev0_647 = __builtin_shufflevector(__s0_647, __s0_647, 3, 2, 1, 0); \
56865		uint32x4_t __rev1_647; __rev1_647 = __builtin_shufflevector(__s1_647, __s1_647, 3, 2, 1, 0); \
56866		__ret_647 = __rev0_647 * __noswap_splatq_laneq_u32(__rev1_647, __p2_647); \
56867		__ret_647 = __builtin_shufflevector(__ret_647, __ret_647, 3, 2, 1, 0); \
56868		__ret_647; \
56869		})
56870		#endif
56871
56872		#ifdef __LITTLE_ENDIAN__
56873		#define vmulq_laneq_u16(__p0_648, __p1_648, __p2_648) __extension__ ({ \
56874		uint16x8_t __ret_648; \
56875		uint16x8_t __s0_648 = __p0_648; \
56876		uint16x8_t __s1_648 = __p1_648; \
56877		__ret_648 = __s0_648 * splatq_laneq_u16(__s1_648, __p2_648); \
56878		__ret_648; \
56879		})
56880		#else
56881		#define vmulq_laneq_u16(__p0_649, __p1_649, __p2_649) __extension__ ({ \
56882		uint16x8_t __ret_649; \
56883		uint16x8_t __s0_649 = __p0_649; \
56884		uint16x8_t __s1_649 = __p1_649; \
56885		uint16x8_t __rev0_649; __rev0_649 = __builtin_shufflevector(__s0_649, __s0_649, 7, 6, 5, 4, 3, 2, 1, 0); \
56886		uint16x8_t __rev1_649; __rev1_649 = __builtin_shufflevector(__s1_649, __s1_649, 7, 6, 5, 4, 3, 2, 1, 0); \
56887		__ret_649 = __rev0_649 * __noswap_splatq_laneq_u16(__rev1_649, __p2_649); \
56888		__ret_649 = __builtin_shufflevector(__ret_649, __ret_649, 7, 6, 5, 4, 3, 2, 1, 0); \
56889		__ret_649; \
56890		})
56891		#endif
56892
56893		#ifdef __LITTLE_ENDIAN__
56894		#define vmulq_laneq_f64(__p0_650, __p1_650, __p2_650) __extension__ ({ \
56895		float64x2_t __ret_650; \
56896		float64x2_t __s0_650 = __p0_650; \
56897		float64x2_t __s1_650 = __p1_650; \
56898		__ret_650 = __s0_650 * splatq_laneq_f64(__s1_650, __p2_650); \
56899		__ret_650; \
56900		})
56901		#else
56902		#define vmulq_laneq_f64(__p0_651, __p1_651, __p2_651) __extension__ ({ \
56903		float64x2_t __ret_651; \
56904		float64x2_t __s0_651 = __p0_651; \
56905		float64x2_t __s1_651 = __p1_651; \
56906		float64x2_t __rev0_651; __rev0_651 = __builtin_shufflevector(__s0_651, __s0_651, 1, 0); \
56907		float64x2_t __rev1_651; __rev1_651 = __builtin_shufflevector(__s1_651, __s1_651, 1, 0); \
56908		__ret_651 = __rev0_651 * __noswap_splatq_laneq_f64(__rev1_651, __p2_651); \
56909		__ret_651 = __builtin_shufflevector(__ret_651, __ret_651, 1, 0); \
56910		__ret_651; \
56911		})
56912		#endif
56913
56914		#ifdef __LITTLE_ENDIAN__
56915		#define vmulq_laneq_f32(__p0_652, __p1_652, __p2_652) __extension__ ({ \
56916		float32x4_t __ret_652; \
56917		float32x4_t __s0_652 = __p0_652; \
56918		float32x4_t __s1_652 = __p1_652; \
56919		__ret_652 = __s0_652 * splatq_laneq_f32(__s1_652, __p2_652); \
56920		__ret_652; \
56921		})
56922		#else
56923		#define vmulq_laneq_f32(__p0_653, __p1_653, __p2_653) __extension__ ({ \
56924		float32x4_t __ret_653; \
56925		float32x4_t __s0_653 = __p0_653; \
56926		float32x4_t __s1_653 = __p1_653; \
56927		float32x4_t __rev0_653; __rev0_653 = __builtin_shufflevector(__s0_653, __s0_653, 3, 2, 1, 0); \
56928		float32x4_t __rev1_653; __rev1_653 = __builtin_shufflevector(__s1_653, __s1_653, 3, 2, 1, 0); \
56929		__ret_653 = __rev0_653 * __noswap_splatq_laneq_f32(__rev1_653, __p2_653); \
56930		__ret_653 = __builtin_shufflevector(__ret_653, __ret_653, 3, 2, 1, 0); \
56931		__ret_653; \
56932		})
56933		#endif
56934
56935		#ifdef __LITTLE_ENDIAN__
56936		#define vmulq_laneq_s32(__p0_654, __p1_654, __p2_654) __extension__ ({ \
56937		int32x4_t __ret_654; \
56938		int32x4_t __s0_654 = __p0_654; \
56939		int32x4_t __s1_654 = __p1_654; \
56940		__ret_654 = __s0_654 * splatq_laneq_s32(__s1_654, __p2_654); \
56941		__ret_654; \
56942		})
56943		#else
56944		#define vmulq_laneq_s32(__p0_655, __p1_655, __p2_655) __extension__ ({ \
56945		int32x4_t __ret_655; \
56946		int32x4_t __s0_655 = __p0_655; \
56947		int32x4_t __s1_655 = __p1_655; \
56948		int32x4_t __rev0_655; __rev0_655 = __builtin_shufflevector(__s0_655, __s0_655, 3, 2, 1, 0); \
56949		int32x4_t __rev1_655; __rev1_655 = __builtin_shufflevector(__s1_655, __s1_655, 3, 2, 1, 0); \
56950		__ret_655 = __rev0_655 * __noswap_splatq_laneq_s32(__rev1_655, __p2_655); \
56951		__ret_655 = __builtin_shufflevector(__ret_655, __ret_655, 3, 2, 1, 0); \
56952		__ret_655; \
56953		})
56954		#endif
56955
56956		#ifdef __LITTLE_ENDIAN__
56957		#define vmulq_laneq_s16(__p0_656, __p1_656, __p2_656) __extension__ ({ \
56958		int16x8_t __ret_656; \
56959		int16x8_t __s0_656 = __p0_656; \
56960		int16x8_t __s1_656 = __p1_656; \
56961		__ret_656 = __s0_656 * splatq_laneq_s16(__s1_656, __p2_656); \
56962		__ret_656; \
56963		})
56964		#else
56965		#define vmulq_laneq_s16(__p0_657, __p1_657, __p2_657) __extension__ ({ \
56966		int16x8_t __ret_657; \
56967		int16x8_t __s0_657 = __p0_657; \
56968		int16x8_t __s1_657 = __p1_657; \
56969		int16x8_t __rev0_657; __rev0_657 = __builtin_shufflevector(__s0_657, __s0_657, 7, 6, 5, 4, 3, 2, 1, 0); \
56970		int16x8_t __rev1_657; __rev1_657 = __builtin_shufflevector(__s1_657, __s1_657, 7, 6, 5, 4, 3, 2, 1, 0); \
56971		__ret_657 = __rev0_657 * __noswap_splatq_laneq_s16(__rev1_657, __p2_657); \
56972		__ret_657 = __builtin_shufflevector(__ret_657, __ret_657, 7, 6, 5, 4, 3, 2, 1, 0); \
56973		__ret_657; \
56974		})
56975		#endif
56976
56977		#ifdef __LITTLE_ENDIAN__
56978		#define vmul_laneq_u32(__p0_658, __p1_658, __p2_658) __extension__ ({ \
56979		uint32x2_t __ret_658; \
56980		uint32x2_t __s0_658 = __p0_658; \
56981		uint32x4_t __s1_658 = __p1_658; \
56982		__ret_658 = __s0_658 * splat_laneq_u32(__s1_658, __p2_658); \
56983		__ret_658; \
56984		})
56985		#else
56986		#define vmul_laneq_u32(__p0_659, __p1_659, __p2_659) __extension__ ({ \
56987		uint32x2_t __ret_659; \
56988		uint32x2_t __s0_659 = __p0_659; \
56989		uint32x4_t __s1_659 = __p1_659; \
56990		uint32x2_t __rev0_659; __rev0_659 = __builtin_shufflevector(__s0_659, __s0_659, 1, 0); \
56991		uint32x4_t __rev1_659; __rev1_659 = __builtin_shufflevector(__s1_659, __s1_659, 3, 2, 1, 0); \
56992		__ret_659 = __rev0_659 * __noswap_splat_laneq_u32(__rev1_659, __p2_659); \
56993		__ret_659 = __builtin_shufflevector(__ret_659, __ret_659, 1, 0); \
56994		__ret_659; \
56995		})
56996		#endif
56997
56998		#ifdef __LITTLE_ENDIAN__
56999		#define vmul_laneq_u16(__p0_660, __p1_660, __p2_660) __extension__ ({ \
57000		uint16x4_t __ret_660; \
57001		uint16x4_t __s0_660 = __p0_660; \
57002		uint16x8_t __s1_660 = __p1_660; \
57003		__ret_660 = __s0_660 * splat_laneq_u16(__s1_660, __p2_660); \
57004		__ret_660; \
57005		})
57006		#else
57007		#define vmul_laneq_u16(__p0_661, __p1_661, __p2_661) __extension__ ({ \
57008		uint16x4_t __ret_661; \
57009		uint16x4_t __s0_661 = __p0_661; \
57010		uint16x8_t __s1_661 = __p1_661; \
57011		uint16x4_t __rev0_661; __rev0_661 = __builtin_shufflevector(__s0_661, __s0_661, 3, 2, 1, 0); \
57012		uint16x8_t __rev1_661; __rev1_661 = __builtin_shufflevector(__s1_661, __s1_661, 7, 6, 5, 4, 3, 2, 1, 0); \
57013		__ret_661 = __rev0_661 * __noswap_splat_laneq_u16(__rev1_661, __p2_661); \
57014		__ret_661 = __builtin_shufflevector(__ret_661, __ret_661, 3, 2, 1, 0); \
57015		__ret_661; \
57016		})
57017		#endif
57018
57019		#ifdef __LITTLE_ENDIAN__
57020		#define vmul_laneq_f32(__p0_662, __p1_662, __p2_662) __extension__ ({ \
57021		float32x2_t __ret_662; \
57022		float32x2_t __s0_662 = __p0_662; \
57023		float32x4_t __s1_662 = __p1_662; \
57024		__ret_662 = __s0_662 * splat_laneq_f32(__s1_662, __p2_662); \
57025		__ret_662; \
57026		})
57027		#else
57028		#define vmul_laneq_f32(__p0_663, __p1_663, __p2_663) __extension__ ({ \
57029		float32x2_t __ret_663; \
57030		float32x2_t __s0_663 = __p0_663; \
57031		float32x4_t __s1_663 = __p1_663; \
57032		float32x2_t __rev0_663; __rev0_663 = __builtin_shufflevector(__s0_663, __s0_663, 1, 0); \
57033		float32x4_t __rev1_663; __rev1_663 = __builtin_shufflevector(__s1_663, __s1_663, 3, 2, 1, 0); \
57034		__ret_663 = __rev0_663 * __noswap_splat_laneq_f32(__rev1_663, __p2_663); \
57035		__ret_663 = __builtin_shufflevector(__ret_663, __ret_663, 1, 0); \
57036		__ret_663; \
57037		})
57038		#endif
57039
57040		#ifdef __LITTLE_ENDIAN__
57041		#define vmul_laneq_s32(__p0_664, __p1_664, __p2_664) __extension__ ({ \
57042		int32x2_t __ret_664; \
57043		int32x2_t __s0_664 = __p0_664; \
57044		int32x4_t __s1_664 = __p1_664; \
57045		__ret_664 = __s0_664 * splat_laneq_s32(__s1_664, __p2_664); \
57046		__ret_664; \
57047		})
57048		#else
57049		#define vmul_laneq_s32(__p0_665, __p1_665, __p2_665) __extension__ ({ \
57050		int32x2_t __ret_665; \
57051		int32x2_t __s0_665 = __p0_665; \
57052		int32x4_t __s1_665 = __p1_665; \
57053		int32x2_t __rev0_665; __rev0_665 = __builtin_shufflevector(__s0_665, __s0_665, 1, 0); \
57054		int32x4_t __rev1_665; __rev1_665 = __builtin_shufflevector(__s1_665, __s1_665, 3, 2, 1, 0); \
57055		__ret_665 = __rev0_665 * __noswap_splat_laneq_s32(__rev1_665, __p2_665); \
57056		__ret_665 = __builtin_shufflevector(__ret_665, __ret_665, 1, 0); \
57057		__ret_665; \
57058		})
57059		#endif
57060
57061		#ifdef __LITTLE_ENDIAN__
57062		#define vmul_laneq_s16(__p0_666, __p1_666, __p2_666) __extension__ ({ \
57063		int16x4_t __ret_666; \
57064		int16x4_t __s0_666 = __p0_666; \
57065		int16x8_t __s1_666 = __p1_666; \
57066		__ret_666 = __s0_666 * splat_laneq_s16(__s1_666, __p2_666); \
57067		__ret_666; \
57068		})
57069		#else
57070		#define vmul_laneq_s16(__p0_667, __p1_667, __p2_667) __extension__ ({ \
57071		int16x4_t __ret_667; \
57072		int16x4_t __s0_667 = __p0_667; \
57073		int16x8_t __s1_667 = __p1_667; \
57074		int16x4_t __rev0_667; __rev0_667 = __builtin_shufflevector(__s0_667, __s0_667, 3, 2, 1, 0); \
57075		int16x8_t __rev1_667; __rev1_667 = __builtin_shufflevector(__s1_667, __s1_667, 7, 6, 5, 4, 3, 2, 1, 0); \
57076		__ret_667 = __rev0_667 * __noswap_splat_laneq_s16(__rev1_667, __p2_667); \
57077		__ret_667 = __builtin_shufflevector(__ret_667, __ret_667, 3, 2, 1, 0); \
57078		__ret_667; \
57079		})
57080		#endif
57081
57082		__ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
57083		float64x1_t __ret;
57084		__ret = (float64x1_t) __builtin_neon_vmul_n_f64((float64x1_t)__p0, __p1);
	57517	__ai int16x8_t vreinterpretq_s16_f64(float64x2_t __p0) {
	57518	int16x8_t __ret;
	57519	__ret = (int16x8_t)(__p0);
	57520	return __ret;
	57521	}
	57522	__ai int16x8_t vreinterpretq_s16_f32(float32x4_t __p0) {
	57523	int16x8_t __ret;
	57524	__ret = (int16x8_t)(__p0);
	57525	return __ret;
	57526	}
	57527	__ai int16x8_t vreinterpretq_s16_f16(float16x8_t __p0) {
	57528	int16x8_t __ret;
	57529	__ret = (int16x8_t)(__p0);
	57530	return __ret;
	57531	}
	57532	__ai int16x8_t vreinterpretq_s16_s32(int32x4_t __p0) {
	57533	int16x8_t __ret;
	57534	__ret = (int16x8_t)(__p0);
	57535	return __ret;
	57536	}
	57537	__ai int16x8_t vreinterpretq_s16_s64(int64x2_t __p0) {
	57538	int16x8_t __ret;
	57539	__ret = (int16x8_t)(__p0);
	57540	return __ret;
	57541	}
	57542	__ai uint8x8_t vreinterpret_u8_p8(poly8x8_t __p0) {
	57543	uint8x8_t __ret;
	57544	__ret = (uint8x8_t)(__p0);
	57545	return __ret;
	57546	}
	57547	__ai uint8x8_t vreinterpret_u8_p64(poly64x1_t __p0) {
	57548	uint8x8_t __ret;
	57549	__ret = (uint8x8_t)(__p0);
	57550	return __ret;
	57551	}
	57552	__ai uint8x8_t vreinterpret_u8_p16(poly16x4_t __p0) {
	57553	uint8x8_t __ret;
	57554	__ret = (uint8x8_t)(__p0);
	57555	return __ret;
	57556	}
	57557	__ai uint8x8_t vreinterpret_u8_u32(uint32x2_t __p0) {
	57558	uint8x8_t __ret;
	57559	__ret = (uint8x8_t)(__p0);
57085	57560	return __ret;
57086	57561	}
57087		#ifdef __LITTLE_ENDIAN__
57088		__ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
57089		float64x2_t __ret;
57090		__ret = __p0 * (float64x2_t) {__p1, __p1};
	57562	__ai uint8x8_t vreinterpret_u8_u64(uint64x1_t __p0) {
	57563	uint8x8_t __ret;
	57564	__ret = (uint8x8_t)(__p0);
57091	57565	return __ret;
57092	57566	}
57093		#else
57094		__ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
57095		float64x2_t __ret;
57096		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57097		__ret = __rev0 * (float64x2_t) {__p1, __p1};
57098		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57567	__ai uint8x8_t vreinterpret_u8_u16(uint16x4_t __p0) {
	57568	uint8x8_t __ret;
	57569	__ret = (uint8x8_t)(__p0);
57099	57570	return __ret;
57100	57571	}
57101		#endif
57102
57103		__ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
57104		poly128_t __ret;
57105		__ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
	57572	__ai uint8x8_t vreinterpret_u8_s8(int8x8_t __p0) {
	57573	uint8x8_t __ret;
	57574	__ret = (uint8x8_t)(__p0);
57106	57575	return __ret;
57107	57576	}
57108		#ifdef __LITTLE_ENDIAN__
57109		__ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
57110		poly16x8_t __ret;
57111		__ret = vmull_p8(vget_high_p8(__p0), vget_high_p8(__p1));
	57577	__ai uint8x8_t vreinterpret_u8_f64(float64x1_t __p0) {
	57578	uint8x8_t __ret;
	57579	__ret = (uint8x8_t)(__p0);
57112	57580	return __ret;
57113	57581	}
57114		#else
57115		__ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
57116		poly16x8_t __ret;
57117		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57118		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57119		__ret = __noswap_vmull_p8(__noswap_vget_high_p8(__rev0), __noswap_vget_high_p8(__rev1));
57120		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57582	__ai uint8x8_t vreinterpret_u8_f32(float32x2_t __p0) {
	57583	uint8x8_t __ret;
	57584	__ret = (uint8x8_t)(__p0);
57121	57585	return __ret;
57122	57586	}
57123		#endif
57124
57125		#ifdef __LITTLE_ENDIAN__
57126		__ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
57127		uint16x8_t __ret;
57128		__ret = vmull_u8(vget_high_u8(__p0), vget_high_u8(__p1));
	57587	__ai uint8x8_t vreinterpret_u8_f16(float16x4_t __p0) {
	57588	uint8x8_t __ret;
	57589	__ret = (uint8x8_t)(__p0);
57129	57590	return __ret;
57130	57591	}
57131		#else
57132		__ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
57133		uint16x8_t __ret;
57134		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57135		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57136		__ret = __noswap_vmull_u8(__noswap_vget_high_u8(__rev0), __noswap_vget_high_u8(__rev1));
57137		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57592	__ai uint8x8_t vreinterpret_u8_s32(int32x2_t __p0) {
	57593	uint8x8_t __ret;
	57594	__ret = (uint8x8_t)(__p0);
57138	57595	return __ret;
57139	57596	}
57140		#endif
57141
57142		#ifdef __LITTLE_ENDIAN__
57143		__ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
57144		uint64x2_t __ret;
57145		__ret = vmull_u32(vget_high_u32(__p0), vget_high_u32(__p1));
	57597	__ai uint8x8_t vreinterpret_u8_s64(int64x1_t __p0) {
	57598	uint8x8_t __ret;
	57599	__ret = (uint8x8_t)(__p0);
57146	57600	return __ret;
57147	57601	}
57148		#else
57149		__ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
57150		uint64x2_t __ret;
57151		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57152		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57153		__ret = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0), __noswap_vget_high_u32(__rev1));
57154		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57602	__ai uint8x8_t vreinterpret_u8_s16(int16x4_t __p0) {
	57603	uint8x8_t __ret;
	57604	__ret = (uint8x8_t)(__p0);
57155	57605	return __ret;
57156	57606	}
57157		#endif
57158
57159		#ifdef __LITTLE_ENDIAN__
57160		__ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
57161		uint32x4_t __ret;
57162		__ret = vmull_u16(vget_high_u16(__p0), vget_high_u16(__p1));
	57607	__ai uint32x2_t vreinterpret_u32_p8(poly8x8_t __p0) {
	57608	uint32x2_t __ret;
	57609	__ret = (uint32x2_t)(__p0);
57163	57610	return __ret;
57164	57611	}
57165		#else
57166		__ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
57167		uint32x4_t __ret;
57168		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
57169		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57170		__ret = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0), __noswap_vget_high_u16(__rev1));
57171		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57612	__ai uint32x2_t vreinterpret_u32_p64(poly64x1_t __p0) {
	57613	uint32x2_t __ret;
	57614	__ret = (uint32x2_t)(__p0);
57172	57615	return __ret;
57173	57616	}
57174		#endif
57175
57176		#ifdef __LITTLE_ENDIAN__
57177		__ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
57178		int16x8_t __ret;
57179		__ret = vmull_s8(vget_high_s8(__p0), vget_high_s8(__p1));
	57617	__ai uint32x2_t vreinterpret_u32_p16(poly16x4_t __p0) {
	57618	uint32x2_t __ret;
	57619	__ret = (uint32x2_t)(__p0);
57180	57620	return __ret;
57181	57621	}
57182		#else
57183		__ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
57184		int16x8_t __ret;
57185		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57186		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57187		__ret = __noswap_vmull_s8(__noswap_vget_high_s8(__rev0), __noswap_vget_high_s8(__rev1));
57188		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57622	__ai uint32x2_t vreinterpret_u32_u8(uint8x8_t __p0) {
	57623	uint32x2_t __ret;
	57624	__ret = (uint32x2_t)(__p0);
57189	57625	return __ret;
57190	57626	}
57191		#endif
57192
57193		#ifdef __LITTLE_ENDIAN__
57194		__ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
57195		int64x2_t __ret;
57196		__ret = vmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
	57627	__ai uint32x2_t vreinterpret_u32_u64(uint64x1_t __p0) {
	57628	uint32x2_t __ret;
	57629	__ret = (uint32x2_t)(__p0);
57197	57630	return __ret;
57198	57631	}
57199		#else
57200		__ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
57201		int64x2_t __ret;
57202		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57203		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57204		__ret = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
57205		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57632	__ai uint32x2_t vreinterpret_u32_u16(uint16x4_t __p0) {
	57633	uint32x2_t __ret;
	57634	__ret = (uint32x2_t)(__p0);
57206	57635	return __ret;
57207	57636	}
57208		#endif
57209
57210		#ifdef __LITTLE_ENDIAN__
57211		__ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
57212		int32x4_t __ret;
57213		__ret = vmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
	57637	__ai uint32x2_t vreinterpret_u32_s8(int8x8_t __p0) {
	57638	uint32x2_t __ret;
	57639	__ret = (uint32x2_t)(__p0);
57214	57640	return __ret;
57215	57641	}
57216		#else
57217		__ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
57218		int32x4_t __ret;
57219		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
57220		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57221		__ret = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
57222		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57642	__ai uint32x2_t vreinterpret_u32_f64(float64x1_t __p0) {
	57643	uint32x2_t __ret;
	57644	__ret = (uint32x2_t)(__p0);
57223	57645	return __ret;
57224	57646	}
57225		#endif
57226
57227		#ifdef __LITTLE_ENDIAN__
57228		__ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
57229		poly128_t __ret;
57230		__ret = vmull_p64((poly64_t)(vget_high_p64(__p0)), (poly64_t)(vget_high_p64(__p1)));
	57647	__ai uint32x2_t vreinterpret_u32_f32(float32x2_t __p0) {
	57648	uint32x2_t __ret;
	57649	__ret = (uint32x2_t)(__p0);
57231	57650	return __ret;
57232	57651	}
57233		#else
57234		__ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
57235		poly128_t __ret;
57236		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57237		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57238		__ret = vmull_p64((poly64_t)(__noswap_vget_high_p64(__rev0)), (poly64_t)(__noswap_vget_high_p64(__rev1)));
	57652	__ai uint32x2_t vreinterpret_u32_f16(float16x4_t __p0) {
	57653	uint32x2_t __ret;
	57654	__ret = (uint32x2_t)(__p0);
57239	57655	return __ret;
57240	57656	}
57241		#endif
57242
57243		#ifdef __LITTLE_ENDIAN__
57244		#define vmull_high_lane_u32(__p0_668, __p1_668, __p2_668) __extension__ ({ \
57245		uint64x2_t __ret_668; \
57246		uint32x4_t __s0_668 = __p0_668; \
57247		uint32x2_t __s1_668 = __p1_668; \
57248		__ret_668 = vmull_u32(vget_high_u32(__s0_668), splat_lane_u32(__s1_668, __p2_668)); \
57249		__ret_668; \
57250		})
57251		#else
57252		#define vmull_high_lane_u32(__p0_669, __p1_669, __p2_669) __extension__ ({ \
57253		uint64x2_t __ret_669; \
57254		uint32x4_t __s0_669 = __p0_669; \
57255		uint32x2_t __s1_669 = __p1_669; \
57256		uint32x4_t __rev0_669; __rev0_669 = __builtin_shufflevector(__s0_669, __s0_669, 3, 2, 1, 0); \
57257		uint32x2_t __rev1_669; __rev1_669 = __builtin_shufflevector(__s1_669, __s1_669, 1, 0); \
57258		__ret_669 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_669), __noswap_splat_lane_u32(__rev1_669, __p2_669)); \
57259		__ret_669 = __builtin_shufflevector(__ret_669, __ret_669, 1, 0); \
57260		__ret_669; \
57261		})
57262		#endif
57263
57264		#ifdef __LITTLE_ENDIAN__
57265		#define vmull_high_lane_u16(__p0_670, __p1_670, __p2_670) __extension__ ({ \
57266		uint32x4_t __ret_670; \
57267		uint16x8_t __s0_670 = __p0_670; \
57268		uint16x4_t __s1_670 = __p1_670; \
57269		__ret_670 = vmull_u16(vget_high_u16(__s0_670), splat_lane_u16(__s1_670, __p2_670)); \
57270		__ret_670; \
57271		})
57272		#else
57273		#define vmull_high_lane_u16(__p0_671, __p1_671, __p2_671) __extension__ ({ \
57274		uint32x4_t __ret_671; \
57275		uint16x8_t __s0_671 = __p0_671; \
57276		uint16x4_t __s1_671 = __p1_671; \
57277		uint16x8_t __rev0_671; __rev0_671 = __builtin_shufflevector(__s0_671, __s0_671, 7, 6, 5, 4, 3, 2, 1, 0); \
57278		uint16x4_t __rev1_671; __rev1_671 = __builtin_shufflevector(__s1_671, __s1_671, 3, 2, 1, 0); \
57279		__ret_671 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_671), __noswap_splat_lane_u16(__rev1_671, __p2_671)); \
57280		__ret_671 = __builtin_shufflevector(__ret_671, __ret_671, 3, 2, 1, 0); \
57281		__ret_671; \
57282		})
57283		#endif
57284
57285		#ifdef __LITTLE_ENDIAN__
57286		#define vmull_high_lane_s32(__p0_672, __p1_672, __p2_672) __extension__ ({ \
57287		int64x2_t __ret_672; \
57288		int32x4_t __s0_672 = __p0_672; \
57289		int32x2_t __s1_672 = __p1_672; \
57290		__ret_672 = vmull_s32(vget_high_s32(__s0_672), splat_lane_s32(__s1_672, __p2_672)); \
57291		__ret_672; \
57292		})
57293		#else
57294		#define vmull_high_lane_s32(__p0_673, __p1_673, __p2_673) __extension__ ({ \
57295		int64x2_t __ret_673; \
57296		int32x4_t __s0_673 = __p0_673; \
57297		int32x2_t __s1_673 = __p1_673; \
57298		int32x4_t __rev0_673; __rev0_673 = __builtin_shufflevector(__s0_673, __s0_673, 3, 2, 1, 0); \
57299		int32x2_t __rev1_673; __rev1_673 = __builtin_shufflevector(__s1_673, __s1_673, 1, 0); \
57300		__ret_673 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_673), __noswap_splat_lane_s32(__rev1_673, __p2_673)); \
57301		__ret_673 = __builtin_shufflevector(__ret_673, __ret_673, 1, 0); \
57302		__ret_673; \
57303		})
57304		#endif
57305
57306		#ifdef __LITTLE_ENDIAN__
57307		#define vmull_high_lane_s16(__p0_674, __p1_674, __p2_674) __extension__ ({ \
57308		int32x4_t __ret_674; \
57309		int16x8_t __s0_674 = __p0_674; \
57310		int16x4_t __s1_674 = __p1_674; \
57311		__ret_674 = vmull_s16(vget_high_s16(__s0_674), splat_lane_s16(__s1_674, __p2_674)); \
57312		__ret_674; \
57313		})
57314		#else
57315		#define vmull_high_lane_s16(__p0_675, __p1_675, __p2_675) __extension__ ({ \
57316		int32x4_t __ret_675; \
57317		int16x8_t __s0_675 = __p0_675; \
57318		int16x4_t __s1_675 = __p1_675; \
57319		int16x8_t __rev0_675; __rev0_675 = __builtin_shufflevector(__s0_675, __s0_675, 7, 6, 5, 4, 3, 2, 1, 0); \
57320		int16x4_t __rev1_675; __rev1_675 = __builtin_shufflevector(__s1_675, __s1_675, 3, 2, 1, 0); \
57321		__ret_675 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_675), __noswap_splat_lane_s16(__rev1_675, __p2_675)); \
57322		__ret_675 = __builtin_shufflevector(__ret_675, __ret_675, 3, 2, 1, 0); \
57323		__ret_675; \
57324		})
57325		#endif
57326
57327		#ifdef __LITTLE_ENDIAN__
57328		#define vmull_high_laneq_u32(__p0_676, __p1_676, __p2_676) __extension__ ({ \
57329		uint64x2_t __ret_676; \
57330		uint32x4_t __s0_676 = __p0_676; \
57331		uint32x4_t __s1_676 = __p1_676; \
57332		__ret_676 = vmull_u32(vget_high_u32(__s0_676), splat_laneq_u32(__s1_676, __p2_676)); \
57333		__ret_676; \
57334		})
57335		#else
57336		#define vmull_high_laneq_u32(__p0_677, __p1_677, __p2_677) __extension__ ({ \
57337		uint64x2_t __ret_677; \
57338		uint32x4_t __s0_677 = __p0_677; \
57339		uint32x4_t __s1_677 = __p1_677; \
57340		uint32x4_t __rev0_677; __rev0_677 = __builtin_shufflevector(__s0_677, __s0_677, 3, 2, 1, 0); \
57341		uint32x4_t __rev1_677; __rev1_677 = __builtin_shufflevector(__s1_677, __s1_677, 3, 2, 1, 0); \
57342		__ret_677 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_677), __noswap_splat_laneq_u32(__rev1_677, __p2_677)); \
57343		__ret_677 = __builtin_shufflevector(__ret_677, __ret_677, 1, 0); \
57344		__ret_677; \
57345		})
57346		#endif
57347
57348		#ifdef __LITTLE_ENDIAN__
57349		#define vmull_high_laneq_u16(__p0_678, __p1_678, __p2_678) __extension__ ({ \
57350		uint32x4_t __ret_678; \
57351		uint16x8_t __s0_678 = __p0_678; \
57352		uint16x8_t __s1_678 = __p1_678; \
57353		__ret_678 = vmull_u16(vget_high_u16(__s0_678), splat_laneq_u16(__s1_678, __p2_678)); \
57354		__ret_678; \
57355		})
57356		#else
57357		#define vmull_high_laneq_u16(__p0_679, __p1_679, __p2_679) __extension__ ({ \
57358		uint32x4_t __ret_679; \
57359		uint16x8_t __s0_679 = __p0_679; \
57360		uint16x8_t __s1_679 = __p1_679; \
57361		uint16x8_t __rev0_679; __rev0_679 = __builtin_shufflevector(__s0_679, __s0_679, 7, 6, 5, 4, 3, 2, 1, 0); \
57362		uint16x8_t __rev1_679; __rev1_679 = __builtin_shufflevector(__s1_679, __s1_679, 7, 6, 5, 4, 3, 2, 1, 0); \
57363		__ret_679 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_679), __noswap_splat_laneq_u16(__rev1_679, __p2_679)); \
57364		__ret_679 = __builtin_shufflevector(__ret_679, __ret_679, 3, 2, 1, 0); \
57365		__ret_679; \
57366		})
57367		#endif
57368
57369		#ifdef __LITTLE_ENDIAN__
57370		#define vmull_high_laneq_s32(__p0_680, __p1_680, __p2_680) __extension__ ({ \
57371		int64x2_t __ret_680; \
57372		int32x4_t __s0_680 = __p0_680; \
57373		int32x4_t __s1_680 = __p1_680; \
57374		__ret_680 = vmull_s32(vget_high_s32(__s0_680), splat_laneq_s32(__s1_680, __p2_680)); \
57375		__ret_680; \
57376		})
57377		#else
57378		#define vmull_high_laneq_s32(__p0_681, __p1_681, __p2_681) __extension__ ({ \
57379		int64x2_t __ret_681; \
57380		int32x4_t __s0_681 = __p0_681; \
57381		int32x4_t __s1_681 = __p1_681; \
57382		int32x4_t __rev0_681; __rev0_681 = __builtin_shufflevector(__s0_681, __s0_681, 3, 2, 1, 0); \
57383		int32x4_t __rev1_681; __rev1_681 = __builtin_shufflevector(__s1_681, __s1_681, 3, 2, 1, 0); \
57384		__ret_681 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_681), __noswap_splat_laneq_s32(__rev1_681, __p2_681)); \
57385		__ret_681 = __builtin_shufflevector(__ret_681, __ret_681, 1, 0); \
57386		__ret_681; \
57387		})
57388		#endif
57389
57390		#ifdef __LITTLE_ENDIAN__
57391		#define vmull_high_laneq_s16(__p0_682, __p1_682, __p2_682) __extension__ ({ \
57392		int32x4_t __ret_682; \
57393		int16x8_t __s0_682 = __p0_682; \
57394		int16x8_t __s1_682 = __p1_682; \
57395		__ret_682 = vmull_s16(vget_high_s16(__s0_682), splat_laneq_s16(__s1_682, __p2_682)); \
57396		__ret_682; \
57397		})
57398		#else
57399		#define vmull_high_laneq_s16(__p0_683, __p1_683, __p2_683) __extension__ ({ \
57400		int32x4_t __ret_683; \
57401		int16x8_t __s0_683 = __p0_683; \
57402		int16x8_t __s1_683 = __p1_683; \
57403		int16x8_t __rev0_683; __rev0_683 = __builtin_shufflevector(__s0_683, __s0_683, 7, 6, 5, 4, 3, 2, 1, 0); \
57404		int16x8_t __rev1_683; __rev1_683 = __builtin_shufflevector(__s1_683, __s1_683, 7, 6, 5, 4, 3, 2, 1, 0); \
57405		__ret_683 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_683), __noswap_splat_laneq_s16(__rev1_683, __p2_683)); \
57406		__ret_683 = __builtin_shufflevector(__ret_683, __ret_683, 3, 2, 1, 0); \
57407		__ret_683; \
57408		})
57409		#endif
57410
57411		#ifdef __LITTLE_ENDIAN__
57412		__ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
57413		uint64x2_t __ret;
57414		__ret = vmull_n_u32(vget_high_u32(__p0), __p1);
	57657	__ai uint32x2_t vreinterpret_u32_s32(int32x2_t __p0) {
	57658	uint32x2_t __ret;
	57659	__ret = (uint32x2_t)(__p0);
57415	57660	return __ret;
57416	57661	}
57417		#else
57418		__ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
57419		uint64x2_t __ret;
57420		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57421		__ret = __noswap_vmull_n_u32(__noswap_vget_high_u32(__rev0), __p1);
57422		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57662	__ai uint32x2_t vreinterpret_u32_s64(int64x1_t __p0) {
	57663	uint32x2_t __ret;
	57664	__ret = (uint32x2_t)(__p0);
57423	57665	return __ret;
57424	57666	}
57425		#endif
57426
57427		#ifdef __LITTLE_ENDIAN__
57428		__ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
57429		uint32x4_t __ret;
57430		__ret = vmull_n_u16(vget_high_u16(__p0), __p1);
	57667	__ai uint32x2_t vreinterpret_u32_s16(int16x4_t __p0) {
	57668	uint32x2_t __ret;
	57669	__ret = (uint32x2_t)(__p0);
57431	57670	return __ret;
57432	57671	}
57433		#else
57434		__ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
57435		uint32x4_t __ret;
57436		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
57437		__ret = __noswap_vmull_n_u16(__noswap_vget_high_u16(__rev0), __p1);
57438		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57672	__ai uint64x1_t vreinterpret_u64_p8(poly8x8_t __p0) {
	57673	uint64x1_t __ret;
	57674	__ret = (uint64x1_t)(__p0);
57439	57675	return __ret;
57440	57676	}
57441		#endif
57442
57443		#ifdef __LITTLE_ENDIAN__
57444		__ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
57445		int64x2_t __ret;
57446		__ret = vmull_n_s32(vget_high_s32(__p0), __p1);
	57677	__ai uint64x1_t vreinterpret_u64_p64(poly64x1_t __p0) {
	57678	uint64x1_t __ret;
	57679	__ret = (uint64x1_t)(__p0);
57447	57680	return __ret;
57448	57681	}
57449		#else
57450		__ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
57451		int64x2_t __ret;
57452		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57453		__ret = __noswap_vmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
57454		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57682	__ai uint64x1_t vreinterpret_u64_p16(poly16x4_t __p0) {
	57683	uint64x1_t __ret;
	57684	__ret = (uint64x1_t)(__p0);
57455	57685	return __ret;
57456	57686	}
57457		#endif
57458
57459		#ifdef __LITTLE_ENDIAN__
57460		__ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
57461		int32x4_t __ret;
57462		__ret = vmull_n_s16(vget_high_s16(__p0), __p1);
	57687	__ai uint64x1_t vreinterpret_u64_u8(uint8x8_t __p0) {
	57688	uint64x1_t __ret;
	57689	__ret = (uint64x1_t)(__p0);
57463	57690	return __ret;
57464	57691	}
57465		#else
57466		__ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
57467		int32x4_t __ret;
57468		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
57469		__ret = __noswap_vmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
57470		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57692	__ai uint64x1_t vreinterpret_u64_u32(uint32x2_t __p0) {
	57693	uint64x1_t __ret;
	57694	__ret = (uint64x1_t)(__p0);
57471	57695	return __ret;
57472	57696	}
57473		#endif
57474
57475		#ifdef __LITTLE_ENDIAN__
57476		#define vmull_laneq_u32(__p0_684, __p1_684, __p2_684) __extension__ ({ \
57477		uint64x2_t __ret_684; \
57478		uint32x2_t __s0_684 = __p0_684; \
57479		uint32x4_t __s1_684 = __p1_684; \
57480		__ret_684 = vmull_u32(__s0_684, splat_laneq_u32(__s1_684, __p2_684)); \
57481		__ret_684; \
57482		})
57483		#else
57484		#define vmull_laneq_u32(__p0_685, __p1_685, __p2_685) __extension__ ({ \
57485		uint64x2_t __ret_685; \
57486		uint32x2_t __s0_685 = __p0_685; \
57487		uint32x4_t __s1_685 = __p1_685; \
57488		uint32x2_t __rev0_685; __rev0_685 = __builtin_shufflevector(__s0_685, __s0_685, 1, 0); \
57489		uint32x4_t __rev1_685; __rev1_685 = __builtin_shufflevector(__s1_685, __s1_685, 3, 2, 1, 0); \
57490		__ret_685 = __noswap_vmull_u32(__rev0_685, __noswap_splat_laneq_u32(__rev1_685, __p2_685)); \
57491		__ret_685 = __builtin_shufflevector(__ret_685, __ret_685, 1, 0); \
57492		__ret_685; \
57493		})
57494		#endif
57495
57496		#ifdef __LITTLE_ENDIAN__
57497		#define vmull_laneq_u16(__p0_686, __p1_686, __p2_686) __extension__ ({ \
57498		uint32x4_t __ret_686; \
57499		uint16x4_t __s0_686 = __p0_686; \
57500		uint16x8_t __s1_686 = __p1_686; \
57501		__ret_686 = vmull_u16(__s0_686, splat_laneq_u16(__s1_686, __p2_686)); \
57502		__ret_686; \
57503		})
57504		#else
57505		#define vmull_laneq_u16(__p0_687, __p1_687, __p2_687) __extension__ ({ \
57506		uint32x4_t __ret_687; \
57507		uint16x4_t __s0_687 = __p0_687; \
57508		uint16x8_t __s1_687 = __p1_687; \
57509		uint16x4_t __rev0_687; __rev0_687 = __builtin_shufflevector(__s0_687, __s0_687, 3, 2, 1, 0); \
57510		uint16x8_t __rev1_687; __rev1_687 = __builtin_shufflevector(__s1_687, __s1_687, 7, 6, 5, 4, 3, 2, 1, 0); \
57511		__ret_687 = __noswap_vmull_u16(__rev0_687, __noswap_splat_laneq_u16(__rev1_687, __p2_687)); \
57512		__ret_687 = __builtin_shufflevector(__ret_687, __ret_687, 3, 2, 1, 0); \
57513		__ret_687; \
57514		})
57515		#endif
57516
57517		#ifdef __LITTLE_ENDIAN__
57518		#define vmull_laneq_s32(__p0_688, __p1_688, __p2_688) __extension__ ({ \
57519		int64x2_t __ret_688; \
57520		int32x2_t __s0_688 = __p0_688; \
57521		int32x4_t __s1_688 = __p1_688; \
57522		__ret_688 = vmull_s32(__s0_688, splat_laneq_s32(__s1_688, __p2_688)); \
57523		__ret_688; \
57524		})
57525		#else
57526		#define vmull_laneq_s32(__p0_689, __p1_689, __p2_689) __extension__ ({ \
57527		int64x2_t __ret_689; \
57528		int32x2_t __s0_689 = __p0_689; \
57529		int32x4_t __s1_689 = __p1_689; \
57530		int32x2_t __rev0_689; __rev0_689 = __builtin_shufflevector(__s0_689, __s0_689, 1, 0); \
57531		int32x4_t __rev1_689; __rev1_689 = __builtin_shufflevector(__s1_689, __s1_689, 3, 2, 1, 0); \
57532		__ret_689 = __noswap_vmull_s32(__rev0_689, __noswap_splat_laneq_s32(__rev1_689, __p2_689)); \
57533		__ret_689 = __builtin_shufflevector(__ret_689, __ret_689, 1, 0); \
57534		__ret_689; \
57535		})
57536		#endif
57537
57538		#ifdef __LITTLE_ENDIAN__
57539		#define vmull_laneq_s16(__p0_690, __p1_690, __p2_690) __extension__ ({ \
57540		int32x4_t __ret_690; \
57541		int16x4_t __s0_690 = __p0_690; \
57542		int16x8_t __s1_690 = __p1_690; \
57543		__ret_690 = vmull_s16(__s0_690, splat_laneq_s16(__s1_690, __p2_690)); \
57544		__ret_690; \
57545		})
57546		#else
57547		#define vmull_laneq_s16(__p0_691, __p1_691, __p2_691) __extension__ ({ \
57548		int32x4_t __ret_691; \
57549		int16x4_t __s0_691 = __p0_691; \
57550		int16x8_t __s1_691 = __p1_691; \
57551		int16x4_t __rev0_691; __rev0_691 = __builtin_shufflevector(__s0_691, __s0_691, 3, 2, 1, 0); \
57552		int16x8_t __rev1_691; __rev1_691 = __builtin_shufflevector(__s1_691, __s1_691, 7, 6, 5, 4, 3, 2, 1, 0); \
57553		__ret_691 = __noswap_vmull_s16(__rev0_691, __noswap_splat_laneq_s16(__rev1_691, __p2_691)); \
57554		__ret_691 = __builtin_shufflevector(__ret_691, __ret_691, 3, 2, 1, 0); \
57555		__ret_691; \
57556		})
57557		#endif
57558
57559		#ifdef __LITTLE_ENDIAN__
57560		__ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
57561		float64x2_t __ret;
57562		__ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	57697	__ai uint64x1_t vreinterpret_u64_u16(uint16x4_t __p0) {
	57698	uint64x1_t __ret;
	57699	__ret = (uint64x1_t)(__p0);
57563	57700	return __ret;
57564	57701	}
57565		#else
57566		__ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
57567		float64x2_t __ret;
57568		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57569		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57570		__ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
57571		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57702	__ai uint64x1_t vreinterpret_u64_s8(int8x8_t __p0) {
	57703	uint64x1_t __ret;
	57704	__ret = (uint64x1_t)(__p0);
57572	57705	return __ret;
57573	57706	}
57574		__ai float64x2_t __noswap_vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
57575		float64x2_t __ret;
57576		__ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	57707	__ai uint64x1_t vreinterpret_u64_f64(float64x1_t __p0) {
	57708	uint64x1_t __ret;
	57709	__ret = (uint64x1_t)(__p0);
57577	57710	return __ret;
57578	57711	}
57579		#endif
57580
57581		#ifdef __LITTLE_ENDIAN__
57582		__ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
57583		float32x4_t __ret;
57584		__ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	57712	__ai uint64x1_t vreinterpret_u64_f32(float32x2_t __p0) {
	57713	uint64x1_t __ret;
	57714	__ret = (uint64x1_t)(__p0);
57585	57715	return __ret;
57586	57716	}
57587		#else
57588		__ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
57589		float32x4_t __ret;
57590		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57591		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57592		__ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
57593		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57717	__ai uint64x1_t vreinterpret_u64_f16(float16x4_t __p0) {
	57718	uint64x1_t __ret;
	57719	__ret = (uint64x1_t)(__p0);
57594	57720	return __ret;
57595	57721	}
57596		__ai float32x4_t __noswap_vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
57597		float32x4_t __ret;
57598		__ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	57722	__ai uint64x1_t vreinterpret_u64_s32(int32x2_t __p0) {
	57723	uint64x1_t __ret;
	57724	__ret = (uint64x1_t)(__p0);
57599	57725	return __ret;
57600	57726	}
57601		#endif
57602
57603		__ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
57604		float64x1_t __ret;
57605		__ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	57727	__ai uint64x1_t vreinterpret_u64_s64(int64x1_t __p0) {
	57728	uint64x1_t __ret;
	57729	__ret = (uint64x1_t)(__p0);
57606	57730	return __ret;
57607	57731	}
57608		#ifdef __LITTLE_ENDIAN__
57609		__ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
57610		float32x2_t __ret;
57611		__ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	57732	__ai uint64x1_t vreinterpret_u64_s16(int16x4_t __p0) {
	57733	uint64x1_t __ret;
	57734	__ret = (uint64x1_t)(__p0);
57612	57735	return __ret;
57613	57736	}
57614		#else
57615		__ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
57616		float32x2_t __ret;
57617		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57618		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57619		__ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
57620		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57737	__ai uint16x4_t vreinterpret_u16_p8(poly8x8_t __p0) {
	57738	uint16x4_t __ret;
	57739	__ret = (uint16x4_t)(__p0);
57621	57740	return __ret;
57622	57741	}
57623		__ai float32x2_t __noswap_vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
57624		float32x2_t __ret;
57625		__ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	57742	__ai uint16x4_t vreinterpret_u16_p64(poly64x1_t __p0) {
	57743	uint16x4_t __ret;
	57744	__ret = (uint16x4_t)(__p0);
	57745	return __ret;
	57746	}
	57747	__ai uint16x4_t vreinterpret_u16_p16(poly16x4_t __p0) {
	57748	uint16x4_t __ret;
	57749	__ret = (uint16x4_t)(__p0);
	57750	return __ret;
	57751	}
	57752	__ai uint16x4_t vreinterpret_u16_u8(uint8x8_t __p0) {
	57753	uint16x4_t __ret;
	57754	__ret = (uint16x4_t)(__p0);
	57755	return __ret;
	57756	}
	57757	__ai uint16x4_t vreinterpret_u16_u32(uint32x2_t __p0) {
	57758	uint16x4_t __ret;
	57759	__ret = (uint16x4_t)(__p0);
	57760	return __ret;
	57761	}
	57762	__ai uint16x4_t vreinterpret_u16_u64(uint64x1_t __p0) {
	57763	uint16x4_t __ret;
	57764	__ret = (uint16x4_t)(__p0);
	57765	return __ret;
	57766	}
	57767	__ai uint16x4_t vreinterpret_u16_s8(int8x8_t __p0) {
	57768	uint16x4_t __ret;
	57769	__ret = (uint16x4_t)(__p0);
	57770	return __ret;
	57771	}
	57772	__ai uint16x4_t vreinterpret_u16_f64(float64x1_t __p0) {
	57773	uint16x4_t __ret;
	57774	__ret = (uint16x4_t)(__p0);
	57775	return __ret;
	57776	}
	57777	__ai uint16x4_t vreinterpret_u16_f32(float32x2_t __p0) {
	57778	uint16x4_t __ret;
	57779	__ret = (uint16x4_t)(__p0);
	57780	return __ret;
	57781	}
	57782	__ai uint16x4_t vreinterpret_u16_f16(float16x4_t __p0) {
	57783	uint16x4_t __ret;
	57784	__ret = (uint16x4_t)(__p0);
	57785	return __ret;
	57786	}
	57787	__ai uint16x4_t vreinterpret_u16_s32(int32x2_t __p0) {
	57788	uint16x4_t __ret;
	57789	__ret = (uint16x4_t)(__p0);
	57790	return __ret;
	57791	}
	57792	__ai uint16x4_t vreinterpret_u16_s64(int64x1_t __p0) {
	57793	uint16x4_t __ret;
	57794	__ret = (uint16x4_t)(__p0);
	57795	return __ret;
	57796	}
	57797	__ai uint16x4_t vreinterpret_u16_s16(int16x4_t __p0) {
	57798	uint16x4_t __ret;
	57799	__ret = (uint16x4_t)(__p0);
	57800	return __ret;
	57801	}
	57802	__ai int8x8_t vreinterpret_s8_p8(poly8x8_t __p0) {
	57803	int8x8_t __ret;
	57804	__ret = (int8x8_t)(__p0);
	57805	return __ret;
	57806	}
	57807	__ai int8x8_t vreinterpret_s8_p64(poly64x1_t __p0) {
	57808	int8x8_t __ret;
	57809	__ret = (int8x8_t)(__p0);
	57810	return __ret;
	57811	}
	57812	__ai int8x8_t vreinterpret_s8_p16(poly16x4_t __p0) {
	57813	int8x8_t __ret;
	57814	__ret = (int8x8_t)(__p0);
	57815	return __ret;
	57816	}
	57817	__ai int8x8_t vreinterpret_s8_u8(uint8x8_t __p0) {
	57818	int8x8_t __ret;
	57819	__ret = (int8x8_t)(__p0);
	57820	return __ret;
	57821	}
	57822	__ai int8x8_t vreinterpret_s8_u32(uint32x2_t __p0) {
	57823	int8x8_t __ret;
	57824	__ret = (int8x8_t)(__p0);
	57825	return __ret;
	57826	}
	57827	__ai int8x8_t vreinterpret_s8_u64(uint64x1_t __p0) {
	57828	int8x8_t __ret;
	57829	__ret = (int8x8_t)(__p0);
57626	57830	return __ret;
57627	57831	}
57628		#endif
57629
57630		__ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
57631		float64_t __ret;
57632		__ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
	57832	__ai int8x8_t vreinterpret_s8_u16(uint16x4_t __p0) {
	57833	int8x8_t __ret;
	57834	__ret = (int8x8_t)(__p0);
57633	57835	return __ret;
57634	57836	}
57635		__ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
57636		float32_t __ret;
57637		__ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
	57837	__ai int8x8_t vreinterpret_s8_f64(float64x1_t __p0) {
	57838	int8x8_t __ret;
	57839	__ret = (int8x8_t)(__p0);
57638	57840	return __ret;
57639	57841	}
57640		#define vmulxd_lane_f64(__p0_692, __p1_692, __p2_692) __extension__ ({ \
57641		float64_t __ret_692; \
57642		float64_t __s0_692 = __p0_692; \
57643		float64x1_t __s1_692 = __p1_692; \
57644		__ret_692 = vmulxd_f64(__s0_692, vget_lane_f64(__s1_692, __p2_692)); \
57645		__ret_692; \
57646		})
57647		#ifdef __LITTLE_ENDIAN__
57648		#define vmulxs_lane_f32(__p0_693, __p1_693, __p2_693) __extension__ ({ \
57649		float32_t __ret_693; \
57650		float32_t __s0_693 = __p0_693; \
57651		float32x2_t __s1_693 = __p1_693; \
57652		__ret_693 = vmulxs_f32(__s0_693, vget_lane_f32(__s1_693, __p2_693)); \
57653		__ret_693; \
57654		})
57655		#else
57656		#define vmulxs_lane_f32(__p0_694, __p1_694, __p2_694) __extension__ ({ \
57657		float32_t __ret_694; \
57658		float32_t __s0_694 = __p0_694; \
57659		float32x2_t __s1_694 = __p1_694; \
57660		float32x2_t __rev1_694; __rev1_694 = __builtin_shufflevector(__s1_694, __s1_694, 1, 0); \
57661		__ret_694 = vmulxs_f32(__s0_694, __noswap_vget_lane_f32(__rev1_694, __p2_694)); \
57662		__ret_694; \
57663		})
57664		#endif
57665
57666		#ifdef __LITTLE_ENDIAN__
57667		#define vmulxq_lane_f64(__p0_695, __p1_695, __p2_695) __extension__ ({ \
57668		float64x2_t __ret_695; \
57669		float64x2_t __s0_695 = __p0_695; \
57670		float64x1_t __s1_695 = __p1_695; \
57671		__ret_695 = vmulxq_f64(__s0_695, splatq_lane_f64(__s1_695, __p2_695)); \
57672		__ret_695; \
57673		})
57674		#else
57675		#define vmulxq_lane_f64(__p0_696, __p1_696, __p2_696) __extension__ ({ \
57676		float64x2_t __ret_696; \
57677		float64x2_t __s0_696 = __p0_696; \
57678		float64x1_t __s1_696 = __p1_696; \
57679		float64x2_t __rev0_696; __rev0_696 = __builtin_shufflevector(__s0_696, __s0_696, 1, 0); \
57680		__ret_696 = __noswap_vmulxq_f64(__rev0_696, __noswap_splatq_lane_f64(__s1_696, __p2_696)); \
57681		__ret_696 = __builtin_shufflevector(__ret_696, __ret_696, 1, 0); \
57682		__ret_696; \
57683		})
57684		#endif
57685
57686		#ifdef __LITTLE_ENDIAN__
57687		#define vmulxq_lane_f32(__p0_697, __p1_697, __p2_697) __extension__ ({ \
57688		float32x4_t __ret_697; \
57689		float32x4_t __s0_697 = __p0_697; \
57690		float32x2_t __s1_697 = __p1_697; \
57691		__ret_697 = vmulxq_f32(__s0_697, splatq_lane_f32(__s1_697, __p2_697)); \
57692		__ret_697; \
57693		})
57694		#else
57695		#define vmulxq_lane_f32(__p0_698, __p1_698, __p2_698) __extension__ ({ \
57696		float32x4_t __ret_698; \
57697		float32x4_t __s0_698 = __p0_698; \
57698		float32x2_t __s1_698 = __p1_698; \
57699		float32x4_t __rev0_698; __rev0_698 = __builtin_shufflevector(__s0_698, __s0_698, 3, 2, 1, 0); \
57700		float32x2_t __rev1_698; __rev1_698 = __builtin_shufflevector(__s1_698, __s1_698, 1, 0); \
57701		__ret_698 = __noswap_vmulxq_f32(__rev0_698, __noswap_splatq_lane_f32(__rev1_698, __p2_698)); \
57702		__ret_698 = __builtin_shufflevector(__ret_698, __ret_698, 3, 2, 1, 0); \
57703		__ret_698; \
57704		})
57705		#endif
57706
57707		#ifdef __LITTLE_ENDIAN__
57708		#define vmulx_lane_f32(__p0_699, __p1_699, __p2_699) __extension__ ({ \
57709		float32x2_t __ret_699; \
57710		float32x2_t __s0_699 = __p0_699; \
57711		float32x2_t __s1_699 = __p1_699; \
57712		__ret_699 = vmulx_f32(__s0_699, splat_lane_f32(__s1_699, __p2_699)); \
57713		__ret_699; \
57714		})
57715		#else
57716		#define vmulx_lane_f32(__p0_700, __p1_700, __p2_700) __extension__ ({ \
57717		float32x2_t __ret_700; \
57718		float32x2_t __s0_700 = __p0_700; \
57719		float32x2_t __s1_700 = __p1_700; \
57720		float32x2_t __rev0_700; __rev0_700 = __builtin_shufflevector(__s0_700, __s0_700, 1, 0); \
57721		float32x2_t __rev1_700; __rev1_700 = __builtin_shufflevector(__s1_700, __s1_700, 1, 0); \
57722		__ret_700 = __noswap_vmulx_f32(__rev0_700, __noswap_splat_lane_f32(__rev1_700, __p2_700)); \
57723		__ret_700 = __builtin_shufflevector(__ret_700, __ret_700, 1, 0); \
57724		__ret_700; \
57725		})
57726		#endif
57727
57728		#ifdef __LITTLE_ENDIAN__
57729		#define vmulxd_laneq_f64(__p0_701, __p1_701, __p2_701) __extension__ ({ \
57730		float64_t __ret_701; \
57731		float64_t __s0_701 = __p0_701; \
57732		float64x2_t __s1_701 = __p1_701; \
57733		__ret_701 = vmulxd_f64(__s0_701, vgetq_lane_f64(__s1_701, __p2_701)); \
57734		__ret_701; \
57735		})
57736		#else
57737		#define vmulxd_laneq_f64(__p0_702, __p1_702, __p2_702) __extension__ ({ \
57738		float64_t __ret_702; \
57739		float64_t __s0_702 = __p0_702; \
57740		float64x2_t __s1_702 = __p1_702; \
57741		float64x2_t __rev1_702; __rev1_702 = __builtin_shufflevector(__s1_702, __s1_702, 1, 0); \
57742		__ret_702 = vmulxd_f64(__s0_702, __noswap_vgetq_lane_f64(__rev1_702, __p2_702)); \
57743		__ret_702; \
57744		})
57745		#endif
57746
57747		#ifdef __LITTLE_ENDIAN__
57748		#define vmulxs_laneq_f32(__p0_703, __p1_703, __p2_703) __extension__ ({ \
57749		float32_t __ret_703; \
57750		float32_t __s0_703 = __p0_703; \
57751		float32x4_t __s1_703 = __p1_703; \
57752		__ret_703 = vmulxs_f32(__s0_703, vgetq_lane_f32(__s1_703, __p2_703)); \
57753		__ret_703; \
57754		})
57755		#else
57756		#define vmulxs_laneq_f32(__p0_704, __p1_704, __p2_704) __extension__ ({ \
57757		float32_t __ret_704; \
57758		float32_t __s0_704 = __p0_704; \
57759		float32x4_t __s1_704 = __p1_704; \
57760		float32x4_t __rev1_704; __rev1_704 = __builtin_shufflevector(__s1_704, __s1_704, 3, 2, 1, 0); \
57761		__ret_704 = vmulxs_f32(__s0_704, __noswap_vgetq_lane_f32(__rev1_704, __p2_704)); \
57762		__ret_704; \
57763		})
57764		#endif
57765
57766		#ifdef __LITTLE_ENDIAN__
57767		#define vmulxq_laneq_f64(__p0_705, __p1_705, __p2_705) __extension__ ({ \
57768		float64x2_t __ret_705; \
57769		float64x2_t __s0_705 = __p0_705; \
57770		float64x2_t __s1_705 = __p1_705; \
57771		__ret_705 = vmulxq_f64(__s0_705, splatq_laneq_f64(__s1_705, __p2_705)); \
57772		__ret_705; \
57773		})
57774		#else
57775		#define vmulxq_laneq_f64(__p0_706, __p1_706, __p2_706) __extension__ ({ \
57776		float64x2_t __ret_706; \
57777		float64x2_t __s0_706 = __p0_706; \
57778		float64x2_t __s1_706 = __p1_706; \
57779		float64x2_t __rev0_706; __rev0_706 = __builtin_shufflevector(__s0_706, __s0_706, 1, 0); \
57780		float64x2_t __rev1_706; __rev1_706 = __builtin_shufflevector(__s1_706, __s1_706, 1, 0); \
57781		__ret_706 = __noswap_vmulxq_f64(__rev0_706, __noswap_splatq_laneq_f64(__rev1_706, __p2_706)); \
57782		__ret_706 = __builtin_shufflevector(__ret_706, __ret_706, 1, 0); \
57783		__ret_706; \
57784		})
57785		#endif
57786
57787		#ifdef __LITTLE_ENDIAN__
57788		#define vmulxq_laneq_f32(__p0_707, __p1_707, __p2_707) __extension__ ({ \
57789		float32x4_t __ret_707; \
57790		float32x4_t __s0_707 = __p0_707; \
57791		float32x4_t __s1_707 = __p1_707; \
57792		__ret_707 = vmulxq_f32(__s0_707, splatq_laneq_f32(__s1_707, __p2_707)); \
57793		__ret_707; \
57794		})
57795		#else
57796		#define vmulxq_laneq_f32(__p0_708, __p1_708, __p2_708) __extension__ ({ \
57797		float32x4_t __ret_708; \
57798		float32x4_t __s0_708 = __p0_708; \
57799		float32x4_t __s1_708 = __p1_708; \
57800		float32x4_t __rev0_708; __rev0_708 = __builtin_shufflevector(__s0_708, __s0_708, 3, 2, 1, 0); \
57801		float32x4_t __rev1_708; __rev1_708 = __builtin_shufflevector(__s1_708, __s1_708, 3, 2, 1, 0); \
57802		__ret_708 = __noswap_vmulxq_f32(__rev0_708, __noswap_splatq_laneq_f32(__rev1_708, __p2_708)); \
57803		__ret_708 = __builtin_shufflevector(__ret_708, __ret_708, 3, 2, 1, 0); \
57804		__ret_708; \
57805		})
57806		#endif
57807
57808		#ifdef __LITTLE_ENDIAN__
57809		#define vmulx_laneq_f32(__p0_709, __p1_709, __p2_709) __extension__ ({ \
57810		float32x2_t __ret_709; \
57811		float32x2_t __s0_709 = __p0_709; \
57812		float32x4_t __s1_709 = __p1_709; \
57813		__ret_709 = vmulx_f32(__s0_709, splat_laneq_f32(__s1_709, __p2_709)); \
57814		__ret_709; \
57815		})
57816		#else
57817		#define vmulx_laneq_f32(__p0_710, __p1_710, __p2_710) __extension__ ({ \
57818		float32x2_t __ret_710; \
57819		float32x2_t __s0_710 = __p0_710; \
57820		float32x4_t __s1_710 = __p1_710; \
57821		float32x2_t __rev0_710; __rev0_710 = __builtin_shufflevector(__s0_710, __s0_710, 1, 0); \
57822		float32x4_t __rev1_710; __rev1_710 = __builtin_shufflevector(__s1_710, __s1_710, 3, 2, 1, 0); \
57823		__ret_710 = __noswap_vmulx_f32(__rev0_710, __noswap_splat_laneq_f32(__rev1_710, __p2_710)); \
57824		__ret_710 = __builtin_shufflevector(__ret_710, __ret_710, 1, 0); \
57825		__ret_710; \
57826		})
57827		#endif
57828
57829		#ifdef __LITTLE_ENDIAN__
57830		__ai float64x2_t vnegq_f64(float64x2_t __p0) {
57831		float64x2_t __ret;
57832		__ret = -__p0;
	57842	__ai int8x8_t vreinterpret_s8_f32(float32x2_t __p0) {
	57843	int8x8_t __ret;
	57844	__ret = (int8x8_t)(__p0);
57833	57845	return __ret;
57834	57846	}
57835		#else
57836		__ai float64x2_t vnegq_f64(float64x2_t __p0) {
57837		float64x2_t __ret;
57838		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57839		__ret = -__rev0;
57840		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57847	__ai int8x8_t vreinterpret_s8_f16(float16x4_t __p0) {
	57848	int8x8_t __ret;
	57849	__ret = (int8x8_t)(__p0);
57841	57850	return __ret;
57842	57851	}
57843		#endif
57844
57845		#ifdef __LITTLE_ENDIAN__
57846		__ai int64x2_t vnegq_s64(int64x2_t __p0) {
57847		int64x2_t __ret;
57848		__ret = -__p0;
	57852	__ai int8x8_t vreinterpret_s8_s32(int32x2_t __p0) {
	57853	int8x8_t __ret;
	57854	__ret = (int8x8_t)(__p0);
57849	57855	return __ret;
57850	57856	}
57851		#else
57852		__ai int64x2_t vnegq_s64(int64x2_t __p0) {
57853		int64x2_t __ret;
57854		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57855		__ret = -__rev0;
57856		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57857	__ai int8x8_t vreinterpret_s8_s64(int64x1_t __p0) {
	57858	int8x8_t __ret;
	57859	__ret = (int8x8_t)(__p0);
57857	57860	return __ret;
57858	57861	}
57859		#endif
57860
57861		__ai float64x1_t vneg_f64(float64x1_t __p0) {
57862		float64x1_t __ret;
57863		__ret = -__p0;
	57862	__ai int8x8_t vreinterpret_s8_s16(int16x4_t __p0) {
	57863	int8x8_t __ret;
	57864	__ret = (int8x8_t)(__p0);
57864	57865	return __ret;
57865	57866	}
57866		__ai int64x1_t vneg_s64(int64x1_t __p0) {
57867		int64x1_t __ret;
57868		__ret = -__p0;
	57867	__ai float64x1_t vreinterpret_f64_p8(poly8x8_t __p0) {
	57868	float64x1_t __ret;
	57869	__ret = (float64x1_t)(__p0);
57869	57870	return __ret;
57870	57871	}
57871		__ai int64_t vnegd_s64(int64_t __p0) {
57872		int64_t __ret;
57873		__ret = (int64_t) __builtin_neon_vnegd_s64(__p0);
	57872	__ai float64x1_t vreinterpret_f64_p64(poly64x1_t __p0) {
	57873	float64x1_t __ret;
	57874	__ret = (float64x1_t)(__p0);
57874	57875	return __ret;
57875	57876	}
57876		#ifdef __LITTLE_ENDIAN__
57877		__ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
57878		uint8x16_t __ret;
57879		__ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	57877	__ai float64x1_t vreinterpret_f64_p16(poly16x4_t __p0) {
	57878	float64x1_t __ret;
	57879	__ret = (float64x1_t)(__p0);
57880	57880	return __ret;
57881	57881	}
57882		#else
57883		__ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
57884		uint8x16_t __ret;
57885		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57886		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57887		__ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
57888		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	57882	__ai float64x1_t vreinterpret_f64_u8(uint8x8_t __p0) {
	57883	float64x1_t __ret;
	57884	__ret = (float64x1_t)(__p0);
57889	57885	return __ret;
57890	57886	}
57891		#endif
57892
57893		#ifdef __LITTLE_ENDIAN__
57894		__ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
57895		uint32x4_t __ret;
57896		__ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	57887	__ai float64x1_t vreinterpret_f64_u32(uint32x2_t __p0) {
	57888	float64x1_t __ret;
	57889	__ret = (float64x1_t)(__p0);
57897	57890	return __ret;
57898	57891	}
57899		#else
57900		__ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
57901		uint32x4_t __ret;
57902		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57903		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57904		__ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
57905		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57892	__ai float64x1_t vreinterpret_f64_u64(uint64x1_t __p0) {
	57893	float64x1_t __ret;
	57894	__ret = (float64x1_t)(__p0);
57906	57895	return __ret;
57907	57896	}
57908		#endif
57909
57910		#ifdef __LITTLE_ENDIAN__
57911		__ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
57912		uint64x2_t __ret;
57913		__ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	57897	__ai float64x1_t vreinterpret_f64_u16(uint16x4_t __p0) {
	57898	float64x1_t __ret;
	57899	__ret = (float64x1_t)(__p0);
57914	57900	return __ret;
57915	57901	}
57916		#else
57917		__ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
57918		uint64x2_t __ret;
57919		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57920		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57921		__ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
57922		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57902	__ai float64x1_t vreinterpret_f64_s8(int8x8_t __p0) {
	57903	float64x1_t __ret;
	57904	__ret = (float64x1_t)(__p0);
57923	57905	return __ret;
57924	57906	}
57925		#endif
57926
57927		#ifdef __LITTLE_ENDIAN__
57928		__ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
57929		uint16x8_t __ret;
57930		__ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	57907	__ai float64x1_t vreinterpret_f64_f32(float32x2_t __p0) {
	57908	float64x1_t __ret;
	57909	__ret = (float64x1_t)(__p0);
57931	57910	return __ret;
57932	57911	}
57933		#else
57934		__ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
57935		uint16x8_t __ret;
57936		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
57937		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57938		__ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
57939		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57912	__ai float64x1_t vreinterpret_f64_f16(float16x4_t __p0) {
	57913	float64x1_t __ret;
	57914	__ret = (float64x1_t)(__p0);
57940	57915	return __ret;
57941	57916	}
57942		#endif
57943
57944		#ifdef __LITTLE_ENDIAN__
57945		__ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
57946		int8x16_t __ret;
57947		__ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	57917	__ai float64x1_t vreinterpret_f64_s32(int32x2_t __p0) {
	57918	float64x1_t __ret;
	57919	__ret = (float64x1_t)(__p0);
57948	57920	return __ret;
57949	57921	}
57950		#else
57951		__ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
57952		int8x16_t __ret;
57953		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57954		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57955		__ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
57956		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	57922	__ai float64x1_t vreinterpret_f64_s64(int64x1_t __p0) {
	57923	float64x1_t __ret;
	57924	__ret = (float64x1_t)(__p0);
57957	57925	return __ret;
57958	57926	}
57959		#endif
57960
57961		#ifdef __LITTLE_ENDIAN__
57962		__ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
57963		float64x2_t __ret;
57964		__ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	57927	__ai float64x1_t vreinterpret_f64_s16(int16x4_t __p0) {
	57928	float64x1_t __ret;
	57929	__ret = (float64x1_t)(__p0);
57965	57930	return __ret;
57966	57931	}
57967		#else
57968		__ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
57969		float64x2_t __ret;
57970		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57971		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57972		__ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
57973		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57932	__ai float32x2_t vreinterpret_f32_p8(poly8x8_t __p0) {
	57933	float32x2_t __ret;
	57934	__ret = (float32x2_t)(__p0);
57974	57935	return __ret;
57975	57936	}
57976		#endif
57977
57978		#ifdef __LITTLE_ENDIAN__
57979		__ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
57980		float32x4_t __ret;
57981		__ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	57937	__ai float32x2_t vreinterpret_f32_p64(poly64x1_t __p0) {
	57938	float32x2_t __ret;
	57939	__ret = (float32x2_t)(__p0);
57982	57940	return __ret;
57983	57941	}
57984		#else
57985		__ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
57986		float32x4_t __ret;
57987		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57988		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57989		__ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
57990		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57942	__ai float32x2_t vreinterpret_f32_p16(poly16x4_t __p0) {
	57943	float32x2_t __ret;
	57944	__ret = (float32x2_t)(__p0);
57991	57945	return __ret;
57992	57946	}
57993		#endif
57994
57995		#ifdef __LITTLE_ENDIAN__
57996		__ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
57997		int32x4_t __ret;
57998		__ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
	57947	__ai float32x2_t vreinterpret_f32_u8(uint8x8_t __p0) {
	57948	float32x2_t __ret;
	57949	__ret = (float32x2_t)(__p0);
57999	57950	return __ret;
58000	57951	}
58001		#else
58002		__ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
58003		int32x4_t __ret;
58004		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58005		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58006		__ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
58007		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	57952	__ai float32x2_t vreinterpret_f32_u32(uint32x2_t __p0) {
	57953	float32x2_t __ret;
	57954	__ret = (float32x2_t)(__p0);
58008	57955	return __ret;
58009	57956	}
58010		#endif
58011
58012		#ifdef __LITTLE_ENDIAN__
58013		__ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
58014		int64x2_t __ret;
58015		__ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
	57957	__ai float32x2_t vreinterpret_f32_u64(uint64x1_t __p0) {
	57958	float32x2_t __ret;
	57959	__ret = (float32x2_t)(__p0);
58016	57960	return __ret;
58017	57961	}
58018		#else
58019		__ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
58020		int64x2_t __ret;
58021		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58022		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58023		__ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
58024		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	57962	__ai float32x2_t vreinterpret_f32_u16(uint16x4_t __p0) {
	57963	float32x2_t __ret;
	57964	__ret = (float32x2_t)(__p0);
58025	57965	return __ret;
58026	57966	}
58027		#endif
58028
58029		#ifdef __LITTLE_ENDIAN__
58030		__ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
58031		int16x8_t __ret;
58032		__ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
	57967	__ai float32x2_t vreinterpret_f32_s8(int8x8_t __p0) {
	57968	float32x2_t __ret;
	57969	__ret = (float32x2_t)(__p0);
58033	57970	return __ret;
58034	57971	}
58035		#else
58036		__ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
58037		int16x8_t __ret;
58038		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
58039		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58040		__ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
58041		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	57972	__ai float32x2_t vreinterpret_f32_f64(float64x1_t __p0) {
	57973	float32x2_t __ret;
	57974	__ret = (float32x2_t)(__p0);
58042	57975	return __ret;
58043	57976	}
58044		#endif
58045
58046		#ifdef __LITTLE_ENDIAN__
58047		__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
58048		uint64_t __ret;
58049		__ret = (uint64_t) __builtin_neon_vpaddd_u64(__p0);
	57977	__ai float32x2_t vreinterpret_f32_f16(float16x4_t __p0) {
	57978	float32x2_t __ret;
	57979	__ret = (float32x2_t)(__p0);
58050	57980	return __ret;
58051	57981	}
58052		#else
58053		__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
58054		uint64_t __ret;
58055		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58056		__ret = (uint64_t) __builtin_neon_vpaddd_u64(__rev0);
	57982	__ai float32x2_t vreinterpret_f32_s32(int32x2_t __p0) {
	57983	float32x2_t __ret;
	57984	__ret = (float32x2_t)(__p0);
58057	57985	return __ret;
58058	57986	}
58059		#endif
58060
58061		#ifdef __LITTLE_ENDIAN__
58062		__ai float64_t vpaddd_f64(float64x2_t __p0) {
58063		float64_t __ret;
58064		__ret = (float64_t) __builtin_neon_vpaddd_f64(__p0);
	57987	__ai float32x2_t vreinterpret_f32_s64(int64x1_t __p0) {
	57988	float32x2_t __ret;
	57989	__ret = (float32x2_t)(__p0);
58065	57990	return __ret;
58066	57991	}
58067		#else
58068		__ai float64_t vpaddd_f64(float64x2_t __p0) {
58069		float64_t __ret;
58070		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58071		__ret = (float64_t) __builtin_neon_vpaddd_f64(__rev0);
	57992	__ai float32x2_t vreinterpret_f32_s16(int16x4_t __p0) {
	57993	float32x2_t __ret;
	57994	__ret = (float32x2_t)(__p0);
58072	57995	return __ret;
58073	57996	}
58074		#endif
58075
58076		#ifdef __LITTLE_ENDIAN__
58077		__ai int64_t vpaddd_s64(int64x2_t __p0) {
58078		int64_t __ret;
58079		__ret = (int64_t) __builtin_neon_vpaddd_s64(__p0);
	57997	__ai float16x4_t vreinterpret_f16_p8(poly8x8_t __p0) {
	57998	float16x4_t __ret;
	57999	__ret = (float16x4_t)(__p0);
58080	58000	return __ret;
58081	58001	}
58082		#else
58083		__ai int64_t vpaddd_s64(int64x2_t __p0) {
58084		int64_t __ret;
58085		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58086		__ret = (int64_t) __builtin_neon_vpaddd_s64(__rev0);
	58002	__ai float16x4_t vreinterpret_f16_p64(poly64x1_t __p0) {
	58003	float16x4_t __ret;
	58004	__ret = (float16x4_t)(__p0);
58087	58005	return __ret;
58088	58006	}
58089		#endif
58090
58091		#ifdef __LITTLE_ENDIAN__
58092		__ai float32_t vpadds_f32(float32x2_t __p0) {
58093		float32_t __ret;
58094		__ret = (float32_t) __builtin_neon_vpadds_f32(__p0);
	58007	__ai float16x4_t vreinterpret_f16_p16(poly16x4_t __p0) {
	58008	float16x4_t __ret;
	58009	__ret = (float16x4_t)(__p0);
58095	58010	return __ret;
58096	58011	}
58097		#else
58098		__ai float32_t vpadds_f32(float32x2_t __p0) {
58099		float32_t __ret;
58100		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58101		__ret = (float32_t) __builtin_neon_vpadds_f32(__rev0);
	58012	__ai float16x4_t vreinterpret_f16_u8(uint8x8_t __p0) {
	58013	float16x4_t __ret;
	58014	__ret = (float16x4_t)(__p0);
58102	58015	return __ret;
58103	58016	}
58104		#endif
58105
58106		#ifdef __LITTLE_ENDIAN__
58107		__ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
58108		uint8x16_t __ret;
58109		__ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	58017	__ai float16x4_t vreinterpret_f16_u32(uint32x2_t __p0) {
	58018	float16x4_t __ret;
	58019	__ret = (float16x4_t)(__p0);
58110	58020	return __ret;
58111	58021	}
58112		#else
58113		__ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
58114		uint8x16_t __ret;
58115		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58116		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58117		__ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
58118		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	58022	__ai float16x4_t vreinterpret_f16_u64(uint64x1_t __p0) {
	58023	float16x4_t __ret;
	58024	__ret = (float16x4_t)(__p0);
58119	58025	return __ret;
58120	58026	}
58121		#endif
58122
58123		#ifdef __LITTLE_ENDIAN__
58124		__ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
58125		uint32x4_t __ret;
58126		__ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	58027	__ai float16x4_t vreinterpret_f16_u16(uint16x4_t __p0) {
	58028	float16x4_t __ret;
	58029	__ret = (float16x4_t)(__p0);
58127	58030	return __ret;
58128	58031	}
58129		#else
58130		__ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
58131		uint32x4_t __ret;
58132		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58133		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58134		__ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
58135		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58032	__ai float16x4_t vreinterpret_f16_s8(int8x8_t __p0) {
	58033	float16x4_t __ret;
	58034	__ret = (float16x4_t)(__p0);
58136	58035	return __ret;
58137	58036	}
58138		#endif
58139
58140		#ifdef __LITTLE_ENDIAN__
58141		__ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
58142		uint16x8_t __ret;
58143		__ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	58037	__ai float16x4_t vreinterpret_f16_f64(float64x1_t __p0) {
	58038	float16x4_t __ret;
	58039	__ret = (float16x4_t)(__p0);
58144	58040	return __ret;
58145	58041	}
58146		#else
58147		__ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
58148		uint16x8_t __ret;
58149		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
58150		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58151		__ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
58152		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	58042	__ai float16x4_t vreinterpret_f16_f32(float32x2_t __p0) {
	58043	float16x4_t __ret;
	58044	__ret = (float16x4_t)(__p0);
58153	58045	return __ret;
58154	58046	}
58155		#endif
58156
58157		#ifdef __LITTLE_ENDIAN__
58158		__ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
58159		int8x16_t __ret;
58160		__ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	58047	__ai float16x4_t vreinterpret_f16_s32(int32x2_t __p0) {
	58048	float16x4_t __ret;
	58049	__ret = (float16x4_t)(__p0);
58161	58050	return __ret;
58162	58051	}
58163		#else
58164		__ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
58165		int8x16_t __ret;
58166		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58167		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58168		__ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
58169		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	58052	__ai float16x4_t vreinterpret_f16_s64(int64x1_t __p0) {
	58053	float16x4_t __ret;
	58054	__ret = (float16x4_t)(__p0);
58170	58055	return __ret;
58171	58056	}
58172		#endif
58173
58174		#ifdef __LITTLE_ENDIAN__
58175		__ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
58176		float64x2_t __ret;
58177		__ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	58057	__ai float16x4_t vreinterpret_f16_s16(int16x4_t __p0) {
	58058	float16x4_t __ret;
	58059	__ret = (float16x4_t)(__p0);
58178	58060	return __ret;
58179	58061	}
58180		#else
58181		__ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
58182		float64x2_t __ret;
58183		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58184		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58185		__ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
58186		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	58062	__ai int32x2_t vreinterpret_s32_p8(poly8x8_t __p0) {
	58063	int32x2_t __ret;
	58064	__ret = (int32x2_t)(__p0);
58187	58065	return __ret;
58188	58066	}
58189		#endif
58190
58191		#ifdef __LITTLE_ENDIAN__
58192		__ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
58193		float32x4_t __ret;
58194		__ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	58067	__ai int32x2_t vreinterpret_s32_p64(poly64x1_t __p0) {
	58068	int32x2_t __ret;
	58069	__ret = (int32x2_t)(__p0);
58195	58070	return __ret;
58196	58071	}
58197		#else
58198		__ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
58199		float32x4_t __ret;
58200		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58201		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58202		__ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
58203		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58072	__ai int32x2_t vreinterpret_s32_p16(poly16x4_t __p0) {
	58073	int32x2_t __ret;
	58074	__ret = (int32x2_t)(__p0);
58204	58075	return __ret;
58205	58076	}
58206		#endif
58207
58208		#ifdef __LITTLE_ENDIAN__
58209		__ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
58210		int32x4_t __ret;
58211		__ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
	58077	__ai int32x2_t vreinterpret_s32_u8(uint8x8_t __p0) {
	58078	int32x2_t __ret;
	58079	__ret = (int32x2_t)(__p0);
58212	58080	return __ret;
58213	58081	}
58214		#else
58215		__ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
58216		int32x4_t __ret;
58217		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58218		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58219		__ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
58220		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58082	__ai int32x2_t vreinterpret_s32_u32(uint32x2_t __p0) {
	58083	int32x2_t __ret;
	58084	__ret = (int32x2_t)(__p0);
58221	58085	return __ret;
58222	58086	}
58223		#endif
58224
58225		#ifdef __LITTLE_ENDIAN__
58226		__ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
58227		int16x8_t __ret;
58228		__ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
	58087	__ai int32x2_t vreinterpret_s32_u64(uint64x1_t __p0) {
	58088	int32x2_t __ret;
	58089	__ret = (int32x2_t)(__p0);
58229	58090	return __ret;
58230	58091	}
58231		#else
58232		__ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
58233		int16x8_t __ret;
58234		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
58235		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58236		__ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
58237		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	58092	__ai int32x2_t vreinterpret_s32_u16(uint16x4_t __p0) {
	58093	int32x2_t __ret;
	58094	__ret = (int32x2_t)(__p0);
58238	58095	return __ret;
58239	58096	}
58240		#endif
58241
58242		#ifdef __LITTLE_ENDIAN__
58243		__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
58244		float64_t __ret;
58245		__ret = (float64_t) __builtin_neon_vpmaxqd_f64(__p0);
	58097	__ai int32x2_t vreinterpret_s32_s8(int8x8_t __p0) {
	58098	int32x2_t __ret;
	58099	__ret = (int32x2_t)(__p0);
58246	58100	return __ret;
58247	58101	}
58248		#else
58249		__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
58250		float64_t __ret;
58251		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58252		__ret = (float64_t) __builtin_neon_vpmaxqd_f64(__rev0);
	58102	__ai int32x2_t vreinterpret_s32_f64(float64x1_t __p0) {
	58103	int32x2_t __ret;
	58104	__ret = (int32x2_t)(__p0);
58253	58105	return __ret;
58254	58106	}
58255		#endif
58256
58257		#ifdef __LITTLE_ENDIAN__
58258		__ai float32_t vpmaxs_f32(float32x2_t __p0) {
58259		float32_t __ret;
58260		__ret = (float32_t) __builtin_neon_vpmaxs_f32(__p0);
	58107	__ai int32x2_t vreinterpret_s32_f32(float32x2_t __p0) {
	58108	int32x2_t __ret;
	58109	__ret = (int32x2_t)(__p0);
58261	58110	return __ret;
58262	58111	}
58263		#else
58264		__ai float32_t vpmaxs_f32(float32x2_t __p0) {
58265		float32_t __ret;
58266		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58267		__ret = (float32_t) __builtin_neon_vpmaxs_f32(__rev0);
	58112	__ai int32x2_t vreinterpret_s32_f16(float16x4_t __p0) {
	58113	int32x2_t __ret;
	58114	__ret = (int32x2_t)(__p0);
58268	58115	return __ret;
58269	58116	}
58270		#endif
58271
58272		#ifdef __LITTLE_ENDIAN__
58273		__ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
58274		float64x2_t __ret;
58275		__ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	58117	__ai int32x2_t vreinterpret_s32_s64(int64x1_t __p0) {
	58118	int32x2_t __ret;
	58119	__ret = (int32x2_t)(__p0);
58276	58120	return __ret;
58277	58121	}
58278		#else
58279		__ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
58280		float64x2_t __ret;
58281		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58282		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58283		__ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
58284		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	58122	__ai int32x2_t vreinterpret_s32_s16(int16x4_t __p0) {
	58123	int32x2_t __ret;
	58124	__ret = (int32x2_t)(__p0);
58285	58125	return __ret;
58286	58126	}
58287		#endif
58288
58289		#ifdef __LITTLE_ENDIAN__
58290		__ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
58291		float32x4_t __ret;
58292		__ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	58127	__ai int64x1_t vreinterpret_s64_p8(poly8x8_t __p0) {
	58128	int64x1_t __ret;
	58129	__ret = (int64x1_t)(__p0);
58293	58130	return __ret;
58294	58131	}
58295		#else
58296		__ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
58297		float32x4_t __ret;
58298		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58299		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58300		__ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
58301		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58132	__ai int64x1_t vreinterpret_s64_p64(poly64x1_t __p0) {
	58133	int64x1_t __ret;
	58134	__ret = (int64x1_t)(__p0);
58302	58135	return __ret;
58303	58136	}
58304		#endif
58305
58306		#ifdef __LITTLE_ENDIAN__
58307		__ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
58308		float32x2_t __ret;
58309		__ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
	58137	__ai int64x1_t vreinterpret_s64_p16(poly16x4_t __p0) {
	58138	int64x1_t __ret;
	58139	__ret = (int64x1_t)(__p0);
58310	58140	return __ret;
58311	58141	}
58312		#else
58313		__ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
58314		float32x2_t __ret;
58315		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58316		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58317		__ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
58318		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	58142	__ai int64x1_t vreinterpret_s64_u8(uint8x8_t __p0) {
	58143	int64x1_t __ret;
	58144	__ret = (int64x1_t)(__p0);
58319	58145	return __ret;
58320	58146	}
58321		#endif
58322
58323		#ifdef __LITTLE_ENDIAN__
58324		__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
58325		float64_t __ret;
58326		__ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__p0);
	58147	__ai int64x1_t vreinterpret_s64_u32(uint32x2_t __p0) {
	58148	int64x1_t __ret;
	58149	__ret = (int64x1_t)(__p0);
58327	58150	return __ret;
58328	58151	}
58329		#else
58330		__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
58331		float64_t __ret;
58332		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58333		__ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__rev0);
	58152	__ai int64x1_t vreinterpret_s64_u64(uint64x1_t __p0) {
	58153	int64x1_t __ret;
	58154	__ret = (int64x1_t)(__p0);
58334	58155	return __ret;
58335	58156	}
58336		#endif
58337
58338		#ifdef __LITTLE_ENDIAN__
58339		__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
58340		float32_t __ret;
58341		__ret = (float32_t) __builtin_neon_vpmaxnms_f32(__p0);
	58157	__ai int64x1_t vreinterpret_s64_u16(uint16x4_t __p0) {
	58158	int64x1_t __ret;
	58159	__ret = (int64x1_t)(__p0);
58342	58160	return __ret;
58343	58161	}
58344		#else
58345		__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
58346		float32_t __ret;
58347		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58348		__ret = (float32_t) __builtin_neon_vpmaxnms_f32(__rev0);
	58162	__ai int64x1_t vreinterpret_s64_s8(int8x8_t __p0) {
	58163	int64x1_t __ret;
	58164	__ret = (int64x1_t)(__p0);
58349	58165	return __ret;
58350	58166	}
58351		#endif
58352
58353		#ifdef __LITTLE_ENDIAN__
58354		__ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
58355		uint8x16_t __ret;
58356		__ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	58167	__ai int64x1_t vreinterpret_s64_f64(float64x1_t __p0) {
	58168	int64x1_t __ret;
	58169	__ret = (int64x1_t)(__p0);
58357	58170	return __ret;
58358	58171	}
58359		#else
58360		__ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
58361		uint8x16_t __ret;
58362		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58363		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58364		__ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
58365		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	58172	__ai int64x1_t vreinterpret_s64_f32(float32x2_t __p0) {
	58173	int64x1_t __ret;
	58174	__ret = (int64x1_t)(__p0);
58366	58175	return __ret;
58367	58176	}
58368		#endif
58369
58370		#ifdef __LITTLE_ENDIAN__
58371		__ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
58372		uint32x4_t __ret;
58373		__ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	58177	__ai int64x1_t vreinterpret_s64_f16(float16x4_t __p0) {
	58178	int64x1_t __ret;
	58179	__ret = (int64x1_t)(__p0);
58374	58180	return __ret;
58375	58181	}
58376		#else
58377		__ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
58378		uint32x4_t __ret;
58379		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58380		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58381		__ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
58382		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58182	__ai int64x1_t vreinterpret_s64_s32(int32x2_t __p0) {
	58183	int64x1_t __ret;
	58184	__ret = (int64x1_t)(__p0);
58383	58185	return __ret;
58384	58186	}
58385		#endif
58386
58387		#ifdef __LITTLE_ENDIAN__
58388		__ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
58389		uint16x8_t __ret;
58390		__ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	58187	__ai int64x1_t vreinterpret_s64_s16(int16x4_t __p0) {
	58188	int64x1_t __ret;
	58189	__ret = (int64x1_t)(__p0);
58391	58190	return __ret;
58392	58191	}
58393		#else
58394		__ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
58395		uint16x8_t __ret;
58396		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
58397		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58398		__ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
58399		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	58192	__ai int16x4_t vreinterpret_s16_p8(poly8x8_t __p0) {
	58193	int16x4_t __ret;
	58194	__ret = (int16x4_t)(__p0);
58400	58195	return __ret;
58401	58196	}
58402		#endif
58403
58404		#ifdef __LITTLE_ENDIAN__
58405		__ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
58406		int8x16_t __ret;
58407		__ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	58197	__ai int16x4_t vreinterpret_s16_p64(poly64x1_t __p0) {
	58198	int16x4_t __ret;
	58199	__ret = (int16x4_t)(__p0);
58408	58200	return __ret;
58409	58201	}
58410		#else
58411		__ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
58412		int8x16_t __ret;
58413		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58414		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58415		__ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
58416		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	58202	__ai int16x4_t vreinterpret_s16_p16(poly16x4_t __p0) {
	58203	int16x4_t __ret;
	58204	__ret = (int16x4_t)(__p0);
58417	58205	return __ret;
58418	58206	}
58419		#endif
58420
58421		#ifdef __LITTLE_ENDIAN__
58422		__ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
58423		float64x2_t __ret;
58424		__ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	58207	__ai int16x4_t vreinterpret_s16_u8(uint8x8_t __p0) {
	58208	int16x4_t __ret;
	58209	__ret = (int16x4_t)(__p0);
58425	58210	return __ret;
58426	58211	}
58427		#else
58428		__ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
58429		float64x2_t __ret;
58430		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58431		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58432		__ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
58433		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	58212	__ai int16x4_t vreinterpret_s16_u32(uint32x2_t __p0) {
	58213	int16x4_t __ret;
	58214	__ret = (int16x4_t)(__p0);
58434	58215	return __ret;
58435	58216	}
58436		#endif
58437
58438		#ifdef __LITTLE_ENDIAN__
58439		__ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
58440		float32x4_t __ret;
58441		__ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
	58217	__ai int16x4_t vreinterpret_s16_u64(uint64x1_t __p0) {
	58218	int16x4_t __ret;
	58219	__ret = (int16x4_t)(__p0);
58442	58220	return __ret;
58443	58221	}
58444		#else
58445		__ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
58446		float32x4_t __ret;
58447		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58448		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58449		__ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
58450		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58222	__ai int16x4_t vreinterpret_s16_u16(uint16x4_t __p0) {
	58223	int16x4_t __ret;
	58224	__ret = (int16x4_t)(__p0);
58451	58225	return __ret;
58452	58226	}
58453		#endif
58454
58455		#ifdef __LITTLE_ENDIAN__
58456		__ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
58457		int32x4_t __ret;
58458		__ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
	58227	__ai int16x4_t vreinterpret_s16_s8(int8x8_t __p0) {
	58228	int16x4_t __ret;
	58229	__ret = (int16x4_t)(__p0);
58459	58230	return __ret;
58460	58231	}
58461		#else
58462		__ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
58463		int32x4_t __ret;
58464		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58465		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58466		__ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
58467		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58232	__ai int16x4_t vreinterpret_s16_f64(float64x1_t __p0) {
	58233	int16x4_t __ret;
	58234	__ret = (int16x4_t)(__p0);
58468	58235	return __ret;
58469	58236	}
58470		#endif
58471
58472		#ifdef __LITTLE_ENDIAN__
58473		__ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
58474		int16x8_t __ret;
58475		__ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
	58237	__ai int16x4_t vreinterpret_s16_f32(float32x2_t __p0) {
	58238	int16x4_t __ret;
	58239	__ret = (int16x4_t)(__p0);
58476	58240	return __ret;
58477	58241	}
58478		#else
58479		__ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
58480		int16x8_t __ret;
58481		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
58482		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58483		__ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
58484		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	58242	__ai int16x4_t vreinterpret_s16_f16(float16x4_t __p0) {
	58243	int16x4_t __ret;
	58244	__ret = (int16x4_t)(__p0);
58485	58245	return __ret;
58486	58246	}
58487		#endif
58488
58489		#ifdef __LITTLE_ENDIAN__
58490		__ai float64_t vpminqd_f64(float64x2_t __p0) {
58491		float64_t __ret;
58492		__ret = (float64_t) __builtin_neon_vpminqd_f64(__p0);
	58247	__ai int16x4_t vreinterpret_s16_s32(int32x2_t __p0) {
	58248	int16x4_t __ret;
	58249	__ret = (int16x4_t)(__p0);
58493	58250	return __ret;
58494	58251	}
58495		#else
58496		__ai float64_t vpminqd_f64(float64x2_t __p0) {
58497		float64_t __ret;
58498		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58499		__ret = (float64_t) __builtin_neon_vpminqd_f64(__rev0);
	58252	__ai int16x4_t vreinterpret_s16_s64(int64x1_t __p0) {
	58253	int16x4_t __ret;
	58254	__ret = (int16x4_t)(__p0);
58500	58255	return __ret;
58501	58256	}
58502		#endif
58503
58504		#ifdef __LITTLE_ENDIAN__
58505		__ai float32_t vpmins_f32(float32x2_t __p0) {
58506		float32_t __ret;
58507		__ret = (float32_t) __builtin_neon_vpmins_f32(__p0);
	58257	__ai uint64_t vrshld_u64(uint64_t __p0, int64_t __p1) {
	58258	uint64_t __ret;
	58259	__ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
58508	58260	return __ret;
58509	58261	}
58510		#else
58511		__ai float32_t vpmins_f32(float32x2_t __p0) {
58512		float32_t __ret;
58513		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58514		__ret = (float32_t) __builtin_neon_vpmins_f32(__rev0);
	58262	__ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
	58263	int64_t __ret;
	58264	__ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
58515	58265	return __ret;
58516	58266	}
	58267	#define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
	58268	uint64_t __ret; \
	58269	uint64_t __s0 = __p0; \
	58270	__ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
	58271	__ret; \
	58272	})
	58273	#define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
	58274	int64_t __ret; \
	58275	int64_t __s0 = __p0; \
	58276	__ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
	58277	__ret; \
	58278	})
	58279	#ifdef __LITTLE_ENDIAN__
	58280	#define vrshrn_high_n_u32(__p0_703, __p1_703, __p2_703) __extension__ ({ \
	58281	uint16x8_t __ret_703; \
	58282	uint16x4_t __s0_703 = __p0_703; \
	58283	uint32x4_t __s1_703 = __p1_703; \
	58284	__ret_703 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_703), (uint16x4_t)(vrshrn_n_u32(__s1_703, __p2_703)))); \
	58285	__ret_703; \
	58286	})
	58287	#else
	58288	#define vrshrn_high_n_u32(__p0_704, __p1_704, __p2_704) __extension__ ({ \
	58289	uint16x8_t __ret_704; \
	58290	uint16x4_t __s0_704 = __p0_704; \
	58291	uint32x4_t __s1_704 = __p1_704; \
	58292	uint16x4_t __rev0_704; __rev0_704 = __builtin_shufflevector(__s0_704, __s0_704, 3, 2, 1, 0); \
	58293	uint32x4_t __rev1_704; __rev1_704 = __builtin_shufflevector(__s1_704, __s1_704, 3, 2, 1, 0); \
	58294	__ret_704 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_704), (uint16x4_t)(__noswap_vrshrn_n_u32(__rev1_704, __p2_704)))); \
	58295	__ret_704 = __builtin_shufflevector(__ret_704, __ret_704, 7, 6, 5, 4, 3, 2, 1, 0); \
	58296	__ret_704; \
	58297	})
58517	58298	#endif
58518	58299
58519	58300	#ifdef __LITTLE_ENDIAN__
58520		__ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
58521		float64x2_t __ret;
58522		__ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
58523		return __ret;
58524		}
	58301	#define vrshrn_high_n_u64(__p0_705, __p1_705, __p2_705) __extension__ ({ \
	58302	uint32x4_t __ret_705; \
	58303	uint32x2_t __s0_705 = __p0_705; \
	58304	uint64x2_t __s1_705 = __p1_705; \
	58305	__ret_705 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_705), (uint32x2_t)(vrshrn_n_u64(__s1_705, __p2_705)))); \
	58306	__ret_705; \
	58307	})
58525	58308	#else
58526		__ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
58527		float64x2_t __ret;
58528		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58529		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58530		__ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
58531		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
58532		return __ret;
58533		}
	58309	#define vrshrn_high_n_u64(__p0_706, __p1_706, __p2_706) __extension__ ({ \
	58310	uint32x4_t __ret_706; \
	58311	uint32x2_t __s0_706 = __p0_706; \
	58312	uint64x2_t __s1_706 = __p1_706; \
	58313	uint32x2_t __rev0_706; __rev0_706 = __builtin_shufflevector(__s0_706, __s0_706, 1, 0); \
	58314	uint64x2_t __rev1_706; __rev1_706 = __builtin_shufflevector(__s1_706, __s1_706, 1, 0); \
	58315	__ret_706 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_706), (uint32x2_t)(__noswap_vrshrn_n_u64(__rev1_706, __p2_706)))); \
	58316	__ret_706 = __builtin_shufflevector(__ret_706, __ret_706, 3, 2, 1, 0); \
	58317	__ret_706; \
	58318	})
58534	58319	#endif
58535	58320
58536	58321	#ifdef __LITTLE_ENDIAN__
58537		__ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
58538		float32x4_t __ret;
58539		__ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
58540		return __ret;
58541		}
	58322	#define vrshrn_high_n_u16(__p0_707, __p1_707, __p2_707) __extension__ ({ \
	58323	uint8x16_t __ret_707; \
	58324	uint8x8_t __s0_707 = __p0_707; \
	58325	uint16x8_t __s1_707 = __p1_707; \
	58326	__ret_707 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_707), (uint8x8_t)(vrshrn_n_u16(__s1_707, __p2_707)))); \
	58327	__ret_707; \
	58328	})
58542	58329	#else
58543		__ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
58544		float32x4_t __ret;
58545		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58546		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58547		__ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
58548		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
58549		return __ret;
58550		}
	58330	#define vrshrn_high_n_u16(__p0_708, __p1_708, __p2_708) __extension__ ({ \
	58331	uint8x16_t __ret_708; \
	58332	uint8x8_t __s0_708 = __p0_708; \
	58333	uint16x8_t __s1_708 = __p1_708; \
	58334	uint8x8_t __rev0_708; __rev0_708 = __builtin_shufflevector(__s0_708, __s0_708, 7, 6, 5, 4, 3, 2, 1, 0); \
	58335	uint16x8_t __rev1_708; __rev1_708 = __builtin_shufflevector(__s1_708, __s1_708, 7, 6, 5, 4, 3, 2, 1, 0); \
	58336	__ret_708 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_708), (uint8x8_t)(__noswap_vrshrn_n_u16(__rev1_708, __p2_708)))); \
	58337	__ret_708 = __builtin_shufflevector(__ret_708, __ret_708, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	58338	__ret_708; \
	58339	})
58551	58340	#endif
58552	58341
58553	58342	#ifdef __LITTLE_ENDIAN__
58554		__ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
58555		float32x2_t __ret;
58556		__ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
58557		return __ret;
58558		}
	58343	#define vrshrn_high_n_s32(__p0_709, __p1_709, __p2_709) __extension__ ({ \
	58344	int16x8_t __ret_709; \
	58345	int16x4_t __s0_709 = __p0_709; \
	58346	int32x4_t __s1_709 = __p1_709; \
	58347	__ret_709 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_709), (int16x4_t)(vrshrn_n_s32(__s1_709, __p2_709)))); \
	58348	__ret_709; \
	58349	})
58559	58350	#else
58560		__ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
58561		float32x2_t __ret;
58562		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58563		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58564		__ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
58565		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
58566		return __ret;
58567		}
	58351	#define vrshrn_high_n_s32(__p0_710, __p1_710, __p2_710) __extension__ ({ \
	58352	int16x8_t __ret_710; \
	58353	int16x4_t __s0_710 = __p0_710; \
	58354	int32x4_t __s1_710 = __p1_710; \
	58355	int16x4_t __rev0_710; __rev0_710 = __builtin_shufflevector(__s0_710, __s0_710, 3, 2, 1, 0); \
	58356	int32x4_t __rev1_710; __rev1_710 = __builtin_shufflevector(__s1_710, __s1_710, 3, 2, 1, 0); \
	58357	__ret_710 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_710), (int16x4_t)(__noswap_vrshrn_n_s32(__rev1_710, __p2_710)))); \
	58358	__ret_710 = __builtin_shufflevector(__ret_710, __ret_710, 7, 6, 5, 4, 3, 2, 1, 0); \
	58359	__ret_710; \
	58360	})
58568	58361	#endif
58569	58362
58570	58363	#ifdef __LITTLE_ENDIAN__
58571		__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
58572		float64_t __ret;
58573		__ret = (float64_t) __builtin_neon_vpminnmqd_f64(__p0);
58574		return __ret;
58575		}
	58364	#define vrshrn_high_n_s64(__p0_711, __p1_711, __p2_711) __extension__ ({ \
	58365	int32x4_t __ret_711; \
	58366	int32x2_t __s0_711 = __p0_711; \
	58367	int64x2_t __s1_711 = __p1_711; \
	58368	__ret_711 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_711), (int32x2_t)(vrshrn_n_s64(__s1_711, __p2_711)))); \
	58369	__ret_711; \
	58370	})
58576	58371	#else
58577		__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
58578		float64_t __ret;
58579		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58580		__ret = (float64_t) __builtin_neon_vpminnmqd_f64(__rev0);
58581		return __ret;
58582		}
	58372	#define vrshrn_high_n_s64(__p0_712, __p1_712, __p2_712) __extension__ ({ \
	58373	int32x4_t __ret_712; \
	58374	int32x2_t __s0_712 = __p0_712; \
	58375	int64x2_t __s1_712 = __p1_712; \
	58376	int32x2_t __rev0_712; __rev0_712 = __builtin_shufflevector(__s0_712, __s0_712, 1, 0); \
	58377	int64x2_t __rev1_712; __rev1_712 = __builtin_shufflevector(__s1_712, __s1_712, 1, 0); \
	58378	__ret_712 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_712), (int32x2_t)(__noswap_vrshrn_n_s64(__rev1_712, __p2_712)))); \
	58379	__ret_712 = __builtin_shufflevector(__ret_712, __ret_712, 3, 2, 1, 0); \
	58380	__ret_712; \
	58381	})
58583	58382	#endif
58584	58383
58585	58384	#ifdef __LITTLE_ENDIAN__
58586		__ai float32_t vpminnms_f32(float32x2_t __p0) {
58587		float32_t __ret;
58588		__ret = (float32_t) __builtin_neon_vpminnms_f32(__p0);
58589		return __ret;
58590		}
	58385	#define vrshrn_high_n_s16(__p0_713, __p1_713, __p2_713) __extension__ ({ \
	58386	int8x16_t __ret_713; \
	58387	int8x8_t __s0_713 = __p0_713; \
	58388	int16x8_t __s1_713 = __p1_713; \
	58389	__ret_713 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_713), (int8x8_t)(vrshrn_n_s16(__s1_713, __p2_713)))); \
	58390	__ret_713; \
	58391	})
58591	58392	#else
58592		__ai float32_t vpminnms_f32(float32x2_t __p0) {
58593		float32_t __ret;
58594		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58595		__ret = (float32_t) __builtin_neon_vpminnms_f32(__rev0);
58596		return __ret;
58597		}
	58393	#define vrshrn_high_n_s16(__p0_714, __p1_714, __p2_714) __extension__ ({ \
	58394	int8x16_t __ret_714; \
	58395	int8x8_t __s0_714 = __p0_714; \
	58396	int16x8_t __s1_714 = __p1_714; \
	58397	int8x8_t __rev0_714; __rev0_714 = __builtin_shufflevector(__s0_714, __s0_714, 7, 6, 5, 4, 3, 2, 1, 0); \
	58398	int16x8_t __rev1_714; __rev1_714 = __builtin_shufflevector(__s1_714, __s1_714, 7, 6, 5, 4, 3, 2, 1, 0); \
	58399	__ret_714 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_714), (int8x8_t)(__noswap_vrshrn_n_s16(__rev1_714, __p2_714)))); \
	58400	__ret_714 = __builtin_shufflevector(__ret_714, __ret_714, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	58401	__ret_714; \
	58402	})
58598	58403	#endif
58599	58404
58600	58405	#ifdef __LITTLE_ENDIAN__
58601		__ai int64x2_t vqabsq_s64(int64x2_t __p0) {
58602		int64x2_t __ret;
58603		__ret = (int64x2_t) __builtin_neon_vqabsq_v((int8x16_t)__p0, 35);
	58406	__ai float64x2_t vrsqrteq_f64(float64x2_t __p0) {
	58407	float64x2_t __ret;
	58408	__ret = (float64x2_t) __builtin_neon_vrsqrteq_v((int8x16_t)__p0, 42);
58604	58409	return __ret;
58605	58410	}
58606	58411	#else
58607		__ai int64x2_t vqabsq_s64(int64x2_t __p0) {
58608		int64x2_t __ret;
58609		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58610		__ret = (int64x2_t) __builtin_neon_vqabsq_v((int8x16_t)__rev0, 35);
	58412	__ai float64x2_t vrsqrteq_f64(float64x2_t __p0) {
	58413	float64x2_t __ret;
	58414	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	58415	__ret = (float64x2_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 42);
58611	58416	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
58612	58417	return __ret;
58613	58418	}
58614	58419	#endif
58615	58420
58616		__ai int64x1_t vqabs_s64(int64x1_t __p0) {
58617		int64x1_t __ret;
58618		__ret = (int64x1_t) __builtin_neon_vqabs_v((int8x8_t)__p0, 3);
58619		return __ret;
58620		}
58621		__ai int8_t vqabsb_s8(int8_t __p0) {
58622		int8_t __ret;
58623		__ret = (int8_t) __builtin_neon_vqabsb_s8(__p0);
	58421	__ai float64x1_t vrsqrte_f64(float64x1_t __p0) {
	58422	float64x1_t __ret;
	58423	__ret = (float64x1_t) __builtin_neon_vrsqrte_v((int8x8_t)__p0, 10);
58624	58424	return __ret;
58625	58425	}
58626		__ai int32_t vqabss_s32(int32_t __p0) {
58627		int32_t __ret;
58628		__ret = (int32_t) __builtin_neon_vqabss_s32(__p0);
	58426	__ai float64_t vrsqrted_f64(float64_t __p0) {
	58427	float64_t __ret;
	58428	__ret = (float64_t) __builtin_neon_vrsqrted_f64(__p0);
58629	58429	return __ret;
58630	58430	}
58631		__ai int64_t vqabsd_s64(int64_t __p0) {
58632		int64_t __ret;
58633		__ret = (int64_t) __builtin_neon_vqabsd_s64(__p0);
	58431	__ai float32_t vrsqrtes_f32(float32_t __p0) {
	58432	float32_t __ret;
	58433	__ret = (float32_t) __builtin_neon_vrsqrtes_f32(__p0);
58634	58434	return __ret;
58635	58435	}
58636		__ai int16_t vqabsh_s16(int16_t __p0) {
58637		int16_t __ret;
58638		__ret = (int16_t) __builtin_neon_vqabsh_s16(__p0);
	58436	#ifdef __LITTLE_ENDIAN__
	58437	__ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
	58438	float64x2_t __ret;
	58439	__ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
58639	58440	return __ret;
58640	58441	}
58641		__ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
58642		uint8_t __ret;
58643		__ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
	58442	#else
	58443	__ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
	58444	float64x2_t __ret;
	58445	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	58446	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	58447	__ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	58448	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
58644	58449	return __ret;
58645	58450	}
58646		__ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
58647		uint32_t __ret;
58648		__ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
	58451	#endif
	58452
	58453	__ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
	58454	float64x1_t __ret;
	58455	__ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
58649	58456	return __ret;
58650	58457	}
58651		__ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
58652		uint64_t __ret;
58653		__ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
	58458	__ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
	58459	float64_t __ret;
	58460	__ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
58654	58461	return __ret;
58655	58462	}
58656		__ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
58657		uint16_t __ret;
58658		__ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
	58463	__ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
	58464	float32_t __ret;
	58465	__ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
58659	58466	return __ret;
58660	58467	}
58661		__ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
58662		int8_t __ret;
58663		__ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
	58468	#define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
	58469	uint64_t __ret; \
	58470	uint64_t __s0 = __p0; \
	58471	uint64_t __s1 = __p1; \
	58472	__ret = (uint64_t) __builtin_neon_vrsrad_n_u64(__s0, __s1, __p2); \
	58473	__ret; \
	58474	})
	58475	#define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
	58476	int64_t __ret; \
	58477	int64_t __s0 = __p0; \
	58478	int64_t __s1 = __p1; \
	58479	__ret = (int64_t) __builtin_neon_vrsrad_n_s64(__s0, __s1, __p2); \
	58480	__ret; \
	58481	})
	58482	#ifdef __LITTLE_ENDIAN__
	58483	__ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	58484	uint16x8_t __ret;
	58485	__ret = vcombine_u16(__p0, vrsubhn_u32(__p1, __p2));
58664	58486	return __ret;
58665	58487	}
58666		__ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
58667		int32_t __ret;
58668		__ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
	58488	#else
	58489	__ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	58490	uint16x8_t __ret;
	58491	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	58492	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	58493	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	58494	__ret = __noswap_vcombine_u16(__rev0, __noswap_vrsubhn_u32(__rev1, __rev2));
	58495	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
58669	58496	return __ret;
58670	58497	}
58671		__ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
58672		int64_t __ret;
58673		__ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
	58498	#endif
	58499
	58500	#ifdef __LITTLE_ENDIAN__
	58501	__ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	58502	uint32x4_t __ret;
	58503	__ret = vcombine_u32(__p0, vrsubhn_u64(__p1, __p2));
58674	58504	return __ret;
58675	58505	}
58676		__ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
58677		int16_t __ret;
58678		__ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
	58506	#else
	58507	__ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	58508	uint32x4_t __ret;
	58509	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	58510	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	58511	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	58512	__ret = __noswap_vcombine_u32(__rev0, __noswap_vrsubhn_u64(__rev1, __rev2));
	58513	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
58679	58514	return __ret;
58680	58515	}
58681		__ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
58682		int64_t __ret;
58683		__ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
	58516	#endif
	58517
	58518	#ifdef __LITTLE_ENDIAN__
	58519	__ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	58520	uint8x16_t __ret;
	58521	__ret = vcombine_u8(__p0, vrsubhn_u16(__p1, __p2));
58684	58522	return __ret;
58685	58523	}
58686		__ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
58687		int32_t __ret;
58688		__ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
	58524	#else
	58525	__ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	58526	uint8x16_t __ret;
	58527	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	58528	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	58529	uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	58530	__ret = __noswap_vcombine_u8(__rev0, __noswap_vrsubhn_u16(__rev1, __rev2));
	58531	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58689	58532	return __ret;
58690	58533	}
	58534	#endif
	58535
58691	58536	#ifdef __LITTLE_ENDIAN__
58692		__ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
58693		int64x2_t __ret;
58694		__ret = vqdmlal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
	58537	__ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	58538	int16x8_t __ret;
	58539	__ret = vcombine_s16(__p0, vrsubhn_s32(__p1, __p2));
58695	58540	return __ret;
58696	58541	}
58697	58542	#else
58698		__ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
58699		int64x2_t __ret;
58700		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	58543	__ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	58544	int16x8_t __ret;
	58545	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58701	58546	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58702	58547	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
58703		__ret = __noswap_vqdmlal_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
58704		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	58548	__ret = __noswap_vcombine_s16(__rev0, __noswap_vrsubhn_s32(__rev1, __rev2));
	58549	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
58705	58550	return __ret;
58706	58551	}
58707	58552	#endif
58708	58553
58709	58554	#ifdef __LITTLE_ENDIAN__
58710		__ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
	58555	__ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
58711	58556	int32x4_t __ret;
58712		__ret = vqdmlal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
	58557	__ret = vcombine_s32(__p0, vrsubhn_s64(__p1, __p2));
58713	58558	return __ret;
58714	58559	}
58715	58560	#else
58716		__ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
	58561	__ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
58717	58562	int32x4_t __ret;
58718		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	58563	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	58564	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	58565	int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	58566	__ret = __noswap_vcombine_s32(__rev0, __noswap_vrsubhn_s64(__rev1, __rev2));
	58567	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58568	return __ret;
	58569	}
	58570	#endif
	58571
	58572	#ifdef __LITTLE_ENDIAN__
	58573	__ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	58574	int8x16_t __ret;
	58575	__ret = vcombine_s8(__p0, vrsubhn_s16(__p1, __p2));
	58576	return __ret;
	58577	}
	58578	#else
	58579	__ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	58580	int8x16_t __ret;
	58581	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
58719	58582	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58720	58583	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
58721		__ret = __noswap_vqdmlal_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
58722		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	58584	__ret = __noswap_vcombine_s8(__rev0, __noswap_vrsubhn_s16(__rev1, __rev2));
	58585	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58723	58586	return __ret;
58724	58587	}
58725	58588	#endif
58726	58589
	58590	#define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	58591	poly64x1_t __ret; \
	58592	poly64_t __s0 = __p0; \
	58593	poly64x1_t __s1 = __p1; \
	58594	__ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (poly64x1_t)__s1, __p2); \
	58595	__ret; \
	58596	})
58727	58597	#ifdef __LITTLE_ENDIAN__
58728		#define vqdmlal_high_lane_s32(__p0_711, __p1_711, __p2_711, __p3_711) __extension__ ({ \
58729		int64x2_t __ret_711; \
58730		int64x2_t __s0_711 = __p0_711; \
58731		int32x4_t __s1_711 = __p1_711; \
58732		int32x2_t __s2_711 = __p2_711; \
58733		__ret_711 = vqdmlal_s32(__s0_711, vget_high_s32(__s1_711), splat_lane_s32(__s2_711, __p3_711)); \
58734		__ret_711; \
	58598	#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	58599	poly64x2_t __ret; \
	58600	poly64_t __s0 = __p0; \
	58601	poly64x2_t __s1 = __p1; \
	58602	__ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
	58603	__ret; \
58735	58604	})
58736	58605	#else
58737		#define vqdmlal_high_lane_s32(__p0_712, __p1_712, __p2_712, __p3_712) __extension__ ({ \
58738		int64x2_t __ret_712; \
58739		int64x2_t __s0_712 = __p0_712; \
58740		int32x4_t __s1_712 = __p1_712; \
58741		int32x2_t __s2_712 = __p2_712; \
58742		int64x2_t __rev0_712; __rev0_712 = __builtin_shufflevector(__s0_712, __s0_712, 1, 0); \
58743		int32x4_t __rev1_712; __rev1_712 = __builtin_shufflevector(__s1_712, __s1_712, 3, 2, 1, 0); \
58744		int32x2_t __rev2_712; __rev2_712 = __builtin_shufflevector(__s2_712, __s2_712, 1, 0); \
58745		__ret_712 = __noswap_vqdmlal_s32(__rev0_712, __noswap_vget_high_s32(__rev1_712), __noswap_splat_lane_s32(__rev2_712, __p3_712)); \
58746		__ret_712 = __builtin_shufflevector(__ret_712, __ret_712, 1, 0); \
58747		__ret_712; \
	58606	#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	58607	poly64x2_t __ret; \
	58608	poly64_t __s0 = __p0; \
	58609	poly64x2_t __s1 = __p1; \
	58610	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	58611	__ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__rev1, __p2); \
	58612	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	58613	__ret; \
	58614	})
	58615	#define __noswap_vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	58616	poly64x2_t __ret; \
	58617	poly64_t __s0 = __p0; \
	58618	poly64x2_t __s1 = __p1; \
	58619	__ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
	58620	__ret; \
58748	58621	})
58749	58622	#endif
58750	58623
58751	58624	#ifdef __LITTLE_ENDIAN__
58752		#define vqdmlal_high_lane_s16(__p0_713, __p1_713, __p2_713, __p3_713) __extension__ ({ \
58753		int32x4_t __ret_713; \
58754		int32x4_t __s0_713 = __p0_713; \
58755		int16x8_t __s1_713 = __p1_713; \
58756		int16x4_t __s2_713 = __p2_713; \
58757		__ret_713 = vqdmlal_s16(__s0_713, vget_high_s16(__s1_713), splat_lane_s16(__s2_713, __p3_713)); \
58758		__ret_713; \
	58625	#define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	58626	float64x2_t __ret; \
	58627	float64_t __s0 = __p0; \
	58628	float64x2_t __s1 = __p1; \
	58629	__ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
	58630	__ret; \
58759	58631	})
58760	58632	#else
58761		#define vqdmlal_high_lane_s16(__p0_714, __p1_714, __p2_714, __p3_714) __extension__ ({ \
58762		int32x4_t __ret_714; \
58763		int32x4_t __s0_714 = __p0_714; \
58764		int16x8_t __s1_714 = __p1_714; \
58765		int16x4_t __s2_714 = __p2_714; \
58766		int32x4_t __rev0_714; __rev0_714 = __builtin_shufflevector(__s0_714, __s0_714, 3, 2, 1, 0); \
58767		int16x8_t __rev1_714; __rev1_714 = __builtin_shufflevector(__s1_714, __s1_714, 7, 6, 5, 4, 3, 2, 1, 0); \
58768		int16x4_t __rev2_714; __rev2_714 = __builtin_shufflevector(__s2_714, __s2_714, 3, 2, 1, 0); \
58769		__ret_714 = __noswap_vqdmlal_s16(__rev0_714, __noswap_vget_high_s16(__rev1_714), __noswap_splat_lane_s16(__rev2_714, __p3_714)); \
58770		__ret_714 = __builtin_shufflevector(__ret_714, __ret_714, 3, 2, 1, 0); \
58771		__ret_714; \
	58633	#define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	58634	float64x2_t __ret; \
	58635	float64_t __s0 = __p0; \
	58636	float64x2_t __s1 = __p1; \
	58637	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	58638	__ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__rev1, __p2); \
	58639	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	58640	__ret; \
	58641	})
	58642	#define __noswap_vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	58643	float64x2_t __ret; \
	58644	float64_t __s0 = __p0; \
	58645	float64x2_t __s1 = __p1; \
	58646	__ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
	58647	__ret; \
58772	58648	})
58773	58649	#endif
58774	58650
	58651	#define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	58652	float64x1_t __ret; \
	58653	float64_t __s0 = __p0; \
	58654	float64x1_t __s1 = __p1; \
	58655	__ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (float64x1_t)__s1, __p2); \
	58656	__ret; \
	58657	})
	58658	__ai uint64_t vshld_u64(uint64_t __p0, int64_t __p1) {
	58659	uint64_t __ret;
	58660	__ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
	58661	return __ret;
	58662	}
	58663	__ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
	58664	int64_t __ret;
	58665	__ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
	58666	return __ret;
	58667	}
	58668	#define vshld_n_u64(__p0, __p1) __extension__ ({ \
	58669	uint64_t __ret; \
	58670	uint64_t __s0 = __p0; \
	58671	__ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
	58672	__ret; \
	58673	})
	58674	#define vshld_n_s64(__p0, __p1) __extension__ ({ \
	58675	int64_t __ret; \
	58676	int64_t __s0 = __p0; \
	58677	__ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
	58678	__ret; \
	58679	})
58775	58680	#ifdef __LITTLE_ENDIAN__
58776		#define vqdmlal_high_laneq_s32(__p0_715, __p1_715, __p2_715, __p3_715) __extension__ ({ \
58777		int64x2_t __ret_715; \
58778		int64x2_t __s0_715 = __p0_715; \
58779		int32x4_t __s1_715 = __p1_715; \
58780		int32x4_t __s2_715 = __p2_715; \
58781		__ret_715 = vqdmlal_s32(__s0_715, vget_high_s32(__s1_715), splat_laneq_s32(__s2_715, __p3_715)); \
	58681	#define vshll_high_n_u8(__p0_715, __p1_715) __extension__ ({ \
	58682	uint16x8_t __ret_715; \
	58683	uint8x16_t __s0_715 = __p0_715; \
	58684	__ret_715 = (uint16x8_t)(vshll_n_u8(vget_high_u8(__s0_715), __p1_715)); \
58782	58685	__ret_715; \
58783	58686	})
58784	58687	#else
58785		#define vqdmlal_high_laneq_s32(__p0_716, __p1_716, __p2_716, __p3_716) __extension__ ({ \
58786		int64x2_t __ret_716; \
58787		int64x2_t __s0_716 = __p0_716; \
58788		int32x4_t __s1_716 = __p1_716; \
58789		int32x4_t __s2_716 = __p2_716; \
58790		int64x2_t __rev0_716; __rev0_716 = __builtin_shufflevector(__s0_716, __s0_716, 1, 0); \
58791		int32x4_t __rev1_716; __rev1_716 = __builtin_shufflevector(__s1_716, __s1_716, 3, 2, 1, 0); \
58792		int32x4_t __rev2_716; __rev2_716 = __builtin_shufflevector(__s2_716, __s2_716, 3, 2, 1, 0); \
58793		__ret_716 = __noswap_vqdmlal_s32(__rev0_716, __noswap_vget_high_s32(__rev1_716), __noswap_splat_laneq_s32(__rev2_716, __p3_716)); \
58794		__ret_716 = __builtin_shufflevector(__ret_716, __ret_716, 1, 0); \
	58688	#define vshll_high_n_u8(__p0_716, __p1_716) __extension__ ({ \
	58689	uint16x8_t __ret_716; \
	58690	uint8x16_t __s0_716 = __p0_716; \
	58691	uint8x16_t __rev0_716; __rev0_716 = __builtin_shufflevector(__s0_716, __s0_716, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	58692	__ret_716 = (uint16x8_t)(__noswap_vshll_n_u8(__noswap_vget_high_u8(__rev0_716), __p1_716)); \
	58693	__ret_716 = __builtin_shufflevector(__ret_716, __ret_716, 7, 6, 5, 4, 3, 2, 1, 0); \
58795	58694	__ret_716; \
58796	58695	})
58797	58696	#endif
58798	58697
58799	58698	#ifdef __LITTLE_ENDIAN__
58800		#define vqdmlal_high_laneq_s16(__p0_717, __p1_717, __p2_717, __p3_717) __extension__ ({ \
58801		int32x4_t __ret_717; \
58802		int32x4_t __s0_717 = __p0_717; \
58803		int16x8_t __s1_717 = __p1_717; \
58804		int16x8_t __s2_717 = __p2_717; \
58805		__ret_717 = vqdmlal_s16(__s0_717, vget_high_s16(__s1_717), splat_laneq_s16(__s2_717, __p3_717)); \
	58699	#define vshll_high_n_u32(__p0_717, __p1_717) __extension__ ({ \
	58700	uint64x2_t __ret_717; \
	58701	uint32x4_t __s0_717 = __p0_717; \
	58702	__ret_717 = (uint64x2_t)(vshll_n_u32(vget_high_u32(__s0_717), __p1_717)); \
58806	58703	__ret_717; \
58807	58704	})
58808	58705	#else
58809		#define vqdmlal_high_laneq_s16(__p0_718, __p1_718, __p2_718, __p3_718) __extension__ ({ \
58810		int32x4_t __ret_718; \
58811		int32x4_t __s0_718 = __p0_718; \
58812		int16x8_t __s1_718 = __p1_718; \
58813		int16x8_t __s2_718 = __p2_718; \
58814		int32x4_t __rev0_718; __rev0_718 = __builtin_shufflevector(__s0_718, __s0_718, 3, 2, 1, 0); \
58815		int16x8_t __rev1_718; __rev1_718 = __builtin_shufflevector(__s1_718, __s1_718, 7, 6, 5, 4, 3, 2, 1, 0); \
58816		int16x8_t __rev2_718; __rev2_718 = __builtin_shufflevector(__s2_718, __s2_718, 7, 6, 5, 4, 3, 2, 1, 0); \
58817		__ret_718 = __noswap_vqdmlal_s16(__rev0_718, __noswap_vget_high_s16(__rev1_718), __noswap_splat_laneq_s16(__rev2_718, __p3_718)); \
58818		__ret_718 = __builtin_shufflevector(__ret_718, __ret_718, 3, 2, 1, 0); \
	58706	#define vshll_high_n_u32(__p0_718, __p1_718) __extension__ ({ \
	58707	uint64x2_t __ret_718; \
	58708	uint32x4_t __s0_718 = __p0_718; \
	58709	uint32x4_t __rev0_718; __rev0_718 = __builtin_shufflevector(__s0_718, __s0_718, 3, 2, 1, 0); \
	58710	__ret_718 = (uint64x2_t)(__noswap_vshll_n_u32(__noswap_vget_high_u32(__rev0_718), __p1_718)); \
	58711	__ret_718 = __builtin_shufflevector(__ret_718, __ret_718, 1, 0); \
58819	58712	__ret_718; \
58820	58713	})
58821	58714	#endif
58822	58715
58823	58716	#ifdef __LITTLE_ENDIAN__
58824		__ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
58825		int64x2_t __ret;
58826		__ret = vqdmlal_n_s32(__p0, vget_high_s32(__p1), __p2);
58827		return __ret;
58828		}
	58717	#define vshll_high_n_u16(__p0_719, __p1_719) __extension__ ({ \
	58718	uint32x4_t __ret_719; \
	58719	uint16x8_t __s0_719 = __p0_719; \
	58720	__ret_719 = (uint32x4_t)(vshll_n_u16(vget_high_u16(__s0_719), __p1_719)); \
	58721	__ret_719; \
	58722	})
58829	58723	#else
58830		__ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
58831		int64x2_t __ret;
58832		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
58833		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58834		__ret = __noswap_vqdmlal_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
58835		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
58836		return __ret;
58837		}
	58724	#define vshll_high_n_u16(__p0_720, __p1_720) __extension__ ({ \
	58725	uint32x4_t __ret_720; \
	58726	uint16x8_t __s0_720 = __p0_720; \
	58727	uint16x8_t __rev0_720; __rev0_720 = __builtin_shufflevector(__s0_720, __s0_720, 7, 6, 5, 4, 3, 2, 1, 0); \
	58728	__ret_720 = (uint32x4_t)(__noswap_vshll_n_u16(__noswap_vget_high_u16(__rev0_720), __p1_720)); \
	58729	__ret_720 = __builtin_shufflevector(__ret_720, __ret_720, 3, 2, 1, 0); \
	58730	__ret_720; \
	58731	})
58838	58732	#endif
58839	58733
58840	58734	#ifdef __LITTLE_ENDIAN__
58841		__ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
58842		int32x4_t __ret;
58843		__ret = vqdmlal_n_s16(__p0, vget_high_s16(__p1), __p2);
58844		return __ret;
58845		}
	58735	#define vshll_high_n_s8(__p0_721, __p1_721) __extension__ ({ \
	58736	int16x8_t __ret_721; \
	58737	int8x16_t __s0_721 = __p0_721; \
	58738	__ret_721 = (int16x8_t)(vshll_n_s8(vget_high_s8(__s0_721), __p1_721)); \
	58739	__ret_721; \
	58740	})
58846	58741	#else
58847		__ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
58848		int32x4_t __ret;
58849		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58850		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58851		__ret = __noswap_vqdmlal_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
58852		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
58853		return __ret;
58854		}
	58742	#define vshll_high_n_s8(__p0_722, __p1_722) __extension__ ({ \
	58743	int16x8_t __ret_722; \
	58744	int8x16_t __s0_722 = __p0_722; \
	58745	int8x16_t __rev0_722; __rev0_722 = __builtin_shufflevector(__s0_722, __s0_722, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	58746	__ret_722 = (int16x8_t)(__noswap_vshll_n_s8(__noswap_vget_high_s8(__rev0_722), __p1_722)); \
	58747	__ret_722 = __builtin_shufflevector(__ret_722, __ret_722, 7, 6, 5, 4, 3, 2, 1, 0); \
	58748	__ret_722; \
	58749	})
58855	58750	#endif
58856	58751
58857	58752	#ifdef __LITTLE_ENDIAN__
58858		#define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58859		int64_t __ret; \
58860		int64_t __s0 = __p0; \
58861		int32_t __s1 = __p1; \
58862		int32x2_t __s2 = __p2; \
58863		__ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __s2, __p3); \
58864		__ret; \
	58753	#define vshll_high_n_s32(__p0_723, __p1_723) __extension__ ({ \
	58754	int64x2_t __ret_723; \
	58755	int32x4_t __s0_723 = __p0_723; \
	58756	__ret_723 = (int64x2_t)(vshll_n_s32(vget_high_s32(__s0_723), __p1_723)); \
	58757	__ret_723; \
58865	58758	})
58866	58759	#else
58867		#define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	58760	#define vshll_high_n_s32(__p0_724, __p1_724) __extension__ ({ \
	58761	int64x2_t __ret_724; \
	58762	int32x4_t __s0_724 = __p0_724; \
	58763	int32x4_t __rev0_724; __rev0_724 = __builtin_shufflevector(__s0_724, __s0_724, 3, 2, 1, 0); \
	58764	__ret_724 = (int64x2_t)(__noswap_vshll_n_s32(__noswap_vget_high_s32(__rev0_724), __p1_724)); \
	58765	__ret_724 = __builtin_shufflevector(__ret_724, __ret_724, 1, 0); \
	58766	__ret_724; \
	58767	})
	58768	#endif
	58769
	58770	#ifdef __LITTLE_ENDIAN__
	58771	#define vshll_high_n_s16(__p0_725, __p1_725) __extension__ ({ \
	58772	int32x4_t __ret_725; \
	58773	int16x8_t __s0_725 = __p0_725; \
	58774	__ret_725 = (int32x4_t)(vshll_n_s16(vget_high_s16(__s0_725), __p1_725)); \
	58775	__ret_725; \
	58776	})
	58777	#else
	58778	#define vshll_high_n_s16(__p0_726, __p1_726) __extension__ ({ \
	58779	int32x4_t __ret_726; \
	58780	int16x8_t __s0_726 = __p0_726; \
	58781	int16x8_t __rev0_726; __rev0_726 = __builtin_shufflevector(__s0_726, __s0_726, 7, 6, 5, 4, 3, 2, 1, 0); \
	58782	__ret_726 = (int32x4_t)(__noswap_vshll_n_s16(__noswap_vget_high_s16(__rev0_726), __p1_726)); \
	58783	__ret_726 = __builtin_shufflevector(__ret_726, __ret_726, 3, 2, 1, 0); \
	58784	__ret_726; \
	58785	})
	58786	#endif
	58787
	58788	#define vshrd_n_u64(__p0, __p1) __extension__ ({ \
	58789	uint64_t __ret; \
	58790	uint64_t __s0 = __p0; \
	58791	__ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
	58792	__ret; \
	58793	})
	58794	#define vshrd_n_s64(__p0, __p1) __extension__ ({ \
58868	58795	int64_t __ret; \
58869	58796	int64_t __s0 = __p0; \
58870		int32_t __s1 = __p1; \
58871		int32x2_t __s2 = __p2; \
58872		int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
58873		__ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __rev2, __p3); \
	58797	__ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
58874	58798	__ret; \
58875	58799	})
	58800	#ifdef __LITTLE_ENDIAN__
	58801	#define vshrn_high_n_u32(__p0_727, __p1_727, __p2_727) __extension__ ({ \
	58802	uint16x8_t __ret_727; \
	58803	uint16x4_t __s0_727 = __p0_727; \
	58804	uint32x4_t __s1_727 = __p1_727; \
	58805	__ret_727 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_727), (uint16x4_t)(vshrn_n_u32(__s1_727, __p2_727)))); \
	58806	__ret_727; \
	58807	})
	58808	#else
	58809	#define vshrn_high_n_u32(__p0_728, __p1_728, __p2_728) __extension__ ({ \
	58810	uint16x8_t __ret_728; \
	58811	uint16x4_t __s0_728 = __p0_728; \
	58812	uint32x4_t __s1_728 = __p1_728; \
	58813	uint16x4_t __rev0_728; __rev0_728 = __builtin_shufflevector(__s0_728, __s0_728, 3, 2, 1, 0); \
	58814	uint32x4_t __rev1_728; __rev1_728 = __builtin_shufflevector(__s1_728, __s1_728, 3, 2, 1, 0); \
	58815	__ret_728 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_728), (uint16x4_t)(__noswap_vshrn_n_u32(__rev1_728, __p2_728)))); \
	58816	__ret_728 = __builtin_shufflevector(__ret_728, __ret_728, 7, 6, 5, 4, 3, 2, 1, 0); \
	58817	__ret_728; \
	58818	})
58876	58819	#endif
58877	58820
58878	58821	#ifdef __LITTLE_ENDIAN__
58879		#define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58880		int32_t __ret; \
58881		int32_t __s0 = __p0; \
58882		int16_t __s1 = __p1; \
58883		int16x4_t __s2 = __p2; \
58884		__ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __s2, __p3); \
58885		__ret; \
	58822	#define vshrn_high_n_u64(__p0_729, __p1_729, __p2_729) __extension__ ({ \
	58823	uint32x4_t __ret_729; \
	58824	uint32x2_t __s0_729 = __p0_729; \
	58825	uint64x2_t __s1_729 = __p1_729; \
	58826	__ret_729 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_729), (uint32x2_t)(vshrn_n_u64(__s1_729, __p2_729)))); \
	58827	__ret_729; \
58886	58828	})
58887	58829	#else
58888		#define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58889		int32_t __ret; \
58890		int32_t __s0 = __p0; \
58891		int16_t __s1 = __p1; \
58892		int16x4_t __s2 = __p2; \
58893		int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
58894		__ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __rev2, __p3); \
58895		__ret; \
	58830	#define vshrn_high_n_u64(__p0_730, __p1_730, __p2_730) __extension__ ({ \
	58831	uint32x4_t __ret_730; \
	58832	uint32x2_t __s0_730 = __p0_730; \
	58833	uint64x2_t __s1_730 = __p1_730; \
	58834	uint32x2_t __rev0_730; __rev0_730 = __builtin_shufflevector(__s0_730, __s0_730, 1, 0); \
	58835	uint64x2_t __rev1_730; __rev1_730 = __builtin_shufflevector(__s1_730, __s1_730, 1, 0); \
	58836	__ret_730 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_730), (uint32x2_t)(__noswap_vshrn_n_u64(__rev1_730, __p2_730)))); \
	58837	__ret_730 = __builtin_shufflevector(__ret_730, __ret_730, 3, 2, 1, 0); \
	58838	__ret_730; \
58896	58839	})
58897	58840	#endif
58898	58841
58899	58842	#ifdef __LITTLE_ENDIAN__
58900		#define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58901		int64_t __ret; \
58902		int64_t __s0 = __p0; \
58903		int32_t __s1 = __p1; \
58904		int32x4_t __s2 = __p2; \
58905		__ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __s2, __p3); \
58906		__ret; \
	58843	#define vshrn_high_n_u16(__p0_731, __p1_731, __p2_731) __extension__ ({ \
	58844	uint8x16_t __ret_731; \
	58845	uint8x8_t __s0_731 = __p0_731; \
	58846	uint16x8_t __s1_731 = __p1_731; \
	58847	__ret_731 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_731), (uint8x8_t)(vshrn_n_u16(__s1_731, __p2_731)))); \
	58848	__ret_731; \
58907	58849	})
58908	58850	#else
58909		#define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58910		int64_t __ret; \
58911		int64_t __s0 = __p0; \
58912		int32_t __s1 = __p1; \
58913		int32x4_t __s2 = __p2; \
58914		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
58915		__ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __rev2, __p3); \
58916		__ret; \
	58851	#define vshrn_high_n_u16(__p0_732, __p1_732, __p2_732) __extension__ ({ \
	58852	uint8x16_t __ret_732; \
	58853	uint8x8_t __s0_732 = __p0_732; \
	58854	uint16x8_t __s1_732 = __p1_732; \
	58855	uint8x8_t __rev0_732; __rev0_732 = __builtin_shufflevector(__s0_732, __s0_732, 7, 6, 5, 4, 3, 2, 1, 0); \
	58856	uint16x8_t __rev1_732; __rev1_732 = __builtin_shufflevector(__s1_732, __s1_732, 7, 6, 5, 4, 3, 2, 1, 0); \
	58857	__ret_732 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_732), (uint8x8_t)(__noswap_vshrn_n_u16(__rev1_732, __p2_732)))); \
	58858	__ret_732 = __builtin_shufflevector(__ret_732, __ret_732, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	58859	__ret_732; \
58917	58860	})
58918	58861	#endif
58919	58862
58920	58863	#ifdef __LITTLE_ENDIAN__
58921		#define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58922		int32_t __ret; \
58923		int32_t __s0 = __p0; \
58924		int16_t __s1 = __p1; \
58925		int16x8_t __s2 = __p2; \
58926		__ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __s2, __p3); \
58927		__ret; \
	58864	#define vshrn_high_n_s32(__p0_733, __p1_733, __p2_733) __extension__ ({ \
	58865	int16x8_t __ret_733; \
	58866	int16x4_t __s0_733 = __p0_733; \
	58867	int32x4_t __s1_733 = __p1_733; \
	58868	__ret_733 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_733), (int16x4_t)(vshrn_n_s32(__s1_733, __p2_733)))); \
	58869	__ret_733; \
58928	58870	})
58929	58871	#else
58930		#define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58931		int32_t __ret; \
58932		int32_t __s0 = __p0; \
58933		int16_t __s1 = __p1; \
58934		int16x8_t __s2 = __p2; \
58935		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
58936		__ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __rev2, __p3); \
58937		__ret; \
	58872	#define vshrn_high_n_s32(__p0_734, __p1_734, __p2_734) __extension__ ({ \
	58873	int16x8_t __ret_734; \
	58874	int16x4_t __s0_734 = __p0_734; \
	58875	int32x4_t __s1_734 = __p1_734; \
	58876	int16x4_t __rev0_734; __rev0_734 = __builtin_shufflevector(__s0_734, __s0_734, 3, 2, 1, 0); \
	58877	int32x4_t __rev1_734; __rev1_734 = __builtin_shufflevector(__s1_734, __s1_734, 3, 2, 1, 0); \
	58878	__ret_734 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_734), (int16x4_t)(__noswap_vshrn_n_s32(__rev1_734, __p2_734)))); \
	58879	__ret_734 = __builtin_shufflevector(__ret_734, __ret_734, 7, 6, 5, 4, 3, 2, 1, 0); \
	58880	__ret_734; \
58938	58881	})
58939	58882	#endif
58940	58883
58941	58884	#ifdef __LITTLE_ENDIAN__
58942		#define vqdmlal_laneq_s32(__p0_719, __p1_719, __p2_719, __p3_719) __extension__ ({ \
58943		int64x2_t __ret_719; \
58944		int64x2_t __s0_719 = __p0_719; \
58945		int32x2_t __s1_719 = __p1_719; \
58946		int32x4_t __s2_719 = __p2_719; \
58947		__ret_719 = vqdmlal_s32(__s0_719, __s1_719, splat_laneq_s32(__s2_719, __p3_719)); \
58948		__ret_719; \
	58885	#define vshrn_high_n_s64(__p0_735, __p1_735, __p2_735) __extension__ ({ \
	58886	int32x4_t __ret_735; \
	58887	int32x2_t __s0_735 = __p0_735; \
	58888	int64x2_t __s1_735 = __p1_735; \
	58889	__ret_735 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_735), (int32x2_t)(vshrn_n_s64(__s1_735, __p2_735)))); \
	58890	__ret_735; \
58949	58891	})
58950	58892	#else
58951		#define vqdmlal_laneq_s32(__p0_720, __p1_720, __p2_720, __p3_720) __extension__ ({ \
58952		int64x2_t __ret_720; \
58953		int64x2_t __s0_720 = __p0_720; \
58954		int32x2_t __s1_720 = __p1_720; \
58955		int32x4_t __s2_720 = __p2_720; \
58956		int64x2_t __rev0_720; __rev0_720 = __builtin_shufflevector(__s0_720, __s0_720, 1, 0); \
58957		int32x2_t __rev1_720; __rev1_720 = __builtin_shufflevector(__s1_720, __s1_720, 1, 0); \
58958		int32x4_t __rev2_720; __rev2_720 = __builtin_shufflevector(__s2_720, __s2_720, 3, 2, 1, 0); \
58959		__ret_720 = __noswap_vqdmlal_s32(__rev0_720, __rev1_720, __noswap_splat_laneq_s32(__rev2_720, __p3_720)); \
58960		__ret_720 = __builtin_shufflevector(__ret_720, __ret_720, 1, 0); \
58961		__ret_720; \
	58893	#define vshrn_high_n_s64(__p0_736, __p1_736, __p2_736) __extension__ ({ \
	58894	int32x4_t __ret_736; \
	58895	int32x2_t __s0_736 = __p0_736; \
	58896	int64x2_t __s1_736 = __p1_736; \
	58897	int32x2_t __rev0_736; __rev0_736 = __builtin_shufflevector(__s0_736, __s0_736, 1, 0); \
	58898	int64x2_t __rev1_736; __rev1_736 = __builtin_shufflevector(__s1_736, __s1_736, 1, 0); \
	58899	__ret_736 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_736), (int32x2_t)(__noswap_vshrn_n_s64(__rev1_736, __p2_736)))); \
	58900	__ret_736 = __builtin_shufflevector(__ret_736, __ret_736, 3, 2, 1, 0); \
	58901	__ret_736; \
58962	58902	})
58963	58903	#endif
58964	58904
58965	58905	#ifdef __LITTLE_ENDIAN__
58966		#define vqdmlal_laneq_s16(__p0_721, __p1_721, __p2_721, __p3_721) __extension__ ({ \
58967		int32x4_t __ret_721; \
58968		int32x4_t __s0_721 = __p0_721; \
58969		int16x4_t __s1_721 = __p1_721; \
58970		int16x8_t __s2_721 = __p2_721; \
58971		__ret_721 = vqdmlal_s16(__s0_721, __s1_721, splat_laneq_s16(__s2_721, __p3_721)); \
58972		__ret_721; \
	58906	#define vshrn_high_n_s16(__p0_737, __p1_737, __p2_737) __extension__ ({ \
	58907	int8x16_t __ret_737; \
	58908	int8x8_t __s0_737 = __p0_737; \
	58909	int16x8_t __s1_737 = __p1_737; \
	58910	__ret_737 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_737), (int8x8_t)(vshrn_n_s16(__s1_737, __p2_737)))); \
	58911	__ret_737; \
58973	58912	})
58974	58913	#else
58975		#define vqdmlal_laneq_s16(__p0_722, __p1_722, __p2_722, __p3_722) __extension__ ({ \
58976		int32x4_t __ret_722; \
58977		int32x4_t __s0_722 = __p0_722; \
58978		int16x4_t __s1_722 = __p1_722; \
58979		int16x8_t __s2_722 = __p2_722; \
58980		int32x4_t __rev0_722; __rev0_722 = __builtin_shufflevector(__s0_722, __s0_722, 3, 2, 1, 0); \
58981		int16x4_t __rev1_722; __rev1_722 = __builtin_shufflevector(__s1_722, __s1_722, 3, 2, 1, 0); \
58982		int16x8_t __rev2_722; __rev2_722 = __builtin_shufflevector(__s2_722, __s2_722, 7, 6, 5, 4, 3, 2, 1, 0); \
58983		__ret_722 = __noswap_vqdmlal_s16(__rev0_722, __rev1_722, __noswap_splat_laneq_s16(__rev2_722, __p3_722)); \
58984		__ret_722 = __builtin_shufflevector(__ret_722, __ret_722, 3, 2, 1, 0); \
58985		__ret_722; \
	58914	#define vshrn_high_n_s16(__p0_738, __p1_738, __p2_738) __extension__ ({ \
	58915	int8x16_t __ret_738; \
	58916	int8x8_t __s0_738 = __p0_738; \
	58917	int16x8_t __s1_738 = __p1_738; \
	58918	int8x8_t __rev0_738; __rev0_738 = __builtin_shufflevector(__s0_738, __s0_738, 7, 6, 5, 4, 3, 2, 1, 0); \
	58919	int16x8_t __rev1_738; __rev1_738 = __builtin_shufflevector(__s1_738, __s1_738, 7, 6, 5, 4, 3, 2, 1, 0); \
	58920	__ret_738 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_738), (int8x8_t)(__noswap_vshrn_n_s16(__rev1_738, __p2_738)))); \
	58921	__ret_738 = __builtin_shufflevector(__ret_738, __ret_738, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	58922	__ret_738; \
58986	58923	})
58987	58924	#endif
58988	58925
58989		__ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
58990		int64_t __ret;
58991		__ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
	58926	#define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
	58927	uint64_t __ret; \
	58928	uint64_t __s0 = __p0; \
	58929	uint64_t __s1 = __p1; \
	58930	__ret = (uint64_t) __builtin_neon_vslid_n_u64(__s0, __s1, __p2); \
	58931	__ret; \
	58932	})
	58933	#define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
	58934	int64_t __ret; \
	58935	int64_t __s0 = __p0; \
	58936	int64_t __s1 = __p1; \
	58937	__ret = (int64_t) __builtin_neon_vslid_n_s64(__s0, __s1, __p2); \
	58938	__ret; \
	58939	})
	58940	#define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
	58941	poly64x1_t __ret; \
	58942	poly64x1_t __s0 = __p0; \
	58943	poly64x1_t __s1 = __p1; \
	58944	__ret = (poly64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
	58945	__ret; \
	58946	})
	58947	#ifdef __LITTLE_ENDIAN__
	58948	#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
	58949	poly64x2_t __ret; \
	58950	poly64x2_t __s0 = __p0; \
	58951	poly64x2_t __s1 = __p1; \
	58952	__ret = (poly64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
	58953	__ret; \
	58954	})
	58955	#else
	58956	#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
	58957	poly64x2_t __ret; \
	58958	poly64x2_t __s0 = __p0; \
	58959	poly64x2_t __s1 = __p1; \
	58960	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	58961	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	58962	__ret = (poly64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
	58963	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	58964	__ret; \
	58965	})
	58966	#endif
	58967
	58968	__ai uint8_t vsqaddb_u8(uint8_t __p0, int8_t __p1) {
	58969	uint8_t __ret;
	58970	__ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
58992	58971	return __ret;
58993	58972	}
58994		__ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
58995		int32_t __ret;
58996		__ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
	58973	__ai uint32_t vsqadds_u32(uint32_t __p0, int32_t __p1) {
	58974	uint32_t __ret;
	58975	__ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
	58976	return __ret;
	58977	}
	58978	__ai uint64_t vsqaddd_u64(uint64_t __p0, int64_t __p1) {
	58979	uint64_t __ret;
	58980	__ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
	58981	return __ret;
	58982	}
	58983	__ai uint16_t vsqaddh_u16(uint16_t __p0, int16_t __p1) {
	58984	uint16_t __ret;
	58985	__ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
58997	58986	return __ret;
58998	58987	}
58999	58988	#ifdef __LITTLE_ENDIAN__
59000		__ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
59001		int64x2_t __ret;
59002		__ret = vqdmlsl_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
	58989	__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
	58990	uint8x16_t __ret;
	58991	__ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
59003	58992	return __ret;
59004	58993	}
59005	58994	#else
59006		__ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
59007		int64x2_t __ret;
59008		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59009		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59010		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
59011		__ret = __noswap_vqdmlsl_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
59012		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	58995	__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
	58996	uint8x16_t __ret;
	58997	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	58998	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	58999	__ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
	59000	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59013	59001	return __ret;
59014	59002	}
59015	59003	#endif
59016	59004
59017	59005	#ifdef __LITTLE_ENDIAN__
59018		__ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
59019		int32x4_t __ret;
59020		__ret = vqdmlsl_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
	59006	__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
	59007	uint32x4_t __ret;
	59008	__ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
59021	59009	return __ret;
59022	59010	}
59023	59011	#else
59024		__ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
59025		int32x4_t __ret;
59026		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59027		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59028		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
59029		__ret = __noswap_vqdmlsl_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
	59012	__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
	59013	uint32x4_t __ret;
	59014	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	59015	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	59016	__ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
59030	59017	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59031	59018	return __ret;
59032	59019	}
59033	59020	#endif
59034	59021
59035	59022	#ifdef __LITTLE_ENDIAN__
59036		#define vqdmlsl_high_lane_s32(__p0_723, __p1_723, __p2_723, __p3_723) __extension__ ({ \
59037		int64x2_t __ret_723; \
59038		int64x2_t __s0_723 = __p0_723; \
59039		int32x4_t __s1_723 = __p1_723; \
59040		int32x2_t __s2_723 = __p2_723; \
59041		__ret_723 = vqdmlsl_s32(__s0_723, vget_high_s32(__s1_723), splat_lane_s32(__s2_723, __p3_723)); \
59042		__ret_723; \
59043		})
	59023	__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
	59024	uint64x2_t __ret;
	59025	__ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	59026	return __ret;
	59027	}
59044	59028	#else
59045		#define vqdmlsl_high_lane_s32(__p0_724, __p1_724, __p2_724, __p3_724) __extension__ ({ \
59046		int64x2_t __ret_724; \
59047		int64x2_t __s0_724 = __p0_724; \
59048		int32x4_t __s1_724 = __p1_724; \
59049		int32x2_t __s2_724 = __p2_724; \
59050		int64x2_t __rev0_724; __rev0_724 = __builtin_shufflevector(__s0_724, __s0_724, 1, 0); \
59051		int32x4_t __rev1_724; __rev1_724 = __builtin_shufflevector(__s1_724, __s1_724, 3, 2, 1, 0); \
59052		int32x2_t __rev2_724; __rev2_724 = __builtin_shufflevector(__s2_724, __s2_724, 1, 0); \
59053		__ret_724 = __noswap_vqdmlsl_s32(__rev0_724, __noswap_vget_high_s32(__rev1_724), __noswap_splat_lane_s32(__rev2_724, __p3_724)); \
59054		__ret_724 = __builtin_shufflevector(__ret_724, __ret_724, 1, 0); \
59055		__ret_724; \
59056		})
	59029	__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
	59030	uint64x2_t __ret;
	59031	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	59032	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	59033	__ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	59034	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	59035	return __ret;
	59036	}
59057	59037	#endif
59058	59038
59059	59039	#ifdef __LITTLE_ENDIAN__
59060		#define vqdmlsl_high_lane_s16(__p0_725, __p1_725, __p2_725, __p3_725) __extension__ ({ \
59061		int32x4_t __ret_725; \
59062		int32x4_t __s0_725 = __p0_725; \
59063		int16x8_t __s1_725 = __p1_725; \
59064		int16x4_t __s2_725 = __p2_725; \
59065		__ret_725 = vqdmlsl_s16(__s0_725, vget_high_s16(__s1_725), splat_lane_s16(__s2_725, __p3_725)); \
59066		__ret_725; \
59067		})
	59040	__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
	59041	uint16x8_t __ret;
	59042	__ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	59043	return __ret;
	59044	}
59068	59045	#else
59069		#define vqdmlsl_high_lane_s16(__p0_726, __p1_726, __p2_726, __p3_726) __extension__ ({ \
59070		int32x4_t __ret_726; \
59071		int32x4_t __s0_726 = __p0_726; \
59072		int16x8_t __s1_726 = __p1_726; \
59073		int16x4_t __s2_726 = __p2_726; \
59074		int32x4_t __rev0_726; __rev0_726 = __builtin_shufflevector(__s0_726, __s0_726, 3, 2, 1, 0); \
59075		int16x8_t __rev1_726; __rev1_726 = __builtin_shufflevector(__s1_726, __s1_726, 7, 6, 5, 4, 3, 2, 1, 0); \
59076		int16x4_t __rev2_726; __rev2_726 = __builtin_shufflevector(__s2_726, __s2_726, 3, 2, 1, 0); \
59077		__ret_726 = __noswap_vqdmlsl_s16(__rev0_726, __noswap_vget_high_s16(__rev1_726), __noswap_splat_lane_s16(__rev2_726, __p3_726)); \
59078		__ret_726 = __builtin_shufflevector(__ret_726, __ret_726, 3, 2, 1, 0); \
59079		__ret_726; \
59080		})
	59046	__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
	59047	uint16x8_t __ret;
	59048	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	59049	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	59050	__ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	59051	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	59052	return __ret;
	59053	}
	59054	#endif
	59055
	59056	#ifdef __LITTLE_ENDIAN__
	59057	__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
	59058	uint8x8_t __ret;
	59059	__ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
	59060	return __ret;
	59061	}
	59062	#else
	59063	__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
	59064	uint8x8_t __ret;
	59065	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	59066	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	59067	__ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
	59068	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	59069	return __ret;
	59070	}
59081	59071	#endif
59082	59072
59083	59073	#ifdef __LITTLE_ENDIAN__
59084		#define vqdmlsl_high_laneq_s32(__p0_727, __p1_727, __p2_727, __p3_727) __extension__ ({ \
59085		int64x2_t __ret_727; \
59086		int64x2_t __s0_727 = __p0_727; \
59087		int32x4_t __s1_727 = __p1_727; \
59088		int32x4_t __s2_727 = __p2_727; \
59089		__ret_727 = vqdmlsl_s32(__s0_727, vget_high_s32(__s1_727), splat_laneq_s32(__s2_727, __p3_727)); \
59090		__ret_727; \
59091		})
	59074	__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
	59075	uint32x2_t __ret;
	59076	__ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
	59077	return __ret;
	59078	}
59092	59079	#else
59093		#define vqdmlsl_high_laneq_s32(__p0_728, __p1_728, __p2_728, __p3_728) __extension__ ({ \
59094		int64x2_t __ret_728; \
59095		int64x2_t __s0_728 = __p0_728; \
59096		int32x4_t __s1_728 = __p1_728; \
59097		int32x4_t __s2_728 = __p2_728; \
59098		int64x2_t __rev0_728; __rev0_728 = __builtin_shufflevector(__s0_728, __s0_728, 1, 0); \
59099		int32x4_t __rev1_728; __rev1_728 = __builtin_shufflevector(__s1_728, __s1_728, 3, 2, 1, 0); \
59100		int32x4_t __rev2_728; __rev2_728 = __builtin_shufflevector(__s2_728, __s2_728, 3, 2, 1, 0); \
59101		__ret_728 = __noswap_vqdmlsl_s32(__rev0_728, __noswap_vget_high_s32(__rev1_728), __noswap_splat_laneq_s32(__rev2_728, __p3_728)); \
59102		__ret_728 = __builtin_shufflevector(__ret_728, __ret_728, 1, 0); \
59103		__ret_728; \
59104		})
	59080	__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
	59081	uint32x2_t __ret;
	59082	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	59083	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	59084	__ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
	59085	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	59086	return __ret;
	59087	}
59105	59088	#endif
59106	59089
	59090	__ai uint64x1_t vsqadd_u64(uint64x1_t __p0, int64x1_t __p1) {
	59091	uint64x1_t __ret;
	59092	__ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
	59093	return __ret;
	59094	}
59107	59095	#ifdef __LITTLE_ENDIAN__
59108		#define vqdmlsl_high_laneq_s16(__p0_729, __p1_729, __p2_729, __p3_729) __extension__ ({ \
59109		int32x4_t __ret_729; \
59110		int32x4_t __s0_729 = __p0_729; \
59111		int16x8_t __s1_729 = __p1_729; \
59112		int16x8_t __s2_729 = __p2_729; \
59113		__ret_729 = vqdmlsl_s16(__s0_729, vget_high_s16(__s1_729), splat_laneq_s16(__s2_729, __p3_729)); \
59114		__ret_729; \
59115		})
	59096	__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
	59097	uint16x4_t __ret;
	59098	__ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
	59099	return __ret;
	59100	}
59116	59101	#else
59117		#define vqdmlsl_high_laneq_s16(__p0_730, __p1_730, __p2_730, __p3_730) __extension__ ({ \
59118		int32x4_t __ret_730; \
59119		int32x4_t __s0_730 = __p0_730; \
59120		int16x8_t __s1_730 = __p1_730; \
59121		int16x8_t __s2_730 = __p2_730; \
59122		int32x4_t __rev0_730; __rev0_730 = __builtin_shufflevector(__s0_730, __s0_730, 3, 2, 1, 0); \
59123		int16x8_t __rev1_730; __rev1_730 = __builtin_shufflevector(__s1_730, __s1_730, 7, 6, 5, 4, 3, 2, 1, 0); \
59124		int16x8_t __rev2_730; __rev2_730 = __builtin_shufflevector(__s2_730, __s2_730, 7, 6, 5, 4, 3, 2, 1, 0); \
59125		__ret_730 = __noswap_vqdmlsl_s16(__rev0_730, __noswap_vget_high_s16(__rev1_730), __noswap_splat_laneq_s16(__rev2_730, __p3_730)); \
59126		__ret_730 = __builtin_shufflevector(__ret_730, __ret_730, 3, 2, 1, 0); \
59127		__ret_730; \
59128		})
	59102	__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
	59103	uint16x4_t __ret;
	59104	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	59105	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	59106	__ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
	59107	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	59108	return __ret;
	59109	}
59129	59110	#endif
59130	59111
59131	59112	#ifdef __LITTLE_ENDIAN__
59132		__ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
59133		int64x2_t __ret;
59134		__ret = vqdmlsl_n_s32(__p0, vget_high_s32(__p1), __p2);
	59113	__ai float64x2_t vsqrtq_f64(float64x2_t __p0) {
	59114	float64x2_t __ret;
	59115	__ret = (float64x2_t) __builtin_neon_vsqrtq_v((int8x16_t)__p0, 42);
59135	59116	return __ret;
59136	59117	}
59137	59118	#else
59138		__ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
59139		int64x2_t __ret;
59140		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59141		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59142		__ret = __noswap_vqdmlsl_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
	59119	__ai float64x2_t vsqrtq_f64(float64x2_t __p0) {
	59120	float64x2_t __ret;
	59121	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	59122	__ret = (float64x2_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 42);
59143	59123	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
59144	59124	return __ret;
59145	59125	}
59146	59126	#endif
59147	59127
59148	59128	#ifdef __LITTLE_ENDIAN__
59149		__ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
59150		int32x4_t __ret;
59151		__ret = vqdmlsl_n_s16(__p0, vget_high_s16(__p1), __p2);
	59129	__ai float32x4_t vsqrtq_f32(float32x4_t __p0) {
	59130	float32x4_t __ret;
	59131	__ret = (float32x4_t) __builtin_neon_vsqrtq_v((int8x16_t)__p0, 41);
59152	59132	return __ret;
59153	59133	}
59154	59134	#else
59155		__ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
59156		int32x4_t __ret;
59157		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59158		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59159		__ret = __noswap_vqdmlsl_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
	59135	__ai float32x4_t vsqrtq_f32(float32x4_t __p0) {
	59136	float32x4_t __ret;
	59137	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	59138	__ret = (float32x4_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 41);
59160	59139	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59161	59140	return __ret;
59162	59141	}
59163	59142	#endif
59164	59143
	59144	__ai float64x1_t vsqrt_f64(float64x1_t __p0) {
	59145	float64x1_t __ret;
	59146	__ret = (float64x1_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 10);
	59147	return __ret;
	59148	}
59165	59149	#ifdef __LITTLE_ENDIAN__
59166		#define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
59167		int64_t __ret; \
59168		int64_t __s0 = __p0; \
59169		int32_t __s1 = __p1; \
59170		int32x2_t __s2 = __p2; \
59171		__ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __s2, __p3); \
	59150	__ai float32x2_t vsqrt_f32(float32x2_t __p0) {
	59151	float32x2_t __ret;
	59152	__ret = (float32x2_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 9);
	59153	return __ret;
	59154	}
	59155	#else
	59156	__ai float32x2_t vsqrt_f32(float32x2_t __p0) {
	59157	float32x2_t __ret;
	59158	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	59159	__ret = (float32x2_t) __builtin_neon_vsqrt_v((int8x8_t)__rev0, 9);
	59160	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	59161	return __ret;
	59162	}
	59163	#endif
	59164
	59165	#define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
	59166	uint64_t __ret; \
	59167	uint64_t __s0 = __p0; \
	59168	uint64_t __s1 = __p1; \
	59169	__ret = (uint64_t) __builtin_neon_vsrad_n_u64(__s0, __s1, __p2); \
59172	59170	__ret; \
59173	59171	})
59174		#else
59175		#define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	59172	#define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
59176	59173	int64_t __ret; \
59177	59174	int64_t __s0 = __p0; \
59178		int32_t __s1 = __p1; \
59179		int32x2_t __s2 = __p2; \
59180		int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
59181		__ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __rev2, __p3); \
59182		__ret; \
59183		})
59184		#endif
59185
59186		#ifdef __LITTLE_ENDIAN__
59187		#define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59188		int32_t __ret; \
59189		int32_t __s0 = __p0; \
59190		int16_t __s1 = __p1; \
59191		int16x4_t __s2 = __p2; \
59192		__ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __s2, __p3); \
	59175	int64_t __s1 = __p1; \
	59176	__ret = (int64_t) __builtin_neon_vsrad_n_s64(__s0, __s1, __p2); \
59193	59177	__ret; \
59194	59178	})
59195		#else
59196		#define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59197		int32_t __ret; \
59198		int32_t __s0 = __p0; \
59199		int16_t __s1 = __p1; \
59200		int16x4_t __s2 = __p2; \
59201		int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
59202		__ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __rev2, __p3); \
	59179	#define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
	59180	uint64_t __ret; \
	59181	uint64_t __s0 = __p0; \
	59182	uint64_t __s1 = __p1; \
	59183	__ret = (uint64_t) __builtin_neon_vsrid_n_u64(__s0, __s1, __p2); \
59203	59184	__ret; \
59204	59185	})
59205		#endif
59206
59207		#ifdef __LITTLE_ENDIAN__
59208		#define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
	59186	#define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
59209	59187	int64_t __ret; \
59210	59188	int64_t __s0 = __p0; \
59211		int32_t __s1 = __p1; \
59212		int32x4_t __s2 = __p2; \
59213		__ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __s2, __p3); \
	59189	int64_t __s1 = __p1; \
	59190	__ret = (int64_t) __builtin_neon_vsrid_n_s64(__s0, __s1, __p2); \
59214	59191	__ret; \
59215	59192	})
59216		#else
59217		#define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
59218		int64_t __ret; \
59219		int64_t __s0 = __p0; \
59220		int32_t __s1 = __p1; \
59221		int32x4_t __s2 = __p2; \
59222		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
59223		__ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __rev2, __p3); \
	59193	#define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
	59194	poly64x1_t __ret; \
	59195	poly64x1_t __s0 = __p0; \
	59196	poly64x1_t __s1 = __p1; \
	59197	__ret = (poly64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
59224	59198	__ret; \
59225	59199	})
59226		#endif
59227
59228	59200	#ifdef __LITTLE_ENDIAN__
59229		#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59230		int32_t __ret; \
59231		int32_t __s0 = __p0; \
59232		int16_t __s1 = __p1; \
59233		int16x8_t __s2 = __p2; \
59234		__ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __s2, __p3); \
	59201	#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
	59202	poly64x2_t __ret; \
	59203	poly64x2_t __s0 = __p0; \
	59204	poly64x2_t __s1 = __p1; \
	59205	__ret = (poly64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
59235	59206	__ret; \
59236	59207	})
59237	59208	#else
59238		#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59239		int32_t __ret; \
59240		int32_t __s0 = __p0; \
59241		int16_t __s1 = __p1; \
59242		int16x8_t __s2 = __p2; \
59243		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
59244		__ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __rev2, __p3); \
	59209	#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
	59210	poly64x2_t __ret; \
	59211	poly64x2_t __s0 = __p0; \
	59212	poly64x2_t __s1 = __p1; \
	59213	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	59214	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	59215	__ret = (poly64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
	59216	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
59245	59217	__ret; \
59246	59218	})
59247	59219	#endif
59248	59220
59249		#ifdef __LITTLE_ENDIAN__
59250		#define vqdmlsl_laneq_s32(__p0_731, __p1_731, __p2_731, __p3_731) __extension__ ({ \
59251		int64x2_t __ret_731; \
59252		int64x2_t __s0_731 = __p0_731; \
59253		int32x2_t __s1_731 = __p1_731; \
59254		int32x4_t __s2_731 = __p2_731; \
59255		__ret_731 = vqdmlsl_s32(__s0_731, __s1_731, splat_laneq_s32(__s2_731, __p3_731)); \
59256		__ret_731; \
59257		})
59258		#else
59259		#define vqdmlsl_laneq_s32(__p0_732, __p1_732, __p2_732, __p3_732) __extension__ ({ \
59260		int64x2_t __ret_732; \
59261		int64x2_t __s0_732 = __p0_732; \
59262		int32x2_t __s1_732 = __p1_732; \
59263		int32x4_t __s2_732 = __p2_732; \
59264		int64x2_t __rev0_732; __rev0_732 = __builtin_shufflevector(__s0_732, __s0_732, 1, 0); \
59265		int32x2_t __rev1_732; __rev1_732 = __builtin_shufflevector(__s1_732, __s1_732, 1, 0); \
59266		int32x4_t __rev2_732; __rev2_732 = __builtin_shufflevector(__s2_732, __s2_732, 3, 2, 1, 0); \
59267		__ret_732 = __noswap_vqdmlsl_s32(__rev0_732, __rev1_732, __noswap_splat_laneq_s32(__rev2_732, __p3_732)); \
59268		__ret_732 = __builtin_shufflevector(__ret_732, __ret_732, 1, 0); \
59269		__ret_732; \
	59221	#define vst1_p64(__p0, __p1) __extension__ ({ \
	59222	poly64x1_t __s1 = __p1; \
	59223	__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 6); \
59270	59224	})
59271		#endif
59272
59273	59225	#ifdef __LITTLE_ENDIAN__
59274		#define vqdmlsl_laneq_s16(__p0_733, __p1_733, __p2_733, __p3_733) __extension__ ({ \
59275		int32x4_t __ret_733; \
59276		int32x4_t __s0_733 = __p0_733; \
59277		int16x4_t __s1_733 = __p1_733; \
59278		int16x8_t __s2_733 = __p2_733; \
59279		__ret_733 = vqdmlsl_s16(__s0_733, __s1_733, splat_laneq_s16(__s2_733, __p3_733)); \
59280		__ret_733; \
	59226	#define vst1q_p64(__p0, __p1) __extension__ ({ \
	59227	poly64x2_t __s1 = __p1; \
	59228	__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 38); \
59281	59229	})
59282	59230	#else
59283		#define vqdmlsl_laneq_s16(__p0_734, __p1_734, __p2_734, __p3_734) __extension__ ({ \
59284		int32x4_t __ret_734; \
59285		int32x4_t __s0_734 = __p0_734; \
59286		int16x4_t __s1_734 = __p1_734; \
59287		int16x8_t __s2_734 = __p2_734; \
59288		int32x4_t __rev0_734; __rev0_734 = __builtin_shufflevector(__s0_734, __s0_734, 3, 2, 1, 0); \
59289		int16x4_t __rev1_734; __rev1_734 = __builtin_shufflevector(__s1_734, __s1_734, 3, 2, 1, 0); \
59290		int16x8_t __rev2_734; __rev2_734 = __builtin_shufflevector(__s2_734, __s2_734, 7, 6, 5, 4, 3, 2, 1, 0); \
59291		__ret_734 = __noswap_vqdmlsl_s16(__rev0_734, __rev1_734, __noswap_splat_laneq_s16(__rev2_734, __p3_734)); \
59292		__ret_734 = __builtin_shufflevector(__ret_734, __ret_734, 3, 2, 1, 0); \
59293		__ret_734; \
	59231	#define vst1q_p64(__p0, __p1) __extension__ ({ \
	59232	poly64x2_t __s1 = __p1; \
	59233	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	59234	__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 38); \
59294	59235	})
59295	59236	#endif
59296	59237
59297		__ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
59298		int32_t __ret;
59299		__ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
59300		return __ret;
59301		}
59302		__ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
59303		int16_t __ret;
59304		__ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
59305		return __ret;
59306		}
59307	59238	#ifdef __LITTLE_ENDIAN__
59308		#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59309		int32x4_t __ret; \
59310		int32x4_t __s0 = __p0; \
59311		int32x2_t __s1 = __p1; \
59312		__ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
59313		__ret; \
	59239	#define vst1q_f64(__p0, __p1) __extension__ ({ \
	59240	float64x2_t __s1 = __p1; \
	59241	__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 42); \
59314	59242	})
59315	59243	#else
59316		#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59317		int32x4_t __ret; \
59318		int32x4_t __s0 = __p0; \
59319		int32x2_t __s1 = __p1; \
59320		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
59321		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
59322		__ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
59323		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
59324		__ret; \
	59244	#define vst1q_f64(__p0, __p1) __extension__ ({ \
	59245	float64x2_t __s1 = __p1; \
	59246	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	59247	__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 42); \
59325	59248	})
59326	59249	#endif
59327	59250
59328		#ifdef __LITTLE_ENDIAN__
59329		#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59330		int16x8_t __ret; \
59331		int16x8_t __s0 = __p0; \
59332		int16x4_t __s1 = __p1; \
59333		__ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
59334		__ret; \
	59251	#define vst1_f64(__p0, __p1) __extension__ ({ \
	59252	float64x1_t __s1 = __p1; \
	59253	__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 10); \
59335	59254	})
59336		#else
59337		#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59338		int16x8_t __ret; \
59339		int16x8_t __s0 = __p0; \
59340		int16x4_t __s1 = __p1; \
59341		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
59342		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59343		__ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
59344		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
59345		__ret; \
	59255	#define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59256	poly64x1_t __s1 = __p1; \
	59257	__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
59346	59258	})
59347		#endif
59348
59349	59259	#ifdef __LITTLE_ENDIAN__
59350		#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59351		int32x2_t __ret; \
59352		int32x2_t __s0 = __p0; \
59353		int32x2_t __s1 = __p1; \
59354		__ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
59355		__ret; \
	59260	#define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59261	poly64x2_t __s1 = __p1; \
	59262	__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 38); \
59356	59263	})
59357	59264	#else
59358		#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59359		int32x2_t __ret; \
59360		int32x2_t __s0 = __p0; \
59361		int32x2_t __s1 = __p1; \
59362		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
59363		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
59364		__ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
59365		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
59366		__ret; \
	59265	#define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59266	poly64x2_t __s1 = __p1; \
	59267	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	59268	__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 38); \
59367	59269	})
59368	59270	#endif
59369	59271
59370	59272	#ifdef __LITTLE_ENDIAN__
59371		#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59372		int16x4_t __ret; \
59373		int16x4_t __s0 = __p0; \
59374		int16x4_t __s1 = __p1; \
59375		__ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
59376		__ret; \
	59273	#define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59274	float64x2_t __s1 = __p1; \
	59275	__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 42); \
59377	59276	})
59378	59277	#else
59379		#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59380		int16x4_t __ret; \
59381		int16x4_t __s0 = __p0; \
59382		int16x4_t __s1 = __p1; \
59383		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
59384		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59385		__ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
59386		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
59387		__ret; \
	59278	#define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59279	float64x2_t __s1 = __p1; \
	59280	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	59281	__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 42); \
59388	59282	})
59389	59283	#endif
59390	59284
59391		#ifdef __LITTLE_ENDIAN__
59392		#define vqdmulhs_lane_s32(__p0_735, __p1_735, __p2_735) __extension__ ({ \
59393		int32_t __ret_735; \
59394		int32_t __s0_735 = __p0_735; \
59395		int32x2_t __s1_735 = __p1_735; \
59396		__ret_735 = vqdmulhs_s32(__s0_735, vget_lane_s32(__s1_735, __p2_735)); \
59397		__ret_735; \
	59285	#define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59286	float64x1_t __s1 = __p1; \
	59287	__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
59398	59288	})
59399		#else
59400		#define vqdmulhs_lane_s32(__p0_736, __p1_736, __p2_736) __extension__ ({ \
59401		int32_t __ret_736; \
59402		int32_t __s0_736 = __p0_736; \
59403		int32x2_t __s1_736 = __p1_736; \
59404		int32x2_t __rev1_736; __rev1_736 = __builtin_shufflevector(__s1_736, __s1_736, 1, 0); \
59405		__ret_736 = vqdmulhs_s32(__s0_736, __noswap_vget_lane_s32(__rev1_736, __p2_736)); \
59406		__ret_736; \
	59289	#define vst1_p64_x2(__p0, __p1) __extension__ ({ \
	59290	poly64x1x2_t __s1 = __p1; \
	59291	__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
59407	59292	})
59408		#endif
59409
59410	59293	#ifdef __LITTLE_ENDIAN__
59411		#define vqdmulhh_lane_s16(__p0_737, __p1_737, __p2_737) __extension__ ({ \
59412		int16_t __ret_737; \
59413		int16_t __s0_737 = __p0_737; \
59414		int16x4_t __s1_737 = __p1_737; \
59415		__ret_737 = vqdmulhh_s16(__s0_737, vget_lane_s16(__s1_737, __p2_737)); \
59416		__ret_737; \
	59294	#define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
	59295	poly64x2x2_t __s1 = __p1; \
	59296	__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 38); \
59417	59297	})
59418	59298	#else
59419		#define vqdmulhh_lane_s16(__p0_738, __p1_738, __p2_738) __extension__ ({ \
59420		int16_t __ret_738; \
59421		int16_t __s0_738 = __p0_738; \
59422		int16x4_t __s1_738 = __p1_738; \
59423		int16x4_t __rev1_738; __rev1_738 = __builtin_shufflevector(__s1_738, __s1_738, 3, 2, 1, 0); \
59424		__ret_738 = vqdmulhh_s16(__s0_738, __noswap_vget_lane_s16(__rev1_738, __p2_738)); \
59425		__ret_738; \
	59299	#define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
	59300	poly64x2x2_t __s1 = __p1; \
	59301	poly64x2x2_t __rev1; \
	59302	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59303	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59304	__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 38); \
59426	59305	})
59427	59306	#endif
59428	59307
59429	59308	#ifdef __LITTLE_ENDIAN__
59430		#define vqdmulhs_laneq_s32(__p0_739, __p1_739, __p2_739) __extension__ ({ \
59431		int32_t __ret_739; \
59432		int32_t __s0_739 = __p0_739; \
59433		int32x4_t __s1_739 = __p1_739; \
59434		__ret_739 = vqdmulhs_s32(__s0_739, vgetq_lane_s32(__s1_739, __p2_739)); \
59435		__ret_739; \
	59309	#define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
	59310	float64x2x2_t __s1 = __p1; \
	59311	__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 42); \
59436	59312	})
59437	59313	#else
59438		#define vqdmulhs_laneq_s32(__p0_740, __p1_740, __p2_740) __extension__ ({ \
59439		int32_t __ret_740; \
59440		int32_t __s0_740 = __p0_740; \
59441		int32x4_t __s1_740 = __p1_740; \
59442		int32x4_t __rev1_740; __rev1_740 = __builtin_shufflevector(__s1_740, __s1_740, 3, 2, 1, 0); \
59443		__ret_740 = vqdmulhs_s32(__s0_740, __noswap_vgetq_lane_s32(__rev1_740, __p2_740)); \
59444		__ret_740; \
	59314	#define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
	59315	float64x2x2_t __s1 = __p1; \
	59316	float64x2x2_t __rev1; \
	59317	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59318	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59319	__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 42); \
59445	59320	})
59446	59321	#endif
59447	59322
59448		#ifdef __LITTLE_ENDIAN__
59449		#define vqdmulhh_laneq_s16(__p0_741, __p1_741, __p2_741) __extension__ ({ \
59450		int16_t __ret_741; \
59451		int16_t __s0_741 = __p0_741; \
59452		int16x8_t __s1_741 = __p1_741; \
59453		__ret_741 = vqdmulhh_s16(__s0_741, vgetq_lane_s16(__s1_741, __p2_741)); \
59454		__ret_741; \
	59323	#define vst1_f64_x2(__p0, __p1) __extension__ ({ \
	59324	float64x1x2_t __s1 = __p1; \
	59325	__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 10); \
59455	59326	})
59456		#else
59457		#define vqdmulhh_laneq_s16(__p0_742, __p1_742, __p2_742) __extension__ ({ \
59458		int16_t __ret_742; \
59459		int16_t __s0_742 = __p0_742; \
59460		int16x8_t __s1_742 = __p1_742; \
59461		int16x8_t __rev1_742; __rev1_742 = __builtin_shufflevector(__s1_742, __s1_742, 7, 6, 5, 4, 3, 2, 1, 0); \
59462		__ret_742 = vqdmulhh_s16(__s0_742, __noswap_vgetq_lane_s16(__rev1_742, __p2_742)); \
59463		__ret_742; \
	59327	#define vst1_p64_x3(__p0, __p1) __extension__ ({ \
	59328	poly64x1x3_t __s1 = __p1; \
	59329	__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
59464	59330	})
59465		#endif
59466
59467	59331	#ifdef __LITTLE_ENDIAN__
59468		#define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59469		int32x4_t __ret; \
59470		int32x4_t __s0 = __p0; \
59471		int32x4_t __s1 = __p1; \
59472		__ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
59473		__ret; \
	59332	#define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
	59333	poly64x2x3_t __s1 = __p1; \
	59334	__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 38); \
59474	59335	})
59475	59336	#else
59476		#define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59477		int32x4_t __ret; \
59478		int32x4_t __s0 = __p0; \
59479		int32x4_t __s1 = __p1; \
59480		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
59481		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59482		__ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
59483		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
59484		__ret; \
	59337	#define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
	59338	poly64x2x3_t __s1 = __p1; \
	59339	poly64x2x3_t __rev1; \
	59340	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59341	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59342	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59343	__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 38); \
59485	59344	})
59486	59345	#endif
59487	59346
59488	59347	#ifdef __LITTLE_ENDIAN__
59489		#define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59490		int16x8_t __ret; \
59491		int16x8_t __s0 = __p0; \
59492		int16x8_t __s1 = __p1; \
59493		__ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
59494		__ret; \
	59348	#define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
	59349	float64x2x3_t __s1 = __p1; \
	59350	__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 42); \
59495	59351	})
59496	59352	#else
59497		#define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59498		int16x8_t __ret; \
59499		int16x8_t __s0 = __p0; \
59500		int16x8_t __s1 = __p1; \
59501		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
59502		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
59503		__ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
59504		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
59505		__ret; \
	59353	#define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
	59354	float64x2x3_t __s1 = __p1; \
	59355	float64x2x3_t __rev1; \
	59356	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59357	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59358	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59359	__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 42); \
59506	59360	})
59507	59361	#endif
59508	59362
59509		#ifdef __LITTLE_ENDIAN__
59510		#define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59511		int32x2_t __ret; \
59512		int32x2_t __s0 = __p0; \
59513		int32x4_t __s1 = __p1; \
59514		__ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
59515		__ret; \
	59363	#define vst1_f64_x3(__p0, __p1) __extension__ ({ \
	59364	float64x1x3_t __s1 = __p1; \
	59365	__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 10); \
59516	59366	})
59517		#else
59518		#define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59519		int32x2_t __ret; \
59520		int32x2_t __s0 = __p0; \
59521		int32x4_t __s1 = __p1; \
59522		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
59523		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59524		__ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
59525		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
59526		__ret; \
	59367	#define vst1_p64_x4(__p0, __p1) __extension__ ({ \
	59368	poly64x1x4_t __s1 = __p1; \
	59369	__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
59527	59370	})
59528		#endif
59529
59530	59371	#ifdef __LITTLE_ENDIAN__
59531		#define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59532		int16x4_t __ret; \
59533		int16x4_t __s0 = __p0; \
59534		int16x8_t __s1 = __p1; \
59535		__ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
59536		__ret; \
	59372	#define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
	59373	poly64x2x4_t __s1 = __p1; \
	59374	__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 38); \
59537	59375	})
59538	59376	#else
59539		#define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59540		int16x4_t __ret; \
59541		int16x4_t __s0 = __p0; \
59542		int16x8_t __s1 = __p1; \
59543		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
59544		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
59545		__ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
59546		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
59547		__ret; \
	59377	#define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
	59378	poly64x2x4_t __s1 = __p1; \
	59379	poly64x2x4_t __rev1; \
	59380	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59381	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59382	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59383	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59384	__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 38); \
59548	59385	})
59549	59386	#endif
59550	59387
59551		__ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
59552		int64_t __ret;
59553		__ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
59554		return __ret;
59555		}
59556		__ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
59557		int32_t __ret;
59558		__ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
59559		return __ret;
59560		}
59561		#ifdef __LITTLE_ENDIAN__
59562		__ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
59563		int64x2_t __ret;
59564		__ret = vqdmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
59565		return __ret;
59566		}
59567		#else
59568		__ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
59569		int64x2_t __ret;
59570		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59571		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59572		__ret = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
59573		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
59574		return __ret;
59575		}
59576		#endif
59577
59578		#ifdef __LITTLE_ENDIAN__
59579		__ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
59580		int32x4_t __ret;
59581		__ret = vqdmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
59582		return __ret;
59583		}
59584		#else
59585		__ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
59586		int32x4_t __ret;
59587		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
59588		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59589		__ret = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
59590		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59591		return __ret;
59592		}
59593		#endif
59594
59595	59388	#ifdef __LITTLE_ENDIAN__
59596		#define vqdmull_high_lane_s32(__p0_743, __p1_743, __p2_743) __extension__ ({ \
59597		int64x2_t __ret_743; \
59598		int32x4_t __s0_743 = __p0_743; \
59599		int32x2_t __s1_743 = __p1_743; \
59600		__ret_743 = vqdmull_s32(vget_high_s32(__s0_743), splat_lane_s32(__s1_743, __p2_743)); \
59601		__ret_743; \
	59389	#define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
	59390	float64x2x4_t __s1 = __p1; \
	59391	__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 42); \
59602	59392	})
59603	59393	#else
59604		#define vqdmull_high_lane_s32(__p0_744, __p1_744, __p2_744) __extension__ ({ \
59605		int64x2_t __ret_744; \
59606		int32x4_t __s0_744 = __p0_744; \
59607		int32x2_t __s1_744 = __p1_744; \
59608		int32x4_t __rev0_744; __rev0_744 = __builtin_shufflevector(__s0_744, __s0_744, 3, 2, 1, 0); \
59609		int32x2_t __rev1_744; __rev1_744 = __builtin_shufflevector(__s1_744, __s1_744, 1, 0); \
59610		__ret_744 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_744), __noswap_splat_lane_s32(__rev1_744, __p2_744)); \
59611		__ret_744 = __builtin_shufflevector(__ret_744, __ret_744, 1, 0); \
59612		__ret_744; \
	59394	#define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
	59395	float64x2x4_t __s1 = __p1; \
	59396	float64x2x4_t __rev1; \
	59397	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59398	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59399	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59400	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59401	__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 42); \
59613	59402	})
59614	59403	#endif
59615	59404
	59405	#define vst1_f64_x4(__p0, __p1) __extension__ ({ \
	59406	float64x1x4_t __s1 = __p1; \
	59407	__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 10); \
	59408	})
	59409	#define vst2_p64(__p0, __p1) __extension__ ({ \
	59410	poly64x1x2_t __s1 = __p1; \
	59411	__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
	59412	})
59616	59413	#ifdef __LITTLE_ENDIAN__
59617		#define vqdmull_high_lane_s16(__p0_745, __p1_745, __p2_745) __extension__ ({ \
59618		int32x4_t __ret_745; \
59619		int16x8_t __s0_745 = __p0_745; \
59620		int16x4_t __s1_745 = __p1_745; \
59621		__ret_745 = vqdmull_s16(vget_high_s16(__s0_745), splat_lane_s16(__s1_745, __p2_745)); \
59622		__ret_745; \
	59414	#define vst2q_p64(__p0, __p1) __extension__ ({ \
	59415	poly64x2x2_t __s1 = __p1; \
	59416	__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 38); \
59623	59417	})
59624	59418	#else
59625		#define vqdmull_high_lane_s16(__p0_746, __p1_746, __p2_746) __extension__ ({ \
59626		int32x4_t __ret_746; \
59627		int16x8_t __s0_746 = __p0_746; \
59628		int16x4_t __s1_746 = __p1_746; \
59629		int16x8_t __rev0_746; __rev0_746 = __builtin_shufflevector(__s0_746, __s0_746, 7, 6, 5, 4, 3, 2, 1, 0); \
59630		int16x4_t __rev1_746; __rev1_746 = __builtin_shufflevector(__s1_746, __s1_746, 3, 2, 1, 0); \
59631		__ret_746 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_746), __noswap_splat_lane_s16(__rev1_746, __p2_746)); \
59632		__ret_746 = __builtin_shufflevector(__ret_746, __ret_746, 3, 2, 1, 0); \
59633		__ret_746; \
	59419	#define vst2q_p64(__p0, __p1) __extension__ ({ \
	59420	poly64x2x2_t __s1 = __p1; \
	59421	poly64x2x2_t __rev1; \
	59422	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59423	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59424	__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 38); \
59634	59425	})
59635	59426	#endif
59636	59427
59637	59428	#ifdef __LITTLE_ENDIAN__
59638		#define vqdmull_high_laneq_s32(__p0_747, __p1_747, __p2_747) __extension__ ({ \
59639		int64x2_t __ret_747; \
59640		int32x4_t __s0_747 = __p0_747; \
59641		int32x4_t __s1_747 = __p1_747; \
59642		__ret_747 = vqdmull_s32(vget_high_s32(__s0_747), splat_laneq_s32(__s1_747, __p2_747)); \
59643		__ret_747; \
	59429	#define vst2q_u64(__p0, __p1) __extension__ ({ \
	59430	uint64x2x2_t __s1 = __p1; \
	59431	__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 51); \
59644	59432	})
59645	59433	#else
59646		#define vqdmull_high_laneq_s32(__p0_748, __p1_748, __p2_748) __extension__ ({ \
59647		int64x2_t __ret_748; \
59648		int32x4_t __s0_748 = __p0_748; \
59649		int32x4_t __s1_748 = __p1_748; \
59650		int32x4_t __rev0_748; __rev0_748 = __builtin_shufflevector(__s0_748, __s0_748, 3, 2, 1, 0); \
59651		int32x4_t __rev1_748; __rev1_748 = __builtin_shufflevector(__s1_748, __s1_748, 3, 2, 1, 0); \
59652		__ret_748 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_748), __noswap_splat_laneq_s32(__rev1_748, __p2_748)); \
59653		__ret_748 = __builtin_shufflevector(__ret_748, __ret_748, 1, 0); \
59654		__ret_748; \
	59434	#define vst2q_u64(__p0, __p1) __extension__ ({ \
	59435	uint64x2x2_t __s1 = __p1; \
	59436	uint64x2x2_t __rev1; \
	59437	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59438	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59439	__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 51); \
59655	59440	})
59656	59441	#endif
59657	59442
59658	59443	#ifdef __LITTLE_ENDIAN__
59659		#define vqdmull_high_laneq_s16(__p0_749, __p1_749, __p2_749) __extension__ ({ \
59660		int32x4_t __ret_749; \
59661		int16x8_t __s0_749 = __p0_749; \
59662		int16x8_t __s1_749 = __p1_749; \
59663		__ret_749 = vqdmull_s16(vget_high_s16(__s0_749), splat_laneq_s16(__s1_749, __p2_749)); \
59664		__ret_749; \
	59444	#define vst2q_f64(__p0, __p1) __extension__ ({ \
	59445	float64x2x2_t __s1 = __p1; \
	59446	__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 42); \
59665	59447	})
59666	59448	#else
59667		#define vqdmull_high_laneq_s16(__p0_750, __p1_750, __p2_750) __extension__ ({ \
59668		int32x4_t __ret_750; \
59669		int16x8_t __s0_750 = __p0_750; \
59670		int16x8_t __s1_750 = __p1_750; \
59671		int16x8_t __rev0_750; __rev0_750 = __builtin_shufflevector(__s0_750, __s0_750, 7, 6, 5, 4, 3, 2, 1, 0); \
59672		int16x8_t __rev1_750; __rev1_750 = __builtin_shufflevector(__s1_750, __s1_750, 7, 6, 5, 4, 3, 2, 1, 0); \
59673		__ret_750 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_750), __noswap_splat_laneq_s16(__rev1_750, __p2_750)); \
59674		__ret_750 = __builtin_shufflevector(__ret_750, __ret_750, 3, 2, 1, 0); \
59675		__ret_750; \
	59449	#define vst2q_f64(__p0, __p1) __extension__ ({ \
	59450	float64x2x2_t __s1 = __p1; \
	59451	float64x2x2_t __rev1; \
	59452	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59453	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59454	__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 42); \
59676	59455	})
59677	59456	#endif
59678	59457
59679	59458	#ifdef __LITTLE_ENDIAN__
59680		__ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
59681		int64x2_t __ret;
59682		__ret = vqdmull_n_s32(vget_high_s32(__p0), __p1);
59683		return __ret;
59684		}
	59459	#define vst2q_s64(__p0, __p1) __extension__ ({ \
	59460	int64x2x2_t __s1 = __p1; \
	59461	__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 35); \
	59462	})
59685	59463	#else
59686		__ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
59687		int64x2_t __ret;
59688		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59689		__ret = __noswap_vqdmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
59690		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
59691		return __ret;
59692		}
	59464	#define vst2q_s64(__p0, __p1) __extension__ ({ \
	59465	int64x2x2_t __s1 = __p1; \
	59466	int64x2x2_t __rev1; \
	59467	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59468	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59469	__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 35); \
	59470	})
59693	59471	#endif
59694	59472
	59473	#define vst2_f64(__p0, __p1) __extension__ ({ \
	59474	float64x1x2_t __s1 = __p1; \
	59475	__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 10); \
	59476	})
	59477	#define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59478	poly64x1x2_t __s1 = __p1; \
	59479	__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
	59480	})
59695	59481	#ifdef __LITTLE_ENDIAN__
59696		__ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
59697		int32x4_t __ret;
59698		__ret = vqdmull_n_s16(vget_high_s16(__p0), __p1);
59699		return __ret;
59700		}
	59482	#define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	59483	poly8x16x2_t __s1 = __p1; \
	59484	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 36); \
	59485	})
59701	59486	#else
59702		__ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
59703		int32x4_t __ret;
59704		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
59705		__ret = __noswap_vqdmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
59706		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59707		return __ret;
59708		}
	59487	#define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	59488	poly8x16x2_t __s1 = __p1; \
	59489	poly8x16x2_t __rev1; \
	59490	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59491	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59492	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 36); \
	59493	})
59709	59494	#endif
59710	59495
59711	59496	#ifdef __LITTLE_ENDIAN__
59712		#define vqdmulls_lane_s32(__p0_751, __p1_751, __p2_751) __extension__ ({ \
59713		int64_t __ret_751; \
59714		int32_t __s0_751 = __p0_751; \
59715		int32x2_t __s1_751 = __p1_751; \
59716		__ret_751 = vqdmulls_s32(__s0_751, vget_lane_s32(__s1_751, __p2_751)); \
59717		__ret_751; \
	59497	#define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59498	poly64x2x2_t __s1 = __p1; \
	59499	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 38); \
59718	59500	})
59719	59501	#else
59720		#define vqdmulls_lane_s32(__p0_752, __p1_752, __p2_752) __extension__ ({ \
59721		int64_t __ret_752; \
59722		int32_t __s0_752 = __p0_752; \
59723		int32x2_t __s1_752 = __p1_752; \
59724		int32x2_t __rev1_752; __rev1_752 = __builtin_shufflevector(__s1_752, __s1_752, 1, 0); \
59725		__ret_752 = vqdmulls_s32(__s0_752, __noswap_vget_lane_s32(__rev1_752, __p2_752)); \
59726		__ret_752; \
	59502	#define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59503	poly64x2x2_t __s1 = __p1; \
	59504	poly64x2x2_t __rev1; \
	59505	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59506	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59507	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 38); \
59727	59508	})
59728	59509	#endif
59729	59510
59730	59511	#ifdef __LITTLE_ENDIAN__
59731		#define vqdmullh_lane_s16(__p0_753, __p1_753, __p2_753) __extension__ ({ \
59732		int32_t __ret_753; \
59733		int16_t __s0_753 = __p0_753; \
59734		int16x4_t __s1_753 = __p1_753; \
59735		__ret_753 = vqdmullh_s16(__s0_753, vget_lane_s16(__s1_753, __p2_753)); \
59736		__ret_753; \
	59512	#define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	59513	uint8x16x2_t __s1 = __p1; \
	59514	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 48); \
59737	59515	})
59738	59516	#else
59739		#define vqdmullh_lane_s16(__p0_754, __p1_754, __p2_754) __extension__ ({ \
59740		int32_t __ret_754; \
59741		int16_t __s0_754 = __p0_754; \
59742		int16x4_t __s1_754 = __p1_754; \
59743		int16x4_t __rev1_754; __rev1_754 = __builtin_shufflevector(__s1_754, __s1_754, 3, 2, 1, 0); \
59744		__ret_754 = vqdmullh_s16(__s0_754, __noswap_vget_lane_s16(__rev1_754, __p2_754)); \
59745		__ret_754; \
	59517	#define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	59518	uint8x16x2_t __s1 = __p1; \
	59519	uint8x16x2_t __rev1; \
	59520	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59521	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59522	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 48); \
59746	59523	})
59747	59524	#endif
59748	59525
59749	59526	#ifdef __LITTLE_ENDIAN__
59750		#define vqdmulls_laneq_s32(__p0_755, __p1_755, __p2_755) __extension__ ({ \
59751		int64_t __ret_755; \
59752		int32_t __s0_755 = __p0_755; \
59753		int32x4_t __s1_755 = __p1_755; \
59754		__ret_755 = vqdmulls_s32(__s0_755, vgetq_lane_s32(__s1_755, __p2_755)); \
59755		__ret_755; \
	59527	#define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59528	uint64x2x2_t __s1 = __p1; \
	59529	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 51); \
59756	59530	})
59757	59531	#else
59758		#define vqdmulls_laneq_s32(__p0_756, __p1_756, __p2_756) __extension__ ({ \
59759		int64_t __ret_756; \
59760		int32_t __s0_756 = __p0_756; \
59761		int32x4_t __s1_756 = __p1_756; \
59762		int32x4_t __rev1_756; __rev1_756 = __builtin_shufflevector(__s1_756, __s1_756, 3, 2, 1, 0); \
59763		__ret_756 = vqdmulls_s32(__s0_756, __noswap_vgetq_lane_s32(__rev1_756, __p2_756)); \
59764		__ret_756; \
	59532	#define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59533	uint64x2x2_t __s1 = __p1; \
	59534	uint64x2x2_t __rev1; \
	59535	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59536	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59537	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 51); \
59765	59538	})
59766	59539	#endif
59767	59540
59768	59541	#ifdef __LITTLE_ENDIAN__
59769		#define vqdmullh_laneq_s16(__p0_757, __p1_757, __p2_757) __extension__ ({ \
59770		int32_t __ret_757; \
59771		int16_t __s0_757 = __p0_757; \
59772		int16x8_t __s1_757 = __p1_757; \
59773		__ret_757 = vqdmullh_s16(__s0_757, vgetq_lane_s16(__s1_757, __p2_757)); \
59774		__ret_757; \
	59542	#define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	59543	int8x16x2_t __s1 = __p1; \
	59544	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 32); \
59775	59545	})
59776	59546	#else
59777		#define vqdmullh_laneq_s16(__p0_758, __p1_758, __p2_758) __extension__ ({ \
59778		int32_t __ret_758; \
59779		int16_t __s0_758 = __p0_758; \
59780		int16x8_t __s1_758 = __p1_758; \
59781		int16x8_t __rev1_758; __rev1_758 = __builtin_shufflevector(__s1_758, __s1_758, 7, 6, 5, 4, 3, 2, 1, 0); \
59782		__ret_758 = vqdmullh_s16(__s0_758, __noswap_vgetq_lane_s16(__rev1_758, __p2_758)); \
59783		__ret_758; \
	59547	#define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	59548	int8x16x2_t __s1 = __p1; \
	59549	int8x16x2_t __rev1; \
	59550	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59551	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59552	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 32); \
59784	59553	})
59785	59554	#endif
59786	59555
59787	59556	#ifdef __LITTLE_ENDIAN__
59788		#define vqdmull_laneq_s32(__p0_759, __p1_759, __p2_759) __extension__ ({ \
59789		int64x2_t __ret_759; \
59790		int32x2_t __s0_759 = __p0_759; \
59791		int32x4_t __s1_759 = __p1_759; \
59792		__ret_759 = vqdmull_s32(__s0_759, splat_laneq_s32(__s1_759, __p2_759)); \
59793		__ret_759; \
	59557	#define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59558	float64x2x2_t __s1 = __p1; \
	59559	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
59794	59560	})
59795	59561	#else
59796		#define vqdmull_laneq_s32(__p0_760, __p1_760, __p2_760) __extension__ ({ \
59797		int64x2_t __ret_760; \
59798		int32x2_t __s0_760 = __p0_760; \
59799		int32x4_t __s1_760 = __p1_760; \
59800		int32x2_t __rev0_760; __rev0_760 = __builtin_shufflevector(__s0_760, __s0_760, 1, 0); \
59801		int32x4_t __rev1_760; __rev1_760 = __builtin_shufflevector(__s1_760, __s1_760, 3, 2, 1, 0); \
59802		__ret_760 = __noswap_vqdmull_s32(__rev0_760, __noswap_splat_laneq_s32(__rev1_760, __p2_760)); \
59803		__ret_760 = __builtin_shufflevector(__ret_760, __ret_760, 1, 0); \
59804		__ret_760; \
	59562	#define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59563	float64x2x2_t __s1 = __p1; \
	59564	float64x2x2_t __rev1; \
	59565	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59566	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59567	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
59805	59568	})
59806	59569	#endif
59807	59570
59808	59571	#ifdef __LITTLE_ENDIAN__
59809		#define vqdmull_laneq_s16(__p0_761, __p1_761, __p2_761) __extension__ ({ \
59810		int32x4_t __ret_761; \
59811		int16x4_t __s0_761 = __p0_761; \
59812		int16x8_t __s1_761 = __p1_761; \
59813		__ret_761 = vqdmull_s16(__s0_761, splat_laneq_s16(__s1_761, __p2_761)); \
59814		__ret_761; \
	59572	#define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59573	int64x2x2_t __s1 = __p1; \
	59574	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
59815	59575	})
59816	59576	#else
59817		#define vqdmull_laneq_s16(__p0_762, __p1_762, __p2_762) __extension__ ({ \
59818		int32x4_t __ret_762; \
59819		int16x4_t __s0_762 = __p0_762; \
59820		int16x8_t __s1_762 = __p1_762; \
59821		int16x4_t __rev0_762; __rev0_762 = __builtin_shufflevector(__s0_762, __s0_762, 3, 2, 1, 0); \
59822		int16x8_t __rev1_762; __rev1_762 = __builtin_shufflevector(__s1_762, __s1_762, 7, 6, 5, 4, 3, 2, 1, 0); \
59823		__ret_762 = __noswap_vqdmull_s16(__rev0_762, __noswap_splat_laneq_s16(__rev1_762, __p2_762)); \
59824		__ret_762 = __builtin_shufflevector(__ret_762, __ret_762, 3, 2, 1, 0); \
59825		__ret_762; \
	59577	#define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59578	int64x2x2_t __s1 = __p1; \
	59579	int64x2x2_t __rev1; \
	59580	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59581	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59582	__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
59826	59583	})
59827	59584	#endif
59828	59585
59829		__ai int16_t vqmovns_s32(int32_t __p0) {
59830		int16_t __ret;
59831		__ret = (int16_t) __builtin_neon_vqmovns_s32(__p0);
59832		return __ret;
59833		}
59834		__ai int32_t vqmovnd_s64(int64_t __p0) {
59835		int32_t __ret;
59836		__ret = (int32_t) __builtin_neon_vqmovnd_s64(__p0);
59837		return __ret;
59838		}
59839		__ai int8_t vqmovnh_s16(int16_t __p0) {
59840		int8_t __ret;
59841		__ret = (int8_t) __builtin_neon_vqmovnh_s16(__p0);
59842		return __ret;
59843		}
59844		__ai uint16_t vqmovns_u32(uint32_t __p0) {
59845		uint16_t __ret;
59846		__ret = (uint16_t) __builtin_neon_vqmovns_u32(__p0);
59847		return __ret;
59848		}
59849		__ai uint32_t vqmovnd_u64(uint64_t __p0) {
59850		uint32_t __ret;
59851		__ret = (uint32_t) __builtin_neon_vqmovnd_u64(__p0);
59852		return __ret;
59853		}
59854		__ai uint8_t vqmovnh_u16(uint16_t __p0) {
59855		uint8_t __ret;
59856		__ret = (uint8_t) __builtin_neon_vqmovnh_u16(__p0);
59857		return __ret;
59858		}
	59586	#define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59587	uint64x1x2_t __s1 = __p1; \
	59588	__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
	59589	})
	59590	#define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59591	float64x1x2_t __s1 = __p1; \
	59592	__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
	59593	})
	59594	#define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59595	int64x1x2_t __s1 = __p1; \
	59596	__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
	59597	})
	59598	#define vst3_p64(__p0, __p1) __extension__ ({ \
	59599	poly64x1x3_t __s1 = __p1; \
	59600	__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
	59601	})
59859	59602	#ifdef __LITTLE_ENDIAN__
59860		__ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
59861		uint16x8_t __ret;
59862		__ret = vcombine_u16(__p0, vqmovn_u32(__p1));
59863		return __ret;
59864		}
	59603	#define vst3q_p64(__p0, __p1) __extension__ ({ \
	59604	poly64x2x3_t __s1 = __p1; \
	59605	__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 38); \
	59606	})
59865	59607	#else
59866		__ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
59867		uint16x8_t __ret;
59868		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59869		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59870		__ret = __noswap_vcombine_u16(__rev0, __noswap_vqmovn_u32(__rev1));
59871		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
59872		return __ret;
59873		}
	59608	#define vst3q_p64(__p0, __p1) __extension__ ({ \
	59609	poly64x2x3_t __s1 = __p1; \
	59610	poly64x2x3_t __rev1; \
	59611	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59612	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59613	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59614	__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 38); \
	59615	})
59874	59616	#endif
59875	59617
59876	59618	#ifdef __LITTLE_ENDIAN__
59877		__ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
59878		uint32x4_t __ret;
59879		__ret = vcombine_u32(__p0, vqmovn_u64(__p1));
59880		return __ret;
59881		}
	59619	#define vst3q_u64(__p0, __p1) __extension__ ({ \
	59620	uint64x2x3_t __s1 = __p1; \
	59621	__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 51); \
	59622	})
59882	59623	#else
59883		__ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
59884		uint32x4_t __ret;
59885		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59886		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59887		__ret = __noswap_vcombine_u32(__rev0, __noswap_vqmovn_u64(__rev1));
59888		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59889		return __ret;
59890		}
	59624	#define vst3q_u64(__p0, __p1) __extension__ ({ \
	59625	uint64x2x3_t __s1 = __p1; \
	59626	uint64x2x3_t __rev1; \
	59627	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59628	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59629	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59630	__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 51); \
	59631	})
59891	59632	#endif
59892	59633
59893	59634	#ifdef __LITTLE_ENDIAN__
59894		__ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
59895		uint8x16_t __ret;
59896		__ret = vcombine_u8(__p0, vqmovn_u16(__p1));
59897		return __ret;
59898		}
	59635	#define vst3q_f64(__p0, __p1) __extension__ ({ \
	59636	float64x2x3_t __s1 = __p1; \
	59637	__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 42); \
	59638	})
59899	59639	#else
59900		__ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
59901		uint8x16_t __ret;
59902		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
59903		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59904		__ret = __noswap_vcombine_u8(__rev0, __noswap_vqmovn_u16(__rev1));
59905		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59906		return __ret;
59907		}
	59640	#define vst3q_f64(__p0, __p1) __extension__ ({ \
	59641	float64x2x3_t __s1 = __p1; \
	59642	float64x2x3_t __rev1; \
	59643	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59644	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59645	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59646	__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 42); \
	59647	})
59908	59648	#endif
59909	59649
59910	59650	#ifdef __LITTLE_ENDIAN__
59911		__ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
59912		int16x8_t __ret;
59913		__ret = vcombine_s16(__p0, vqmovn_s32(__p1));
59914		return __ret;
59915		}
	59651	#define vst3q_s64(__p0, __p1) __extension__ ({ \
	59652	int64x2x3_t __s1 = __p1; \
	59653	__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 35); \
	59654	})
59916	59655	#else
59917		__ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
59918		int16x8_t __ret;
59919		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59920		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59921		__ret = __noswap_vcombine_s16(__rev0, __noswap_vqmovn_s32(__rev1));
59922		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
59923		return __ret;
59924		}
	59656	#define vst3q_s64(__p0, __p1) __extension__ ({ \
	59657	int64x2x3_t __s1 = __p1; \
	59658	int64x2x3_t __rev1; \
	59659	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59660	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59661	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59662	__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 35); \
	59663	})
59925	59664	#endif
59926	59665
	59666	#define vst3_f64(__p0, __p1) __extension__ ({ \
	59667	float64x1x3_t __s1 = __p1; \
	59668	__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 10); \
	59669	})
	59670	#define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59671	poly64x1x3_t __s1 = __p1; \
	59672	__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
	59673	})
59927	59674	#ifdef __LITTLE_ENDIAN__
59928		__ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
59929		int32x4_t __ret;
59930		__ret = vcombine_s32(__p0, vqmovn_s64(__p1));
59931		return __ret;
59932		}
	59675	#define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	59676	poly8x16x3_t __s1 = __p1; \
	59677	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 36); \
	59678	})
59933	59679	#else
59934		__ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
59935		int32x4_t __ret;
59936		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59937		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59938		__ret = __noswap_vcombine_s32(__rev0, __noswap_vqmovn_s64(__rev1));
59939		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59940		return __ret;
59941		}
	59680	#define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	59681	poly8x16x3_t __s1 = __p1; \
	59682	poly8x16x3_t __rev1; \
	59683	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59684	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59685	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59686	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 36); \
	59687	})
59942	59688	#endif
59943	59689
59944	59690	#ifdef __LITTLE_ENDIAN__
59945		__ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
59946		int8x16_t __ret;
59947		__ret = vcombine_s8(__p0, vqmovn_s16(__p1));
59948		return __ret;
59949		}
	59691	#define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59692	poly64x2x3_t __s1 = __p1; \
	59693	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 38); \
	59694	})
59950	59695	#else
59951		__ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
59952		int8x16_t __ret;
59953		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
59954		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59955		__ret = __noswap_vcombine_s8(__rev0, __noswap_vqmovn_s16(__rev1));
59956		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59957		return __ret;
59958		}
	59696	#define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59697	poly64x2x3_t __s1 = __p1; \
	59698	poly64x2x3_t __rev1; \
	59699	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59700	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59701	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59702	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 38); \
	59703	})
59959	59704	#endif
59960	59705
59961		__ai uint16_t vqmovuns_s32(int32_t __p0) {
59962		uint16_t __ret;
59963		__ret = (uint16_t) __builtin_neon_vqmovuns_s32(__p0);
59964		return __ret;
59965		}
59966		__ai uint32_t vqmovund_s64(int64_t __p0) {
59967		uint32_t __ret;
59968		__ret = (uint32_t) __builtin_neon_vqmovund_s64(__p0);
59969		return __ret;
59970		}
59971		__ai uint8_t vqmovunh_s16(int16_t __p0) {
59972		uint8_t __ret;
59973		__ret = (uint8_t) __builtin_neon_vqmovunh_s16(__p0);
59974		return __ret;
59975		}
59976	59706	#ifdef __LITTLE_ENDIAN__
59977		__ai uint16x8_t vqmovun_high_s32(uint16x4_t __p0, int32x4_t __p1) {
59978		uint16x8_t __ret;
59979		__ret = vcombine_u16((uint16x4_t)(__p0), vqmovun_s32(__p1));
59980		return __ret;
59981		}
	59707	#define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	59708	uint8x16x3_t __s1 = __p1; \
	59709	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 48); \
	59710	})
59982	59711	#else
59983		__ai uint16x8_t vqmovun_high_s32(uint16x4_t __p0, int32x4_t __p1) {
59984		uint16x8_t __ret;
59985		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59986		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59987		__ret = __noswap_vcombine_u16((uint16x4_t)(__rev0), __noswap_vqmovun_s32(__rev1));
59988		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
59989		return __ret;
59990		}
	59712	#define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	59713	uint8x16x3_t __s1 = __p1; \
	59714	uint8x16x3_t __rev1; \
	59715	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59716	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59717	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59718	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 48); \
	59719	})
59991	59720	#endif
59992	59721
59993	59722	#ifdef __LITTLE_ENDIAN__
59994		__ai uint32x4_t vqmovun_high_s64(uint32x2_t __p0, int64x2_t __p1) {
59995		uint32x4_t __ret;
59996		__ret = vcombine_u32((uint32x2_t)(__p0), vqmovun_s64(__p1));
59997		return __ret;
59998		}
	59723	#define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59724	uint64x2x3_t __s1 = __p1; \
	59725	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 51); \
	59726	})
59999	59727	#else
60000		__ai uint32x4_t vqmovun_high_s64(uint32x2_t __p0, int64x2_t __p1) {
60001		uint32x4_t __ret;
60002		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
60003		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60004		__ret = __noswap_vcombine_u32((uint32x2_t)(__rev0), __noswap_vqmovun_s64(__rev1));
60005		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
60006		return __ret;
60007		}
	59728	#define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59729	uint64x2x3_t __s1 = __p1; \
	59730	uint64x2x3_t __rev1; \
	59731	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59732	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59733	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59734	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 51); \
	59735	})
60008	59736	#endif
60009	59737
60010	59738	#ifdef __LITTLE_ENDIAN__
60011		__ai uint8x16_t vqmovun_high_s16(uint8x8_t __p0, int16x8_t __p1) {
60012		uint8x16_t __ret;
60013		__ret = vcombine_u8((uint8x8_t)(__p0), vqmovun_s16(__p1));
60014		return __ret;
60015		}
	59739	#define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	59740	int8x16x3_t __s1 = __p1; \
	59741	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 32); \
	59742	})
60016	59743	#else
60017		__ai uint8x16_t vqmovun_high_s16(uint8x8_t __p0, int16x8_t __p1) {
60018		uint8x16_t __ret;
60019		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
60020		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60021		__ret = __noswap_vcombine_u8((uint8x8_t)(__rev0), __noswap_vqmovun_s16(__rev1));
60022		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
60023		return __ret;
60024		}
	59744	#define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	59745	int8x16x3_t __s1 = __p1; \
	59746	int8x16x3_t __rev1; \
	59747	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59748	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59749	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59750	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 32); \
	59751	})
60025	59752	#endif
60026	59753
60027	59754	#ifdef __LITTLE_ENDIAN__
60028		__ai int64x2_t vqnegq_s64(int64x2_t __p0) {
60029		int64x2_t __ret;
60030		__ret = (int64x2_t) __builtin_neon_vqnegq_v((int8x16_t)__p0, 35);
60031		return __ret;
60032		}
	59755	#define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59756	float64x2x3_t __s1 = __p1; \
	59757	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
	59758	})
60033	59759	#else
60034		__ai int64x2_t vqnegq_s64(int64x2_t __p0) {
60035		int64x2_t __ret;
60036		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
60037		__ret = (int64x2_t) __builtin_neon_vqnegq_v((int8x16_t)__rev0, 35);
60038		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
60039		return __ret;
60040		}
60041		#endif
60042
60043		__ai int64x1_t vqneg_s64(int64x1_t __p0) {
60044		int64x1_t __ret;
60045		__ret = (int64x1_t) __builtin_neon_vqneg_v((int8x8_t)__p0, 3);
60046		return __ret;
60047		}
60048		__ai int8_t vqnegb_s8(int8_t __p0) {
60049		int8_t __ret;
60050		__ret = (int8_t) __builtin_neon_vqnegb_s8(__p0);
60051		return __ret;
60052		}
60053		__ai int32_t vqnegs_s32(int32_t __p0) {
60054		int32_t __ret;
60055		__ret = (int32_t) __builtin_neon_vqnegs_s32(__p0);
60056		return __ret;
60057		}
60058		__ai int64_t vqnegd_s64(int64_t __p0) {
60059		int64_t __ret;
60060		__ret = (int64_t) __builtin_neon_vqnegd_s64(__p0);
60061		return __ret;
60062		}
60063		__ai int16_t vqnegh_s16(int16_t __p0) {
60064		int16_t __ret;
60065		__ret = (int16_t) __builtin_neon_vqnegh_s16(__p0);
60066		return __ret;
60067		}
60068		__ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
60069		int32_t __ret;
60070		__ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
60071		return __ret;
60072		}
60073		__ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
60074		int16_t __ret;
60075		__ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
60076		return __ret;
60077		}
	59760	#define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59761	float64x2x3_t __s1 = __p1; \
	59762	float64x2x3_t __rev1; \
	59763	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59764	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59765	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59766	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
	59767	})
	59768	#endif
	59769
60078	59770	#ifdef __LITTLE_ENDIAN__
60079		#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60080		int32x4_t __ret; \
60081		int32x4_t __s0 = __p0; \
60082		int32x2_t __s1 = __p1; \
60083		__ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
60084		__ret; \
	59771	#define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59772	int64x2x3_t __s1 = __p1; \
	59773	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
60085	59774	})
60086	59775	#else
60087		#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60088		int32x4_t __ret; \
60089		int32x4_t __s0 = __p0; \
60090		int32x2_t __s1 = __p1; \
60091		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
60092		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
60093		__ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
60094		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
60095		__ret; \
	59776	#define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59777	int64x2x3_t __s1 = __p1; \
	59778	int64x2x3_t __rev1; \
	59779	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59780	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59781	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59782	__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
60096	59783	})
60097	59784	#endif
60098	59785
	59786	#define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59787	uint64x1x3_t __s1 = __p1; \
	59788	__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
	59789	})
	59790	#define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59791	float64x1x3_t __s1 = __p1; \
	59792	__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
	59793	})
	59794	#define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59795	int64x1x3_t __s1 = __p1; \
	59796	__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
	59797	})
	59798	#define vst4_p64(__p0, __p1) __extension__ ({ \
	59799	poly64x1x4_t __s1 = __p1; \
	59800	__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
	59801	})
60099	59802	#ifdef __LITTLE_ENDIAN__
60100		#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60101		int16x8_t __ret; \
60102		int16x8_t __s0 = __p0; \
60103		int16x4_t __s1 = __p1; \
60104		__ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
60105		__ret; \
	59803	#define vst4q_p64(__p0, __p1) __extension__ ({ \
	59804	poly64x2x4_t __s1 = __p1; \
	59805	__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 38); \
60106	59806	})
60107	59807	#else
60108		#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60109		int16x8_t __ret; \
60110		int16x8_t __s0 = __p0; \
60111		int16x4_t __s1 = __p1; \
60112		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
60113		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60114		__ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
60115		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
60116		__ret; \
	59808	#define vst4q_p64(__p0, __p1) __extension__ ({ \
	59809	poly64x2x4_t __s1 = __p1; \
	59810	poly64x2x4_t __rev1; \
	59811	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59812	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59813	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59814	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59815	__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 38); \
60117	59816	})
60118	59817	#endif
60119	59818
60120	59819	#ifdef __LITTLE_ENDIAN__
60121		#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60122		int32x2_t __ret; \
60123		int32x2_t __s0 = __p0; \
60124		int32x2_t __s1 = __p1; \
60125		__ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
60126		__ret; \
	59820	#define vst4q_u64(__p0, __p1) __extension__ ({ \
	59821	uint64x2x4_t __s1 = __p1; \
	59822	__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 51); \
60127	59823	})
60128	59824	#else
60129		#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60130		int32x2_t __ret; \
60131		int32x2_t __s0 = __p0; \
60132		int32x2_t __s1 = __p1; \
60133		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
60134		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
60135		__ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
60136		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
60137		__ret; \
	59825	#define vst4q_u64(__p0, __p1) __extension__ ({ \
	59826	uint64x2x4_t __s1 = __p1; \
	59827	uint64x2x4_t __rev1; \
	59828	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59829	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59830	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59831	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59832	__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 51); \
60138	59833	})
60139	59834	#endif
60140	59835
60141	59836	#ifdef __LITTLE_ENDIAN__
60142		#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60143		int16x4_t __ret; \
60144		int16x4_t __s0 = __p0; \
60145		int16x4_t __s1 = __p1; \
60146		__ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
60147		__ret; \
	59837	#define vst4q_f64(__p0, __p1) __extension__ ({ \
	59838	float64x2x4_t __s1 = __p1; \
	59839	__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 42); \
60148	59840	})
60149	59841	#else
60150		#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60151		int16x4_t __ret; \
60152		int16x4_t __s0 = __p0; \
60153		int16x4_t __s1 = __p1; \
60154		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
60155		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60156		__ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
60157		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
60158		__ret; \
	59842	#define vst4q_f64(__p0, __p1) __extension__ ({ \
	59843	float64x2x4_t __s1 = __p1; \
	59844	float64x2x4_t __rev1; \
	59845	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59846	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59847	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59848	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59849	__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 42); \
60159	59850	})
60160	59851	#endif
60161	59852
60162	59853	#ifdef __LITTLE_ENDIAN__
60163		#define vqrdmulhs_lane_s32(__p0_763, __p1_763, __p2_763) __extension__ ({ \
60164		int32_t __ret_763; \
60165		int32_t __s0_763 = __p0_763; \
60166		int32x2_t __s1_763 = __p1_763; \
60167		__ret_763 = vqrdmulhs_s32(__s0_763, vget_lane_s32(__s1_763, __p2_763)); \
60168		__ret_763; \
	59854	#define vst4q_s64(__p0, __p1) __extension__ ({ \
	59855	int64x2x4_t __s1 = __p1; \
	59856	__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 35); \
60169	59857	})
60170	59858	#else
60171		#define vqrdmulhs_lane_s32(__p0_764, __p1_764, __p2_764) __extension__ ({ \
60172		int32_t __ret_764; \
60173		int32_t __s0_764 = __p0_764; \
60174		int32x2_t __s1_764 = __p1_764; \
60175		int32x2_t __rev1_764; __rev1_764 = __builtin_shufflevector(__s1_764, __s1_764, 1, 0); \
60176		__ret_764 = vqrdmulhs_s32(__s0_764, __noswap_vget_lane_s32(__rev1_764, __p2_764)); \
60177		__ret_764; \
	59859	#define vst4q_s64(__p0, __p1) __extension__ ({ \
	59860	int64x2x4_t __s1 = __p1; \
	59861	int64x2x4_t __rev1; \
	59862	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59863	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59864	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59865	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59866	__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 35); \
60178	59867	})
60179	59868	#endif
60180	59869
	59870	#define vst4_f64(__p0, __p1) __extension__ ({ \
	59871	float64x1x4_t __s1 = __p1; \
	59872	__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 10); \
	59873	})
	59874	#define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59875	poly64x1x4_t __s1 = __p1; \
	59876	__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
	59877	})
60181	59878	#ifdef __LITTLE_ENDIAN__
60182		#define vqrdmulhh_lane_s16(__p0_765, __p1_765, __p2_765) __extension__ ({ \
60183		int16_t __ret_765; \
60184		int16_t __s0_765 = __p0_765; \
60185		int16x4_t __s1_765 = __p1_765; \
60186		__ret_765 = vqrdmulhh_s16(__s0_765, vget_lane_s16(__s1_765, __p2_765)); \
60187		__ret_765; \
	59879	#define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	59880	poly8x16x4_t __s1 = __p1; \
	59881	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 36); \
60188	59882	})
60189	59883	#else
60190		#define vqrdmulhh_lane_s16(__p0_766, __p1_766, __p2_766) __extension__ ({ \
60191		int16_t __ret_766; \
60192		int16_t __s0_766 = __p0_766; \
60193		int16x4_t __s1_766 = __p1_766; \
60194		int16x4_t __rev1_766; __rev1_766 = __builtin_shufflevector(__s1_766, __s1_766, 3, 2, 1, 0); \
60195		__ret_766 = vqrdmulhh_s16(__s0_766, __noswap_vget_lane_s16(__rev1_766, __p2_766)); \
60196		__ret_766; \
	59884	#define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
	59885	poly8x16x4_t __s1 = __p1; \
	59886	poly8x16x4_t __rev1; \
	59887	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59888	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59889	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59890	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59891	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 36); \
60197	59892	})
60198	59893	#endif
60199	59894
60200	59895	#ifdef __LITTLE_ENDIAN__
60201		#define vqrdmulhs_laneq_s32(__p0_767, __p1_767, __p2_767) __extension__ ({ \
60202		int32_t __ret_767; \
60203		int32_t __s0_767 = __p0_767; \
60204		int32x4_t __s1_767 = __p1_767; \
60205		__ret_767 = vqrdmulhs_s32(__s0_767, vgetq_lane_s32(__s1_767, __p2_767)); \
60206		__ret_767; \
	59896	#define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59897	poly64x2x4_t __s1 = __p1; \
	59898	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 38); \
60207	59899	})
60208	59900	#else
60209		#define vqrdmulhs_laneq_s32(__p0_768, __p1_768, __p2_768) __extension__ ({ \
60210		int32_t __ret_768; \
60211		int32_t __s0_768 = __p0_768; \
60212		int32x4_t __s1_768 = __p1_768; \
60213		int32x4_t __rev1_768; __rev1_768 = __builtin_shufflevector(__s1_768, __s1_768, 3, 2, 1, 0); \
60214		__ret_768 = vqrdmulhs_s32(__s0_768, __noswap_vgetq_lane_s32(__rev1_768, __p2_768)); \
60215		__ret_768; \
	59901	#define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
	59902	poly64x2x4_t __s1 = __p1; \
	59903	poly64x2x4_t __rev1; \
	59904	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59905	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59906	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59907	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59908	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 38); \
60216	59909	})
60217	59910	#endif
60218	59911
60219	59912	#ifdef __LITTLE_ENDIAN__
60220		#define vqrdmulhh_laneq_s16(__p0_769, __p1_769, __p2_769) __extension__ ({ \
60221		int16_t __ret_769; \
60222		int16_t __s0_769 = __p0_769; \
60223		int16x8_t __s1_769 = __p1_769; \
60224		__ret_769 = vqrdmulhh_s16(__s0_769, vgetq_lane_s16(__s1_769, __p2_769)); \
60225		__ret_769; \
	59913	#define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	59914	uint8x16x4_t __s1 = __p1; \
	59915	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 48); \
60226	59916	})
60227	59917	#else
60228		#define vqrdmulhh_laneq_s16(__p0_770, __p1_770, __p2_770) __extension__ ({ \
60229		int16_t __ret_770; \
60230		int16_t __s0_770 = __p0_770; \
60231		int16x8_t __s1_770 = __p1_770; \
60232		int16x8_t __rev1_770; __rev1_770 = __builtin_shufflevector(__s1_770, __s1_770, 7, 6, 5, 4, 3, 2, 1, 0); \
60233		__ret_770 = vqrdmulhh_s16(__s0_770, __noswap_vgetq_lane_s16(__rev1_770, __p2_770)); \
60234		__ret_770; \
	59918	#define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
	59919	uint8x16x4_t __s1 = __p1; \
	59920	uint8x16x4_t __rev1; \
	59921	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59922	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59923	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59924	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59925	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 48); \
60235	59926	})
60236	59927	#endif
60237	59928
60238	59929	#ifdef __LITTLE_ENDIAN__
60239		#define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60240		int32x4_t __ret; \
60241		int32x4_t __s0 = __p0; \
60242		int32x4_t __s1 = __p1; \
60243		__ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
60244		__ret; \
	59930	#define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59931	uint64x2x4_t __s1 = __p1; \
	59932	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 51); \
60245	59933	})
60246	59934	#else
60247		#define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60248		int32x4_t __ret; \
60249		int32x4_t __s0 = __p0; \
60250		int32x4_t __s1 = __p1; \
60251		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
60252		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60253		__ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
60254		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
60255		__ret; \
	59935	#define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59936	uint64x2x4_t __s1 = __p1; \
	59937	uint64x2x4_t __rev1; \
	59938	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59939	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59940	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59941	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59942	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 51); \
60256	59943	})
60257	59944	#endif
60258	59945
60259	59946	#ifdef __LITTLE_ENDIAN__
60260		#define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60261		int16x8_t __ret; \
60262		int16x8_t __s0 = __p0; \
60263		int16x8_t __s1 = __p1; \
60264		__ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
60265		__ret; \
	59947	#define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	59948	int8x16x4_t __s1 = __p1; \
	59949	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 32); \
60266	59950	})
60267	59951	#else
60268		#define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60269		int16x8_t __ret; \
60270		int16x8_t __s0 = __p0; \
60271		int16x8_t __s1 = __p1; \
60272		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
60273		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
60274		__ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
60275		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
60276		__ret; \
	59952	#define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
	59953	int8x16x4_t __s1 = __p1; \
	59954	int8x16x4_t __rev1; \
	59955	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59956	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59957	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59958	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	59959	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 32); \
60277	59960	})
60278	59961	#endif
60279	59962
60280	59963	#ifdef __LITTLE_ENDIAN__
60281		#define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60282		int32x2_t __ret; \
60283		int32x2_t __s0 = __p0; \
60284		int32x4_t __s1 = __p1; \
60285		__ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
60286		__ret; \
	59964	#define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59965	float64x2x4_t __s1 = __p1; \
	59966	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 42); \
60287	59967	})
60288	59968	#else
60289		#define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60290		int32x2_t __ret; \
60291		int32x2_t __s0 = __p0; \
60292		int32x4_t __s1 = __p1; \
60293		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
60294		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60295		__ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
60296		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
60297		__ret; \
	59969	#define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	59970	float64x2x4_t __s1 = __p1; \
	59971	float64x2x4_t __rev1; \
	59972	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59973	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59974	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59975	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59976	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 42); \
60298	59977	})
60299	59978	#endif
60300	59979
60301	59980	#ifdef __LITTLE_ENDIAN__
60302		#define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60303		int16x4_t __ret; \
60304		int16x4_t __s0 = __p0; \
60305		int16x8_t __s1 = __p1; \
60306		__ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
60307		__ret; \
	59981	#define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59982	int64x2x4_t __s1 = __p1; \
	59983	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 35); \
60308	59984	})
60309	59985	#else
60310		#define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60311		int16x4_t __ret; \
60312		int16x4_t __s0 = __p0; \
60313		int16x8_t __s1 = __p1; \
60314		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
60315		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
60316		__ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
60317		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
60318		__ret; \
	59986	#define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	59987	int64x2x4_t __s1 = __p1; \
	59988	int64x2x4_t __rev1; \
	59989	__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
	59990	__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
	59991	__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
	59992	__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
	59993	__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 35); \
60319	59994	})
60320	59995	#endif
60321	59996
60322		__ai uint8_t vqrshlb_u8(uint8_t __p0, int8_t __p1) {
60323		uint8_t __ret;
60324		__ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
	59997	#define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
	59998	uint64x1x4_t __s1 = __p1; \
	59999	__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
	60000	})
	60001	#define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
	60002	float64x1x4_t __s1 = __p1; \
	60003	__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 10); \
	60004	})
	60005	#define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
	60006	int64x1x4_t __s1 = __p1; \
	60007	__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 3); \
	60008	})
	60009	#define vstrq_p128(__p0, __p1) __extension__ ({ \
	60010	poly128_t __s1 = __p1; \
	60011	__builtin_neon_vstrq_p128(__p0, __s1); \
	60012	})
	60013	__ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
	60014	uint64_t __ret;
	60015	__ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
60325	60016	return __ret;
60326	60017	}
60327		__ai uint32_t vqrshls_u32(uint32_t __p0, int32_t __p1) {
60328		uint32_t __ret;
60329		__ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
	60018	__ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
	60019	int64_t __ret;
	60020	__ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
60330	60021	return __ret;
60331	60022	}
60332		__ai uint64_t vqrshld_u64(uint64_t __p0, int64_t __p1) {
60333		uint64_t __ret;
60334		__ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
	60023	#ifdef __LITTLE_ENDIAN__
	60024	__ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
	60025	float64x2_t __ret;
	60026	__ret = __p0 - __p1;
60335	60027	return __ret;
60336	60028	}
60337		__ai uint16_t vqrshlh_u16(uint16_t __p0, int16_t __p1) {
60338		uint16_t __ret;
60339		__ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
	60029	#else
	60030	__ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
	60031	float64x2_t __ret;
	60032	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60033	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60034	__ret = __rev0 - __rev1;
	60035	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
60340	60036	return __ret;
60341	60037	}
60342		__ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
60343		int8_t __ret;
60344		__ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
	60038	#endif
	60039
	60040	__ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
	60041	float64x1_t __ret;
	60042	__ret = __p0 - __p1;
60345	60043	return __ret;
60346	60044	}
60347		__ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
60348		int32_t __ret;
60349		__ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
	60045	#ifdef __LITTLE_ENDIAN__
	60046	__ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	60047	uint16x8_t __ret;
	60048	__ret = vcombine_u16(__p0, vsubhn_u32(__p1, __p2));
60350	60049	return __ret;
60351	60050	}
60352		__ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
60353		int64_t __ret;
60354		__ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
	60051	#else
	60052	__ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	60053	uint16x8_t __ret;
	60054	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60055	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60056	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	60057	__ret = __noswap_vcombine_u16(__rev0, __noswap_vsubhn_u32(__rev1, __rev2));
	60058	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
60355	60059	return __ret;
60356	60060	}
60357		__ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
60358		int16_t __ret;
60359		__ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
	60061	#endif
	60062
	60063	#ifdef __LITTLE_ENDIAN__
	60064	__ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	60065	uint32x4_t __ret;
	60066	__ret = vcombine_u32(__p0, vsubhn_u64(__p1, __p2));
60360	60067	return __ret;
60361	60068	}
60362		#ifdef __LITTLE_ENDIAN__
60363		#define vqrshrn_high_n_u32(__p0_771, __p1_771, __p2_771) __extension__ ({ \
60364		uint16x8_t __ret_771; \
60365		uint16x4_t __s0_771 = __p0_771; \
60366		uint32x4_t __s1_771 = __p1_771; \
60367		__ret_771 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_771), (uint16x4_t)(vqrshrn_n_u32(__s1_771, __p2_771)))); \
60368		__ret_771; \
60369		})
60370	60069	#else
60371		#define vqrshrn_high_n_u32(__p0_772, __p1_772, __p2_772) __extension__ ({ \
60372		uint16x8_t __ret_772; \
60373		uint16x4_t __s0_772 = __p0_772; \
60374		uint32x4_t __s1_772 = __p1_772; \
60375		uint16x4_t __rev0_772; __rev0_772 = __builtin_shufflevector(__s0_772, __s0_772, 3, 2, 1, 0); \
60376		uint32x4_t __rev1_772; __rev1_772 = __builtin_shufflevector(__s1_772, __s1_772, 3, 2, 1, 0); \
60377		__ret_772 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_772), (uint16x4_t)(__noswap_vqrshrn_n_u32(__rev1_772, __p2_772)))); \
60378		__ret_772 = __builtin_shufflevector(__ret_772, __ret_772, 7, 6, 5, 4, 3, 2, 1, 0); \
60379		__ret_772; \
60380		})
	60070	__ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	60071	uint32x4_t __ret;
	60072	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60073	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60074	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	60075	__ret = __noswap_vcombine_u32(__rev0, __noswap_vsubhn_u64(__rev1, __rev2));
	60076	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60077	return __ret;
	60078	}
60381	60079	#endif
60382	60080
60383	60081	#ifdef __LITTLE_ENDIAN__
60384		#define vqrshrn_high_n_u64(__p0_773, __p1_773, __p2_773) __extension__ ({ \
60385		uint32x4_t __ret_773; \
60386		uint32x2_t __s0_773 = __p0_773; \
60387		uint64x2_t __s1_773 = __p1_773; \
60388		__ret_773 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_773), (uint32x2_t)(vqrshrn_n_u64(__s1_773, __p2_773)))); \
60389		__ret_773; \
60390		})
	60082	__ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	60083	uint8x16_t __ret;
	60084	__ret = vcombine_u8(__p0, vsubhn_u16(__p1, __p2));
	60085	return __ret;
	60086	}
60391	60087	#else
60392		#define vqrshrn_high_n_u64(__p0_774, __p1_774, __p2_774) __extension__ ({ \
60393		uint32x4_t __ret_774; \
60394		uint32x2_t __s0_774 = __p0_774; \
60395		uint64x2_t __s1_774 = __p1_774; \
60396		uint32x2_t __rev0_774; __rev0_774 = __builtin_shufflevector(__s0_774, __s0_774, 1, 0); \
60397		uint64x2_t __rev1_774; __rev1_774 = __builtin_shufflevector(__s1_774, __s1_774, 1, 0); \
60398		__ret_774 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_774), (uint32x2_t)(__noswap_vqrshrn_n_u64(__rev1_774, __p2_774)))); \
60399		__ret_774 = __builtin_shufflevector(__ret_774, __ret_774, 3, 2, 1, 0); \
60400		__ret_774; \
60401		})
	60088	__ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	60089	uint8x16_t __ret;
	60090	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60091	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60092	uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	60093	__ret = __noswap_vcombine_u8(__rev0, __noswap_vsubhn_u16(__rev1, __rev2));
	60094	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60095	return __ret;
	60096	}
60402	60097	#endif
60403	60098
60404	60099	#ifdef __LITTLE_ENDIAN__
60405		#define vqrshrn_high_n_u16(__p0_775, __p1_775, __p2_775) __extension__ ({ \
60406		uint8x16_t __ret_775; \
60407		uint8x8_t __s0_775 = __p0_775; \
60408		uint16x8_t __s1_775 = __p1_775; \
60409		__ret_775 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_775), (uint8x8_t)(vqrshrn_n_u16(__s1_775, __p2_775)))); \
60410		__ret_775; \
60411		})
	60100	__ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	60101	int16x8_t __ret;
	60102	__ret = vcombine_s16(__p0, vsubhn_s32(__p1, __p2));
	60103	return __ret;
	60104	}
60412	60105	#else
60413		#define vqrshrn_high_n_u16(__p0_776, __p1_776, __p2_776) __extension__ ({ \
60414		uint8x16_t __ret_776; \
60415		uint8x8_t __s0_776 = __p0_776; \
60416		uint16x8_t __s1_776 = __p1_776; \
60417		uint8x8_t __rev0_776; __rev0_776 = __builtin_shufflevector(__s0_776, __s0_776, 7, 6, 5, 4, 3, 2, 1, 0); \
60418		uint16x8_t __rev1_776; __rev1_776 = __builtin_shufflevector(__s1_776, __s1_776, 7, 6, 5, 4, 3, 2, 1, 0); \
60419		__ret_776 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_776), (uint8x8_t)(__noswap_vqrshrn_n_u16(__rev1_776, __p2_776)))); \
60420		__ret_776 = __builtin_shufflevector(__ret_776, __ret_776, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60421		__ret_776; \
60422		})
	60106	__ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	60107	int16x8_t __ret;
	60108	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60109	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60110	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	60111	__ret = __noswap_vcombine_s16(__rev0, __noswap_vsubhn_s32(__rev1, __rev2));
	60112	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60113	return __ret;
	60114	}
60423	60115	#endif
60424	60116
60425	60117	#ifdef __LITTLE_ENDIAN__
60426		#define vqrshrn_high_n_s32(__p0_777, __p1_777, __p2_777) __extension__ ({ \
60427		int16x8_t __ret_777; \
60428		int16x4_t __s0_777 = __p0_777; \
60429		int32x4_t __s1_777 = __p1_777; \
60430		__ret_777 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_777), (int16x4_t)(vqrshrn_n_s32(__s1_777, __p2_777)))); \
60431		__ret_777; \
60432		})
	60118	__ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	60119	int32x4_t __ret;
	60120	__ret = vcombine_s32(__p0, vsubhn_s64(__p1, __p2));
	60121	return __ret;
	60122	}
60433	60123	#else
60434		#define vqrshrn_high_n_s32(__p0_778, __p1_778, __p2_778) __extension__ ({ \
60435		int16x8_t __ret_778; \
60436		int16x4_t __s0_778 = __p0_778; \
60437		int32x4_t __s1_778 = __p1_778; \
60438		int16x4_t __rev0_778; __rev0_778 = __builtin_shufflevector(__s0_778, __s0_778, 3, 2, 1, 0); \
60439		int32x4_t __rev1_778; __rev1_778 = __builtin_shufflevector(__s1_778, __s1_778, 3, 2, 1, 0); \
60440		__ret_778 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_778), (int16x4_t)(__noswap_vqrshrn_n_s32(__rev1_778, __p2_778)))); \
60441		__ret_778 = __builtin_shufflevector(__ret_778, __ret_778, 7, 6, 5, 4, 3, 2, 1, 0); \
60442		__ret_778; \
60443		})
	60124	__ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	60125	int32x4_t __ret;
	60126	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60127	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60128	int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	60129	__ret = __noswap_vcombine_s32(__rev0, __noswap_vsubhn_s64(__rev1, __rev2));
	60130	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60131	return __ret;
	60132	}
60444	60133	#endif
60445	60134
60446	60135	#ifdef __LITTLE_ENDIAN__
60447		#define vqrshrn_high_n_s64(__p0_779, __p1_779, __p2_779) __extension__ ({ \
60448		int32x4_t __ret_779; \
60449		int32x2_t __s0_779 = __p0_779; \
60450		int64x2_t __s1_779 = __p1_779; \
60451		__ret_779 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_779), (int32x2_t)(vqrshrn_n_s64(__s1_779, __p2_779)))); \
60452		__ret_779; \
60453		})
	60136	__ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	60137	int8x16_t __ret;
	60138	__ret = vcombine_s8(__p0, vsubhn_s16(__p1, __p2));
	60139	return __ret;
	60140	}
60454	60141	#else
60455		#define vqrshrn_high_n_s64(__p0_780, __p1_780, __p2_780) __extension__ ({ \
60456		int32x4_t __ret_780; \
60457		int32x2_t __s0_780 = __p0_780; \
60458		int64x2_t __s1_780 = __p1_780; \
60459		int32x2_t __rev0_780; __rev0_780 = __builtin_shufflevector(__s0_780, __s0_780, 1, 0); \
60460		int64x2_t __rev1_780; __rev1_780 = __builtin_shufflevector(__s1_780, __s1_780, 1, 0); \
60461		__ret_780 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_780), (int32x2_t)(__noswap_vqrshrn_n_s64(__rev1_780, __p2_780)))); \
60462		__ret_780 = __builtin_shufflevector(__ret_780, __ret_780, 3, 2, 1, 0); \
60463		__ret_780; \
60464		})
	60142	__ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	60143	int8x16_t __ret;
	60144	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60145	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60146	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	60147	__ret = __noswap_vcombine_s8(__rev0, __noswap_vsubhn_s16(__rev1, __rev2));
	60148	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60149	return __ret;
	60150	}
60465	60151	#endif
60466	60152
60467	60153	#ifdef __LITTLE_ENDIAN__
60468		#define vqrshrn_high_n_s16(__p0_781, __p1_781, __p2_781) __extension__ ({ \
60469		int8x16_t __ret_781; \
60470		int8x8_t __s0_781 = __p0_781; \
60471		int16x8_t __s1_781 = __p1_781; \
60472		__ret_781 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_781), (int8x8_t)(vqrshrn_n_s16(__s1_781, __p2_781)))); \
60473		__ret_781; \
60474		})
	60154	__ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
	60155	uint16x8_t __ret;
	60156	__ret = vmovl_high_u8(__p0) - vmovl_high_u8(__p1);
	60157	return __ret;
	60158	}
60475	60159	#else
60476		#define vqrshrn_high_n_s16(__p0_782, __p1_782, __p2_782) __extension__ ({ \
60477		int8x16_t __ret_782; \
60478		int8x8_t __s0_782 = __p0_782; \
60479		int16x8_t __s1_782 = __p1_782; \
60480		int8x8_t __rev0_782; __rev0_782 = __builtin_shufflevector(__s0_782, __s0_782, 7, 6, 5, 4, 3, 2, 1, 0); \
60481		int16x8_t __rev1_782; __rev1_782 = __builtin_shufflevector(__s1_782, __s1_782, 7, 6, 5, 4, 3, 2, 1, 0); \
60482		__ret_782 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_782), (int8x8_t)(__noswap_vqrshrn_n_s16(__rev1_782, __p2_782)))); \
60483		__ret_782 = __builtin_shufflevector(__ret_782, __ret_782, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60484		__ret_782; \
60485		})
60486		#endif
60487
60488		#define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
60489		uint16_t __ret; \
60490		uint32_t __s0 = __p0; \
60491		__ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
60492		__ret; \
60493		})
60494		#define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
60495		uint32_t __ret; \
60496		uint64_t __s0 = __p0; \
60497		__ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
60498		__ret; \
60499		})
60500		#define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
60501		uint8_t __ret; \
60502		uint16_t __s0 = __p0; \
60503		__ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
60504		__ret; \
60505		})
60506		#define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
60507		int16_t __ret; \
60508		int32_t __s0 = __p0; \
60509		__ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
60510		__ret; \
60511		})
60512		#define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
60513		int32_t __ret; \
60514		int64_t __s0 = __p0; \
60515		__ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
60516		__ret; \
60517		})
60518		#define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
60519		int8_t __ret; \
60520		int16_t __s0 = __p0; \
60521		__ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
60522		__ret; \
60523		})
	60160	__ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
	60161	uint16x8_t __ret;
	60162	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60163	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60164	__ret = __noswap_vmovl_high_u8(__rev0) - __noswap_vmovl_high_u8(__rev1);
	60165	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60166	return __ret;
	60167	}
	60168	#endif
	60169
60524	60170	#ifdef __LITTLE_ENDIAN__
60525		#define vqrshrun_high_n_s32(__p0_783, __p1_783, __p2_783) __extension__ ({ \
60526		int16x8_t __ret_783; \
60527		int16x4_t __s0_783 = __p0_783; \
60528		int32x4_t __s1_783 = __p1_783; \
60529		__ret_783 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_783), (int16x4_t)(vqrshrun_n_s32(__s1_783, __p2_783)))); \
60530		__ret_783; \
60531		})
	60171	__ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
	60172	uint64x2_t __ret;
	60173	__ret = vmovl_high_u32(__p0) - vmovl_high_u32(__p1);
	60174	return __ret;
	60175	}
60532	60176	#else
60533		#define vqrshrun_high_n_s32(__p0_784, __p1_784, __p2_784) __extension__ ({ \
60534		int16x8_t __ret_784; \
60535		int16x4_t __s0_784 = __p0_784; \
60536		int32x4_t __s1_784 = __p1_784; \
60537		int16x4_t __rev0_784; __rev0_784 = __builtin_shufflevector(__s0_784, __s0_784, 3, 2, 1, 0); \
60538		int32x4_t __rev1_784; __rev1_784 = __builtin_shufflevector(__s1_784, __s1_784, 3, 2, 1, 0); \
60539		__ret_784 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_784), (int16x4_t)(__noswap_vqrshrun_n_s32(__rev1_784, __p2_784)))); \
60540		__ret_784 = __builtin_shufflevector(__ret_784, __ret_784, 7, 6, 5, 4, 3, 2, 1, 0); \
60541		__ret_784; \
60542		})
	60177	__ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
	60178	uint64x2_t __ret;
	60179	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60180	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60181	__ret = __noswap_vmovl_high_u32(__rev0) - __noswap_vmovl_high_u32(__rev1);
	60182	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	60183	return __ret;
	60184	}
60543	60185	#endif
60544	60186
60545	60187	#ifdef __LITTLE_ENDIAN__
60546		#define vqrshrun_high_n_s64(__p0_785, __p1_785, __p2_785) __extension__ ({ \
60547		int32x4_t __ret_785; \
60548		int32x2_t __s0_785 = __p0_785; \
60549		int64x2_t __s1_785 = __p1_785; \
60550		__ret_785 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_785), (int32x2_t)(vqrshrun_n_s64(__s1_785, __p2_785)))); \
60551		__ret_785; \
60552		})
	60188	__ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
	60189	uint32x4_t __ret;
	60190	__ret = vmovl_high_u16(__p0) - vmovl_high_u16(__p1);
	60191	return __ret;
	60192	}
60553	60193	#else
60554		#define vqrshrun_high_n_s64(__p0_786, __p1_786, __p2_786) __extension__ ({ \
60555		int32x4_t __ret_786; \
60556		int32x2_t __s0_786 = __p0_786; \
60557		int64x2_t __s1_786 = __p1_786; \
60558		int32x2_t __rev0_786; __rev0_786 = __builtin_shufflevector(__s0_786, __s0_786, 1, 0); \
60559		int64x2_t __rev1_786; __rev1_786 = __builtin_shufflevector(__s1_786, __s1_786, 1, 0); \
60560		__ret_786 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_786), (int32x2_t)(__noswap_vqrshrun_n_s64(__rev1_786, __p2_786)))); \
60561		__ret_786 = __builtin_shufflevector(__ret_786, __ret_786, 3, 2, 1, 0); \
60562		__ret_786; \
60563		})
	60194	__ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
	60195	uint32x4_t __ret;
	60196	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60197	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60198	__ret = __noswap_vmovl_high_u16(__rev0) - __noswap_vmovl_high_u16(__rev1);
	60199	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60200	return __ret;
	60201	}
60564	60202	#endif
60565	60203
60566	60204	#ifdef __LITTLE_ENDIAN__
60567		#define vqrshrun_high_n_s16(__p0_787, __p1_787, __p2_787) __extension__ ({ \
60568		int8x16_t __ret_787; \
60569		int8x8_t __s0_787 = __p0_787; \
60570		int16x8_t __s1_787 = __p1_787; \
60571		__ret_787 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_787), (int8x8_t)(vqrshrun_n_s16(__s1_787, __p2_787)))); \
60572		__ret_787; \
60573		})
	60205	__ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
	60206	int16x8_t __ret;
	60207	__ret = vmovl_high_s8(__p0) - vmovl_high_s8(__p1);
	60208	return __ret;
	60209	}
60574	60210	#else
60575		#define vqrshrun_high_n_s16(__p0_788, __p1_788, __p2_788) __extension__ ({ \
60576		int8x16_t __ret_788; \
60577		int8x8_t __s0_788 = __p0_788; \
60578		int16x8_t __s1_788 = __p1_788; \
60579		int8x8_t __rev0_788; __rev0_788 = __builtin_shufflevector(__s0_788, __s0_788, 7, 6, 5, 4, 3, 2, 1, 0); \
60580		int16x8_t __rev1_788; __rev1_788 = __builtin_shufflevector(__s1_788, __s1_788, 7, 6, 5, 4, 3, 2, 1, 0); \
60581		__ret_788 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_788), (int8x8_t)(__noswap_vqrshrun_n_s16(__rev1_788, __p2_788)))); \
60582		__ret_788 = __builtin_shufflevector(__ret_788, __ret_788, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60583		__ret_788; \
60584		})
	60211	__ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
	60212	int16x8_t __ret;
	60213	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60214	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60215	__ret = __noswap_vmovl_high_s8(__rev0) - __noswap_vmovl_high_s8(__rev1);
	60216	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60217	return __ret;
	60218	}
60585	60219	#endif
60586	60220
60587		#define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
60588		int16_t __ret; \
60589		int32_t __s0 = __p0; \
60590		__ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
60591		__ret; \
60592		})
60593		#define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
60594		int32_t __ret; \
60595		int64_t __s0 = __p0; \
60596		__ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
60597		__ret; \
60598		})
60599		#define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
60600		int8_t __ret; \
60601		int16_t __s0 = __p0; \
60602		__ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
60603		__ret; \
60604		})
60605		__ai uint8_t vqshlb_u8(uint8_t __p0, int8_t __p1) {
60606		uint8_t __ret;
60607		__ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
	60221	#ifdef __LITTLE_ENDIAN__
	60222	__ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
	60223	int64x2_t __ret;
	60224	__ret = vmovl_high_s32(__p0) - vmovl_high_s32(__p1);
60608	60225	return __ret;
60609	60226	}
60610		__ai uint32_t vqshls_u32(uint32_t __p0, int32_t __p1) {
60611		uint32_t __ret;
60612		__ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
	60227	#else
	60228	__ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
	60229	int64x2_t __ret;
	60230	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60231	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60232	__ret = __noswap_vmovl_high_s32(__rev0) - __noswap_vmovl_high_s32(__rev1);
	60233	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
60613	60234	return __ret;
60614	60235	}
60615		__ai uint64_t vqshld_u64(uint64_t __p0, int64_t __p1) {
60616		uint64_t __ret;
60617		__ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
	60236	#endif
	60237
	60238	#ifdef __LITTLE_ENDIAN__
	60239	__ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
	60240	int32x4_t __ret;
	60241	__ret = vmovl_high_s16(__p0) - vmovl_high_s16(__p1);
60618	60242	return __ret;
60619	60243	}
60620		__ai uint16_t vqshlh_u16(uint16_t __p0, int16_t __p1) {
60621		uint16_t __ret;
60622		__ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
	60244	#else
	60245	__ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
	60246	int32x4_t __ret;
	60247	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60248	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60249	__ret = __noswap_vmovl_high_s16(__rev0) - __noswap_vmovl_high_s16(__rev1);
	60250	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
60623	60251	return __ret;
60624	60252	}
60625		__ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
60626		int8_t __ret;
60627		__ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
	60253	#endif
	60254
	60255	#ifdef __LITTLE_ENDIAN__
	60256	__ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
	60257	uint16x8_t __ret;
	60258	__ret = __p0 - vmovl_high_u8(__p1);
60628	60259	return __ret;
60629	60260	}
60630		__ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
60631		int32_t __ret;
60632		__ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
	60261	#else
	60262	__ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
	60263	uint16x8_t __ret;
	60264	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60265	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60266	__ret = __rev0 - __noswap_vmovl_high_u8(__rev1);
	60267	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
60633	60268	return __ret;
60634	60269	}
60635		__ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
60636		int64_t __ret;
60637		__ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
	60270	#endif
	60271
	60272	#ifdef __LITTLE_ENDIAN__
	60273	__ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
	60274	uint64x2_t __ret;
	60275	__ret = __p0 - vmovl_high_u32(__p1);
60638	60276	return __ret;
60639	60277	}
60640		__ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
60641		int16_t __ret;
60642		__ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
	60278	#else
	60279	__ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
	60280	uint64x2_t __ret;
	60281	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60282	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60283	__ret = __rev0 - __noswap_vmovl_high_u32(__rev1);
	60284	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
60643	60285	return __ret;
60644	60286	}
60645		#define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
60646		uint8_t __ret; \
60647		uint8_t __s0 = __p0; \
60648		__ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
60649		__ret; \
60650		})
60651		#define vqshls_n_u32(__p0, __p1) __extension__ ({ \
60652		uint32_t __ret; \
60653		uint32_t __s0 = __p0; \
60654		__ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
60655		__ret; \
60656		})
60657		#define vqshld_n_u64(__p0, __p1) __extension__ ({ \
60658		uint64_t __ret; \
60659		uint64_t __s0 = __p0; \
60660		__ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
60661		__ret; \
60662		})
60663		#define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
60664		uint16_t __ret; \
60665		uint16_t __s0 = __p0; \
60666		__ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
60667		__ret; \
60668		})
60669		#define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
60670		int8_t __ret; \
60671		int8_t __s0 = __p0; \
60672		__ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
60673		__ret; \
60674		})
60675		#define vqshls_n_s32(__p0, __p1) __extension__ ({ \
60676		int32_t __ret; \
60677		int32_t __s0 = __p0; \
60678		__ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
60679		__ret; \
60680		})
60681		#define vqshld_n_s64(__p0, __p1) __extension__ ({ \
60682		int64_t __ret; \
60683		int64_t __s0 = __p0; \
60684		__ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
60685		__ret; \
60686		})
60687		#define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
60688		int16_t __ret; \
60689		int16_t __s0 = __p0; \
60690		__ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
60691		__ret; \
60692		})
60693		#define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
60694		int8_t __ret; \
60695		int8_t __s0 = __p0; \
60696		__ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
60697		__ret; \
60698		})
60699		#define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
60700		int32_t __ret; \
60701		int32_t __s0 = __p0; \
60702		__ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
60703		__ret; \
60704		})
60705		#define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
60706		int64_t __ret; \
60707		int64_t __s0 = __p0; \
60708		__ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
60709		__ret; \
60710		})
60711		#define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
60712		int16_t __ret; \
60713		int16_t __s0 = __p0; \
60714		__ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
60715		__ret; \
60716		})
	60287	#endif
	60288
60717	60289	#ifdef __LITTLE_ENDIAN__
60718		#define vqshrn_high_n_u32(__p0_789, __p1_789, __p2_789) __extension__ ({ \
60719		uint16x8_t __ret_789; \
60720		uint16x4_t __s0_789 = __p0_789; \
60721		uint32x4_t __s1_789 = __p1_789; \
60722		__ret_789 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_789), (uint16x4_t)(vqshrn_n_u32(__s1_789, __p2_789)))); \
60723		__ret_789; \
60724		})
	60290	__ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
	60291	uint32x4_t __ret;
	60292	__ret = __p0 - vmovl_high_u16(__p1);
	60293	return __ret;
	60294	}
60725	60295	#else
60726		#define vqshrn_high_n_u32(__p0_790, __p1_790, __p2_790) __extension__ ({ \
60727		uint16x8_t __ret_790; \
60728		uint16x4_t __s0_790 = __p0_790; \
60729		uint32x4_t __s1_790 = __p1_790; \
60730		uint16x4_t __rev0_790; __rev0_790 = __builtin_shufflevector(__s0_790, __s0_790, 3, 2, 1, 0); \
60731		uint32x4_t __rev1_790; __rev1_790 = __builtin_shufflevector(__s1_790, __s1_790, 3, 2, 1, 0); \
60732		__ret_790 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_790), (uint16x4_t)(__noswap_vqshrn_n_u32(__rev1_790, __p2_790)))); \
60733		__ret_790 = __builtin_shufflevector(__ret_790, __ret_790, 7, 6, 5, 4, 3, 2, 1, 0); \
60734		__ret_790; \
60735		})
	60296	__ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
	60297	uint32x4_t __ret;
	60298	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60299	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60300	__ret = __rev0 - __noswap_vmovl_high_u16(__rev1);
	60301	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60302	return __ret;
	60303	}
60736	60304	#endif
60737	60305
60738	60306	#ifdef __LITTLE_ENDIAN__
60739		#define vqshrn_high_n_u64(__p0_791, __p1_791, __p2_791) __extension__ ({ \
60740		uint32x4_t __ret_791; \
60741		uint32x2_t __s0_791 = __p0_791; \
60742		uint64x2_t __s1_791 = __p1_791; \
60743		__ret_791 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_791), (uint32x2_t)(vqshrn_n_u64(__s1_791, __p2_791)))); \
60744		__ret_791; \
60745		})
	60307	__ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
	60308	int16x8_t __ret;
	60309	__ret = __p0 - vmovl_high_s8(__p1);
	60310	return __ret;
	60311	}
60746	60312	#else
60747		#define vqshrn_high_n_u64(__p0_792, __p1_792, __p2_792) __extension__ ({ \
60748		uint32x4_t __ret_792; \
60749		uint32x2_t __s0_792 = __p0_792; \
60750		uint64x2_t __s1_792 = __p1_792; \
60751		uint32x2_t __rev0_792; __rev0_792 = __builtin_shufflevector(__s0_792, __s0_792, 1, 0); \
60752		uint64x2_t __rev1_792; __rev1_792 = __builtin_shufflevector(__s1_792, __s1_792, 1, 0); \
60753		__ret_792 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_792), (uint32x2_t)(__noswap_vqshrn_n_u64(__rev1_792, __p2_792)))); \
60754		__ret_792 = __builtin_shufflevector(__ret_792, __ret_792, 3, 2, 1, 0); \
60755		__ret_792; \
60756		})
	60313	__ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
	60314	int16x8_t __ret;
	60315	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60316	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60317	__ret = __rev0 - __noswap_vmovl_high_s8(__rev1);
	60318	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60319	return __ret;
	60320	}
60757	60321	#endif
60758	60322
60759	60323	#ifdef __LITTLE_ENDIAN__
60760		#define vqshrn_high_n_u16(__p0_793, __p1_793, __p2_793) __extension__ ({ \
60761		uint8x16_t __ret_793; \
60762		uint8x8_t __s0_793 = __p0_793; \
60763		uint16x8_t __s1_793 = __p1_793; \
60764		__ret_793 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_793), (uint8x8_t)(vqshrn_n_u16(__s1_793, __p2_793)))); \
60765		__ret_793; \
60766		})
	60324	__ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
	60325	int64x2_t __ret;
	60326	__ret = __p0 - vmovl_high_s32(__p1);
	60327	return __ret;
	60328	}
60767	60329	#else
60768		#define vqshrn_high_n_u16(__p0_794, __p1_794, __p2_794) __extension__ ({ \
60769		uint8x16_t __ret_794; \
60770		uint8x8_t __s0_794 = __p0_794; \
60771		uint16x8_t __s1_794 = __p1_794; \
60772		uint8x8_t __rev0_794; __rev0_794 = __builtin_shufflevector(__s0_794, __s0_794, 7, 6, 5, 4, 3, 2, 1, 0); \
60773		uint16x8_t __rev1_794; __rev1_794 = __builtin_shufflevector(__s1_794, __s1_794, 7, 6, 5, 4, 3, 2, 1, 0); \
60774		__ret_794 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_794), (uint8x8_t)(__noswap_vqshrn_n_u16(__rev1_794, __p2_794)))); \
60775		__ret_794 = __builtin_shufflevector(__ret_794, __ret_794, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60776		__ret_794; \
60777		})
	60330	__ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
	60331	int64x2_t __ret;
	60332	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60333	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60334	__ret = __rev0 - __noswap_vmovl_high_s32(__rev1);
	60335	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	60336	return __ret;
	60337	}
60778	60338	#endif
60779	60339
60780	60340	#ifdef __LITTLE_ENDIAN__
60781		#define vqshrn_high_n_s32(__p0_795, __p1_795, __p2_795) __extension__ ({ \
60782		int16x8_t __ret_795; \
60783		int16x4_t __s0_795 = __p0_795; \
60784		int32x4_t __s1_795 = __p1_795; \
60785		__ret_795 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_795), (int16x4_t)(vqshrn_n_s32(__s1_795, __p2_795)))); \
60786		__ret_795; \
60787		})
	60341	__ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
	60342	int32x4_t __ret;
	60343	__ret = __p0 - vmovl_high_s16(__p1);
	60344	return __ret;
	60345	}
60788	60346	#else
60789		#define vqshrn_high_n_s32(__p0_796, __p1_796, __p2_796) __extension__ ({ \
60790		int16x8_t __ret_796; \
60791		int16x4_t __s0_796 = __p0_796; \
60792		int32x4_t __s1_796 = __p1_796; \
60793		int16x4_t __rev0_796; __rev0_796 = __builtin_shufflevector(__s0_796, __s0_796, 3, 2, 1, 0); \
60794		int32x4_t __rev1_796; __rev1_796 = __builtin_shufflevector(__s1_796, __s1_796, 3, 2, 1, 0); \
60795		__ret_796 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_796), (int16x4_t)(__noswap_vqshrn_n_s32(__rev1_796, __p2_796)))); \
60796		__ret_796 = __builtin_shufflevector(__ret_796, __ret_796, 7, 6, 5, 4, 3, 2, 1, 0); \
60797		__ret_796; \
60798		})
	60347	__ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
	60348	int32x4_t __ret;
	60349	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60350	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60351	__ret = __rev0 - __noswap_vmovl_high_s16(__rev1);
	60352	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60353	return __ret;
	60354	}
60799	60355	#endif
60800	60356
60801	60357	#ifdef __LITTLE_ENDIAN__
60802		#define vqshrn_high_n_s64(__p0_797, __p1_797, __p2_797) __extension__ ({ \
60803		int32x4_t __ret_797; \
60804		int32x2_t __s0_797 = __p0_797; \
60805		int64x2_t __s1_797 = __p1_797; \
60806		__ret_797 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_797), (int32x2_t)(vqshrn_n_s64(__s1_797, __p2_797)))); \
60807		__ret_797; \
60808		})
	60358	__ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
	60359	poly8x8_t __ret;
	60360	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
	60361	return __ret;
	60362	}
60809	60363	#else
60810		#define vqshrn_high_n_s64(__p0_798, __p1_798, __p2_798) __extension__ ({ \
60811		int32x4_t __ret_798; \
60812		int32x2_t __s0_798 = __p0_798; \
60813		int64x2_t __s1_798 = __p1_798; \
60814		int32x2_t __rev0_798; __rev0_798 = __builtin_shufflevector(__s0_798, __s0_798, 1, 0); \
60815		int64x2_t __rev1_798; __rev1_798 = __builtin_shufflevector(__s1_798, __s1_798, 1, 0); \
60816		__ret_798 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_798), (int32x2_t)(__noswap_vqshrn_n_s64(__rev1_798, __p2_798)))); \
60817		__ret_798 = __builtin_shufflevector(__ret_798, __ret_798, 3, 2, 1, 0); \
60818		__ret_798; \
60819		})
	60364	__ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
	60365	poly8x8_t __ret;
	60366	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60367	poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60368	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
	60369	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60370	return __ret;
	60371	}
60820	60372	#endif
60821	60373
60822	60374	#ifdef __LITTLE_ENDIAN__
60823		#define vqshrn_high_n_s16(__p0_799, __p1_799, __p2_799) __extension__ ({ \
60824		int8x16_t __ret_799; \
60825		int8x8_t __s0_799 = __p0_799; \
60826		int16x8_t __s1_799 = __p1_799; \
60827		__ret_799 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_799), (int8x8_t)(vqshrn_n_s16(__s1_799, __p2_799)))); \
60828		__ret_799; \
60829		})
	60375	__ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
	60376	poly16x4_t __ret;
	60377	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
	60378	return __ret;
	60379	}
60830	60380	#else
60831		#define vqshrn_high_n_s16(__p0_800, __p1_800, __p2_800) __extension__ ({ \
60832		int8x16_t __ret_800; \
60833		int8x8_t __s0_800 = __p0_800; \
60834		int16x8_t __s1_800 = __p1_800; \
60835		int8x8_t __rev0_800; __rev0_800 = __builtin_shufflevector(__s0_800, __s0_800, 7, 6, 5, 4, 3, 2, 1, 0); \
60836		int16x8_t __rev1_800; __rev1_800 = __builtin_shufflevector(__s1_800, __s1_800, 7, 6, 5, 4, 3, 2, 1, 0); \
60837		__ret_800 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_800), (int8x8_t)(__noswap_vqshrn_n_s16(__rev1_800, __p2_800)))); \
60838		__ret_800 = __builtin_shufflevector(__ret_800, __ret_800, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60839		__ret_800; \
60840		})
	60381	__ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
	60382	poly16x4_t __ret;
	60383	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60384	poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60385	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	60386	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60387	return __ret;
	60388	}
60841	60389	#endif
60842	60390
60843		#define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
60844		uint16_t __ret; \
60845		uint32_t __s0 = __p0; \
60846		__ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
60847		__ret; \
60848		})
60849		#define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
60850		uint32_t __ret; \
60851		uint64_t __s0 = __p0; \
60852		__ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
60853		__ret; \
60854		})
60855		#define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
60856		uint8_t __ret; \
60857		uint16_t __s0 = __p0; \
60858		__ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
60859		__ret; \
60860		})
60861		#define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
60862		int16_t __ret; \
60863		int32_t __s0 = __p0; \
60864		__ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
60865		__ret; \
60866		})
60867		#define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
60868		int32_t __ret; \
60869		int64_t __s0 = __p0; \
60870		__ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
60871		__ret; \
60872		})
60873		#define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
60874		int8_t __ret; \
60875		int16_t __s0 = __p0; \
60876		__ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
60877		__ret; \
60878		})
60879	60391	#ifdef __LITTLE_ENDIAN__
60880		#define vqshrun_high_n_s32(__p0_801, __p1_801, __p2_801) __extension__ ({ \
60881		int16x8_t __ret_801; \
60882		int16x4_t __s0_801 = __p0_801; \
60883		int32x4_t __s1_801 = __p1_801; \
60884		__ret_801 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_801), (int16x4_t)(vqshrun_n_s32(__s1_801, __p2_801)))); \
60885		__ret_801; \
60886		})
	60392	__ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	60393	poly8x16_t __ret;
	60394	__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
	60395	return __ret;
	60396	}
60887	60397	#else
60888		#define vqshrun_high_n_s32(__p0_802, __p1_802, __p2_802) __extension__ ({ \
60889		int16x8_t __ret_802; \
60890		int16x4_t __s0_802 = __p0_802; \
60891		int32x4_t __s1_802 = __p1_802; \
60892		int16x4_t __rev0_802; __rev0_802 = __builtin_shufflevector(__s0_802, __s0_802, 3, 2, 1, 0); \
60893		int32x4_t __rev1_802; __rev1_802 = __builtin_shufflevector(__s1_802, __s1_802, 3, 2, 1, 0); \
60894		__ret_802 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_802), (int16x4_t)(__noswap_vqshrun_n_s32(__rev1_802, __p2_802)))); \
60895		__ret_802 = __builtin_shufflevector(__ret_802, __ret_802, 7, 6, 5, 4, 3, 2, 1, 0); \
60896		__ret_802; \
60897		})
	60398	__ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	60399	poly8x16_t __ret;
	60400	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60401	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60402	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
	60403	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60404	return __ret;
	60405	}
60898	60406	#endif
60899	60407
60900	60408	#ifdef __LITTLE_ENDIAN__
60901		#define vqshrun_high_n_s64(__p0_803, __p1_803, __p2_803) __extension__ ({ \
60902		int32x4_t __ret_803; \
60903		int32x2_t __s0_803 = __p0_803; \
60904		int64x2_t __s1_803 = __p1_803; \
60905		__ret_803 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_803), (int32x2_t)(vqshrun_n_s64(__s1_803, __p2_803)))); \
60906		__ret_803; \
60907		})
	60409	__ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	60410	poly64x2_t __ret;
	60411	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	60412	return __ret;
	60413	}
60908	60414	#else
60909		#define vqshrun_high_n_s64(__p0_804, __p1_804, __p2_804) __extension__ ({ \
60910		int32x4_t __ret_804; \
60911		int32x2_t __s0_804 = __p0_804; \
60912		int64x2_t __s1_804 = __p1_804; \
60913		int32x2_t __rev0_804; __rev0_804 = __builtin_shufflevector(__s0_804, __s0_804, 1, 0); \
60914		int64x2_t __rev1_804; __rev1_804 = __builtin_shufflevector(__s1_804, __s1_804, 1, 0); \
60915		__ret_804 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_804), (int32x2_t)(__noswap_vqshrun_n_s64(__rev1_804, __p2_804)))); \
60916		__ret_804 = __builtin_shufflevector(__ret_804, __ret_804, 3, 2, 1, 0); \
60917		__ret_804; \
60918		})
	60415	__ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	60416	poly64x2_t __ret;
	60417	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60418	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60419	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60420	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	60421	return __ret;
	60422	}
60919	60423	#endif
60920	60424
60921	60425	#ifdef __LITTLE_ENDIAN__
60922		#define vqshrun_high_n_s16(__p0_805, __p1_805, __p2_805) __extension__ ({ \
60923		int8x16_t __ret_805; \
60924		int8x8_t __s0_805 = __p0_805; \
60925		int16x8_t __s1_805 = __p1_805; \
60926		__ret_805 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_805), (int8x8_t)(vqshrun_n_s16(__s1_805, __p2_805)))); \
60927		__ret_805; \
60928		})
	60426	__ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	60427	poly16x8_t __ret;
	60428	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
	60429	return __ret;
	60430	}
60929	60431	#else
60930		#define vqshrun_high_n_s16(__p0_806, __p1_806, __p2_806) __extension__ ({ \
60931		int8x16_t __ret_806; \
60932		int8x8_t __s0_806 = __p0_806; \
60933		int16x8_t __s1_806 = __p1_806; \
60934		int8x8_t __rev0_806; __rev0_806 = __builtin_shufflevector(__s0_806, __s0_806, 7, 6, 5, 4, 3, 2, 1, 0); \
60935		int16x8_t __rev1_806; __rev1_806 = __builtin_shufflevector(__s1_806, __s1_806, 7, 6, 5, 4, 3, 2, 1, 0); \
60936		__ret_806 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_806), (int8x8_t)(__noswap_vqshrun_n_s16(__rev1_806, __p2_806)))); \
60937		__ret_806 = __builtin_shufflevector(__ret_806, __ret_806, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60938		__ret_806; \
60939		})
	60432	__ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	60433	poly16x8_t __ret;
	60434	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60435	poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60436	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
	60437	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60438	return __ret;
	60439	}
60940	60440	#endif
60941	60441
60942		#define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
60943		int16_t __ret; \
60944		int32_t __s0 = __p0; \
60945		__ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
60946		__ret; \
60947		})
60948		#define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
60949		int32_t __ret; \
60950		int64_t __s0 = __p0; \
60951		__ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
60952		__ret; \
60953		})
60954		#define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
60955		int8_t __ret; \
60956		int16_t __s0 = __p0; \
60957		__ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
60958		__ret; \
60959		})
60960		__ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
60961		uint8_t __ret;
60962		__ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
	60442	#ifdef __LITTLE_ENDIAN__
	60443	__ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	60444	uint8x16_t __ret;
	60445	__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
60963	60446	return __ret;
60964	60447	}
60965		__ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
60966		uint32_t __ret;
60967		__ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
	60448	#else
	60449	__ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	60450	uint8x16_t __ret;
	60451	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60452	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60453	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
	60454	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
60968	60455	return __ret;
60969	60456	}
60970		__ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
60971		uint64_t __ret;
60972		__ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
	60457	#endif
	60458
	60459	#ifdef __LITTLE_ENDIAN__
	60460	__ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	60461	uint32x4_t __ret;
	60462	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
60973	60463	return __ret;
60974	60464	}
60975		__ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
60976		uint16_t __ret;
60977		__ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
	60465	#else
	60466	__ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	60467	uint32x4_t __ret;
	60468	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60469	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60470	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	60471	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
60978	60472	return __ret;
60979	60473	}
60980		__ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
60981		int8_t __ret;
60982		__ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
	60474	#endif
	60475
	60476	#ifdef __LITTLE_ENDIAN__
	60477	__ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	60478	uint64x2_t __ret;
	60479	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
60983	60480	return __ret;
60984	60481	}
60985		__ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
60986		int32_t __ret;
60987		__ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
	60482	#else
	60483	__ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	60484	uint64x2_t __ret;
	60485	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60486	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60487	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60488	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	60489	return __ret;
	60490	}
	60491	#endif
	60492
	60493	#ifdef __LITTLE_ENDIAN__
	60494	__ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	60495	uint16x8_t __ret;
	60496	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
	60497	return __ret;
	60498	}
	60499	#else
	60500	__ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	60501	uint16x8_t __ret;
	60502	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60503	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60504	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
	60505	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
60988	60506	return __ret;
60989	60507	}
60990		__ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
60991		int64_t __ret;
60992		__ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
	60508	#endif
	60509
	60510	#ifdef __LITTLE_ENDIAN__
	60511	__ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
	60512	int8x16_t __ret;
	60513	__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
60993	60514	return __ret;
60994	60515	}
60995		__ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
60996		int16_t __ret;
60997		__ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
	60516	#else
	60517	__ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
	60518	int8x16_t __ret;
	60519	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60520	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60521	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
	60522	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
60998	60523	return __ret;
60999	60524	}
	60525	#endif
	60526
61000	60527	#ifdef __LITTLE_ENDIAN__
61001		__ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
61002		poly8x8_t __ret;
61003		__ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 4);
	60528	__ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
	60529	float64x2_t __ret;
	60530	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61004	60531	return __ret;
61005	60532	}
61006	60533	#else
61007		__ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
61008		poly8x8_t __ret;
61009		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61010		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61011		__ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 4);
61012		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60534	__ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
	60535	float64x2_t __ret;
	60536	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60537	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60538	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60539	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61013	60540	return __ret;
61014	60541	}
61015	60542	#endif
61016	60543
61017	60544	#ifdef __LITTLE_ENDIAN__
61018		__ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
61019		poly8x16_t __ret;
61020		__ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 36);
	60545	__ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
	60546	float32x4_t __ret;
	60547	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
61021	60548	return __ret;
61022	60549	}
61023	60550	#else
61024		__ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
61025		poly8x16_t __ret;
61026		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61027		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61028		__ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 36);
61029		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60551	__ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
	60552	float32x4_t __ret;
	60553	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60554	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60555	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	60556	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61030	60557	return __ret;
61031	60558	}
61032	60559	#endif
61033	60560
61034	60561	#ifdef __LITTLE_ENDIAN__
61035		__ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
61036		uint8x16_t __ret;
61037		__ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	60562	__ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
	60563	int32x4_t __ret;
	60564	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
61038	60565	return __ret;
61039	60566	}
61040	60567	#else
61041		__ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
61042		uint8x16_t __ret;
61043		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61044		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61045		__ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
61046		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60568	__ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
	60569	int32x4_t __ret;
	60570	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60571	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60572	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	60573	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61047	60574	return __ret;
61048	60575	}
61049	60576	#endif
61050	60577
61051	60578	#ifdef __LITTLE_ENDIAN__
61052		__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
61053		int8x16_t __ret;
61054		__ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	60579	__ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
	60580	int64x2_t __ret;
	60581	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61055	60582	return __ret;
61056	60583	}
61057	60584	#else
61058		__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
61059		int8x16_t __ret;
61060		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61061		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61062		__ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
61063		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60585	__ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
	60586	int64x2_t __ret;
	60587	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60588	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60589	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60590	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61064	60591	return __ret;
61065	60592	}
61066	60593	#endif
61067	60594
61068	60595	#ifdef __LITTLE_ENDIAN__
61069		__ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
61070		uint8x8_t __ret;
61071		__ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 16);
	60596	__ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
	60597	int16x8_t __ret;
	60598	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
61072	60599	return __ret;
61073	60600	}
61074	60601	#else
61075		__ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
61076		uint8x8_t __ret;
61077		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61078		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61079		__ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 16);
	60602	__ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
	60603	int16x8_t __ret;
	60604	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60605	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60606	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
61080	60607	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61081	60608	return __ret;
61082	60609	}
61083	60610	#endif
61084	60611
61085	60612	#ifdef __LITTLE_ENDIAN__
61086		__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
61087		int8x8_t __ret;
61088		__ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 0);
	60613	__ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
	60614	uint8x8_t __ret;
	60615	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
61089	60616	return __ret;
61090	60617	}
61091	60618	#else
61092		__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
61093		int8x8_t __ret;
61094		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60619	__ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
	60620	uint8x8_t __ret;
	60621	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61095	60622	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61096		__ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 0);
	60623	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
61097	60624	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61098	60625	return __ret;
61099	60626	}
61100	60627	#endif
61101	60628
61102	60629	#ifdef __LITTLE_ENDIAN__
61103		__ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
61104		poly8x8_t __ret;
61105		__ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 4);
	60630	__ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
	60631	uint32x2_t __ret;
	60632	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61106	60633	return __ret;
61107	60634	}
61108	60635	#else
61109		__ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
61110		poly8x8_t __ret;
61111		poly8x16x2_t __rev0;
61112		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61113		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61114		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61115		__ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 4);
61116		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60636	__ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
	60637	uint32x2_t __ret;
	60638	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60639	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60640	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60641	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61117	60642	return __ret;
61118	60643	}
61119	60644	#endif
61120	60645
61121	60646	#ifdef __LITTLE_ENDIAN__
61122		__ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
61123		poly8x16_t __ret;
61124		__ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 36);
	60647	__ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
	60648	uint16x4_t __ret;
	60649	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
61125	60650	return __ret;
61126	60651	}
61127	60652	#else
61128		__ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
61129		poly8x16_t __ret;
61130		poly8x16x2_t __rev0;
61131		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61132		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61133		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61134		__ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 36);
61135		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60653	__ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
	60654	uint16x4_t __ret;
	60655	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60656	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60657	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	60658	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61136	60659	return __ret;
61137	60660	}
61138	60661	#endif
61139	60662
61140	60663	#ifdef __LITTLE_ENDIAN__
61141		__ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
61142		uint8x16_t __ret;
61143		__ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 48);
	60664	__ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
	60665	int8x8_t __ret;
	60666	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
61144	60667	return __ret;
61145	60668	}
61146	60669	#else
61147		__ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
61148		uint8x16_t __ret;
61149		uint8x16x2_t __rev0;
61150		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61151		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61152		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61153		__ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 48);
61154		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60670	__ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
	60671	int8x8_t __ret;
	60672	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60673	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60674	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
	60675	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61155	60676	return __ret;
61156	60677	}
61157	60678	#endif
61158	60679
61159	60680	#ifdef __LITTLE_ENDIAN__
61160		__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
61161		int8x16_t __ret;
61162		__ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 32);
	60681	__ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
	60682	float32x2_t __ret;
	60683	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61163	60684	return __ret;
61164	60685	}
61165	60686	#else
61166		__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
61167		int8x16_t __ret;
61168		int8x16x2_t __rev0;
61169		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61170		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61171		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61172		__ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 32);
61173		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60687	__ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
	60688	float32x2_t __ret;
	60689	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60690	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60691	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60692	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61174	60693	return __ret;
61175	60694	}
61176	60695	#endif
61177	60696
61178	60697	#ifdef __LITTLE_ENDIAN__
61179		__ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
61180		uint8x8_t __ret;
61181		__ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 16);
	60698	__ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
	60699	int32x2_t __ret;
	60700	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61182	60701	return __ret;
61183	60702	}
61184	60703	#else
61185		__ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
61186		uint8x8_t __ret;
61187		uint8x16x2_t __rev0;
61188		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61189		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61190		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61191		__ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 16);
61192		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60704	__ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
	60705	int32x2_t __ret;
	60706	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60707	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60708	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	60709	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61193	60710	return __ret;
61194	60711	}
61195	60712	#endif
61196	60713
61197	60714	#ifdef __LITTLE_ENDIAN__
61198		__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
61199		int8x8_t __ret;
61200		__ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 0);
	60715	__ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
	60716	int16x4_t __ret;
	60717	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
61201	60718	return __ret;
61202	60719	}
61203	60720	#else
61204		__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
61205		int8x8_t __ret;
61206		int8x16x2_t __rev0;
61207		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61208		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61209		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61210		__ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 0);
61211		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60721	__ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
	60722	int16x4_t __ret;
	60723	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60724	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60725	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	60726	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61212	60727	return __ret;
61213	60728	}
61214	60729	#endif
61215	60730
61216	60731	#ifdef __LITTLE_ENDIAN__
61217		__ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
	60732	__ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
61218	60733	poly8x8_t __ret;
61219		__ret = (poly8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 4);
	60734	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
61220	60735	return __ret;
61221	60736	}
61222	60737	#else
61223		__ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
	60738	__ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
61224	60739	poly8x8_t __ret;
61225		poly8x16x3_t __rev0;
61226		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61227		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61228		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61229		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61230		__ret = (poly8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 4);
	60740	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60741	poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60742	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
61231	60743	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61232	60744	return __ret;
61233	60745	}
61234	60746	#endif
61235	60747
61236	60748	#ifdef __LITTLE_ENDIAN__
61237		__ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
61238		poly8x16_t __ret;
61239		__ret = (poly8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 36);
	60749	__ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
	60750	poly16x4_t __ret;
	60751	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
61240	60752	return __ret;
61241	60753	}
61242	60754	#else
61243		__ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
61244		poly8x16_t __ret;
61245		poly8x16x3_t __rev0;
61246		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61247		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61248		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61249		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61250		__ret = (poly8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 36);
61251		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60755	__ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
	60756	poly16x4_t __ret;
	60757	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60758	poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60759	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	60760	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61252	60761	return __ret;
61253	60762	}
61254	60763	#endif
61255	60764
61256	60765	#ifdef __LITTLE_ENDIAN__
61257		__ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
61258		uint8x16_t __ret;
61259		__ret = (uint8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 48);
	60766	__ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	60767	poly8x16_t __ret;
	60768	__ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
61260	60769	return __ret;
61261	60770	}
61262	60771	#else
61263		__ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
61264		uint8x16_t __ret;
61265		uint8x16x3_t __rev0;
61266		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61267		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61268		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61269		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61270		__ret = (uint8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 48);
	60772	__ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	60773	poly8x16_t __ret;
	60774	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60775	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60776	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
61271	60777	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61272	60778	return __ret;
61273	60779	}
61274	60780	#endif
61275	60781
61276	60782	#ifdef __LITTLE_ENDIAN__
61277		__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
61278		int8x16_t __ret;
61279		__ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 32);
	60783	__ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	60784	poly64x2_t __ret;
	60785	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
61280	60786	return __ret;
61281	60787	}
61282	60788	#else
61283		__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
61284		int8x16_t __ret;
61285		int8x16x3_t __rev0;
61286		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61287		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61288		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61289		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61290		__ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev1, 32);
61291		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60789	__ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	60790	poly64x2_t __ret;
	60791	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60792	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60793	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	60794	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61292	60795	return __ret;
61293	60796	}
61294	60797	#endif
61295	60798
61296	60799	#ifdef __LITTLE_ENDIAN__
61297		__ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
61298		uint8x8_t __ret;
61299		__ret = (uint8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 16);
	60800	__ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	60801	poly16x8_t __ret;
	60802	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
61300	60803	return __ret;
61301	60804	}
61302	60805	#else
61303		__ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
61304		uint8x8_t __ret;
61305		uint8x16x3_t __rev0;
61306		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61307		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61308		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61309		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61310		__ret = (uint8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 16);
	60806	__ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	60807	poly16x8_t __ret;
	60808	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60809	poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60810	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
61311	60811	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61312	60812	return __ret;
61313	60813	}
61314	60814	#endif
61315	60815
61316	60816	#ifdef __LITTLE_ENDIAN__
61317		__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
61318		int8x8_t __ret;
61319		__ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 0);
	60817	__ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	60818	uint8x16_t __ret;
	60819	__ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
61320	60820	return __ret;
61321	60821	}
61322	60822	#else
61323		__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
61324		int8x8_t __ret;
61325		int8x16x3_t __rev0;
61326		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61327		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61328		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61329		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61330		__ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x8_t)__rev1, 0);
61331		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60823	__ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	60824	uint8x16_t __ret;
	60825	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60826	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60827	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
	60828	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61332	60829	return __ret;
61333	60830	}
61334	60831	#endif
61335	60832
61336	60833	#ifdef __LITTLE_ENDIAN__
61337		__ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
61338		poly8x8_t __ret;
61339		__ret = (poly8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 4);
	60834	__ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	60835	uint32x4_t __ret;
	60836	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
61340	60837	return __ret;
61341	60838	}
61342	60839	#else
61343		__ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
61344		poly8x8_t __ret;
61345		poly8x16x4_t __rev0;
61346		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61347		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61348		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61349		__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61350		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61351		__ret = (poly8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 4);
61352		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60840	__ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	60841	uint32x4_t __ret;
	60842	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60843	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60844	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	60845	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61353	60846	return __ret;
61354	60847	}
61355	60848	#endif
61356	60849
61357	60850	#ifdef __LITTLE_ENDIAN__
61358		__ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
61359		poly8x16_t __ret;
61360		__ret = (poly8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 36);
	60851	__ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	60852	uint64x2_t __ret;
	60853	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
61361	60854	return __ret;
61362	60855	}
61363	60856	#else
61364		__ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
61365		poly8x16_t __ret;
61366		poly8x16x4_t __rev0;
61367		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61368		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61369		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61370		__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61371		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61372		__ret = (poly8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 36);
61373		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60857	__ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	60858	uint64x2_t __ret;
	60859	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60860	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60861	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	60862	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61374	60863	return __ret;
61375	60864	}
61376	60865	#endif
61377	60866
61378	60867	#ifdef __LITTLE_ENDIAN__
61379		__ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
61380		uint8x16_t __ret;
61381		__ret = (uint8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 48);
	60868	__ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	60869	uint16x8_t __ret;
	60870	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
61382	60871	return __ret;
61383	60872	}
61384	60873	#else
61385		__ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
61386		uint8x16_t __ret;
61387		uint8x16x4_t __rev0;
61388		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61389		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61390		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61391		__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61392		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61393		__ret = (uint8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 48);
61394		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60874	__ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	60875	uint16x8_t __ret;
	60876	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60877	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60878	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	60879	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61395	60880	return __ret;
61396	60881	}
61397	60882	#endif
61398	60883
61399	60884	#ifdef __LITTLE_ENDIAN__
61400		__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
	60885	__ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
61401	60886	int8x16_t __ret;
61402		__ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 32);
	60887	__ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
61403	60888	return __ret;
61404	60889	}
61405	60890	#else
61406		__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
	60891	__ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
61407	60892	int8x16_t __ret;
61408		int8x16x4_t __rev0;
61409		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61410		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61411		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61412		__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61413		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61414		__ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x16_t)__rev1, 32);
	60893	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60894	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60895	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
61415	60896	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61416	60897	return __ret;
61417	60898	}
61418	60899	#endif
61419	60900
61420	60901	#ifdef __LITTLE_ENDIAN__
61421		__ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
61422		uint8x8_t __ret;
61423		__ret = (uint8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 16);
	60902	__ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
	60903	float64x2_t __ret;
	60904	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
61424	60905	return __ret;
61425	60906	}
61426	60907	#else
61427		__ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
61428		uint8x8_t __ret;
61429		uint8x16x4_t __rev0;
61430		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61431		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61432		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61433		__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61434		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61435		__ret = (uint8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 16);
61436		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60908	__ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
	60909	float64x2_t __ret;
	60910	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60911	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60912	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	60913	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61437	60914	return __ret;
61438	60915	}
61439	60916	#endif
61440	60917
61441	60918	#ifdef __LITTLE_ENDIAN__
61442		__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
61443		int8x8_t __ret;
61444		__ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 0);
	60919	__ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
	60920	float32x4_t __ret;
	60921	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
61445	60922	return __ret;
61446	60923	}
61447	60924	#else
61448		__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
61449		int8x8_t __ret;
61450		int8x16x4_t __rev0;
61451		__rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61452		__rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61453		__rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61454		__rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61455		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61456		__ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev0.val[2], (int8x16_t)__rev0.val[3], (int8x8_t)__rev1, 0);
61457		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60925	__ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
	60926	float32x4_t __ret;
	60927	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60928	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60929	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	60930	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61458	60931	return __ret;
61459	60932	}
61460	60933	#endif
61461	60934
61462	60935	#ifdef __LITTLE_ENDIAN__
61463		__ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
61464		poly8x8_t __ret;
61465		__ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 4);
	60936	__ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
	60937	int32x4_t __ret;
	60938	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
61466	60939	return __ret;
61467	60940	}
61468	60941	#else
61469		__ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
61470		poly8x8_t __ret;
61471		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61472		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61473		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61474		__ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 4);
61475		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	60942	__ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
	60943	int32x4_t __ret;
	60944	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	60945	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	60946	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	60947	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	60948	return __ret;
	60949	}
	60950	#endif
	60951
	60952	#ifdef __LITTLE_ENDIAN__
	60953	__ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
	60954	int64x2_t __ret;
	60955	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	60956	return __ret;
	60957	}
	60958	#else
	60959	__ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
	60960	int64x2_t __ret;
	60961	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	60962	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	60963	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	60964	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61476	60965	return __ret;
61477	60966	}
61478	60967	#endif
61479	60968
61480	60969	#ifdef __LITTLE_ENDIAN__
61481		__ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
61482		poly8x16_t __ret;
61483		__ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 36);
	60970	__ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
	60971	int16x8_t __ret;
	60972	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
61484	60973	return __ret;
61485	60974	}
61486	60975	#else
61487		__ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
61488		poly8x16_t __ret;
61489		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61490		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61491		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61492		__ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 36);
61493		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60976	__ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
	60977	int16x8_t __ret;
	60978	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60979	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60980	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	60981	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61494	60982	return __ret;
61495	60983	}
61496	60984	#endif
61497	60985
61498	60986	#ifdef __LITTLE_ENDIAN__
61499		__ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
61500		uint8x16_t __ret;
61501		__ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
	60987	__ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
	60988	uint8x8_t __ret;
	60989	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
61502	60990	return __ret;
61503	60991	}
61504	60992	#else
61505		__ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
61506		uint8x16_t __ret;
61507		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61508		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61509		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61510		__ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
61511		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	60993	__ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
	60994	uint8x8_t __ret;
	60995	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	60996	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	60997	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	60998	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61512	60999	return __ret;
61513	61000	}
61514	61001	#endif
61515	61002
61516	61003	#ifdef __LITTLE_ENDIAN__
61517		__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
61518		int8x16_t __ret;
61519		__ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
	61004	__ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
	61005	uint32x2_t __ret;
	61006	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
61520	61007	return __ret;
61521	61008	}
61522	61009	#else
61523		__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
61524		int8x16_t __ret;
61525		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61526		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61527		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61528		__ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
61529		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61010	__ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
	61011	uint32x2_t __ret;
	61012	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61013	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61014	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	61015	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61530	61016	return __ret;
61531	61017	}
61532	61018	#endif
61533	61019
61534	61020	#ifdef __LITTLE_ENDIAN__
61535		__ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
61536		uint8x8_t __ret;
61537		__ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 16);
	61021	__ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
	61022	uint16x4_t __ret;
	61023	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
61538	61024	return __ret;
61539	61025	}
61540	61026	#else
61541		__ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
61542		uint8x8_t __ret;
61543		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61544		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61545		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61546		__ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 16);
61547		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61027	__ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
	61028	uint16x4_t __ret;
	61029	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61030	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61031	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	61032	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61548	61033	return __ret;
61549	61034	}
61550	61035	#endif
61551	61036
61552	61037	#ifdef __LITTLE_ENDIAN__
61553		__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
	61038	__ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
61554	61039	int8x8_t __ret;
61555		__ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 0);
	61040	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
61556	61041	return __ret;
61557	61042	}
61558	61043	#else
61559		__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
	61044	__ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
61560	61045	int8x8_t __ret;
61561	61046	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61562		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61563		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61564		__ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 0);
	61047	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61048	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
61565	61049	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61566	61050	return __ret;
61567	61051	}
61568	61052	#endif
61569	61053
61570	61054	#ifdef __LITTLE_ENDIAN__
61571		__ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
61572		poly8x8_t __ret;
61573		__ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 4);
	61055	__ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
	61056	float32x2_t __ret;
	61057	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
61574	61058	return __ret;
61575	61059	}
61576	61060	#else
61577		__ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
61578		poly8x8_t __ret;
61579		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61580		poly8x16x2_t __rev1;
61581		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61582		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61583		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61584		__ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 4);
61585		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61061	__ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
	61062	float32x2_t __ret;
	61063	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61064	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61065	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	61066	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61586	61067	return __ret;
61587	61068	}
61588	61069	#endif
61589	61070
61590	61071	#ifdef __LITTLE_ENDIAN__
61591		__ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
61592		poly8x16_t __ret;
61593		__ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 36);
	61072	__ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
	61073	int32x2_t __ret;
	61074	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
61594	61075	return __ret;
61595	61076	}
61596	61077	#else
61597		__ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
61598		poly8x16_t __ret;
61599		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61600		poly8x16x2_t __rev1;
61601		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61602		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61603		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61604		__ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 36);
61605		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61078	__ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
	61079	int32x2_t __ret;
	61080	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61081	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61082	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	61083	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61606	61084	return __ret;
61607	61085	}
61608	61086	#endif
61609	61087
61610	61088	#ifdef __LITTLE_ENDIAN__
61611		__ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
61612		uint8x16_t __ret;
61613		__ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 48);
	61089	__ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
	61090	int16x4_t __ret;
	61091	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
61614	61092	return __ret;
61615	61093	}
61616	61094	#else
61617		__ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
61618		uint8x16_t __ret;
61619		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61620		uint8x16x2_t __rev1;
61621		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61622		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61623		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61624		__ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 48);
61625		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61095	__ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
	61096	int16x4_t __ret;
	61097	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61098	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61099	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	61100	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61626	61101	return __ret;
61627	61102	}
61628	61103	#endif
61629	61104
	61105	__ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
	61106	uint64x1_t __ret;
	61107	__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
	61108	return __ret;
	61109	}
61630	61110	#ifdef __LITTLE_ENDIAN__
61631		__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
61632		int8x16_t __ret;
61633		__ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 32);
	61111	__ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
	61112	uint64x2_t __ret;
	61113	__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
61634	61114	return __ret;
61635	61115	}
61636	61116	#else
61637		__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
61638		int8x16_t __ret;
61639		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61640		int8x16x2_t __rev1;
61641		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61642		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61643		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61644		__ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 32);
61645		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61117	__ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
	61118	uint64x2_t __ret;
	61119	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61120	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61121	__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	61122	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61646	61123	return __ret;
61647	61124	}
61648	61125	#endif
61649	61126
61650	61127	#ifdef __LITTLE_ENDIAN__
61651		__ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
61652		uint8x8_t __ret;
61653		__ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 16);
	61128	__ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	61129	uint64x2_t __ret;
	61130	__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
61654	61131	return __ret;
61655	61132	}
61656	61133	#else
61657		__ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
61658		uint8x8_t __ret;
61659		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61660		uint8x16x2_t __rev1;
61661		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61662		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61663		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61664		__ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 16);
61665		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61134	__ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
	61135	uint64x2_t __ret;
	61136	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61137	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61138	__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	61139	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61666	61140	return __ret;
61667	61141	}
61668	61142	#endif
61669	61143
61670	61144	#ifdef __LITTLE_ENDIAN__
61671		__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
61672		int8x8_t __ret;
61673		__ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 0);
	61145	__ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
	61146	uint64x2_t __ret;
	61147	__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
61674	61148	return __ret;
61675	61149	}
61676	61150	#else
61677		__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
61678		int8x8_t __ret;
61679		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61680		int8x16x2_t __rev1;
61681		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61682		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61683		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61684		__ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 0);
61685		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61151	__ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
	61152	uint64x2_t __ret;
	61153	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61154	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61155	__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	61156	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61686	61157	return __ret;
61687	61158	}
61688	61159	#endif
61689	61160
61690		#ifdef __LITTLE_ENDIAN__
61691		__ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
61692		poly8x8_t __ret;
61693		__ret = (poly8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 4);
	61161	__ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
	61162	uint64x1_t __ret;
	61163	__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
61694	61164	return __ret;
61695	61165	}
61696		#else
61697		__ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
61698		poly8x8_t __ret;
61699		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61700		poly8x16x3_t __rev1;
61701		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61702		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61703		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61704		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61705		__ret = (poly8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 4);
61706		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61166	__ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
	61167	uint64x1_t __ret;
	61168	__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
	61169	return __ret;
	61170	}
	61171	__ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
	61172	uint64_t __ret;
	61173	__ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
	61174	return __ret;
	61175	}
	61176	__ai uint64_t vtstd_s64(int64_t __p0, int64_t __p1) {
	61177	uint64_t __ret;
	61178	__ret = (uint64_t) __builtin_neon_vtstd_s64(__p0, __p1);
	61179	return __ret;
	61180	}
	61181	__ai int8_t vuqaddb_s8(int8_t __p0, uint8_t __p1) {
	61182	int8_t __ret;
	61183	__ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
	61184	return __ret;
	61185	}
	61186	__ai int32_t vuqadds_s32(int32_t __p0, uint32_t __p1) {
	61187	int32_t __ret;
	61188	__ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
	61189	return __ret;
	61190	}
	61191	__ai int64_t vuqaddd_s64(int64_t __p0, uint64_t __p1) {
	61192	int64_t __ret;
	61193	__ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
	61194	return __ret;
	61195	}
	61196	__ai int16_t vuqaddh_s16(int16_t __p0, uint16_t __p1) {
	61197	int16_t __ret;
	61198	__ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
61707	61199	return __ret;
61708	61200	}
61709		#endif
61710
61711	61201	#ifdef __LITTLE_ENDIAN__
61712		__ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
61713		poly8x16_t __ret;
61714		__ret = (poly8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 36);
	61202	__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
	61203	int8x16_t __ret;
	61204	__ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
61715	61205	return __ret;
61716	61206	}
61717	61207	#else
61718		__ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
61719		poly8x16_t __ret;
61720		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61721		poly8x16x3_t __rev1;
61722		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61723		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61724		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61725		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61726		__ret = (poly8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 36);
	61208	__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
	61209	int8x16_t __ret;
	61210	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61211	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61212	__ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
61727	61213	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61728	61214	return __ret;
61729	61215	}
61730	61216	#endif
61731	61217
61732	61218	#ifdef __LITTLE_ENDIAN__
61733		__ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
61734		uint8x16_t __ret;
61735		__ret = (uint8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 48);
	61219	__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
	61220	int32x4_t __ret;
	61221	__ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
61736	61222	return __ret;
61737	61223	}
61738	61224	#else
61739		__ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
61740		uint8x16_t __ret;
61741		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61742		uint8x16x3_t __rev1;
61743		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61744		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61745		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61746		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61747		__ret = (uint8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 48);
61748		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61225	__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
	61226	int32x4_t __ret;
	61227	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61228	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61229	__ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
	61230	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61749	61231	return __ret;
61750	61232	}
61751	61233	#endif
61752	61234
61753	61235	#ifdef __LITTLE_ENDIAN__
61754		__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
61755		int8x16_t __ret;
61756		__ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 32);
	61236	__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
	61237	int64x2_t __ret;
	61238	__ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
61757	61239	return __ret;
61758	61240	}
61759	61241	#else
61760		__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
61761		int8x16_t __ret;
61762		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61763		int8x16x3_t __rev1;
61764		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61765		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61766		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61767		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61768		__ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev2, 32);
61769		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61242	__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
	61243	int64x2_t __ret;
	61244	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61245	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61246	__ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
	61247	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61770	61248	return __ret;
61771	61249	}
61772	61250	#endif
61773	61251
61774	61252	#ifdef __LITTLE_ENDIAN__
61775		__ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
61776		uint8x8_t __ret;
61777		__ret = (uint8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 16);
	61253	__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
	61254	int16x8_t __ret;
	61255	__ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
61778	61256	return __ret;
61779	61257	}
61780	61258	#else
61781		__ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
61782		uint8x8_t __ret;
61783		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61784		uint8x16x3_t __rev1;
61785		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61786		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61787		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61788		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61789		__ret = (uint8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 16);
	61259	__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
	61260	int16x8_t __ret;
	61261	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61262	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61263	__ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
61790	61264	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61791	61265	return __ret;
61792	61266	}
61793	61267	#endif
61794	61268
61795	61269	#ifdef __LITTLE_ENDIAN__
61796		__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
	61270	__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
61797	61271	int8x8_t __ret;
61798		__ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 0);
	61272	__ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
61799	61273	return __ret;
61800	61274	}
61801	61275	#else
61802		__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
	61276	__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
61803	61277	int8x8_t __ret;
61804	61278	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61805		int8x16x3_t __rev1;
61806		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61807		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61808		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61809		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61810		__ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x8_t)__rev2, 0);
	61279	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61280	__ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
61811	61281	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61812	61282	return __ret;
61813	61283	}
61814	61284	#endif
61815	61285
61816	61286	#ifdef __LITTLE_ENDIAN__
61817		__ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
61818		poly8x8_t __ret;
61819		__ret = (poly8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 4);
	61287	__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
	61288	int32x2_t __ret;
	61289	__ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
61820	61290	return __ret;
61821	61291	}
61822	61292	#else
61823		__ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
61824		poly8x8_t __ret;
61825		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61826		poly8x16x4_t __rev1;
61827		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61828		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61829		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61830		__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61831		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61832		__ret = (poly8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 4);
61833		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61293	__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
	61294	int32x2_t __ret;
	61295	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61296	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61297	__ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
	61298	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61834	61299	return __ret;
61835	61300	}
61836	61301	#endif
61837	61302
	61303	__ai int64x1_t vuqadd_s64(int64x1_t __p0, uint64x1_t __p1) {
	61304	int64x1_t __ret;
	61305	__ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
	61306	return __ret;
	61307	}
61838	61308	#ifdef __LITTLE_ENDIAN__
61839		__ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
61840		poly8x16_t __ret;
61841		__ret = (poly8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 36);
	61309	__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
	61310	int16x4_t __ret;
	61311	__ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
61842	61312	return __ret;
61843	61313	}
61844	61314	#else
61845		__ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
61846		poly8x16_t __ret;
61847		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61848		poly8x16x4_t __rev1;
61849		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61850		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61851		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61852		__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61853		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61854		__ret = (poly8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 36);
61855		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61315	__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
	61316	int16x4_t __ret;
	61317	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61318	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61319	__ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
	61320	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61856	61321	return __ret;
61857	61322	}
61858	61323	#endif
61859	61324
61860	61325	#ifdef __LITTLE_ENDIAN__
61861		__ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
61862		uint8x16_t __ret;
61863		__ret = (uint8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 48);
	61326	__ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
	61327	poly8x8_t __ret;
	61328	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
61864	61329	return __ret;
61865	61330	}
61866	61331	#else
61867		__ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
61868		uint8x16_t __ret;
61869		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61870		uint8x16x4_t __rev1;
61871		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61872		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61873		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61874		__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61875		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61876		__ret = (uint8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 48);
61877		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61332	__ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
	61333	poly8x8_t __ret;
	61334	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61335	poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61336	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	61337	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61878	61338	return __ret;
61879	61339	}
61880	61340	#endif
61881	61341
61882	61342	#ifdef __LITTLE_ENDIAN__
61883		__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
61884		int8x16_t __ret;
61885		__ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 32);
	61343	__ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
	61344	poly16x4_t __ret;
	61345	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
61886	61346	return __ret;
61887	61347	}
61888	61348	#else
61889		__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
61890		int8x16_t __ret;
61891		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61892		int8x16x4_t __rev1;
61893		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61894		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61895		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61896		__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61897		uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61898		__ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x16_t)__rev2, 32);
	61349	__ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
	61350	poly16x4_t __ret;
	61351	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61352	poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61353	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	61354	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	61355	return __ret;
	61356	}
	61357	#endif
	61358
	61359	#ifdef __LITTLE_ENDIAN__
	61360	__ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	61361	poly8x16_t __ret;
	61362	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	61363	return __ret;
	61364	}
	61365	#else
	61366	__ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	61367	poly8x16_t __ret;
	61368	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61369	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61370	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
61899	61371	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61900	61372	return __ret;
61901	61373	}
61902	61374	#endif
61903	61375
61904	61376	#ifdef __LITTLE_ENDIAN__
61905		__ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
61906		uint8x8_t __ret;
61907		__ret = (uint8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 16);
	61377	__ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	61378	poly64x2_t __ret;
	61379	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61908	61380	return __ret;
61909	61381	}
61910	61382	#else
61911		__ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
61912		uint8x8_t __ret;
61913		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61914		uint8x16x4_t __rev1;
61915		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61916		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61917		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61918		__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61919		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61920		__ret = (uint8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 16);
61921		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61383	__ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	61384	poly64x2_t __ret;
	61385	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61386	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61387	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	61388	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61922	61389	return __ret;
61923	61390	}
61924	61391	#endif
61925	61392
61926	61393	#ifdef __LITTLE_ENDIAN__
61927		__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
61928		int8x8_t __ret;
61929		__ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 0);
	61394	__ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	61395	poly16x8_t __ret;
	61396	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
61930	61397	return __ret;
61931	61398	}
61932	61399	#else
61933		__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
61934		int8x8_t __ret;
61935		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61936		int8x16x4_t __rev1;
61937		__rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61938		__rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61939		__rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61940		__rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61941		uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61942		__ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], (int8x8_t)__rev2, 0);
	61400	__ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	61401	poly16x8_t __ret;
	61402	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61403	poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61404	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
61943	61405	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
61944	61406	return __ret;
61945	61407	}
61946	61408	#endif
61947	61409
61948	61410	#ifdef __LITTLE_ENDIAN__
61949		__ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
61950		uint16x8_t __ret;
61951		__ret = vcombine_u16(__p0, vraddhn_u32(__p1, __p2));
	61411	__ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	61412	uint8x16_t __ret;
	61413	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
61952	61414	return __ret;
61953	61415	}
61954	61416	#else
61955		__ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
61956		uint16x8_t __ret;
61957		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
61958		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
61959		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
61960		__ret = __noswap_vcombine_u16(__rev0, __noswap_vraddhn_u32(__rev1, __rev2));
61961		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61417	__ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	61418	uint8x16_t __ret;
	61419	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61420	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61421	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	61422	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61962	61423	return __ret;
61963	61424	}
61964	61425	#endif
61965	61426
61966	61427	#ifdef __LITTLE_ENDIAN__
61967		__ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	61428	__ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
61968	61429	uint32x4_t __ret;
61969		__ret = vcombine_u32(__p0, vraddhn_u64(__p1, __p2));
	61430	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
61970	61431	return __ret;
61971	61432	}
61972	61433	#else
61973		__ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	61434	__ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
61974	61435	uint32x4_t __ret;
61975		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
61976		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
61977		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
61978		__ret = __noswap_vcombine_u32(__rev0, __noswap_vraddhn_u64(__rev1, __rev2));
	61436	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61437	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61438	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
61979	61439	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
61980	61440	return __ret;
61981	61441	}
61982	61442	#endif
61983	61443
61984	61444	#ifdef __LITTLE_ENDIAN__
61985		__ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
61986		uint8x16_t __ret;
61987		__ret = vcombine_u8(__p0, vraddhn_u16(__p1, __p2));
	61445	__ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	61446	uint64x2_t __ret;
	61447	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
61988	61448	return __ret;
61989	61449	}
61990	61450	#else
61991		__ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
61992		uint8x16_t __ret;
61993		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
61994		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61995		uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61996		__ret = __noswap_vcombine_u8(__rev0, __noswap_vraddhn_u16(__rev1, __rev2));
61997		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61451	__ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	61452	uint64x2_t __ret;
	61453	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61454	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61455	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	61456	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
61998	61457	return __ret;
61999	61458	}
62000	61459	#endif
62001	61460
62002	61461	#ifdef __LITTLE_ENDIAN__
62003		__ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
62004		int16x8_t __ret;
62005		__ret = vcombine_s16(__p0, vraddhn_s32(__p1, __p2));
	61462	__ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	61463	uint16x8_t __ret;
	61464	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
62006	61465	return __ret;
62007	61466	}
62008	61467	#else
62009		__ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
62010		int16x8_t __ret;
62011		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
62012		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
62013		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
62014		__ret = __noswap_vcombine_s16(__rev0, __noswap_vraddhn_s32(__rev1, __rev2));
	61468	__ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	61469	uint16x8_t __ret;
	61470	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61471	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61472	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
62015	61473	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62016	61474	return __ret;
62017	61475	}
62018	61476	#endif
62019	61477
62020	61478	#ifdef __LITTLE_ENDIAN__
62021		__ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
62022		int32x4_t __ret;
62023		__ret = vcombine_s32(__p0, vraddhn_s64(__p1, __p2));
	61479	__ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
	61480	int8x16_t __ret;
	61481	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
62024	61482	return __ret;
62025	61483	}
62026	61484	#else
62027		__ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
62028		int32x4_t __ret;
62029		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62030		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62031		int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
62032		__ret = __noswap_vcombine_s32(__rev0, __noswap_vraddhn_s64(__rev1, __rev2));
62033		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	61485	__ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
	61486	int8x16_t __ret;
	61487	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61488	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61489	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	61490	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62034	61491	return __ret;
62035	61492	}
62036	61493	#endif
62037	61494
62038	61495	#ifdef __LITTLE_ENDIAN__
62039		__ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
62040		int8x16_t __ret;
62041		__ret = vcombine_s8(__p0, vraddhn_s16(__p1, __p2));
	61496	__ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
	61497	float64x2_t __ret;
	61498	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62042	61499	return __ret;
62043	61500	}
62044	61501	#else
62045		__ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
62046		int8x16_t __ret;
62047		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
62048		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
62049		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
62050		__ret = __noswap_vcombine_s8(__rev0, __noswap_vraddhn_s16(__rev1, __rev2));
62051		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61502	__ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
	61503	float64x2_t __ret;
	61504	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61505	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61506	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	61507	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62052	61508	return __ret;
62053	61509	}
62054	61510	#endif
62055	61511
62056	61512	#ifdef __LITTLE_ENDIAN__
62057		__ai poly8x8_t vrbit_p8(poly8x8_t __p0) {
62058		poly8x8_t __ret;
62059		__ret = (poly8x8_t) __builtin_neon_vrbit_v((int8x8_t)__p0, 4);
	61513	__ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
	61514	float32x4_t __ret;
	61515	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
62060	61516	return __ret;
62061	61517	}
62062	61518	#else
62063		__ai poly8x8_t vrbit_p8(poly8x8_t __p0) {
62064		poly8x8_t __ret;
62065		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
62066		__ret = (poly8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 4);
62067		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61519	__ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
	61520	float32x4_t __ret;
	61521	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61522	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61523	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	61524	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62068	61525	return __ret;
62069	61526	}
62070	61527	#endif
62071	61528
62072	61529	#ifdef __LITTLE_ENDIAN__
62073		__ai poly8x16_t vrbitq_p8(poly8x16_t __p0) {
62074		poly8x16_t __ret;
62075		__ret = (poly8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__p0, 36);
	61530	__ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
	61531	int32x4_t __ret;
	61532	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
62076	61533	return __ret;
62077	61534	}
62078	61535	#else
62079		__ai poly8x16_t vrbitq_p8(poly8x16_t __p0) {
62080		poly8x16_t __ret;
62081		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62082		__ret = (poly8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 36);
62083		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61536	__ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
	61537	int32x4_t __ret;
	61538	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61539	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61540	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	61541	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62084	61542	return __ret;
62085	61543	}
62086	61544	#endif
62087	61545
62088	61546	#ifdef __LITTLE_ENDIAN__
62089		__ai uint8x16_t vrbitq_u8(uint8x16_t __p0) {
62090		uint8x16_t __ret;
62091		__ret = (uint8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__p0, 48);
	61547	__ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
	61548	int64x2_t __ret;
	61549	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62092	61550	return __ret;
62093	61551	}
62094	61552	#else
62095		__ai uint8x16_t vrbitq_u8(uint8x16_t __p0) {
62096		uint8x16_t __ret;
62097		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62098		__ret = (uint8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 48);
62099		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61553	__ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
	61554	int64x2_t __ret;
	61555	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61556	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61557	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	61558	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62100	61559	return __ret;
62101	61560	}
62102	61561	#endif
62103	61562
62104	61563	#ifdef __LITTLE_ENDIAN__
62105		__ai int8x16_t vrbitq_s8(int8x16_t __p0) {
62106		int8x16_t __ret;
62107		__ret = (int8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__p0, 32);
	61564	__ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
	61565	int16x8_t __ret;
	61566	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
62108	61567	return __ret;
62109	61568	}
62110	61569	#else
62111		__ai int8x16_t vrbitq_s8(int8x16_t __p0) {
62112		int8x16_t __ret;
62113		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62114		__ret = (int8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 32);
62115		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61570	__ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
	61571	int16x8_t __ret;
	61572	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61573	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61574	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	61575	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62116	61576	return __ret;
62117	61577	}
62118	61578	#endif
62119	61579
62120	61580	#ifdef __LITTLE_ENDIAN__
62121		__ai uint8x8_t vrbit_u8(uint8x8_t __p0) {
	61581	__ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
62122	61582	uint8x8_t __ret;
62123		__ret = (uint8x8_t) __builtin_neon_vrbit_v((int8x8_t)__p0, 16);
	61583	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
62124	61584	return __ret;
62125	61585	}
62126	61586	#else
62127		__ai uint8x8_t vrbit_u8(uint8x8_t __p0) {
	61587	__ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
62128	61588	uint8x8_t __ret;
62129	61589	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
62130		__ret = (uint8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 16);
	61590	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61591	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
62131	61592	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62132	61593	return __ret;
62133	61594	}
62134	61595	#endif
62135	61596
62136	61597	#ifdef __LITTLE_ENDIAN__
62137		__ai int8x8_t vrbit_s8(int8x8_t __p0) {
62138		int8x8_t __ret;
62139		__ret = (int8x8_t) __builtin_neon_vrbit_v((int8x8_t)__p0, 0);
	61598	__ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
	61599	uint32x2_t __ret;
	61600	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62140	61601	return __ret;
62141	61602	}
62142	61603	#else
62143		__ai int8x8_t vrbit_s8(int8x8_t __p0) {
62144		int8x8_t __ret;
62145		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
62146		__ret = (int8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 0);
62147		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61604	__ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
	61605	uint32x2_t __ret;
	61606	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61607	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61608	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	61609	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62148	61610	return __ret;
62149	61611	}
62150	61612	#endif
62151	61613
62152	61614	#ifdef __LITTLE_ENDIAN__
62153		__ai float64x2_t vrecpeq_f64(float64x2_t __p0) {
62154		float64x2_t __ret;
62155		__ret = (float64x2_t) __builtin_neon_vrecpeq_v((int8x16_t)__p0, 42);
	61615	__ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
	61616	uint16x4_t __ret;
	61617	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
62156	61618	return __ret;
62157	61619	}
62158	61620	#else
62159		__ai float64x2_t vrecpeq_f64(float64x2_t __p0) {
62160		float64x2_t __ret;
62161		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62162		__ret = (float64x2_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 42);
62163		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	61621	__ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
	61622	uint16x4_t __ret;
	61623	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61624	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61625	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	61626	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62164	61627	return __ret;
62165	61628	}
62166	61629	#endif
62167	61630
62168		__ai float64x1_t vrecpe_f64(float64x1_t __p0) {
62169		float64x1_t __ret;
62170		__ret = (float64x1_t) __builtin_neon_vrecpe_v((int8x8_t)__p0, 10);
62171		return __ret;
62172		}
62173		__ai float64_t vrecped_f64(float64_t __p0) {
62174		float64_t __ret;
62175		__ret = (float64_t) __builtin_neon_vrecped_f64(__p0);
	61631	#ifdef __LITTLE_ENDIAN__
	61632	__ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
	61633	int8x8_t __ret;
	61634	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
62176	61635	return __ret;
62177	61636	}
62178		__ai float32_t vrecpes_f32(float32_t __p0) {
62179		float32_t __ret;
62180		__ret = (float32_t) __builtin_neon_vrecpes_f32(__p0);
	61637	#else
	61638	__ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
	61639	int8x8_t __ret;
	61640	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61641	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61642	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	61643	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62181	61644	return __ret;
62182	61645	}
	61646	#endif
	61647
62183	61648	#ifdef __LITTLE_ENDIAN__
62184		__ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
62185		float64x2_t __ret;
62186		__ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	61649	__ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
	61650	float32x2_t __ret;
	61651	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62187	61652	return __ret;
62188	61653	}
62189	61654	#else
62190		__ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
62191		float64x2_t __ret;
62192		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62193		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62194		__ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	61655	__ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
	61656	float32x2_t __ret;
	61657	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61658	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61659	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
62195	61660	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62196	61661	return __ret;
62197	61662	}
62198	61663	#endif
62199	61664
62200		__ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
62201		float64x1_t __ret;
62202		__ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	61665	#ifdef __LITTLE_ENDIAN__
	61666	__ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
	61667	int32x2_t __ret;
	61668	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62203	61669	return __ret;
62204	61670	}
62205		__ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
62206		float64_t __ret;
62207		__ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
	61671	#else
	61672	__ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
	61673	int32x2_t __ret;
	61674	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61675	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61676	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	61677	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62208	61678	return __ret;
62209	61679	}
62210		__ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
62211		float32_t __ret;
62212		__ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
	61680	#endif
	61681
	61682	#ifdef __LITTLE_ENDIAN__
	61683	__ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
	61684	int16x4_t __ret;
	61685	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
62213	61686	return __ret;
62214	61687	}
62215		__ai float64_t vrecpxd_f64(float64_t __p0) {
62216		float64_t __ret;
62217		__ret = (float64_t) __builtin_neon_vrecpxd_f64(__p0);
	61688	#else
	61689	__ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
	61690	int16x4_t __ret;
	61691	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61692	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61693	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	61694	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62218	61695	return __ret;
62219	61696	}
62220		__ai float32_t vrecpxs_f32(float32_t __p0) {
62221		float32_t __ret;
62222		__ret = (float32_t) __builtin_neon_vrecpxs_f32(__p0);
	61697	#endif
	61698
	61699	#ifdef __LITTLE_ENDIAN__
	61700	__ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
	61701	poly8x8_t __ret;
	61702	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
62223	61703	return __ret;
62224	61704	}
62225		__ai uint64_t vrshld_u64(uint64_t __p0, int64_t __p1) {
62226		uint64_t __ret;
62227		__ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
	61705	#else
	61706	__ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
	61707	poly8x8_t __ret;
	61708	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61709	poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61710	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
	61711	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62228	61712	return __ret;
62229	61713	}
62230		__ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
62231		int64_t __ret;
62232		__ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
	61714	#endif
	61715
	61716	#ifdef __LITTLE_ENDIAN__
	61717	__ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
	61718	poly16x4_t __ret;
	61719	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
62233	61720	return __ret;
62234	61721	}
62235		#define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
62236		uint64_t __ret; \
62237		uint64_t __s0 = __p0; \
62238		__ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
62239		__ret; \
62240		})
62241		#define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
62242		int64_t __ret; \
62243		int64_t __s0 = __p0; \
62244		__ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
62245		__ret; \
62246		})
62247		#ifdef __LITTLE_ENDIAN__
62248		#define vrshrn_high_n_u32(__p0_807, __p1_807, __p2_807) __extension__ ({ \
62249		uint16x8_t __ret_807; \
62250		uint16x4_t __s0_807 = __p0_807; \
62251		uint32x4_t __s1_807 = __p1_807; \
62252		__ret_807 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_807), (uint16x4_t)(vrshrn_n_u32(__s1_807, __p2_807)))); \
62253		__ret_807; \
62254		})
62255	61722	#else
62256		#define vrshrn_high_n_u32(__p0_808, __p1_808, __p2_808) __extension__ ({ \
62257		uint16x8_t __ret_808; \
62258		uint16x4_t __s0_808 = __p0_808; \
62259		uint32x4_t __s1_808 = __p1_808; \
62260		uint16x4_t __rev0_808; __rev0_808 = __builtin_shufflevector(__s0_808, __s0_808, 3, 2, 1, 0); \
62261		uint32x4_t __rev1_808; __rev1_808 = __builtin_shufflevector(__s1_808, __s1_808, 3, 2, 1, 0); \
62262		__ret_808 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_808), (uint16x4_t)(__noswap_vrshrn_n_u32(__rev1_808, __p2_808)))); \
62263		__ret_808 = __builtin_shufflevector(__ret_808, __ret_808, 7, 6, 5, 4, 3, 2, 1, 0); \
62264		__ret_808; \
62265		})
	61723	__ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
	61724	poly16x4_t __ret;
	61725	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61726	poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61727	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	61728	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	61729	return __ret;
	61730	}
62266	61731	#endif
62267	61732
62268	61733	#ifdef __LITTLE_ENDIAN__
62269		#define vrshrn_high_n_u64(__p0_809, __p1_809, __p2_809) __extension__ ({ \
62270		uint32x4_t __ret_809; \
62271		uint32x2_t __s0_809 = __p0_809; \
62272		uint64x2_t __s1_809 = __p1_809; \
62273		__ret_809 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_809), (uint32x2_t)(vrshrn_n_u64(__s1_809, __p2_809)))); \
62274		__ret_809; \
62275		})
	61734	__ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	61735	poly8x16_t __ret;
	61736	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	61737	return __ret;
	61738	}
62276	61739	#else
62277		#define vrshrn_high_n_u64(__p0_810, __p1_810, __p2_810) __extension__ ({ \
62278		uint32x4_t __ret_810; \
62279		uint32x2_t __s0_810 = __p0_810; \
62280		uint64x2_t __s1_810 = __p1_810; \
62281		uint32x2_t __rev0_810; __rev0_810 = __builtin_shufflevector(__s0_810, __s0_810, 1, 0); \
62282		uint64x2_t __rev1_810; __rev1_810 = __builtin_shufflevector(__s1_810, __s1_810, 1, 0); \
62283		__ret_810 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_810), (uint32x2_t)(__noswap_vrshrn_n_u64(__rev1_810, __p2_810)))); \
62284		__ret_810 = __builtin_shufflevector(__ret_810, __ret_810, 3, 2, 1, 0); \
62285		__ret_810; \
62286		})
	61740	__ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	61741	poly8x16_t __ret;
	61742	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61743	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61744	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	61745	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61746	return __ret;
	61747	}
62287	61748	#endif
62288	61749
62289	61750	#ifdef __LITTLE_ENDIAN__
62290		#define vrshrn_high_n_u16(__p0_811, __p1_811, __p2_811) __extension__ ({ \
62291		uint8x16_t __ret_811; \
62292		uint8x8_t __s0_811 = __p0_811; \
62293		uint16x8_t __s1_811 = __p1_811; \
62294		__ret_811 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_811), (uint8x8_t)(vrshrn_n_u16(__s1_811, __p2_811)))); \
62295		__ret_811; \
62296		})
	61751	__ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	61752	poly64x2_t __ret;
	61753	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	61754	return __ret;
	61755	}
62297	61756	#else
62298		#define vrshrn_high_n_u16(__p0_812, __p1_812, __p2_812) __extension__ ({ \
62299		uint8x16_t __ret_812; \
62300		uint8x8_t __s0_812 = __p0_812; \
62301		uint16x8_t __s1_812 = __p1_812; \
62302		uint8x8_t __rev0_812; __rev0_812 = __builtin_shufflevector(__s0_812, __s0_812, 7, 6, 5, 4, 3, 2, 1, 0); \
62303		uint16x8_t __rev1_812; __rev1_812 = __builtin_shufflevector(__s1_812, __s1_812, 7, 6, 5, 4, 3, 2, 1, 0); \
62304		__ret_812 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_812), (uint8x8_t)(__noswap_vrshrn_n_u16(__rev1_812, __p2_812)))); \
62305		__ret_812 = __builtin_shufflevector(__ret_812, __ret_812, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
62306		__ret_812; \
62307		})
	61757	__ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	61758	poly64x2_t __ret;
	61759	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61760	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61761	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	61762	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	61763	return __ret;
	61764	}
62308	61765	#endif
62309	61766
62310	61767	#ifdef __LITTLE_ENDIAN__
62311		#define vrshrn_high_n_s32(__p0_813, __p1_813, __p2_813) __extension__ ({ \
62312		int16x8_t __ret_813; \
62313		int16x4_t __s0_813 = __p0_813; \
62314		int32x4_t __s1_813 = __p1_813; \
62315		__ret_813 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_813), (int16x4_t)(vrshrn_n_s32(__s1_813, __p2_813)))); \
62316		__ret_813; \
62317		})
	61768	__ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	61769	poly16x8_t __ret;
	61770	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
	61771	return __ret;
	61772	}
62318	61773	#else
62319		#define vrshrn_high_n_s32(__p0_814, __p1_814, __p2_814) __extension__ ({ \
62320		int16x8_t __ret_814; \
62321		int16x4_t __s0_814 = __p0_814; \
62322		int32x4_t __s1_814 = __p1_814; \
62323		int16x4_t __rev0_814; __rev0_814 = __builtin_shufflevector(__s0_814, __s0_814, 3, 2, 1, 0); \
62324		int32x4_t __rev1_814; __rev1_814 = __builtin_shufflevector(__s1_814, __s1_814, 3, 2, 1, 0); \
62325		__ret_814 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_814), (int16x4_t)(__noswap_vrshrn_n_s32(__rev1_814, __p2_814)))); \
62326		__ret_814 = __builtin_shufflevector(__ret_814, __ret_814, 7, 6, 5, 4, 3, 2, 1, 0); \
62327		__ret_814; \
62328		})
	61774	__ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	61775	poly16x8_t __ret;
	61776	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61777	poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61778	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
	61779	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61780	return __ret;
	61781	}
62329	61782	#endif
62330	61783
62331	61784	#ifdef __LITTLE_ENDIAN__
62332		#define vrshrn_high_n_s64(__p0_815, __p1_815, __p2_815) __extension__ ({ \
62333		int32x4_t __ret_815; \
62334		int32x2_t __s0_815 = __p0_815; \
62335		int64x2_t __s1_815 = __p1_815; \
62336		__ret_815 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_815), (int32x2_t)(vrshrn_n_s64(__s1_815, __p2_815)))); \
62337		__ret_815; \
62338		})
	61785	__ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	61786	uint8x16_t __ret;
	61787	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	61788	return __ret;
	61789	}
62339	61790	#else
62340		#define vrshrn_high_n_s64(__p0_816, __p1_816, __p2_816) __extension__ ({ \
62341		int32x4_t __ret_816; \
62342		int32x2_t __s0_816 = __p0_816; \
62343		int64x2_t __s1_816 = __p1_816; \
62344		int32x2_t __rev0_816; __rev0_816 = __builtin_shufflevector(__s0_816, __s0_816, 1, 0); \
62345		int64x2_t __rev1_816; __rev1_816 = __builtin_shufflevector(__s1_816, __s1_816, 1, 0); \
62346		__ret_816 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_816), (int32x2_t)(__noswap_vrshrn_n_s64(__rev1_816, __p2_816)))); \
62347		__ret_816 = __builtin_shufflevector(__ret_816, __ret_816, 3, 2, 1, 0); \
62348		__ret_816; \
62349		})
	61791	__ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	61792	uint8x16_t __ret;
	61793	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61794	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61795	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	61796	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61797	return __ret;
	61798	}
62350	61799	#endif
62351	61800
62352	61801	#ifdef __LITTLE_ENDIAN__
62353		#define vrshrn_high_n_s16(__p0_817, __p1_817, __p2_817) __extension__ ({ \
62354		int8x16_t __ret_817; \
62355		int8x8_t __s0_817 = __p0_817; \
62356		int16x8_t __s1_817 = __p1_817; \
62357		__ret_817 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_817), (int8x8_t)(vrshrn_n_s16(__s1_817, __p2_817)))); \
62358		__ret_817; \
62359		})
	61802	__ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	61803	uint32x4_t __ret;
	61804	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	61805	return __ret;
	61806	}
62360	61807	#else
62361		#define vrshrn_high_n_s16(__p0_818, __p1_818, __p2_818) __extension__ ({ \
62362		int8x16_t __ret_818; \
62363		int8x8_t __s0_818 = __p0_818; \
62364		int16x8_t __s1_818 = __p1_818; \
62365		int8x8_t __rev0_818; __rev0_818 = __builtin_shufflevector(__s0_818, __s0_818, 7, 6, 5, 4, 3, 2, 1, 0); \
62366		int16x8_t __rev1_818; __rev1_818 = __builtin_shufflevector(__s1_818, __s1_818, 7, 6, 5, 4, 3, 2, 1, 0); \
62367		__ret_818 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_818), (int8x8_t)(__noswap_vrshrn_n_s16(__rev1_818, __p2_818)))); \
62368		__ret_818 = __builtin_shufflevector(__ret_818, __ret_818, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
62369		__ret_818; \
62370		})
	61808	__ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	61809	uint32x4_t __ret;
	61810	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61811	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61812	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	61813	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	61814	return __ret;
	61815	}
62371	61816	#endif
62372	61817
62373	61818	#ifdef __LITTLE_ENDIAN__
62374		__ai float64x2_t vrsqrteq_f64(float64x2_t __p0) {
62375		float64x2_t __ret;
62376		__ret = (float64x2_t) __builtin_neon_vrsqrteq_v((int8x16_t)__p0, 42);
	61819	__ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	61820	uint64x2_t __ret;
	61821	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
62377	61822	return __ret;
62378	61823	}
62379	61824	#else
62380		__ai float64x2_t vrsqrteq_f64(float64x2_t __p0) {
62381		float64x2_t __ret;
62382		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62383		__ret = (float64x2_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 42);
	61825	__ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	61826	uint64x2_t __ret;
	61827	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61828	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61829	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
62384	61830	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62385	61831	return __ret;
62386	61832	}
62387	61833	#endif
62388	61834
62389		__ai float64x1_t vrsqrte_f64(float64x1_t __p0) {
62390		float64x1_t __ret;
62391		__ret = (float64x1_t) __builtin_neon_vrsqrte_v((int8x8_t)__p0, 10);
	61835	#ifdef __LITTLE_ENDIAN__
	61836	__ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	61837	uint16x8_t __ret;
	61838	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
62392	61839	return __ret;
62393	61840	}
62394		__ai float64_t vrsqrted_f64(float64_t __p0) {
62395		float64_t __ret;
62396		__ret = (float64_t) __builtin_neon_vrsqrted_f64(__p0);
	61841	#else
	61842	__ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	61843	uint16x8_t __ret;
	61844	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61845	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61846	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
	61847	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62397	61848	return __ret;
62398	61849	}
62399		__ai float32_t vrsqrtes_f32(float32_t __p0) {
62400		float32_t __ret;
62401		__ret = (float32_t) __builtin_neon_vrsqrtes_f32(__p0);
	61850	#endif
	61851
	61852	#ifdef __LITTLE_ENDIAN__
	61853	__ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
	61854	int8x16_t __ret;
	61855	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	61856	return __ret;
	61857	}
	61858	#else
	61859	__ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
	61860	int8x16_t __ret;
	61861	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61862	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61863	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	61864	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62402	61865	return __ret;
62403	61866	}
	61867	#endif
	61868
62404	61869	#ifdef __LITTLE_ENDIAN__
62405		__ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
	61870	__ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
62406	61871	float64x2_t __ret;
62407		__ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
	61872	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
62408	61873	return __ret;
62409	61874	}
62410	61875	#else
62411		__ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
	61876	__ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
62412	61877	float64x2_t __ret;
62413	61878	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62414	61879	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62415		__ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	61880	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
62416	61881	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62417	61882	return __ret;
62418	61883	}
62419	61884	#endif
62420	61885
62421		__ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
62422		float64x1_t __ret;
62423		__ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
62424		return __ret;
62425		}
62426		__ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
62427		float64_t __ret;
62428		__ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
62429		return __ret;
62430		}
62431		__ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
62432		float32_t __ret;
62433		__ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
62434		return __ret;
62435		}
62436		#define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
62437		uint64_t __ret; \
62438		uint64_t __s0 = __p0; \
62439		uint64_t __s1 = __p1; \
62440		__ret = (uint64_t) __builtin_neon_vrsrad_n_u64(__s0, __s1, __p2); \
62441		__ret; \
62442		})
62443		#define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
62444		int64_t __ret; \
62445		int64_t __s0 = __p0; \
62446		int64_t __s1 = __p1; \
62447		__ret = (int64_t) __builtin_neon_vrsrad_n_s64(__s0, __s1, __p2); \
62448		__ret; \
62449		})
62450	61886	#ifdef __LITTLE_ENDIAN__
62451		__ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
62452		uint16x8_t __ret;
62453		__ret = vcombine_u16(__p0, vrsubhn_u32(__p1, __p2));
	61887	__ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
	61888	float32x4_t __ret;
	61889	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
62454	61890	return __ret;
62455	61891	}
62456	61892	#else
62457		__ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
62458		uint16x8_t __ret;
62459		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
62460		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
62461		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
62462		__ret = __noswap_vcombine_u16(__rev0, __noswap_vrsubhn_u32(__rev1, __rev2));
62463		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	61893	__ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
	61894	float32x4_t __ret;
	61895	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61896	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61897	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	61898	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62464	61899	return __ret;
62465	61900	}
62466	61901	#endif
62467	61902
62468	61903	#ifdef __LITTLE_ENDIAN__
62469		__ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
62470		uint32x4_t __ret;
62471		__ret = vcombine_u32(__p0, vrsubhn_u64(__p1, __p2));
	61904	__ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
	61905	int32x4_t __ret;
	61906	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
62472	61907	return __ret;
62473	61908	}
62474	61909	#else
62475		__ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
62476		uint32x4_t __ret;
62477		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62478		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62479		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
62480		__ret = __noswap_vcombine_u32(__rev0, __noswap_vrsubhn_u64(__rev1, __rev2));
	61910	__ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
	61911	int32x4_t __ret;
	61912	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61913	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61914	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
62481	61915	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62482	61916	return __ret;
62483	61917	}
62484	61918	#endif
62485	61919
62486	61920	#ifdef __LITTLE_ENDIAN__
62487		__ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
62488		uint8x16_t __ret;
62489		__ret = vcombine_u8(__p0, vrsubhn_u16(__p1, __p2));
	61921	__ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
	61922	int64x2_t __ret;
	61923	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
62490	61924	return __ret;
62491	61925	}
62492	61926	#else
62493		__ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
62494		uint8x16_t __ret;
62495		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
62496		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
62497		uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
62498		__ret = __noswap_vcombine_u8(__rev0, __noswap_vrsubhn_u16(__rev1, __rev2));
62499		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61927	__ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
	61928	int64x2_t __ret;
	61929	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61930	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61931	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	61932	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62500	61933	return __ret;
62501	61934	}
62502	61935	#endif
62503	61936
62504	61937	#ifdef __LITTLE_ENDIAN__
62505		__ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	61938	__ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
62506	61939	int16x8_t __ret;
62507		__ret = vcombine_s16(__p0, vrsubhn_s32(__p1, __p2));
	61940	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
62508	61941	return __ret;
62509	61942	}
62510	61943	#else
62511		__ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	61944	__ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
62512	61945	int16x8_t __ret;
62513		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
62514		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
62515		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
62516		__ret = __noswap_vcombine_s16(__rev0, __noswap_vrsubhn_s32(__rev1, __rev2));
	61946	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61947	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61948	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
62517	61949	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62518	61950	return __ret;
62519	61951	}
62520	61952	#endif
62521	61953
62522	61954	#ifdef __LITTLE_ENDIAN__
62523		__ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
62524		int32x4_t __ret;
62525		__ret = vcombine_s32(__p0, vrsubhn_s64(__p1, __p2));
	61955	__ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
	61956	uint8x8_t __ret;
	61957	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
62526	61958	return __ret;
62527	61959	}
62528	61960	#else
62529		__ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
62530		int32x4_t __ret;
62531		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
62532		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62533		int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
62534		__ret = __noswap_vcombine_s32(__rev0, __noswap_vrsubhn_s64(__rev1, __rev2));
62535		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	61961	__ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
	61962	uint8x8_t __ret;
	61963	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	61964	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	61965	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
	61966	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62536	61967	return __ret;
62537	61968	}
62538	61969	#endif
62539	61970
62540	61971	#ifdef __LITTLE_ENDIAN__
62541		__ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
62542		int8x16_t __ret;
62543		__ret = vcombine_s8(__p0, vrsubhn_s16(__p1, __p2));
	61972	__ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
	61973	uint32x2_t __ret;
	61974	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
62544	61975	return __ret;
62545	61976	}
62546	61977	#else
62547		__ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
62548		int8x16_t __ret;
62549		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
62550		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
62551		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
62552		__ret = __noswap_vcombine_s8(__rev0, __noswap_vrsubhn_s16(__rev1, __rev2));
62553		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	61978	__ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
	61979	uint32x2_t __ret;
	61980	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	61981	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	61982	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	61983	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62554	61984	return __ret;
62555	61985	}
62556	61986	#endif
62557	61987
62558		#define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62559		poly64x1_t __ret; \
62560		poly64_t __s0 = __p0; \
62561		poly64x1_t __s1 = __p1; \
62562		__ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (poly64x1_t)__s1, __p2); \
62563		__ret; \
62564		})
62565	61988	#ifdef __LITTLE_ENDIAN__
62566		#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62567		poly64x2_t __ret; \
62568		poly64_t __s0 = __p0; \
62569		poly64x2_t __s1 = __p1; \
62570		__ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
62571		__ret; \
62572		})
	61989	__ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
	61990	uint16x4_t __ret;
	61991	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	61992	return __ret;
	61993	}
62573	61994	#else
62574		#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62575		poly64x2_t __ret; \
62576		poly64_t __s0 = __p0; \
62577		poly64x2_t __s1 = __p1; \
62578		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
62579		__ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__rev1, __p2); \
62580		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
62581		__ret; \
62582		})
62583		#define __noswap_vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62584		poly64x2_t __ret; \
62585		poly64_t __s0 = __p0; \
62586		poly64x2_t __s1 = __p1; \
62587		__ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
62588		__ret; \
62589		})
	61995	__ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
	61996	uint16x4_t __ret;
	61997	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	61998	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	61999	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	62000	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62001	return __ret;
	62002	}
62590	62003	#endif
62591	62004
62592	62005	#ifdef __LITTLE_ENDIAN__
62593		#define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62594		float64x2_t __ret; \
62595		float64_t __s0 = __p0; \
62596		float64x2_t __s1 = __p1; \
62597		__ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
62598		__ret; \
62599		})
62600		#else
62601		#define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62602		float64x2_t __ret; \
62603		float64_t __s0 = __p0; \
62604		float64x2_t __s1 = __p1; \
62605		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
62606		__ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__rev1, __p2); \
62607		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
62608		__ret; \
62609		})
62610		#define __noswap_vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62611		float64x2_t __ret; \
62612		float64_t __s0 = __p0; \
62613		float64x2_t __s1 = __p1; \
62614		__ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
62615		__ret; \
62616		})
62617		#endif
62618
62619		#define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62620		float64x1_t __ret; \
62621		float64_t __s0 = __p0; \
62622		float64x1_t __s1 = __p1; \
62623		__ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (float64x1_t)__s1, __p2); \
62624		__ret; \
62625		})
62626		__ai uint64_t vshld_u64(uint64_t __p0, int64_t __p1) {
62627		uint64_t __ret;
62628		__ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
	62006	__ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
	62007	int8x8_t __ret;
	62008	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
62629	62009	return __ret;
62630	62010	}
62631		__ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
62632		int64_t __ret;
62633		__ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
	62011	#else
	62012	__ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
	62013	int8x8_t __ret;
	62014	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62015	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62016	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
	62017	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
62634	62018	return __ret;
62635	62019	}
62636		#define vshld_n_u64(__p0, __p1) __extension__ ({ \
62637		uint64_t __ret; \
62638		uint64_t __s0 = __p0; \
62639		__ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
62640		__ret; \
62641		})
62642		#define vshld_n_s64(__p0, __p1) __extension__ ({ \
62643		int64_t __ret; \
62644		int64_t __s0 = __p0; \
62645		__ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
62646		__ret; \
62647		})
62648		#ifdef __LITTLE_ENDIAN__
62649		#define vshll_high_n_u8(__p0_819, __p1_819) __extension__ ({ \
62650		uint16x8_t __ret_819; \
62651		uint8x16_t __s0_819 = __p0_819; \
62652		__ret_819 = (uint16x8_t)(vshll_n_u8(vget_high_u8(__s0_819), __p1_819)); \
62653		__ret_819; \
62654		})
62655		#else
62656		#define vshll_high_n_u8(__p0_820, __p1_820) __extension__ ({ \
62657		uint16x8_t __ret_820; \
62658		uint8x16_t __s0_820 = __p0_820; \
62659		uint8x16_t __rev0_820; __rev0_820 = __builtin_shufflevector(__s0_820, __s0_820, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
62660		__ret_820 = (uint16x8_t)(__noswap_vshll_n_u8(__noswap_vget_high_u8(__rev0_820), __p1_820)); \
62661		__ret_820 = __builtin_shufflevector(__ret_820, __ret_820, 7, 6, 5, 4, 3, 2, 1, 0); \
62662		__ret_820; \
62663		})
62664	62020	#endif
62665	62021
62666	62022	#ifdef __LITTLE_ENDIAN__
62667		#define vshll_high_n_u32(__p0_821, __p1_821) __extension__ ({ \
62668		uint64x2_t __ret_821; \
62669		uint32x4_t __s0_821 = __p0_821; \
62670		__ret_821 = (uint64x2_t)(vshll_n_u32(vget_high_u32(__s0_821), __p1_821)); \
62671		__ret_821; \
62672		})
	62023	__ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
	62024	float32x2_t __ret;
	62025	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62026	return __ret;
	62027	}
62673	62028	#else
62674		#define vshll_high_n_u32(__p0_822, __p1_822) __extension__ ({ \
62675		uint64x2_t __ret_822; \
62676		uint32x4_t __s0_822 = __p0_822; \
62677		uint32x4_t __rev0_822; __rev0_822 = __builtin_shufflevector(__s0_822, __s0_822, 3, 2, 1, 0); \
62678		__ret_822 = (uint64x2_t)(__noswap_vshll_n_u32(__noswap_vget_high_u32(__rev0_822), __p1_822)); \
62679		__ret_822 = __builtin_shufflevector(__ret_822, __ret_822, 1, 0); \
62680		__ret_822; \
62681		})
	62029	__ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
	62030	float32x2_t __ret;
	62031	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62032	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62033	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62034	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62035	return __ret;
	62036	}
62682	62037	#endif
62683	62038
62684	62039	#ifdef __LITTLE_ENDIAN__
62685		#define vshll_high_n_u16(__p0_823, __p1_823) __extension__ ({ \
62686		uint32x4_t __ret_823; \
62687		uint16x8_t __s0_823 = __p0_823; \
62688		__ret_823 = (uint32x4_t)(vshll_n_u16(vget_high_u16(__s0_823), __p1_823)); \
62689		__ret_823; \
62690		})
	62040	__ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
	62041	int32x2_t __ret;
	62042	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62043	return __ret;
	62044	}
62691	62045	#else
62692		#define vshll_high_n_u16(__p0_824, __p1_824) __extension__ ({ \
62693		uint32x4_t __ret_824; \
62694		uint16x8_t __s0_824 = __p0_824; \
62695		uint16x8_t __rev0_824; __rev0_824 = __builtin_shufflevector(__s0_824, __s0_824, 7, 6, 5, 4, 3, 2, 1, 0); \
62696		__ret_824 = (uint32x4_t)(__noswap_vshll_n_u16(__noswap_vget_high_u16(__rev0_824), __p1_824)); \
62697		__ret_824 = __builtin_shufflevector(__ret_824, __ret_824, 3, 2, 1, 0); \
62698		__ret_824; \
62699		})
	62046	__ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
	62047	int32x2_t __ret;
	62048	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62049	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62050	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62051	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62052	return __ret;
	62053	}
62700	62054	#endif
62701	62055
62702	62056	#ifdef __LITTLE_ENDIAN__
62703		#define vshll_high_n_s8(__p0_825, __p1_825) __extension__ ({ \
62704		int16x8_t __ret_825; \
62705		int8x16_t __s0_825 = __p0_825; \
62706		__ret_825 = (int16x8_t)(vshll_n_s8(vget_high_s8(__s0_825), __p1_825)); \
62707		__ret_825; \
62708		})
	62057	__ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
	62058	int16x4_t __ret;
	62059	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	62060	return __ret;
	62061	}
62709	62062	#else
62710		#define vshll_high_n_s8(__p0_826, __p1_826) __extension__ ({ \
62711		int16x8_t __ret_826; \
62712		int8x16_t __s0_826 = __p0_826; \
62713		int8x16_t __rev0_826; __rev0_826 = __builtin_shufflevector(__s0_826, __s0_826, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
62714		__ret_826 = (int16x8_t)(__noswap_vshll_n_s8(__noswap_vget_high_s8(__rev0_826), __p1_826)); \
62715		__ret_826 = __builtin_shufflevector(__ret_826, __ret_826, 7, 6, 5, 4, 3, 2, 1, 0); \
62716		__ret_826; \
62717		})
	62063	__ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
	62064	int16x4_t __ret;
	62065	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62066	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62067	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	62068	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62069	return __ret;
	62070	}
62718	62071	#endif
62719	62072
62720	62073	#ifdef __LITTLE_ENDIAN__
62721		#define vshll_high_n_s32(__p0_827, __p1_827) __extension__ ({ \
62722		int64x2_t __ret_827; \
62723		int32x4_t __s0_827 = __p0_827; \
62724		__ret_827 = (int64x2_t)(vshll_n_s32(vget_high_s32(__s0_827), __p1_827)); \
62725		__ret_827; \
62726		})
	62074	__ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
	62075	poly8x8_t __ret;
	62076	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
	62077	return __ret;
	62078	}
62727	62079	#else
62728		#define vshll_high_n_s32(__p0_828, __p1_828) __extension__ ({ \
62729		int64x2_t __ret_828; \
62730		int32x4_t __s0_828 = __p0_828; \
62731		int32x4_t __rev0_828; __rev0_828 = __builtin_shufflevector(__s0_828, __s0_828, 3, 2, 1, 0); \
62732		__ret_828 = (int64x2_t)(__noswap_vshll_n_s32(__noswap_vget_high_s32(__rev0_828), __p1_828)); \
62733		__ret_828 = __builtin_shufflevector(__ret_828, __ret_828, 1, 0); \
62734		__ret_828; \
62735		})
	62080	__ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
	62081	poly8x8_t __ret;
	62082	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62083	poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62084	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
	62085	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62086	return __ret;
	62087	}
62736	62088	#endif
62737	62089
62738	62090	#ifdef __LITTLE_ENDIAN__
62739		#define vshll_high_n_s16(__p0_829, __p1_829) __extension__ ({ \
62740		int32x4_t __ret_829; \
62741		int16x8_t __s0_829 = __p0_829; \
62742		__ret_829 = (int32x4_t)(vshll_n_s16(vget_high_s16(__s0_829), __p1_829)); \
62743		__ret_829; \
62744		})
	62091	__ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
	62092	poly16x4_t __ret;
	62093	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
	62094	return __ret;
	62095	}
62745	62096	#else
62746		#define vshll_high_n_s16(__p0_830, __p1_830) __extension__ ({ \
62747		int32x4_t __ret_830; \
62748		int16x8_t __s0_830 = __p0_830; \
62749		int16x8_t __rev0_830; __rev0_830 = __builtin_shufflevector(__s0_830, __s0_830, 7, 6, 5, 4, 3, 2, 1, 0); \
62750		__ret_830 = (int32x4_t)(__noswap_vshll_n_s16(__noswap_vget_high_s16(__rev0_830), __p1_830)); \
62751		__ret_830 = __builtin_shufflevector(__ret_830, __ret_830, 3, 2, 1, 0); \
62752		__ret_830; \
62753		})
	62097	__ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
	62098	poly16x4_t __ret;
	62099	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62100	poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62101	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
	62102	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62103	return __ret;
	62104	}
62754	62105	#endif
62755	62106
62756		#define vshrd_n_u64(__p0, __p1) __extension__ ({ \
62757		uint64_t __ret; \
62758		uint64_t __s0 = __p0; \
62759		__ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
62760		__ret; \
62761		})
62762		#define vshrd_n_s64(__p0, __p1) __extension__ ({ \
62763		int64_t __ret; \
62764		int64_t __s0 = __p0; \
62765		__ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
62766		__ret; \
62767		})
62768	62107	#ifdef __LITTLE_ENDIAN__
62769		#define vshrn_high_n_u32(__p0_831, __p1_831, __p2_831) __extension__ ({ \
62770		uint16x8_t __ret_831; \
62771		uint16x4_t __s0_831 = __p0_831; \
62772		uint32x4_t __s1_831 = __p1_831; \
62773		__ret_831 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_831), (uint16x4_t)(vshrn_n_u32(__s1_831, __p2_831)))); \
62774		__ret_831; \
62775		})
	62108	__ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	62109	poly8x16_t __ret;
	62110	__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
	62111	return __ret;
	62112	}
62776	62113	#else
62777		#define vshrn_high_n_u32(__p0_832, __p1_832, __p2_832) __extension__ ({ \
62778		uint16x8_t __ret_832; \
62779		uint16x4_t __s0_832 = __p0_832; \
62780		uint32x4_t __s1_832 = __p1_832; \
62781		uint16x4_t __rev0_832; __rev0_832 = __builtin_shufflevector(__s0_832, __s0_832, 3, 2, 1, 0); \
62782		uint32x4_t __rev1_832; __rev1_832 = __builtin_shufflevector(__s1_832, __s1_832, 3, 2, 1, 0); \
62783		__ret_832 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_832), (uint16x4_t)(__noswap_vshrn_n_u32(__rev1_832, __p2_832)))); \
62784		__ret_832 = __builtin_shufflevector(__ret_832, __ret_832, 7, 6, 5, 4, 3, 2, 1, 0); \
62785		__ret_832; \
62786		})
	62114	__ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	62115	poly8x16_t __ret;
	62116	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62117	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62118	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
	62119	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62120	return __ret;
	62121	}
62787	62122	#endif
62788	62123
62789	62124	#ifdef __LITTLE_ENDIAN__
62790		#define vshrn_high_n_u64(__p0_833, __p1_833, __p2_833) __extension__ ({ \
62791		uint32x4_t __ret_833; \
62792		uint32x2_t __s0_833 = __p0_833; \
62793		uint64x2_t __s1_833 = __p1_833; \
62794		__ret_833 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_833), (uint32x2_t)(vshrn_n_u64(__s1_833, __p2_833)))); \
62795		__ret_833; \
62796		})
	62125	__ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	62126	poly64x2_t __ret;
	62127	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	62128	return __ret;
	62129	}
62797	62130	#else
62798		#define vshrn_high_n_u64(__p0_834, __p1_834, __p2_834) __extension__ ({ \
62799		uint32x4_t __ret_834; \
62800		uint32x2_t __s0_834 = __p0_834; \
62801		uint64x2_t __s1_834 = __p1_834; \
62802		uint32x2_t __rev0_834; __rev0_834 = __builtin_shufflevector(__s0_834, __s0_834, 1, 0); \
62803		uint64x2_t __rev1_834; __rev1_834 = __builtin_shufflevector(__s1_834, __s1_834, 1, 0); \
62804		__ret_834 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_834), (uint32x2_t)(__noswap_vshrn_n_u64(__rev1_834, __p2_834)))); \
62805		__ret_834 = __builtin_shufflevector(__ret_834, __ret_834, 3, 2, 1, 0); \
62806		__ret_834; \
62807		})
	62131	__ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	62132	poly64x2_t __ret;
	62133	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62134	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62135	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	62136	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62137	return __ret;
	62138	}
62808	62139	#endif
62809	62140
62810	62141	#ifdef __LITTLE_ENDIAN__
62811		#define vshrn_high_n_u16(__p0_835, __p1_835, __p2_835) __extension__ ({ \
62812		uint8x16_t __ret_835; \
62813		uint8x8_t __s0_835 = __p0_835; \
62814		uint16x8_t __s1_835 = __p1_835; \
62815		__ret_835 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_835), (uint8x8_t)(vshrn_n_u16(__s1_835, __p2_835)))); \
62816		__ret_835; \
62817		})
	62142	__ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	62143	poly16x8_t __ret;
	62144	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
	62145	return __ret;
	62146	}
62818	62147	#else
62819		#define vshrn_high_n_u16(__p0_836, __p1_836, __p2_836) __extension__ ({ \
62820		uint8x16_t __ret_836; \
62821		uint8x8_t __s0_836 = __p0_836; \
62822		uint16x8_t __s1_836 = __p1_836; \
62823		uint8x8_t __rev0_836; __rev0_836 = __builtin_shufflevector(__s0_836, __s0_836, 7, 6, 5, 4, 3, 2, 1, 0); \
62824		uint16x8_t __rev1_836; __rev1_836 = __builtin_shufflevector(__s1_836, __s1_836, 7, 6, 5, 4, 3, 2, 1, 0); \
62825		__ret_836 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_836), (uint8x8_t)(__noswap_vshrn_n_u16(__rev1_836, __p2_836)))); \
62826		__ret_836 = __builtin_shufflevector(__ret_836, __ret_836, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
62827		__ret_836; \
62828		})
	62148	__ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	62149	poly16x8_t __ret;
	62150	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62151	poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62152	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
	62153	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62154	return __ret;
	62155	}
62829	62156	#endif
62830	62157
62831	62158	#ifdef __LITTLE_ENDIAN__
62832		#define vshrn_high_n_s32(__p0_837, __p1_837, __p2_837) __extension__ ({ \
62833		int16x8_t __ret_837; \
62834		int16x4_t __s0_837 = __p0_837; \
62835		int32x4_t __s1_837 = __p1_837; \
62836		__ret_837 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_837), (int16x4_t)(vshrn_n_s32(__s1_837, __p2_837)))); \
62837		__ret_837; \
62838		})
	62159	__ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	62160	uint8x16_t __ret;
	62161	__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
	62162	return __ret;
	62163	}
62839	62164	#else
62840		#define vshrn_high_n_s32(__p0_838, __p1_838, __p2_838) __extension__ ({ \
62841		int16x8_t __ret_838; \
62842		int16x4_t __s0_838 = __p0_838; \
62843		int32x4_t __s1_838 = __p1_838; \
62844		int16x4_t __rev0_838; __rev0_838 = __builtin_shufflevector(__s0_838, __s0_838, 3, 2, 1, 0); \
62845		int32x4_t __rev1_838; __rev1_838 = __builtin_shufflevector(__s1_838, __s1_838, 3, 2, 1, 0); \
62846		__ret_838 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_838), (int16x4_t)(__noswap_vshrn_n_s32(__rev1_838, __p2_838)))); \
62847		__ret_838 = __builtin_shufflevector(__ret_838, __ret_838, 7, 6, 5, 4, 3, 2, 1, 0); \
62848		__ret_838; \
62849		})
	62165	__ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	62166	uint8x16_t __ret;
	62167	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62168	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62169	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
	62170	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62171	return __ret;
	62172	}
62850	62173	#endif
62851	62174
62852	62175	#ifdef __LITTLE_ENDIAN__
62853		#define vshrn_high_n_s64(__p0_839, __p1_839, __p2_839) __extension__ ({ \
62854		int32x4_t __ret_839; \
62855		int32x2_t __s0_839 = __p0_839; \
62856		int64x2_t __s1_839 = __p1_839; \
62857		__ret_839 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_839), (int32x2_t)(vshrn_n_s64(__s1_839, __p2_839)))); \
62858		__ret_839; \
62859		})
	62176	__ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	62177	uint32x4_t __ret;
	62178	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
	62179	return __ret;
	62180	}
62860	62181	#else
62861		#define vshrn_high_n_s64(__p0_840, __p1_840, __p2_840) __extension__ ({ \
62862		int32x4_t __ret_840; \
62863		int32x2_t __s0_840 = __p0_840; \
62864		int64x2_t __s1_840 = __p1_840; \
62865		int32x2_t __rev0_840; __rev0_840 = __builtin_shufflevector(__s0_840, __s0_840, 1, 0); \
62866		int64x2_t __rev1_840; __rev1_840 = __builtin_shufflevector(__s1_840, __s1_840, 1, 0); \
62867		__ret_840 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_840), (int32x2_t)(__noswap_vshrn_n_s64(__rev1_840, __p2_840)))); \
62868		__ret_840 = __builtin_shufflevector(__ret_840, __ret_840, 3, 2, 1, 0); \
62869		__ret_840; \
62870		})
	62182	__ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	62183	uint32x4_t __ret;
	62184	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62185	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62186	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
	62187	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62188	return __ret;
	62189	}
62871	62190	#endif
62872	62191
62873	62192	#ifdef __LITTLE_ENDIAN__
62874		#define vshrn_high_n_s16(__p0_841, __p1_841, __p2_841) __extension__ ({ \
62875		int8x16_t __ret_841; \
62876		int8x8_t __s0_841 = __p0_841; \
62877		int16x8_t __s1_841 = __p1_841; \
62878		__ret_841 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_841), (int8x8_t)(vshrn_n_s16(__s1_841, __p2_841)))); \
62879		__ret_841; \
62880		})
	62193	__ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	62194	uint64x2_t __ret;
	62195	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	62196	return __ret;
	62197	}
62881	62198	#else
62882		#define vshrn_high_n_s16(__p0_842, __p1_842, __p2_842) __extension__ ({ \
62883		int8x16_t __ret_842; \
62884		int8x8_t __s0_842 = __p0_842; \
62885		int16x8_t __s1_842 = __p1_842; \
62886		int8x8_t __rev0_842; __rev0_842 = __builtin_shufflevector(__s0_842, __s0_842, 7, 6, 5, 4, 3, 2, 1, 0); \
62887		int16x8_t __rev1_842; __rev1_842 = __builtin_shufflevector(__s1_842, __s1_842, 7, 6, 5, 4, 3, 2, 1, 0); \
62888		__ret_842 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_842), (int8x8_t)(__noswap_vshrn_n_s16(__rev1_842, __p2_842)))); \
62889		__ret_842 = __builtin_shufflevector(__ret_842, __ret_842, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
62890		__ret_842; \
62891		})
	62199	__ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	62200	uint64x2_t __ret;
	62201	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62202	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62203	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	62204	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62205	return __ret;
	62206	}
62892	62207	#endif
62893	62208
62894		#define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
62895		uint64_t __ret; \
62896		uint64_t __s0 = __p0; \
62897		uint64_t __s1 = __p1; \
62898		__ret = (uint64_t) __builtin_neon_vslid_n_u64(__s0, __s1, __p2); \
62899		__ret; \
62900		})
62901		#define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
62902		int64_t __ret; \
62903		int64_t __s0 = __p0; \
62904		int64_t __s1 = __p1; \
62905		__ret = (int64_t) __builtin_neon_vslid_n_s64(__s0, __s1, __p2); \
62906		__ret; \
62907		})
62908		#define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
62909		poly64x1_t __ret; \
62910		poly64x1_t __s0 = __p0; \
62911		poly64x1_t __s1 = __p1; \
62912		__ret = (poly64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
62913		__ret; \
62914		})
62915	62209	#ifdef __LITTLE_ENDIAN__
62916		#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
62917		poly64x2_t __ret; \
62918		poly64x2_t __s0 = __p0; \
62919		poly64x2_t __s1 = __p1; \
62920		__ret = (poly64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
62921		__ret; \
62922		})
	62210	__ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	62211	uint16x8_t __ret;
	62212	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
	62213	return __ret;
	62214	}
62923	62215	#else
62924		#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
62925		poly64x2_t __ret; \
62926		poly64x2_t __s0 = __p0; \
62927		poly64x2_t __s1 = __p1; \
62928		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
62929		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
62930		__ret = (poly64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
62931		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
62932		__ret; \
62933		})
	62216	__ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	62217	uint16x8_t __ret;
	62218	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62219	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62220	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
	62221	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62222	return __ret;
	62223	}
62934	62224	#endif
62935	62225
62936		__ai uint8_t vsqaddb_u8(uint8_t __p0, int8_t __p1) {
62937		uint8_t __ret;
62938		__ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
	62226	#ifdef __LITTLE_ENDIAN__
	62227	__ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
	62228	int8x16_t __ret;
	62229	__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
62939	62230	return __ret;
62940	62231	}
62941		__ai uint32_t vsqadds_u32(uint32_t __p0, int32_t __p1) {
62942		uint32_t __ret;
62943		__ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
	62232	#else
	62233	__ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
	62234	int8x16_t __ret;
	62235	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62236	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62237	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
	62238	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62944	62239	return __ret;
62945	62240	}
62946		__ai uint64_t vsqaddd_u64(uint64_t __p0, int64_t __p1) {
62947		uint64_t __ret;
62948		__ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
	62241	#endif
	62242
	62243	#ifdef __LITTLE_ENDIAN__
	62244	__ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
	62245	float64x2_t __ret;
	62246	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62949	62247	return __ret;
62950	62248	}
62951		__ai uint16_t vsqaddh_u16(uint16_t __p0, int16_t __p1) {
62952		uint16_t __ret;
62953		__ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
	62249	#else
	62250	__ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
	62251	float64x2_t __ret;
	62252	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62253	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62254	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	62255	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
62954	62256	return __ret;
62955	62257	}
	62258	#endif
	62259
62956	62260	#ifdef __LITTLE_ENDIAN__
62957		__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
62958		uint8x16_t __ret;
62959		__ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
	62261	__ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
	62262	float32x4_t __ret;
	62263	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
62960	62264	return __ret;
62961	62265	}
62962	62266	#else
62963		__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
62964		uint8x16_t __ret;
62965		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62966		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62967		__ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
62968		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62267	__ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
	62268	float32x4_t __ret;
	62269	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62270	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62271	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
	62272	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62969	62273	return __ret;
62970	62274	}
62971	62275	#endif
62972	62276
62973	62277	#ifdef __LITTLE_ENDIAN__
62974		__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
62975		uint32x4_t __ret;
62976		__ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
	62278	__ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
	62279	int32x4_t __ret;
	62280	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
62977	62281	return __ret;
62978	62282	}
62979	62283	#else
62980		__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
62981		uint32x4_t __ret;
62982		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62284	__ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
	62285	int32x4_t __ret;
	62286	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
62983	62287	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
62984		__ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	62288	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
62985	62289	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
62986	62290	return __ret;
62987	62291	}
62988	62292	#endif
62989	62293
62990	62294	#ifdef __LITTLE_ENDIAN__
62991		__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
62992		uint64x2_t __ret;
62993		__ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	62295	__ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
	62296	int64x2_t __ret;
	62297	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
62994	62298	return __ret;
62995	62299	}
62996	62300	#else
62997		__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
62998		uint64x2_t __ret;
62999		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62301	__ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
	62302	int64x2_t __ret;
	62303	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
63000	62304	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63001		__ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	62305	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
63002	62306	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
63003	62307	return __ret;
63004	62308	}
63005	62309	#endif
63006	62310
63007	62311	#ifdef __LITTLE_ENDIAN__
63008		__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
63009		uint16x8_t __ret;
63010		__ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
	62312	__ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
	62313	int16x8_t __ret;
	62314	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
63011	62315	return __ret;
63012	62316	}
63013	62317	#else
63014		__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
63015		uint16x8_t __ret;
63016		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62318	__ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
	62319	int16x8_t __ret;
	62320	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
63017	62321	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63018		__ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
	62322	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
63019	62323	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
63020	62324	return __ret;
63021	62325	}
63022	62326	#endif
63023	62327
63024	62328	#ifdef __LITTLE_ENDIAN__
63025		__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
	62329	__ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
63026	62330	uint8x8_t __ret;
63027		__ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
	62331	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
63028	62332	return __ret;
63029	62333	}
63030	62334	#else
63031		__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
	62335	__ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
63032	62336	uint8x8_t __ret;
63033	62337	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
63034		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63035		__ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
	62338	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62339	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
63036	62340	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
63037	62341	return __ret;
63038	62342	}
63039	62343	#endif
63040	62344
63041	62345	#ifdef __LITTLE_ENDIAN__
63042		__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
	62346	__ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
63043	62347	uint32x2_t __ret;
63044		__ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
	62348	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63045	62349	return __ret;
63046	62350	}
63047	62351	#else
63048		__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
	62352	__ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
63049	62353	uint32x2_t __ret;
63050	62354	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
63051		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63052		__ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
	62355	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62356	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
63053	62357	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
63054	62358	return __ret;
63055	62359	}
63056	62360	#endif
63057	62361
63058		__ai uint64x1_t vsqadd_u64(uint64x1_t __p0, int64x1_t __p1) {
63059		uint64x1_t __ret;
63060		__ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
63061		return __ret;
63062		}
63063	62362	#ifdef __LITTLE_ENDIAN__
63064		__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
	62363	__ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
63065	62364	uint16x4_t __ret;
63066		__ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
	62365	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
63067	62366	return __ret;
63068	62367	}
63069	62368	#else
63070		__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
	62369	__ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
63071	62370	uint16x4_t __ret;
63072	62371	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
63073		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63074		__ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
	62372	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62373	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
63075	62374	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
63076	62375	return __ret;
63077	62376	}
63078	62377	#endif
63079	62378
63080	62379	#ifdef __LITTLE_ENDIAN__
63081		__ai float64x2_t vsqrtq_f64(float64x2_t __p0) {
63082		float64x2_t __ret;
63083		__ret = (float64x2_t) __builtin_neon_vsqrtq_v((int8x16_t)__p0, 42);
	62380	__ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
	62381	int8x8_t __ret;
	62382	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
63084	62383	return __ret;
63085	62384	}
63086	62385	#else
63087		__ai float64x2_t vsqrtq_f64(float64x2_t __p0) {
63088		float64x2_t __ret;
63089		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
63090		__ret = (float64x2_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 42);
63091		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62386	__ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
	62387	int8x8_t __ret;
	62388	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62389	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62390	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
	62391	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
63092	62392	return __ret;
63093	62393	}
63094	62394	#endif
63095	62395
63096	62396	#ifdef __LITTLE_ENDIAN__
63097		__ai float32x4_t vsqrtq_f32(float32x4_t __p0) {
63098		float32x4_t __ret;
63099		__ret = (float32x4_t) __builtin_neon_vsqrtq_v((int8x16_t)__p0, 41);
	62397	__ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
	62398	float32x2_t __ret;
	62399	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63100	62400	return __ret;
63101	62401	}
63102	62402	#else
63103		__ai float32x4_t vsqrtq_f32(float32x4_t __p0) {
63104		float32x4_t __ret;
63105		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
63106		__ret = (float32x4_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 41);
63107		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62403	__ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
	62404	float32x2_t __ret;
	62405	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62406	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62407	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	62408	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
63108	62409	return __ret;
63109	62410	}
63110	62411	#endif
63111	62412
63112		__ai float64x1_t vsqrt_f64(float64x1_t __p0) {
63113		float64x1_t __ret;
63114		__ret = (float64x1_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 10);
63115		return __ret;
63116		}
63117	62413	#ifdef __LITTLE_ENDIAN__
63118		__ai float32x2_t vsqrt_f32(float32x2_t __p0) {
63119		float32x2_t __ret;
63120		__ret = (float32x2_t) __builtin_neon_vsqrt_v((int8x8_t)__p0, 9);
	62414	__ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
	62415	int32x2_t __ret;
	62416	__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63121	62417	return __ret;
63122	62418	}
63123	62419	#else
63124		__ai float32x2_t vsqrt_f32(float32x2_t __p0) {
63125		float32x2_t __ret;
63126		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
63127		__ret = (float32x2_t) __builtin_neon_vsqrt_v((int8x8_t)__rev0, 9);
	62420	__ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
	62421	int32x2_t __ret;
	62422	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62423	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62424	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
63128	62425	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
63129	62426	return __ret;
63130	62427	}
63131	62428	#endif
63132	62429
63133		#define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
63134		uint64_t __ret; \
63135		uint64_t __s0 = __p0; \
63136		uint64_t __s1 = __p1; \
63137		__ret = (uint64_t) __builtin_neon_vsrad_n_u64(__s0, __s1, __p2); \
63138		__ret; \
63139		})
63140		#define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
63141		int64_t __ret; \
63142		int64_t __s0 = __p0; \
63143		int64_t __s1 = __p1; \
63144		__ret = (int64_t) __builtin_neon_vsrad_n_s64(__s0, __s1, __p2); \
63145		__ret; \
63146		})
63147		#define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
63148		uint64_t __ret; \
63149		uint64_t __s0 = __p0; \
63150		uint64_t __s1 = __p1; \
63151		__ret = (uint64_t) __builtin_neon_vsrid_n_u64(__s0, __s1, __p2); \
63152		__ret; \
63153		})
63154		#define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
63155		int64_t __ret; \
63156		int64_t __s0 = __p0; \
63157		int64_t __s1 = __p1; \
63158		__ret = (int64_t) __builtin_neon_vsrid_n_s64(__s0, __s1, __p2); \
63159		__ret; \
63160		})
63161		#define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
63162		poly64x1_t __ret; \
63163		poly64x1_t __s0 = __p0; \
63164		poly64x1_t __s1 = __p1; \
63165		__ret = (poly64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
63166		__ret; \
63167		})
63168		#ifdef __LITTLE_ENDIAN__
63169		#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
63170		poly64x2_t __ret; \
63171		poly64x2_t __s0 = __p0; \
63172		poly64x2_t __s1 = __p1; \
63173		__ret = (poly64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
63174		__ret; \
63175		})
63176		#else
63177		#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
63178		poly64x2_t __ret; \
63179		poly64x2_t __s0 = __p0; \
63180		poly64x2_t __s1 = __p1; \
63181		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
63182		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
63183		__ret = (poly64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
63184		__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
63185		__ret; \
63186		})
63187		#endif
63188
63189		#define vst1_p64(__p0, __p1) __extension__ ({ \
63190		poly64x1_t __s1 = __p1; \
63191		__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 6); \
63192		})
63193		#ifdef __LITTLE_ENDIAN__
63194		#define vst1q_p64(__p0, __p1) __extension__ ({ \
63195		poly64x2_t __s1 = __p1; \
63196		__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 38); \
63197		})
63198		#else
63199		#define vst1q_p64(__p0, __p1) __extension__ ({ \
63200		poly64x2_t __s1 = __p1; \
63201		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
63202		__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 38); \
63203		})
63204		#endif
63205
63206		#ifdef __LITTLE_ENDIAN__
63207		#define vst1q_f64(__p0, __p1) __extension__ ({ \
63208		float64x2_t __s1 = __p1; \
63209		__builtin_neon_vst1q_v(__p0, (int8x16_t)__s1, 42); \
63210		})
63211		#else
63212		#define vst1q_f64(__p0, __p1) __extension__ ({ \
63213		float64x2_t __s1 = __p1; \
63214		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
63215		__builtin_neon_vst1q_v(__p0, (int8x16_t)__rev1, 42); \
63216		})
63217		#endif
63218
63219		#define vst1_f64(__p0, __p1) __extension__ ({ \
63220		float64x1_t __s1 = __p1; \
63221		__builtin_neon_vst1_v(__p0, (int8x8_t)__s1, 10); \
63222		})
63223		#define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63224		poly64x1_t __s1 = __p1; \
63225		__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
63226		})
63227		#ifdef __LITTLE_ENDIAN__
63228		#define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63229		poly64x2_t __s1 = __p1; \
63230		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 38); \
63231		})
63232		#else
63233		#define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63234		poly64x2_t __s1 = __p1; \
63235		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
63236		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 38); \
63237		})
63238		#endif
63239
63240		#ifdef __LITTLE_ENDIAN__
63241		#define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63242		float64x2_t __s1 = __p1; \
63243		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__s1, __p2, 42); \
63244		})
63245		#else
63246		#define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63247		float64x2_t __s1 = __p1; \
63248		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
63249		__builtin_neon_vst1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 42); \
63250		})
63251		#endif
63252
63253		#define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63254		float64x1_t __s1 = __p1; \
63255		__builtin_neon_vst1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
63256		})
63257		#define vst1_p64_x2(__p0, __p1) __extension__ ({ \
63258		poly64x1x2_t __s1 = __p1; \
63259		__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
63260		})
63261		#ifdef __LITTLE_ENDIAN__
63262		#define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
63263		poly64x2x2_t __s1 = __p1; \
63264		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 38); \
63265		})
63266		#else
63267		#define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
63268		poly64x2x2_t __s1 = __p1; \
63269		poly64x2x2_t __rev1; \
63270		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63271		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63272		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 38); \
63273		})
63274		#endif
63275
63276		#ifdef __LITTLE_ENDIAN__
63277		#define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
63278		float64x2x2_t __s1 = __p1; \
63279		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 42); \
63280		})
63281		#else
63282		#define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
63283		float64x2x2_t __s1 = __p1; \
63284		float64x2x2_t __rev1; \
63285		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63286		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63287		__builtin_neon_vst1q_x2_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 42); \
63288		})
63289		#endif
63290
63291		#define vst1_f64_x2(__p0, __p1) __extension__ ({ \
63292		float64x1x2_t __s1 = __p1; \
63293		__builtin_neon_vst1_x2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 10); \
63294		})
63295		#define vst1_p64_x3(__p0, __p1) __extension__ ({ \
63296		poly64x1x3_t __s1 = __p1; \
63297		__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
63298		})
63299		#ifdef __LITTLE_ENDIAN__
63300		#define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
63301		poly64x2x3_t __s1 = __p1; \
63302		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 38); \
63303		})
63304		#else
63305		#define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
63306		poly64x2x3_t __s1 = __p1; \
63307		poly64x2x3_t __rev1; \
63308		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63309		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63310		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63311		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 38); \
63312		})
63313		#endif
63314
63315		#ifdef __LITTLE_ENDIAN__
63316		#define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
63317		float64x2x3_t __s1 = __p1; \
63318		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 42); \
63319		})
63320		#else
63321		#define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
63322		float64x2x3_t __s1 = __p1; \
63323		float64x2x3_t __rev1; \
63324		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63325		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63326		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63327		__builtin_neon_vst1q_x3_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 42); \
63328		})
63329		#endif
63330
63331		#define vst1_f64_x3(__p0, __p1) __extension__ ({ \
63332		float64x1x3_t __s1 = __p1; \
63333		__builtin_neon_vst1_x3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 10); \
63334		})
63335		#define vst1_p64_x4(__p0, __p1) __extension__ ({ \
63336		poly64x1x4_t __s1 = __p1; \
63337		__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
63338		})
63339		#ifdef __LITTLE_ENDIAN__
63340		#define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
63341		poly64x2x4_t __s1 = __p1; \
63342		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 38); \
63343		})
63344		#else
63345		#define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
63346		poly64x2x4_t __s1 = __p1; \
63347		poly64x2x4_t __rev1; \
63348		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63349		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63350		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63351		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63352		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 38); \
63353		})
63354		#endif
63355
63356		#ifdef __LITTLE_ENDIAN__
63357		#define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
63358		float64x2x4_t __s1 = __p1; \
63359		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 42); \
63360		})
63361		#else
63362		#define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
63363		float64x2x4_t __s1 = __p1; \
63364		float64x2x4_t __rev1; \
63365		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63366		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63367		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63368		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63369		__builtin_neon_vst1q_x4_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 42); \
63370		})
63371		#endif
63372
63373		#define vst1_f64_x4(__p0, __p1) __extension__ ({ \
63374		float64x1x4_t __s1 = __p1; \
63375		__builtin_neon_vst1_x4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 10); \
63376		})
63377		#define vst2_p64(__p0, __p1) __extension__ ({ \
63378		poly64x1x2_t __s1 = __p1; \
63379		__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 6); \
63380		})
63381		#ifdef __LITTLE_ENDIAN__
63382		#define vst2q_p64(__p0, __p1) __extension__ ({ \
63383		poly64x2x2_t __s1 = __p1; \
63384		__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 38); \
63385		})
63386		#else
63387		#define vst2q_p64(__p0, __p1) __extension__ ({ \
63388		poly64x2x2_t __s1 = __p1; \
63389		poly64x2x2_t __rev1; \
63390		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63391		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63392		__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 38); \
63393		})
63394		#endif
63395
63396	62430	#ifdef __LITTLE_ENDIAN__
63397		#define vst2q_u64(__p0, __p1) __extension__ ({ \
63398		uint64x2x2_t __s1 = __p1; \
63399		__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 51); \
63400		})
	62431	__ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
	62432	int16x4_t __ret;
	62433	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
	62434	return __ret;
	62435	}
63401	62436	#else
63402		#define vst2q_u64(__p0, __p1) __extension__ ({ \
63403		uint64x2x2_t __s1 = __p1; \
63404		uint64x2x2_t __rev1; \
63405		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63406		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63407		__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 51); \
63408		})
	62437	__ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
	62438	int16x4_t __ret;
	62439	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62440	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62441	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
	62442	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62443	return __ret;
	62444	}
63409	62445	#endif
63410	62446
63411	62447	#ifdef __LITTLE_ENDIAN__
63412		#define vst2q_f64(__p0, __p1) __extension__ ({ \
63413		float64x2x2_t __s1 = __p1; \
63414		__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 42); \
63415		})
	62448	__ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
	62449	poly8x8_t __ret;
	62450	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	62451	return __ret;
	62452	}
63416	62453	#else
63417		#define vst2q_f64(__p0, __p1) __extension__ ({ \
63418		float64x2x2_t __s1 = __p1; \
63419		float64x2x2_t __rev1; \
63420		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63421		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63422		__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 42); \
63423		})
	62454	__ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
	62455	poly8x8_t __ret;
	62456	poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62457	poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62458	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
	62459	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62460	return __ret;
	62461	}
63424	62462	#endif
63425	62463
63426	62464	#ifdef __LITTLE_ENDIAN__
63427		#define vst2q_s64(__p0, __p1) __extension__ ({ \
63428		int64x2x2_t __s1 = __p1; \
63429		__builtin_neon_vst2q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], 35); \
63430		})
	62465	__ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
	62466	poly16x4_t __ret;
	62467	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	62468	return __ret;
	62469	}
63431	62470	#else
63432		#define vst2q_s64(__p0, __p1) __extension__ ({ \
63433		int64x2x2_t __s1 = __p1; \
63434		int64x2x2_t __rev1; \
63435		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63436		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63437		__builtin_neon_vst2q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], 35); \
63438		})
	62471	__ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
	62472	poly16x4_t __ret;
	62473	poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62474	poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62475	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	62476	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62477	return __ret;
	62478	}
63439	62479	#endif
63440	62480
63441		#define vst2_f64(__p0, __p1) __extension__ ({ \
63442		float64x1x2_t __s1 = __p1; \
63443		__builtin_neon_vst2_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], 10); \
63444		})
63445		#define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63446		poly64x1x2_t __s1 = __p1; \
63447		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
63448		})
63449	62481	#ifdef __LITTLE_ENDIAN__
63450		#define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63451		poly8x16x2_t __s1 = __p1; \
63452		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 36); \
63453		})
	62482	__ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	62483	poly8x16_t __ret;
	62484	__ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	62485	return __ret;
	62486	}
63454	62487	#else
63455		#define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63456		poly8x16x2_t __s1 = __p1; \
63457		poly8x16x2_t __rev1; \
63458		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63459		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63460		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 36); \
63461		})
	62488	__ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
	62489	poly8x16_t __ret;
	62490	poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62491	poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62492	__ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	62493	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62494	return __ret;
	62495	}
63462	62496	#endif
63463	62497
63464	62498	#ifdef __LITTLE_ENDIAN__
63465		#define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63466		poly64x2x2_t __s1 = __p1; \
63467		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 38); \
63468		})
	62499	__ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	62500	poly64x2_t __ret;
	62501	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62502	return __ret;
	62503	}
63469	62504	#else
63470		#define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63471		poly64x2x2_t __s1 = __p1; \
63472		poly64x2x2_t __rev1; \
63473		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63474		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63475		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 38); \
63476		})
	62505	__ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
	62506	poly64x2_t __ret;
	62507	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62508	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62509	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62510	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62511	return __ret;
	62512	}
63477	62513	#endif
63478	62514
63479	62515	#ifdef __LITTLE_ENDIAN__
63480		#define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63481		uint8x16x2_t __s1 = __p1; \
63482		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 48); \
63483		})
	62516	__ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	62517	poly16x8_t __ret;
	62518	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	62519	return __ret;
	62520	}
63484	62521	#else
63485		#define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63486		uint8x16x2_t __s1 = __p1; \
63487		uint8x16x2_t __rev1; \
63488		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63489		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63490		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 48); \
63491		})
	62522	__ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
	62523	poly16x8_t __ret;
	62524	poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62525	poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62526	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
	62527	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62528	return __ret;
	62529	}
63492	62530	#endif
63493	62531
63494	62532	#ifdef __LITTLE_ENDIAN__
63495		#define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63496		uint64x2x2_t __s1 = __p1; \
63497		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 51); \
63498		})
	62533	__ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	62534	uint8x16_t __ret;
	62535	__ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	62536	return __ret;
	62537	}
63499	62538	#else
63500		#define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63501		uint64x2x2_t __s1 = __p1; \
63502		uint64x2x2_t __rev1; \
63503		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63504		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63505		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 51); \
63506		})
	62539	__ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	62540	uint8x16_t __ret;
	62541	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62542	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62543	__ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	62544	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62545	return __ret;
	62546	}
63507	62547	#endif
63508	62548
63509	62549	#ifdef __LITTLE_ENDIAN__
63510		#define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63511		int8x16x2_t __s1 = __p1; \
63512		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 32); \
63513		})
	62550	__ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	62551	uint32x4_t __ret;
	62552	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	62553	return __ret;
	62554	}
63514	62555	#else
63515		#define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63516		int8x16x2_t __s1 = __p1; \
63517		int8x16x2_t __rev1; \
63518		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63519		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63520		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 32); \
63521		})
	62556	__ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	62557	uint32x4_t __ret;
	62558	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62559	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62560	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	62561	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62562	return __ret;
	62563	}
63522	62564	#endif
63523	62565
63524	62566	#ifdef __LITTLE_ENDIAN__
63525		#define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63526		float64x2x2_t __s1 = __p1; \
63527		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
63528		})
	62567	__ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	62568	uint64x2_t __ret;
	62569	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62570	return __ret;
	62571	}
63529	62572	#else
63530		#define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63531		float64x2x2_t __s1 = __p1; \
63532		float64x2x2_t __rev1; \
63533		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63534		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63535		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
63536		})
	62573	__ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	62574	uint64x2_t __ret;
	62575	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62576	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62577	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62578	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62579	return __ret;
	62580	}
63537	62581	#endif
63538	62582
63539	62583	#ifdef __LITTLE_ENDIAN__
63540		#define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63541		int64x2x2_t __s1 = __p1; \
63542		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
63543		})
	62584	__ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	62585	uint16x8_t __ret;
	62586	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	62587	return __ret;
	62588	}
63544	62589	#else
63545		#define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63546		int64x2x2_t __s1 = __p1; \
63547		int64x2x2_t __rev1; \
63548		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63549		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63550		__builtin_neon_vst2q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
63551		})
	62590	__ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	62591	uint16x8_t __ret;
	62592	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62593	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62594	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
	62595	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62596	return __ret;
	62597	}
63552	62598	#endif
63553	62599
63554		#define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63555		uint64x1x2_t __s1 = __p1; \
63556		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
63557		})
63558		#define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63559		float64x1x2_t __s1 = __p1; \
63560		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
63561		})
63562		#define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63563		int64x1x2_t __s1 = __p1; \
63564		__builtin_neon_vst2_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
63565		})
63566		#define vst3_p64(__p0, __p1) __extension__ ({ \
63567		poly64x1x3_t __s1 = __p1; \
63568		__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 6); \
63569		})
63570	62600	#ifdef __LITTLE_ENDIAN__
63571		#define vst3q_p64(__p0, __p1) __extension__ ({ \
63572		poly64x2x3_t __s1 = __p1; \
63573		__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 38); \
63574		})
	62601	__ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
	62602	int8x16_t __ret;
	62603	__ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	62604	return __ret;
	62605	}
63575	62606	#else
63576		#define vst3q_p64(__p0, __p1) __extension__ ({ \
63577		poly64x2x3_t __s1 = __p1; \
63578		poly64x2x3_t __rev1; \
63579		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63580		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63581		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63582		__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 38); \
63583		})
	62607	__ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
	62608	int8x16_t __ret;
	62609	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62610	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62611	__ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	62612	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	62613	return __ret;
	62614	}
63584	62615	#endif
63585	62616
63586	62617	#ifdef __LITTLE_ENDIAN__
63587		#define vst3q_u64(__p0, __p1) __extension__ ({ \
63588		uint64x2x3_t __s1 = __p1; \
63589		__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 51); \
63590		})
	62618	__ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
	62619	float64x2_t __ret;
	62620	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62621	return __ret;
	62622	}
63591	62623	#else
63592		#define vst3q_u64(__p0, __p1) __extension__ ({ \
63593		uint64x2x3_t __s1 = __p1; \
63594		uint64x2x3_t __rev1; \
63595		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63596		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63597		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63598		__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 51); \
63599		})
	62624	__ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
	62625	float64x2_t __ret;
	62626	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62627	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62628	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62629	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62630	return __ret;
	62631	}
63600	62632	#endif
63601	62633
63602	62634	#ifdef __LITTLE_ENDIAN__
63603		#define vst3q_f64(__p0, __p1) __extension__ ({ \
63604		float64x2x3_t __s1 = __p1; \
63605		__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 42); \
63606		})
	62635	__ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
	62636	float32x4_t __ret;
	62637	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	62638	return __ret;
	62639	}
63607	62640	#else
63608		#define vst3q_f64(__p0, __p1) __extension__ ({ \
63609		float64x2x3_t __s1 = __p1; \
63610		float64x2x3_t __rev1; \
63611		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63612		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63613		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63614		__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 42); \
63615		})
	62641	__ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
	62642	float32x4_t __ret;
	62643	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62644	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62645	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	62646	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62647	return __ret;
	62648	}
63616	62649	#endif
63617	62650
63618	62651	#ifdef __LITTLE_ENDIAN__
63619		#define vst3q_s64(__p0, __p1) __extension__ ({ \
63620		int64x2x3_t __s1 = __p1; \
63621		__builtin_neon_vst3q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], 35); \
63622		})
	62652	__ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
	62653	int32x4_t __ret;
	62654	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	62655	return __ret;
	62656	}
63623	62657	#else
63624		#define vst3q_s64(__p0, __p1) __extension__ ({ \
63625		int64x2x3_t __s1 = __p1; \
63626		int64x2x3_t __rev1; \
63627		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63628		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63629		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63630		__builtin_neon_vst3q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], 35); \
63631		})
	62658	__ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
	62659	int32x4_t __ret;
	62660	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62661	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62662	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	62663	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62664	return __ret;
	62665	}
63632	62666	#endif
63633	62667
63634		#define vst3_f64(__p0, __p1) __extension__ ({ \
63635		float64x1x3_t __s1 = __p1; \
63636		__builtin_neon_vst3_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], 10); \
63637		})
63638		#define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63639		poly64x1x3_t __s1 = __p1; \
63640		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
63641		})
63642	62668	#ifdef __LITTLE_ENDIAN__
63643		#define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63644		poly8x16x3_t __s1 = __p1; \
63645		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 36); \
63646		})
	62669	__ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
	62670	int64x2_t __ret;
	62671	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62672	return __ret;
	62673	}
63647	62674	#else
63648		#define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63649		poly8x16x3_t __s1 = __p1; \
63650		poly8x16x3_t __rev1; \
63651		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63652		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63653		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63654		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 36); \
63655		})
	62675	__ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
	62676	int64x2_t __ret;
	62677	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62678	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62679	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62680	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62681	return __ret;
	62682	}
63656	62683	#endif
63657	62684
63658	62685	#ifdef __LITTLE_ENDIAN__
63659		#define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63660		poly64x2x3_t __s1 = __p1; \
63661		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 38); \
63662		})
	62686	__ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
	62687	int16x8_t __ret;
	62688	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	62689	return __ret;
	62690	}
63663	62691	#else
63664		#define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63665		poly64x2x3_t __s1 = __p1; \
63666		poly64x2x3_t __rev1; \
63667		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63668		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63669		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63670		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 38); \
63671		})
	62692	__ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
	62693	int16x8_t __ret;
	62694	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62695	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62696	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
	62697	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62698	return __ret;
	62699	}
63672	62700	#endif
63673	62701
63674	62702	#ifdef __LITTLE_ENDIAN__
63675		#define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63676		uint8x16x3_t __s1 = __p1; \
63677		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 48); \
63678		})
	62703	__ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
	62704	uint8x8_t __ret;
	62705	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	62706	return __ret;
	62707	}
63679	62708	#else
63680		#define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63681		uint8x16x3_t __s1 = __p1; \
63682		uint8x16x3_t __rev1; \
63683		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63684		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63685		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63686		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 48); \
63687		})
	62709	__ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
	62710	uint8x8_t __ret;
	62711	uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62712	uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62713	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
	62714	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62715	return __ret;
	62716	}
63688	62717	#endif
63689	62718
63690	62719	#ifdef __LITTLE_ENDIAN__
63691		#define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63692		uint64x2x3_t __s1 = __p1; \
63693		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 51); \
63694		})
	62720	__ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
	62721	uint32x2_t __ret;
	62722	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62723	return __ret;
	62724	}
63695	62725	#else
63696		#define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63697		uint64x2x3_t __s1 = __p1; \
63698		uint64x2x3_t __rev1; \
63699		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63700		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63701		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63702		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 51); \
63703		})
	62726	__ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
	62727	uint32x2_t __ret;
	62728	uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62729	uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62730	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62731	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62732	return __ret;
	62733	}
63704	62734	#endif
63705	62735
63706	62736	#ifdef __LITTLE_ENDIAN__
63707		#define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63708		int8x16x3_t __s1 = __p1; \
63709		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 32); \
63710		})
	62737	__ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
	62738	uint16x4_t __ret;
	62739	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	62740	return __ret;
	62741	}
63711	62742	#else
63712		#define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63713		int8x16x3_t __s1 = __p1; \
63714		int8x16x3_t __rev1; \
63715		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63716		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63717		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63718		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 32); \
63719		})
	62743	__ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
	62744	uint16x4_t __ret;
	62745	uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62746	uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62747	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	62748	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62749	return __ret;
	62750	}
63720	62751	#endif
63721	62752
63722	62753	#ifdef __LITTLE_ENDIAN__
63723		#define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63724		float64x2x3_t __s1 = __p1; \
63725		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
63726		})
	62754	__ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
	62755	int8x8_t __ret;
	62756	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	62757	return __ret;
	62758	}
63727	62759	#else
63728		#define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63729		float64x2x3_t __s1 = __p1; \
63730		float64x2x3_t __rev1; \
63731		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63732		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63733		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63734		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
63735		})
	62760	__ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
	62761	int8x8_t __ret;
	62762	int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62763	int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	62764	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
	62765	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62766	return __ret;
	62767	}
63736	62768	#endif
63737	62769
63738	62770	#ifdef __LITTLE_ENDIAN__
63739		#define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63740		int64x2x3_t __s1 = __p1; \
63741		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
63742		})
	62771	__ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
	62772	float32x2_t __ret;
	62773	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62774	return __ret;
	62775	}
63743	62776	#else
63744		#define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63745		int64x2x3_t __s1 = __p1; \
63746		int64x2x3_t __rev1; \
63747		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63748		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63749		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63750		__builtin_neon_vst3q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
63751		})
	62777	__ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
	62778	float32x2_t __ret;
	62779	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62780	float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62781	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62782	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62783	return __ret;
	62784	}
63752	62785	#endif
63753	62786
63754		#define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63755		uint64x1x3_t __s1 = __p1; \
63756		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
63757		})
63758		#define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63759		float64x1x3_t __s1 = __p1; \
63760		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
63761		})
63762		#define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63763		int64x1x3_t __s1 = __p1; \
63764		__builtin_neon_vst3_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
63765		})
63766		#define vst4_p64(__p0, __p1) __extension__ ({ \
63767		poly64x1x4_t __s1 = __p1; \
63768		__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 6); \
63769		})
63770	62787	#ifdef __LITTLE_ENDIAN__
63771		#define vst4q_p64(__p0, __p1) __extension__ ({ \
63772		poly64x2x4_t __s1 = __p1; \
63773		__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 38); \
63774		})
63775		#else
63776		#define vst4q_p64(__p0, __p1) __extension__ ({ \
63777		poly64x2x4_t __s1 = __p1; \
63778		poly64x2x4_t __rev1; \
63779		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63780		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63781		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63782		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63783		__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 38); \
63784		})
	62788	__ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
	62789	int32x2_t __ret;
	62790	__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	62791	return __ret;
	62792	}
	62793	#else
	62794	__ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
	62795	int32x2_t __ret;
	62796	int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62797	int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62798	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	62799	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	62800	return __ret;
	62801	}
63785	62802	#endif
63786	62803
63787	62804	#ifdef __LITTLE_ENDIAN__
63788		#define vst4q_u64(__p0, __p1) __extension__ ({ \
63789		uint64x2x4_t __s1 = __p1; \
63790		__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 51); \
63791		})
	62805	__ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
	62806	int16x4_t __ret;
	62807	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	62808	return __ret;
	62809	}
63792	62810	#else
63793		#define vst4q_u64(__p0, __p1) __extension__ ({ \
63794		uint64x2x4_t __s1 = __p1; \
63795		uint64x2x4_t __rev1; \
63796		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63797		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63798		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63799		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63800		__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 51); \
63801		})
	62811	__ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
	62812	int16x4_t __ret;
	62813	int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62814	int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62815	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	62816	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62817	return __ret;
	62818	}
63802	62819	#endif
63803	62820
	62821	__ai __attribute__((target("aes"))) poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
	62822	poly128_t __ret;
	62823	__ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
	62824	return __ret;
	62825	}
63804	62826	#ifdef __LITTLE_ENDIAN__
63805		#define vst4q_f64(__p0, __p1) __extension__ ({ \
63806		float64x2x4_t __s1 = __p1; \
63807		__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 42); \
63808		})
	62827	__ai __attribute__((target("aes"))) poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
	62828	poly128_t __ret;
	62829	__ret = vmull_p64((poly64_t)(vget_high_p64(__p0)), (poly64_t)(vget_high_p64(__p1)));
	62830	return __ret;
	62831	}
63809	62832	#else
63810		#define vst4q_f64(__p0, __p1) __extension__ ({ \
63811		float64x2x4_t __s1 = __p1; \
63812		float64x2x4_t __rev1; \
63813		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63814		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63815		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63816		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63817		__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 42); \
63818		})
	62833	__ai __attribute__((target("aes"))) poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
	62834	poly128_t __ret;
	62835	poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	62836	poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	62837	__ret = vmull_p64((poly64_t)(__noswap_vget_high_p64(__rev0)), (poly64_t)(__noswap_vget_high_p64(__rev1)));
	62838	return __ret;
	62839	}
63819	62840	#endif
63820	62841
63821	62842	#ifdef __LITTLE_ENDIAN__
63822		#define vst4q_s64(__p0, __p1) __extension__ ({ \
63823		int64x2x4_t __s1 = __p1; \
63824		__builtin_neon_vst4q_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], 35); \
63825		})
	62843	__ai __attribute__((target("bf16"))) bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
	62844	bfloat16x8_t __ret;
	62845	__ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_f32((int8x16_t)__p0, 43);
	62846	return __ret;
	62847	}
63826	62848	#else
63827		#define vst4q_s64(__p0, __p1) __extension__ ({ \
63828		int64x2x4_t __s1 = __p1; \
63829		int64x2x4_t __rev1; \
63830		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63831		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63832		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63833		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63834		__builtin_neon_vst4q_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], 35); \
63835		})
	62849	__ai __attribute__((target("bf16"))) bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
	62850	bfloat16x8_t __ret;
	62851	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62852	__ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_f32((int8x16_t)__rev0, 43);
	62853	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62854	return __ret;
	62855	}
	62856	__ai __attribute__((target("bf16"))) bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
	62857	bfloat16x8_t __ret;
	62858	__ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_f32((int8x16_t)__p0, 43);
	62859	return __ret;
	62860	}
63836	62861	#endif
63837	62862
63838		#define vst4_f64(__p0, __p1) __extension__ ({ \
63839		float64x1x4_t __s1 = __p1; \
63840		__builtin_neon_vst4_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], 10); \
63841		})
63842		#define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63843		poly64x1x4_t __s1 = __p1; \
63844		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 6); \
63845		})
63846	62863	#ifdef __LITTLE_ENDIAN__
63847		#define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63848		poly8x16x4_t __s1 = __p1; \
63849		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 36); \
	62864	#define vcopyq_lane_bf16(__p0_739, __p1_739, __p2_739, __p3_739) __extension__ ({ \
	62865	bfloat16x8_t __ret_739; \
	62866	bfloat16x8_t __s0_739 = __p0_739; \
	62867	bfloat16x4_t __s2_739 = __p2_739; \
	62868	__ret_739 = vsetq_lane_bf16(vget_lane_bf16(__s2_739, __p3_739), __s0_739, __p1_739); \
	62869	__ret_739; \
63850	62870	})
63851	62871	#else
63852		#define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63853		poly8x16x4_t __s1 = __p1; \
63854		poly8x16x4_t __rev1; \
63855		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63856		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63857		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63858		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63859		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 36); \
	62872	#define vcopyq_lane_bf16(__p0_740, __p1_740, __p2_740, __p3_740) __extension__ ({ \
	62873	bfloat16x8_t __ret_740; \
	62874	bfloat16x8_t __s0_740 = __p0_740; \
	62875	bfloat16x4_t __s2_740 = __p2_740; \
	62876	bfloat16x8_t __rev0_740; __rev0_740 = __builtin_shufflevector(__s0_740, __s0_740, 7, 6, 5, 4, 3, 2, 1, 0); \
	62877	bfloat16x4_t __rev2_740; __rev2_740 = __builtin_shufflevector(__s2_740, __s2_740, 3, 2, 1, 0); \
	62878	__ret_740 = __noswap_vsetq_lane_bf16(__noswap_vget_lane_bf16(__rev2_740, __p3_740), __rev0_740, __p1_740); \
	62879	__ret_740 = __builtin_shufflevector(__ret_740, __ret_740, 7, 6, 5, 4, 3, 2, 1, 0); \
	62880	__ret_740; \
63860	62881	})
63861	62882	#endif
63862	62883
63863	62884	#ifdef __LITTLE_ENDIAN__
63864		#define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63865		poly64x2x4_t __s1 = __p1; \
63866		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 38); \
	62885	#define vcopy_lane_bf16(__p0_741, __p1_741, __p2_741, __p3_741) __extension__ ({ \
	62886	bfloat16x4_t __ret_741; \
	62887	bfloat16x4_t __s0_741 = __p0_741; \
	62888	bfloat16x4_t __s2_741 = __p2_741; \
	62889	__ret_741 = vset_lane_bf16(vget_lane_bf16(__s2_741, __p3_741), __s0_741, __p1_741); \
	62890	__ret_741; \
63867	62891	})
63868	62892	#else
63869		#define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63870		poly64x2x4_t __s1 = __p1; \
63871		poly64x2x4_t __rev1; \
63872		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63873		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63874		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63875		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63876		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 38); \
	62893	#define vcopy_lane_bf16(__p0_742, __p1_742, __p2_742, __p3_742) __extension__ ({ \
	62894	bfloat16x4_t __ret_742; \
	62895	bfloat16x4_t __s0_742 = __p0_742; \
	62896	bfloat16x4_t __s2_742 = __p2_742; \
	62897	bfloat16x4_t __rev0_742; __rev0_742 = __builtin_shufflevector(__s0_742, __s0_742, 3, 2, 1, 0); \
	62898	bfloat16x4_t __rev2_742; __rev2_742 = __builtin_shufflevector(__s2_742, __s2_742, 3, 2, 1, 0); \
	62899	__ret_742 = __noswap_vset_lane_bf16(__noswap_vget_lane_bf16(__rev2_742, __p3_742), __rev0_742, __p1_742); \
	62900	__ret_742 = __builtin_shufflevector(__ret_742, __ret_742, 3, 2, 1, 0); \
	62901	__ret_742; \
63877	62902	})
63878	62903	#endif
63879	62904
63880	62905	#ifdef __LITTLE_ENDIAN__
63881		#define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63882		uint8x16x4_t __s1 = __p1; \
63883		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 48); \
	62906	#define vcopyq_laneq_bf16(__p0_743, __p1_743, __p2_743, __p3_743) __extension__ ({ \
	62907	bfloat16x8_t __ret_743; \
	62908	bfloat16x8_t __s0_743 = __p0_743; \
	62909	bfloat16x8_t __s2_743 = __p2_743; \
	62910	__ret_743 = vsetq_lane_bf16(vgetq_lane_bf16(__s2_743, __p3_743), __s0_743, __p1_743); \
	62911	__ret_743; \
63884	62912	})
63885	62913	#else
63886		#define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63887		uint8x16x4_t __s1 = __p1; \
63888		uint8x16x4_t __rev1; \
63889		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63890		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63891		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63892		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63893		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 48); \
	62914	#define vcopyq_laneq_bf16(__p0_744, __p1_744, __p2_744, __p3_744) __extension__ ({ \
	62915	bfloat16x8_t __ret_744; \
	62916	bfloat16x8_t __s0_744 = __p0_744; \
	62917	bfloat16x8_t __s2_744 = __p2_744; \
	62918	bfloat16x8_t __rev0_744; __rev0_744 = __builtin_shufflevector(__s0_744, __s0_744, 7, 6, 5, 4, 3, 2, 1, 0); \
	62919	bfloat16x8_t __rev2_744; __rev2_744 = __builtin_shufflevector(__s2_744, __s2_744, 7, 6, 5, 4, 3, 2, 1, 0); \
	62920	__ret_744 = __noswap_vsetq_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_744, __p3_744), __rev0_744, __p1_744); \
	62921	__ret_744 = __builtin_shufflevector(__ret_744, __ret_744, 7, 6, 5, 4, 3, 2, 1, 0); \
	62922	__ret_744; \
63894	62923	})
63895	62924	#endif
63896	62925
63897	62926	#ifdef __LITTLE_ENDIAN__
63898		#define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63899		uint64x2x4_t __s1 = __p1; \
63900		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 51); \
	62927	#define vcopy_laneq_bf16(__p0_745, __p1_745, __p2_745, __p3_745) __extension__ ({ \
	62928	bfloat16x4_t __ret_745; \
	62929	bfloat16x4_t __s0_745 = __p0_745; \
	62930	bfloat16x8_t __s2_745 = __p2_745; \
	62931	__ret_745 = vset_lane_bf16(vgetq_lane_bf16(__s2_745, __p3_745), __s0_745, __p1_745); \
	62932	__ret_745; \
63901	62933	})
63902	62934	#else
63903		#define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63904		uint64x2x4_t __s1 = __p1; \
63905		uint64x2x4_t __rev1; \
63906		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63907		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63908		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63909		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63910		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 51); \
	62935	#define vcopy_laneq_bf16(__p0_746, __p1_746, __p2_746, __p3_746) __extension__ ({ \
	62936	bfloat16x4_t __ret_746; \
	62937	bfloat16x4_t __s0_746 = __p0_746; \
	62938	bfloat16x8_t __s2_746 = __p2_746; \
	62939	bfloat16x4_t __rev0_746; __rev0_746 = __builtin_shufflevector(__s0_746, __s0_746, 3, 2, 1, 0); \
	62940	bfloat16x8_t __rev2_746; __rev2_746 = __builtin_shufflevector(__s2_746, __s2_746, 7, 6, 5, 4, 3, 2, 1, 0); \
	62941	__ret_746 = __noswap_vset_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_746, __p3_746), __rev0_746, __p1_746); \
	62942	__ret_746 = __builtin_shufflevector(__ret_746, __ret_746, 3, 2, 1, 0); \
	62943	__ret_746; \
63911	62944	})
63912	62945	#endif
63913	62946
63914	62947	#ifdef __LITTLE_ENDIAN__
63915		#define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63916		int8x16x4_t __s1 = __p1; \
63917		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 32); \
63918		})
	62948	__ai __attribute__((target("bf16"))) bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
	62949	bfloat16x4_t __ret;
	62950	__ret = vget_low_bf16(__a64_vcvtq_low_bf16_f32(__p0));
	62951	return __ret;
	62952	}
63919	62953	#else
63920		#define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63921		int8x16x4_t __s1 = __p1; \
63922		int8x16x4_t __rev1; \
63923		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63924		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63925		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63926		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
63927		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 32); \
63928		})
	62954	__ai __attribute__((target("bf16"))) bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
	62955	bfloat16x4_t __ret;
	62956	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62957	__ret = __noswap_vget_low_bf16(__noswap___a64_vcvtq_low_bf16_f32(__rev0));
	62958	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	62959	return __ret;
	62960	}
63929	62961	#endif
63930	62962
63931	62963	#ifdef __LITTLE_ENDIAN__
63932		#define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63933		float64x2x4_t __s1 = __p1; \
63934		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 42); \
63935		})
	62964	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
	62965	bfloat16x8_t __ret;
	62966	__ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_f32((int8x16_t)__p0, (int8x16_t)__p1, 43);
	62967	return __ret;
	62968	}
63936	62969	#else
63937		#define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63938		float64x2x4_t __s1 = __p1; \
63939		float64x2x4_t __rev1; \
63940		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63941		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63942		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63943		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63944		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 42); \
63945		})
	62970	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
	62971	bfloat16x8_t __ret;
	62972	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	62973	float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	62974	__ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_f32((int8x16_t)__rev0, (int8x16_t)__rev1, 43);
	62975	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62976	return __ret;
	62977	}
63946	62978	#endif
63947	62979
63948	62980	#ifdef __LITTLE_ENDIAN__
63949		#define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63950		int64x2x4_t __s1 = __p1; \
63951		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], (int8x16_t)__s1.val[3], __p2, 35); \
63952		})
	62981	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
	62982	bfloat16x8_t __ret;
	62983	__ret = __a64_vcvtq_low_bf16_f32(__p0);
	62984	return __ret;
	62985	}
63953	62986	#else
63954		#define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63955		int64x2x4_t __s1 = __p1; \
63956		int64x2x4_t __rev1; \
63957		__rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
63958		__rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
63959		__rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
63960		__rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
63961		__builtin_neon_vst4q_lane_v(__p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], (int8x16_t)__rev1.val[3], __p2, 35); \
63962		})
	62987	__ai __attribute__((target("bf16"))) bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
	62988	bfloat16x8_t __ret;
	62989	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	62990	__ret = __noswap___a64_vcvtq_low_bf16_f32(__rev0);
	62991	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	62992	return __ret;
	62993	}
63963	62994	#endif
63964	62995
63965		#define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63966		uint64x1x4_t __s1 = __p1; \
63967		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 19); \
63968		})
63969		#define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63970		float64x1x4_t __s1 = __p1; \
63971		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 10); \
63972		})
63973		#define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63974		int64x1x4_t __s1 = __p1; \
63975		__builtin_neon_vst4_lane_v(__p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], (int8x8_t)__s1.val[3], __p2, 3); \
63976		})
63977		#define vstrq_p128(__p0, __p1) __extension__ ({ \
63978		poly128_t __s1 = __p1; \
63979		__builtin_neon_vstrq_p128(__p0, __s1); \
63980		})
63981		__ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
63982		uint64_t __ret;
63983		__ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
	62996	__ai __attribute__((target("bf16"))) poly8x8_t vreinterpret_p8_bf16(bfloat16x4_t __p0) {
	62997	poly8x8_t __ret;
	62998	__ret = (poly8x8_t)(__p0);
63984	62999	return __ret;
63985	63000	}
63986		__ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
63987		int64_t __ret;
63988		__ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
	63001	__ai __attribute__((target("bf16"))) poly64x1_t vreinterpret_p64_bf16(bfloat16x4_t __p0) {
	63002	poly64x1_t __ret;
	63003	__ret = (poly64x1_t)(__p0);
63989	63004	return __ret;
63990	63005	}
63991		#ifdef __LITTLE_ENDIAN__
63992		__ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
63993		float64x2_t __ret;
63994		__ret = __p0 - __p1;
	63006	__ai __attribute__((target("bf16"))) poly16x4_t vreinterpret_p16_bf16(bfloat16x4_t __p0) {
	63007	poly16x4_t __ret;
	63008	__ret = (poly16x4_t)(__p0);
63995	63009	return __ret;
63996	63010	}
63997		#else
63998		__ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
	63011	__ai __attribute__((target("bf16"))) poly8x16_t vreinterpretq_p8_bf16(bfloat16x8_t __p0) {
	63012	poly8x16_t __ret;
	63013	__ret = (poly8x16_t)(__p0);
	63014	return __ret;
	63015	}
	63016	__ai __attribute__((target("bf16"))) poly128_t vreinterpretq_p128_bf16(bfloat16x8_t __p0) {
	63017	poly128_t __ret;
	63018	__ret = (poly128_t)(__p0);
	63019	return __ret;
	63020	}
	63021	__ai __attribute__((target("bf16"))) poly64x2_t vreinterpretq_p64_bf16(bfloat16x8_t __p0) {
	63022	poly64x2_t __ret;
	63023	__ret = (poly64x2_t)(__p0);
	63024	return __ret;
	63025	}
	63026	__ai __attribute__((target("bf16"))) poly16x8_t vreinterpretq_p16_bf16(bfloat16x8_t __p0) {
	63027	poly16x8_t __ret;
	63028	__ret = (poly16x8_t)(__p0);
	63029	return __ret;
	63030	}
	63031	__ai __attribute__((target("bf16"))) uint8x16_t vreinterpretq_u8_bf16(bfloat16x8_t __p0) {
	63032	uint8x16_t __ret;
	63033	__ret = (uint8x16_t)(__p0);
	63034	return __ret;
	63035	}
	63036	__ai __attribute__((target("bf16"))) uint32x4_t vreinterpretq_u32_bf16(bfloat16x8_t __p0) {
	63037	uint32x4_t __ret;
	63038	__ret = (uint32x4_t)(__p0);
	63039	return __ret;
	63040	}
	63041	__ai __attribute__((target("bf16"))) uint64x2_t vreinterpretq_u64_bf16(bfloat16x8_t __p0) {
	63042	uint64x2_t __ret;
	63043	__ret = (uint64x2_t)(__p0);
	63044	return __ret;
	63045	}
	63046	__ai __attribute__((target("bf16"))) uint16x8_t vreinterpretq_u16_bf16(bfloat16x8_t __p0) {
	63047	uint16x8_t __ret;
	63048	__ret = (uint16x8_t)(__p0);
	63049	return __ret;
	63050	}
	63051	__ai __attribute__((target("bf16"))) int8x16_t vreinterpretq_s8_bf16(bfloat16x8_t __p0) {
	63052	int8x16_t __ret;
	63053	__ret = (int8x16_t)(__p0);
	63054	return __ret;
	63055	}
	63056	__ai __attribute__((target("bf16"))) float64x2_t vreinterpretq_f64_bf16(bfloat16x8_t __p0) {
63999	63057	float64x2_t __ret;
64000		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64001		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64002		__ret = __rev0 - __rev1;
64003		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	63058	__ret = (float64x2_t)(__p0);
64004	63059	return __ret;
64005	63060	}
64006		#endif
64007
64008		__ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
	63061	__ai __attribute__((target("bf16"))) float32x4_t vreinterpretq_f32_bf16(bfloat16x8_t __p0) {
	63062	float32x4_t __ret;
	63063	__ret = (float32x4_t)(__p0);
	63064	return __ret;
	63065	}
	63066	__ai __attribute__((target("bf16"))) float16x8_t vreinterpretq_f16_bf16(bfloat16x8_t __p0) {
	63067	float16x8_t __ret;
	63068	__ret = (float16x8_t)(__p0);
	63069	return __ret;
	63070	}
	63071	__ai __attribute__((target("bf16"))) int32x4_t vreinterpretq_s32_bf16(bfloat16x8_t __p0) {
	63072	int32x4_t __ret;
	63073	__ret = (int32x4_t)(__p0);
	63074	return __ret;
	63075	}
	63076	__ai __attribute__((target("bf16"))) int64x2_t vreinterpretq_s64_bf16(bfloat16x8_t __p0) {
	63077	int64x2_t __ret;
	63078	__ret = (int64x2_t)(__p0);
	63079	return __ret;
	63080	}
	63081	__ai __attribute__((target("bf16"))) int16x8_t vreinterpretq_s16_bf16(bfloat16x8_t __p0) {
	63082	int16x8_t __ret;
	63083	__ret = (int16x8_t)(__p0);
	63084	return __ret;
	63085	}
	63086	__ai __attribute__((target("bf16"))) uint8x8_t vreinterpret_u8_bf16(bfloat16x4_t __p0) {
	63087	uint8x8_t __ret;
	63088	__ret = (uint8x8_t)(__p0);
	63089	return __ret;
	63090	}
	63091	__ai __attribute__((target("bf16"))) uint32x2_t vreinterpret_u32_bf16(bfloat16x4_t __p0) {
	63092	uint32x2_t __ret;
	63093	__ret = (uint32x2_t)(__p0);
	63094	return __ret;
	63095	}
	63096	__ai __attribute__((target("bf16"))) uint64x1_t vreinterpret_u64_bf16(bfloat16x4_t __p0) {
	63097	uint64x1_t __ret;
	63098	__ret = (uint64x1_t)(__p0);
	63099	return __ret;
	63100	}
	63101	__ai __attribute__((target("bf16"))) uint16x4_t vreinterpret_u16_bf16(bfloat16x4_t __p0) {
	63102	uint16x4_t __ret;
	63103	__ret = (uint16x4_t)(__p0);
	63104	return __ret;
	63105	}
	63106	__ai __attribute__((target("bf16"))) int8x8_t vreinterpret_s8_bf16(bfloat16x4_t __p0) {
	63107	int8x8_t __ret;
	63108	__ret = (int8x8_t)(__p0);
	63109	return __ret;
	63110	}
	63111	__ai __attribute__((target("bf16"))) float64x1_t vreinterpret_f64_bf16(bfloat16x4_t __p0) {
64009	63112	float64x1_t __ret;
64010		__ret = __p0 - __p1;
	63113	__ret = (float64x1_t)(__p0);
64011	63114	return __ret;
64012	63115	}
64013		#ifdef __LITTLE_ENDIAN__
64014		__ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
64015		uint16x8_t __ret;
64016		__ret = vcombine_u16(__p0, vsubhn_u32(__p1, __p2));
	63116	__ai __attribute__((target("bf16"))) float32x2_t vreinterpret_f32_bf16(bfloat16x4_t __p0) {
	63117	float32x2_t __ret;
	63118	__ret = (float32x2_t)(__p0);
64017	63119	return __ret;
64018	63120	}
64019		#else
64020		__ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
64021		uint16x8_t __ret;
64022		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64023		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64024		uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
64025		__ret = __noswap_vcombine_u16(__rev0, __noswap_vsubhn_u32(__rev1, __rev2));
64026		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	63121	__ai __attribute__((target("bf16"))) float16x4_t vreinterpret_f16_bf16(bfloat16x4_t __p0) {
	63122	float16x4_t __ret;
	63123	__ret = (float16x4_t)(__p0);
64027	63124	return __ret;
64028	63125	}
64029		#endif
64030
64031		#ifdef __LITTLE_ENDIAN__
64032		__ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
64033		uint32x4_t __ret;
64034		__ret = vcombine_u32(__p0, vsubhn_u64(__p1, __p2));
	63126	__ai __attribute__((target("bf16"))) int32x2_t vreinterpret_s32_bf16(bfloat16x4_t __p0) {
	63127	int32x2_t __ret;
	63128	__ret = (int32x2_t)(__p0);
64035	63129	return __ret;
64036	63130	}
64037		#else
64038		__ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
64039		uint32x4_t __ret;
64040		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64041		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64042		uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
64043		__ret = __noswap_vcombine_u32(__rev0, __noswap_vsubhn_u64(__rev1, __rev2));
64044		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	63131	__ai __attribute__((target("bf16"))) int64x1_t vreinterpret_s64_bf16(bfloat16x4_t __p0) {
	63132	int64x1_t __ret;
	63133	__ret = (int64x1_t)(__p0);
	63134	return __ret;
	63135	}
	63136	__ai __attribute__((target("bf16"))) int16x4_t vreinterpret_s16_bf16(bfloat16x4_t __p0) {
	63137	int16x4_t __ret;
	63138	__ret = (int16x4_t)(__p0);
	63139	return __ret;
	63140	}
	63141	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p8(poly8x16_t __p0) {
	63142	bfloat16x8_t __ret;
	63143	__ret = (bfloat16x8_t)(__p0);
	63144	return __ret;
	63145	}
	63146	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p128(poly128_t __p0) {
	63147	bfloat16x8_t __ret;
	63148	__ret = (bfloat16x8_t)(__p0);
	63149	return __ret;
	63150	}
	63151	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p64(poly64x2_t __p0) {
	63152	bfloat16x8_t __ret;
	63153	__ret = (bfloat16x8_t)(__p0);
	63154	return __ret;
	63155	}
	63156	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_p16(poly16x8_t __p0) {
	63157	bfloat16x8_t __ret;
	63158	__ret = (bfloat16x8_t)(__p0);
	63159	return __ret;
	63160	}
	63161	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u8(uint8x16_t __p0) {
	63162	bfloat16x8_t __ret;
	63163	__ret = (bfloat16x8_t)(__p0);
	63164	return __ret;
	63165	}
	63166	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u32(uint32x4_t __p0) {
	63167	bfloat16x8_t __ret;
	63168	__ret = (bfloat16x8_t)(__p0);
	63169	return __ret;
	63170	}
	63171	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u64(uint64x2_t __p0) {
	63172	bfloat16x8_t __ret;
	63173	__ret = (bfloat16x8_t)(__p0);
	63174	return __ret;
	63175	}
	63176	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_u16(uint16x8_t __p0) {
	63177	bfloat16x8_t __ret;
	63178	__ret = (bfloat16x8_t)(__p0);
	63179	return __ret;
	63180	}
	63181	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s8(int8x16_t __p0) {
	63182	bfloat16x8_t __ret;
	63183	__ret = (bfloat16x8_t)(__p0);
	63184	return __ret;
	63185	}
	63186	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_f64(float64x2_t __p0) {
	63187	bfloat16x8_t __ret;
	63188	__ret = (bfloat16x8_t)(__p0);
	63189	return __ret;
	63190	}
	63191	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_f32(float32x4_t __p0) {
	63192	bfloat16x8_t __ret;
	63193	__ret = (bfloat16x8_t)(__p0);
	63194	return __ret;
	63195	}
	63196	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_f16(float16x8_t __p0) {
	63197	bfloat16x8_t __ret;
	63198	__ret = (bfloat16x8_t)(__p0);
	63199	return __ret;
	63200	}
	63201	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s32(int32x4_t __p0) {
	63202	bfloat16x8_t __ret;
	63203	__ret = (bfloat16x8_t)(__p0);
	63204	return __ret;
	63205	}
	63206	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s64(int64x2_t __p0) {
	63207	bfloat16x8_t __ret;
	63208	__ret = (bfloat16x8_t)(__p0);
	63209	return __ret;
	63210	}
	63211	__ai __attribute__((target("bf16"))) bfloat16x8_t vreinterpretq_bf16_s16(int16x8_t __p0) {
	63212	bfloat16x8_t __ret;
	63213	__ret = (bfloat16x8_t)(__p0);
	63214	return __ret;
	63215	}
	63216	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_p8(poly8x8_t __p0) {
	63217	bfloat16x4_t __ret;
	63218	__ret = (bfloat16x4_t)(__p0);
	63219	return __ret;
	63220	}
	63221	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_p64(poly64x1_t __p0) {
	63222	bfloat16x4_t __ret;
	63223	__ret = (bfloat16x4_t)(__p0);
	63224	return __ret;
	63225	}
	63226	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_p16(poly16x4_t __p0) {
	63227	bfloat16x4_t __ret;
	63228	__ret = (bfloat16x4_t)(__p0);
	63229	return __ret;
	63230	}
	63231	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u8(uint8x8_t __p0) {
	63232	bfloat16x4_t __ret;
	63233	__ret = (bfloat16x4_t)(__p0);
	63234	return __ret;
	63235	}
	63236	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u32(uint32x2_t __p0) {
	63237	bfloat16x4_t __ret;
	63238	__ret = (bfloat16x4_t)(__p0);
	63239	return __ret;
	63240	}
	63241	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u64(uint64x1_t __p0) {
	63242	bfloat16x4_t __ret;
	63243	__ret = (bfloat16x4_t)(__p0);
	63244	return __ret;
	63245	}
	63246	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_u16(uint16x4_t __p0) {
	63247	bfloat16x4_t __ret;
	63248	__ret = (bfloat16x4_t)(__p0);
	63249	return __ret;
	63250	}
	63251	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s8(int8x8_t __p0) {
	63252	bfloat16x4_t __ret;
	63253	__ret = (bfloat16x4_t)(__p0);
	63254	return __ret;
	63255	}
	63256	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_f64(float64x1_t __p0) {
	63257	bfloat16x4_t __ret;
	63258	__ret = (bfloat16x4_t)(__p0);
	63259	return __ret;
	63260	}
	63261	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_f32(float32x2_t __p0) {
	63262	bfloat16x4_t __ret;
	63263	__ret = (bfloat16x4_t)(__p0);
	63264	return __ret;
	63265	}
	63266	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_f16(float16x4_t __p0) {
	63267	bfloat16x4_t __ret;
	63268	__ret = (bfloat16x4_t)(__p0);
	63269	return __ret;
	63270	}
	63271	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s32(int32x2_t __p0) {
	63272	bfloat16x4_t __ret;
	63273	__ret = (bfloat16x4_t)(__p0);
	63274	return __ret;
	63275	}
	63276	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s64(int64x1_t __p0) {
	63277	bfloat16x4_t __ret;
	63278	__ret = (bfloat16x4_t)(__p0);
	63279	return __ret;
	63280	}
	63281	__ai __attribute__((target("bf16"))) bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
	63282	bfloat16x4_t __ret;
	63283	__ret = (bfloat16x4_t)(__p0);
64045	63284	return __ret;
64046	63285	}
	63286	#ifdef __LITTLE_ENDIAN__
	63287	#define vdotq_laneq_u32(__p0_747, __p1_747, __p2_747, __p3_747) __extension__ ({ \
	63288	uint32x4_t __ret_747; \
	63289	uint32x4_t __s0_747 = __p0_747; \
	63290	uint8x16_t __s1_747 = __p1_747; \
	63291	uint8x16_t __s2_747 = __p2_747; \
	63292	uint8x16_t __reint_747 = __s2_747; \
	63293	uint32x4_t __reint1_747 = splatq_laneq_u32((uint32x4_t ) &__reint_747, __p3_747); \
	63294	__ret_747 = vdotq_u32(__s0_747, __s1_747, (uint8x16_t ) &__reint1_747); \
	63295	__ret_747; \
	63296	})
	63297	#else
	63298	#define vdotq_laneq_u32(__p0_748, __p1_748, __p2_748, __p3_748) __extension__ ({ \
	63299	uint32x4_t __ret_748; \
	63300	uint32x4_t __s0_748 = __p0_748; \
	63301	uint8x16_t __s1_748 = __p1_748; \
	63302	uint8x16_t __s2_748 = __p2_748; \
	63303	uint32x4_t __rev0_748; __rev0_748 = __builtin_shufflevector(__s0_748, __s0_748, 3, 2, 1, 0); \
	63304	uint8x16_t __rev1_748; __rev1_748 = __builtin_shufflevector(__s1_748, __s1_748, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	63305	uint8x16_t __rev2_748; __rev2_748 = __builtin_shufflevector(__s2_748, __s2_748, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	63306	uint8x16_t __reint_748 = __rev2_748; \
	63307	uint32x4_t __reint1_748 = __noswap_splatq_laneq_u32((uint32x4_t ) &__reint_748, __p3_748); \
	63308	__ret_748 = __noswap_vdotq_u32(__rev0_748, __rev1_748, (uint8x16_t ) &__reint1_748); \
	63309	__ret_748 = __builtin_shufflevector(__ret_748, __ret_748, 3, 2, 1, 0); \
	63310	__ret_748; \
	63311	})
64047	63312	#endif
64048	63313
64049	63314	#ifdef __LITTLE_ENDIAN__
64050		__ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
64051		uint8x16_t __ret;
64052		__ret = vcombine_u8(__p0, vsubhn_u16(__p1, __p2));
64053		return __ret;
64054		}
	63315	#define vdotq_laneq_s32(__p0_749, __p1_749, __p2_749, __p3_749) __extension__ ({ \
	63316	int32x4_t __ret_749; \
	63317	int32x4_t __s0_749 = __p0_749; \
	63318	int8x16_t __s1_749 = __p1_749; \
	63319	int8x16_t __s2_749 = __p2_749; \
	63320	int8x16_t __reint_749 = __s2_749; \
	63321	int32x4_t __reint1_749 = splatq_laneq_s32((int32x4_t ) &__reint_749, __p3_749); \
	63322	__ret_749 = vdotq_s32(__s0_749, __s1_749, (int8x16_t ) &__reint1_749); \
	63323	__ret_749; \
	63324	})
	63325	#else
	63326	#define vdotq_laneq_s32(__p0_750, __p1_750, __p2_750, __p3_750) __extension__ ({ \
	63327	int32x4_t __ret_750; \
	63328	int32x4_t __s0_750 = __p0_750; \
	63329	int8x16_t __s1_750 = __p1_750; \
	63330	int8x16_t __s2_750 = __p2_750; \
	63331	int32x4_t __rev0_750; __rev0_750 = __builtin_shufflevector(__s0_750, __s0_750, 3, 2, 1, 0); \
	63332	int8x16_t __rev1_750; __rev1_750 = __builtin_shufflevector(__s1_750, __s1_750, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	63333	int8x16_t __rev2_750; __rev2_750 = __builtin_shufflevector(__s2_750, __s2_750, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	63334	int8x16_t __reint_750 = __rev2_750; \
	63335	int32x4_t __reint1_750 = __noswap_splatq_laneq_s32((int32x4_t ) &__reint_750, __p3_750); \
	63336	__ret_750 = __noswap_vdotq_s32(__rev0_750, __rev1_750, (int8x16_t ) &__reint1_750); \
	63337	__ret_750 = __builtin_shufflevector(__ret_750, __ret_750, 3, 2, 1, 0); \
	63338	__ret_750; \
	63339	})
	63340	#endif
	63341
	63342	#ifdef __LITTLE_ENDIAN__
	63343	#define vdot_laneq_u32(__p0_751, __p1_751, __p2_751, __p3_751) __extension__ ({ \
	63344	uint32x2_t __ret_751; \
	63345	uint32x2_t __s0_751 = __p0_751; \
	63346	uint8x8_t __s1_751 = __p1_751; \
	63347	uint8x16_t __s2_751 = __p2_751; \
	63348	uint8x16_t __reint_751 = __s2_751; \
	63349	uint32x2_t __reint1_751 = splat_laneq_u32((uint32x4_t ) &__reint_751, __p3_751); \
	63350	__ret_751 = vdot_u32(__s0_751, __s1_751, (uint8x8_t ) &__reint1_751); \
	63351	__ret_751; \
	63352	})
64055	63353	#else
64056		__ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
64057		uint8x16_t __ret;
64058		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64059		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64060		uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
64061		__ret = __noswap_vcombine_u8(__rev0, __noswap_vsubhn_u16(__rev1, __rev2));
64062		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64063		return __ret;
64064		}
	63354	#define vdot_laneq_u32(__p0_752, __p1_752, __p2_752, __p3_752) __extension__ ({ \
	63355	uint32x2_t __ret_752; \
	63356	uint32x2_t __s0_752 = __p0_752; \
	63357	uint8x8_t __s1_752 = __p1_752; \
	63358	uint8x16_t __s2_752 = __p2_752; \
	63359	uint32x2_t __rev0_752; __rev0_752 = __builtin_shufflevector(__s0_752, __s0_752, 1, 0); \
	63360	uint8x8_t __rev1_752; __rev1_752 = __builtin_shufflevector(__s1_752, __s1_752, 7, 6, 5, 4, 3, 2, 1, 0); \
	63361	uint8x16_t __rev2_752; __rev2_752 = __builtin_shufflevector(__s2_752, __s2_752, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	63362	uint8x16_t __reint_752 = __rev2_752; \
	63363	uint32x2_t __reint1_752 = __noswap_splat_laneq_u32((uint32x4_t ) &__reint_752, __p3_752); \
	63364	__ret_752 = __noswap_vdot_u32(__rev0_752, __rev1_752, (uint8x8_t ) &__reint1_752); \
	63365	__ret_752 = __builtin_shufflevector(__ret_752, __ret_752, 1, 0); \
	63366	__ret_752; \
	63367	})
64065	63368	#endif
64066	63369
64067	63370	#ifdef __LITTLE_ENDIAN__
64068		__ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
64069		int16x8_t __ret;
64070		__ret = vcombine_s16(__p0, vsubhn_s32(__p1, __p2));
64071		return __ret;
64072		}
	63371	#define vdot_laneq_s32(__p0_753, __p1_753, __p2_753, __p3_753) __extension__ ({ \
	63372	int32x2_t __ret_753; \
	63373	int32x2_t __s0_753 = __p0_753; \
	63374	int8x8_t __s1_753 = __p1_753; \
	63375	int8x16_t __s2_753 = __p2_753; \
	63376	int8x16_t __reint_753 = __s2_753; \
	63377	int32x2_t __reint1_753 = splat_laneq_s32((int32x4_t ) &__reint_753, __p3_753); \
	63378	__ret_753 = vdot_s32(__s0_753, __s1_753, (int8x8_t ) &__reint1_753); \
	63379	__ret_753; \
	63380	})
64073	63381	#else
64074		__ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
64075		int16x8_t __ret;
64076		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64077		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64078		int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
64079		__ret = __noswap_vcombine_s16(__rev0, __noswap_vsubhn_s32(__rev1, __rev2));
64080		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64081		return __ret;
64082		}
	63382	#define vdot_laneq_s32(__p0_754, __p1_754, __p2_754, __p3_754) __extension__ ({ \
	63383	int32x2_t __ret_754; \
	63384	int32x2_t __s0_754 = __p0_754; \
	63385	int8x8_t __s1_754 = __p1_754; \
	63386	int8x16_t __s2_754 = __p2_754; \
	63387	int32x2_t __rev0_754; __rev0_754 = __builtin_shufflevector(__s0_754, __s0_754, 1, 0); \
	63388	int8x8_t __rev1_754; __rev1_754 = __builtin_shufflevector(__s1_754, __s1_754, 7, 6, 5, 4, 3, 2, 1, 0); \
	63389	int8x16_t __rev2_754; __rev2_754 = __builtin_shufflevector(__s2_754, __s2_754, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	63390	int8x16_t __reint_754 = __rev2_754; \
	63391	int32x2_t __reint1_754 = __noswap_splat_laneq_s32((int32x4_t ) &__reint_754, __p3_754); \
	63392	__ret_754 = __noswap_vdot_s32(__rev0_754, __rev1_754, (int8x8_t ) &__reint1_754); \
	63393	__ret_754 = __builtin_shufflevector(__ret_754, __ret_754, 1, 0); \
	63394	__ret_754; \
	63395	})
64083	63396	#endif
64084	63397
64085	63398	#ifdef __LITTLE_ENDIAN__
64086		__ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
64087		int32x4_t __ret;
64088		__ret = vcombine_s32(__p0, vsubhn_s64(__p1, __p2));
	63399	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63400	float32x4_t __ret;
	63401	__ret = (float32x4_t) __builtin_neon_vfmlalq_high_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
64089	63402	return __ret;
64090	63403	}
64091	63404	#else
64092		__ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
64093		int32x4_t __ret;
64094		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64095		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64096		int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
64097		__ret = __noswap_vcombine_s32(__rev0, __noswap_vsubhn_s64(__rev1, __rev2));
	63405	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63406	float32x4_t __ret;
	63407	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	63408	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	63409	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	63410	__ret = (float32x4_t) __builtin_neon_vfmlalq_high_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
64098	63411	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64099	63412	return __ret;
64100	63413	}
	63414	__ai __attribute__((target("fp16fml"))) float32x4_t __noswap_vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63415	float32x4_t __ret;
	63416	__ret = (float32x4_t) __builtin_neon_vfmlalq_high_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	63417	return __ret;
	63418	}
64101	63419	#endif
64102	63420
64103	63421	#ifdef __LITTLE_ENDIAN__
64104		__ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
64105		int8x16_t __ret;
64106		__ret = vcombine_s8(__p0, vsubhn_s16(__p1, __p2));
	63422	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63423	float32x2_t __ret;
	63424	__ret = (float32x2_t) __builtin_neon_vfmlal_high_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
64107	63425	return __ret;
64108	63426	}
64109	63427	#else
64110		__ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
64111		int8x16_t __ret;
64112		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64113		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64114		int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
64115		__ret = __noswap_vcombine_s8(__rev0, __noswap_vsubhn_s16(__rev1, __rev2));
64116		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	63428	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63429	float32x2_t __ret;
	63430	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	63431	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	63432	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	63433	__ret = (float32x2_t) __builtin_neon_vfmlal_high_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	63434	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	63435	return __ret;
	63436	}
	63437	__ai __attribute__((target("fp16fml"))) float32x2_t __noswap_vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63438	float32x2_t __ret;
	63439	__ret = (float32x2_t) __builtin_neon_vfmlal_high_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
64117	63440	return __ret;
64118	63441	}
64119	63442	#endif
64120	63443
64121	63444	#ifdef __LITTLE_ENDIAN__
64122		__ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
64123		uint16x8_t __ret;
64124		__ret = vmovl_high_u8(__p0) - vmovl_high_u8(__p1);
	63445	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63446	float32x4_t __ret;
	63447	__ret = (float32x4_t) __builtin_neon_vfmlalq_low_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
64125	63448	return __ret;
64126	63449	}
64127	63450	#else
64128		__ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
64129		uint16x8_t __ret;
64130		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64131		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64132		__ret = __noswap_vmovl_high_u8(__rev0) - __noswap_vmovl_high_u8(__rev1);
64133		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	63451	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63452	float32x4_t __ret;
	63453	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	63454	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	63455	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	63456	__ret = (float32x4_t) __builtin_neon_vfmlalq_low_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
	63457	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	63458	return __ret;
	63459	}
	63460	__ai __attribute__((target("fp16fml"))) float32x4_t __noswap_vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63461	float32x4_t __ret;
	63462	__ret = (float32x4_t) __builtin_neon_vfmlalq_low_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
64134	63463	return __ret;
64135	63464	}
64136	63465	#endif
64137	63466
64138	63467	#ifdef __LITTLE_ENDIAN__
64139		__ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
64140		uint64x2_t __ret;
64141		__ret = vmovl_high_u32(__p0) - vmovl_high_u32(__p1);
	63468	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63469	float32x2_t __ret;
	63470	__ret = (float32x2_t) __builtin_neon_vfmlal_low_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
64142	63471	return __ret;
64143	63472	}
64144	63473	#else
64145		__ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
64146		uint64x2_t __ret;
64147		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64148		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64149		__ret = __noswap_vmovl_high_u32(__rev0) - __noswap_vmovl_high_u32(__rev1);
	63474	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63475	float32x2_t __ret;
	63476	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	63477	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	63478	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	63479	__ret = (float32x2_t) __builtin_neon_vfmlal_low_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
64150	63480	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64151	63481	return __ret;
64152	63482	}
	63483	__ai __attribute__((target("fp16fml"))) float32x2_t __noswap_vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63484	float32x2_t __ret;
	63485	__ret = (float32x2_t) __builtin_neon_vfmlal_low_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	63486	return __ret;
	63487	}
64153	63488	#endif
64154	63489
64155	63490	#ifdef __LITTLE_ENDIAN__
64156		__ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
64157		uint32x4_t __ret;
64158		__ret = vmovl_high_u16(__p0) - vmovl_high_u16(__p1);
	63491	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63492	float32x4_t __ret;
	63493	__ret = (float32x4_t) __builtin_neon_vfmlslq_high_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
64159	63494	return __ret;
64160	63495	}
64161	63496	#else
64162		__ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
64163		uint32x4_t __ret;
64164		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64165		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64166		__ret = __noswap_vmovl_high_u16(__rev0) - __noswap_vmovl_high_u16(__rev1);
	63497	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63498	float32x4_t __ret;
	63499	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	63500	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	63501	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	63502	__ret = (float32x4_t) __builtin_neon_vfmlslq_high_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
64167	63503	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64168	63504	return __ret;
64169	63505	}
64170		#endif
64171
64172		#ifdef __LITTLE_ENDIAN__
64173		__ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
64174		int16x8_t __ret;
64175		__ret = vmovl_high_s8(__p0) - vmovl_high_s8(__p1);
64176		return __ret;
64177		}
64178		#else
64179		__ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
64180		int16x8_t __ret;
64181		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64182		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64183		__ret = __noswap_vmovl_high_s8(__rev0) - __noswap_vmovl_high_s8(__rev1);
64184		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	63506	__ai __attribute__((target("fp16fml"))) float32x4_t __noswap_vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63507	float32x4_t __ret;
	63508	__ret = (float32x4_t) __builtin_neon_vfmlslq_high_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
64185	63509	return __ret;
64186	63510	}
64187	63511	#endif
64188	63512
64189	63513	#ifdef __LITTLE_ENDIAN__
64190		__ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
64191		int64x2_t __ret;
64192		__ret = vmovl_high_s32(__p0) - vmovl_high_s32(__p1);
	63514	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63515	float32x2_t __ret;
	63516	__ret = (float32x2_t) __builtin_neon_vfmlsl_high_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
64193	63517	return __ret;
64194	63518	}
64195	63519	#else
64196		__ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
64197		int64x2_t __ret;
64198		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64199		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64200		__ret = __noswap_vmovl_high_s32(__rev0) - __noswap_vmovl_high_s32(__rev1);
	63520	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63521	float32x2_t __ret;
	63522	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	63523	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	63524	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	63525	__ret = (float32x2_t) __builtin_neon_vfmlsl_high_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
64201	63526	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64202	63527	return __ret;
64203	63528	}
	63529	__ai __attribute__((target("fp16fml"))) float32x2_t __noswap_vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63530	float32x2_t __ret;
	63531	__ret = (float32x2_t) __builtin_neon_vfmlsl_high_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
	63532	return __ret;
	63533	}
64204	63534	#endif
64205	63535
64206	63536	#ifdef __LITTLE_ENDIAN__
64207		__ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
64208		int32x4_t __ret;
64209		__ret = vmovl_high_s16(__p0) - vmovl_high_s16(__p1);
	63537	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63538	float32x4_t __ret;
	63539	__ret = (float32x4_t) __builtin_neon_vfmlslq_low_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
64210	63540	return __ret;
64211	63541	}
64212	63542	#else
64213		__ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
64214		int32x4_t __ret;
64215		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64216		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64217		__ret = __noswap_vmovl_high_s16(__rev0) - __noswap_vmovl_high_s16(__rev1);
	63543	__ai __attribute__((target("fp16fml"))) float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63544	float32x4_t __ret;
	63545	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	63546	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	63547	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	63548	__ret = (float32x4_t) __builtin_neon_vfmlslq_low_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
64218	63549	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64219	63550	return __ret;
64220	63551	}
	63552	__ai __attribute__((target("fp16fml"))) float32x4_t __noswap_vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
	63553	float32x4_t __ret;
	63554	__ret = (float32x4_t) __builtin_neon_vfmlslq_low_f16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
	63555	return __ret;
	63556	}
64221	63557	#endif
64222	63558
64223	63559	#ifdef __LITTLE_ENDIAN__
64224		__ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
64225		uint16x8_t __ret;
64226		__ret = __p0 - vmovl_high_u8(__p1);
	63560	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63561	float32x2_t __ret;
	63562	__ret = (float32x2_t) __builtin_neon_vfmlsl_low_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
64227	63563	return __ret;
64228	63564	}
64229	63565	#else
64230		__ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
64231		uint16x8_t __ret;
64232		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64233		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64234		__ret = __rev0 - __noswap_vmovl_high_u8(__rev1);
64235		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	63566	__ai __attribute__((target("fp16fml"))) float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63567	float32x2_t __ret;
	63568	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	63569	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	63570	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	63571	__ret = (float32x2_t) __builtin_neon_vfmlsl_low_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
	63572	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	63573	return __ret;
	63574	}
	63575	__ai __attribute__((target("fp16fml"))) float32x2_t __noswap_vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
	63576	float32x2_t __ret;
	63577	__ret = (float32x2_t) __builtin_neon_vfmlsl_low_f16((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
64236	63578	return __ret;
64237	63579	}
64238	63580	#endif
64239	63581
64240	63582	#ifdef __LITTLE_ENDIAN__
64241		__ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
64242		uint64x2_t __ret;
64243		__ret = __p0 - vmovl_high_u32(__p1);
	63583	__ai __attribute__((target("fullfp16"))) float16x8_t vdivq_f16(float16x8_t __p0, float16x8_t __p1) {
	63584	float16x8_t __ret;
	63585	__ret = __p0 / __p1;
64244	63586	return __ret;
64245	63587	}
64246	63588	#else
64247		__ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
64248		uint64x2_t __ret;
64249		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64250		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64251		__ret = __rev0 - __noswap_vmovl_high_u32(__rev1);
64252		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	63589	__ai __attribute__((target("fullfp16"))) float16x8_t vdivq_f16(float16x8_t __p0, float16x8_t __p1) {
	63590	float16x8_t __ret;
	63591	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	63592	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	63593	__ret = __rev0 / __rev1;
	63594	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64253	63595	return __ret;
64254	63596	}
64255	63597	#endif
64256	63598
64257	63599	#ifdef __LITTLE_ENDIAN__
64258		__ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
64259		uint32x4_t __ret;
64260		__ret = __p0 - vmovl_high_u16(__p1);
	63600	__ai __attribute__((target("fullfp16"))) float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
	63601	float16x4_t __ret;
	63602	__ret = __p0 / __p1;
64261	63603	return __ret;
64262	63604	}
64263	63605	#else
64264		__ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
64265		uint32x4_t __ret;
64266		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64267		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64268		__ret = __rev0 - __noswap_vmovl_high_u16(__rev1);
	63606	__ai __attribute__((target("fullfp16"))) float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
	63607	float16x4_t __ret;
	63608	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	63609	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	63610	__ret = __rev0 / __rev1;
64269	63611	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64270	63612	return __ret;
64271	63613	}
64272	63614	#endif
64273	63615
64274	63616	#ifdef __LITTLE_ENDIAN__
64275		__ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
64276		int16x8_t __ret;
64277		__ret = __p0 - vmovl_high_s8(__p1);
64278		return __ret;
64279		}
	63617	#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
	63618	float16_t __ret; \
	63619	float16x4_t __s0 = __p0; \
	63620	__ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__s0, __p1); \
	63621	__ret; \
	63622	})
64280	63623	#else
64281		__ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
64282		int16x8_t __ret;
64283		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64284		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64285		__ret = __rev0 - __noswap_vmovl_high_s8(__rev1);
64286		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64287		return __ret;
64288		}
	63624	#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
	63625	float16_t __ret; \
	63626	float16x4_t __s0 = __p0; \
	63627	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	63628	__ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__rev0, __p1); \
	63629	__ret; \
	63630	})
64289	63631	#endif
64290	63632
64291	63633	#ifdef __LITTLE_ENDIAN__
64292		__ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
64293		int64x2_t __ret;
64294		__ret = __p0 - vmovl_high_s32(__p1);
64295		return __ret;
64296		}
	63634	#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
	63635	float16_t __ret; \
	63636	float16x8_t __s0 = __p0; \
	63637	__ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__s0, __p1); \
	63638	__ret; \
	63639	})
64297	63640	#else
64298		__ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
64299		int64x2_t __ret;
64300		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64301		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64302		__ret = __rev0 - __noswap_vmovl_high_s32(__rev1);
64303		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64304		return __ret;
64305		}
	63641	#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
	63642	float16_t __ret; \
	63643	float16x8_t __s0 = __p0; \
	63644	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	63645	__ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__rev0, __p1); \
	63646	__ret; \
	63647	})
64306	63648	#endif
64307	63649
64308	63650	#ifdef __LITTLE_ENDIAN__
64309		__ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
64310		int32x4_t __ret;
64311		__ret = __p0 - vmovl_high_s16(__p1);
64312		return __ret;
64313		}
	63651	#define vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63652	float16_t __ret; \
	63653	float16_t __s0 = __p0; \
	63654	float16_t __s1 = __p1; \
	63655	float16x4_t __s2 = __p2; \
	63656	__ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
	63657	__ret; \
	63658	})
64314	63659	#else
64315		__ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
64316		int32x4_t __ret;
64317		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64318		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64319		__ret = __rev0 - __noswap_vmovl_high_s16(__rev1);
64320		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64321		return __ret;
64322		}
	63660	#define vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63661	float16_t __ret; \
	63662	float16_t __s0 = __p0; \
	63663	float16_t __s1 = __p1; \
	63664	float16x4_t __s2 = __p2; \
	63665	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	63666	__ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__rev2, __p3); \
	63667	__ret; \
	63668	})
	63669	#define __noswap_vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63670	float16_t __ret; \
	63671	float16_t __s0 = __p0; \
	63672	float16_t __s1 = __p1; \
	63673	float16x4_t __s2 = __p2; \
	63674	__ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
	63675	__ret; \
	63676	})
64323	63677	#endif
64324	63678
64325	63679	#ifdef __LITTLE_ENDIAN__
64326		#define vsudotq_laneq_s32(__p0_843, __p1_843, __p2_843, __p3_843) __extension__ ({ \
64327		int32x4_t __ret_843; \
64328		int32x4_t __s0_843 = __p0_843; \
64329		int8x16_t __s1_843 = __p1_843; \
64330		uint8x16_t __s2_843 = __p2_843; \
64331		uint8x16_t __reint_843 = __s2_843; \
64332		__ret_843 = vusdotq_s32(__s0_843, (uint8x16_t)(splatq_laneq_s32((int32x4_t ) &__reint_843, __p3_843)), __s1_843); \
64333		__ret_843; \
	63680	#define vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63681	float16x8_t __ret; \
	63682	float16x8_t __s0 = __p0; \
	63683	float16x8_t __s1 = __p1; \
	63684	float16x4_t __s2 = __p2; \
	63685	__ret = (float16x8_t) __builtin_neon_vfmaq_lane_f16((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 40); \
	63686	__ret; \
64334	63687	})
64335	63688	#else
64336		#define vsudotq_laneq_s32(__p0_844, __p1_844, __p2_844, __p3_844) __extension__ ({ \
64337		int32x4_t __ret_844; \
64338		int32x4_t __s0_844 = __p0_844; \
64339		int8x16_t __s1_844 = __p1_844; \
64340		uint8x16_t __s2_844 = __p2_844; \
64341		int32x4_t __rev0_844; __rev0_844 = __builtin_shufflevector(__s0_844, __s0_844, 3, 2, 1, 0); \
64342		int8x16_t __rev1_844; __rev1_844 = __builtin_shufflevector(__s1_844, __s1_844, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
64343		uint8x16_t __rev2_844; __rev2_844 = __builtin_shufflevector(__s2_844, __s2_844, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
64344		uint8x16_t __reint_844 = __rev2_844; \
64345		__ret_844 = __noswap_vusdotq_s32(__rev0_844, (uint8x16_t)(__noswap_splatq_laneq_s32((int32x4_t ) &__reint_844, __p3_844)), __rev1_844); \
64346		__ret_844 = __builtin_shufflevector(__ret_844, __ret_844, 3, 2, 1, 0); \
64347		__ret_844; \
	63689	#define vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63690	float16x8_t __ret; \
	63691	float16x8_t __s0 = __p0; \
	63692	float16x8_t __s1 = __p1; \
	63693	float16x4_t __s2 = __p2; \
	63694	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	63695	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	63696	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	63697	__ret = (float16x8_t) __builtin_neon_vfmaq_lane_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, __p3, 40); \
	63698	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	63699	__ret; \
	63700	})
	63701	#define __noswap_vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63702	float16x8_t __ret; \
	63703	float16x8_t __s0 = __p0; \
	63704	float16x8_t __s1 = __p1; \
	63705	float16x4_t __s2 = __p2; \
	63706	__ret = (float16x8_t) __builtin_neon_vfmaq_lane_f16((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 40); \
	63707	__ret; \
64348	63708	})
64349	63709	#endif
64350	63710
64351	63711	#ifdef __LITTLE_ENDIAN__
64352		#define vsudot_laneq_s32(__p0_845, __p1_845, __p2_845, __p3_845) __extension__ ({ \
64353		int32x2_t __ret_845; \
64354		int32x2_t __s0_845 = __p0_845; \
64355		int8x8_t __s1_845 = __p1_845; \
64356		uint8x16_t __s2_845 = __p2_845; \
64357		uint8x16_t __reint_845 = __s2_845; \
64358		__ret_845 = vusdot_s32(__s0_845, (uint8x8_t)(splat_laneq_s32((int32x4_t ) &__reint_845, __p3_845)), __s1_845); \
64359		__ret_845; \
	63712	#define vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63713	float16x4_t __ret; \
	63714	float16x4_t __s0 = __p0; \
	63715	float16x4_t __s1 = __p1; \
	63716	float16x4_t __s2 = __p2; \
	63717	__ret = (float16x4_t) __builtin_neon_vfma_lane_f16((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 8); \
	63718	__ret; \
64360	63719	})
64361	63720	#else
64362		#define vsudot_laneq_s32(__p0_846, __p1_846, __p2_846, __p3_846) __extension__ ({ \
64363		int32x2_t __ret_846; \
64364		int32x2_t __s0_846 = __p0_846; \
64365		int8x8_t __s1_846 = __p1_846; \
64366		uint8x16_t __s2_846 = __p2_846; \
64367		int32x2_t __rev0_846; __rev0_846 = __builtin_shufflevector(__s0_846, __s0_846, 1, 0); \
64368		int8x8_t __rev1_846; __rev1_846 = __builtin_shufflevector(__s1_846, __s1_846, 7, 6, 5, 4, 3, 2, 1, 0); \
64369		uint8x16_t __rev2_846; __rev2_846 = __builtin_shufflevector(__s2_846, __s2_846, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
64370		uint8x16_t __reint_846 = __rev2_846; \
64371		__ret_846 = __noswap_vusdot_s32(__rev0_846, (uint8x8_t)(__noswap_splat_laneq_s32((int32x4_t ) &__reint_846, __p3_846)), __rev1_846); \
64372		__ret_846 = __builtin_shufflevector(__ret_846, __ret_846, 1, 0); \
64373		__ret_846; \
	63721	#define vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63722	float16x4_t __ret; \
	63723	float16x4_t __s0 = __p0; \
	63724	float16x4_t __s1 = __p1; \
	63725	float16x4_t __s2 = __p2; \
	63726	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	63727	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	63728	float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	63729	__ret = (float16x4_t) __builtin_neon_vfma_lane_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, __p3, 8); \
	63730	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	63731	__ret; \
	63732	})
	63733	#define __noswap_vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63734	float16x4_t __ret; \
	63735	float16x4_t __s0 = __p0; \
	63736	float16x4_t __s1 = __p1; \
	63737	float16x4_t __s2 = __p2; \
	63738	__ret = (float16x4_t) __builtin_neon_vfma_lane_f16((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 8); \
	63739	__ret; \
64374	63740	})
64375	63741	#endif
64376	63742
64377	63743	#ifdef __LITTLE_ENDIAN__
64378		__ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64379		poly8x8_t __ret;
64380		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64381		return __ret;
64382		}
	63744	#define vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63745	float16_t __ret; \
	63746	float16_t __s0 = __p0; \
	63747	float16_t __s1 = __p1; \
	63748	float16x8_t __s2 = __p2; \
	63749	__ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
	63750	__ret; \
	63751	})
64383	63752	#else
64384		__ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64385		poly8x8_t __ret;
64386		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64387		poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64388		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
64389		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64390		return __ret;
64391		}
	63753	#define vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63754	float16_t __ret; \
	63755	float16_t __s0 = __p0; \
	63756	float16_t __s1 = __p1; \
	63757	float16x8_t __s2 = __p2; \
	63758	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
	63759	__ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__rev2, __p3); \
	63760	__ret; \
	63761	})
	63762	#define __noswap_vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63763	float16_t __ret; \
	63764	float16_t __s0 = __p0; \
	63765	float16_t __s1 = __p1; \
	63766	float16x8_t __s2 = __p2; \
	63767	__ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
	63768	__ret; \
	63769	})
64392	63770	#endif
64393	63771
64394	63772	#ifdef __LITTLE_ENDIAN__
64395		__ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64396		poly16x4_t __ret;
64397		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64398		return __ret;
64399		}
	63773	#define vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63774	float16x8_t __ret; \
	63775	float16x8_t __s0 = __p0; \
	63776	float16x8_t __s1 = __p1; \
	63777	float16x8_t __s2 = __p2; \
	63778	__ret = (float16x8_t) __builtin_neon_vfmaq_laneq_f16((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 40); \
	63779	__ret; \
	63780	})
64400	63781	#else
64401		__ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64402		poly16x4_t __ret;
64403		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64404		poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64405		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
64406		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64407		return __ret;
64408		}
	63782	#define vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63783	float16x8_t __ret; \
	63784	float16x8_t __s0 = __p0; \
	63785	float16x8_t __s1 = __p1; \
	63786	float16x8_t __s2 = __p2; \
	63787	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	63788	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	63789	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
	63790	__ret = (float16x8_t) __builtin_neon_vfmaq_laneq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 40); \
	63791	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	63792	__ret; \
	63793	})
	63794	#define __noswap_vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63795	float16x8_t __ret; \
	63796	float16x8_t __s0 = __p0; \
	63797	float16x8_t __s1 = __p1; \
	63798	float16x8_t __s2 = __p2; \
	63799	__ret = (float16x8_t) __builtin_neon_vfmaq_laneq_f16((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 40); \
	63800	__ret; \
	63801	})
64409	63802	#endif
64410	63803
64411	63804	#ifdef __LITTLE_ENDIAN__
64412		__ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64413		poly8x16_t __ret;
64414		__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64415		return __ret;
64416		}
	63805	#define vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63806	float16x4_t __ret; \
	63807	float16x4_t __s0 = __p0; \
	63808	float16x4_t __s1 = __p1; \
	63809	float16x8_t __s2 = __p2; \
	63810	__ret = (float16x4_t) __builtin_neon_vfma_laneq_f16((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 8); \
	63811	__ret; \
	63812	})
64417	63813	#else
64418		__ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64419		poly8x16_t __ret;
64420		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64421		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64422		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64423		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64424		return __ret;
64425		}
	63814	#define vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63815	float16x4_t __ret; \
	63816	float16x4_t __s0 = __p0; \
	63817	float16x4_t __s1 = __p1; \
	63818	float16x8_t __s2 = __p2; \
	63819	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	63820	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	63821	float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
	63822	__ret = (float16x4_t) __builtin_neon_vfma_laneq_f16((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x16_t)__rev2, __p3, 8); \
	63823	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	63824	__ret; \
	63825	})
	63826	#define __noswap_vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
	63827	float16x4_t __ret; \
	63828	float16x4_t __s0 = __p0; \
	63829	float16x4_t __s1 = __p1; \
	63830	float16x8_t __s2 = __p2; \
	63831	__ret = (float16x4_t) __builtin_neon_vfma_laneq_f16((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 8); \
	63832	__ret; \
	63833	})
64426	63834	#endif
64427	63835
64428	63836	#ifdef __LITTLE_ENDIAN__
64429		__ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64430		poly64x2_t __ret;
64431		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64432		return __ret;
64433		}
	63837	#define vfmaq_n_f16(__p0, __p1, __p2) __extension__ ({ \
	63838	float16x8_t __ret; \
	63839	float16x8_t __s0 = __p0; \
	63840	float16x8_t __s1 = __p1; \
	63841	float16_t __s2 = __p2; \
	63842	__ret = vfmaq_f16(__s0, __s1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
	63843	__ret; \
	63844	})
64434	63845	#else
64435		__ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64436		poly64x2_t __ret;
64437		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64438		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64439		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64440		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64441		return __ret;
64442		}
	63846	#define vfmaq_n_f16(__p0, __p1, __p2) __extension__ ({ \
	63847	float16x8_t __ret; \
	63848	float16x8_t __s0 = __p0; \
	63849	float16x8_t __s1 = __p1; \
	63850	float16_t __s2 = __p2; \
	63851	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	63852	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	63853	__ret = __noswap_vfmaq_f16(__rev0, __rev1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
	63854	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	63855	__ret; \
	63856	})
	63857	#endif
	63858
	63859	#ifdef __LITTLE_ENDIAN__
	63860	#define vfma_n_f16(__p0, __p1, __p2) __extension__ ({ \
	63861	float16x4_t __ret; \
	63862	float16x4_t __s0 = __p0; \
	63863	float16x4_t __s1 = __p1; \
	63864	float16_t __s2 = __p2; \
	63865	__ret = vfma_f16(__s0, __s1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
	63866	__ret; \
	63867	})
	63868	#else
	63869	#define vfma_n_f16(__p0, __p1, __p2) __extension__ ({ \
	63870	float16x4_t __ret; \
	63871	float16x4_t __s0 = __p0; \
	63872	float16x4_t __s1 = __p1; \
	63873	float16_t __s2 = __p2; \
	63874	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	63875	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	63876	__ret = __noswap_vfma_f16(__rev0, __rev1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
	63877	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	63878	__ret; \
	63879	})
64443	63880	#endif
64444	63881
64445	63882	#ifdef __LITTLE_ENDIAN__
64446		__ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64447		poly16x8_t __ret;
64448		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64449		return __ret;
64450		}
	63883	#define vfmsh_lane_f16(__p0_755, __p1_755, __p2_755, __p3_755) __extension__ ({ \
	63884	float16_t __ret_755; \
	63885	float16_t __s0_755 = __p0_755; \
	63886	float16_t __s1_755 = __p1_755; \
	63887	float16x4_t __s2_755 = __p2_755; \
	63888	__ret_755 = vfmah_lane_f16(__s0_755, -__s1_755, __s2_755, __p3_755); \
	63889	__ret_755; \
	63890	})
64451	63891	#else
64452		__ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64453		poly16x8_t __ret;
64454		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64455		poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64456		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
64457		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64458		return __ret;
64459		}
	63892	#define vfmsh_lane_f16(__p0_756, __p1_756, __p2_756, __p3_756) __extension__ ({ \
	63893	float16_t __ret_756; \
	63894	float16_t __s0_756 = __p0_756; \
	63895	float16_t __s1_756 = __p1_756; \
	63896	float16x4_t __s2_756 = __p2_756; \
	63897	float16x4_t __rev2_756; __rev2_756 = __builtin_shufflevector(__s2_756, __s2_756, 3, 2, 1, 0); \
	63898	__ret_756 = __noswap_vfmah_lane_f16(__s0_756, -__s1_756, __rev2_756, __p3_756); \
	63899	__ret_756; \
	63900	})
64460	63901	#endif
64461	63902
64462	63903	#ifdef __LITTLE_ENDIAN__
64463		__ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64464		uint8x16_t __ret;
64465		__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64466		return __ret;
64467		}
	63904	#define vfmsq_lane_f16(__p0_757, __p1_757, __p2_757, __p3_757) __extension__ ({ \
	63905	float16x8_t __ret_757; \
	63906	float16x8_t __s0_757 = __p0_757; \
	63907	float16x8_t __s1_757 = __p1_757; \
	63908	float16x4_t __s2_757 = __p2_757; \
	63909	__ret_757 = vfmaq_lane_f16(__s0_757, -__s1_757, __s2_757, __p3_757); \
	63910	__ret_757; \
	63911	})
64468	63912	#else
64469		__ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64470		uint8x16_t __ret;
64471		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64472		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64473		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64474		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64475		return __ret;
64476		}
	63913	#define vfmsq_lane_f16(__p0_758, __p1_758, __p2_758, __p3_758) __extension__ ({ \
	63914	float16x8_t __ret_758; \
	63915	float16x8_t __s0_758 = __p0_758; \
	63916	float16x8_t __s1_758 = __p1_758; \
	63917	float16x4_t __s2_758 = __p2_758; \
	63918	float16x8_t __rev0_758; __rev0_758 = __builtin_shufflevector(__s0_758, __s0_758, 7, 6, 5, 4, 3, 2, 1, 0); \
	63919	float16x8_t __rev1_758; __rev1_758 = __builtin_shufflevector(__s1_758, __s1_758, 7, 6, 5, 4, 3, 2, 1, 0); \
	63920	float16x4_t __rev2_758; __rev2_758 = __builtin_shufflevector(__s2_758, __s2_758, 3, 2, 1, 0); \
	63921	__ret_758 = __noswap_vfmaq_lane_f16(__rev0_758, -__rev1_758, __rev2_758, __p3_758); \
	63922	__ret_758 = __builtin_shufflevector(__ret_758, __ret_758, 7, 6, 5, 4, 3, 2, 1, 0); \
	63923	__ret_758; \
	63924	})
64477	63925	#endif
64478	63926
64479	63927	#ifdef __LITTLE_ENDIAN__
64480		__ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64481		uint32x4_t __ret;
64482		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64483		return __ret;
64484		}
	63928	#define vfms_lane_f16(__p0_759, __p1_759, __p2_759, __p3_759) __extension__ ({ \
	63929	float16x4_t __ret_759; \
	63930	float16x4_t __s0_759 = __p0_759; \
	63931	float16x4_t __s1_759 = __p1_759; \
	63932	float16x4_t __s2_759 = __p2_759; \
	63933	__ret_759 = vfma_lane_f16(__s0_759, -__s1_759, __s2_759, __p3_759); \
	63934	__ret_759; \
	63935	})
64485	63936	#else
64486		__ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64487		uint32x4_t __ret;
64488		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64489		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64490		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
64491		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64492		return __ret;
64493		}
	63937	#define vfms_lane_f16(__p0_760, __p1_760, __p2_760, __p3_760) __extension__ ({ \
	63938	float16x4_t __ret_760; \
	63939	float16x4_t __s0_760 = __p0_760; \
	63940	float16x4_t __s1_760 = __p1_760; \
	63941	float16x4_t __s2_760 = __p2_760; \
	63942	float16x4_t __rev0_760; __rev0_760 = __builtin_shufflevector(__s0_760, __s0_760, 3, 2, 1, 0); \
	63943	float16x4_t __rev1_760; __rev1_760 = __builtin_shufflevector(__s1_760, __s1_760, 3, 2, 1, 0); \
	63944	float16x4_t __rev2_760; __rev2_760 = __builtin_shufflevector(__s2_760, __s2_760, 3, 2, 1, 0); \
	63945	__ret_760 = __noswap_vfma_lane_f16(__rev0_760, -__rev1_760, __rev2_760, __p3_760); \
	63946	__ret_760 = __builtin_shufflevector(__ret_760, __ret_760, 3, 2, 1, 0); \
	63947	__ret_760; \
	63948	})
64494	63949	#endif
64495	63950
64496	63951	#ifdef __LITTLE_ENDIAN__
64497		__ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64498		uint64x2_t __ret;
64499		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64500		return __ret;
64501		}
	63952	#define vfmsh_laneq_f16(__p0_761, __p1_761, __p2_761, __p3_761) __extension__ ({ \
	63953	float16_t __ret_761; \
	63954	float16_t __s0_761 = __p0_761; \
	63955	float16_t __s1_761 = __p1_761; \
	63956	float16x8_t __s2_761 = __p2_761; \
	63957	__ret_761 = vfmah_laneq_f16(__s0_761, -__s1_761, __s2_761, __p3_761); \
	63958	__ret_761; \
	63959	})
64502	63960	#else
64503		__ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64504		uint64x2_t __ret;
64505		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64506		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64507		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64508		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64509		return __ret;
64510		}
	63961	#define vfmsh_laneq_f16(__p0_762, __p1_762, __p2_762, __p3_762) __extension__ ({ \
	63962	float16_t __ret_762; \
	63963	float16_t __s0_762 = __p0_762; \
	63964	float16_t __s1_762 = __p1_762; \
	63965	float16x8_t __s2_762 = __p2_762; \
	63966	float16x8_t __rev2_762; __rev2_762 = __builtin_shufflevector(__s2_762, __s2_762, 7, 6, 5, 4, 3, 2, 1, 0); \
	63967	__ret_762 = __noswap_vfmah_laneq_f16(__s0_762, -__s1_762, __rev2_762, __p3_762); \
	63968	__ret_762; \
	63969	})
64511	63970	#endif
64512	63971
64513	63972	#ifdef __LITTLE_ENDIAN__
64514		__ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64515		uint16x8_t __ret;
64516		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64517		return __ret;
64518		}
	63973	#define vfmsq_laneq_f16(__p0_763, __p1_763, __p2_763, __p3_763) __extension__ ({ \
	63974	float16x8_t __ret_763; \
	63975	float16x8_t __s0_763 = __p0_763; \
	63976	float16x8_t __s1_763 = __p1_763; \
	63977	float16x8_t __s2_763 = __p2_763; \
	63978	__ret_763 = vfmaq_laneq_f16(__s0_763, -__s1_763, __s2_763, __p3_763); \
	63979	__ret_763; \
	63980	})
64519	63981	#else
64520		__ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64521		uint16x8_t __ret;
64522		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64523		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64524		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
64525		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64526		return __ret;
64527		}
	63982	#define vfmsq_laneq_f16(__p0_764, __p1_764, __p2_764, __p3_764) __extension__ ({ \
	63983	float16x8_t __ret_764; \
	63984	float16x8_t __s0_764 = __p0_764; \
	63985	float16x8_t __s1_764 = __p1_764; \
	63986	float16x8_t __s2_764 = __p2_764; \
	63987	float16x8_t __rev0_764; __rev0_764 = __builtin_shufflevector(__s0_764, __s0_764, 7, 6, 5, 4, 3, 2, 1, 0); \
	63988	float16x8_t __rev1_764; __rev1_764 = __builtin_shufflevector(__s1_764, __s1_764, 7, 6, 5, 4, 3, 2, 1, 0); \
	63989	float16x8_t __rev2_764; __rev2_764 = __builtin_shufflevector(__s2_764, __s2_764, 7, 6, 5, 4, 3, 2, 1, 0); \
	63990	__ret_764 = __noswap_vfmaq_laneq_f16(__rev0_764, -__rev1_764, __rev2_764, __p3_764); \
	63991	__ret_764 = __builtin_shufflevector(__ret_764, __ret_764, 7, 6, 5, 4, 3, 2, 1, 0); \
	63992	__ret_764; \
	63993	})
64528	63994	#endif
64529	63995
64530	63996	#ifdef __LITTLE_ENDIAN__
64531		__ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
64532		int8x16_t __ret;
64533		__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64534		return __ret;
64535		}
	63997	#define vfms_laneq_f16(__p0_765, __p1_765, __p2_765, __p3_765) __extension__ ({ \
	63998	float16x4_t __ret_765; \
	63999	float16x4_t __s0_765 = __p0_765; \
	64000	float16x4_t __s1_765 = __p1_765; \
	64001	float16x8_t __s2_765 = __p2_765; \
	64002	__ret_765 = vfma_laneq_f16(__s0_765, -__s1_765, __s2_765, __p3_765); \
	64003	__ret_765; \
	64004	})
64536	64005	#else
64537		__ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
64538		int8x16_t __ret;
64539		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64540		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64541		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64542		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64543		return __ret;
64544		}
	64006	#define vfms_laneq_f16(__p0_766, __p1_766, __p2_766, __p3_766) __extension__ ({ \
	64007	float16x4_t __ret_766; \
	64008	float16x4_t __s0_766 = __p0_766; \
	64009	float16x4_t __s1_766 = __p1_766; \
	64010	float16x8_t __s2_766 = __p2_766; \
	64011	float16x4_t __rev0_766; __rev0_766 = __builtin_shufflevector(__s0_766, __s0_766, 3, 2, 1, 0); \
	64012	float16x4_t __rev1_766; __rev1_766 = __builtin_shufflevector(__s1_766, __s1_766, 3, 2, 1, 0); \
	64013	float16x8_t __rev2_766; __rev2_766 = __builtin_shufflevector(__s2_766, __s2_766, 7, 6, 5, 4, 3, 2, 1, 0); \
	64014	__ret_766 = __noswap_vfma_laneq_f16(__rev0_766, -__rev1_766, __rev2_766, __p3_766); \
	64015	__ret_766 = __builtin_shufflevector(__ret_766, __ret_766, 3, 2, 1, 0); \
	64016	__ret_766; \
	64017	})
64545	64018	#endif
64546	64019
64547	64020	#ifdef __LITTLE_ENDIAN__
64548		__ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
64549		float64x2_t __ret;
64550		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64551		return __ret;
64552		}
	64021	#define vfmsq_n_f16(__p0, __p1, __p2) __extension__ ({ \
	64022	float16x8_t __ret; \
	64023	float16x8_t __s0 = __p0; \
	64024	float16x8_t __s1 = __p1; \
	64025	float16_t __s2 = __p2; \
	64026	__ret = vfmaq_f16(__s0, -__s1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
	64027	__ret; \
	64028	})
64553	64029	#else
64554		__ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
64555		float64x2_t __ret;
64556		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64557		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64558		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64559		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64560		return __ret;
64561		}
	64030	#define vfmsq_n_f16(__p0, __p1, __p2) __extension__ ({ \
	64031	float16x8_t __ret; \
	64032	float16x8_t __s0 = __p0; \
	64033	float16x8_t __s1 = __p1; \
	64034	float16_t __s2 = __p2; \
	64035	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	64036	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	64037	__ret = __noswap_vfmaq_f16(__rev0, -__rev1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
	64038	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	64039	__ret; \
	64040	})
64562	64041	#endif
64563	64042
64564	64043	#ifdef __LITTLE_ENDIAN__
64565		__ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
64566		float32x4_t __ret;
64567		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64568		return __ret;
64569		}
	64044	#define vfms_n_f16(__p0, __p1, __p2) __extension__ ({ \
	64045	float16x4_t __ret; \
	64046	float16x4_t __s0 = __p0; \
	64047	float16x4_t __s1 = __p1; \
	64048	float16_t __s2 = __p2; \
	64049	__ret = vfma_f16(__s0, -__s1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
	64050	__ret; \
	64051	})
64570	64052	#else
64571		__ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
64572		float32x4_t __ret;
64573		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64574		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64575		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
64576		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64577		return __ret;
64578		}
	64053	#define vfms_n_f16(__p0, __p1, __p2) __extension__ ({ \
	64054	float16x4_t __ret; \
	64055	float16x4_t __s0 = __p0; \
	64056	float16x4_t __s1 = __p1; \
	64057	float16_t __s2 = __p2; \
	64058	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	64059	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	64060	__ret = __noswap_vfma_f16(__rev0, -__rev1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
	64061	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	64062	__ret; \
	64063	})
64579	64064	#endif
64580	64065
64581	64066	#ifdef __LITTLE_ENDIAN__
64582		__ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
64583		int32x4_t __ret;
64584		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64585		return __ret;
64586		}
	64067	#define vmaxnmvq_f16(__p0) __extension__ ({ \
	64068	float16_t __ret; \
	64069	float16x8_t __s0 = __p0; \
	64070	__ret = (float16_t) __builtin_neon_vmaxnmvq_f16((int8x16_t)__s0); \
	64071	__ret; \
	64072	})
64587	64073	#else
64588		__ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
64589		int32x4_t __ret;
64590		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64591		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64592		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
64593		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64594		return __ret;
64595		}
	64074	#define vmaxnmvq_f16(__p0) __extension__ ({ \
	64075	float16_t __ret; \
	64076	float16x8_t __s0 = __p0; \
	64077	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	64078	__ret = (float16_t) __builtin_neon_vmaxnmvq_f16((int8x16_t)__rev0); \
	64079	__ret; \
	64080	})
64596	64081	#endif
64597	64082
64598	64083	#ifdef __LITTLE_ENDIAN__
64599		__ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
64600		int64x2_t __ret;
64601		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64602		return __ret;
64603		}
	64084	#define vmaxnmv_f16(__p0) __extension__ ({ \
	64085	float16_t __ret; \
	64086	float16x4_t __s0 = __p0; \
	64087	__ret = (float16_t) __builtin_neon_vmaxnmv_f16((int8x8_t)__s0); \
	64088	__ret; \
	64089	})
64604	64090	#else
64605		__ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
64606		int64x2_t __ret;
64607		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64608		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64609		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64610		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64611		return __ret;
64612		}
	64091	#define vmaxnmv_f16(__p0) __extension__ ({ \
	64092	float16_t __ret; \
	64093	float16x4_t __s0 = __p0; \
	64094	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	64095	__ret = (float16_t) __builtin_neon_vmaxnmv_f16((int8x8_t)__rev0); \
	64096	__ret; \
	64097	})
64613	64098	#endif
64614	64099
64615	64100	#ifdef __LITTLE_ENDIAN__
64616		__ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
64617		int16x8_t __ret;
64618		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64619		return __ret;
64620		}
	64101	#define vmaxvq_f16(__p0) __extension__ ({ \
	64102	float16_t __ret; \
	64103	float16x8_t __s0 = __p0; \
	64104	__ret = (float16_t) __builtin_neon_vmaxvq_f16((int8x16_t)__s0); \
	64105	__ret; \
	64106	})
64621	64107	#else
64622		__ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
64623		int16x8_t __ret;
64624		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64625		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64626		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
64627		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64628		return __ret;
64629		}
	64108	#define vmaxvq_f16(__p0) __extension__ ({ \
	64109	float16_t __ret; \
	64110	float16x8_t __s0 = __p0; \
	64111	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	64112	__ret = (float16_t) __builtin_neon_vmaxvq_f16((int8x16_t)__rev0); \
	64113	__ret; \
	64114	})
64630	64115	#endif
64631	64116
64632	64117	#ifdef __LITTLE_ENDIAN__
64633		__ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64634		uint8x8_t __ret;
64635		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64636		return __ret;
64637		}
	64118	#define vmaxv_f16(__p0) __extension__ ({ \
	64119	float16_t __ret; \
	64120	float16x4_t __s0 = __p0; \
	64121	__ret = (float16_t) __builtin_neon_vmaxv_f16((int8x8_t)__s0); \
	64122	__ret; \
	64123	})
64638	64124	#else
64639		__ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64640		uint8x8_t __ret;
64641		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64642		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64643		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
64644		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64645		return __ret;
64646		}
	64125	#define vmaxv_f16(__p0) __extension__ ({ \
	64126	float16_t __ret; \
	64127	float16x4_t __s0 = __p0; \
	64128	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	64129	__ret = (float16_t) __builtin_neon_vmaxv_f16((int8x8_t)__rev0); \
	64130	__ret; \
	64131	})
64647	64132	#endif
64648	64133
64649	64134	#ifdef __LITTLE_ENDIAN__
64650		__ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64651		uint32x2_t __ret;
64652		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64653		return __ret;
64654		}
	64135	#define vminnmvq_f16(__p0) __extension__ ({ \
	64136	float16_t __ret; \
	64137	float16x8_t __s0 = __p0; \
	64138	__ret = (float16_t) __builtin_neon_vminnmvq_f16((int8x16_t)__s0); \
	64139	__ret; \
	64140	})
64655	64141	#else
64656		__ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64657		uint32x2_t __ret;
64658		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64659		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64660		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64661		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64662		return __ret;
64663		}
	64142	#define vminnmvq_f16(__p0) __extension__ ({ \
	64143	float16_t __ret; \
	64144	float16x8_t __s0 = __p0; \
	64145	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	64146	__ret = (float16_t) __builtin_neon_vminnmvq_f16((int8x16_t)__rev0); \
	64147	__ret; \
	64148	})
64664	64149	#endif
64665	64150
64666	64151	#ifdef __LITTLE_ENDIAN__
64667		__ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64668		uint16x4_t __ret;
64669		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64670		return __ret;
64671		}
	64152	#define vminnmv_f16(__p0) __extension__ ({ \
	64153	float16_t __ret; \
	64154	float16x4_t __s0 = __p0; \
	64155	__ret = (float16_t) __builtin_neon_vminnmv_f16((int8x8_t)__s0); \
	64156	__ret; \
	64157	})
64672	64158	#else
64673		__ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64674		uint16x4_t __ret;
64675		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64676		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64677		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
64678		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64679		return __ret;
64680		}
	64159	#define vminnmv_f16(__p0) __extension__ ({ \
	64160	float16_t __ret; \
	64161	float16x4_t __s0 = __p0; \
	64162	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	64163	__ret = (float16_t) __builtin_neon_vminnmv_f16((int8x8_t)__rev0); \
	64164	__ret; \
	64165	})
64681	64166	#endif
64682	64167
64683	64168	#ifdef __LITTLE_ENDIAN__
64684		__ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
64685		int8x8_t __ret;
64686		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64687		return __ret;
64688		}
	64169	#define vminvq_f16(__p0) __extension__ ({ \
	64170	float16_t __ret; \
	64171	float16x8_t __s0 = __p0; \
	64172	__ret = (float16_t) __builtin_neon_vminvq_f16((int8x16_t)__s0); \
	64173	__ret; \
	64174	})
64689	64175	#else
64690		__ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
64691		int8x8_t __ret;
64692		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64693		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64694		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
64695		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64696		return __ret;
64697		}
	64176	#define vminvq_f16(__p0) __extension__ ({ \
	64177	float16_t __ret; \
	64178	float16x8_t __s0 = __p0; \
	64179	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	64180	__ret = (float16_t) __builtin_neon_vminvq_f16((int8x16_t)__rev0); \
	64181	__ret; \
	64182	})
64698	64183	#endif
64699	64184
64700	64185	#ifdef __LITTLE_ENDIAN__
64701		__ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
64702		float32x2_t __ret;
64703		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64704		return __ret;
64705		}
	64186	#define vminv_f16(__p0) __extension__ ({ \
	64187	float16_t __ret; \
	64188	float16x4_t __s0 = __p0; \
	64189	__ret = (float16_t) __builtin_neon_vminv_f16((int8x8_t)__s0); \
	64190	__ret; \
	64191	})
64706	64192	#else
64707		__ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
64708		float32x2_t __ret;
64709		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64710		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64711		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64712		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64713		return __ret;
64714		}
	64193	#define vminv_f16(__p0) __extension__ ({ \
	64194	float16_t __ret; \
	64195	float16x4_t __s0 = __p0; \
	64196	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	64197	__ret = (float16_t) __builtin_neon_vminv_f16((int8x8_t)__rev0); \
	64198	__ret; \
	64199	})
64715	64200	#endif
64716	64201
64717	64202	#ifdef __LITTLE_ENDIAN__
64718		__ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
64719		int32x2_t __ret;
64720		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64721		return __ret;
64722		}
	64203	#define vmulq_laneq_f16(__p0_767, __p1_767, __p2_767) __extension__ ({ \
	64204	float16x8_t __ret_767; \
	64205	float16x8_t __s0_767 = __p0_767; \
	64206	float16x8_t __s1_767 = __p1_767; \
	64207	__ret_767 = __s0_767 * splatq_laneq_f16(__s1_767, __p2_767); \
	64208	__ret_767; \
	64209	})
64723	64210	#else
64724		__ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
64725		int32x2_t __ret;
64726		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64727		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64728		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
64729		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64730		return __ret;
64731		}
	64211	#define vmulq_laneq_f16(__p0_768, __p1_768, __p2_768) __extension__ ({ \
	64212	float16x8_t __ret_768; \
	64213	float16x8_t __s0_768 = __p0_768; \
	64214	float16x8_t __s1_768 = __p1_768; \
	64215	float16x8_t __rev0_768; __rev0_768 = __builtin_shufflevector(__s0_768, __s0_768, 7, 6, 5, 4, 3, 2, 1, 0); \
	64216	float16x8_t __rev1_768; __rev1_768 = __builtin_shufflevector(__s1_768, __s1_768, 7, 6, 5, 4, 3, 2, 1, 0); \
	64217	__ret_768 = __rev0_768 * __noswap_splatq_laneq_f16(__rev1_768, __p2_768); \
	64218	__ret_768 = __builtin_shufflevector(__ret_768, __ret_768, 7, 6, 5, 4, 3, 2, 1, 0); \
	64219	__ret_768; \
	64220	})
64732	64221	#endif
64733	64222
64734	64223	#ifdef __LITTLE_ENDIAN__
64735		__ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
64736		int16x4_t __ret;
64737		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64738		return __ret;
64739		}
	64224	#define vmul_laneq_f16(__p0_769, __p1_769, __p2_769) __extension__ ({ \
	64225	float16x4_t __ret_769; \
	64226	float16x4_t __s0_769 = __p0_769; \
	64227	float16x8_t __s1_769 = __p1_769; \
	64228	__ret_769 = __s0_769 * splat_laneq_f16(__s1_769, __p2_769); \
	64229	__ret_769; \
	64230	})
64740	64231	#else
64741		__ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
64742		int16x4_t __ret;
64743		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64744		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64745		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
64746		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64747		return __ret;
64748		}
	64232	#define vmul_laneq_f16(__p0_770, __p1_770, __p2_770) __extension__ ({ \
	64233	float16x4_t __ret_770; \
	64234	float16x4_t __s0_770 = __p0_770; \
	64235	float16x8_t __s1_770 = __p1_770; \
	64236	float16x4_t __rev0_770; __rev0_770 = __builtin_shufflevector(__s0_770, __s0_770, 3, 2, 1, 0); \
	64237	float16x8_t __rev1_770; __rev1_770 = __builtin_shufflevector(__s1_770, __s1_770, 7, 6, 5, 4, 3, 2, 1, 0); \
	64238	__ret_770 = __rev0_770 * __noswap_splat_laneq_f16(__rev1_770, __p2_770); \
	64239	__ret_770 = __builtin_shufflevector(__ret_770, __ret_770, 3, 2, 1, 0); \
	64240	__ret_770; \
	64241	})
64749	64242	#endif
64750	64243
64751	64244	#ifdef __LITTLE_ENDIAN__
64752		__ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64753		poly8x8_t __ret;
64754		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
	64245	__ai __attribute__((target("fullfp16"))) float16x8_t vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
	64246	float16x8_t __ret;
	64247	__ret = (float16x8_t) __builtin_neon_vmulxq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
64755	64248	return __ret;
64756	64249	}
64757	64250	#else
64758		__ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64759		poly8x8_t __ret;
64760		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64761		poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64762		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	64251	__ai __attribute__((target("fullfp16"))) float16x8_t vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
	64252	float16x8_t __ret;
	64253	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64254	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64255	__ret = (float16x8_t) __builtin_neon_vmulxq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
64763	64256	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64764	64257	return __ret;
64765	64258	}
64766		#endif
64767
64768		#ifdef __LITTLE_ENDIAN__
64769		__ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64770		poly16x4_t __ret;
64771		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64772		return __ret;
64773		}
64774		#else
64775		__ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64776		poly16x4_t __ret;
64777		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64778		poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64779		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
64780		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	64259	__ai __attribute__((target("fullfp16"))) float16x8_t __noswap_vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
	64260	float16x8_t __ret;
	64261	__ret = (float16x8_t) __builtin_neon_vmulxq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
64781	64262	return __ret;
64782	64263	}
64783	64264	#endif
64784	64265
64785	64266	#ifdef __LITTLE_ENDIAN__
64786		__ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64787		poly8x16_t __ret;
64788		__ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
	64267	__ai __attribute__((target("fullfp16"))) float16x4_t vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
	64268	float16x4_t __ret;
	64269	__ret = (float16x4_t) __builtin_neon_vmulx_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
64789	64270	return __ret;
64790	64271	}
64791	64272	#else
64792		__ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64793		poly8x16_t __ret;
64794		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64795		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64796		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64797		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64798		return __ret;
64799		}
64800		#endif
64801
64802		#ifdef __LITTLE_ENDIAN__
64803		__ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64804		poly64x2_t __ret;
64805		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	64273	__ai __attribute__((target("fullfp16"))) float16x4_t vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
	64274	float16x4_t __ret;
	64275	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64276	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64277	__ret = (float16x4_t) __builtin_neon_vmulx_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	64278	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64806	64279	return __ret;
64807	64280	}
64808		#else
64809		__ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64810		poly64x2_t __ret;
64811		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64812		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64813		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
64814		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64281	__ai __attribute__((target("fullfp16"))) float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
	64282	float16x4_t __ret;
	64283	__ret = (float16x4_t) __builtin_neon_vmulx_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
64815	64284	return __ret;
64816	64285	}
64817	64286	#endif
64818	64287
64819	64288	#ifdef __LITTLE_ENDIAN__
64820		__ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64821		poly16x8_t __ret;
64822		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64823		return __ret;
64824		}
	64289	#define vmulxh_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	64290	float16_t __ret; \
	64291	float16_t __s0 = __p0; \
	64292	float16x4_t __s1 = __p1; \
	64293	__ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__s1, __p2); \
	64294	__ret; \
	64295	})
64825	64296	#else
64826		__ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64827		poly16x8_t __ret;
64828		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64829		poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64830		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
64831		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64832		return __ret;
64833		}
	64297	#define vmulxh_lane_f16(__p0, __p1, __p2) __extension__ ({ \
	64298	float16_t __ret; \
	64299	float16_t __s0 = __p0; \
	64300	float16x4_t __s1 = __p1; \
	64301	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	64302	__ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__rev1, __p2); \
	64303	__ret; \
	64304	})
64834	64305	#endif
64835	64306
64836	64307	#ifdef __LITTLE_ENDIAN__
64837		__ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64838		uint8x16_t __ret;
64839		__ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64840		return __ret;
64841		}
	64308	#define vmulxq_lane_f16(__p0_771, __p1_771, __p2_771) __extension__ ({ \
	64309	float16x8_t __ret_771; \
	64310	float16x8_t __s0_771 = __p0_771; \
	64311	float16x4_t __s1_771 = __p1_771; \
	64312	__ret_771 = vmulxq_f16(__s0_771, splatq_lane_f16(__s1_771, __p2_771)); \
	64313	__ret_771; \
	64314	})
	64315	#else
	64316	#define vmulxq_lane_f16(__p0_772, __p1_772, __p2_772) __extension__ ({ \
	64317	float16x8_t __ret_772; \
	64318	float16x8_t __s0_772 = __p0_772; \
	64319	float16x4_t __s1_772 = __p1_772; \
	64320	float16x8_t __rev0_772; __rev0_772 = __builtin_shufflevector(__s0_772, __s0_772, 7, 6, 5, 4, 3, 2, 1, 0); \
	64321	float16x4_t __rev1_772; __rev1_772 = __builtin_shufflevector(__s1_772, __s1_772, 3, 2, 1, 0); \
	64322	__ret_772 = __noswap_vmulxq_f16(__rev0_772, __noswap_splatq_lane_f16(__rev1_772, __p2_772)); \
	64323	__ret_772 = __builtin_shufflevector(__ret_772, __ret_772, 7, 6, 5, 4, 3, 2, 1, 0); \
	64324	__ret_772; \
	64325	})
	64326	#endif
	64327
	64328	#ifdef __LITTLE_ENDIAN__
	64329	#define vmulx_lane_f16(__p0_773, __p1_773, __p2_773) __extension__ ({ \
	64330	float16x4_t __ret_773; \
	64331	float16x4_t __s0_773 = __p0_773; \
	64332	float16x4_t __s1_773 = __p1_773; \
	64333	__ret_773 = vmulx_f16(__s0_773, splat_lane_f16(__s1_773, __p2_773)); \
	64334	__ret_773; \
	64335	})
64842	64336	#else
64843		__ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64844		uint8x16_t __ret;
64845		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64846		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64847		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64848		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64849		return __ret;
64850		}
	64337	#define vmulx_lane_f16(__p0_774, __p1_774, __p2_774) __extension__ ({ \
	64338	float16x4_t __ret_774; \
	64339	float16x4_t __s0_774 = __p0_774; \
	64340	float16x4_t __s1_774 = __p1_774; \
	64341	float16x4_t __rev0_774; __rev0_774 = __builtin_shufflevector(__s0_774, __s0_774, 3, 2, 1, 0); \
	64342	float16x4_t __rev1_774; __rev1_774 = __builtin_shufflevector(__s1_774, __s1_774, 3, 2, 1, 0); \
	64343	__ret_774 = __noswap_vmulx_f16(__rev0_774, __noswap_splat_lane_f16(__rev1_774, __p2_774)); \
	64344	__ret_774 = __builtin_shufflevector(__ret_774, __ret_774, 3, 2, 1, 0); \
	64345	__ret_774; \
	64346	})
64851	64347	#endif
64852	64348
64853	64349	#ifdef __LITTLE_ENDIAN__
64854		__ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64855		uint32x4_t __ret;
64856		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64857		return __ret;
64858		}
	64350	#define vmulxh_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
	64351	float16_t __ret; \
	64352	float16_t __s0 = __p0; \
	64353	float16x8_t __s1 = __p1; \
	64354	__ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__s1, __p2); \
	64355	__ret; \
	64356	})
64859	64357	#else
64860		__ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64861		uint32x4_t __ret;
64862		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64863		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64864		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
64865		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
64866		return __ret;
64867		}
	64358	#define vmulxh_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
	64359	float16_t __ret; \
	64360	float16_t __s0 = __p0; \
	64361	float16x8_t __s1 = __p1; \
	64362	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
	64363	__ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__rev1, __p2); \
	64364	__ret; \
	64365	})
64868	64366	#endif
64869	64367
64870	64368	#ifdef __LITTLE_ENDIAN__
64871		__ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64872		uint64x2_t __ret;
64873		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64874		return __ret;
64875		}
	64369	#define vmulxq_laneq_f16(__p0_775, __p1_775, __p2_775) __extension__ ({ \
	64370	float16x8_t __ret_775; \
	64371	float16x8_t __s0_775 = __p0_775; \
	64372	float16x8_t __s1_775 = __p1_775; \
	64373	__ret_775 = vmulxq_f16(__s0_775, splatq_laneq_f16(__s1_775, __p2_775)); \
	64374	__ret_775; \
	64375	})
64876	64376	#else
64877		__ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64878		uint64x2_t __ret;
64879		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64880		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64881		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
64882		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64883		return __ret;
64884		}
	64377	#define vmulxq_laneq_f16(__p0_776, __p1_776, __p2_776) __extension__ ({ \
	64378	float16x8_t __ret_776; \
	64379	float16x8_t __s0_776 = __p0_776; \
	64380	float16x8_t __s1_776 = __p1_776; \
	64381	float16x8_t __rev0_776; __rev0_776 = __builtin_shufflevector(__s0_776, __s0_776, 7, 6, 5, 4, 3, 2, 1, 0); \
	64382	float16x8_t __rev1_776; __rev1_776 = __builtin_shufflevector(__s1_776, __s1_776, 7, 6, 5, 4, 3, 2, 1, 0); \
	64383	__ret_776 = __noswap_vmulxq_f16(__rev0_776, __noswap_splatq_laneq_f16(__rev1_776, __p2_776)); \
	64384	__ret_776 = __builtin_shufflevector(__ret_776, __ret_776, 7, 6, 5, 4, 3, 2, 1, 0); \
	64385	__ret_776; \
	64386	})
64885	64387	#endif
64886	64388
64887	64389	#ifdef __LITTLE_ENDIAN__
64888		__ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64889		uint16x8_t __ret;
64890		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64891		return __ret;
64892		}
	64390	#define vmulx_laneq_f16(__p0_777, __p1_777, __p2_777) __extension__ ({ \
	64391	float16x4_t __ret_777; \
	64392	float16x4_t __s0_777 = __p0_777; \
	64393	float16x8_t __s1_777 = __p1_777; \
	64394	__ret_777 = vmulx_f16(__s0_777, splat_laneq_f16(__s1_777, __p2_777)); \
	64395	__ret_777; \
	64396	})
64893	64397	#else
64894		__ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64895		uint16x8_t __ret;
64896		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64897		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64898		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
64899		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64900		return __ret;
64901		}
	64398	#define vmulx_laneq_f16(__p0_778, __p1_778, __p2_778) __extension__ ({ \
	64399	float16x4_t __ret_778; \
	64400	float16x4_t __s0_778 = __p0_778; \
	64401	float16x8_t __s1_778 = __p1_778; \
	64402	float16x4_t __rev0_778; __rev0_778 = __builtin_shufflevector(__s0_778, __s0_778, 3, 2, 1, 0); \
	64403	float16x8_t __rev1_778; __rev1_778 = __builtin_shufflevector(__s1_778, __s1_778, 7, 6, 5, 4, 3, 2, 1, 0); \
	64404	__ret_778 = __noswap_vmulx_f16(__rev0_778, __noswap_splat_laneq_f16(__rev1_778, __p2_778)); \
	64405	__ret_778 = __builtin_shufflevector(__ret_778, __ret_778, 3, 2, 1, 0); \
	64406	__ret_778; \
	64407	})
64902	64408	#endif
64903	64409
64904	64410	#ifdef __LITTLE_ENDIAN__
64905		__ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64906		int8x16_t __ret;
64907		__ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64908		return __ret;
64909		}
	64411	#define vmulxq_n_f16(__p0, __p1) __extension__ ({ \
	64412	float16x8_t __ret; \
	64413	float16x8_t __s0 = __p0; \
	64414	float16_t __s1 = __p1; \
	64415	__ret = vmulxq_f16(__s0, (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}); \
	64416	__ret; \
	64417	})
64910	64418	#else
64911		__ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64912		int8x16_t __ret;
64913		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64914		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64915		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64916		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64917		return __ret;
64918		}
	64419	#define vmulxq_n_f16(__p0, __p1) __extension__ ({ \
	64420	float16x8_t __ret; \
	64421	float16x8_t __s0 = __p0; \
	64422	float16_t __s1 = __p1; \
	64423	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
	64424	__ret = __noswap_vmulxq_f16(__rev0, (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}); \
	64425	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
	64426	__ret; \
	64427	})
64919	64428	#endif
64920	64429
64921	64430	#ifdef __LITTLE_ENDIAN__
64922		__ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64923		float64x2_t __ret;
64924		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64925		return __ret;
64926		}
	64431	#define vmulx_n_f16(__p0, __p1) __extension__ ({ \
	64432	float16x4_t __ret; \
	64433	float16x4_t __s0 = __p0; \
	64434	float16_t __s1 = __p1; \
	64435	__ret = vmulx_f16(__s0, (float16x4_t) {__s1, __s1, __s1, __s1}); \
	64436	__ret; \
	64437	})
64927	64438	#else
64928		__ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64929		float64x2_t __ret;
64930		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64931		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64932		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
64933		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
64934		return __ret;
64935		}
	64439	#define vmulx_n_f16(__p0, __p1) __extension__ ({ \
	64440	float16x4_t __ret; \
	64441	float16x4_t __s0 = __p0; \
	64442	float16_t __s1 = __p1; \
	64443	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	64444	__ret = __noswap_vmulx_f16(__rev0, (float16x4_t) {__s1, __s1, __s1, __s1}); \
	64445	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	64446	__ret; \
	64447	})
64936	64448	#endif
64937	64449
64938	64450	#ifdef __LITTLE_ENDIAN__
64939		__ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64940		float32x4_t __ret;
64941		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
	64451	__ai __attribute__((target("fullfp16"))) float16x8_t vpaddq_f16(float16x8_t __p0, float16x8_t __p1) {
	64452	float16x8_t __ret;
	64453	__ret = (float16x8_t) __builtin_neon_vpaddq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
64942	64454	return __ret;
64943	64455	}
64944	64456	#else
64945		__ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64946		float32x4_t __ret;
64947		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64948		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64949		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
64950		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	64457	__ai __attribute__((target("fullfp16"))) float16x8_t vpaddq_f16(float16x8_t __p0, float16x8_t __p1) {
	64458	float16x8_t __ret;
	64459	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64460	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64461	__ret = (float16x8_t) __builtin_neon_vpaddq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	64462	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64951	64463	return __ret;
64952	64464	}
64953	64465	#endif
64954	64466
64955	64467	#ifdef __LITTLE_ENDIAN__
64956		__ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64957		int32x4_t __ret;
64958		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
	64468	__ai __attribute__((target("fullfp16"))) float16x8_t vpmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
	64469	float16x8_t __ret;
	64470	__ret = (float16x8_t) __builtin_neon_vpmaxq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
64959	64471	return __ret;
64960	64472	}
64961	64473	#else
64962		__ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64963		int32x4_t __ret;
64964		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
64965		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64966		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
64967		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	64474	__ai __attribute__((target("fullfp16"))) float16x8_t vpmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
	64475	float16x8_t __ret;
	64476	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64477	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64478	__ret = (float16x8_t) __builtin_neon_vpmaxq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	64479	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64968	64480	return __ret;
64969	64481	}
64970	64482	#endif
64971	64483
64972	64484	#ifdef __LITTLE_ENDIAN__
64973		__ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64974		int64x2_t __ret;
64975		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	64485	__ai __attribute__((target("fullfp16"))) float16x8_t vpmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	64486	float16x8_t __ret;
	64487	__ret = (float16x8_t) __builtin_neon_vpmaxnmq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
64976	64488	return __ret;
64977	64489	}
64978	64490	#else
64979		__ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64980		int64x2_t __ret;
64981		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
64982		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64983		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
64984		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64491	__ai __attribute__((target("fullfp16"))) float16x8_t vpmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	64492	float16x8_t __ret;
	64493	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64494	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64495	__ret = (float16x8_t) __builtin_neon_vpmaxnmq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	64496	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
64985	64497	return __ret;
64986	64498	}
64987	64499	#endif
64988	64500
64989	64501	#ifdef __LITTLE_ENDIAN__
64990		__ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64991		int16x8_t __ret;
64992		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
	64502	__ai __attribute__((target("fullfp16"))) float16x4_t vpmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
	64503	float16x4_t __ret;
	64504	__ret = (float16x4_t) __builtin_neon_vpmaxnm_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
64993	64505	return __ret;
64994	64506	}
64995	64507	#else
64996		__ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64997		int16x8_t __ret;
64998		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
64999		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65000		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
65001		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	64508	__ai __attribute__((target("fullfp16"))) float16x4_t vpmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
	64509	float16x4_t __ret;
	64510	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64511	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64512	__ret = (float16x4_t) __builtin_neon_vpmaxnm_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
	64513	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65002	64514	return __ret;
65003	64515	}
65004	64516	#endif
65005	64517
65006	64518	#ifdef __LITTLE_ENDIAN__
65007		__ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65008		uint8x8_t __ret;
65009		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
	64519	__ai __attribute__((target("fullfp16"))) float16x8_t vpminq_f16(float16x8_t __p0, float16x8_t __p1) {
	64520	float16x8_t __ret;
	64521	__ret = (float16x8_t) __builtin_neon_vpminq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
65010	64522	return __ret;
65011	64523	}
65012	64524	#else
65013		__ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65014		uint8x8_t __ret;
65015		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65016		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65017		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	64525	__ai __attribute__((target("fullfp16"))) float16x8_t vpminq_f16(float16x8_t __p0, float16x8_t __p1) {
	64526	float16x8_t __ret;
	64527	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64528	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64529	__ret = (float16x8_t) __builtin_neon_vpminq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
65018	64530	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65019	64531	return __ret;
65020	64532	}
65021	64533	#endif
65022	64534
65023	64535	#ifdef __LITTLE_ENDIAN__
65024		__ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65025		uint32x2_t __ret;
65026		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	64536	__ai __attribute__((target("fullfp16"))) float16x8_t vpminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	64537	float16x8_t __ret;
	64538	__ret = (float16x8_t) __builtin_neon_vpminnmq_f16((int8x16_t)__p0, (int8x16_t)__p1, 40);
65027	64539	return __ret;
65028	64540	}
65029	64541	#else
65030		__ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65031		uint32x2_t __ret;
65032		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65033		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65034		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
65035		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64542	__ai __attribute__((target("fullfp16"))) float16x8_t vpminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
	64543	float16x8_t __ret;
	64544	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64545	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64546	__ret = (float16x8_t) __builtin_neon_vpminnmq_f16((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
	64547	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65036	64548	return __ret;
65037	64549	}
65038	64550	#endif
65039	64551
65040	64552	#ifdef __LITTLE_ENDIAN__
65041		__ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65042		uint16x4_t __ret;
65043		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
	64553	__ai __attribute__((target("fullfp16"))) float16x4_t vpminnm_f16(float16x4_t __p0, float16x4_t __p1) {
	64554	float16x4_t __ret;
	64555	__ret = (float16x4_t) __builtin_neon_vpminnm_f16((int8x8_t)__p0, (int8x8_t)__p1, 8);
65044	64556	return __ret;
65045	64557	}
65046	64558	#else
65047		__ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65048		uint16x4_t __ret;
65049		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65050		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65051		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	64559	__ai __attribute__((target("fullfp16"))) float16x4_t vpminnm_f16(float16x4_t __p0, float16x4_t __p1) {
	64560	float16x4_t __ret;
	64561	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64562	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64563	__ret = (float16x4_t) __builtin_neon_vpminnm_f16((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
65052	64564	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65053	64565	return __ret;
65054	64566	}
65055	64567	#endif
65056	64568
65057	64569	#ifdef __LITTLE_ENDIAN__
65058		__ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
65059		int8x8_t __ret;
65060		__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
	64570	__ai __attribute__((target("fullfp16"))) float16x8_t vrndiq_f16(float16x8_t __p0) {
	64571	float16x8_t __ret;
	64572	__ret = (float16x8_t) __builtin_neon_vrndiq_f16((int8x16_t)__p0, 40);
65061	64573	return __ret;
65062	64574	}
65063	64575	#else
65064		__ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
65065		int8x8_t __ret;
65066		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65067		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65068		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	64576	__ai __attribute__((target("fullfp16"))) float16x8_t vrndiq_f16(float16x8_t __p0) {
	64577	float16x8_t __ret;
	64578	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64579	__ret = (float16x8_t) __builtin_neon_vrndiq_f16((int8x16_t)__rev0, 40);
65069	64580	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65070	64581	return __ret;
65071	64582	}
65072	64583	#endif
65073	64584
65074	64585	#ifdef __LITTLE_ENDIAN__
65075		__ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
65076		float32x2_t __ret;
65077		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	64586	__ai __attribute__((target("fullfp16"))) float16x4_t vrndi_f16(float16x4_t __p0) {
	64587	float16x4_t __ret;
	64588	__ret = (float16x4_t) __builtin_neon_vrndi_f16((int8x8_t)__p0, 8);
65078	64589	return __ret;
65079	64590	}
65080	64591	#else
65081		__ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
65082		float32x2_t __ret;
65083		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65084		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65085		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
65086		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64592	__ai __attribute__((target("fullfp16"))) float16x4_t vrndi_f16(float16x4_t __p0) {
	64593	float16x4_t __ret;
	64594	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64595	__ret = (float16x4_t) __builtin_neon_vrndi_f16((int8x8_t)__rev0, 8);
	64596	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65087	64597	return __ret;
65088	64598	}
65089	64599	#endif
65090	64600
65091	64601	#ifdef __LITTLE_ENDIAN__
65092		__ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
65093		int32x2_t __ret;
65094		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	64602	__ai __attribute__((target("fullfp16"))) float16x8_t vsqrtq_f16(float16x8_t __p0) {
	64603	float16x8_t __ret;
	64604	__ret = (float16x8_t) __builtin_neon_vsqrtq_f16((int8x16_t)__p0, 40);
65095	64605	return __ret;
65096	64606	}
65097	64607	#else
65098		__ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
65099		int32x2_t __ret;
65100		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65101		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65102		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
65103		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64608	__ai __attribute__((target("fullfp16"))) float16x8_t vsqrtq_f16(float16x8_t __p0) {
	64609	float16x8_t __ret;
	64610	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64611	__ret = (float16x8_t) __builtin_neon_vsqrtq_f16((int8x16_t)__rev0, 40);
	64612	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65104	64613	return __ret;
65105	64614	}
65106	64615	#endif
65107	64616
65108	64617	#ifdef __LITTLE_ENDIAN__
65109		__ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
65110		int16x4_t __ret;
65111		__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
	64618	__ai __attribute__((target("fullfp16"))) float16x4_t vsqrt_f16(float16x4_t __p0) {
	64619	float16x4_t __ret;
	64620	__ret = (float16x4_t) __builtin_neon_vsqrt_f16((int8x8_t)__p0, 8);
65112	64621	return __ret;
65113	64622	}
65114	64623	#else
65115		__ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
65116		int16x4_t __ret;
65117		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65118		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65119		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	64624	__ai __attribute__((target("fullfp16"))) float16x4_t vsqrt_f16(float16x4_t __p0) {
	64625	float16x4_t __ret;
	64626	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64627	__ret = (float16x4_t) __builtin_neon_vsqrt_f16((int8x8_t)__rev0, 8);
65120	64628	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65121	64629	return __ret;
65122	64630	}
65123	64631	#endif
65124	64632
65125		__ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
65126		uint64x1_t __ret;
65127		__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
65128		return __ret;
65129		}
65130	64633	#ifdef __LITTLE_ENDIAN__
65131		__ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
65132		uint64x2_t __ret;
65133		__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	64634	__ai __attribute__((target("fullfp16"))) float16x8_t vtrn1q_f16(float16x8_t __p0, float16x8_t __p1) {
	64635	float16x8_t __ret;
	64636	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
65134	64637	return __ret;
65135	64638	}
65136	64639	#else
65137		__ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
65138		uint64x2_t __ret;
65139		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65140		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65141		__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
65142		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64640	__ai __attribute__((target("fullfp16"))) float16x8_t vtrn1q_f16(float16x8_t __p0, float16x8_t __p1) {
	64641	float16x8_t __ret;
	64642	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64643	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64644	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
	64645	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65143	64646	return __ret;
65144	64647	}
65145	64648	#endif
65146	64649
65147	64650	#ifdef __LITTLE_ENDIAN__
65148		__ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
65149		uint64x2_t __ret;
65150		__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	64651	__ai __attribute__((target("fullfp16"))) float16x4_t vtrn1_f16(float16x4_t __p0, float16x4_t __p1) {
	64652	float16x4_t __ret;
	64653	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
65151	64654	return __ret;
65152	64655	}
65153	64656	#else
65154		__ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
65155		uint64x2_t __ret;
65156		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65157		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65158		__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
65159		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64657	__ai __attribute__((target("fullfp16"))) float16x4_t vtrn1_f16(float16x4_t __p0, float16x4_t __p1) {
	64658	float16x4_t __ret;
	64659	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64660	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64661	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
	64662	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65160	64663	return __ret;
65161	64664	}
65162	64665	#endif
65163	64666
65164	64667	#ifdef __LITTLE_ENDIAN__
65165		__ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
65166		uint64x2_t __ret;
65167		__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
	64668	__ai __attribute__((target("fullfp16"))) float16x8_t vtrn2q_f16(float16x8_t __p0, float16x8_t __p1) {
	64669	float16x8_t __ret;
	64670	__ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
65168	64671	return __ret;
65169	64672	}
65170	64673	#else
65171		__ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
65172		uint64x2_t __ret;
65173		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65174		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65175		__ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
65176		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64674	__ai __attribute__((target("fullfp16"))) float16x8_t vtrn2q_f16(float16x8_t __p0, float16x8_t __p1) {
	64675	float16x8_t __ret;
	64676	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64677	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64678	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
	64679	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65177	64680	return __ret;
65178	64681	}
65179	64682	#endif
65180	64683
65181		__ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
65182		uint64x1_t __ret;
65183		__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
65184		return __ret;
65185		}
65186		__ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
65187		uint64x1_t __ret;
65188		__ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
65189		return __ret;
65190		}
65191		__ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
65192		uint64_t __ret;
65193		__ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
65194		return __ret;
65195		}
65196		__ai uint64_t vtstd_s64(int64_t __p0, int64_t __p1) {
65197		uint64_t __ret;
65198		__ret = (uint64_t) __builtin_neon_vtstd_s64(__p0, __p1);
65199		return __ret;
65200		}
65201		__ai int8_t vuqaddb_s8(int8_t __p0, uint8_t __p1) {
65202		int8_t __ret;
65203		__ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
65204		return __ret;
65205		}
65206		__ai int32_t vuqadds_s32(int32_t __p0, uint32_t __p1) {
65207		int32_t __ret;
65208		__ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
65209		return __ret;
65210		}
65211		__ai int64_t vuqaddd_s64(int64_t __p0, uint64_t __p1) {
65212		int64_t __ret;
65213		__ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
65214		return __ret;
65215		}
65216		__ai int16_t vuqaddh_s16(int16_t __p0, uint16_t __p1) {
65217		int16_t __ret;
65218		__ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
65219		return __ret;
65220		}
65221	64684	#ifdef __LITTLE_ENDIAN__
65222		__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
65223		int8x16_t __ret;
65224		__ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
	64685	__ai __attribute__((target("fullfp16"))) float16x4_t vtrn2_f16(float16x4_t __p0, float16x4_t __p1) {
	64686	float16x4_t __ret;
	64687	__ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
65225	64688	return __ret;
65226	64689	}
65227	64690	#else
65228		__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
65229		int8x16_t __ret;
65230		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65231		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65232		__ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
65233		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	64691	__ai __attribute__((target("fullfp16"))) float16x4_t vtrn2_f16(float16x4_t __p0, float16x4_t __p1) {
	64692	float16x4_t __ret;
	64693	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64694	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64695	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
	64696	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65234	64697	return __ret;
65235	64698	}
65236	64699	#endif
65237	64700
65238	64701	#ifdef __LITTLE_ENDIAN__
65239		__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
65240		int32x4_t __ret;
65241		__ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
	64702	__ai __attribute__((target("fullfp16"))) float16x8_t vuzp1q_f16(float16x8_t __p0, float16x8_t __p1) {
	64703	float16x8_t __ret;
	64704	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65242	64705	return __ret;
65243	64706	}
65244	64707	#else
65245		__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
65246		int32x4_t __ret;
65247		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65248		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65249		__ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
65250		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	64708	__ai __attribute__((target("fullfp16"))) float16x8_t vuzp1q_f16(float16x8_t __p0, float16x8_t __p1) {
	64709	float16x8_t __ret;
	64710	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64711	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64712	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	64713	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65251	64714	return __ret;
65252	64715	}
65253	64716	#endif
65254	64717
65255	64718	#ifdef __LITTLE_ENDIAN__
65256		__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
65257		int64x2_t __ret;
65258		__ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
	64719	__ai __attribute__((target("fullfp16"))) float16x4_t vuzp1_f16(float16x4_t __p0, float16x4_t __p1) {
	64720	float16x4_t __ret;
	64721	__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65259	64722	return __ret;
65260	64723	}
65261	64724	#else
65262		__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
65263		int64x2_t __ret;
65264		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65265		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65266		__ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
65267		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64725	__ai __attribute__((target("fullfp16"))) float16x4_t vuzp1_f16(float16x4_t __p0, float16x4_t __p1) {
	64726	float16x4_t __ret;
	64727	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64728	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64729	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	64730	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65268	64731	return __ret;
65269	64732	}
65270	64733	#endif
65271	64734
65272	64735	#ifdef __LITTLE_ENDIAN__
65273		__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
65274		int16x8_t __ret;
65275		__ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
	64736	__ai __attribute__((target("fullfp16"))) float16x8_t vuzp2q_f16(float16x8_t __p0, float16x8_t __p1) {
	64737	float16x8_t __ret;
	64738	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65276	64739	return __ret;
65277	64740	}
65278	64741	#else
65279		__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
65280		int16x8_t __ret;
65281		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65282		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65283		__ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
	64742	__ai __attribute__((target("fullfp16"))) float16x8_t vuzp2q_f16(float16x8_t __p0, float16x8_t __p1) {
	64743	float16x8_t __ret;
	64744	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64745	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64746	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
65284	64747	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65285	64748	return __ret;
65286	64749	}
65287	64750	#endif
65288	64751
65289	64752	#ifdef __LITTLE_ENDIAN__
65290		__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
65291		int8x8_t __ret;
65292		__ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
	64753	__ai __attribute__((target("fullfp16"))) float16x4_t vuzp2_f16(float16x4_t __p0, float16x4_t __p1) {
	64754	float16x4_t __ret;
	64755	__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65293	64756	return __ret;
65294	64757	}
65295	64758	#else
65296		__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
65297		int8x8_t __ret;
65298		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65299		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65300		__ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
65301		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	64759	__ai __attribute__((target("fullfp16"))) float16x4_t vuzp2_f16(float16x4_t __p0, float16x4_t __p1) {
	64760	float16x4_t __ret;
	64761	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64762	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64763	__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	64764	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65302	64765	return __ret;
65303	64766	}
65304	64767	#endif
65305	64768
65306	64769	#ifdef __LITTLE_ENDIAN__
65307		__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
65308		int32x2_t __ret;
65309		__ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
	64770	__ai __attribute__((target("fullfp16"))) float16x8_t vzip1q_f16(float16x8_t __p0, float16x8_t __p1) {
	64771	float16x8_t __ret;
	64772	__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65310	64773	return __ret;
65311	64774	}
65312	64775	#else
65313		__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
65314		int32x2_t __ret;
65315		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65316		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65317		__ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
65318		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	64776	__ai __attribute__((target("fullfp16"))) float16x8_t vzip1q_f16(float16x8_t __p0, float16x8_t __p1) {
	64777	float16x8_t __ret;
	64778	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64779	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64780	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
	64781	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65319	64782	return __ret;
65320	64783	}
65321	64784	#endif
65322	64785
65323		__ai int64x1_t vuqadd_s64(int64x1_t __p0, uint64x1_t __p1) {
65324		int64x1_t __ret;
65325		__ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
65326		return __ret;
65327		}
65328	64786	#ifdef __LITTLE_ENDIAN__
65329		__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
65330		int16x4_t __ret;
65331		__ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
	64787	__ai __attribute__((target("fullfp16"))) float16x4_t vzip1_f16(float16x4_t __p0, float16x4_t __p1) {
	64788	float16x4_t __ret;
	64789	__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65332	64790	return __ret;
65333	64791	}
65334	64792	#else
65335		__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
65336		int16x4_t __ret;
65337		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65338		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65339		__ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
	64793	__ai __attribute__((target("fullfp16"))) float16x4_t vzip1_f16(float16x4_t __p0, float16x4_t __p1) {
	64794	float16x4_t __ret;
	64795	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64796	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64797	__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
65340	64798	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65341	64799	return __ret;
65342	64800	}
65343	64801	#endif
65344	64802
65345	64803	#ifdef __LITTLE_ENDIAN__
65346		#define vusdotq_laneq_s32(__p0_847, __p1_847, __p2_847, __p3_847) __extension__ ({ \
65347		int32x4_t __ret_847; \
65348		int32x4_t __s0_847 = __p0_847; \
65349		uint8x16_t __s1_847 = __p1_847; \
65350		int8x16_t __s2_847 = __p2_847; \
65351		int8x16_t __reint_847 = __s2_847; \
65352		__ret_847 = vusdotq_s32(__s0_847, __s1_847, (int8x16_t)(splatq_laneq_s32((int32x4_t ) &__reint_847, __p3_847))); \
65353		__ret_847; \
65354		})
65355		#else
65356		#define vusdotq_laneq_s32(__p0_848, __p1_848, __p2_848, __p3_848) __extension__ ({ \
65357		int32x4_t __ret_848; \
65358		int32x4_t __s0_848 = __p0_848; \
65359		uint8x16_t __s1_848 = __p1_848; \
65360		int8x16_t __s2_848 = __p2_848; \
65361		int32x4_t __rev0_848; __rev0_848 = __builtin_shufflevector(__s0_848, __s0_848, 3, 2, 1, 0); \
65362		uint8x16_t __rev1_848; __rev1_848 = __builtin_shufflevector(__s1_848, __s1_848, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
65363		int8x16_t __rev2_848; __rev2_848 = __builtin_shufflevector(__s2_848, __s2_848, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
65364		int8x16_t __reint_848 = __rev2_848; \
65365		__ret_848 = __noswap_vusdotq_s32(__rev0_848, __rev1_848, (int8x16_t)(__noswap_splatq_laneq_s32((int32x4_t ) &__reint_848, __p3_848))); \
65366		__ret_848 = __builtin_shufflevector(__ret_848, __ret_848, 3, 2, 1, 0); \
65367		__ret_848; \
65368		})
65369		#endif
65370
65371		#ifdef __LITTLE_ENDIAN__
65372		#define vusdot_laneq_s32(__p0_849, __p1_849, __p2_849, __p3_849) __extension__ ({ \
65373		int32x2_t __ret_849; \
65374		int32x2_t __s0_849 = __p0_849; \
65375		uint8x8_t __s1_849 = __p1_849; \
65376		int8x16_t __s2_849 = __p2_849; \
65377		int8x16_t __reint_849 = __s2_849; \
65378		__ret_849 = vusdot_s32(__s0_849, __s1_849, (int8x8_t)(splat_laneq_s32((int32x4_t ) &__reint_849, __p3_849))); \
65379		__ret_849; \
65380		})
65381		#else
65382		#define vusdot_laneq_s32(__p0_850, __p1_850, __p2_850, __p3_850) __extension__ ({ \
65383		int32x2_t __ret_850; \
65384		int32x2_t __s0_850 = __p0_850; \
65385		uint8x8_t __s1_850 = __p1_850; \
65386		int8x16_t __s2_850 = __p2_850; \
65387		int32x2_t __rev0_850; __rev0_850 = __builtin_shufflevector(__s0_850, __s0_850, 1, 0); \
65388		uint8x8_t __rev1_850; __rev1_850 = __builtin_shufflevector(__s1_850, __s1_850, 7, 6, 5, 4, 3, 2, 1, 0); \
65389		int8x16_t __rev2_850; __rev2_850 = __builtin_shufflevector(__s2_850, __s2_850, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
65390		int8x16_t __reint_850 = __rev2_850; \
65391		__ret_850 = __noswap_vusdot_s32(__rev0_850, __rev1_850, (int8x8_t)(__noswap_splat_laneq_s32((int32x4_t ) &__reint_850, __p3_850))); \
65392		__ret_850 = __builtin_shufflevector(__ret_850, __ret_850, 1, 0); \
65393		__ret_850; \
65394		})
65395		#endif
65396
65397		#ifdef __LITTLE_ENDIAN__
65398		__ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65399		poly8x8_t __ret;
65400		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
	64804	__ai __attribute__((target("fullfp16"))) float16x8_t vzip2q_f16(float16x8_t __p0, float16x8_t __p1) {
	64805	float16x8_t __ret;
	64806	__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
65401	64807	return __ret;
65402	64808	}
65403		#else
65404		__ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65405		poly8x8_t __ret;
65406		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65407		poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65408		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	64809	#else
	64810	__ai __attribute__((target("fullfp16"))) float16x8_t vzip2q_f16(float16x8_t __p0, float16x8_t __p1) {
	64811	float16x8_t __ret;
	64812	float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	64813	float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	64814	__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
65409	64815	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65410	64816	return __ret;
65411	64817	}
65412	64818	#endif
65413	64819
65414	64820	#ifdef __LITTLE_ENDIAN__
65415		__ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65416		poly16x4_t __ret;
65417		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
	64821	__ai __attribute__((target("fullfp16"))) float16x4_t vzip2_f16(float16x4_t __p0, float16x4_t __p1) {
	64822	float16x4_t __ret;
	64823	__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
65418	64824	return __ret;
65419	64825	}
65420	64826	#else
65421		__ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65422		poly16x4_t __ret;
65423		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65424		poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65425		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	64827	__ai __attribute__((target("fullfp16"))) float16x4_t vzip2_f16(float16x4_t __p0, float16x4_t __p1) {
	64828	float16x4_t __ret;
	64829	float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	64830	float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	64831	__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
65426	64832	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65427	64833	return __ret;
65428	64834	}
65429	64835	#endif
65430	64836
65431	64837	#ifdef __LITTLE_ENDIAN__
65432		__ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65433		poly8x16_t __ret;
65434		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
65435		return __ret;
65436		}
	64838	#define vsudotq_laneq_s32(__p0_779, __p1_779, __p2_779, __p3_779) __extension__ ({ \
	64839	int32x4_t __ret_779; \
	64840	int32x4_t __s0_779 = __p0_779; \
	64841	int8x16_t __s1_779 = __p1_779; \
	64842	uint8x16_t __s2_779 = __p2_779; \
	64843	uint8x16_t __reint_779 = __s2_779; \
	64844	__ret_779 = vusdotq_s32(__s0_779, (uint8x16_t)(splatq_laneq_s32((int32x4_t ) &__reint_779, __p3_779)), __s1_779); \
	64845	__ret_779; \
	64846	})
65437	64847	#else
65438		__ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65439		poly8x16_t __ret;
65440		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65441		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65442		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
65443		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65444		return __ret;
65445		}
	64848	#define vsudotq_laneq_s32(__p0_780, __p1_780, __p2_780, __p3_780) __extension__ ({ \
	64849	int32x4_t __ret_780; \
	64850	int32x4_t __s0_780 = __p0_780; \
	64851	int8x16_t __s1_780 = __p1_780; \
	64852	uint8x16_t __s2_780 = __p2_780; \
	64853	int32x4_t __rev0_780; __rev0_780 = __builtin_shufflevector(__s0_780, __s0_780, 3, 2, 1, 0); \
	64854	int8x16_t __rev1_780; __rev1_780 = __builtin_shufflevector(__s1_780, __s1_780, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	64855	uint8x16_t __rev2_780; __rev2_780 = __builtin_shufflevector(__s2_780, __s2_780, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	64856	uint8x16_t __reint_780 = __rev2_780; \
	64857	__ret_780 = __noswap_vusdotq_s32(__rev0_780, (uint8x16_t)(__noswap_splatq_laneq_s32((int32x4_t ) &__reint_780, __p3_780)), __rev1_780); \
	64858	__ret_780 = __builtin_shufflevector(__ret_780, __ret_780, 3, 2, 1, 0); \
	64859	__ret_780; \
	64860	})
65446	64861	#endif
65447	64862
65448	64863	#ifdef __LITTLE_ENDIAN__
65449		__ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65450		poly64x2_t __ret;
65451		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65452		return __ret;
65453		}
	64864	#define vsudot_laneq_s32(__p0_781, __p1_781, __p2_781, __p3_781) __extension__ ({ \
	64865	int32x2_t __ret_781; \
	64866	int32x2_t __s0_781 = __p0_781; \
	64867	int8x8_t __s1_781 = __p1_781; \
	64868	uint8x16_t __s2_781 = __p2_781; \
	64869	uint8x16_t __reint_781 = __s2_781; \
	64870	__ret_781 = vusdot_s32(__s0_781, (uint8x8_t)(splat_laneq_s32((int32x4_t ) &__reint_781, __p3_781)), __s1_781); \
	64871	__ret_781; \
	64872	})
65454	64873	#else
65455		__ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65456		poly64x2_t __ret;
65457		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65458		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65459		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
65460		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65461		return __ret;
65462		}
	64874	#define vsudot_laneq_s32(__p0_782, __p1_782, __p2_782, __p3_782) __extension__ ({ \
	64875	int32x2_t __ret_782; \
	64876	int32x2_t __s0_782 = __p0_782; \
	64877	int8x8_t __s1_782 = __p1_782; \
	64878	uint8x16_t __s2_782 = __p2_782; \
	64879	int32x2_t __rev0_782; __rev0_782 = __builtin_shufflevector(__s0_782, __s0_782, 1, 0); \
	64880	int8x8_t __rev1_782; __rev1_782 = __builtin_shufflevector(__s1_782, __s1_782, 7, 6, 5, 4, 3, 2, 1, 0); \
	64881	uint8x16_t __rev2_782; __rev2_782 = __builtin_shufflevector(__s2_782, __s2_782, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	64882	uint8x16_t __reint_782 = __rev2_782; \
	64883	__ret_782 = __noswap_vusdot_s32(__rev0_782, (uint8x8_t)(__noswap_splat_laneq_s32((int32x4_t ) &__reint_782, __p3_782)), __rev1_782); \
	64884	__ret_782 = __builtin_shufflevector(__ret_782, __ret_782, 1, 0); \
	64885	__ret_782; \
	64886	})
65463	64887	#endif
65464	64888
65465	64889	#ifdef __LITTLE_ENDIAN__
65466		__ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65467		poly16x8_t __ret;
65468		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65469		return __ret;
65470		}
	64890	#define vusdotq_laneq_s32(__p0_783, __p1_783, __p2_783, __p3_783) __extension__ ({ \
	64891	int32x4_t __ret_783; \
	64892	int32x4_t __s0_783 = __p0_783; \
	64893	uint8x16_t __s1_783 = __p1_783; \
	64894	int8x16_t __s2_783 = __p2_783; \
	64895	int8x16_t __reint_783 = __s2_783; \
	64896	__ret_783 = vusdotq_s32(__s0_783, __s1_783, (int8x16_t)(splatq_laneq_s32((int32x4_t ) &__reint_783, __p3_783))); \
	64897	__ret_783; \
	64898	})
65471	64899	#else
65472		__ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65473		poly16x8_t __ret;
65474		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65475		poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65476		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
65477		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65478		return __ret;
65479		}
	64900	#define vusdotq_laneq_s32(__p0_784, __p1_784, __p2_784, __p3_784) __extension__ ({ \
	64901	int32x4_t __ret_784; \
	64902	int32x4_t __s0_784 = __p0_784; \
	64903	uint8x16_t __s1_784 = __p1_784; \
	64904	int8x16_t __s2_784 = __p2_784; \
	64905	int32x4_t __rev0_784; __rev0_784 = __builtin_shufflevector(__s0_784, __s0_784, 3, 2, 1, 0); \
	64906	uint8x16_t __rev1_784; __rev1_784 = __builtin_shufflevector(__s1_784, __s1_784, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	64907	int8x16_t __rev2_784; __rev2_784 = __builtin_shufflevector(__s2_784, __s2_784, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	64908	int8x16_t __reint_784 = __rev2_784; \
	64909	__ret_784 = __noswap_vusdotq_s32(__rev0_784, __rev1_784, (int8x16_t)(__noswap_splatq_laneq_s32((int32x4_t ) &__reint_784, __p3_784))); \
	64910	__ret_784 = __builtin_shufflevector(__ret_784, __ret_784, 3, 2, 1, 0); \
	64911	__ret_784; \
	64912	})
65480	64913	#endif
65481	64914
65482	64915	#ifdef __LITTLE_ENDIAN__
65483		__ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	64916	#define vusdot_laneq_s32(__p0_785, __p1_785, __p2_785, __p3_785) __extension__ ({ \
	64917	int32x2_t __ret_785; \
	64918	int32x2_t __s0_785 = __p0_785; \
	64919	uint8x8_t __s1_785 = __p1_785; \
	64920	int8x16_t __s2_785 = __p2_785; \
	64921	int8x16_t __reint_785 = __s2_785; \
	64922	__ret_785 = vusdot_s32(__s0_785, __s1_785, (int8x8_t)(splat_laneq_s32((int32x4_t ) &__reint_785, __p3_785))); \
	64923	__ret_785; \
	64924	})
	64925	#else
	64926	#define vusdot_laneq_s32(__p0_786, __p1_786, __p2_786, __p3_786) __extension__ ({ \
	64927	int32x2_t __ret_786; \
	64928	int32x2_t __s0_786 = __p0_786; \
	64929	uint8x8_t __s1_786 = __p1_786; \
	64930	int8x16_t __s2_786 = __p2_786; \
	64931	int32x2_t __rev0_786; __rev0_786 = __builtin_shufflevector(__s0_786, __s0_786, 1, 0); \
	64932	uint8x8_t __rev1_786; __rev1_786 = __builtin_shufflevector(__s1_786, __s1_786, 7, 6, 5, 4, 3, 2, 1, 0); \
	64933	int8x16_t __rev2_786; __rev2_786 = __builtin_shufflevector(__s2_786, __s2_786, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	64934	int8x16_t __reint_786 = __rev2_786; \
	64935	__ret_786 = __noswap_vusdot_s32(__rev0_786, __rev1_786, (int8x8_t)(__noswap_splat_laneq_s32((int32x4_t ) &__reint_786, __p3_786))); \
	64936	__ret_786 = __builtin_shufflevector(__ret_786, __ret_786, 1, 0); \
	64937	__ret_786; \
	64938	})
	64939	#endif
	64940
	64941	#ifdef __LITTLE_ENDIAN__
	64942	__ai __attribute__((target("sha3"))) uint8x16_t vbcaxq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
65484	64943	uint8x16_t __ret;
65485		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	64944	__ret = (uint8x16_t) __builtin_neon_vbcaxq_u8((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
65486	64945	return __ret;
65487	64946	}
65488	64947	#else
65489		__ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
	64948	__ai __attribute__((target("sha3"))) uint8x16_t vbcaxq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
65490	64949	uint8x16_t __ret;
65491	64950	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65492	64951	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65493		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	64952	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	64953	__ret = (uint8x16_t) __builtin_neon_vbcaxq_u8((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
65494	64954	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65495	64955	return __ret;
65496	64956	}
65497	64957	#endif
65498	64958
65499	64959	#ifdef __LITTLE_ENDIAN__
65500		__ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	64960	__ai __attribute__((target("sha3"))) uint32x4_t vbcaxq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
65501	64961	uint32x4_t __ret;
65502		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
	64962	__ret = (uint32x4_t) __builtin_neon_vbcaxq_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
65503	64963	return __ret;
65504	64964	}
65505	64965	#else
65506		__ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	64966	__ai __attribute__((target("sha3"))) uint32x4_t vbcaxq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
65507	64967	uint32x4_t __ret;
65508	64968	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65509	64969	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65510		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	64970	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	64971	__ret = (uint32x4_t) __builtin_neon_vbcaxq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
65511	64972	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65512	64973	return __ret;
65513	64974	}
65514	64975	#endif
65515	64976
65516	64977	#ifdef __LITTLE_ENDIAN__
65517		__ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	64978	__ai __attribute__((target("sha3"))) uint64x2_t vbcaxq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
65518	64979	uint64x2_t __ret;
65519		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	64980	__ret = (uint64x2_t) __builtin_neon_vbcaxq_u64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
65520	64981	return __ret;
65521	64982	}
65522	64983	#else
65523		__ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	64984	__ai __attribute__((target("sha3"))) uint64x2_t vbcaxq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
65524	64985	uint64x2_t __ret;
65525	64986	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65526	64987	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65527		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	64988	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	64989	__ret = (uint64x2_t) __builtin_neon_vbcaxq_u64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
65528	64990	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65529	64991	return __ret;
65530	64992	}
65531	64993	#endif
65532	64994
65533	64995	#ifdef __LITTLE_ENDIAN__
65534		__ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	64996	__ai __attribute__((target("sha3"))) uint16x8_t vbcaxq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
65535	64997	uint16x8_t __ret;
65536		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
	64998	__ret = (uint16x8_t) __builtin_neon_vbcaxq_u16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 49);
65537	64999	return __ret;
65538	65000	}
65539	65001	#else
65540		__ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
	65002	__ai __attribute__((target("sha3"))) uint16x8_t vbcaxq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
65541	65003	uint16x8_t __ret;
65542	65004	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65543	65005	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65544		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	65006	uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	65007	__ret = (uint16x8_t) __builtin_neon_vbcaxq_u16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
65545	65008	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65546	65009	return __ret;
65547	65010	}
65548	65011	#endif
65549	65012
65550	65013	#ifdef __LITTLE_ENDIAN__
65551		__ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
	65014	__ai __attribute__((target("sha3"))) int8x16_t vbcaxq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
65552	65015	int8x16_t __ret;
65553		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	65016	__ret = (int8x16_t) __builtin_neon_vbcaxq_s8((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
65554	65017	return __ret;
65555	65018	}
65556	65019	#else
65557		__ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
	65020	__ai __attribute__((target("sha3"))) int8x16_t vbcaxq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
65558	65021	int8x16_t __ret;
65559	65022	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65560	65023	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65561		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
	65024	int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65025	__ret = (int8x16_t) __builtin_neon_vbcaxq_s8((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
65562	65026	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65563	65027	return __ret;
65564	65028	}
65565	65029	#endif
65566	65030
65567	65031	#ifdef __LITTLE_ENDIAN__
65568		__ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
65569		float64x2_t __ret;
65570		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65571		return __ret;
65572		}
65573		#else
65574		__ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
65575		float64x2_t __ret;
65576		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65577		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65578		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
65579		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65580		return __ret;
65581		}
65582		#endif
65583
65584		#ifdef __LITTLE_ENDIAN__
65585		__ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
65586		float32x4_t __ret;
65587		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65588		return __ret;
65589		}
65590		#else
65591		__ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
65592		float32x4_t __ret;
65593		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65594		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65595		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
65596		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65597		return __ret;
65598		}
65599		#endif
65600
65601		#ifdef __LITTLE_ENDIAN__
65602		__ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
	65032	__ai __attribute__((target("sha3"))) int32x4_t vbcaxq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
65603	65033	int32x4_t __ret;
65604		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
	65034	__ret = (int32x4_t) __builtin_neon_vbcaxq_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
65605	65035	return __ret;
65606	65036	}
65607	65037	#else
65608		__ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
	65038	__ai __attribute__((target("sha3"))) int32x4_t vbcaxq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
65609	65039	int32x4_t __ret;
65610	65040	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65611	65041	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65612		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
	65042	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	65043	__ret = (int32x4_t) __builtin_neon_vbcaxq_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
65613	65044	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65614	65045	return __ret;
65615	65046	}
65616	65047	#endif
65617	65048
65618	65049	#ifdef __LITTLE_ENDIAN__
65619		__ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
	65050	__ai __attribute__((target("sha3"))) int64x2_t vbcaxq_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
65620	65051	int64x2_t __ret;
65621		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	65052	__ret = (int64x2_t) __builtin_neon_vbcaxq_s64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 35);
65622	65053	return __ret;
65623	65054	}
65624	65055	#else
65625		__ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
	65056	__ai __attribute__((target("sha3"))) int64x2_t vbcaxq_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
65626	65057	int64x2_t __ret;
65627	65058	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65628	65059	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65629		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	65060	int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65061	__ret = (int64x2_t) __builtin_neon_vbcaxq_s64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
65630	65062	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65631	65063	return __ret;
65632	65064	}
65633	65065	#endif
65634	65066
65635	65067	#ifdef __LITTLE_ENDIAN__
65636		__ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
	65068	__ai __attribute__((target("sha3"))) int16x8_t vbcaxq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
65637	65069	int16x8_t __ret;
65638		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
	65070	__ret = (int16x8_t) __builtin_neon_vbcaxq_s16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
65639	65071	return __ret;
65640	65072	}
65641	65073	#else
65642		__ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
	65074	__ai __attribute__((target("sha3"))) int16x8_t vbcaxq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
65643	65075	int16x8_t __ret;
65644	65076	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65645	65077	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65646		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	65078	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	65079	__ret = (int16x8_t) __builtin_neon_vbcaxq_s16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
65647	65080	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65648	65081	return __ret;
65649	65082	}
65650	65083	#endif
65651	65084
65652	65085	#ifdef __LITTLE_ENDIAN__
65653		__ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65654		uint8x8_t __ret;
65655		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
	65086	__ai __attribute__((target("sha3"))) uint8x16_t veor3q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	65087	uint8x16_t __ret;
	65088	__ret = (uint8x16_t) __builtin_neon_veor3q_u8((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
65656	65089	return __ret;
65657	65090	}
65658	65091	#else
65659		__ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65660		uint8x8_t __ret;
65661		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65662		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65663		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
65664		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	65092	__ai __attribute__((target("sha3"))) uint8x16_t veor3q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
	65093	uint8x16_t __ret;
	65094	uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65095	uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65096	uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65097	__ret = (uint8x16_t) __builtin_neon_veor3q_u8((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
	65098	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65665	65099	return __ret;
65666	65100	}
65667	65101	#endif
65668	65102
65669	65103	#ifdef __LITTLE_ENDIAN__
65670		__ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65671		uint32x2_t __ret;
65672		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	65104	__ai __attribute__((target("sha3"))) uint32x4_t veor3q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	65105	uint32x4_t __ret;
	65106	__ret = (uint32x4_t) __builtin_neon_veor3q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
65673	65107	return __ret;
65674	65108	}
65675	65109	#else
65676		__ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65677		uint32x2_t __ret;
65678		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65679		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65680		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
65681		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	65110	__ai __attribute__((target("sha3"))) uint32x4_t veor3q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	65111	uint32x4_t __ret;
	65112	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	65113	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	65114	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	65115	__ret = (uint32x4_t) __builtin_neon_veor3q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	65116	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65682	65117	return __ret;
65683	65118	}
65684	65119	#endif
65685	65120
65686	65121	#ifdef __LITTLE_ENDIAN__
65687		__ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65688		uint16x4_t __ret;
65689		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
	65122	__ai __attribute__((target("sha3"))) uint64x2_t veor3q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65123	uint64x2_t __ret;
	65124	__ret = (uint64x2_t) __builtin_neon_veor3q_u64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
65690	65125	return __ret;
65691	65126	}
65692	65127	#else
65693		__ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65694		uint16x4_t __ret;
65695		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65696		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65697		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
65698		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	65128	__ai __attribute__((target("sha3"))) uint64x2_t veor3q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65129	uint64x2_t __ret;
	65130	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65131	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65132	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65133	__ret = (uint64x2_t) __builtin_neon_veor3q_u64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
	65134	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65699	65135	return __ret;
65700	65136	}
65701	65137	#endif
65702	65138
65703	65139	#ifdef __LITTLE_ENDIAN__
65704		__ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
65705		int8x8_t __ret;
65706		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
	65140	__ai __attribute__((target("sha3"))) uint16x8_t veor3q_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	65141	uint16x8_t __ret;
	65142	__ret = (uint16x8_t) __builtin_neon_veor3q_u16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 49);
65707	65143	return __ret;
65708	65144	}
65709	65145	#else
65710		__ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
65711		int8x8_t __ret;
65712		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65713		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65714		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
	65146	__ai __attribute__((target("sha3"))) uint16x8_t veor3q_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
	65147	uint16x8_t __ret;
	65148	uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	65149	uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	65150	uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	65151	__ret = (uint16x8_t) __builtin_neon_veor3q_u16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
65715	65152	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65716	65153	return __ret;
65717	65154	}
65718	65155	#endif
65719	65156
65720	65157	#ifdef __LITTLE_ENDIAN__
65721		__ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
65722		float32x2_t __ret;
65723		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	65158	__ai __attribute__((target("sha3"))) int8x16_t veor3q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
	65159	int8x16_t __ret;
	65160	__ret = (int8x16_t) __builtin_neon_veor3q_s8((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
65724	65161	return __ret;
65725	65162	}
65726	65163	#else
65727		__ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
65728		float32x2_t __ret;
65729		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65730		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65731		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
65732		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	65164	__ai __attribute__((target("sha3"))) int8x16_t veor3q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
	65165	int8x16_t __ret;
	65166	int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65167	int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65168	int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65169	__ret = (int8x16_t) __builtin_neon_veor3q_s8((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
	65170	__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65733	65171	return __ret;
65734	65172	}
65735	65173	#endif
65736	65174
65737	65175	#ifdef __LITTLE_ENDIAN__
65738		__ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
65739		int32x2_t __ret;
65740		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	65176	__ai __attribute__((target("sha3"))) int32x4_t veor3q_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	65177	int32x4_t __ret;
	65178	__ret = (int32x4_t) __builtin_neon_veor3q_s32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
65741	65179	return __ret;
65742	65180	}
65743	65181	#else
65744		__ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
65745		int32x2_t __ret;
65746		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65747		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65748		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
65749		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	65182	__ai __attribute__((target("sha3"))) int32x4_t veor3q_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
	65183	int32x4_t __ret;
	65184	int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	65185	int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	65186	int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	65187	__ret = (int32x4_t) __builtin_neon_veor3q_s32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
	65188	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65750	65189	return __ret;
65751	65190	}
65752	65191	#endif
65753	65192
65754	65193	#ifdef __LITTLE_ENDIAN__
65755		__ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
65756		int16x4_t __ret;
65757		__ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
	65194	__ai __attribute__((target("sha3"))) int64x2_t veor3q_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	65195	int64x2_t __ret;
	65196	__ret = (int64x2_t) __builtin_neon_veor3q_s64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 35);
65758	65197	return __ret;
65759	65198	}
65760	65199	#else
65761		__ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
65762		int16x4_t __ret;
65763		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65764		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65765		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
65766		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	65200	__ai __attribute__((target("sha3"))) int64x2_t veor3q_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
	65201	int64x2_t __ret;
	65202	int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65203	int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65204	int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65205	__ret = (int64x2_t) __builtin_neon_veor3q_s64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
	65206	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65767	65207	return __ret;
65768	65208	}
65769	65209	#endif
65770	65210
65771	65211	#ifdef __LITTLE_ENDIAN__
65772		__ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65773		poly8x8_t __ret;
65774		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
	65212	__ai __attribute__((target("sha3"))) int16x8_t veor3q_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	65213	int16x8_t __ret;
	65214	__ret = (int16x8_t) __builtin_neon_veor3q_s16((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
65775	65215	return __ret;
65776	65216	}
65777	65217	#else
65778		__ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65779		poly8x8_t __ret;
65780		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65781		poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65782		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
	65218	__ai __attribute__((target("sha3"))) int16x8_t veor3q_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
	65219	int16x8_t __ret;
	65220	int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
	65221	int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
	65222	int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
	65223	__ret = (int16x8_t) __builtin_neon_veor3q_s16((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
65783	65224	__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
65784	65225	return __ret;
65785	65226	}
65786	65227	#endif
65787	65228
65788	65229	#ifdef __LITTLE_ENDIAN__
65789		__ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65790		poly16x4_t __ret;
65791		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	65230	__ai __attribute__((target("sha3"))) uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	65231	uint64x2_t __ret;
	65232	__ret = (uint64x2_t) __builtin_neon_vrax1q_u64((int8x16_t)__p0, (int8x16_t)__p1, 51);
65792	65233	return __ret;
65793	65234	}
65794	65235	#else
65795		__ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65796		poly16x4_t __ret;
65797		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65798		poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65799		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
65800		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	65236	__ai __attribute__((target("sha3"))) uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	65237	uint64x2_t __ret;
	65238	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65239	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65240	__ret = (uint64x2_t) __builtin_neon_vrax1q_u64((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	65241	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65801	65242	return __ret;
65802	65243	}
65803	65244	#endif
65804	65245
65805	65246	#ifdef __LITTLE_ENDIAN__
65806		__ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65807		poly8x16_t __ret;
65808		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	65247	__ai __attribute__((target("sha3"))) uint64x2_t vsha512hq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65248	uint64x2_t __ret;
	65249	__ret = (uint64x2_t) __builtin_neon_vsha512hq_u64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
65809	65250	return __ret;
65810	65251	}
65811	65252	#else
65812		__ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65813		poly8x16_t __ret;
65814		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65815		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65816		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65817		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65253	__ai __attribute__((target("sha3"))) uint64x2_t vsha512hq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65254	uint64x2_t __ret;
	65255	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65256	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65257	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65258	__ret = (uint64x2_t) __builtin_neon_vsha512hq_u64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
	65259	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65818	65260	return __ret;
65819	65261	}
65820	65262	#endif
65821	65263
65822	65264	#ifdef __LITTLE_ENDIAN__
65823		__ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65824		poly64x2_t __ret;
65825		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	65265	__ai __attribute__((target("sha3"))) uint64x2_t vsha512h2q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65266	uint64x2_t __ret;
	65267	__ret = (uint64x2_t) __builtin_neon_vsha512h2q_u64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
65826	65268	return __ret;
65827	65269	}
65828	65270	#else
65829		__ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65830		poly64x2_t __ret;
65831		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65832		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65833		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	65271	__ai __attribute__((target("sha3"))) uint64x2_t vsha512h2q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65272	uint64x2_t __ret;
	65273	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65274	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65275	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65276	__ret = (uint64x2_t) __builtin_neon_vsha512h2q_u64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
65834	65277	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65835	65278	return __ret;
65836	65279	}
65837	65280	#endif
65838	65281
65839	65282	#ifdef __LITTLE_ENDIAN__
65840		__ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65841		poly16x8_t __ret;
65842		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
	65283	__ai __attribute__((target("sha3"))) uint64x2_t vsha512su0q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	65284	uint64x2_t __ret;
	65285	__ret = (uint64x2_t) __builtin_neon_vsha512su0q_u64((int8x16_t)__p0, (int8x16_t)__p1, 51);
65843	65286	return __ret;
65844	65287	}
65845	65288	#else
65846		__ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65847		poly16x8_t __ret;
65848		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65849		poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65850		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
65851		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	65289	__ai __attribute__((target("sha3"))) uint64x2_t vsha512su0q_u64(uint64x2_t __p0, uint64x2_t __p1) {
	65290	uint64x2_t __ret;
	65291	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65292	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65293	__ret = (uint64x2_t) __builtin_neon_vsha512su0q_u64((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
	65294	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65852	65295	return __ret;
65853	65296	}
65854	65297	#endif
65855	65298
65856	65299	#ifdef __LITTLE_ENDIAN__
65857		__ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65858		uint8x16_t __ret;
65859		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
	65300	__ai __attribute__((target("sha3"))) uint64x2_t vsha512su1q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65301	uint64x2_t __ret;
	65302	__ret = (uint64x2_t) __builtin_neon_vsha512su1q_u64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
65860	65303	return __ret;
65861	65304	}
65862	65305	#else
65863		__ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65864		uint8x16_t __ret;
65865		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65866		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65867		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65868		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65306	__ai __attribute__((target("sha3"))) uint64x2_t vsha512su1q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
	65307	uint64x2_t __ret;
	65308	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65309	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65310	uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65311	__ret = (uint64x2_t) __builtin_neon_vsha512su1q_u64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
	65312	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65869	65313	return __ret;
65870	65314	}
65871	65315	#endif
65872	65316
65873	65317	#ifdef __LITTLE_ENDIAN__
65874		__ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	65318	#define vxarq_u64(__p0, __p1, __p2) __extension__ ({ \
	65319	uint64x2_t __ret; \
	65320	uint64x2_t __s0 = __p0; \
	65321	uint64x2_t __s1 = __p1; \
	65322	__ret = (uint64x2_t) __builtin_neon_vxarq_u64((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
	65323	__ret; \
	65324	})
	65325	#else
	65326	#define vxarq_u64(__p0, __p1, __p2) __extension__ ({ \
	65327	uint64x2_t __ret; \
	65328	uint64x2_t __s0 = __p0; \
	65329	uint64x2_t __s1 = __p1; \
	65330	uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
	65331	uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
	65332	__ret = (uint64x2_t) __builtin_neon_vxarq_u64((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
	65333	__ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
	65334	__ret; \
	65335	})
	65336	#endif
	65337
	65338	#ifdef __LITTLE_ENDIAN__
	65339	__ai __attribute__((target("sm4"))) uint32x4_t vsm3partw1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
65875	65340	uint32x4_t __ret;
65876		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	65341	__ret = (uint32x4_t) __builtin_neon_vsm3partw1q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
65877	65342	return __ret;
65878	65343	}
65879	65344	#else
65880		__ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
	65345	__ai __attribute__((target("sm4"))) uint32x4_t vsm3partw1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
65881	65346	uint32x4_t __ret;
65882	65347	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65883	65348	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65884		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	65349	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	65350	__ret = (uint32x4_t) __builtin_neon_vsm3partw1q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
65885	65351	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65886	65352	return __ret;
65887	65353	}
65888	65354	#endif
65889	65355
65890	65356	#ifdef __LITTLE_ENDIAN__
65891		__ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65892		uint64x2_t __ret;
65893		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	65357	__ai __attribute__((target("sm4"))) uint32x4_t vsm3partw2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	65358	uint32x4_t __ret;
	65359	__ret = (uint32x4_t) __builtin_neon_vsm3partw2q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
65894	65360	return __ret;
65895	65361	}
65896	65362	#else
65897		__ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65898		uint64x2_t __ret;
65899		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65900		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65901		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
65902		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	65363	__ai __attribute__((target("sm4"))) uint32x4_t vsm3partw2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	65364	uint32x4_t __ret;
	65365	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	65366	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	65367	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	65368	__ret = (uint32x4_t) __builtin_neon_vsm3partw2q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	65369	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65903	65370	return __ret;
65904	65371	}
65905	65372	#endif
65906	65373
65907	65374	#ifdef __LITTLE_ENDIAN__
65908		__ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65909		uint16x8_t __ret;
65910		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
	65375	__ai __attribute__((target("sm4"))) uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	65376	uint32x4_t __ret;
	65377	__ret = (uint32x4_t) __builtin_neon_vsm3ss1q_u32((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
65911	65378	return __ret;
65912	65379	}
65913	65380	#else
65914		__ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65915		uint16x8_t __ret;
65916		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
65917		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65918		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
65919		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	65381	__ai __attribute__((target("sm4"))) uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
	65382	uint32x4_t __ret;
	65383	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	65384	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	65385	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
	65386	__ret = (uint32x4_t) __builtin_neon_vsm3ss1q_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
	65387	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65920	65388	return __ret;
65921	65389	}
65922	65390	#endif
65923	65391
65924	65392	#ifdef __LITTLE_ENDIAN__
65925		__ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65926		int8x16_t __ret;
65927		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65928		return __ret;
65929		}
	65393	#define vsm3tt1aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65394	uint32x4_t __ret; \
	65395	uint32x4_t __s0 = __p0; \
	65396	uint32x4_t __s1 = __p1; \
	65397	uint32x4_t __s2 = __p2; \
	65398	__ret = (uint32x4_t) __builtin_neon_vsm3tt1aq_u32((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
	65399	__ret; \
	65400	})
65930	65401	#else
65931		__ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65932		int8x16_t __ret;
65933		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65934		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65935		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65936		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65937		return __ret;
65938		}
	65402	#define vsm3tt1aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65403	uint32x4_t __ret; \
	65404	uint32x4_t __s0 = __p0; \
	65405	uint32x4_t __s1 = __p1; \
	65406	uint32x4_t __s2 = __p2; \
	65407	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	65408	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	65409	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	65410	__ret = (uint32x4_t) __builtin_neon_vsm3tt1aq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
	65411	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	65412	__ret; \
	65413	})
65939	65414	#endif
65940	65415
65941	65416	#ifdef __LITTLE_ENDIAN__
65942		__ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65943		float64x2_t __ret;
65944		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65945		return __ret;
65946		}
	65417	#define vsm3tt1bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65418	uint32x4_t __ret; \
	65419	uint32x4_t __s0 = __p0; \
	65420	uint32x4_t __s1 = __p1; \
	65421	uint32x4_t __s2 = __p2; \
	65422	__ret = (uint32x4_t) __builtin_neon_vsm3tt1bq_u32((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
	65423	__ret; \
	65424	})
65947	65425	#else
65948		__ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65949		float64x2_t __ret;
65950		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
65951		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65952		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
65953		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
65954		return __ret;
65955		}
	65426	#define vsm3tt1bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65427	uint32x4_t __ret; \
	65428	uint32x4_t __s0 = __p0; \
	65429	uint32x4_t __s1 = __p1; \
	65430	uint32x4_t __s2 = __p2; \
	65431	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	65432	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	65433	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	65434	__ret = (uint32x4_t) __builtin_neon_vsm3tt1bq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
	65435	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	65436	__ret; \
	65437	})
65956	65438	#endif
65957	65439
65958	65440	#ifdef __LITTLE_ENDIAN__
65959		__ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65960		float32x4_t __ret;
65961		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65962		return __ret;
65963		}
	65441	#define vsm3tt2aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65442	uint32x4_t __ret; \
	65443	uint32x4_t __s0 = __p0; \
	65444	uint32x4_t __s1 = __p1; \
	65445	uint32x4_t __s2 = __p2; \
	65446	__ret = (uint32x4_t) __builtin_neon_vsm3tt2aq_u32((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
	65447	__ret; \
	65448	})
65964	65449	#else
65965		__ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65966		float32x4_t __ret;
65967		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65968		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65969		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
65970		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65971		return __ret;
65972		}
	65450	#define vsm3tt2aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65451	uint32x4_t __ret; \
	65452	uint32x4_t __s0 = __p0; \
	65453	uint32x4_t __s1 = __p1; \
	65454	uint32x4_t __s2 = __p2; \
	65455	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	65456	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	65457	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	65458	__ret = (uint32x4_t) __builtin_neon_vsm3tt2aq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
	65459	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	65460	__ret; \
	65461	})
	65462	#endif
	65463
	65464	#ifdef __LITTLE_ENDIAN__
	65465	#define vsm3tt2bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65466	uint32x4_t __ret; \
	65467	uint32x4_t __s0 = __p0; \
	65468	uint32x4_t __s1 = __p1; \
	65469	uint32x4_t __s2 = __p2; \
	65470	__ret = (uint32x4_t) __builtin_neon_vsm3tt2bq_u32((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
	65471	__ret; \
	65472	})
	65473	#else
	65474	#define vsm3tt2bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
	65475	uint32x4_t __ret; \
	65476	uint32x4_t __s0 = __p0; \
	65477	uint32x4_t __s1 = __p1; \
	65478	uint32x4_t __s2 = __p2; \
	65479	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
	65480	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
	65481	uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
	65482	__ret = (uint32x4_t) __builtin_neon_vsm3tt2bq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
	65483	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
	65484	__ret; \
	65485	})
65973	65486	#endif
65974	65487
65975	65488	#ifdef __LITTLE_ENDIAN__
65976		__ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65977		int32x4_t __ret;
65978		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
	65489	__ai __attribute__((target("sm4"))) uint32x4_t vsm4eq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	65490	uint32x4_t __ret;
	65491	__ret = (uint32x4_t) __builtin_neon_vsm4eq_u32((int8x16_t)__p0, (int8x16_t)__p1, 50);
65979	65492	return __ret;
65980	65493	}
65981	65494	#else
65982		__ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65983		int32x4_t __ret;
65984		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
65985		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65986		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
	65495	__ai __attribute__((target("sm4"))) uint32x4_t vsm4eq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	65496	uint32x4_t __ret;
	65497	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	65498	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	65499	__ret = (uint32x4_t) __builtin_neon_vsm4eq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
65987	65500	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
65988	65501	return __ret;
65989	65502	}
65990	65503	#endif
65991	65504
65992	65505	#ifdef __LITTLE_ENDIAN__
65993		__ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
65994		int64x2_t __ret;
65995		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	65506	__ai __attribute__((target("sm4"))) uint32x4_t vsm4ekeyq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	65507	uint32x4_t __ret;
	65508	__ret = (uint32x4_t) __builtin_neon_vsm4ekeyq_u32((int8x16_t)__p0, (int8x16_t)__p1, 50);
65996	65509	return __ret;
65997	65510	}
65998	65511	#else
65999		__ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
66000		int64x2_t __ret;
66001		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66002		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66003		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
66004		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	65512	__ai __attribute__((target("sm4"))) uint32x4_t vsm4ekeyq_u32(uint32x4_t __p0, uint32x4_t __p1) {
	65513	uint32x4_t __ret;
	65514	uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	65515	uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
	65516	__ret = (uint32x4_t) __builtin_neon_vsm4ekeyq_u32((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
	65517	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66005	65518	return __ret;
66006	65519	}
66007	65520	#endif
66008	65521
66009		#ifdef __LITTLE_ENDIAN__
66010		__ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
66011		int16x8_t __ret;
66012		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
	65522	__ai __attribute__((target("v8.1a"))) int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
	65523	int32_t __ret;
	65524	__ret = (int32_t) __builtin_neon_vqrdmlahs_s32(__p0, __p1, __p2);
66013	65525	return __ret;
66014	65526	}
66015		#else
66016		__ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
66017		int16x8_t __ret;
66018		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66019		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66020		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
66021		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	65527	__ai __attribute__((target("v8.1a"))) int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
	65528	int16_t __ret;
	65529	__ret = (int16_t) __builtin_neon_vqrdmlahh_s16(__p0, __p1, __p2);
66022	65530	return __ret;
66023	65531	}
66024		#endif
66025
66026	65532	#ifdef __LITTLE_ENDIAN__
66027		__ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66028		uint8x8_t __ret;
66029		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
66030		return __ret;
66031		}
	65533	#define vqrdmlahs_lane_s32(__p0_787, __p1_787, __p2_787, __p3_787) __extension__ ({ \
	65534	int32_t __ret_787; \
	65535	int32_t __s0_787 = __p0_787; \
	65536	int32_t __s1_787 = __p1_787; \
	65537	int32x2_t __s2_787 = __p2_787; \
	65538	__ret_787 = vqrdmlahs_s32(__s0_787, __s1_787, vget_lane_s32(__s2_787, __p3_787)); \
	65539	__ret_787; \
	65540	})
66032	65541	#else
66033		__ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66034		uint8x8_t __ret;
66035		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66036		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66037		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
66038		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66039		return __ret;
66040		}
	65542	#define vqrdmlahs_lane_s32(__p0_788, __p1_788, __p2_788, __p3_788) __extension__ ({ \
	65543	int32_t __ret_788; \
	65544	int32_t __s0_788 = __p0_788; \
	65545	int32_t __s1_788 = __p1_788; \
	65546	int32x2_t __s2_788 = __p2_788; \
	65547	int32x2_t __rev2_788; __rev2_788 = __builtin_shufflevector(__s2_788, __s2_788, 1, 0); \
	65548	__ret_788 = vqrdmlahs_s32(__s0_788, __s1_788, __noswap_vget_lane_s32(__rev2_788, __p3_788)); \
	65549	__ret_788; \
	65550	})
66041	65551	#endif
66042	65552
66043	65553	#ifdef __LITTLE_ENDIAN__
66044		__ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66045		uint32x2_t __ret;
66046		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66047		return __ret;
66048		}
	65554	#define vqrdmlahh_lane_s16(__p0_789, __p1_789, __p2_789, __p3_789) __extension__ ({ \
	65555	int16_t __ret_789; \
	65556	int16_t __s0_789 = __p0_789; \
	65557	int16_t __s1_789 = __p1_789; \
	65558	int16x4_t __s2_789 = __p2_789; \
	65559	__ret_789 = vqrdmlahh_s16(__s0_789, __s1_789, vget_lane_s16(__s2_789, __p3_789)); \
	65560	__ret_789; \
	65561	})
66049	65562	#else
66050		__ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66051		uint32x2_t __ret;
66052		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66053		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66054		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
66055		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66056		return __ret;
66057		}
	65563	#define vqrdmlahh_lane_s16(__p0_790, __p1_790, __p2_790, __p3_790) __extension__ ({ \
	65564	int16_t __ret_790; \
	65565	int16_t __s0_790 = __p0_790; \
	65566	int16_t __s1_790 = __p1_790; \
	65567	int16x4_t __s2_790 = __p2_790; \
	65568	int16x4_t __rev2_790; __rev2_790 = __builtin_shufflevector(__s2_790, __s2_790, 3, 2, 1, 0); \
	65569	__ret_790 = vqrdmlahh_s16(__s0_790, __s1_790, __noswap_vget_lane_s16(__rev2_790, __p3_790)); \
	65570	__ret_790; \
	65571	})
66058	65572	#endif
66059	65573
66060	65574	#ifdef __LITTLE_ENDIAN__
66061		__ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66062		uint16x4_t __ret;
66063		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
66064		return __ret;
66065		}
	65575	#define vqrdmlahs_laneq_s32(__p0_791, __p1_791, __p2_791, __p3_791) __extension__ ({ \
	65576	int32_t __ret_791; \
	65577	int32_t __s0_791 = __p0_791; \
	65578	int32_t __s1_791 = __p1_791; \
	65579	int32x4_t __s2_791 = __p2_791; \
	65580	__ret_791 = vqrdmlahs_s32(__s0_791, __s1_791, vgetq_lane_s32(__s2_791, __p3_791)); \
	65581	__ret_791; \
	65582	})
66066	65583	#else
66067		__ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66068		uint16x4_t __ret;
66069		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66070		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66071		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
66072		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66073		return __ret;
66074		}
	65584	#define vqrdmlahs_laneq_s32(__p0_792, __p1_792, __p2_792, __p3_792) __extension__ ({ \
	65585	int32_t __ret_792; \
	65586	int32_t __s0_792 = __p0_792; \
	65587	int32_t __s1_792 = __p1_792; \
	65588	int32x4_t __s2_792 = __p2_792; \
	65589	int32x4_t __rev2_792; __rev2_792 = __builtin_shufflevector(__s2_792, __s2_792, 3, 2, 1, 0); \
	65590	__ret_792 = vqrdmlahs_s32(__s0_792, __s1_792, __noswap_vgetq_lane_s32(__rev2_792, __p3_792)); \
	65591	__ret_792; \
	65592	})
66075	65593	#endif
66076	65594
66077	65595	#ifdef __LITTLE_ENDIAN__
66078		__ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
66079		int8x8_t __ret;
66080		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
66081		return __ret;
66082		}
	65596	#define vqrdmlahh_laneq_s16(__p0_793, __p1_793, __p2_793, __p3_793) __extension__ ({ \
	65597	int16_t __ret_793; \
	65598	int16_t __s0_793 = __p0_793; \
	65599	int16_t __s1_793 = __p1_793; \
	65600	int16x8_t __s2_793 = __p2_793; \
	65601	__ret_793 = vqrdmlahh_s16(__s0_793, __s1_793, vgetq_lane_s16(__s2_793, __p3_793)); \
	65602	__ret_793; \
	65603	})
66083	65604	#else
66084		__ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
66085		int8x8_t __ret;
66086		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66087		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66088		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
66089		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66090		return __ret;
66091		}
	65605	#define vqrdmlahh_laneq_s16(__p0_794, __p1_794, __p2_794, __p3_794) __extension__ ({ \
	65606	int16_t __ret_794; \
	65607	int16_t __s0_794 = __p0_794; \
	65608	int16_t __s1_794 = __p1_794; \
	65609	int16x8_t __s2_794 = __p2_794; \
	65610	int16x8_t __rev2_794; __rev2_794 = __builtin_shufflevector(__s2_794, __s2_794, 7, 6, 5, 4, 3, 2, 1, 0); \
	65611	__ret_794 = vqrdmlahh_s16(__s0_794, __s1_794, __noswap_vgetq_lane_s16(__rev2_794, __p3_794)); \
	65612	__ret_794; \
	65613	})
66092	65614	#endif
66093	65615
66094	65616	#ifdef __LITTLE_ENDIAN__
66095		__ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
66096		float32x2_t __ret;
66097		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66098		return __ret;
66099		}
	65617	#define vqrdmlahq_laneq_s32(__p0_795, __p1_795, __p2_795, __p3_795) __extension__ ({ \
	65618	int32x4_t __ret_795; \
	65619	int32x4_t __s0_795 = __p0_795; \
	65620	int32x4_t __s1_795 = __p1_795; \
	65621	int32x4_t __s2_795 = __p2_795; \
	65622	__ret_795 = vqrdmlahq_s32(__s0_795, __s1_795, splatq_laneq_s32(__s2_795, __p3_795)); \
	65623	__ret_795; \
	65624	})
66100	65625	#else
66101		__ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
66102		float32x2_t __ret;
66103		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66104		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66105		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
66106		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66107		return __ret;
66108		}
	65626	#define vqrdmlahq_laneq_s32(__p0_796, __p1_796, __p2_796, __p3_796) __extension__ ({ \
	65627	int32x4_t __ret_796; \
	65628	int32x4_t __s0_796 = __p0_796; \
	65629	int32x4_t __s1_796 = __p1_796; \
	65630	int32x4_t __s2_796 = __p2_796; \
	65631	int32x4_t __rev0_796; __rev0_796 = __builtin_shufflevector(__s0_796, __s0_796, 3, 2, 1, 0); \
	65632	int32x4_t __rev1_796; __rev1_796 = __builtin_shufflevector(__s1_796, __s1_796, 3, 2, 1, 0); \
	65633	int32x4_t __rev2_796; __rev2_796 = __builtin_shufflevector(__s2_796, __s2_796, 3, 2, 1, 0); \
	65634	__ret_796 = __noswap_vqrdmlahq_s32(__rev0_796, __rev1_796, __noswap_splatq_laneq_s32(__rev2_796, __p3_796)); \
	65635	__ret_796 = __builtin_shufflevector(__ret_796, __ret_796, 3, 2, 1, 0); \
	65636	__ret_796; \
	65637	})
66109	65638	#endif
66110	65639
66111	65640	#ifdef __LITTLE_ENDIAN__
66112		__ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
66113		int32x2_t __ret;
66114		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66115		return __ret;
66116		}
	65641	#define vqrdmlahq_laneq_s16(__p0_797, __p1_797, __p2_797, __p3_797) __extension__ ({ \
	65642	int16x8_t __ret_797; \
	65643	int16x8_t __s0_797 = __p0_797; \
	65644	int16x8_t __s1_797 = __p1_797; \
	65645	int16x8_t __s2_797 = __p2_797; \
	65646	__ret_797 = vqrdmlahq_s16(__s0_797, __s1_797, splatq_laneq_s16(__s2_797, __p3_797)); \
	65647	__ret_797; \
	65648	})
66117	65649	#else
66118		__ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
66119		int32x2_t __ret;
66120		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66121		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66122		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
66123		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66124		return __ret;
66125		}
	65650	#define vqrdmlahq_laneq_s16(__p0_798, __p1_798, __p2_798, __p3_798) __extension__ ({ \
	65651	int16x8_t __ret_798; \
	65652	int16x8_t __s0_798 = __p0_798; \
	65653	int16x8_t __s1_798 = __p1_798; \
	65654	int16x8_t __s2_798 = __p2_798; \
	65655	int16x8_t __rev0_798; __rev0_798 = __builtin_shufflevector(__s0_798, __s0_798, 7, 6, 5, 4, 3, 2, 1, 0); \
	65656	int16x8_t __rev1_798; __rev1_798 = __builtin_shufflevector(__s1_798, __s1_798, 7, 6, 5, 4, 3, 2, 1, 0); \
	65657	int16x8_t __rev2_798; __rev2_798 = __builtin_shufflevector(__s2_798, __s2_798, 7, 6, 5, 4, 3, 2, 1, 0); \
	65658	__ret_798 = __noswap_vqrdmlahq_s16(__rev0_798, __rev1_798, __noswap_splatq_laneq_s16(__rev2_798, __p3_798)); \
	65659	__ret_798 = __builtin_shufflevector(__ret_798, __ret_798, 7, 6, 5, 4, 3, 2, 1, 0); \
	65660	__ret_798; \
	65661	})
66126	65662	#endif
66127	65663
66128	65664	#ifdef __LITTLE_ENDIAN__
66129		__ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
66130		int16x4_t __ret;
66131		__ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
66132		return __ret;
66133		}
	65665	#define vqrdmlah_laneq_s32(__p0_799, __p1_799, __p2_799, __p3_799) __extension__ ({ \
	65666	int32x2_t __ret_799; \
	65667	int32x2_t __s0_799 = __p0_799; \
	65668	int32x2_t __s1_799 = __p1_799; \
	65669	int32x4_t __s2_799 = __p2_799; \
	65670	__ret_799 = vqrdmlah_s32(__s0_799, __s1_799, splat_laneq_s32(__s2_799, __p3_799)); \
	65671	__ret_799; \
	65672	})
66134	65673	#else
66135		__ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
66136		int16x4_t __ret;
66137		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66138		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66139		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
66140		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66141		return __ret;
66142		}
	65674	#define vqrdmlah_laneq_s32(__p0_800, __p1_800, __p2_800, __p3_800) __extension__ ({ \
	65675	int32x2_t __ret_800; \
	65676	int32x2_t __s0_800 = __p0_800; \
	65677	int32x2_t __s1_800 = __p1_800; \
	65678	int32x4_t __s2_800 = __p2_800; \
	65679	int32x2_t __rev0_800; __rev0_800 = __builtin_shufflevector(__s0_800, __s0_800, 1, 0); \
	65680	int32x2_t __rev1_800; __rev1_800 = __builtin_shufflevector(__s1_800, __s1_800, 1, 0); \
	65681	int32x4_t __rev2_800; __rev2_800 = __builtin_shufflevector(__s2_800, __s2_800, 3, 2, 1, 0); \
	65682	__ret_800 = __noswap_vqrdmlah_s32(__rev0_800, __rev1_800, __noswap_splat_laneq_s32(__rev2_800, __p3_800)); \
	65683	__ret_800 = __builtin_shufflevector(__ret_800, __ret_800, 1, 0); \
	65684	__ret_800; \
	65685	})
66143	65686	#endif
66144	65687
66145	65688	#ifdef __LITTLE_ENDIAN__
66146		__ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
66147		poly8x8_t __ret;
66148		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66149		return __ret;
66150		}
	65689	#define vqrdmlah_laneq_s16(__p0_801, __p1_801, __p2_801, __p3_801) __extension__ ({ \
	65690	int16x4_t __ret_801; \
	65691	int16x4_t __s0_801 = __p0_801; \
	65692	int16x4_t __s1_801 = __p1_801; \
	65693	int16x8_t __s2_801 = __p2_801; \
	65694	__ret_801 = vqrdmlah_s16(__s0_801, __s1_801, splat_laneq_s16(__s2_801, __p3_801)); \
	65695	__ret_801; \
	65696	})
66151	65697	#else
66152		__ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
66153		poly8x8_t __ret;
66154		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66155		poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66156		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
66157		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66158		return __ret;
66159		}
	65698	#define vqrdmlah_laneq_s16(__p0_802, __p1_802, __p2_802, __p3_802) __extension__ ({ \
	65699	int16x4_t __ret_802; \
	65700	int16x4_t __s0_802 = __p0_802; \
	65701	int16x4_t __s1_802 = __p1_802; \
	65702	int16x8_t __s2_802 = __p2_802; \
	65703	int16x4_t __rev0_802; __rev0_802 = __builtin_shufflevector(__s0_802, __s0_802, 3, 2, 1, 0); \
	65704	int16x4_t __rev1_802; __rev1_802 = __builtin_shufflevector(__s1_802, __s1_802, 3, 2, 1, 0); \
	65705	int16x8_t __rev2_802; __rev2_802 = __builtin_shufflevector(__s2_802, __s2_802, 7, 6, 5, 4, 3, 2, 1, 0); \
	65706	__ret_802 = __noswap_vqrdmlah_s16(__rev0_802, __rev1_802, __noswap_splat_laneq_s16(__rev2_802, __p3_802)); \
	65707	__ret_802 = __builtin_shufflevector(__ret_802, __ret_802, 3, 2, 1, 0); \
	65708	__ret_802; \
	65709	})
66160	65710	#endif
66161	65711
66162		#ifdef __LITTLE_ENDIAN__
66163		__ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
66164		poly16x4_t __ret;
66165		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
	65712	__ai __attribute__((target("v8.1a"))) int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
	65713	int32_t __ret;
	65714	__ret = (int32_t) __builtin_neon_vqrdmlshs_s32(__p0, __p1, __p2);
66166	65715	return __ret;
66167	65716	}
66168		#else
66169		__ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
66170		poly16x4_t __ret;
66171		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66172		poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66173		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
66174		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	65717	__ai __attribute__((target("v8.1a"))) int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
	65718	int16_t __ret;
	65719	__ret = (int16_t) __builtin_neon_vqrdmlshh_s16(__p0, __p1, __p2);
66175	65720	return __ret;
66176	65721	}
	65722	#ifdef __LITTLE_ENDIAN__
	65723	#define vqrdmlshs_lane_s32(__p0_803, __p1_803, __p2_803, __p3_803) __extension__ ({ \
	65724	int32_t __ret_803; \
	65725	int32_t __s0_803 = __p0_803; \
	65726	int32_t __s1_803 = __p1_803; \
	65727	int32x2_t __s2_803 = __p2_803; \
	65728	__ret_803 = vqrdmlshs_s32(__s0_803, __s1_803, vget_lane_s32(__s2_803, __p3_803)); \
	65729	__ret_803; \
	65730	})
	65731	#else
	65732	#define vqrdmlshs_lane_s32(__p0_804, __p1_804, __p2_804, __p3_804) __extension__ ({ \
	65733	int32_t __ret_804; \
	65734	int32_t __s0_804 = __p0_804; \
	65735	int32_t __s1_804 = __p1_804; \
	65736	int32x2_t __s2_804 = __p2_804; \
	65737	int32x2_t __rev2_804; __rev2_804 = __builtin_shufflevector(__s2_804, __s2_804, 1, 0); \
	65738	__ret_804 = vqrdmlshs_s32(__s0_804, __s1_804, __noswap_vget_lane_s32(__rev2_804, __p3_804)); \
	65739	__ret_804; \
	65740	})
66177	65741	#endif
66178	65742
66179	65743	#ifdef __LITTLE_ENDIAN__
66180		__ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66181		poly8x16_t __ret;
66182		__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
66183		return __ret;
66184		}
	65744	#define vqrdmlshh_lane_s16(__p0_805, __p1_805, __p2_805, __p3_805) __extension__ ({ \
	65745	int16_t __ret_805; \
	65746	int16_t __s0_805 = __p0_805; \
	65747	int16_t __s1_805 = __p1_805; \
	65748	int16x4_t __s2_805 = __p2_805; \
	65749	__ret_805 = vqrdmlshh_s16(__s0_805, __s1_805, vget_lane_s16(__s2_805, __p3_805)); \
	65750	__ret_805; \
	65751	})
66185	65752	#else
66186		__ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66187		poly8x16_t __ret;
66188		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66189		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66190		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
66191		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66192		return __ret;
66193		}
	65753	#define vqrdmlshh_lane_s16(__p0_806, __p1_806, __p2_806, __p3_806) __extension__ ({ \
	65754	int16_t __ret_806; \
	65755	int16_t __s0_806 = __p0_806; \
	65756	int16_t __s1_806 = __p1_806; \
	65757	int16x4_t __s2_806 = __p2_806; \
	65758	int16x4_t __rev2_806; __rev2_806 = __builtin_shufflevector(__s2_806, __s2_806, 3, 2, 1, 0); \
	65759	__ret_806 = vqrdmlshh_s16(__s0_806, __s1_806, __noswap_vget_lane_s16(__rev2_806, __p3_806)); \
	65760	__ret_806; \
	65761	})
66194	65762	#endif
66195	65763
66196	65764	#ifdef __LITTLE_ENDIAN__
66197		__ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66198		poly64x2_t __ret;
66199		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66200		return __ret;
66201		}
	65765	#define vqrdmlshs_laneq_s32(__p0_807, __p1_807, __p2_807, __p3_807) __extension__ ({ \
	65766	int32_t __ret_807; \
	65767	int32_t __s0_807 = __p0_807; \
	65768	int32_t __s1_807 = __p1_807; \
	65769	int32x4_t __s2_807 = __p2_807; \
	65770	__ret_807 = vqrdmlshs_s32(__s0_807, __s1_807, vgetq_lane_s32(__s2_807, __p3_807)); \
	65771	__ret_807; \
	65772	})
66202	65773	#else
66203		__ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66204		poly64x2_t __ret;
66205		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66206		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66207		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
66208		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66209		return __ret;
66210		}
	65774	#define vqrdmlshs_laneq_s32(__p0_808, __p1_808, __p2_808, __p3_808) __extension__ ({ \
	65775	int32_t __ret_808; \
	65776	int32_t __s0_808 = __p0_808; \
	65777	int32_t __s1_808 = __p1_808; \
	65778	int32x4_t __s2_808 = __p2_808; \
	65779	int32x4_t __rev2_808; __rev2_808 = __builtin_shufflevector(__s2_808, __s2_808, 3, 2, 1, 0); \
	65780	__ret_808 = vqrdmlshs_s32(__s0_808, __s1_808, __noswap_vgetq_lane_s32(__rev2_808, __p3_808)); \
	65781	__ret_808; \
	65782	})
66211	65783	#endif
66212	65784
66213	65785	#ifdef __LITTLE_ENDIAN__
66214		__ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66215		poly16x8_t __ret;
66216		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66217		return __ret;
66218		}
	65786	#define vqrdmlshh_laneq_s16(__p0_809, __p1_809, __p2_809, __p3_809) __extension__ ({ \
	65787	int16_t __ret_809; \
	65788	int16_t __s0_809 = __p0_809; \
	65789	int16_t __s1_809 = __p1_809; \
	65790	int16x8_t __s2_809 = __p2_809; \
	65791	__ret_809 = vqrdmlshh_s16(__s0_809, __s1_809, vgetq_lane_s16(__s2_809, __p3_809)); \
	65792	__ret_809; \
	65793	})
66219	65794	#else
66220		__ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66221		poly16x8_t __ret;
66222		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66223		poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66224		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
66225		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66226		return __ret;
66227		}
	65795	#define vqrdmlshh_laneq_s16(__p0_810, __p1_810, __p2_810, __p3_810) __extension__ ({ \
	65796	int16_t __ret_810; \
	65797	int16_t __s0_810 = __p0_810; \
	65798	int16_t __s1_810 = __p1_810; \
	65799	int16x8_t __s2_810 = __p2_810; \
	65800	int16x8_t __rev2_810; __rev2_810 = __builtin_shufflevector(__s2_810, __s2_810, 7, 6, 5, 4, 3, 2, 1, 0); \
	65801	__ret_810 = vqrdmlshh_s16(__s0_810, __s1_810, __noswap_vgetq_lane_s16(__rev2_810, __p3_810)); \
	65802	__ret_810; \
	65803	})
66228	65804	#endif
66229	65805
66230	65806	#ifdef __LITTLE_ENDIAN__
66231		__ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66232		uint8x16_t __ret;
66233		__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
66234		return __ret;
66235		}
	65807	#define vqrdmlshq_laneq_s32(__p0_811, __p1_811, __p2_811, __p3_811) __extension__ ({ \
	65808	int32x4_t __ret_811; \
	65809	int32x4_t __s0_811 = __p0_811; \
	65810	int32x4_t __s1_811 = __p1_811; \
	65811	int32x4_t __s2_811 = __p2_811; \
	65812	__ret_811 = vqrdmlshq_s32(__s0_811, __s1_811, splatq_laneq_s32(__s2_811, __p3_811)); \
	65813	__ret_811; \
	65814	})
66236	65815	#else
66237		__ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66238		uint8x16_t __ret;
66239		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66240		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66241		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
66242		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66243		return __ret;
66244		}
	65816	#define vqrdmlshq_laneq_s32(__p0_812, __p1_812, __p2_812, __p3_812) __extension__ ({ \
	65817	int32x4_t __ret_812; \
	65818	int32x4_t __s0_812 = __p0_812; \
	65819	int32x4_t __s1_812 = __p1_812; \
	65820	int32x4_t __s2_812 = __p2_812; \
	65821	int32x4_t __rev0_812; __rev0_812 = __builtin_shufflevector(__s0_812, __s0_812, 3, 2, 1, 0); \
	65822	int32x4_t __rev1_812; __rev1_812 = __builtin_shufflevector(__s1_812, __s1_812, 3, 2, 1, 0); \
	65823	int32x4_t __rev2_812; __rev2_812 = __builtin_shufflevector(__s2_812, __s2_812, 3, 2, 1, 0); \
	65824	__ret_812 = __noswap_vqrdmlshq_s32(__rev0_812, __rev1_812, __noswap_splatq_laneq_s32(__rev2_812, __p3_812)); \
	65825	__ret_812 = __builtin_shufflevector(__ret_812, __ret_812, 3, 2, 1, 0); \
	65826	__ret_812; \
	65827	})
66245	65828	#endif
66246	65829
66247	65830	#ifdef __LITTLE_ENDIAN__
66248		__ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66249		uint32x4_t __ret;
66250		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66251		return __ret;
66252		}
	65831	#define vqrdmlshq_laneq_s16(__p0_813, __p1_813, __p2_813, __p3_813) __extension__ ({ \
	65832	int16x8_t __ret_813; \
	65833	int16x8_t __s0_813 = __p0_813; \
	65834	int16x8_t __s1_813 = __p1_813; \
	65835	int16x8_t __s2_813 = __p2_813; \
	65836	__ret_813 = vqrdmlshq_s16(__s0_813, __s1_813, splatq_laneq_s16(__s2_813, __p3_813)); \
	65837	__ret_813; \
	65838	})
66253	65839	#else
66254		__ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66255		uint32x4_t __ret;
66256		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66257		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66258		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
66259		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66260		return __ret;
66261		}
	65840	#define vqrdmlshq_laneq_s16(__p0_814, __p1_814, __p2_814, __p3_814) __extension__ ({ \
	65841	int16x8_t __ret_814; \
	65842	int16x8_t __s0_814 = __p0_814; \
	65843	int16x8_t __s1_814 = __p1_814; \
	65844	int16x8_t __s2_814 = __p2_814; \
	65845	int16x8_t __rev0_814; __rev0_814 = __builtin_shufflevector(__s0_814, __s0_814, 7, 6, 5, 4, 3, 2, 1, 0); \
	65846	int16x8_t __rev1_814; __rev1_814 = __builtin_shufflevector(__s1_814, __s1_814, 7, 6, 5, 4, 3, 2, 1, 0); \
	65847	int16x8_t __rev2_814; __rev2_814 = __builtin_shufflevector(__s2_814, __s2_814, 7, 6, 5, 4, 3, 2, 1, 0); \
	65848	__ret_814 = __noswap_vqrdmlshq_s16(__rev0_814, __rev1_814, __noswap_splatq_laneq_s16(__rev2_814, __p3_814)); \
	65849	__ret_814 = __builtin_shufflevector(__ret_814, __ret_814, 7, 6, 5, 4, 3, 2, 1, 0); \
	65850	__ret_814; \
	65851	})
66262	65852	#endif
66263	65853
66264	65854	#ifdef __LITTLE_ENDIAN__
66265		__ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66266		uint64x2_t __ret;
66267		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66268		return __ret;
66269		}
	65855	#define vqrdmlsh_laneq_s32(__p0_815, __p1_815, __p2_815, __p3_815) __extension__ ({ \
	65856	int32x2_t __ret_815; \
	65857	int32x2_t __s0_815 = __p0_815; \
	65858	int32x2_t __s1_815 = __p1_815; \
	65859	int32x4_t __s2_815 = __p2_815; \
	65860	__ret_815 = vqrdmlsh_s32(__s0_815, __s1_815, splat_laneq_s32(__s2_815, __p3_815)); \
	65861	__ret_815; \
	65862	})
66270	65863	#else
66271		__ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66272		uint64x2_t __ret;
66273		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66274		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66275		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
66276		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66277		return __ret;
66278		}
	65864	#define vqrdmlsh_laneq_s32(__p0_816, __p1_816, __p2_816, __p3_816) __extension__ ({ \
	65865	int32x2_t __ret_816; \
	65866	int32x2_t __s0_816 = __p0_816; \
	65867	int32x2_t __s1_816 = __p1_816; \
	65868	int32x4_t __s2_816 = __p2_816; \
	65869	int32x2_t __rev0_816; __rev0_816 = __builtin_shufflevector(__s0_816, __s0_816, 1, 0); \
	65870	int32x2_t __rev1_816; __rev1_816 = __builtin_shufflevector(__s1_816, __s1_816, 1, 0); \
	65871	int32x4_t __rev2_816; __rev2_816 = __builtin_shufflevector(__s2_816, __s2_816, 3, 2, 1, 0); \
	65872	__ret_816 = __noswap_vqrdmlsh_s32(__rev0_816, __rev1_816, __noswap_splat_laneq_s32(__rev2_816, __p3_816)); \
	65873	__ret_816 = __builtin_shufflevector(__ret_816, __ret_816, 1, 0); \
	65874	__ret_816; \
	65875	})
66279	65876	#endif
66280	65877
66281	65878	#ifdef __LITTLE_ENDIAN__
66282		__ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66283		uint16x8_t __ret;
66284		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66285		return __ret;
66286		}
	65879	#define vqrdmlsh_laneq_s16(__p0_817, __p1_817, __p2_817, __p3_817) __extension__ ({ \
	65880	int16x4_t __ret_817; \
	65881	int16x4_t __s0_817 = __p0_817; \
	65882	int16x4_t __s1_817 = __p1_817; \
	65883	int16x8_t __s2_817 = __p2_817; \
	65884	__ret_817 = vqrdmlsh_s16(__s0_817, __s1_817, splat_laneq_s16(__s2_817, __p3_817)); \
	65885	__ret_817; \
	65886	})
66287	65887	#else
66288		__ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66289		uint16x8_t __ret;
66290		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66291		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66292		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
66293		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66294		return __ret;
66295		}
	65888	#define vqrdmlsh_laneq_s16(__p0_818, __p1_818, __p2_818, __p3_818) __extension__ ({ \
	65889	int16x4_t __ret_818; \
	65890	int16x4_t __s0_818 = __p0_818; \
	65891	int16x4_t __s1_818 = __p1_818; \
	65892	int16x8_t __s2_818 = __p2_818; \
	65893	int16x4_t __rev0_818; __rev0_818 = __builtin_shufflevector(__s0_818, __s0_818, 3, 2, 1, 0); \
	65894	int16x4_t __rev1_818; __rev1_818 = __builtin_shufflevector(__s1_818, __s1_818, 3, 2, 1, 0); \
	65895	int16x8_t __rev2_818; __rev2_818 = __builtin_shufflevector(__s2_818, __s2_818, 7, 6, 5, 4, 3, 2, 1, 0); \
	65896	__ret_818 = __noswap_vqrdmlsh_s16(__rev0_818, __rev1_818, __noswap_splat_laneq_s16(__rev2_818, __p3_818)); \
	65897	__ret_818 = __builtin_shufflevector(__ret_818, __ret_818, 3, 2, 1, 0); \
	65898	__ret_818; \
	65899	})
66296	65900	#endif
66297	65901
66298	65902	#ifdef __LITTLE_ENDIAN__
66299		__ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
66300		int8x16_t __ret;
66301		__ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
	65903	__ai __attribute__((target("v8.3a"))) float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
	65904	float64x2_t __ret;
	65905	__ret = (float64x2_t) __builtin_neon_vcaddq_rot270_f64((int8x16_t)__p0, (int8x16_t)__p1, 42);
66302	65906	return __ret;
66303	65907	}
66304	65908	#else
66305		__ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
66306		int8x16_t __ret;
66307		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66308		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66309		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
66310		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	65909	__ai __attribute__((target("v8.3a"))) float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
	65910	float64x2_t __ret;
	65911	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65912	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65913	__ret = (float64x2_t) __builtin_neon_vcaddq_rot270_f64((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	65914	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66311	65915	return __ret;
66312	65916	}
66313	65917	#endif
66314	65918
66315	65919	#ifdef __LITTLE_ENDIAN__
66316		__ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
	65920	__ai __attribute__((target("v8.3a"))) float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
66317	65921	float64x2_t __ret;
66318		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	65922	__ret = (float64x2_t) __builtin_neon_vcaddq_rot90_f64((int8x16_t)__p0, (int8x16_t)__p1, 42);
66319	65923	return __ret;
66320	65924	}
66321	65925	#else
66322		__ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
	65926	__ai __attribute__((target("v8.3a"))) float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
66323	65927	float64x2_t __ret;
66324	65928	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66325	65929	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66326		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	65930	__ret = (float64x2_t) __builtin_neon_vcaddq_rot90_f64((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
66327	65931	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66328	65932	return __ret;
66329	65933	}
66330	65934	#endif
66331	65935
66332	65936	#ifdef __LITTLE_ENDIAN__
66333		__ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
66334		float32x4_t __ret;
66335		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
	65937	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	65938	float64x2_t __ret;
	65939	__ret = (float64x2_t) __builtin_neon_vcmlaq_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66336	65940	return __ret;
66337	65941	}
66338	65942	#else
66339		__ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
66340		float32x4_t __ret;
66341		float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66342		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66343		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
66344		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	65943	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	65944	float64x2_t __ret;
	65945	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	65946	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	65947	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	65948	__ret = (float64x2_t) __builtin_neon_vcmlaq_f64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	65949	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	65950	return __ret;
	65951	}
	65952	__ai __attribute__((target("v8.3a"))) float64x2_t __noswap_vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	65953	float64x2_t __ret;
	65954	__ret = (float64x2_t) __builtin_neon_vcmlaq_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66345	65955	return __ret;
66346	65956	}
66347	65957	#endif
66348	65958
66349		#ifdef __LITTLE_ENDIAN__
66350		__ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
66351		int32x4_t __ret;
66352		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
	65959	__ai __attribute__((target("v8.3a"))) float64x1_t vcmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	65960	float64x1_t __ret;
	65961	__ret = (float64x1_t) __builtin_neon_vcmla_f64((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
66353	65962	return __ret;
66354	65963	}
	65964	#define vcmla_lane_f64(__p0_819, __p1_819, __p2_819, __p3_819) __extension__ ({ \
	65965	float64x1_t __ret_819; \
	65966	float64x1_t __s0_819 = __p0_819; \
	65967	float64x1_t __s1_819 = __p1_819; \
	65968	float64x1_t __s2_819 = __p2_819; \
	65969	float64x1_t __reint_819 = __s2_819; \
	65970	uint64x2_t __reint1_819 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_819, __p3_819), vgetq_lane_u64((uint64x2_t ) &__reint_819, __p3_819)}; \
	65971	__ret_819 = vcmla_f64(__s0_819, __s1_819, (float64x1_t ) &__reint1_819); \
	65972	__ret_819; \
	65973	})
	65974	#ifdef __LITTLE_ENDIAN__
	65975	#define vcmlaq_lane_f64(__p0_820, __p1_820, __p2_820, __p3_820) __extension__ ({ \
	65976	float64x2_t __ret_820; \
	65977	float64x2_t __s0_820 = __p0_820; \
	65978	float64x2_t __s1_820 = __p1_820; \
	65979	float64x1_t __s2_820 = __p2_820; \
	65980	float64x1_t __reint_820 = __s2_820; \
	65981	uint64x2_t __reint1_820 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_820, __p3_820), vgetq_lane_u64((uint64x2_t ) &__reint_820, __p3_820)}; \
	65982	__ret_820 = vcmlaq_f64(__s0_820, __s1_820, (float64x2_t ) &__reint1_820); \
	65983	__ret_820; \
	65984	})
66355	65985	#else
66356		__ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
66357		int32x4_t __ret;
66358		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66359		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66360		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
66361		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66362		return __ret;
66363		}
	65986	#define vcmlaq_lane_f64(__p0_821, __p1_821, __p2_821, __p3_821) __extension__ ({ \
	65987	float64x2_t __ret_821; \
	65988	float64x2_t __s0_821 = __p0_821; \
	65989	float64x2_t __s1_821 = __p1_821; \
	65990	float64x1_t __s2_821 = __p2_821; \
	65991	float64x2_t __rev0_821; __rev0_821 = __builtin_shufflevector(__s0_821, __s0_821, 1, 0); \
	65992	float64x2_t __rev1_821; __rev1_821 = __builtin_shufflevector(__s1_821, __s1_821, 1, 0); \
	65993	float64x1_t __reint_821 = __s2_821; \
	65994	uint64x2_t __reint1_821 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_821, __p3_821), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_821, __p3_821)}; \
	65995	__ret_821 = __noswap_vcmlaq_f64(__rev0_821, __rev1_821, (float64x2_t ) &__reint1_821); \
	65996	__ret_821 = __builtin_shufflevector(__ret_821, __ret_821, 1, 0); \
	65997	__ret_821; \
	65998	})
66364	65999	#endif
66365	66000
66366	66001	#ifdef __LITTLE_ENDIAN__
66367		__ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
66368		int64x2_t __ret;
66369		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66370		return __ret;
66371		}
	66002	#define vcmla_laneq_f64(__p0_822, __p1_822, __p2_822, __p3_822) __extension__ ({ \
	66003	float64x1_t __ret_822; \
	66004	float64x1_t __s0_822 = __p0_822; \
	66005	float64x1_t __s1_822 = __p1_822; \
	66006	float64x2_t __s2_822 = __p2_822; \
	66007	float64x2_t __reint_822 = __s2_822; \
	66008	uint64x2_t __reint1_822 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_822, __p3_822), vgetq_lane_u64((uint64x2_t ) &__reint_822, __p3_822)}; \
	66009	__ret_822 = vcmla_f64(__s0_822, __s1_822, (float64x1_t ) &__reint1_822); \
	66010	__ret_822; \
	66011	})
66372	66012	#else
66373		__ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
66374		int64x2_t __ret;
66375		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66376		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66377		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
66378		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66379		return __ret;
66380		}
	66013	#define vcmla_laneq_f64(__p0_823, __p1_823, __p2_823, __p3_823) __extension__ ({ \
	66014	float64x1_t __ret_823; \
	66015	float64x1_t __s0_823 = __p0_823; \
	66016	float64x1_t __s1_823 = __p1_823; \
	66017	float64x2_t __s2_823 = __p2_823; \
	66018	float64x2_t __rev2_823; __rev2_823 = __builtin_shufflevector(__s2_823, __s2_823, 1, 0); \
	66019	float64x2_t __reint_823 = __rev2_823; \
	66020	uint64x2_t __reint1_823 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_823, __p3_823), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_823, __p3_823)}; \
	66021	__ret_823 = vcmla_f64(__s0_823, __s1_823, (float64x1_t ) &__reint1_823); \
	66022	__ret_823; \
	66023	})
66381	66024	#endif
66382	66025
66383	66026	#ifdef __LITTLE_ENDIAN__
66384		__ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
66385		int16x8_t __ret;
66386		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66387		return __ret;
66388		}
	66027	#define vcmlaq_laneq_f64(__p0_824, __p1_824, __p2_824, __p3_824) __extension__ ({ \
	66028	float64x2_t __ret_824; \
	66029	float64x2_t __s0_824 = __p0_824; \
	66030	float64x2_t __s1_824 = __p1_824; \
	66031	float64x2_t __s2_824 = __p2_824; \
	66032	float64x2_t __reint_824 = __s2_824; \
	66033	uint64x2_t __reint1_824 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_824, __p3_824), vgetq_lane_u64((uint64x2_t ) &__reint_824, __p3_824)}; \
	66034	__ret_824 = vcmlaq_f64(__s0_824, __s1_824, (float64x2_t ) &__reint1_824); \
	66035	__ret_824; \
	66036	})
66389	66037	#else
66390		__ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
66391		int16x8_t __ret;
66392		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66393		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66394		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
66395		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66396		return __ret;
66397		}
	66038	#define vcmlaq_laneq_f64(__p0_825, __p1_825, __p2_825, __p3_825) __extension__ ({ \
	66039	float64x2_t __ret_825; \
	66040	float64x2_t __s0_825 = __p0_825; \
	66041	float64x2_t __s1_825 = __p1_825; \
	66042	float64x2_t __s2_825 = __p2_825; \
	66043	float64x2_t __rev0_825; __rev0_825 = __builtin_shufflevector(__s0_825, __s0_825, 1, 0); \
	66044	float64x2_t __rev1_825; __rev1_825 = __builtin_shufflevector(__s1_825, __s1_825, 1, 0); \
	66045	float64x2_t __rev2_825; __rev2_825 = __builtin_shufflevector(__s2_825, __s2_825, 1, 0); \
	66046	float64x2_t __reint_825 = __rev2_825; \
	66047	uint64x2_t __reint1_825 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_825, __p3_825), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_825, __p3_825)}; \
	66048	__ret_825 = __noswap_vcmlaq_f64(__rev0_825, __rev1_825, (float64x2_t ) &__reint1_825); \
	66049	__ret_825 = __builtin_shufflevector(__ret_825, __ret_825, 1, 0); \
	66050	__ret_825; \
	66051	})
66398	66052	#endif
66399	66053
66400	66054	#ifdef __LITTLE_ENDIAN__
66401		__ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
66402		uint8x8_t __ret;
66403		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
	66055	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66056	float64x2_t __ret;
	66057	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66404	66058	return __ret;
66405	66059	}
66406	66060	#else
66407		__ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
66408		uint8x8_t __ret;
66409		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66410		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66411		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
66412		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	66061	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66062	float64x2_t __ret;
	66063	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66064	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	66065	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	66066	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_f64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	66067	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	66068	return __ret;
	66069	}
	66070	__ai __attribute__((target("v8.3a"))) float64x2_t __noswap_vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66071	float64x2_t __ret;
	66072	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66413	66073	return __ret;
66414	66074	}
66415	66075	#endif
66416	66076
66417		#ifdef __LITTLE_ENDIAN__
66418		__ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
66419		uint32x2_t __ret;
66420		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	66077	__ai __attribute__((target("v8.3a"))) float64x1_t vcmla_rot180_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	66078	float64x1_t __ret;
	66079	__ret = (float64x1_t) __builtin_neon_vcmla_rot180_f64((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
66421	66080	return __ret;
66422	66081	}
	66082	#define vcmla_rot180_lane_f64(__p0_826, __p1_826, __p2_826, __p3_826) __extension__ ({ \
	66083	float64x1_t __ret_826; \
	66084	float64x1_t __s0_826 = __p0_826; \
	66085	float64x1_t __s1_826 = __p1_826; \
	66086	float64x1_t __s2_826 = __p2_826; \
	66087	float64x1_t __reint_826 = __s2_826; \
	66088	uint64x2_t __reint1_826 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_826, __p3_826), vgetq_lane_u64((uint64x2_t ) &__reint_826, __p3_826)}; \
	66089	__ret_826 = vcmla_rot180_f64(__s0_826, __s1_826, (float64x1_t ) &__reint1_826); \
	66090	__ret_826; \
	66091	})
	66092	#ifdef __LITTLE_ENDIAN__
	66093	#define vcmlaq_rot180_lane_f64(__p0_827, __p1_827, __p2_827, __p3_827) __extension__ ({ \
	66094	float64x2_t __ret_827; \
	66095	float64x2_t __s0_827 = __p0_827; \
	66096	float64x2_t __s1_827 = __p1_827; \
	66097	float64x1_t __s2_827 = __p2_827; \
	66098	float64x1_t __reint_827 = __s2_827; \
	66099	uint64x2_t __reint1_827 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_827, __p3_827), vgetq_lane_u64((uint64x2_t ) &__reint_827, __p3_827)}; \
	66100	__ret_827 = vcmlaq_rot180_f64(__s0_827, __s1_827, (float64x2_t ) &__reint1_827); \
	66101	__ret_827; \
	66102	})
66423	66103	#else
66424		__ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
66425		uint32x2_t __ret;
66426		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66427		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66428		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
66429		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66430		return __ret;
66431		}
	66104	#define vcmlaq_rot180_lane_f64(__p0_828, __p1_828, __p2_828, __p3_828) __extension__ ({ \
	66105	float64x2_t __ret_828; \
	66106	float64x2_t __s0_828 = __p0_828; \
	66107	float64x2_t __s1_828 = __p1_828; \
	66108	float64x1_t __s2_828 = __p2_828; \
	66109	float64x2_t __rev0_828; __rev0_828 = __builtin_shufflevector(__s0_828, __s0_828, 1, 0); \
	66110	float64x2_t __rev1_828; __rev1_828 = __builtin_shufflevector(__s1_828, __s1_828, 1, 0); \
	66111	float64x1_t __reint_828 = __s2_828; \
	66112	uint64x2_t __reint1_828 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_828, __p3_828), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_828, __p3_828)}; \
	66113	__ret_828 = __noswap_vcmlaq_rot180_f64(__rev0_828, __rev1_828, (float64x2_t ) &__reint1_828); \
	66114	__ret_828 = __builtin_shufflevector(__ret_828, __ret_828, 1, 0); \
	66115	__ret_828; \
	66116	})
66432	66117	#endif
66433	66118
66434	66119	#ifdef __LITTLE_ENDIAN__
66435		__ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
66436		uint16x4_t __ret;
66437		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66438		return __ret;
66439		}
	66120	#define vcmla_rot180_laneq_f64(__p0_829, __p1_829, __p2_829, __p3_829) __extension__ ({ \
	66121	float64x1_t __ret_829; \
	66122	float64x1_t __s0_829 = __p0_829; \
	66123	float64x1_t __s1_829 = __p1_829; \
	66124	float64x2_t __s2_829 = __p2_829; \
	66125	float64x2_t __reint_829 = __s2_829; \
	66126	uint64x2_t __reint1_829 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_829, __p3_829), vgetq_lane_u64((uint64x2_t ) &__reint_829, __p3_829)}; \
	66127	__ret_829 = vcmla_rot180_f64(__s0_829, __s1_829, (float64x1_t ) &__reint1_829); \
	66128	__ret_829; \
	66129	})
66440	66130	#else
66441		__ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
66442		uint16x4_t __ret;
66443		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66444		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66445		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
66446		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66447		return __ret;
66448		}
	66131	#define vcmla_rot180_laneq_f64(__p0_830, __p1_830, __p2_830, __p3_830) __extension__ ({ \
	66132	float64x1_t __ret_830; \
	66133	float64x1_t __s0_830 = __p0_830; \
	66134	float64x1_t __s1_830 = __p1_830; \
	66135	float64x2_t __s2_830 = __p2_830; \
	66136	float64x2_t __rev2_830; __rev2_830 = __builtin_shufflevector(__s2_830, __s2_830, 1, 0); \
	66137	float64x2_t __reint_830 = __rev2_830; \
	66138	uint64x2_t __reint1_830 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_830, __p3_830), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_830, __p3_830)}; \
	66139	__ret_830 = vcmla_rot180_f64(__s0_830, __s1_830, (float64x1_t ) &__reint1_830); \
	66140	__ret_830; \
	66141	})
66449	66142	#endif
66450	66143
66451	66144	#ifdef __LITTLE_ENDIAN__
66452		__ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
66453		int8x8_t __ret;
66454		__ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66455		return __ret;
66456		}
	66145	#define vcmlaq_rot180_laneq_f64(__p0_831, __p1_831, __p2_831, __p3_831) __extension__ ({ \
	66146	float64x2_t __ret_831; \
	66147	float64x2_t __s0_831 = __p0_831; \
	66148	float64x2_t __s1_831 = __p1_831; \
	66149	float64x2_t __s2_831 = __p2_831; \
	66150	float64x2_t __reint_831 = __s2_831; \
	66151	uint64x2_t __reint1_831 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_831, __p3_831), vgetq_lane_u64((uint64x2_t ) &__reint_831, __p3_831)}; \
	66152	__ret_831 = vcmlaq_rot180_f64(__s0_831, __s1_831, (float64x2_t ) &__reint1_831); \
	66153	__ret_831; \
	66154	})
66457	66155	#else
66458		__ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
66459		int8x8_t __ret;
66460		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66461		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66462		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
66463		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66464		return __ret;
66465		}
	66156	#define vcmlaq_rot180_laneq_f64(__p0_832, __p1_832, __p2_832, __p3_832) __extension__ ({ \
	66157	float64x2_t __ret_832; \
	66158	float64x2_t __s0_832 = __p0_832; \
	66159	float64x2_t __s1_832 = __p1_832; \
	66160	float64x2_t __s2_832 = __p2_832; \
	66161	float64x2_t __rev0_832; __rev0_832 = __builtin_shufflevector(__s0_832, __s0_832, 1, 0); \
	66162	float64x2_t __rev1_832; __rev1_832 = __builtin_shufflevector(__s1_832, __s1_832, 1, 0); \
	66163	float64x2_t __rev2_832; __rev2_832 = __builtin_shufflevector(__s2_832, __s2_832, 1, 0); \
	66164	float64x2_t __reint_832 = __rev2_832; \
	66165	uint64x2_t __reint1_832 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_832, __p3_832), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_832, __p3_832)}; \
	66166	__ret_832 = __noswap_vcmlaq_rot180_f64(__rev0_832, __rev1_832, (float64x2_t ) &__reint1_832); \
	66167	__ret_832 = __builtin_shufflevector(__ret_832, __ret_832, 1, 0); \
	66168	__ret_832; \
	66169	})
66466	66170	#endif
66467	66171
66468	66172	#ifdef __LITTLE_ENDIAN__
66469		__ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
66470		float32x2_t __ret;
66471		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	66173	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66174	float64x2_t __ret;
	66175	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66472	66176	return __ret;
66473	66177	}
66474	66178	#else
66475		__ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
66476		float32x2_t __ret;
66477		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66478		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66479		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
	66179	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66180	float64x2_t __ret;
	66181	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66182	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	66183	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	66184	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_f64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
66480	66185	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66481	66186	return __ret;
66482	66187	}
	66188	__ai __attribute__((target("v8.3a"))) float64x2_t __noswap_vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66189	float64x2_t __ret;
	66190	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
	66191	return __ret;
	66192	}
66483	66193	#endif
66484	66194
66485		#ifdef __LITTLE_ENDIAN__
66486		__ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
66487		int32x2_t __ret;
66488		__ret = __builtin_shufflevector(__p0, __p1, 0, 2);
	66195	__ai __attribute__((target("v8.3a"))) float64x1_t vcmla_rot270_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	66196	float64x1_t __ret;
	66197	__ret = (float64x1_t) __builtin_neon_vcmla_rot270_f64((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
66489	66198	return __ret;
66490	66199	}
	66200	#define vcmla_rot270_lane_f64(__p0_833, __p1_833, __p2_833, __p3_833) __extension__ ({ \
	66201	float64x1_t __ret_833; \
	66202	float64x1_t __s0_833 = __p0_833; \
	66203	float64x1_t __s1_833 = __p1_833; \
	66204	float64x1_t __s2_833 = __p2_833; \
	66205	float64x1_t __reint_833 = __s2_833; \
	66206	uint64x2_t __reint1_833 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_833, __p3_833), vgetq_lane_u64((uint64x2_t ) &__reint_833, __p3_833)}; \
	66207	__ret_833 = vcmla_rot270_f64(__s0_833, __s1_833, (float64x1_t ) &__reint1_833); \
	66208	__ret_833; \
	66209	})
	66210	#ifdef __LITTLE_ENDIAN__
	66211	#define vcmlaq_rot270_lane_f64(__p0_834, __p1_834, __p2_834, __p3_834) __extension__ ({ \
	66212	float64x2_t __ret_834; \
	66213	float64x2_t __s0_834 = __p0_834; \
	66214	float64x2_t __s1_834 = __p1_834; \
	66215	float64x1_t __s2_834 = __p2_834; \
	66216	float64x1_t __reint_834 = __s2_834; \
	66217	uint64x2_t __reint1_834 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_834, __p3_834), vgetq_lane_u64((uint64x2_t ) &__reint_834, __p3_834)}; \
	66218	__ret_834 = vcmlaq_rot270_f64(__s0_834, __s1_834, (float64x2_t ) &__reint1_834); \
	66219	__ret_834; \
	66220	})
66491	66221	#else
66492		__ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
66493		int32x2_t __ret;
66494		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66495		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66496		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
66497		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66498		return __ret;
66499		}
	66222	#define vcmlaq_rot270_lane_f64(__p0_835, __p1_835, __p2_835, __p3_835) __extension__ ({ \
	66223	float64x2_t __ret_835; \
	66224	float64x2_t __s0_835 = __p0_835; \
	66225	float64x2_t __s1_835 = __p1_835; \
	66226	float64x1_t __s2_835 = __p2_835; \
	66227	float64x2_t __rev0_835; __rev0_835 = __builtin_shufflevector(__s0_835, __s0_835, 1, 0); \
	66228	float64x2_t __rev1_835; __rev1_835 = __builtin_shufflevector(__s1_835, __s1_835, 1, 0); \
	66229	float64x1_t __reint_835 = __s2_835; \
	66230	uint64x2_t __reint1_835 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_835, __p3_835), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_835, __p3_835)}; \
	66231	__ret_835 = __noswap_vcmlaq_rot270_f64(__rev0_835, __rev1_835, (float64x2_t ) &__reint1_835); \
	66232	__ret_835 = __builtin_shufflevector(__ret_835, __ret_835, 1, 0); \
	66233	__ret_835; \
	66234	})
66500	66235	#endif
66501	66236
66502	66237	#ifdef __LITTLE_ENDIAN__
66503		__ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
66504		int16x4_t __ret;
66505		__ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66506		return __ret;
66507		}
	66238	#define vcmla_rot270_laneq_f64(__p0_836, __p1_836, __p2_836, __p3_836) __extension__ ({ \
	66239	float64x1_t __ret_836; \
	66240	float64x1_t __s0_836 = __p0_836; \
	66241	float64x1_t __s1_836 = __p1_836; \
	66242	float64x2_t __s2_836 = __p2_836; \
	66243	float64x2_t __reint_836 = __s2_836; \
	66244	uint64x2_t __reint1_836 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_836, __p3_836), vgetq_lane_u64((uint64x2_t ) &__reint_836, __p3_836)}; \
	66245	__ret_836 = vcmla_rot270_f64(__s0_836, __s1_836, (float64x1_t ) &__reint1_836); \
	66246	__ret_836; \
	66247	})
66508	66248	#else
66509		__ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
66510		int16x4_t __ret;
66511		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66512		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66513		__ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
66514		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66515		return __ret;
66516		}
	66249	#define vcmla_rot270_laneq_f64(__p0_837, __p1_837, __p2_837, __p3_837) __extension__ ({ \
	66250	float64x1_t __ret_837; \
	66251	float64x1_t __s0_837 = __p0_837; \
	66252	float64x1_t __s1_837 = __p1_837; \
	66253	float64x2_t __s2_837 = __p2_837; \
	66254	float64x2_t __rev2_837; __rev2_837 = __builtin_shufflevector(__s2_837, __s2_837, 1, 0); \
	66255	float64x2_t __reint_837 = __rev2_837; \
	66256	uint64x2_t __reint1_837 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_837, __p3_837), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_837, __p3_837)}; \
	66257	__ret_837 = vcmla_rot270_f64(__s0_837, __s1_837, (float64x1_t ) &__reint1_837); \
	66258	__ret_837; \
	66259	})
66517	66260	#endif
66518	66261
66519	66262	#ifdef __LITTLE_ENDIAN__
66520		__ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
66521		poly8x8_t __ret;
66522		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66523		return __ret;
66524		}
	66263	#define vcmlaq_rot270_laneq_f64(__p0_838, __p1_838, __p2_838, __p3_838) __extension__ ({ \
	66264	float64x2_t __ret_838; \
	66265	float64x2_t __s0_838 = __p0_838; \
	66266	float64x2_t __s1_838 = __p1_838; \
	66267	float64x2_t __s2_838 = __p2_838; \
	66268	float64x2_t __reint_838 = __s2_838; \
	66269	uint64x2_t __reint1_838 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_838, __p3_838), vgetq_lane_u64((uint64x2_t ) &__reint_838, __p3_838)}; \
	66270	__ret_838 = vcmlaq_rot270_f64(__s0_838, __s1_838, (float64x2_t ) &__reint1_838); \
	66271	__ret_838; \
	66272	})
66525	66273	#else
66526		__ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
66527		poly8x8_t __ret;
66528		poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66529		poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66530		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
66531		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66532		return __ret;
66533		}
	66274	#define vcmlaq_rot270_laneq_f64(__p0_839, __p1_839, __p2_839, __p3_839) __extension__ ({ \
	66275	float64x2_t __ret_839; \
	66276	float64x2_t __s0_839 = __p0_839; \
	66277	float64x2_t __s1_839 = __p1_839; \
	66278	float64x2_t __s2_839 = __p2_839; \
	66279	float64x2_t __rev0_839; __rev0_839 = __builtin_shufflevector(__s0_839, __s0_839, 1, 0); \
	66280	float64x2_t __rev1_839; __rev1_839 = __builtin_shufflevector(__s1_839, __s1_839, 1, 0); \
	66281	float64x2_t __rev2_839; __rev2_839 = __builtin_shufflevector(__s2_839, __s2_839, 1, 0); \
	66282	float64x2_t __reint_839 = __rev2_839; \
	66283	uint64x2_t __reint1_839 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_839, __p3_839), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_839, __p3_839)}; \
	66284	__ret_839 = __noswap_vcmlaq_rot270_f64(__rev0_839, __rev1_839, (float64x2_t ) &__reint1_839); \
	66285	__ret_839 = __builtin_shufflevector(__ret_839, __ret_839, 1, 0); \
	66286	__ret_839; \
	66287	})
66534	66288	#endif
66535	66289
66536	66290	#ifdef __LITTLE_ENDIAN__
66537		__ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
66538		poly16x4_t __ret;
66539		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	66291	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66292	float64x2_t __ret;
	66293	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66540	66294	return __ret;
66541	66295	}
66542	66296	#else
66543		__ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
66544		poly16x4_t __ret;
66545		poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66546		poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66547		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
66548		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	66297	__ai __attribute__((target("v8.3a"))) float64x2_t vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66298	float64x2_t __ret;
	66299	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66300	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	66301	float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
	66302	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_f64((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
	66303	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	66304	return __ret;
	66305	}
	66306	__ai __attribute__((target("v8.3a"))) float64x2_t __noswap_vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
	66307	float64x2_t __ret;
	66308	__ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_f64((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
66549	66309	return __ret;
66550	66310	}
66551	66311	#endif
66552	66312
66553		#ifdef __LITTLE_ENDIAN__
66554		__ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66555		poly8x16_t __ret;
66556		__ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	66313	__ai __attribute__((target("v8.3a"))) float64x1_t vcmla_rot90_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
	66314	float64x1_t __ret;
	66315	__ret = (float64x1_t) __builtin_neon_vcmla_rot90_f64((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
66557	66316	return __ret;
66558	66317	}
	66318	#define vcmla_rot90_lane_f64(__p0_840, __p1_840, __p2_840, __p3_840) __extension__ ({ \
	66319	float64x1_t __ret_840; \
	66320	float64x1_t __s0_840 = __p0_840; \
	66321	float64x1_t __s1_840 = __p1_840; \
	66322	float64x1_t __s2_840 = __p2_840; \
	66323	float64x1_t __reint_840 = __s2_840; \
	66324	uint64x2_t __reint1_840 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_840, __p3_840), vgetq_lane_u64((uint64x2_t ) &__reint_840, __p3_840)}; \
	66325	__ret_840 = vcmla_rot90_f64(__s0_840, __s1_840, (float64x1_t ) &__reint1_840); \
	66326	__ret_840; \
	66327	})
	66328	#ifdef __LITTLE_ENDIAN__
	66329	#define vcmlaq_rot90_lane_f64(__p0_841, __p1_841, __p2_841, __p3_841) __extension__ ({ \
	66330	float64x2_t __ret_841; \
	66331	float64x2_t __s0_841 = __p0_841; \
	66332	float64x2_t __s1_841 = __p1_841; \
	66333	float64x1_t __s2_841 = __p2_841; \
	66334	float64x1_t __reint_841 = __s2_841; \
	66335	uint64x2_t __reint1_841 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_841, __p3_841), vgetq_lane_u64((uint64x2_t ) &__reint_841, __p3_841)}; \
	66336	__ret_841 = vcmlaq_rot90_f64(__s0_841, __s1_841, (float64x2_t ) &__reint1_841); \
	66337	__ret_841; \
	66338	})
66559	66339	#else
66560		__ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66561		poly8x16_t __ret;
66562		poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66563		poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66564		__ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
66565		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66566		return __ret;
66567		}
	66340	#define vcmlaq_rot90_lane_f64(__p0_842, __p1_842, __p2_842, __p3_842) __extension__ ({ \
	66341	float64x2_t __ret_842; \
	66342	float64x2_t __s0_842 = __p0_842; \
	66343	float64x2_t __s1_842 = __p1_842; \
	66344	float64x1_t __s2_842 = __p2_842; \
	66345	float64x2_t __rev0_842; __rev0_842 = __builtin_shufflevector(__s0_842, __s0_842, 1, 0); \
	66346	float64x2_t __rev1_842; __rev1_842 = __builtin_shufflevector(__s1_842, __s1_842, 1, 0); \
	66347	float64x1_t __reint_842 = __s2_842; \
	66348	uint64x2_t __reint1_842 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_842, __p3_842), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_842, __p3_842)}; \
	66349	__ret_842 = __noswap_vcmlaq_rot90_f64(__rev0_842, __rev1_842, (float64x2_t ) &__reint1_842); \
	66350	__ret_842 = __builtin_shufflevector(__ret_842, __ret_842, 1, 0); \
	66351	__ret_842; \
	66352	})
66568	66353	#endif
66569	66354
66570	66355	#ifdef __LITTLE_ENDIAN__
66571		__ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66572		poly64x2_t __ret;
66573		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66574		return __ret;
66575		}
	66356	#define vcmla_rot90_laneq_f64(__p0_843, __p1_843, __p2_843, __p3_843) __extension__ ({ \
	66357	float64x1_t __ret_843; \
	66358	float64x1_t __s0_843 = __p0_843; \
	66359	float64x1_t __s1_843 = __p1_843; \
	66360	float64x2_t __s2_843 = __p2_843; \
	66361	float64x2_t __reint_843 = __s2_843; \
	66362	uint64x2_t __reint1_843 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_843, __p3_843), vgetq_lane_u64((uint64x2_t ) &__reint_843, __p3_843)}; \
	66363	__ret_843 = vcmla_rot90_f64(__s0_843, __s1_843, (float64x1_t ) &__reint1_843); \
	66364	__ret_843; \
	66365	})
66576	66366	#else
66577		__ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66578		poly64x2_t __ret;
66579		poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66580		poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66581		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
66582		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66583		return __ret;
66584		}
	66367	#define vcmla_rot90_laneq_f64(__p0_844, __p1_844, __p2_844, __p3_844) __extension__ ({ \
	66368	float64x1_t __ret_844; \
	66369	float64x1_t __s0_844 = __p0_844; \
	66370	float64x1_t __s1_844 = __p1_844; \
	66371	float64x2_t __s2_844 = __p2_844; \
	66372	float64x2_t __rev2_844; __rev2_844 = __builtin_shufflevector(__s2_844, __s2_844, 1, 0); \
	66373	float64x2_t __reint_844 = __rev2_844; \
	66374	uint64x2_t __reint1_844 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_844, __p3_844), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_844, __p3_844)}; \
	66375	__ret_844 = vcmla_rot90_f64(__s0_844, __s1_844, (float64x1_t ) &__reint1_844); \
	66376	__ret_844; \
	66377	})
66585	66378	#endif
66586	66379
66587	66380	#ifdef __LITTLE_ENDIAN__
66588		__ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66589		poly16x8_t __ret;
66590		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66591		return __ret;
66592		}
	66381	#define vcmlaq_rot90_laneq_f64(__p0_845, __p1_845, __p2_845, __p3_845) __extension__ ({ \
	66382	float64x2_t __ret_845; \
	66383	float64x2_t __s0_845 = __p0_845; \
	66384	float64x2_t __s1_845 = __p1_845; \
	66385	float64x2_t __s2_845 = __p2_845; \
	66386	float64x2_t __reint_845 = __s2_845; \
	66387	uint64x2_t __reint1_845 = (uint64x2_t) {vgetq_lane_u64((uint64x2_t ) &__reint_845, __p3_845), vgetq_lane_u64((uint64x2_t ) &__reint_845, __p3_845)}; \
	66388	__ret_845 = vcmlaq_rot90_f64(__s0_845, __s1_845, (float64x2_t ) &__reint1_845); \
	66389	__ret_845; \
	66390	})
66593	66391	#else
66594		__ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66595		poly16x8_t __ret;
66596		poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66597		poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66598		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
66599		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
66600		return __ret;
66601		}
	66392	#define vcmlaq_rot90_laneq_f64(__p0_846, __p1_846, __p2_846, __p3_846) __extension__ ({ \
	66393	float64x2_t __ret_846; \
	66394	float64x2_t __s0_846 = __p0_846; \
	66395	float64x2_t __s1_846 = __p1_846; \
	66396	float64x2_t __s2_846 = __p2_846; \
	66397	float64x2_t __rev0_846; __rev0_846 = __builtin_shufflevector(__s0_846, __s0_846, 1, 0); \
	66398	float64x2_t __rev1_846; __rev1_846 = __builtin_shufflevector(__s1_846, __s1_846, 1, 0); \
	66399	float64x2_t __rev2_846; __rev2_846 = __builtin_shufflevector(__s2_846, __s2_846, 1, 0); \
	66400	float64x2_t __reint_846 = __rev2_846; \
	66401	uint64x2_t __reint1_846 = (uint64x2_t) {__noswap_vgetq_lane_u64((uint64x2_t ) &__reint_846, __p3_846), __noswap_vgetq_lane_u64((uint64x2_t ) &__reint_846, __p3_846)}; \
	66402	__ret_846 = __noswap_vcmlaq_rot90_f64(__rev0_846, __rev1_846, (float64x2_t ) &__reint1_846); \
	66403	__ret_846 = __builtin_shufflevector(__ret_846, __ret_846, 1, 0); \
	66404	__ret_846; \
	66405	})
66602	66406	#endif
66603	66407
66604	66408	#ifdef __LITTLE_ENDIAN__
66605		__ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66606		uint8x16_t __ret;
66607		__ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	66409	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd32xq_f32(float32x4_t __p0) {
	66410	float32x4_t __ret;
	66411	__ret = (float32x4_t) __builtin_neon_vrnd32xq_f32((int8x16_t)__p0, 41);
66608	66412	return __ret;
66609	66413	}
66610	66414	#else
66611		__ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66612		uint8x16_t __ret;
66613		uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66614		uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66615		__ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
66616		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	66415	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd32xq_f32(float32x4_t __p0) {
	66416	float32x4_t __ret;
	66417	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	66418	__ret = (float32x4_t) __builtin_neon_vrnd32xq_f32((int8x16_t)__rev0, 41);
	66419	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66617	66420	return __ret;
66618	66421	}
66619	66422	#endif
66620	66423
66621	66424	#ifdef __LITTLE_ENDIAN__
66622		__ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66623		uint32x4_t __ret;
66624		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	66425	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd32x_f32(float32x2_t __p0) {
	66426	float32x2_t __ret;
	66427	__ret = (float32x2_t) __builtin_neon_vrnd32x_f32((int8x8_t)__p0, 9);
66625	66428	return __ret;
66626	66429	}
66627	66430	#else
66628		__ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66629		uint32x4_t __ret;
66630		uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66631		uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66632		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
66633		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	66431	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd32x_f32(float32x2_t __p0) {
	66432	float32x2_t __ret;
	66433	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66434	__ret = (float32x2_t) __builtin_neon_vrnd32x_f32((int8x8_t)__rev0, 9);
	66435	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66634	66436	return __ret;
66635	66437	}
66636	66438	#endif
66637	66439
66638	66440	#ifdef __LITTLE_ENDIAN__
66639		__ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66640		uint64x2_t __ret;
66641		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	66441	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd32zq_f32(float32x4_t __p0) {
	66442	float32x4_t __ret;
	66443	__ret = (float32x4_t) __builtin_neon_vrnd32zq_f32((int8x16_t)__p0, 41);
66642	66444	return __ret;
66643	66445	}
66644	66446	#else
66645		__ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66646		uint64x2_t __ret;
66647		uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66648		uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66649		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
66650		__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
	66447	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd32zq_f32(float32x4_t __p0) {
	66448	float32x4_t __ret;
	66449	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	66450	__ret = (float32x4_t) __builtin_neon_vrnd32zq_f32((int8x16_t)__rev0, 41);
	66451	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66651	66452	return __ret;
66652	66453	}
66653	66454	#endif
66654	66455
66655	66456	#ifdef __LITTLE_ENDIAN__
66656		__ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66657		uint16x8_t __ret;
66658		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	66457	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd32z_f32(float32x2_t __p0) {
	66458	float32x2_t __ret;
	66459	__ret = (float32x2_t) __builtin_neon_vrnd32z_f32((int8x8_t)__p0, 9);
66659	66460	return __ret;
66660	66461	}
66661	66462	#else
66662		__ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66663		uint16x8_t __ret;
66664		uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66665		uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66666		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
66667		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	66463	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd32z_f32(float32x2_t __p0) {
	66464	float32x2_t __ret;
	66465	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66466	__ret = (float32x2_t) __builtin_neon_vrnd32z_f32((int8x8_t)__rev0, 9);
	66467	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66668	66468	return __ret;
66669	66469	}
66670	66470	#endif
66671	66471
66672	66472	#ifdef __LITTLE_ENDIAN__
66673		__ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
66674		int8x16_t __ret;
66675		__ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
	66473	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd64xq_f32(float32x4_t __p0) {
	66474	float32x4_t __ret;
	66475	__ret = (float32x4_t) __builtin_neon_vrnd64xq_f32((int8x16_t)__p0, 41);
66676	66476	return __ret;
66677	66477	}
66678	66478	#else
66679		__ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
66680		int8x16_t __ret;
66681		int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66682		int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66683		__ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
66684		__ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
	66479	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd64xq_f32(float32x4_t __p0) {
	66480	float32x4_t __ret;
	66481	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
	66482	__ret = (float32x4_t) __builtin_neon_vrnd64xq_f32((int8x16_t)__rev0, 41);
	66483	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66685	66484	return __ret;
66686	66485	}
66687	66486	#endif
66688	66487
66689	66488	#ifdef __LITTLE_ENDIAN__
66690		__ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
66691		float64x2_t __ret;
66692		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	66489	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd64x_f32(float32x2_t __p0) {
	66490	float32x2_t __ret;
	66491	__ret = (float32x2_t) __builtin_neon_vrnd64x_f32((int8x8_t)__p0, 9);
66693	66492	return __ret;
66694	66493	}
66695	66494	#else
66696		__ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
66697		float64x2_t __ret;
66698		float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66699		float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66700		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	66495	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd64x_f32(float32x2_t __p0) {
	66496	float32x2_t __ret;
	66497	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66498	__ret = (float32x2_t) __builtin_neon_vrnd64x_f32((int8x8_t)__rev0, 9);
66701	66499	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66702	66500	return __ret;
66703	66501	}
66704	66502	#endif
66705	66503
66706	66504	#ifdef __LITTLE_ENDIAN__
66707		__ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
	66505	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd64zq_f32(float32x4_t __p0) {
66708	66506	float32x4_t __ret;
66709		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	66507	__ret = (float32x4_t) __builtin_neon_vrnd64zq_f32((int8x16_t)__p0, 41);
66710	66508	return __ret;
66711	66509	}
66712	66510	#else
66713		__ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
	66511	__ai __attribute__((target("v8.5a"))) float32x4_t vrnd64zq_f32(float32x4_t __p0) {
66714	66512	float32x4_t __ret;
66715	66513	float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66716		float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66717		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
	66514	__ret = (float32x4_t) __builtin_neon_vrnd64zq_f32((int8x16_t)__rev0, 41);
66718	66515	__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66719	66516	return __ret;
66720	66517	}
66721	66518	#endif
66722	66519
66723	66520	#ifdef __LITTLE_ENDIAN__
66724		__ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
66725		int32x4_t __ret;
66726		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	66521	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd64z_f32(float32x2_t __p0) {
	66522	float32x2_t __ret;
	66523	__ret = (float32x2_t) __builtin_neon_vrnd64z_f32((int8x8_t)__p0, 9);
66727	66524	return __ret;
66728	66525	}
66729	66526	#else
66730		__ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
66731		int32x4_t __ret;
66732		int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66733		int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66734		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
66735		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	66527	__ai __attribute__((target("v8.5a"))) float32x2_t vrnd64z_f32(float32x2_t __p0) {
	66528	float32x2_t __ret;
	66529	float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66530	__ret = (float32x2_t) __builtin_neon_vrnd64z_f32((int8x8_t)__rev0, 9);
	66531	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66736	66532	return __ret;
66737	66533	}
66738	66534	#endif
66739	66535
	66536	#endif
	66537	#if defined(__aarch64__) && defined(__ARM_FEATURE_DIRECTED_ROUNDING)
66740	66538	#ifdef __LITTLE_ENDIAN__
66741		__ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
66742		int64x2_t __ret;
66743		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	66539	__ai float64x2_t vrndq_f64(float64x2_t __p0) {
	66540	float64x2_t __ret;
	66541	__ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__p0, 42);
66744	66542	return __ret;
66745	66543	}
66746	66544	#else
66747		__ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
66748		int64x2_t __ret;
66749		int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66750		int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66751		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	66545	__ai float64x2_t vrndq_f64(float64x2_t __p0) {
	66546	float64x2_t __ret;
	66547	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66548	__ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 42);
66752	66549	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66753	66550	return __ret;
66754	66551	}
66755	66552	#endif
66756	66553
	66554	__ai float64x1_t vrnd_f64(float64x1_t __p0) {
	66555	float64x1_t __ret;
	66556	__ret = (float64x1_t) __builtin_neon_vrnd_v((int8x8_t)__p0, 10);
	66557	return __ret;
	66558	}
66757	66559	#ifdef __LITTLE_ENDIAN__
66758		__ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
66759		int16x8_t __ret;
66760		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	66560	__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
	66561	float64x2_t __ret;
	66562	__ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__p0, 42);
66761	66563	return __ret;
66762	66564	}
66763	66565	#else
66764		__ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
66765		int16x8_t __ret;
66766		int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66767		int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66768		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
66769		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	66566	__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
	66567	float64x2_t __ret;
	66568	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66569	__ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 42);
	66570	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66770	66571	return __ret;
66771	66572	}
66772	66573	#endif
66773	66574
	66575	__ai float64x1_t vrnda_f64(float64x1_t __p0) {
	66576	float64x1_t __ret;
	66577	__ret = (float64x1_t) __builtin_neon_vrnda_v((int8x8_t)__p0, 10);
	66578	return __ret;
	66579	}
66774	66580	#ifdef __LITTLE_ENDIAN__
66775		__ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66776		uint8x8_t __ret;
66777		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	66581	__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
	66582	float64x2_t __ret;
	66583	__ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__p0, 42);
66778	66584	return __ret;
66779	66585	}
66780	66586	#else
66781		__ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66782		uint8x8_t __ret;
66783		uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66784		uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66785		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
66786		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	66587	__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
	66588	float64x2_t __ret;
	66589	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66590	__ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 42);
	66591	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66787	66592	return __ret;
66788	66593	}
66789	66594	#endif
66790	66595
	66596	__ai float64x1_t vrndi_f64(float64x1_t __p0) {
	66597	float64x1_t __ret;
	66598	__ret = (float64x1_t) __builtin_neon_vrndi_v((int8x8_t)__p0, 10);
	66599	return __ret;
	66600	}
66791	66601	#ifdef __LITTLE_ENDIAN__
66792		__ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66793		uint32x2_t __ret;
66794		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	66602	__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
	66603	float64x2_t __ret;
	66604	__ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__p0, 42);
66795	66605	return __ret;
66796	66606	}
66797	66607	#else
66798		__ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66799		uint32x2_t __ret;
66800		uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66801		uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66802		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	66608	__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
	66609	float64x2_t __ret;
	66610	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66611	__ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 42);
66803	66612	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66804	66613	return __ret;
66805	66614	}
66806	66615	#endif
66807	66616
	66617	__ai float64x1_t vrndm_f64(float64x1_t __p0) {
	66618	float64x1_t __ret;
	66619	__ret = (float64x1_t) __builtin_neon_vrndm_v((int8x8_t)__p0, 10);
	66620	return __ret;
	66621	}
66808	66622	#ifdef __LITTLE_ENDIAN__
66809		__ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66810		uint16x4_t __ret;
66811		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	66623	__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
	66624	float64x2_t __ret;
	66625	__ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__p0, 42);
66812	66626	return __ret;
66813	66627	}
66814	66628	#else
66815		__ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66816		uint16x4_t __ret;
66817		uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66818		uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66819		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
66820		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	66629	__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
	66630	float64x2_t __ret;
	66631	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66632	__ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 42);
	66633	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66821	66634	return __ret;
66822	66635	}
66823	66636	#endif
66824	66637
	66638	__ai float64x1_t vrndn_f64(float64x1_t __p0) {
	66639	float64x1_t __ret;
	66640	__ret = (float64x1_t) __builtin_neon_vrndn_v((int8x8_t)__p0, 10);
	66641	return __ret;
	66642	}
66825	66643	#ifdef __LITTLE_ENDIAN__
66826		__ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66827		int8x8_t __ret;
66828		__ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
	66644	__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
	66645	float64x2_t __ret;
	66646	__ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__p0, 42);
66829	66647	return __ret;
66830	66648	}
66831	66649	#else
66832		__ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66833		int8x8_t __ret;
66834		int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
66835		int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66836		__ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
66837		__ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
	66650	__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
	66651	float64x2_t __ret;
	66652	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66653	__ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 42);
	66654	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66838	66655	return __ret;
66839	66656	}
66840	66657	#endif
66841	66658
	66659	__ai float64x1_t vrndp_f64(float64x1_t __p0) {
	66660	float64x1_t __ret;
	66661	__ret = (float64x1_t) __builtin_neon_vrndp_v((int8x8_t)__p0, 10);
	66662	return __ret;
	66663	}
66842	66664	#ifdef __LITTLE_ENDIAN__
66843		__ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66844		float32x2_t __ret;
66845		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	66665	__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
	66666	float64x2_t __ret;
	66667	__ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__p0, 42);
66846	66668	return __ret;
66847	66669	}
66848	66670	#else
66849		__ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66850		float32x2_t __ret;
66851		float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66852		float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66853		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	66671	__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
	66672	float64x2_t __ret;
	66673	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66674	__ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 42);
66854	66675	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66855	66676	return __ret;
66856	66677	}
66857	66678	#endif
66858	66679
	66680	__ai float64x1_t vrndx_f64(float64x1_t __p0) {
	66681	float64x1_t __ret;
	66682	__ret = (float64x1_t) __builtin_neon_vrndx_v((int8x8_t)__p0, 10);
	66683	return __ret;
	66684	}
	66685	#endif
	66686	#if defined(__aarch64__) && defined(__ARM_FEATURE_NUMERIC_MAXMIN)
66859	66687	#ifdef __LITTLE_ENDIAN__
66860		__ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66861		int32x2_t __ret;
66862		__ret = __builtin_shufflevector(__p0, __p1, 1, 3);
	66688	__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	66689	float64x2_t __ret;
	66690	__ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
66863	66691	return __ret;
66864	66692	}
66865	66693	#else
66866		__ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66867		int32x2_t __ret;
66868		int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
66869		int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66870		__ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
	66694	__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	66695	float64x2_t __ret;
	66696	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66697	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	66698	__ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
66871	66699	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66872	66700	return __ret;
66873	66701	}
66874	66702	#endif
66875	66703
	66704	__ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
	66705	float64x1_t __ret;
	66706	__ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	66707	return __ret;
	66708	}
66876	66709	#ifdef __LITTLE_ENDIAN__
66877		__ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66878		int16x4_t __ret;
66879		__ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
	66710	__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	66711	float64x2_t __ret;
	66712	__ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
66880	66713	return __ret;
66881	66714	}
66882	66715	#else
66883		__ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66884		int16x4_t __ret;
66885		int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66886		int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66887		__ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
66888		__ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
	66716	__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
	66717	float64x2_t __ret;
	66718	float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
	66719	float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
	66720	__ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
	66721	__ret = __builtin_shufflevector(__ret, __ret, 1, 0);
66889	66722	return __ret;
66890	66723	}
66891	66724	#endif
66892	66725
	66726	__ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
	66727	float64x1_t __ret;
	66728	__ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
	66729	return __ret;
	66730	}
66893	66731	#endif
66894	66732	#ifdef __LITTLE_ENDIAN__
66895	66733	__ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
...	...	@@ -67444,60 +67282,60 @@ __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
67444	67282	#endif
67445	67283
67446	67284	#ifdef __LITTLE_ENDIAN__
67447		#define vget_lane_f16(__p0_851, __p1_851) __extension__ ({ \
67448		float16_t __ret_851; \
67449		float16x4_t __s0_851 = __p0_851; \
67450		float16x4_t __reint_851 = __s0_851; \
67451		int16_t __reint1_851 = vget_lane_s16((int16x4_t ) &__reint_851, __p1_851); \
67452		__ret_851 = (float16_t ) &__reint1_851; \
67453		__ret_851; \
	67285	#define vget_lane_f16(__p0_847, __p1_847) __extension__ ({ \
	67286	float16_t __ret_847; \
	67287	float16x4_t __s0_847 = __p0_847; \
	67288	float16x4_t __reint_847 = __s0_847; \
	67289	int16_t __reint1_847 = vget_lane_s16((int16x4_t ) &__reint_847, __p1_847); \
	67290	__ret_847 = (float16_t ) &__reint1_847; \
	67291	__ret_847; \
67454	67292	})
67455	67293	#else
67456		#define vget_lane_f16(__p0_852, __p1_852) __extension__ ({ \
67457		float16_t __ret_852; \
67458		float16x4_t __s0_852 = __p0_852; \
67459		float16x4_t __rev0_852; __rev0_852 = __builtin_shufflevector(__s0_852, __s0_852, 3, 2, 1, 0); \
67460		float16x4_t __reint_852 = __rev0_852; \
67461		int16_t __reint1_852 = __noswap_vget_lane_s16((int16x4_t ) &__reint_852, __p1_852); \
67462		__ret_852 = (float16_t ) &__reint1_852; \
67463		__ret_852; \
	67294	#define vget_lane_f16(__p0_848, __p1_848) __extension__ ({ \
	67295	float16_t __ret_848; \
	67296	float16x4_t __s0_848 = __p0_848; \
	67297	float16x4_t __rev0_848; __rev0_848 = __builtin_shufflevector(__s0_848, __s0_848, 3, 2, 1, 0); \
	67298	float16x4_t __reint_848 = __rev0_848; \
	67299	int16_t __reint1_848 = __noswap_vget_lane_s16((int16x4_t ) &__reint_848, __p1_848); \
	67300	__ret_848 = (float16_t ) &__reint1_848; \
	67301	__ret_848; \
67464	67302	})
67465		#define __noswap_vget_lane_f16(__p0_853, __p1_853) __extension__ ({ \
67466		float16_t __ret_853; \
67467		float16x4_t __s0_853 = __p0_853; \
67468		float16x4_t __reint_853 = __s0_853; \
67469		int16_t __reint1_853 = __noswap_vget_lane_s16((int16x4_t ) &__reint_853, __p1_853); \
67470		__ret_853 = (float16_t ) &__reint1_853; \
67471		__ret_853; \
	67303	#define __noswap_vget_lane_f16(__p0_849, __p1_849) __extension__ ({ \
	67304	float16_t __ret_849; \
	67305	float16x4_t __s0_849 = __p0_849; \
	67306	float16x4_t __reint_849 = __s0_849; \
	67307	int16_t __reint1_849 = __noswap_vget_lane_s16((int16x4_t ) &__reint_849, __p1_849); \
	67308	__ret_849 = (float16_t ) &__reint1_849; \
	67309	__ret_849; \
67472	67310	})
67473	67311	#endif
67474	67312
67475	67313	#ifdef __LITTLE_ENDIAN__
67476		#define vgetq_lane_f16(__p0_854, __p1_854) __extension__ ({ \
67477		float16_t __ret_854; \
67478		float16x8_t __s0_854 = __p0_854; \
67479		float16x8_t __reint_854 = __s0_854; \
67480		int16_t __reint1_854 = vgetq_lane_s16((int16x8_t ) &__reint_854, __p1_854); \
67481		__ret_854 = (float16_t ) &__reint1_854; \
67482		__ret_854; \
	67314	#define vgetq_lane_f16(__p0_850, __p1_850) __extension__ ({ \
	67315	float16_t __ret_850; \
	67316	float16x8_t __s0_850 = __p0_850; \
	67317	float16x8_t __reint_850 = __s0_850; \
	67318	int16_t __reint1_850 = vgetq_lane_s16((int16x8_t ) &__reint_850, __p1_850); \
	67319	__ret_850 = (float16_t ) &__reint1_850; \
	67320	__ret_850; \
67483	67321	})
67484	67322	#else
67485		#define vgetq_lane_f16(__p0_855, __p1_855) __extension__ ({ \
67486		float16_t __ret_855; \
67487		float16x8_t __s0_855 = __p0_855; \
67488		float16x8_t __rev0_855; __rev0_855 = __builtin_shufflevector(__s0_855, __s0_855, 7, 6, 5, 4, 3, 2, 1, 0); \
67489		float16x8_t __reint_855 = __rev0_855; \
67490		int16_t __reint1_855 = __noswap_vgetq_lane_s16((int16x8_t ) &__reint_855, __p1_855); \
67491		__ret_855 = (float16_t ) &__reint1_855; \
67492		__ret_855; \
	67323	#define vgetq_lane_f16(__p0_851, __p1_851) __extension__ ({ \
	67324	float16_t __ret_851; \
	67325	float16x8_t __s0_851 = __p0_851; \
	67326	float16x8_t __rev0_851; __rev0_851 = __builtin_shufflevector(__s0_851, __s0_851, 7, 6, 5, 4, 3, 2, 1, 0); \
	67327	float16x8_t __reint_851 = __rev0_851; \
	67328	int16_t __reint1_851 = __noswap_vgetq_lane_s16((int16x8_t ) &__reint_851, __p1_851); \
	67329	__ret_851 = (float16_t ) &__reint1_851; \
	67330	__ret_851; \
67493	67331	})
67494		#define __noswap_vgetq_lane_f16(__p0_856, __p1_856) __extension__ ({ \
67495		float16_t __ret_856; \
67496		float16x8_t __s0_856 = __p0_856; \
67497		float16x8_t __reint_856 = __s0_856; \
67498		int16_t __reint1_856 = __noswap_vgetq_lane_s16((int16x8_t ) &__reint_856, __p1_856); \
67499		__ret_856 = (float16_t ) &__reint1_856; \
67500		__ret_856; \
	67332	#define __noswap_vgetq_lane_f16(__p0_852, __p1_852) __extension__ ({ \
	67333	float16_t __ret_852; \
	67334	float16x8_t __s0_852 = __p0_852; \
	67335	float16x8_t __reint_852 = __s0_852; \
	67336	int16_t __reint1_852 = __noswap_vgetq_lane_s16((int16x8_t ) &__reint_852, __p1_852); \
	67337	__ret_852 = (float16_t ) &__reint1_852; \
	67338	__ret_852; \
67501	67339	})
67502	67340	#endif
67503	67341
...	...	@@ -67640,98 +67478,98 @@ __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
67640	67478	#endif
67641	67479
67642	67480	#ifdef __LITTLE_ENDIAN__
67643		#define vmlal_lane_u32(__p0_857, __p1_857, __p2_857, __p3_857) __extension__ ({ \
67644		uint64x2_t __ret_857; \
67645		uint64x2_t __s0_857 = __p0_857; \
67646		uint32x2_t __s1_857 = __p1_857; \
67647		uint32x2_t __s2_857 = __p2_857; \
67648		__ret_857 = __s0_857 + vmull_u32(__s1_857, splat_lane_u32(__s2_857, __p3_857)); \
67649		__ret_857; \
	67481	#define vmlal_lane_u32(__p0_853, __p1_853, __p2_853, __p3_853) __extension__ ({ \
	67482	uint64x2_t __ret_853; \
	67483	uint64x2_t __s0_853 = __p0_853; \
	67484	uint32x2_t __s1_853 = __p1_853; \
	67485	uint32x2_t __s2_853 = __p2_853; \
	67486	__ret_853 = __s0_853 + vmull_u32(__s1_853, splat_lane_u32(__s2_853, __p3_853)); \
	67487	__ret_853; \
67650	67488	})
67651	67489	#else
67652		#define vmlal_lane_u32(__p0_858, __p1_858, __p2_858, __p3_858) __extension__ ({ \
67653		uint64x2_t __ret_858; \
67654		uint64x2_t __s0_858 = __p0_858; \
67655		uint32x2_t __s1_858 = __p1_858; \
67656		uint32x2_t __s2_858 = __p2_858; \
67657		uint64x2_t __rev0_858; __rev0_858 = __builtin_shufflevector(__s0_858, __s0_858, 1, 0); \
67658		uint32x2_t __rev1_858; __rev1_858 = __builtin_shufflevector(__s1_858, __s1_858, 1, 0); \
67659		uint32x2_t __rev2_858; __rev2_858 = __builtin_shufflevector(__s2_858, __s2_858, 1, 0); \
67660		__ret_858 = __rev0_858 + __noswap_vmull_u32(__rev1_858, __noswap_splat_lane_u32(__rev2_858, __p3_858)); \
67661		__ret_858 = __builtin_shufflevector(__ret_858, __ret_858, 1, 0); \
67662		__ret_858; \
	67490	#define vmlal_lane_u32(__p0_854, __p1_854, __p2_854, __p3_854) __extension__ ({ \
	67491	uint64x2_t __ret_854; \
	67492	uint64x2_t __s0_854 = __p0_854; \
	67493	uint32x2_t __s1_854 = __p1_854; \
	67494	uint32x2_t __s2_854 = __p2_854; \
	67495	uint64x2_t __rev0_854; __rev0_854 = __builtin_shufflevector(__s0_854, __s0_854, 1, 0); \
	67496	uint32x2_t __rev1_854; __rev1_854 = __builtin_shufflevector(__s1_854, __s1_854, 1, 0); \
	67497	uint32x2_t __rev2_854; __rev2_854 = __builtin_shufflevector(__s2_854, __s2_854, 1, 0); \
	67498	__ret_854 = __rev0_854 + __noswap_vmull_u32(__rev1_854, __noswap_splat_lane_u32(__rev2_854, __p3_854)); \
	67499	__ret_854 = __builtin_shufflevector(__ret_854, __ret_854, 1, 0); \
	67500	__ret_854; \
67663	67501	})
67664	67502	#endif
67665	67503
67666	67504	#ifdef __LITTLE_ENDIAN__
67667		#define vmlal_lane_u16(__p0_859, __p1_859, __p2_859, __p3_859) __extension__ ({ \
67668		uint32x4_t __ret_859; \
67669		uint32x4_t __s0_859 = __p0_859; \
67670		uint16x4_t __s1_859 = __p1_859; \
67671		uint16x4_t __s2_859 = __p2_859; \
67672		__ret_859 = __s0_859 + vmull_u16(__s1_859, splat_lane_u16(__s2_859, __p3_859)); \
67673		__ret_859; \
	67505	#define vmlal_lane_u16(__p0_855, __p1_855, __p2_855, __p3_855) __extension__ ({ \
	67506	uint32x4_t __ret_855; \
	67507	uint32x4_t __s0_855 = __p0_855; \
	67508	uint16x4_t __s1_855 = __p1_855; \
	67509	uint16x4_t __s2_855 = __p2_855; \
	67510	__ret_855 = __s0_855 + vmull_u16(__s1_855, splat_lane_u16(__s2_855, __p3_855)); \
	67511	__ret_855; \
67674	67512	})
67675	67513	#else
67676		#define vmlal_lane_u16(__p0_860, __p1_860, __p2_860, __p3_860) __extension__ ({ \
67677		uint32x4_t __ret_860; \
67678		uint32x4_t __s0_860 = __p0_860; \
67679		uint16x4_t __s1_860 = __p1_860; \
67680		uint16x4_t __s2_860 = __p2_860; \
67681		uint32x4_t __rev0_860; __rev0_860 = __builtin_shufflevector(__s0_860, __s0_860, 3, 2, 1, 0); \
67682		uint16x4_t __rev1_860; __rev1_860 = __builtin_shufflevector(__s1_860, __s1_860, 3, 2, 1, 0); \
67683		uint16x4_t __rev2_860; __rev2_860 = __builtin_shufflevector(__s2_860, __s2_860, 3, 2, 1, 0); \
67684		__ret_860 = __rev0_860 + __noswap_vmull_u16(__rev1_860, __noswap_splat_lane_u16(__rev2_860, __p3_860)); \
67685		__ret_860 = __builtin_shufflevector(__ret_860, __ret_860, 3, 2, 1, 0); \
67686		__ret_860; \
	67514	#define vmlal_lane_u16(__p0_856, __p1_856, __p2_856, __p3_856) __extension__ ({ \
	67515	uint32x4_t __ret_856; \
	67516	uint32x4_t __s0_856 = __p0_856; \
	67517	uint16x4_t __s1_856 = __p1_856; \
	67518	uint16x4_t __s2_856 = __p2_856; \
	67519	uint32x4_t __rev0_856; __rev0_856 = __builtin_shufflevector(__s0_856, __s0_856, 3, 2, 1, 0); \
	67520	uint16x4_t __rev1_856; __rev1_856 = __builtin_shufflevector(__s1_856, __s1_856, 3, 2, 1, 0); \
	67521	uint16x4_t __rev2_856; __rev2_856 = __builtin_shufflevector(__s2_856, __s2_856, 3, 2, 1, 0); \
	67522	__ret_856 = __rev0_856 + __noswap_vmull_u16(__rev1_856, __noswap_splat_lane_u16(__rev2_856, __p3_856)); \
	67523	__ret_856 = __builtin_shufflevector(__ret_856, __ret_856, 3, 2, 1, 0); \
	67524	__ret_856; \
67687	67525	})
67688	67526	#endif
67689	67527
67690	67528	#ifdef __LITTLE_ENDIAN__
67691		#define vmlal_lane_s32(__p0_861, __p1_861, __p2_861, __p3_861) __extension__ ({ \
67692		int64x2_t __ret_861; \
67693		int64x2_t __s0_861 = __p0_861; \
67694		int32x2_t __s1_861 = __p1_861; \
67695		int32x2_t __s2_861 = __p2_861; \
67696		__ret_861 = __s0_861 + vmull_s32(__s1_861, splat_lane_s32(__s2_861, __p3_861)); \
67697		__ret_861; \
	67529	#define vmlal_lane_s32(__p0_857, __p1_857, __p2_857, __p3_857) __extension__ ({ \
	67530	int64x2_t __ret_857; \
	67531	int64x2_t __s0_857 = __p0_857; \
	67532	int32x2_t __s1_857 = __p1_857; \
	67533	int32x2_t __s2_857 = __p2_857; \
	67534	__ret_857 = __s0_857 + vmull_s32(__s1_857, splat_lane_s32(__s2_857, __p3_857)); \
	67535	__ret_857; \
67698	67536	})
67699	67537	#else
67700		#define vmlal_lane_s32(__p0_862, __p1_862, __p2_862, __p3_862) __extension__ ({ \
67701		int64x2_t __ret_862; \
67702		int64x2_t __s0_862 = __p0_862; \
67703		int32x2_t __s1_862 = __p1_862; \
67704		int32x2_t __s2_862 = __p2_862; \
67705		int64x2_t __rev0_862; __rev0_862 = __builtin_shufflevector(__s0_862, __s0_862, 1, 0); \
67706		int32x2_t __rev1_862; __rev1_862 = __builtin_shufflevector(__s1_862, __s1_862, 1, 0); \
67707		int32x2_t __rev2_862; __rev2_862 = __builtin_shufflevector(__s2_862, __s2_862, 1, 0); \
67708		__ret_862 = __rev0_862 + __noswap_vmull_s32(__rev1_862, __noswap_splat_lane_s32(__rev2_862, __p3_862)); \
67709		__ret_862 = __builtin_shufflevector(__ret_862, __ret_862, 1, 0); \
67710		__ret_862; \
	67538	#define vmlal_lane_s32(__p0_858, __p1_858, __p2_858, __p3_858) __extension__ ({ \
	67539	int64x2_t __ret_858; \
	67540	int64x2_t __s0_858 = __p0_858; \
	67541	int32x2_t __s1_858 = __p1_858; \
	67542	int32x2_t __s2_858 = __p2_858; \
	67543	int64x2_t __rev0_858; __rev0_858 = __builtin_shufflevector(__s0_858, __s0_858, 1, 0); \
	67544	int32x2_t __rev1_858; __rev1_858 = __builtin_shufflevector(__s1_858, __s1_858, 1, 0); \
	67545	int32x2_t __rev2_858; __rev2_858 = __builtin_shufflevector(__s2_858, __s2_858, 1, 0); \
	67546	__ret_858 = __rev0_858 + __noswap_vmull_s32(__rev1_858, __noswap_splat_lane_s32(__rev2_858, __p3_858)); \
	67547	__ret_858 = __builtin_shufflevector(__ret_858, __ret_858, 1, 0); \
	67548	__ret_858; \
67711	67549	})
67712	67550	#endif
67713	67551
67714	67552	#ifdef __LITTLE_ENDIAN__
67715		#define vmlal_lane_s16(__p0_863, __p1_863, __p2_863, __p3_863) __extension__ ({ \
67716		int32x4_t __ret_863; \
67717		int32x4_t __s0_863 = __p0_863; \
67718		int16x4_t __s1_863 = __p1_863; \
67719		int16x4_t __s2_863 = __p2_863; \
67720		__ret_863 = __s0_863 + vmull_s16(__s1_863, splat_lane_s16(__s2_863, __p3_863)); \
67721		__ret_863; \
	67553	#define vmlal_lane_s16(__p0_859, __p1_859, __p2_859, __p3_859) __extension__ ({ \
	67554	int32x4_t __ret_859; \
	67555	int32x4_t __s0_859 = __p0_859; \
	67556	int16x4_t __s1_859 = __p1_859; \
	67557	int16x4_t __s2_859 = __p2_859; \
	67558	__ret_859 = __s0_859 + vmull_s16(__s1_859, splat_lane_s16(__s2_859, __p3_859)); \
	67559	__ret_859; \
67722	67560	})
67723	67561	#else
67724		#define vmlal_lane_s16(__p0_864, __p1_864, __p2_864, __p3_864) __extension__ ({ \
67725		int32x4_t __ret_864; \
67726		int32x4_t __s0_864 = __p0_864; \
67727		int16x4_t __s1_864 = __p1_864; \
67728		int16x4_t __s2_864 = __p2_864; \
67729		int32x4_t __rev0_864; __rev0_864 = __builtin_shufflevector(__s0_864, __s0_864, 3, 2, 1, 0); \
67730		int16x4_t __rev1_864; __rev1_864 = __builtin_shufflevector(__s1_864, __s1_864, 3, 2, 1, 0); \
67731		int16x4_t __rev2_864; __rev2_864 = __builtin_shufflevector(__s2_864, __s2_864, 3, 2, 1, 0); \
67732		__ret_864 = __rev0_864 + __noswap_vmull_s16(__rev1_864, __noswap_splat_lane_s16(__rev2_864, __p3_864)); \
67733		__ret_864 = __builtin_shufflevector(__ret_864, __ret_864, 3, 2, 1, 0); \
67734		__ret_864; \
	67562	#define vmlal_lane_s16(__p0_860, __p1_860, __p2_860, __p3_860) __extension__ ({ \
	67563	int32x4_t __ret_860; \
	67564	int32x4_t __s0_860 = __p0_860; \
	67565	int16x4_t __s1_860 = __p1_860; \
	67566	int16x4_t __s2_860 = __p2_860; \
	67567	int32x4_t __rev0_860; __rev0_860 = __builtin_shufflevector(__s0_860, __s0_860, 3, 2, 1, 0); \
	67568	int16x4_t __rev1_860; __rev1_860 = __builtin_shufflevector(__s1_860, __s1_860, 3, 2, 1, 0); \
	67569	int16x4_t __rev2_860; __rev2_860 = __builtin_shufflevector(__s2_860, __s2_860, 3, 2, 1, 0); \
	67570	__ret_860 = __rev0_860 + __noswap_vmull_s16(__rev1_860, __noswap_splat_lane_s16(__rev2_860, __p3_860)); \
	67571	__ret_860 = __builtin_shufflevector(__ret_860, __ret_860, 3, 2, 1, 0); \
	67572	__ret_860; \
67735	67573	})
67736	67574	#endif
67737	67575
...	...	@@ -67962,98 +67800,98 @@ __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
67962	67800	#endif
67963	67801
67964	67802	#ifdef __LITTLE_ENDIAN__
67965		#define vmlsl_lane_u32(__p0_865, __p1_865, __p2_865, __p3_865) __extension__ ({ \
67966		uint64x2_t __ret_865; \
67967		uint64x2_t __s0_865 = __p0_865; \
67968		uint32x2_t __s1_865 = __p1_865; \
67969		uint32x2_t __s2_865 = __p2_865; \
67970		__ret_865 = __s0_865 - vmull_u32(__s1_865, splat_lane_u32(__s2_865, __p3_865)); \
67971		__ret_865; \
	67803	#define vmlsl_lane_u32(__p0_861, __p1_861, __p2_861, __p3_861) __extension__ ({ \
	67804	uint64x2_t __ret_861; \
	67805	uint64x2_t __s0_861 = __p0_861; \
	67806	uint32x2_t __s1_861 = __p1_861; \
	67807	uint32x2_t __s2_861 = __p2_861; \
	67808	__ret_861 = __s0_861 - vmull_u32(__s1_861, splat_lane_u32(__s2_861, __p3_861)); \
	67809	__ret_861; \
67972	67810	})
67973	67811	#else
67974		#define vmlsl_lane_u32(__p0_866, __p1_866, __p2_866, __p3_866) __extension__ ({ \
67975		uint64x2_t __ret_866; \
67976		uint64x2_t __s0_866 = __p0_866; \
67977		uint32x2_t __s1_866 = __p1_866; \
67978		uint32x2_t __s2_866 = __p2_866; \
67979		uint64x2_t __rev0_866; __rev0_866 = __builtin_shufflevector(__s0_866, __s0_866, 1, 0); \
67980		uint32x2_t __rev1_866; __rev1_866 = __builtin_shufflevector(__s1_866, __s1_866, 1, 0); \
67981		uint32x2_t __rev2_866; __rev2_866 = __builtin_shufflevector(__s2_866, __s2_866, 1, 0); \
67982		__ret_866 = __rev0_866 - __noswap_vmull_u32(__rev1_866, __noswap_splat_lane_u32(__rev2_866, __p3_866)); \
67983		__ret_866 = __builtin_shufflevector(__ret_866, __ret_866, 1, 0); \
67984		__ret_866; \
	67812	#define vmlsl_lane_u32(__p0_862, __p1_862, __p2_862, __p3_862) __extension__ ({ \
	67813	uint64x2_t __ret_862; \
	67814	uint64x2_t __s0_862 = __p0_862; \
	67815	uint32x2_t __s1_862 = __p1_862; \
	67816	uint32x2_t __s2_862 = __p2_862; \
	67817	uint64x2_t __rev0_862; __rev0_862 = __builtin_shufflevector(__s0_862, __s0_862, 1, 0); \
	67818	uint32x2_t __rev1_862; __rev1_862 = __builtin_shufflevector(__s1_862, __s1_862, 1, 0); \
	67819	uint32x2_t __rev2_862; __rev2_862 = __builtin_shufflevector(__s2_862, __s2_862, 1, 0); \
	67820	__ret_862 = __rev0_862 - __noswap_vmull_u32(__rev1_862, __noswap_splat_lane_u32(__rev2_862, __p3_862)); \
	67821	__ret_862 = __builtin_shufflevector(__ret_862, __ret_862, 1, 0); \
	67822	__ret_862; \
67985	67823	})
67986	67824	#endif
67987	67825
67988	67826	#ifdef __LITTLE_ENDIAN__
67989		#define vmlsl_lane_u16(__p0_867, __p1_867, __p2_867, __p3_867) __extension__ ({ \
67990		uint32x4_t __ret_867; \
67991		uint32x4_t __s0_867 = __p0_867; \
67992		uint16x4_t __s1_867 = __p1_867; \
67993		uint16x4_t __s2_867 = __p2_867; \
67994		__ret_867 = __s0_867 - vmull_u16(__s1_867, splat_lane_u16(__s2_867, __p3_867)); \
67995		__ret_867; \
	67827	#define vmlsl_lane_u16(__p0_863, __p1_863, __p2_863, __p3_863) __extension__ ({ \
	67828	uint32x4_t __ret_863; \
	67829	uint32x4_t __s0_863 = __p0_863; \
	67830	uint16x4_t __s1_863 = __p1_863; \
	67831	uint16x4_t __s2_863 = __p2_863; \
	67832	__ret_863 = __s0_863 - vmull_u16(__s1_863, splat_lane_u16(__s2_863, __p3_863)); \
	67833	__ret_863; \
67996	67834	})
67997	67835	#else
67998		#define vmlsl_lane_u16(__p0_868, __p1_868, __p2_868, __p3_868) __extension__ ({ \
67999		uint32x4_t __ret_868; \
68000		uint32x4_t __s0_868 = __p0_868; \
68001		uint16x4_t __s1_868 = __p1_868; \
68002		uint16x4_t __s2_868 = __p2_868; \
68003		uint32x4_t __rev0_868; __rev0_868 = __builtin_shufflevector(__s0_868, __s0_868, 3, 2, 1, 0); \
68004		uint16x4_t __rev1_868; __rev1_868 = __builtin_shufflevector(__s1_868, __s1_868, 3, 2, 1, 0); \
68005		uint16x4_t __rev2_868; __rev2_868 = __builtin_shufflevector(__s2_868, __s2_868, 3, 2, 1, 0); \
68006		__ret_868 = __rev0_868 - __noswap_vmull_u16(__rev1_868, __noswap_splat_lane_u16(__rev2_868, __p3_868)); \
68007		__ret_868 = __builtin_shufflevector(__ret_868, __ret_868, 3, 2, 1, 0); \
68008		__ret_868; \
	67836	#define vmlsl_lane_u16(__p0_864, __p1_864, __p2_864, __p3_864) __extension__ ({ \
	67837	uint32x4_t __ret_864; \
	67838	uint32x4_t __s0_864 = __p0_864; \
	67839	uint16x4_t __s1_864 = __p1_864; \
	67840	uint16x4_t __s2_864 = __p2_864; \
	67841	uint32x4_t __rev0_864; __rev0_864 = __builtin_shufflevector(__s0_864, __s0_864, 3, 2, 1, 0); \
	67842	uint16x4_t __rev1_864; __rev1_864 = __builtin_shufflevector(__s1_864, __s1_864, 3, 2, 1, 0); \
	67843	uint16x4_t __rev2_864; __rev2_864 = __builtin_shufflevector(__s2_864, __s2_864, 3, 2, 1, 0); \
	67844	__ret_864 = __rev0_864 - __noswap_vmull_u16(__rev1_864, __noswap_splat_lane_u16(__rev2_864, __p3_864)); \
	67845	__ret_864 = __builtin_shufflevector(__ret_864, __ret_864, 3, 2, 1, 0); \
	67846	__ret_864; \
68009	67847	})
68010	67848	#endif
68011	67849
68012	67850	#ifdef __LITTLE_ENDIAN__
68013		#define vmlsl_lane_s32(__p0_869, __p1_869, __p2_869, __p3_869) __extension__ ({ \
68014		int64x2_t __ret_869; \
68015		int64x2_t __s0_869 = __p0_869; \
68016		int32x2_t __s1_869 = __p1_869; \
68017		int32x2_t __s2_869 = __p2_869; \
68018		__ret_869 = __s0_869 - vmull_s32(__s1_869, splat_lane_s32(__s2_869, __p3_869)); \
68019		__ret_869; \
	67851	#define vmlsl_lane_s32(__p0_865, __p1_865, __p2_865, __p3_865) __extension__ ({ \
	67852	int64x2_t __ret_865; \
	67853	int64x2_t __s0_865 = __p0_865; \
	67854	int32x2_t __s1_865 = __p1_865; \
	67855	int32x2_t __s2_865 = __p2_865; \
	67856	__ret_865 = __s0_865 - vmull_s32(__s1_865, splat_lane_s32(__s2_865, __p3_865)); \
	67857	__ret_865; \
68020	67858	})
68021	67859	#else
68022		#define vmlsl_lane_s32(__p0_870, __p1_870, __p2_870, __p3_870) __extension__ ({ \
68023		int64x2_t __ret_870; \
68024		int64x2_t __s0_870 = __p0_870; \
68025		int32x2_t __s1_870 = __p1_870; \
68026		int32x2_t __s2_870 = __p2_870; \
68027		int64x2_t __rev0_870; __rev0_870 = __builtin_shufflevector(__s0_870, __s0_870, 1, 0); \
68028		int32x2_t __rev1_870; __rev1_870 = __builtin_shufflevector(__s1_870, __s1_870, 1, 0); \
68029		int32x2_t __rev2_870; __rev2_870 = __builtin_shufflevector(__s2_870, __s2_870, 1, 0); \
68030		__ret_870 = __rev0_870 - __noswap_vmull_s32(__rev1_870, __noswap_splat_lane_s32(__rev2_870, __p3_870)); \
68031		__ret_870 = __builtin_shufflevector(__ret_870, __ret_870, 1, 0); \
68032		__ret_870; \
	67860	#define vmlsl_lane_s32(__p0_866, __p1_866, __p2_866, __p3_866) __extension__ ({ \
	67861	int64x2_t __ret_866; \
	67862	int64x2_t __s0_866 = __p0_866; \
	67863	int32x2_t __s1_866 = __p1_866; \
	67864	int32x2_t __s2_866 = __p2_866; \
	67865	int64x2_t __rev0_866; __rev0_866 = __builtin_shufflevector(__s0_866, __s0_866, 1, 0); \
	67866	int32x2_t __rev1_866; __rev1_866 = __builtin_shufflevector(__s1_866, __s1_866, 1, 0); \
	67867	int32x2_t __rev2_866; __rev2_866 = __builtin_shufflevector(__s2_866, __s2_866, 1, 0); \
	67868	__ret_866 = __rev0_866 - __noswap_vmull_s32(__rev1_866, __noswap_splat_lane_s32(__rev2_866, __p3_866)); \
	67869	__ret_866 = __builtin_shufflevector(__ret_866, __ret_866, 1, 0); \
	67870	__ret_866; \
68033	67871	})
68034	67872	#endif
68035	67873
68036	67874	#ifdef __LITTLE_ENDIAN__
68037		#define vmlsl_lane_s16(__p0_871, __p1_871, __p2_871, __p3_871) __extension__ ({ \
68038		int32x4_t __ret_871; \
68039		int32x4_t __s0_871 = __p0_871; \
68040		int16x4_t __s1_871 = __p1_871; \
68041		int16x4_t __s2_871 = __p2_871; \
68042		__ret_871 = __s0_871 - vmull_s16(__s1_871, splat_lane_s16(__s2_871, __p3_871)); \
68043		__ret_871; \
	67875	#define vmlsl_lane_s16(__p0_867, __p1_867, __p2_867, __p3_867) __extension__ ({ \
	67876	int32x4_t __ret_867; \
	67877	int32x4_t __s0_867 = __p0_867; \
	67878	int16x4_t __s1_867 = __p1_867; \
	67879	int16x4_t __s2_867 = __p2_867; \
	67880	__ret_867 = __s0_867 - vmull_s16(__s1_867, splat_lane_s16(__s2_867, __p3_867)); \
	67881	__ret_867; \
68044	67882	})
68045	67883	#else
68046		#define vmlsl_lane_s16(__p0_872, __p1_872, __p2_872, __p3_872) __extension__ ({ \
68047		int32x4_t __ret_872; \
68048		int32x4_t __s0_872 = __p0_872; \
68049		int16x4_t __s1_872 = __p1_872; \
68050		int16x4_t __s2_872 = __p2_872; \
68051		int32x4_t __rev0_872; __rev0_872 = __builtin_shufflevector(__s0_872, __s0_872, 3, 2, 1, 0); \
68052		int16x4_t __rev1_872; __rev1_872 = __builtin_shufflevector(__s1_872, __s1_872, 3, 2, 1, 0); \
68053		int16x4_t __rev2_872; __rev2_872 = __builtin_shufflevector(__s2_872, __s2_872, 3, 2, 1, 0); \
68054		__ret_872 = __rev0_872 - __noswap_vmull_s16(__rev1_872, __noswap_splat_lane_s16(__rev2_872, __p3_872)); \
68055		__ret_872 = __builtin_shufflevector(__ret_872, __ret_872, 3, 2, 1, 0); \
68056		__ret_872; \
	67884	#define vmlsl_lane_s16(__p0_868, __p1_868, __p2_868, __p3_868) __extension__ ({ \
	67885	int32x4_t __ret_868; \
	67886	int32x4_t __s0_868 = __p0_868; \
	67887	int16x4_t __s1_868 = __p1_868; \
	67888	int16x4_t __s2_868 = __p2_868; \
	67889	int32x4_t __rev0_868; __rev0_868 = __builtin_shufflevector(__s0_868, __s0_868, 3, 2, 1, 0); \
	67890	int16x4_t __rev1_868; __rev1_868 = __builtin_shufflevector(__s1_868, __s1_868, 3, 2, 1, 0); \
	67891	int16x4_t __rev2_868; __rev2_868 = __builtin_shufflevector(__s2_868, __s2_868, 3, 2, 1, 0); \
	67892	__ret_868 = __rev0_868 - __noswap_vmull_s16(__rev1_868, __noswap_splat_lane_s16(__rev2_868, __p3_868)); \
	67893	__ret_868 = __builtin_shufflevector(__ret_868, __ret_868, 3, 2, 1, 0); \
	67894	__ret_868; \
68057	67895	})
68058	67896	#endif
68059	67897
...	...	@@ -68146,69 +67984,116 @@ __ai int32x4_t __noswap_vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2
68146	67984	#endif
68147	67985
68148	67986	#ifdef __LITTLE_ENDIAN__
68149		#define vset_lane_f16(__p0_873, __p1_873, __p2_873) __extension__ ({ \
68150		float16x4_t __ret_873; \
68151		float16_t __s0_873 = __p0_873; \
68152		float16x4_t __s1_873 = __p1_873; \
68153		float16_t __reint_873 = __s0_873; \
68154		float16x4_t __reint1_873 = __s1_873; \
68155		int16x4_t __reint2_873 = vset_lane_s16((int16_t ) &__reint_873, (int16x4_t ) &__reint1_873, __p2_873); \
68156		__ret_873 = (float16x4_t ) &__reint2_873; \
	67987	#define vset_lane_f16(__p0_869, __p1_869, __p2_869) __extension__ ({ \
	67988	float16x4_t __ret_869; \
	67989	float16_t __s0_869 = __p0_869; \
	67990	float16x4_t __s1_869 = __p1_869; \
	67991	float16_t __reint_869 = __s0_869; \
	67992	float16x4_t __reint1_869 = __s1_869; \
	67993	int16x4_t __reint2_869 = vset_lane_s16((int16_t ) &__reint_869, (int16x4_t ) &__reint1_869, __p2_869); \
	67994	__ret_869 = (float16x4_t ) &__reint2_869; \
	67995	__ret_869; \
	67996	})
	67997	#else
	67998	#define vset_lane_f16(__p0_870, __p1_870, __p2_870) __extension__ ({ \
	67999	float16x4_t __ret_870; \
	68000	float16_t __s0_870 = __p0_870; \
	68001	float16x4_t __s1_870 = __p1_870; \
	68002	float16x4_t __rev1_870; __rev1_870 = __builtin_shufflevector(__s1_870, __s1_870, 3, 2, 1, 0); \
	68003	float16_t __reint_870 = __s0_870; \
	68004	float16x4_t __reint1_870 = __rev1_870; \
	68005	int16x4_t __reint2_870 = __noswap_vset_lane_s16((int16_t ) &__reint_870, (int16x4_t ) &__reint1_870, __p2_870); \
	68006	__ret_870 = (float16x4_t ) &__reint2_870; \
	68007	__ret_870 = __builtin_shufflevector(__ret_870, __ret_870, 3, 2, 1, 0); \
	68008	__ret_870; \
	68009	})
	68010	#endif
	68011
	68012	#ifdef __LITTLE_ENDIAN__
	68013	#define vsetq_lane_f16(__p0_871, __p1_871, __p2_871) __extension__ ({ \
	68014	float16x8_t __ret_871; \
	68015	float16_t __s0_871 = __p0_871; \
	68016	float16x8_t __s1_871 = __p1_871; \
	68017	float16_t __reint_871 = __s0_871; \
	68018	float16x8_t __reint1_871 = __s1_871; \
	68019	int16x8_t __reint2_871 = vsetq_lane_s16((int16_t ) &__reint_871, (int16x8_t ) &__reint1_871, __p2_871); \
	68020	__ret_871 = (float16x8_t ) &__reint2_871; \
	68021	__ret_871; \
	68022	})
	68023	#else
	68024	#define vsetq_lane_f16(__p0_872, __p1_872, __p2_872) __extension__ ({ \
	68025	float16x8_t __ret_872; \
	68026	float16_t __s0_872 = __p0_872; \
	68027	float16x8_t __s1_872 = __p1_872; \
	68028	float16x8_t __rev1_872; __rev1_872 = __builtin_shufflevector(__s1_872, __s1_872, 7, 6, 5, 4, 3, 2, 1, 0); \
	68029	float16_t __reint_872 = __s0_872; \
	68030	float16x8_t __reint1_872 = __rev1_872; \
	68031	int16x8_t __reint2_872 = __noswap_vsetq_lane_s16((int16_t ) &__reint_872, (int16x8_t ) &__reint1_872, __p2_872); \
	68032	__ret_872 = (float16x8_t ) &__reint2_872; \
	68033	__ret_872 = __builtin_shufflevector(__ret_872, __ret_872, 7, 6, 5, 4, 3, 2, 1, 0); \
	68034	__ret_872; \
	68035	})
	68036	#endif
	68037
	68038	#ifdef __LITTLE_ENDIAN__
	68039	#define vbfmlalbq_lane_f32(__p0_873, __p1_873, __p2_873, __p3_873) __extension__ ({ \
	68040	float32x4_t __ret_873; \
	68041	float32x4_t __s0_873 = __p0_873; \
	68042	bfloat16x8_t __s1_873 = __p1_873; \
	68043	bfloat16x4_t __s2_873 = __p2_873; \
	68044	__ret_873 = vbfmlalbq_f32(__s0_873, __s1_873, (bfloat16x8_t) {vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873), vget_lane_bf16(__s2_873, __p3_873)}); \
68157	68045	__ret_873; \
68158	68046	})
68159	68047	#else
68160		#define vset_lane_f16(__p0_874, __p1_874, __p2_874) __extension__ ({ \
68161		float16x4_t __ret_874; \
68162		float16_t __s0_874 = __p0_874; \
68163		float16x4_t __s1_874 = __p1_874; \
68164		float16x4_t __rev1_874; __rev1_874 = __builtin_shufflevector(__s1_874, __s1_874, 3, 2, 1, 0); \
68165		float16_t __reint_874 = __s0_874; \
68166		float16x4_t __reint1_874 = __rev1_874; \
68167		int16x4_t __reint2_874 = __noswap_vset_lane_s16((int16_t ) &__reint_874, (int16x4_t ) &__reint1_874, __p2_874); \
68168		__ret_874 = (float16x4_t ) &__reint2_874; \
	68048	#define vbfmlalbq_lane_f32(__p0_874, __p1_874, __p2_874, __p3_874) __extension__ ({ \
	68049	float32x4_t __ret_874; \
	68050	float32x4_t __s0_874 = __p0_874; \
	68051	bfloat16x8_t __s1_874 = __p1_874; \
	68052	bfloat16x4_t __s2_874 = __p2_874; \
	68053	float32x4_t __rev0_874; __rev0_874 = __builtin_shufflevector(__s0_874, __s0_874, 3, 2, 1, 0); \
	68054	bfloat16x8_t __rev1_874; __rev1_874 = __builtin_shufflevector(__s1_874, __s1_874, 7, 6, 5, 4, 3, 2, 1, 0); \
	68055	bfloat16x4_t __rev2_874; __rev2_874 = __builtin_shufflevector(__s2_874, __s2_874, 3, 2, 1, 0); \
	68056	__ret_874 = __noswap_vbfmlalbq_f32(__rev0_874, __rev1_874, (bfloat16x8_t) {__noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874), __noswap_vget_lane_bf16(__rev2_874, __p3_874)}); \
68169	68057	__ret_874 = __builtin_shufflevector(__ret_874, __ret_874, 3, 2, 1, 0); \
68170	68058	__ret_874; \
68171	68059	})
68172	68060	#endif
68173	68061
68174	68062	#ifdef __LITTLE_ENDIAN__
68175		#define vsetq_lane_f16(__p0_875, __p1_875, __p2_875) __extension__ ({ \
68176		float16x8_t __ret_875; \
68177		float16_t __s0_875 = __p0_875; \
68178		float16x8_t __s1_875 = __p1_875; \
68179		float16_t __reint_875 = __s0_875; \
68180		float16x8_t __reint1_875 = __s1_875; \
68181		int16x8_t __reint2_875 = vsetq_lane_s16((int16_t ) &__reint_875, (int16x8_t ) &__reint1_875, __p2_875); \
68182		__ret_875 = (float16x8_t ) &__reint2_875; \
	68063	#define vbfmlalbq_laneq_f32(__p0_875, __p1_875, __p2_875, __p3_875) __extension__ ({ \
	68064	float32x4_t __ret_875; \
	68065	float32x4_t __s0_875 = __p0_875; \
	68066	bfloat16x8_t __s1_875 = __p1_875; \
	68067	bfloat16x8_t __s2_875 = __p2_875; \
	68068	__ret_875 = vbfmlalbq_f32(__s0_875, __s1_875, (bfloat16x8_t) {vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875), vgetq_lane_bf16(__s2_875, __p3_875)}); \
68183	68069	__ret_875; \
68184	68070	})
68185	68071	#else
68186		#define vsetq_lane_f16(__p0_876, __p1_876, __p2_876) __extension__ ({ \
68187		float16x8_t __ret_876; \
68188		float16_t __s0_876 = __p0_876; \
68189		float16x8_t __s1_876 = __p1_876; \
68190		float16x8_t __rev1_876; __rev1_876 = __builtin_shufflevector(__s1_876, __s1_876, 7, 6, 5, 4, 3, 2, 1, 0); \
68191		float16_t __reint_876 = __s0_876; \
68192		float16x8_t __reint1_876 = __rev1_876; \
68193		int16x8_t __reint2_876 = __noswap_vsetq_lane_s16((int16_t ) &__reint_876, (int16x8_t ) &__reint1_876, __p2_876); \
68194		__ret_876 = (float16x8_t ) &__reint2_876; \
68195		__ret_876 = __builtin_shufflevector(__ret_876, __ret_876, 7, 6, 5, 4, 3, 2, 1, 0); \
	68072	#define vbfmlalbq_laneq_f32(__p0_876, __p1_876, __p2_876, __p3_876) __extension__ ({ \
	68073	float32x4_t __ret_876; \
	68074	float32x4_t __s0_876 = __p0_876; \
	68075	bfloat16x8_t __s1_876 = __p1_876; \
	68076	bfloat16x8_t __s2_876 = __p2_876; \
	68077	float32x4_t __rev0_876; __rev0_876 = __builtin_shufflevector(__s0_876, __s0_876, 3, 2, 1, 0); \
	68078	bfloat16x8_t __rev1_876; __rev1_876 = __builtin_shufflevector(__s1_876, __s1_876, 7, 6, 5, 4, 3, 2, 1, 0); \
	68079	bfloat16x8_t __rev2_876; __rev2_876 = __builtin_shufflevector(__s2_876, __s2_876, 7, 6, 5, 4, 3, 2, 1, 0); \
	68080	__ret_876 = __noswap_vbfmlalbq_f32(__rev0_876, __rev1_876, (bfloat16x8_t) {__noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876), __noswap_vgetq_lane_bf16(__rev2_876, __p3_876)}); \
	68081	__ret_876 = __builtin_shufflevector(__ret_876, __ret_876, 3, 2, 1, 0); \
68196	68082	__ret_876; \
68197	68083	})
68198	68084	#endif
68199	68085
68200		#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC)
68201	68086	#ifdef __LITTLE_ENDIAN__
68202		#define vbfmlalbq_lane_f32(__p0_877, __p1_877, __p2_877, __p3_877) __extension__ ({ \
	68087	#define vbfmlaltq_lane_f32(__p0_877, __p1_877, __p2_877, __p3_877) __extension__ ({ \
68203	68088	float32x4_t __ret_877; \
68204	68089	float32x4_t __s0_877 = __p0_877; \
68205	68090	bfloat16x8_t __s1_877 = __p1_877; \
68206	68091	bfloat16x4_t __s2_877 = __p2_877; \
68207		__ret_877 = vbfmlalbq_f32(__s0_877, __s1_877, (bfloat16x8_t) {vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877)}); \
	68092	__ret_877 = vbfmlaltq_f32(__s0_877, __s1_877, (bfloat16x8_t) {vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877), vget_lane_bf16(__s2_877, __p3_877)}); \
68208	68093	__ret_877; \
68209	68094	})
68210	68095	#else
68211		#define vbfmlalbq_lane_f32(__p0_878, __p1_878, __p2_878, __p3_878) __extension__ ({ \
	68096	#define vbfmlaltq_lane_f32(__p0_878, __p1_878, __p2_878, __p3_878) __extension__ ({ \
68212	68097	float32x4_t __ret_878; \
68213	68098	float32x4_t __s0_878 = __p0_878; \
68214	68099	bfloat16x8_t __s1_878 = __p1_878; \
...	...	@@ -68216,23 +68101,23 @@ int16x8_t __reint2_876 = __noswap_vsetq_lane_s16((int16_t ) &__reint_876, *(in
68216	68101	float32x4_t __rev0_878; __rev0_878 = __builtin_shufflevector(__s0_878, __s0_878, 3, 2, 1, 0); \
68217	68102	bfloat16x8_t __rev1_878; __rev1_878 = __builtin_shufflevector(__s1_878, __s1_878, 7, 6, 5, 4, 3, 2, 1, 0); \
68218	68103	bfloat16x4_t __rev2_878; __rev2_878 = __builtin_shufflevector(__s2_878, __s2_878, 3, 2, 1, 0); \
68219		__ret_878 = __noswap_vbfmlalbq_f32(__rev0_878, __rev1_878, (bfloat16x8_t) {__noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878)}); \
	68104	__ret_878 = __noswap_vbfmlaltq_f32(__rev0_878, __rev1_878, (bfloat16x8_t) {__noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878), __noswap_vget_lane_bf16(__rev2_878, __p3_878)}); \
68220	68105	__ret_878 = __builtin_shufflevector(__ret_878, __ret_878, 3, 2, 1, 0); \
68221	68106	__ret_878; \
68222	68107	})
68223	68108	#endif
68224	68109
68225	68110	#ifdef __LITTLE_ENDIAN__
68226		#define vbfmlalbq_laneq_f32(__p0_879, __p1_879, __p2_879, __p3_879) __extension__ ({ \
	68111	#define vbfmlaltq_laneq_f32(__p0_879, __p1_879, __p2_879, __p3_879) __extension__ ({ \
68227	68112	float32x4_t __ret_879; \
68228	68113	float32x4_t __s0_879 = __p0_879; \
68229	68114	bfloat16x8_t __s1_879 = __p1_879; \
68230	68115	bfloat16x8_t __s2_879 = __p2_879; \
68231		__ret_879 = vbfmlalbq_f32(__s0_879, __s1_879, (bfloat16x8_t) {vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879)}); \
	68116	__ret_879 = vbfmlaltq_f32(__s0_879, __s1_879, (bfloat16x8_t) {vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879), vgetq_lane_bf16(__s2_879, __p3_879)}); \
68232	68117	__ret_879; \
68233	68118	})
68234	68119	#else
68235		#define vbfmlalbq_laneq_f32(__p0_880, __p1_880, __p2_880, __p3_880) __extension__ ({ \
	68120	#define vbfmlaltq_laneq_f32(__p0_880, __p1_880, __p2_880, __p3_880) __extension__ ({ \
68236	68121	float32x4_t __ret_880; \
68237	68122	float32x4_t __s0_880 = __p0_880; \
68238	68123	bfloat16x8_t __s1_880 = __p1_880; \
...	...	@@ -68240,68 +68125,20 @@ int16x8_t __reint2_876 = __noswap_vsetq_lane_s16((int16_t ) &__reint_876, *(in
68240	68125	float32x4_t __rev0_880; __rev0_880 = __builtin_shufflevector(__s0_880, __s0_880, 3, 2, 1, 0); \
68241	68126	bfloat16x8_t __rev1_880; __rev1_880 = __builtin_shufflevector(__s1_880, __s1_880, 7, 6, 5, 4, 3, 2, 1, 0); \
68242	68127	bfloat16x8_t __rev2_880; __rev2_880 = __builtin_shufflevector(__s2_880, __s2_880, 7, 6, 5, 4, 3, 2, 1, 0); \
68243		__ret_880 = __noswap_vbfmlalbq_f32(__rev0_880, __rev1_880, (bfloat16x8_t) {__noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880)}); \
	68128	__ret_880 = __noswap_vbfmlaltq_f32(__rev0_880, __rev1_880, (bfloat16x8_t) {__noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880), __noswap_vgetq_lane_bf16(__rev2_880, __p3_880)}); \
68244	68129	__ret_880 = __builtin_shufflevector(__ret_880, __ret_880, 3, 2, 1, 0); \
68245	68130	__ret_880; \
68246	68131	})
68247	68132	#endif
68248	68133
68249	68134	#ifdef __LITTLE_ENDIAN__
68250		#define vbfmlaltq_lane_f32(__p0_881, __p1_881, __p2_881, __p3_881) __extension__ ({ \
68251		float32x4_t __ret_881; \
68252		float32x4_t __s0_881 = __p0_881; \
68253		bfloat16x8_t __s1_881 = __p1_881; \
68254		bfloat16x4_t __s2_881 = __p2_881; \
68255		__ret_881 = vbfmlaltq_f32(__s0_881, __s1_881, (bfloat16x8_t) {vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881), vget_lane_bf16(__s2_881, __p3_881)}); \
68256		__ret_881; \
68257		})
68258		#else
68259		#define vbfmlaltq_lane_f32(__p0_882, __p1_882, __p2_882, __p3_882) __extension__ ({ \
68260		float32x4_t __ret_882; \
68261		float32x4_t __s0_882 = __p0_882; \
68262		bfloat16x8_t __s1_882 = __p1_882; \
68263		bfloat16x4_t __s2_882 = __p2_882; \
68264		float32x4_t __rev0_882; __rev0_882 = __builtin_shufflevector(__s0_882, __s0_882, 3, 2, 1, 0); \
68265		bfloat16x8_t __rev1_882; __rev1_882 = __builtin_shufflevector(__s1_882, __s1_882, 7, 6, 5, 4, 3, 2, 1, 0); \
68266		bfloat16x4_t __rev2_882; __rev2_882 = __builtin_shufflevector(__s2_882, __s2_882, 3, 2, 1, 0); \
68267		__ret_882 = __noswap_vbfmlaltq_f32(__rev0_882, __rev1_882, (bfloat16x8_t) {__noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882), __noswap_vget_lane_bf16(__rev2_882, __p3_882)}); \
68268		__ret_882 = __builtin_shufflevector(__ret_882, __ret_882, 3, 2, 1, 0); \
68269		__ret_882; \
68270		})
68271		#endif
68272
68273		#ifdef __LITTLE_ENDIAN__
68274		#define vbfmlaltq_laneq_f32(__p0_883, __p1_883, __p2_883, __p3_883) __extension__ ({ \
68275		float32x4_t __ret_883; \
68276		float32x4_t __s0_883 = __p0_883; \
68277		bfloat16x8_t __s1_883 = __p1_883; \
68278		bfloat16x8_t __s2_883 = __p2_883; \
68279		__ret_883 = vbfmlaltq_f32(__s0_883, __s1_883, (bfloat16x8_t) {vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883), vgetq_lane_bf16(__s2_883, __p3_883)}); \
68280		__ret_883; \
68281		})
68282		#else
68283		#define vbfmlaltq_laneq_f32(__p0_884, __p1_884, __p2_884, __p3_884) __extension__ ({ \
68284		float32x4_t __ret_884; \
68285		float32x4_t __s0_884 = __p0_884; \
68286		bfloat16x8_t __s1_884 = __p1_884; \
68287		bfloat16x8_t __s2_884 = __p2_884; \
68288		float32x4_t __rev0_884; __rev0_884 = __builtin_shufflevector(__s0_884, __s0_884, 3, 2, 1, 0); \
68289		bfloat16x8_t __rev1_884; __rev1_884 = __builtin_shufflevector(__s1_884, __s1_884, 7, 6, 5, 4, 3, 2, 1, 0); \
68290		bfloat16x8_t __rev2_884; __rev2_884 = __builtin_shufflevector(__s2_884, __s2_884, 7, 6, 5, 4, 3, 2, 1, 0); \
68291		__ret_884 = __noswap_vbfmlaltq_f32(__rev0_884, __rev1_884, (bfloat16x8_t) {__noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884), __noswap_vgetq_lane_bf16(__rev2_884, __p3_884)}); \
68292		__ret_884 = __builtin_shufflevector(__ret_884, __ret_884, 3, 2, 1, 0); \
68293		__ret_884; \
68294		})
68295		#endif
68296
68297		#ifdef __LITTLE_ENDIAN__
68298		__ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
	68135	__ai __attribute__((target("bf16"))) float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
68299	68136	float32x4_t __ret;
68300	68137	__ret = vcvt_f32_bf16(vget_high_bf16(__p0));
68301	68138	return __ret;
68302	68139	}
68303	68140	#else
68304		__ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
	68141	__ai __attribute__((target("bf16"))) float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
68305	68142	float32x4_t __ret;
68306	68143	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
68307	68144	__ret = __noswap_vcvt_f32_bf16(__noswap_vget_high_bf16(__rev0));
...	...	@@ -68311,13 +68148,13 @@ __ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
68311	68148	#endif
68312	68149
68313	68150	#ifdef __LITTLE_ENDIAN__
68314		__ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
	68151	__ai __attribute__((target("bf16"))) float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
68315	68152	float32x4_t __ret;
68316	68153	__ret = vcvt_f32_bf16(vget_low_bf16(__p0));
68317	68154	return __ret;
68318	68155	}
68319	68156	#else
68320		__ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
	68157	__ai __attribute__((target("bf16"))) float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
68321	68158	float32x4_t __ret;
68322	68159	bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
68323	68160	__ret = __noswap_vcvt_f32_bf16(__noswap_vget_low_bf16(__rev0));
...	...	@@ -68326,487 +68163,58 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
68326	68163	}
68327	68164	#endif
68328	68165
68329		#endif
68330		#if defined(__ARM_FEATURE_FP16_FML) && defined(__aarch64__)
68331		#ifdef __LITTLE_ENDIAN__
68332		#define vfmlalq_lane_high_f16(__p0_885, __p1_885, __p2_885, __p3_885) __extension__ ({ \
68333		float32x4_t __ret_885; \
68334		float32x4_t __s0_885 = __p0_885; \
68335		float16x8_t __s1_885 = __p1_885; \
68336		float16x4_t __s2_885 = __p2_885; \
68337		__ret_885 = vfmlalq_high_f16(__s0_885, __s1_885, (float16x8_t) {vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885), vget_lane_f16(__s2_885, __p3_885)}); \
68338		__ret_885; \
68339		})
68340		#else
68341		#define vfmlalq_lane_high_f16(__p0_886, __p1_886, __p2_886, __p3_886) __extension__ ({ \
68342		float32x4_t __ret_886; \
68343		float32x4_t __s0_886 = __p0_886; \
68344		float16x8_t __s1_886 = __p1_886; \
68345		float16x4_t __s2_886 = __p2_886; \
68346		float32x4_t __rev0_886; __rev0_886 = __builtin_shufflevector(__s0_886, __s0_886, 3, 2, 1, 0); \
68347		float16x8_t __rev1_886; __rev1_886 = __builtin_shufflevector(__s1_886, __s1_886, 7, 6, 5, 4, 3, 2, 1, 0); \
68348		float16x4_t __rev2_886; __rev2_886 = __builtin_shufflevector(__s2_886, __s2_886, 3, 2, 1, 0); \
68349		__ret_886 = __noswap_vfmlalq_high_f16(__rev0_886, __rev1_886, (float16x8_t) {__noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886), __noswap_vget_lane_f16(__rev2_886, __p3_886)}); \
68350		__ret_886 = __builtin_shufflevector(__ret_886, __ret_886, 3, 2, 1, 0); \
68351		__ret_886; \
68352		})
68353		#endif
68354
68355		#ifdef __LITTLE_ENDIAN__
68356		#define vfmlal_lane_high_f16(__p0_887, __p1_887, __p2_887, __p3_887) __extension__ ({ \
68357		float32x2_t __ret_887; \
68358		float32x2_t __s0_887 = __p0_887; \
68359		float16x4_t __s1_887 = __p1_887; \
68360		float16x4_t __s2_887 = __p2_887; \
68361		__ret_887 = vfmlal_high_f16(__s0_887, __s1_887, (float16x4_t) {vget_lane_f16(__s2_887, __p3_887), vget_lane_f16(__s2_887, __p3_887), vget_lane_f16(__s2_887, __p3_887), vget_lane_f16(__s2_887, __p3_887)}); \
68362		__ret_887; \
68363		})
68364		#else
68365		#define vfmlal_lane_high_f16(__p0_888, __p1_888, __p2_888, __p3_888) __extension__ ({ \
68366		float32x2_t __ret_888; \
68367		float32x2_t __s0_888 = __p0_888; \
68368		float16x4_t __s1_888 = __p1_888; \
68369		float16x4_t __s2_888 = __p2_888; \
68370		float32x2_t __rev0_888; __rev0_888 = __builtin_shufflevector(__s0_888, __s0_888, 1, 0); \
68371		float16x4_t __rev1_888; __rev1_888 = __builtin_shufflevector(__s1_888, __s1_888, 3, 2, 1, 0); \
68372		float16x4_t __rev2_888; __rev2_888 = __builtin_shufflevector(__s2_888, __s2_888, 3, 2, 1, 0); \
68373		__ret_888 = __noswap_vfmlal_high_f16(__rev0_888, __rev1_888, (float16x4_t) {__noswap_vget_lane_f16(__rev2_888, __p3_888), __noswap_vget_lane_f16(__rev2_888, __p3_888), __noswap_vget_lane_f16(__rev2_888, __p3_888), __noswap_vget_lane_f16(__rev2_888, __p3_888)}); \
68374		__ret_888 = __builtin_shufflevector(__ret_888, __ret_888, 1, 0); \
68375		__ret_888; \
68376		})
68377		#endif
68378
68379		#ifdef __LITTLE_ENDIAN__
68380		#define vfmlalq_lane_low_f16(__p0_889, __p1_889, __p2_889, __p3_889) __extension__ ({ \
68381		float32x4_t __ret_889; \
68382		float32x4_t __s0_889 = __p0_889; \
68383		float16x8_t __s1_889 = __p1_889; \
68384		float16x4_t __s2_889 = __p2_889; \
68385		__ret_889 = vfmlalq_low_f16(__s0_889, __s1_889, (float16x8_t) {vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889), vget_lane_f16(__s2_889, __p3_889)}); \
68386		__ret_889; \
68387		})
68388		#else
68389		#define vfmlalq_lane_low_f16(__p0_890, __p1_890, __p2_890, __p3_890) __extension__ ({ \
68390		float32x4_t __ret_890; \
68391		float32x4_t __s0_890 = __p0_890; \
68392		float16x8_t __s1_890 = __p1_890; \
68393		float16x4_t __s2_890 = __p2_890; \
68394		float32x4_t __rev0_890; __rev0_890 = __builtin_shufflevector(__s0_890, __s0_890, 3, 2, 1, 0); \
68395		float16x8_t __rev1_890; __rev1_890 = __builtin_shufflevector(__s1_890, __s1_890, 7, 6, 5, 4, 3, 2, 1, 0); \
68396		float16x4_t __rev2_890; __rev2_890 = __builtin_shufflevector(__s2_890, __s2_890, 3, 2, 1, 0); \
68397		__ret_890 = __noswap_vfmlalq_low_f16(__rev0_890, __rev1_890, (float16x8_t) {__noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890), __noswap_vget_lane_f16(__rev2_890, __p3_890)}); \
68398		__ret_890 = __builtin_shufflevector(__ret_890, __ret_890, 3, 2, 1, 0); \
68399		__ret_890; \
68400		})
68401		#endif
68402
68403		#ifdef __LITTLE_ENDIAN__
68404		#define vfmlal_lane_low_f16(__p0_891, __p1_891, __p2_891, __p3_891) __extension__ ({ \
68405		float32x2_t __ret_891; \
68406		float32x2_t __s0_891 = __p0_891; \
68407		float16x4_t __s1_891 = __p1_891; \
68408		float16x4_t __s2_891 = __p2_891; \
68409		__ret_891 = vfmlal_low_f16(__s0_891, __s1_891, (float16x4_t) {vget_lane_f16(__s2_891, __p3_891), vget_lane_f16(__s2_891, __p3_891), vget_lane_f16(__s2_891, __p3_891), vget_lane_f16(__s2_891, __p3_891)}); \
68410		__ret_891; \
68411		})
68412		#else
68413		#define vfmlal_lane_low_f16(__p0_892, __p1_892, __p2_892, __p3_892) __extension__ ({ \
68414		float32x2_t __ret_892; \
68415		float32x2_t __s0_892 = __p0_892; \
68416		float16x4_t __s1_892 = __p1_892; \
68417		float16x4_t __s2_892 = __p2_892; \
68418		float32x2_t __rev0_892; __rev0_892 = __builtin_shufflevector(__s0_892, __s0_892, 1, 0); \
68419		float16x4_t __rev1_892; __rev1_892 = __builtin_shufflevector(__s1_892, __s1_892, 3, 2, 1, 0); \
68420		float16x4_t __rev2_892; __rev2_892 = __builtin_shufflevector(__s2_892, __s2_892, 3, 2, 1, 0); \
68421		__ret_892 = __noswap_vfmlal_low_f16(__rev0_892, __rev1_892, (float16x4_t) {__noswap_vget_lane_f16(__rev2_892, __p3_892), __noswap_vget_lane_f16(__rev2_892, __p3_892), __noswap_vget_lane_f16(__rev2_892, __p3_892), __noswap_vget_lane_f16(__rev2_892, __p3_892)}); \
68422		__ret_892 = __builtin_shufflevector(__ret_892, __ret_892, 1, 0); \
68423		__ret_892; \
68424		})
68425		#endif
68426
68427		#ifdef __LITTLE_ENDIAN__
68428		#define vfmlalq_laneq_high_f16(__p0_893, __p1_893, __p2_893, __p3_893) __extension__ ({ \
68429		float32x4_t __ret_893; \
68430		float32x4_t __s0_893 = __p0_893; \
68431		float16x8_t __s1_893 = __p1_893; \
68432		float16x8_t __s2_893 = __p2_893; \
68433		__ret_893 = vfmlalq_high_f16(__s0_893, __s1_893, (float16x8_t) {vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893), vgetq_lane_f16(__s2_893, __p3_893)}); \
68434		__ret_893; \
68435		})
68436		#else
68437		#define vfmlalq_laneq_high_f16(__p0_894, __p1_894, __p2_894, __p3_894) __extension__ ({ \
68438		float32x4_t __ret_894; \
68439		float32x4_t __s0_894 = __p0_894; \
68440		float16x8_t __s1_894 = __p1_894; \
68441		float16x8_t __s2_894 = __p2_894; \
68442		float32x4_t __rev0_894; __rev0_894 = __builtin_shufflevector(__s0_894, __s0_894, 3, 2, 1, 0); \
68443		float16x8_t __rev1_894; __rev1_894 = __builtin_shufflevector(__s1_894, __s1_894, 7, 6, 5, 4, 3, 2, 1, 0); \
68444		float16x8_t __rev2_894; __rev2_894 = __builtin_shufflevector(__s2_894, __s2_894, 7, 6, 5, 4, 3, 2, 1, 0); \
68445		__ret_894 = __noswap_vfmlalq_high_f16(__rev0_894, __rev1_894, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894), __noswap_vgetq_lane_f16(__rev2_894, __p3_894)}); \
68446		__ret_894 = __builtin_shufflevector(__ret_894, __ret_894, 3, 2, 1, 0); \
68447		__ret_894; \
68448		})
68449		#endif
68450
68451		#ifdef __LITTLE_ENDIAN__
68452		#define vfmlal_laneq_high_f16(__p0_895, __p1_895, __p2_895, __p3_895) __extension__ ({ \
68453		float32x2_t __ret_895; \
68454		float32x2_t __s0_895 = __p0_895; \
68455		float16x4_t __s1_895 = __p1_895; \
68456		float16x8_t __s2_895 = __p2_895; \
68457		__ret_895 = vfmlal_high_f16(__s0_895, __s1_895, (float16x4_t) {vgetq_lane_f16(__s2_895, __p3_895), vgetq_lane_f16(__s2_895, __p3_895), vgetq_lane_f16(__s2_895, __p3_895), vgetq_lane_f16(__s2_895, __p3_895)}); \
68458		__ret_895; \
68459		})
68460		#else
68461		#define vfmlal_laneq_high_f16(__p0_896, __p1_896, __p2_896, __p3_896) __extension__ ({ \
68462		float32x2_t __ret_896; \
68463		float32x2_t __s0_896 = __p0_896; \
68464		float16x4_t __s1_896 = __p1_896; \
68465		float16x8_t __s2_896 = __p2_896; \
68466		float32x2_t __rev0_896; __rev0_896 = __builtin_shufflevector(__s0_896, __s0_896, 1, 0); \
68467		float16x4_t __rev1_896; __rev1_896 = __builtin_shufflevector(__s1_896, __s1_896, 3, 2, 1, 0); \
68468		float16x8_t __rev2_896; __rev2_896 = __builtin_shufflevector(__s2_896, __s2_896, 7, 6, 5, 4, 3, 2, 1, 0); \
68469		__ret_896 = __noswap_vfmlal_high_f16(__rev0_896, __rev1_896, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_896, __p3_896), __noswap_vgetq_lane_f16(__rev2_896, __p3_896), __noswap_vgetq_lane_f16(__rev2_896, __p3_896), __noswap_vgetq_lane_f16(__rev2_896, __p3_896)}); \
68470		__ret_896 = __builtin_shufflevector(__ret_896, __ret_896, 1, 0); \
68471		__ret_896; \
68472		})
68473		#endif
68474
68475		#ifdef __LITTLE_ENDIAN__
68476		#define vfmlalq_laneq_low_f16(__p0_897, __p1_897, __p2_897, __p3_897) __extension__ ({ \
68477		float32x4_t __ret_897; \
68478		float32x4_t __s0_897 = __p0_897; \
68479		float16x8_t __s1_897 = __p1_897; \
68480		float16x8_t __s2_897 = __p2_897; \
68481		__ret_897 = vfmlalq_low_f16(__s0_897, __s1_897, (float16x8_t) {vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897), vgetq_lane_f16(__s2_897, __p3_897)}); \
68482		__ret_897; \
68483		})
68484		#else
68485		#define vfmlalq_laneq_low_f16(__p0_898, __p1_898, __p2_898, __p3_898) __extension__ ({ \
68486		float32x4_t __ret_898; \
68487		float32x4_t __s0_898 = __p0_898; \
68488		float16x8_t __s1_898 = __p1_898; \
68489		float16x8_t __s2_898 = __p2_898; \
68490		float32x4_t __rev0_898; __rev0_898 = __builtin_shufflevector(__s0_898, __s0_898, 3, 2, 1, 0); \
68491		float16x8_t __rev1_898; __rev1_898 = __builtin_shufflevector(__s1_898, __s1_898, 7, 6, 5, 4, 3, 2, 1, 0); \
68492		float16x8_t __rev2_898; __rev2_898 = __builtin_shufflevector(__s2_898, __s2_898, 7, 6, 5, 4, 3, 2, 1, 0); \
68493		__ret_898 = __noswap_vfmlalq_low_f16(__rev0_898, __rev1_898, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898), __noswap_vgetq_lane_f16(__rev2_898, __p3_898)}); \
68494		__ret_898 = __builtin_shufflevector(__ret_898, __ret_898, 3, 2, 1, 0); \
68495		__ret_898; \
68496		})
68497		#endif
68498
68499		#ifdef __LITTLE_ENDIAN__
68500		#define vfmlal_laneq_low_f16(__p0_899, __p1_899, __p2_899, __p3_899) __extension__ ({ \
68501		float32x2_t __ret_899; \
68502		float32x2_t __s0_899 = __p0_899; \
68503		float16x4_t __s1_899 = __p1_899; \
68504		float16x8_t __s2_899 = __p2_899; \
68505		__ret_899 = vfmlal_low_f16(__s0_899, __s1_899, (float16x4_t) {vgetq_lane_f16(__s2_899, __p3_899), vgetq_lane_f16(__s2_899, __p3_899), vgetq_lane_f16(__s2_899, __p3_899), vgetq_lane_f16(__s2_899, __p3_899)}); \
68506		__ret_899; \
68507		})
68508		#else
68509		#define vfmlal_laneq_low_f16(__p0_900, __p1_900, __p2_900, __p3_900) __extension__ ({ \
68510		float32x2_t __ret_900; \
68511		float32x2_t __s0_900 = __p0_900; \
68512		float16x4_t __s1_900 = __p1_900; \
68513		float16x8_t __s2_900 = __p2_900; \
68514		float32x2_t __rev0_900; __rev0_900 = __builtin_shufflevector(__s0_900, __s0_900, 1, 0); \
68515		float16x4_t __rev1_900; __rev1_900 = __builtin_shufflevector(__s1_900, __s1_900, 3, 2, 1, 0); \
68516		float16x8_t __rev2_900; __rev2_900 = __builtin_shufflevector(__s2_900, __s2_900, 7, 6, 5, 4, 3, 2, 1, 0); \
68517		__ret_900 = __noswap_vfmlal_low_f16(__rev0_900, __rev1_900, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_900, __p3_900), __noswap_vgetq_lane_f16(__rev2_900, __p3_900), __noswap_vgetq_lane_f16(__rev2_900, __p3_900), __noswap_vgetq_lane_f16(__rev2_900, __p3_900)}); \
68518		__ret_900 = __builtin_shufflevector(__ret_900, __ret_900, 1, 0); \
68519		__ret_900; \
68520		})
68521		#endif
68522
68523		#ifdef __LITTLE_ENDIAN__
68524		#define vfmlslq_lane_high_f16(__p0_901, __p1_901, __p2_901, __p3_901) __extension__ ({ \
68525		float32x4_t __ret_901; \
68526		float32x4_t __s0_901 = __p0_901; \
68527		float16x8_t __s1_901 = __p1_901; \
68528		float16x4_t __s2_901 = __p2_901; \
68529		__ret_901 = vfmlslq_high_f16(__s0_901, __s1_901, (float16x8_t) {vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901), vget_lane_f16(__s2_901, __p3_901)}); \
68530		__ret_901; \
68531		})
68532		#else
68533		#define vfmlslq_lane_high_f16(__p0_902, __p1_902, __p2_902, __p3_902) __extension__ ({ \
68534		float32x4_t __ret_902; \
68535		float32x4_t __s0_902 = __p0_902; \
68536		float16x8_t __s1_902 = __p1_902; \
68537		float16x4_t __s2_902 = __p2_902; \
68538		float32x4_t __rev0_902; __rev0_902 = __builtin_shufflevector(__s0_902, __s0_902, 3, 2, 1, 0); \
68539		float16x8_t __rev1_902; __rev1_902 = __builtin_shufflevector(__s1_902, __s1_902, 7, 6, 5, 4, 3, 2, 1, 0); \
68540		float16x4_t __rev2_902; __rev2_902 = __builtin_shufflevector(__s2_902, __s2_902, 3, 2, 1, 0); \
68541		__ret_902 = __noswap_vfmlslq_high_f16(__rev0_902, __rev1_902, (float16x8_t) {__noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902), __noswap_vget_lane_f16(__rev2_902, __p3_902)}); \
68542		__ret_902 = __builtin_shufflevector(__ret_902, __ret_902, 3, 2, 1, 0); \
68543		__ret_902; \
68544		})
68545		#endif
68546
68547		#ifdef __LITTLE_ENDIAN__
68548		#define vfmlsl_lane_high_f16(__p0_903, __p1_903, __p2_903, __p3_903) __extension__ ({ \
68549		float32x2_t __ret_903; \
68550		float32x2_t __s0_903 = __p0_903; \
68551		float16x4_t __s1_903 = __p1_903; \
68552		float16x4_t __s2_903 = __p2_903; \
68553		__ret_903 = vfmlsl_high_f16(__s0_903, __s1_903, (float16x4_t) {vget_lane_f16(__s2_903, __p3_903), vget_lane_f16(__s2_903, __p3_903), vget_lane_f16(__s2_903, __p3_903), vget_lane_f16(__s2_903, __p3_903)}); \
68554		__ret_903; \
68555		})
68556		#else
68557		#define vfmlsl_lane_high_f16(__p0_904, __p1_904, __p2_904, __p3_904) __extension__ ({ \
68558		float32x2_t __ret_904; \
68559		float32x2_t __s0_904 = __p0_904; \
68560		float16x4_t __s1_904 = __p1_904; \
68561		float16x4_t __s2_904 = __p2_904; \
68562		float32x2_t __rev0_904; __rev0_904 = __builtin_shufflevector(__s0_904, __s0_904, 1, 0); \
68563		float16x4_t __rev1_904; __rev1_904 = __builtin_shufflevector(__s1_904, __s1_904, 3, 2, 1, 0); \
68564		float16x4_t __rev2_904; __rev2_904 = __builtin_shufflevector(__s2_904, __s2_904, 3, 2, 1, 0); \
68565		__ret_904 = __noswap_vfmlsl_high_f16(__rev0_904, __rev1_904, (float16x4_t) {__noswap_vget_lane_f16(__rev2_904, __p3_904), __noswap_vget_lane_f16(__rev2_904, __p3_904), __noswap_vget_lane_f16(__rev2_904, __p3_904), __noswap_vget_lane_f16(__rev2_904, __p3_904)}); \
68566		__ret_904 = __builtin_shufflevector(__ret_904, __ret_904, 1, 0); \
68567		__ret_904; \
68568		})
68569		#endif
68570
68571		#ifdef __LITTLE_ENDIAN__
68572		#define vfmlslq_lane_low_f16(__p0_905, __p1_905, __p2_905, __p3_905) __extension__ ({ \
68573		float32x4_t __ret_905; \
68574		float32x4_t __s0_905 = __p0_905; \
68575		float16x8_t __s1_905 = __p1_905; \
68576		float16x4_t __s2_905 = __p2_905; \
68577		__ret_905 = vfmlslq_low_f16(__s0_905, __s1_905, (float16x8_t) {vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905), vget_lane_f16(__s2_905, __p3_905)}); \
68578		__ret_905; \
68579		})
68580		#else
68581		#define vfmlslq_lane_low_f16(__p0_906, __p1_906, __p2_906, __p3_906) __extension__ ({ \
68582		float32x4_t __ret_906; \
68583		float32x4_t __s0_906 = __p0_906; \
68584		float16x8_t __s1_906 = __p1_906; \
68585		float16x4_t __s2_906 = __p2_906; \
68586		float32x4_t __rev0_906; __rev0_906 = __builtin_shufflevector(__s0_906, __s0_906, 3, 2, 1, 0); \
68587		float16x8_t __rev1_906; __rev1_906 = __builtin_shufflevector(__s1_906, __s1_906, 7, 6, 5, 4, 3, 2, 1, 0); \
68588		float16x4_t __rev2_906; __rev2_906 = __builtin_shufflevector(__s2_906, __s2_906, 3, 2, 1, 0); \
68589		__ret_906 = __noswap_vfmlslq_low_f16(__rev0_906, __rev1_906, (float16x8_t) {__noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906), __noswap_vget_lane_f16(__rev2_906, __p3_906)}); \
68590		__ret_906 = __builtin_shufflevector(__ret_906, __ret_906, 3, 2, 1, 0); \
68591		__ret_906; \
68592		})
68593		#endif
68594
68595		#ifdef __LITTLE_ENDIAN__
68596		#define vfmlsl_lane_low_f16(__p0_907, __p1_907, __p2_907, __p3_907) __extension__ ({ \
68597		float32x2_t __ret_907; \
68598		float32x2_t __s0_907 = __p0_907; \
68599		float16x4_t __s1_907 = __p1_907; \
68600		float16x4_t __s2_907 = __p2_907; \
68601		__ret_907 = vfmlsl_low_f16(__s0_907, __s1_907, (float16x4_t) {vget_lane_f16(__s2_907, __p3_907), vget_lane_f16(__s2_907, __p3_907), vget_lane_f16(__s2_907, __p3_907), vget_lane_f16(__s2_907, __p3_907)}); \
68602		__ret_907; \
68603		})
68604		#else
68605		#define vfmlsl_lane_low_f16(__p0_908, __p1_908, __p2_908, __p3_908) __extension__ ({ \
68606		float32x2_t __ret_908; \
68607		float32x2_t __s0_908 = __p0_908; \
68608		float16x4_t __s1_908 = __p1_908; \
68609		float16x4_t __s2_908 = __p2_908; \
68610		float32x2_t __rev0_908; __rev0_908 = __builtin_shufflevector(__s0_908, __s0_908, 1, 0); \
68611		float16x4_t __rev1_908; __rev1_908 = __builtin_shufflevector(__s1_908, __s1_908, 3, 2, 1, 0); \
68612		float16x4_t __rev2_908; __rev2_908 = __builtin_shufflevector(__s2_908, __s2_908, 3, 2, 1, 0); \
68613		__ret_908 = __noswap_vfmlsl_low_f16(__rev0_908, __rev1_908, (float16x4_t) {__noswap_vget_lane_f16(__rev2_908, __p3_908), __noswap_vget_lane_f16(__rev2_908, __p3_908), __noswap_vget_lane_f16(__rev2_908, __p3_908), __noswap_vget_lane_f16(__rev2_908, __p3_908)}); \
68614		__ret_908 = __builtin_shufflevector(__ret_908, __ret_908, 1, 0); \
68615		__ret_908; \
68616		})
68617		#endif
68618
68619		#ifdef __LITTLE_ENDIAN__
68620		#define vfmlslq_laneq_high_f16(__p0_909, __p1_909, __p2_909, __p3_909) __extension__ ({ \
68621		float32x4_t __ret_909; \
68622		float32x4_t __s0_909 = __p0_909; \
68623		float16x8_t __s1_909 = __p1_909; \
68624		float16x8_t __s2_909 = __p2_909; \
68625		__ret_909 = vfmlslq_high_f16(__s0_909, __s1_909, (float16x8_t) {vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909), vgetq_lane_f16(__s2_909, __p3_909)}); \
68626		__ret_909; \
68627		})
68628		#else
68629		#define vfmlslq_laneq_high_f16(__p0_910, __p1_910, __p2_910, __p3_910) __extension__ ({ \
68630		float32x4_t __ret_910; \
68631		float32x4_t __s0_910 = __p0_910; \
68632		float16x8_t __s1_910 = __p1_910; \
68633		float16x8_t __s2_910 = __p2_910; \
68634		float32x4_t __rev0_910; __rev0_910 = __builtin_shufflevector(__s0_910, __s0_910, 3, 2, 1, 0); \
68635		float16x8_t __rev1_910; __rev1_910 = __builtin_shufflevector(__s1_910, __s1_910, 7, 6, 5, 4, 3, 2, 1, 0); \
68636		float16x8_t __rev2_910; __rev2_910 = __builtin_shufflevector(__s2_910, __s2_910, 7, 6, 5, 4, 3, 2, 1, 0); \
68637		__ret_910 = __noswap_vfmlslq_high_f16(__rev0_910, __rev1_910, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910), __noswap_vgetq_lane_f16(__rev2_910, __p3_910)}); \
68638		__ret_910 = __builtin_shufflevector(__ret_910, __ret_910, 3, 2, 1, 0); \
68639		__ret_910; \
68640		})
68641		#endif
68642
68643		#ifdef __LITTLE_ENDIAN__
68644		#define vfmlsl_laneq_high_f16(__p0_911, __p1_911, __p2_911, __p3_911) __extension__ ({ \
68645		float32x2_t __ret_911; \
68646		float32x2_t __s0_911 = __p0_911; \
68647		float16x4_t __s1_911 = __p1_911; \
68648		float16x8_t __s2_911 = __p2_911; \
68649		__ret_911 = vfmlsl_high_f16(__s0_911, __s1_911, (float16x4_t) {vgetq_lane_f16(__s2_911, __p3_911), vgetq_lane_f16(__s2_911, __p3_911), vgetq_lane_f16(__s2_911, __p3_911), vgetq_lane_f16(__s2_911, __p3_911)}); \
68650		__ret_911; \
68651		})
68652		#else
68653		#define vfmlsl_laneq_high_f16(__p0_912, __p1_912, __p2_912, __p3_912) __extension__ ({ \
68654		float32x2_t __ret_912; \
68655		float32x2_t __s0_912 = __p0_912; \
68656		float16x4_t __s1_912 = __p1_912; \
68657		float16x8_t __s2_912 = __p2_912; \
68658		float32x2_t __rev0_912; __rev0_912 = __builtin_shufflevector(__s0_912, __s0_912, 1, 0); \
68659		float16x4_t __rev1_912; __rev1_912 = __builtin_shufflevector(__s1_912, __s1_912, 3, 2, 1, 0); \
68660		float16x8_t __rev2_912; __rev2_912 = __builtin_shufflevector(__s2_912, __s2_912, 7, 6, 5, 4, 3, 2, 1, 0); \
68661		__ret_912 = __noswap_vfmlsl_high_f16(__rev0_912, __rev1_912, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_912, __p3_912), __noswap_vgetq_lane_f16(__rev2_912, __p3_912), __noswap_vgetq_lane_f16(__rev2_912, __p3_912), __noswap_vgetq_lane_f16(__rev2_912, __p3_912)}); \
68662		__ret_912 = __builtin_shufflevector(__ret_912, __ret_912, 1, 0); \
68663		__ret_912; \
68664		})
68665		#endif
68666
68667		#ifdef __LITTLE_ENDIAN__
68668		#define vfmlslq_laneq_low_f16(__p0_913, __p1_913, __p2_913, __p3_913) __extension__ ({ \
68669		float32x4_t __ret_913; \
68670		float32x4_t __s0_913 = __p0_913; \
68671		float16x8_t __s1_913 = __p1_913; \
68672		float16x8_t __s2_913 = __p2_913; \
68673		__ret_913 = vfmlslq_low_f16(__s0_913, __s1_913, (float16x8_t) {vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913), vgetq_lane_f16(__s2_913, __p3_913)}); \
68674		__ret_913; \
68675		})
68676		#else
68677		#define vfmlslq_laneq_low_f16(__p0_914, __p1_914, __p2_914, __p3_914) __extension__ ({ \
68678		float32x4_t __ret_914; \
68679		float32x4_t __s0_914 = __p0_914; \
68680		float16x8_t __s1_914 = __p1_914; \
68681		float16x8_t __s2_914 = __p2_914; \
68682		float32x4_t __rev0_914; __rev0_914 = __builtin_shufflevector(__s0_914, __s0_914, 3, 2, 1, 0); \
68683		float16x8_t __rev1_914; __rev1_914 = __builtin_shufflevector(__s1_914, __s1_914, 7, 6, 5, 4, 3, 2, 1, 0); \
68684		float16x8_t __rev2_914; __rev2_914 = __builtin_shufflevector(__s2_914, __s2_914, 7, 6, 5, 4, 3, 2, 1, 0); \
68685		__ret_914 = __noswap_vfmlslq_low_f16(__rev0_914, __rev1_914, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914), __noswap_vgetq_lane_f16(__rev2_914, __p3_914)}); \
68686		__ret_914 = __builtin_shufflevector(__ret_914, __ret_914, 3, 2, 1, 0); \
68687		__ret_914; \
68688		})
68689		#endif
68690
68691	68166	#ifdef __LITTLE_ENDIAN__
68692		#define vfmlsl_laneq_low_f16(__p0_915, __p1_915, __p2_915, __p3_915) __extension__ ({ \
68693		float32x2_t __ret_915; \
68694		float32x2_t __s0_915 = __p0_915; \
68695		float16x4_t __s1_915 = __p1_915; \
68696		float16x8_t __s2_915 = __p2_915; \
68697		__ret_915 = vfmlsl_low_f16(__s0_915, __s1_915, (float16x4_t) {vgetq_lane_f16(__s2_915, __p3_915), vgetq_lane_f16(__s2_915, __p3_915), vgetq_lane_f16(__s2_915, __p3_915), vgetq_lane_f16(__s2_915, __p3_915)}); \
68698		__ret_915; \
68699		})
68700		#else
68701		#define vfmlsl_laneq_low_f16(__p0_916, __p1_916, __p2_916, __p3_916) __extension__ ({ \
68702		float32x2_t __ret_916; \
68703		float32x2_t __s0_916 = __p0_916; \
68704		float16x4_t __s1_916 = __p1_916; \
68705		float16x8_t __s2_916 = __p2_916; \
68706		float32x2_t __rev0_916; __rev0_916 = __builtin_shufflevector(__s0_916, __s0_916, 1, 0); \
68707		float16x4_t __rev1_916; __rev1_916 = __builtin_shufflevector(__s1_916, __s1_916, 3, 2, 1, 0); \
68708		float16x8_t __rev2_916; __rev2_916 = __builtin_shufflevector(__s2_916, __s2_916, 7, 6, 5, 4, 3, 2, 1, 0); \
68709		__ret_916 = __noswap_vfmlsl_low_f16(__rev0_916, __rev1_916, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_916, __p3_916), __noswap_vgetq_lane_f16(__rev2_916, __p3_916), __noswap_vgetq_lane_f16(__rev2_916, __p3_916), __noswap_vgetq_lane_f16(__rev2_916, __p3_916)}); \
68710		__ret_916 = __builtin_shufflevector(__ret_916, __ret_916, 1, 0); \
68711		__ret_916; \
68712		})
68713		#endif
68714
68715		#endif
68716		#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(__aarch64__)
68717		#ifdef __LITTLE_ENDIAN__
68718		#define vmulh_lane_f16(__p0_917, __p1_917, __p2_917) __extension__ ({ \
68719		float16_t __ret_917; \
68720		float16_t __s0_917 = __p0_917; \
68721		float16x4_t __s1_917 = __p1_917; \
68722		__ret_917 = __s0_917 * vget_lane_f16(__s1_917, __p2_917); \
68723		__ret_917; \
68724		})
68725		#else
68726		#define vmulh_lane_f16(__p0_918, __p1_918, __p2_918) __extension__ ({ \
68727		float16_t __ret_918; \
68728		float16_t __s0_918 = __p0_918; \
68729		float16x4_t __s1_918 = __p1_918; \
68730		float16x4_t __rev1_918; __rev1_918 = __builtin_shufflevector(__s1_918, __s1_918, 3, 2, 1, 0); \
68731		__ret_918 = __s0_918 * __noswap_vget_lane_f16(__rev1_918, __p2_918); \
68732		__ret_918; \
68733		})
68734		#endif
68735
68736		#ifdef __LITTLE_ENDIAN__
68737		#define vmulh_laneq_f16(__p0_919, __p1_919, __p2_919) __extension__ ({ \
68738		float16_t __ret_919; \
68739		float16_t __s0_919 = __p0_919; \
68740		float16x8_t __s1_919 = __p1_919; \
68741		__ret_919 = __s0_919 * vgetq_lane_f16(__s1_919, __p2_919); \
68742		__ret_919; \
68743		})
68744		#else
68745		#define vmulh_laneq_f16(__p0_920, __p1_920, __p2_920) __extension__ ({ \
68746		float16_t __ret_920; \
68747		float16_t __s0_920 = __p0_920; \
68748		float16x8_t __s1_920 = __p1_920; \
68749		float16x8_t __rev1_920; __rev1_920 = __builtin_shufflevector(__s1_920, __s1_920, 7, 6, 5, 4, 3, 2, 1, 0); \
68750		__ret_920 = __s0_920 * __noswap_vgetq_lane_f16(__rev1_920, __p2_920); \
68751		__ret_920; \
68752		})
68753		#endif
68754
68755		#endif
68756		#if defined(__ARM_FEATURE_MATMUL_INT8)
68757		#ifdef __LITTLE_ENDIAN__
68758		#define vsudotq_lane_s32(__p0_921, __p1_921, __p2_921, __p3_921) __extension__ ({ \
68759		int32x4_t __ret_921; \
68760		int32x4_t __s0_921 = __p0_921; \
68761		int8x16_t __s1_921 = __p1_921; \
68762		uint8x8_t __s2_921 = __p2_921; \
68763		uint8x8_t __reint_921 = __s2_921; \
68764		__ret_921 = vusdotq_s32(__s0_921, (uint8x16_t)(splatq_lane_s32((int32x2_t ) &__reint_921, __p3_921)), __s1_921); \
68765		__ret_921; \
	68167	#define vsudotq_lane_s32(__p0_881, __p1_881, __p2_881, __p3_881) __extension__ ({ \
	68168	int32x4_t __ret_881; \
	68169	int32x4_t __s0_881 = __p0_881; \
	68170	int8x16_t __s1_881 = __p1_881; \
	68171	uint8x8_t __s2_881 = __p2_881; \
	68172	uint8x8_t __reint_881 = __s2_881; \
	68173	__ret_881 = vusdotq_s32(__s0_881, (uint8x16_t)(splatq_lane_s32((int32x2_t ) &__reint_881, __p3_881)), __s1_881); \
	68174	__ret_881; \
68766	68175	})
68767	68176	#else
68768		#define vsudotq_lane_s32(__p0_922, __p1_922, __p2_922, __p3_922) __extension__ ({ \
68769		int32x4_t __ret_922; \
68770		int32x4_t __s0_922 = __p0_922; \
68771		int8x16_t __s1_922 = __p1_922; \
68772		uint8x8_t __s2_922 = __p2_922; \
68773		int32x4_t __rev0_922; __rev0_922 = __builtin_shufflevector(__s0_922, __s0_922, 3, 2, 1, 0); \
68774		int8x16_t __rev1_922; __rev1_922 = __builtin_shufflevector(__s1_922, __s1_922, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
68775		uint8x8_t __rev2_922; __rev2_922 = __builtin_shufflevector(__s2_922, __s2_922, 7, 6, 5, 4, 3, 2, 1, 0); \
68776		uint8x8_t __reint_922 = __rev2_922; \
68777		__ret_922 = __noswap_vusdotq_s32(__rev0_922, (uint8x16_t)(__noswap_splatq_lane_s32((int32x2_t ) &__reint_922, __p3_922)), __rev1_922); \
68778		__ret_922 = __builtin_shufflevector(__ret_922, __ret_922, 3, 2, 1, 0); \
68779		__ret_922; \
	68177	#define vsudotq_lane_s32(__p0_882, __p1_882, __p2_882, __p3_882) __extension__ ({ \
	68178	int32x4_t __ret_882; \
	68179	int32x4_t __s0_882 = __p0_882; \
	68180	int8x16_t __s1_882 = __p1_882; \
	68181	uint8x8_t __s2_882 = __p2_882; \
	68182	int32x4_t __rev0_882; __rev0_882 = __builtin_shufflevector(__s0_882, __s0_882, 3, 2, 1, 0); \
	68183	int8x16_t __rev1_882; __rev1_882 = __builtin_shufflevector(__s1_882, __s1_882, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
	68184	uint8x8_t __rev2_882; __rev2_882 = __builtin_shufflevector(__s2_882, __s2_882, 7, 6, 5, 4, 3, 2, 1, 0); \
	68185	uint8x8_t __reint_882 = __rev2_882; \
	68186	__ret_882 = __noswap_vusdotq_s32(__rev0_882, (uint8x16_t)(__noswap_splatq_lane_s32((int32x2_t ) &__reint_882, __p3_882)), __rev1_882); \
	68187	__ret_882 = __builtin_shufflevector(__ret_882, __ret_882, 3, 2, 1, 0); \
	68188	__ret_882; \
68780	68189	})
68781	68190	#endif
68782	68191
68783	68192	#ifdef __LITTLE_ENDIAN__
68784		#define vsudot_lane_s32(__p0_923, __p1_923, __p2_923, __p3_923) __extension__ ({ \
68785		int32x2_t __ret_923; \
68786		int32x2_t __s0_923 = __p0_923; \
68787		int8x8_t __s1_923 = __p1_923; \
68788		uint8x8_t __s2_923 = __p2_923; \
68789		uint8x8_t __reint_923 = __s2_923; \
68790		__ret_923 = vusdot_s32(__s0_923, (uint8x8_t)(splat_lane_s32((int32x2_t ) &__reint_923, __p3_923)), __s1_923); \
68791		__ret_923; \
	68193	#define vsudot_lane_s32(__p0_883, __p1_883, __p2_883, __p3_883) __extension__ ({ \
	68194	int32x2_t __ret_883; \
	68195	int32x2_t __s0_883 = __p0_883; \
	68196	int8x8_t __s1_883 = __p1_883; \
	68197	uint8x8_t __s2_883 = __p2_883; \
	68198	uint8x8_t __reint_883 = __s2_883; \
	68199	__ret_883 = vusdot_s32(__s0_883, (uint8x8_t)(splat_lane_s32((int32x2_t ) &__reint_883, __p3_883)), __s1_883); \
	68200	__ret_883; \
68792	68201	})
68793	68202	#else
68794		#define vsudot_lane_s32(__p0_924, __p1_924, __p2_924, __p3_924) __extension__ ({ \
68795		int32x2_t __ret_924; \
68796		int32x2_t __s0_924 = __p0_924; \
68797		int8x8_t __s1_924 = __p1_924; \
68798		uint8x8_t __s2_924 = __p2_924; \
68799		int32x2_t __rev0_924; __rev0_924 = __builtin_shufflevector(__s0_924, __s0_924, 1, 0); \
68800		int8x8_t __rev1_924; __rev1_924 = __builtin_shufflevector(__s1_924, __s1_924, 7, 6, 5, 4, 3, 2, 1, 0); \
68801		uint8x8_t __rev2_924; __rev2_924 = __builtin_shufflevector(__s2_924, __s2_924, 7, 6, 5, 4, 3, 2, 1, 0); \
68802		uint8x8_t __reint_924 = __rev2_924; \
68803		__ret_924 = __noswap_vusdot_s32(__rev0_924, (uint8x8_t)(__noswap_splat_lane_s32((int32x2_t ) &__reint_924, __p3_924)), __rev1_924); \
68804		__ret_924 = __builtin_shufflevector(__ret_924, __ret_924, 1, 0); \
68805		__ret_924; \
	68203	#define vsudot_lane_s32(__p0_884, __p1_884, __p2_884, __p3_884) __extension__ ({ \
	68204	int32x2_t __ret_884; \
	68205	int32x2_t __s0_884 = __p0_884; \
	68206	int8x8_t __s1_884 = __p1_884; \
	68207	uint8x8_t __s2_884 = __p2_884; \
	68208	int32x2_t __rev0_884; __rev0_884 = __builtin_shufflevector(__s0_884, __s0_884, 1, 0); \
	68209	int8x8_t __rev1_884; __rev1_884 = __builtin_shufflevector(__s1_884, __s1_884, 7, 6, 5, 4, 3, 2, 1, 0); \
	68210	uint8x8_t __rev2_884; __rev2_884 = __builtin_shufflevector(__s2_884, __s2_884, 7, 6, 5, 4, 3, 2, 1, 0); \
	68211	uint8x8_t __reint_884 = __rev2_884; \
	68212	__ret_884 = __noswap_vusdot_s32(__rev0_884, (uint8x8_t)(__noswap_splat_lane_s32((int32x2_t ) &__reint_884, __p3_884)), __rev1_884); \
	68213	__ret_884 = __builtin_shufflevector(__ret_884, __ret_884, 1, 0); \
	68214	__ret_884; \
68806	68215	})
68807	68216	#endif
68808	68217
68809		#endif
68810	68218	#if defined(__aarch64__)
68811	68219	#ifdef __LITTLE_ENDIAN__
68812	68220	__ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
...	...	@@ -69115,136 +68523,136 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
69115	68523	#endif
69116	68524
69117	68525	#ifdef __LITTLE_ENDIAN__
69118		#define vcopyq_lane_p64(__p0_925, __p1_925, __p2_925, __p3_925) __extension__ ({ \
69119		poly64x2_t __ret_925; \
69120		poly64x2_t __s0_925 = __p0_925; \
69121		poly64x1_t __s2_925 = __p2_925; \
69122		__ret_925 = vsetq_lane_p64(vget_lane_p64(__s2_925, __p3_925), __s0_925, __p1_925); \
69123		__ret_925; \
	68526	#define vcopyq_lane_p64(__p0_885, __p1_885, __p2_885, __p3_885) __extension__ ({ \
	68527	poly64x2_t __ret_885; \
	68528	poly64x2_t __s0_885 = __p0_885; \
	68529	poly64x1_t __s2_885 = __p2_885; \
	68530	__ret_885 = vsetq_lane_p64(vget_lane_p64(__s2_885, __p3_885), __s0_885, __p1_885); \
	68531	__ret_885; \
69124	68532	})
69125	68533	#else
69126		#define vcopyq_lane_p64(__p0_926, __p1_926, __p2_926, __p3_926) __extension__ ({ \
69127		poly64x2_t __ret_926; \
69128		poly64x2_t __s0_926 = __p0_926; \
69129		poly64x1_t __s2_926 = __p2_926; \
69130		poly64x2_t __rev0_926; __rev0_926 = __builtin_shufflevector(__s0_926, __s0_926, 1, 0); \
69131		__ret_926 = __noswap_vsetq_lane_p64(vget_lane_p64(__s2_926, __p3_926), __rev0_926, __p1_926); \
69132		__ret_926 = __builtin_shufflevector(__ret_926, __ret_926, 1, 0); \
69133		__ret_926; \
	68534	#define vcopyq_lane_p64(__p0_886, __p1_886, __p2_886, __p3_886) __extension__ ({ \
	68535	poly64x2_t __ret_886; \
	68536	poly64x2_t __s0_886 = __p0_886; \
	68537	poly64x1_t __s2_886 = __p2_886; \
	68538	poly64x2_t __rev0_886; __rev0_886 = __builtin_shufflevector(__s0_886, __s0_886, 1, 0); \
	68539	__ret_886 = __noswap_vsetq_lane_p64(vget_lane_p64(__s2_886, __p3_886), __rev0_886, __p1_886); \
	68540	__ret_886 = __builtin_shufflevector(__ret_886, __ret_886, 1, 0); \
	68541	__ret_886; \
69134	68542	})
69135	68543	#endif
69136	68544
69137	68545	#ifdef __LITTLE_ENDIAN__
69138		#define vcopyq_lane_f64(__p0_927, __p1_927, __p2_927, __p3_927) __extension__ ({ \
69139		float64x2_t __ret_927; \
69140		float64x2_t __s0_927 = __p0_927; \
69141		float64x1_t __s2_927 = __p2_927; \
69142		__ret_927 = vsetq_lane_f64(vget_lane_f64(__s2_927, __p3_927), __s0_927, __p1_927); \
69143		__ret_927; \
	68546	#define vcopyq_lane_f64(__p0_887, __p1_887, __p2_887, __p3_887) __extension__ ({ \
	68547	float64x2_t __ret_887; \
	68548	float64x2_t __s0_887 = __p0_887; \
	68549	float64x1_t __s2_887 = __p2_887; \
	68550	__ret_887 = vsetq_lane_f64(vget_lane_f64(__s2_887, __p3_887), __s0_887, __p1_887); \
	68551	__ret_887; \
69144	68552	})
69145	68553	#else
69146		#define vcopyq_lane_f64(__p0_928, __p1_928, __p2_928, __p3_928) __extension__ ({ \
69147		float64x2_t __ret_928; \
69148		float64x2_t __s0_928 = __p0_928; \
69149		float64x1_t __s2_928 = __p2_928; \
69150		float64x2_t __rev0_928; __rev0_928 = __builtin_shufflevector(__s0_928, __s0_928, 1, 0); \
69151		__ret_928 = __noswap_vsetq_lane_f64(vget_lane_f64(__s2_928, __p3_928), __rev0_928, __p1_928); \
69152		__ret_928 = __builtin_shufflevector(__ret_928, __ret_928, 1, 0); \
69153		__ret_928; \
	68554	#define vcopyq_lane_f64(__p0_888, __p1_888, __p2_888, __p3_888) __extension__ ({ \
	68555	float64x2_t __ret_888; \
	68556	float64x2_t __s0_888 = __p0_888; \
	68557	float64x1_t __s2_888 = __p2_888; \
	68558	float64x2_t __rev0_888; __rev0_888 = __builtin_shufflevector(__s0_888, __s0_888, 1, 0); \
	68559	__ret_888 = __noswap_vsetq_lane_f64(vget_lane_f64(__s2_888, __p3_888), __rev0_888, __p1_888); \
	68560	__ret_888 = __builtin_shufflevector(__ret_888, __ret_888, 1, 0); \
	68561	__ret_888; \
69154	68562	})
69155	68563	#endif
69156	68564
69157		#define vcopy_lane_p64(__p0_929, __p1_929, __p2_929, __p3_929) __extension__ ({ \
69158		poly64x1_t __ret_929; \
69159		poly64x1_t __s0_929 = __p0_929; \
69160		poly64x1_t __s2_929 = __p2_929; \
69161		__ret_929 = vset_lane_p64(vget_lane_p64(__s2_929, __p3_929), __s0_929, __p1_929); \
69162		__ret_929; \
	68565	#define vcopy_lane_p64(__p0_889, __p1_889, __p2_889, __p3_889) __extension__ ({ \
	68566	poly64x1_t __ret_889; \
	68567	poly64x1_t __s0_889 = __p0_889; \
	68568	poly64x1_t __s2_889 = __p2_889; \
	68569	__ret_889 = vset_lane_p64(vget_lane_p64(__s2_889, __p3_889), __s0_889, __p1_889); \
	68570	__ret_889; \
69163	68571	})
69164		#define vcopy_lane_f64(__p0_930, __p1_930, __p2_930, __p3_930) __extension__ ({ \
69165		float64x1_t __ret_930; \
69166		float64x1_t __s0_930 = __p0_930; \
69167		float64x1_t __s2_930 = __p2_930; \
69168		__ret_930 = vset_lane_f64(vget_lane_f64(__s2_930, __p3_930), __s0_930, __p1_930); \
69169		__ret_930; \
	68572	#define vcopy_lane_f64(__p0_890, __p1_890, __p2_890, __p3_890) __extension__ ({ \
	68573	float64x1_t __ret_890; \
	68574	float64x1_t __s0_890 = __p0_890; \
	68575	float64x1_t __s2_890 = __p2_890; \
	68576	__ret_890 = vset_lane_f64(vget_lane_f64(__s2_890, __p3_890), __s0_890, __p1_890); \
	68577	__ret_890; \
69170	68578	})
69171	68579	#ifdef __LITTLE_ENDIAN__
69172		#define vcopyq_laneq_p64(__p0_931, __p1_931, __p2_931, __p3_931) __extension__ ({ \
69173		poly64x2_t __ret_931; \
69174		poly64x2_t __s0_931 = __p0_931; \
69175		poly64x2_t __s2_931 = __p2_931; \
69176		__ret_931 = vsetq_lane_p64(vgetq_lane_p64(__s2_931, __p3_931), __s0_931, __p1_931); \
69177		__ret_931; \
	68580	#define vcopyq_laneq_p64(__p0_891, __p1_891, __p2_891, __p3_891) __extension__ ({ \
	68581	poly64x2_t __ret_891; \
	68582	poly64x2_t __s0_891 = __p0_891; \
	68583	poly64x2_t __s2_891 = __p2_891; \
	68584	__ret_891 = vsetq_lane_p64(vgetq_lane_p64(__s2_891, __p3_891), __s0_891, __p1_891); \
	68585	__ret_891; \
69178	68586	})
69179	68587	#else
69180		#define vcopyq_laneq_p64(__p0_932, __p1_932, __p2_932, __p3_932) __extension__ ({ \
69181		poly64x2_t __ret_932; \
69182		poly64x2_t __s0_932 = __p0_932; \
69183		poly64x2_t __s2_932 = __p2_932; \
69184		poly64x2_t __rev0_932; __rev0_932 = __builtin_shufflevector(__s0_932, __s0_932, 1, 0); \
69185		poly64x2_t __rev2_932; __rev2_932 = __builtin_shufflevector(__s2_932, __s2_932, 1, 0); \
69186		__ret_932 = __noswap_vsetq_lane_p64(__noswap_vgetq_lane_p64(__rev2_932, __p3_932), __rev0_932, __p1_932); \
69187		__ret_932 = __builtin_shufflevector(__ret_932, __ret_932, 1, 0); \
69188		__ret_932; \
	68588	#define vcopyq_laneq_p64(__p0_892, __p1_892, __p2_892, __p3_892) __extension__ ({ \
	68589	poly64x2_t __ret_892; \
	68590	poly64x2_t __s0_892 = __p0_892; \
	68591	poly64x2_t __s2_892 = __p2_892; \
	68592	poly64x2_t __rev0_892; __rev0_892 = __builtin_shufflevector(__s0_892, __s0_892, 1, 0); \
	68593	poly64x2_t __rev2_892; __rev2_892 = __builtin_shufflevector(__s2_892, __s2_892, 1, 0); \
	68594	__ret_892 = __noswap_vsetq_lane_p64(__noswap_vgetq_lane_p64(__rev2_892, __p3_892), __rev0_892, __p1_892); \
	68595	__ret_892 = __builtin_shufflevector(__ret_892, __ret_892, 1, 0); \
	68596	__ret_892; \
69189	68597	})
69190	68598	#endif
69191	68599
69192	68600	#ifdef __LITTLE_ENDIAN__
69193		#define vcopyq_laneq_f64(__p0_933, __p1_933, __p2_933, __p3_933) __extension__ ({ \
69194		float64x2_t __ret_933; \
69195		float64x2_t __s0_933 = __p0_933; \
69196		float64x2_t __s2_933 = __p2_933; \
69197		__ret_933 = vsetq_lane_f64(vgetq_lane_f64(__s2_933, __p3_933), __s0_933, __p1_933); \
69198		__ret_933; \
	68601	#define vcopyq_laneq_f64(__p0_893, __p1_893, __p2_893, __p3_893) __extension__ ({ \
	68602	float64x2_t __ret_893; \
	68603	float64x2_t __s0_893 = __p0_893; \
	68604	float64x2_t __s2_893 = __p2_893; \
	68605	__ret_893 = vsetq_lane_f64(vgetq_lane_f64(__s2_893, __p3_893), __s0_893, __p1_893); \
	68606	__ret_893; \
69199	68607	})
69200	68608	#else
69201		#define vcopyq_laneq_f64(__p0_934, __p1_934, __p2_934, __p3_934) __extension__ ({ \
69202		float64x2_t __ret_934; \
69203		float64x2_t __s0_934 = __p0_934; \
69204		float64x2_t __s2_934 = __p2_934; \
69205		float64x2_t __rev0_934; __rev0_934 = __builtin_shufflevector(__s0_934, __s0_934, 1, 0); \
69206		float64x2_t __rev2_934; __rev2_934 = __builtin_shufflevector(__s2_934, __s2_934, 1, 0); \
69207		__ret_934 = __noswap_vsetq_lane_f64(__noswap_vgetq_lane_f64(__rev2_934, __p3_934), __rev0_934, __p1_934); \
69208		__ret_934 = __builtin_shufflevector(__ret_934, __ret_934, 1, 0); \
69209		__ret_934; \
	68609	#define vcopyq_laneq_f64(__p0_894, __p1_894, __p2_894, __p3_894) __extension__ ({ \
	68610	float64x2_t __ret_894; \
	68611	float64x2_t __s0_894 = __p0_894; \
	68612	float64x2_t __s2_894 = __p2_894; \
	68613	float64x2_t __rev0_894; __rev0_894 = __builtin_shufflevector(__s0_894, __s0_894, 1, 0); \
	68614	float64x2_t __rev2_894; __rev2_894 = __builtin_shufflevector(__s2_894, __s2_894, 1, 0); \
	68615	__ret_894 = __noswap_vsetq_lane_f64(__noswap_vgetq_lane_f64(__rev2_894, __p3_894), __rev0_894, __p1_894); \
	68616	__ret_894 = __builtin_shufflevector(__ret_894, __ret_894, 1, 0); \
	68617	__ret_894; \
69210	68618	})
69211	68619	#endif
69212	68620
69213	68621	#ifdef __LITTLE_ENDIAN__
69214		#define vcopy_laneq_p64(__p0_935, __p1_935, __p2_935, __p3_935) __extension__ ({ \
69215		poly64x1_t __ret_935; \
69216		poly64x1_t __s0_935 = __p0_935; \
69217		poly64x2_t __s2_935 = __p2_935; \
69218		__ret_935 = vset_lane_p64(vgetq_lane_p64(__s2_935, __p3_935), __s0_935, __p1_935); \
69219		__ret_935; \
	68622	#define vcopy_laneq_p64(__p0_895, __p1_895, __p2_895, __p3_895) __extension__ ({ \
	68623	poly64x1_t __ret_895; \
	68624	poly64x1_t __s0_895 = __p0_895; \
	68625	poly64x2_t __s2_895 = __p2_895; \
	68626	__ret_895 = vset_lane_p64(vgetq_lane_p64(__s2_895, __p3_895), __s0_895, __p1_895); \
	68627	__ret_895; \
69220	68628	})
69221	68629	#else
69222		#define vcopy_laneq_p64(__p0_936, __p1_936, __p2_936, __p3_936) __extension__ ({ \
69223		poly64x1_t __ret_936; \
69224		poly64x1_t __s0_936 = __p0_936; \
69225		poly64x2_t __s2_936 = __p2_936; \
69226		poly64x2_t __rev2_936; __rev2_936 = __builtin_shufflevector(__s2_936, __s2_936, 1, 0); \
69227		__ret_936 = vset_lane_p64(__noswap_vgetq_lane_p64(__rev2_936, __p3_936), __s0_936, __p1_936); \
69228		__ret_936; \
	68630	#define vcopy_laneq_p64(__p0_896, __p1_896, __p2_896, __p3_896) __extension__ ({ \
	68631	poly64x1_t __ret_896; \
	68632	poly64x1_t __s0_896 = __p0_896; \
	68633	poly64x2_t __s2_896 = __p2_896; \
	68634	poly64x2_t __rev2_896; __rev2_896 = __builtin_shufflevector(__s2_896, __s2_896, 1, 0); \
	68635	__ret_896 = vset_lane_p64(__noswap_vgetq_lane_p64(__rev2_896, __p3_896), __s0_896, __p1_896); \
	68636	__ret_896; \
69229	68637	})
69230	68638	#endif
69231	68639
69232	68640	#ifdef __LITTLE_ENDIAN__
69233		#define vcopy_laneq_f64(__p0_937, __p1_937, __p2_937, __p3_937) __extension__ ({ \
69234		float64x1_t __ret_937; \
69235		float64x1_t __s0_937 = __p0_937; \
69236		float64x2_t __s2_937 = __p2_937; \
69237		__ret_937 = vset_lane_f64(vgetq_lane_f64(__s2_937, __p3_937), __s0_937, __p1_937); \
69238		__ret_937; \
	68641	#define vcopy_laneq_f64(__p0_897, __p1_897, __p2_897, __p3_897) __extension__ ({ \
	68642	float64x1_t __ret_897; \
	68643	float64x1_t __s0_897 = __p0_897; \
	68644	float64x2_t __s2_897 = __p2_897; \
	68645	__ret_897 = vset_lane_f64(vgetq_lane_f64(__s2_897, __p3_897), __s0_897, __p1_897); \
	68646	__ret_897; \
69239	68647	})
69240	68648	#else
69241		#define vcopy_laneq_f64(__p0_938, __p1_938, __p2_938, __p3_938) __extension__ ({ \
69242		float64x1_t __ret_938; \
69243		float64x1_t __s0_938 = __p0_938; \
69244		float64x2_t __s2_938 = __p2_938; \
69245		float64x2_t __rev2_938; __rev2_938 = __builtin_shufflevector(__s2_938, __s2_938, 1, 0); \
69246		__ret_938 = vset_lane_f64(__noswap_vgetq_lane_f64(__rev2_938, __p3_938), __s0_938, __p1_938); \
69247		__ret_938; \
	68649	#define vcopy_laneq_f64(__p0_898, __p1_898, __p2_898, __p3_898) __extension__ ({ \
	68650	float64x1_t __ret_898; \
	68651	float64x1_t __s0_898 = __p0_898; \
	68652	float64x2_t __s2_898 = __p2_898; \
	68653	float64x2_t __rev2_898; __rev2_898 = __builtin_shufflevector(__s2_898, __s2_898, 1, 0); \
	68654	__ret_898 = vset_lane_f64(__noswap_vgetq_lane_f64(__rev2_898, __p3_898), __s0_898, __p1_898); \
	68655	__ret_898; \
69248	68656	})
69249	68657	#endif
69250	68658
...	...	@@ -69600,38 +69008,460 @@ __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
69600	69008	}
69601	69009	#endif
69602	69010
69603		#define vmulx_lane_f64(__p0_939, __p1_939, __p2_939) __extension__ ({ \
69604		float64x1_t __ret_939; \
69605		float64x1_t __s0_939 = __p0_939; \
69606		float64x1_t __s1_939 = __p1_939; \
69607		float64_t __x_939 = vget_lane_f64(__s0_939, 0); \
69608		float64_t __y_939 = vget_lane_f64(__s1_939, __p2_939); \
69609		float64_t __z_939 = vmulxd_f64(__x_939, __y_939); \
69610		__ret_939 = vset_lane_f64(__z_939, __s0_939, __p2_939); \
69611		__ret_939; \
	69011	#define vmulx_lane_f64(__p0_899, __p1_899, __p2_899) __extension__ ({ \
	69012	float64x1_t __ret_899; \
	69013	float64x1_t __s0_899 = __p0_899; \
	69014	float64x1_t __s1_899 = __p1_899; \
	69015	float64_t __x_899 = vget_lane_f64(__s0_899, 0); \
	69016	float64_t __y_899 = vget_lane_f64(__s1_899, __p2_899); \
	69017	float64_t __z_899 = vmulxd_f64(__x_899, __y_899); \
	69018	__ret_899 = vset_lane_f64(__z_899, __s0_899, __p2_899); \
	69019	__ret_899; \
	69020	})
	69021	#ifdef __LITTLE_ENDIAN__
	69022	#define vmulx_laneq_f64(__p0_900, __p1_900, __p2_900) __extension__ ({ \
	69023	float64x1_t __ret_900; \
	69024	float64x1_t __s0_900 = __p0_900; \
	69025	float64x2_t __s1_900 = __p1_900; \
	69026	float64_t __x_900 = vget_lane_f64(__s0_900, 0); \
	69027	float64_t __y_900 = vgetq_lane_f64(__s1_900, __p2_900); \
	69028	float64_t __z_900 = vmulxd_f64(__x_900, __y_900); \
	69029	__ret_900 = vset_lane_f64(__z_900, __s0_900, 0); \
	69030	__ret_900; \
	69031	})
	69032	#else
	69033	#define vmulx_laneq_f64(__p0_901, __p1_901, __p2_901) __extension__ ({ \
	69034	float64x1_t __ret_901; \
	69035	float64x1_t __s0_901 = __p0_901; \
	69036	float64x2_t __s1_901 = __p1_901; \
	69037	float64x2_t __rev1_901; __rev1_901 = __builtin_shufflevector(__s1_901, __s1_901, 1, 0); \
	69038	float64_t __x_901 = vget_lane_f64(__s0_901, 0); \
	69039	float64_t __y_901 = __noswap_vgetq_lane_f64(__rev1_901, __p2_901); \
	69040	float64_t __z_901 = vmulxd_f64(__x_901, __y_901); \
	69041	__ret_901 = vset_lane_f64(__z_901, __s0_901, 0); \
	69042	__ret_901; \
	69043	})
	69044	#endif
	69045
	69046	#ifdef __LITTLE_ENDIAN__
	69047	#define vfmlalq_lane_high_f16(__p0_902, __p1_902, __p2_902, __p3_902) __extension__ ({ \
	69048	float32x4_t __ret_902; \
	69049	float32x4_t __s0_902 = __p0_902; \
	69050	float16x8_t __s1_902 = __p1_902; \
	69051	float16x4_t __s2_902 = __p2_902; \
	69052	__ret_902 = vfmlalq_high_f16(__s0_902, __s1_902, (float16x8_t) {vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902), vget_lane_f16(__s2_902, __p3_902)}); \
	69053	__ret_902; \
	69054	})
	69055	#else
	69056	#define vfmlalq_lane_high_f16(__p0_903, __p1_903, __p2_903, __p3_903) __extension__ ({ \
	69057	float32x4_t __ret_903; \
	69058	float32x4_t __s0_903 = __p0_903; \
	69059	float16x8_t __s1_903 = __p1_903; \
	69060	float16x4_t __s2_903 = __p2_903; \
	69061	float32x4_t __rev0_903; __rev0_903 = __builtin_shufflevector(__s0_903, __s0_903, 3, 2, 1, 0); \
	69062	float16x8_t __rev1_903; __rev1_903 = __builtin_shufflevector(__s1_903, __s1_903, 7, 6, 5, 4, 3, 2, 1, 0); \
	69063	float16x4_t __rev2_903; __rev2_903 = __builtin_shufflevector(__s2_903, __s2_903, 3, 2, 1, 0); \
	69064	__ret_903 = __noswap_vfmlalq_high_f16(__rev0_903, __rev1_903, (float16x8_t) {__noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903), __noswap_vget_lane_f16(__rev2_903, __p3_903)}); \
	69065	__ret_903 = __builtin_shufflevector(__ret_903, __ret_903, 3, 2, 1, 0); \
	69066	__ret_903; \
	69067	})
	69068	#endif
	69069
	69070	#ifdef __LITTLE_ENDIAN__
	69071	#define vfmlal_lane_high_f16(__p0_904, __p1_904, __p2_904, __p3_904) __extension__ ({ \
	69072	float32x2_t __ret_904; \
	69073	float32x2_t __s0_904 = __p0_904; \
	69074	float16x4_t __s1_904 = __p1_904; \
	69075	float16x4_t __s2_904 = __p2_904; \
	69076	__ret_904 = vfmlal_high_f16(__s0_904, __s1_904, (float16x4_t) {vget_lane_f16(__s2_904, __p3_904), vget_lane_f16(__s2_904, __p3_904), vget_lane_f16(__s2_904, __p3_904), vget_lane_f16(__s2_904, __p3_904)}); \
	69077	__ret_904; \
	69078	})
	69079	#else
	69080	#define vfmlal_lane_high_f16(__p0_905, __p1_905, __p2_905, __p3_905) __extension__ ({ \
	69081	float32x2_t __ret_905; \
	69082	float32x2_t __s0_905 = __p0_905; \
	69083	float16x4_t __s1_905 = __p1_905; \
	69084	float16x4_t __s2_905 = __p2_905; \
	69085	float32x2_t __rev0_905; __rev0_905 = __builtin_shufflevector(__s0_905, __s0_905, 1, 0); \
	69086	float16x4_t __rev1_905; __rev1_905 = __builtin_shufflevector(__s1_905, __s1_905, 3, 2, 1, 0); \
	69087	float16x4_t __rev2_905; __rev2_905 = __builtin_shufflevector(__s2_905, __s2_905, 3, 2, 1, 0); \
	69088	__ret_905 = __noswap_vfmlal_high_f16(__rev0_905, __rev1_905, (float16x4_t) {__noswap_vget_lane_f16(__rev2_905, __p3_905), __noswap_vget_lane_f16(__rev2_905, __p3_905), __noswap_vget_lane_f16(__rev2_905, __p3_905), __noswap_vget_lane_f16(__rev2_905, __p3_905)}); \
	69089	__ret_905 = __builtin_shufflevector(__ret_905, __ret_905, 1, 0); \
	69090	__ret_905; \
	69091	})
	69092	#endif
	69093
	69094	#ifdef __LITTLE_ENDIAN__
	69095	#define vfmlalq_lane_low_f16(__p0_906, __p1_906, __p2_906, __p3_906) __extension__ ({ \
	69096	float32x4_t __ret_906; \
	69097	float32x4_t __s0_906 = __p0_906; \
	69098	float16x8_t __s1_906 = __p1_906; \
	69099	float16x4_t __s2_906 = __p2_906; \
	69100	__ret_906 = vfmlalq_low_f16(__s0_906, __s1_906, (float16x8_t) {vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906), vget_lane_f16(__s2_906, __p3_906)}); \
	69101	__ret_906; \
	69102	})
	69103	#else
	69104	#define vfmlalq_lane_low_f16(__p0_907, __p1_907, __p2_907, __p3_907) __extension__ ({ \
	69105	float32x4_t __ret_907; \
	69106	float32x4_t __s0_907 = __p0_907; \
	69107	float16x8_t __s1_907 = __p1_907; \
	69108	float16x4_t __s2_907 = __p2_907; \
	69109	float32x4_t __rev0_907; __rev0_907 = __builtin_shufflevector(__s0_907, __s0_907, 3, 2, 1, 0); \
	69110	float16x8_t __rev1_907; __rev1_907 = __builtin_shufflevector(__s1_907, __s1_907, 7, 6, 5, 4, 3, 2, 1, 0); \
	69111	float16x4_t __rev2_907; __rev2_907 = __builtin_shufflevector(__s2_907, __s2_907, 3, 2, 1, 0); \
	69112	__ret_907 = __noswap_vfmlalq_low_f16(__rev0_907, __rev1_907, (float16x8_t) {__noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907), __noswap_vget_lane_f16(__rev2_907, __p3_907)}); \
	69113	__ret_907 = __builtin_shufflevector(__ret_907, __ret_907, 3, 2, 1, 0); \
	69114	__ret_907; \
	69115	})
	69116	#endif
	69117
	69118	#ifdef __LITTLE_ENDIAN__
	69119	#define vfmlal_lane_low_f16(__p0_908, __p1_908, __p2_908, __p3_908) __extension__ ({ \
	69120	float32x2_t __ret_908; \
	69121	float32x2_t __s0_908 = __p0_908; \
	69122	float16x4_t __s1_908 = __p1_908; \
	69123	float16x4_t __s2_908 = __p2_908; \
	69124	__ret_908 = vfmlal_low_f16(__s0_908, __s1_908, (float16x4_t) {vget_lane_f16(__s2_908, __p3_908), vget_lane_f16(__s2_908, __p3_908), vget_lane_f16(__s2_908, __p3_908), vget_lane_f16(__s2_908, __p3_908)}); \
	69125	__ret_908; \
	69126	})
	69127	#else
	69128	#define vfmlal_lane_low_f16(__p0_909, __p1_909, __p2_909, __p3_909) __extension__ ({ \
	69129	float32x2_t __ret_909; \
	69130	float32x2_t __s0_909 = __p0_909; \
	69131	float16x4_t __s1_909 = __p1_909; \
	69132	float16x4_t __s2_909 = __p2_909; \
	69133	float32x2_t __rev0_909; __rev0_909 = __builtin_shufflevector(__s0_909, __s0_909, 1, 0); \
	69134	float16x4_t __rev1_909; __rev1_909 = __builtin_shufflevector(__s1_909, __s1_909, 3, 2, 1, 0); \
	69135	float16x4_t __rev2_909; __rev2_909 = __builtin_shufflevector(__s2_909, __s2_909, 3, 2, 1, 0); \
	69136	__ret_909 = __noswap_vfmlal_low_f16(__rev0_909, __rev1_909, (float16x4_t) {__noswap_vget_lane_f16(__rev2_909, __p3_909), __noswap_vget_lane_f16(__rev2_909, __p3_909), __noswap_vget_lane_f16(__rev2_909, __p3_909), __noswap_vget_lane_f16(__rev2_909, __p3_909)}); \
	69137	__ret_909 = __builtin_shufflevector(__ret_909, __ret_909, 1, 0); \
	69138	__ret_909; \
	69139	})
	69140	#endif
	69141
	69142	#ifdef __LITTLE_ENDIAN__
	69143	#define vfmlalq_laneq_high_f16(__p0_910, __p1_910, __p2_910, __p3_910) __extension__ ({ \
	69144	float32x4_t __ret_910; \
	69145	float32x4_t __s0_910 = __p0_910; \
	69146	float16x8_t __s1_910 = __p1_910; \
	69147	float16x8_t __s2_910 = __p2_910; \
	69148	__ret_910 = vfmlalq_high_f16(__s0_910, __s1_910, (float16x8_t) {vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910), vgetq_lane_f16(__s2_910, __p3_910)}); \
	69149	__ret_910; \
	69150	})
	69151	#else
	69152	#define vfmlalq_laneq_high_f16(__p0_911, __p1_911, __p2_911, __p3_911) __extension__ ({ \
	69153	float32x4_t __ret_911; \
	69154	float32x4_t __s0_911 = __p0_911; \
	69155	float16x8_t __s1_911 = __p1_911; \
	69156	float16x8_t __s2_911 = __p2_911; \
	69157	float32x4_t __rev0_911; __rev0_911 = __builtin_shufflevector(__s0_911, __s0_911, 3, 2, 1, 0); \
	69158	float16x8_t __rev1_911; __rev1_911 = __builtin_shufflevector(__s1_911, __s1_911, 7, 6, 5, 4, 3, 2, 1, 0); \
	69159	float16x8_t __rev2_911; __rev2_911 = __builtin_shufflevector(__s2_911, __s2_911, 7, 6, 5, 4, 3, 2, 1, 0); \
	69160	__ret_911 = __noswap_vfmlalq_high_f16(__rev0_911, __rev1_911, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911), __noswap_vgetq_lane_f16(__rev2_911, __p3_911)}); \
	69161	__ret_911 = __builtin_shufflevector(__ret_911, __ret_911, 3, 2, 1, 0); \
	69162	__ret_911; \
69612	69163	})
	69164	#endif
	69165
69613	69166	#ifdef __LITTLE_ENDIAN__
69614		#define vmulx_laneq_f64(__p0_940, __p1_940, __p2_940) __extension__ ({ \
69615		float64x1_t __ret_940; \
69616		float64x1_t __s0_940 = __p0_940; \
69617		float64x2_t __s1_940 = __p1_940; \
69618		float64_t __x_940 = vget_lane_f64(__s0_940, 0); \
69619		float64_t __y_940 = vgetq_lane_f64(__s1_940, __p2_940); \
69620		float64_t __z_940 = vmulxd_f64(__x_940, __y_940); \
69621		__ret_940 = vset_lane_f64(__z_940, __s0_940, 0); \
69622		__ret_940; \
	69167	#define vfmlal_laneq_high_f16(__p0_912, __p1_912, __p2_912, __p3_912) __extension__ ({ \
	69168	float32x2_t __ret_912; \
	69169	float32x2_t __s0_912 = __p0_912; \
	69170	float16x4_t __s1_912 = __p1_912; \
	69171	float16x8_t __s2_912 = __p2_912; \
	69172	__ret_912 = vfmlal_high_f16(__s0_912, __s1_912, (float16x4_t) {vgetq_lane_f16(__s2_912, __p3_912), vgetq_lane_f16(__s2_912, __p3_912), vgetq_lane_f16(__s2_912, __p3_912), vgetq_lane_f16(__s2_912, __p3_912)}); \
	69173	__ret_912; \
69623	69174	})
69624	69175	#else
69625		#define vmulx_laneq_f64(__p0_941, __p1_941, __p2_941) __extension__ ({ \
69626		float64x1_t __ret_941; \
69627		float64x1_t __s0_941 = __p0_941; \
69628		float64x2_t __s1_941 = __p1_941; \
69629		float64x2_t __rev1_941; __rev1_941 = __builtin_shufflevector(__s1_941, __s1_941, 1, 0); \
69630		float64_t __x_941 = vget_lane_f64(__s0_941, 0); \
69631		float64_t __y_941 = __noswap_vgetq_lane_f64(__rev1_941, __p2_941); \
69632		float64_t __z_941 = vmulxd_f64(__x_941, __y_941); \
69633		__ret_941 = vset_lane_f64(__z_941, __s0_941, 0); \
69634		__ret_941; \
	69176	#define vfmlal_laneq_high_f16(__p0_913, __p1_913, __p2_913, __p3_913) __extension__ ({ \
	69177	float32x2_t __ret_913; \
	69178	float32x2_t __s0_913 = __p0_913; \
	69179	float16x4_t __s1_913 = __p1_913; \
	69180	float16x8_t __s2_913 = __p2_913; \
	69181	float32x2_t __rev0_913; __rev0_913 = __builtin_shufflevector(__s0_913, __s0_913, 1, 0); \
	69182	float16x4_t __rev1_913; __rev1_913 = __builtin_shufflevector(__s1_913, __s1_913, 3, 2, 1, 0); \
	69183	float16x8_t __rev2_913; __rev2_913 = __builtin_shufflevector(__s2_913, __s2_913, 7, 6, 5, 4, 3, 2, 1, 0); \
	69184	__ret_913 = __noswap_vfmlal_high_f16(__rev0_913, __rev1_913, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_913, __p3_913), __noswap_vgetq_lane_f16(__rev2_913, __p3_913), __noswap_vgetq_lane_f16(__rev2_913, __p3_913), __noswap_vgetq_lane_f16(__rev2_913, __p3_913)}); \
	69185	__ret_913 = __builtin_shufflevector(__ret_913, __ret_913, 1, 0); \
	69186	__ret_913; \
	69187	})
	69188	#endif
	69189
	69190	#ifdef __LITTLE_ENDIAN__
	69191	#define vfmlalq_laneq_low_f16(__p0_914, __p1_914, __p2_914, __p3_914) __extension__ ({ \
	69192	float32x4_t __ret_914; \
	69193	float32x4_t __s0_914 = __p0_914; \
	69194	float16x8_t __s1_914 = __p1_914; \
	69195	float16x8_t __s2_914 = __p2_914; \
	69196	__ret_914 = vfmlalq_low_f16(__s0_914, __s1_914, (float16x8_t) {vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914), vgetq_lane_f16(__s2_914, __p3_914)}); \
	69197	__ret_914; \
	69198	})
	69199	#else
	69200	#define vfmlalq_laneq_low_f16(__p0_915, __p1_915, __p2_915, __p3_915) __extension__ ({ \
	69201	float32x4_t __ret_915; \
	69202	float32x4_t __s0_915 = __p0_915; \
	69203	float16x8_t __s1_915 = __p1_915; \
	69204	float16x8_t __s2_915 = __p2_915; \
	69205	float32x4_t __rev0_915; __rev0_915 = __builtin_shufflevector(__s0_915, __s0_915, 3, 2, 1, 0); \
	69206	float16x8_t __rev1_915; __rev1_915 = __builtin_shufflevector(__s1_915, __s1_915, 7, 6, 5, 4, 3, 2, 1, 0); \
	69207	float16x8_t __rev2_915; __rev2_915 = __builtin_shufflevector(__s2_915, __s2_915, 7, 6, 5, 4, 3, 2, 1, 0); \
	69208	__ret_915 = __noswap_vfmlalq_low_f16(__rev0_915, __rev1_915, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915), __noswap_vgetq_lane_f16(__rev2_915, __p3_915)}); \
	69209	__ret_915 = __builtin_shufflevector(__ret_915, __ret_915, 3, 2, 1, 0); \
	69210	__ret_915; \
	69211	})
	69212	#endif
	69213
	69214	#ifdef __LITTLE_ENDIAN__
	69215	#define vfmlal_laneq_low_f16(__p0_916, __p1_916, __p2_916, __p3_916) __extension__ ({ \
	69216	float32x2_t __ret_916; \
	69217	float32x2_t __s0_916 = __p0_916; \
	69218	float16x4_t __s1_916 = __p1_916; \
	69219	float16x8_t __s2_916 = __p2_916; \
	69220	__ret_916 = vfmlal_low_f16(__s0_916, __s1_916, (float16x4_t) {vgetq_lane_f16(__s2_916, __p3_916), vgetq_lane_f16(__s2_916, __p3_916), vgetq_lane_f16(__s2_916, __p3_916), vgetq_lane_f16(__s2_916, __p3_916)}); \
	69221	__ret_916; \
	69222	})
	69223	#else
	69224	#define vfmlal_laneq_low_f16(__p0_917, __p1_917, __p2_917, __p3_917) __extension__ ({ \
	69225	float32x2_t __ret_917; \
	69226	float32x2_t __s0_917 = __p0_917; \
	69227	float16x4_t __s1_917 = __p1_917; \
	69228	float16x8_t __s2_917 = __p2_917; \
	69229	float32x2_t __rev0_917; __rev0_917 = __builtin_shufflevector(__s0_917, __s0_917, 1, 0); \
	69230	float16x4_t __rev1_917; __rev1_917 = __builtin_shufflevector(__s1_917, __s1_917, 3, 2, 1, 0); \
	69231	float16x8_t __rev2_917; __rev2_917 = __builtin_shufflevector(__s2_917, __s2_917, 7, 6, 5, 4, 3, 2, 1, 0); \
	69232	__ret_917 = __noswap_vfmlal_low_f16(__rev0_917, __rev1_917, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_917, __p3_917), __noswap_vgetq_lane_f16(__rev2_917, __p3_917), __noswap_vgetq_lane_f16(__rev2_917, __p3_917), __noswap_vgetq_lane_f16(__rev2_917, __p3_917)}); \
	69233	__ret_917 = __builtin_shufflevector(__ret_917, __ret_917, 1, 0); \
	69234	__ret_917; \
	69235	})
	69236	#endif
	69237
	69238	#ifdef __LITTLE_ENDIAN__
	69239	#define vfmlslq_lane_high_f16(__p0_918, __p1_918, __p2_918, __p3_918) __extension__ ({ \
	69240	float32x4_t __ret_918; \
	69241	float32x4_t __s0_918 = __p0_918; \
	69242	float16x8_t __s1_918 = __p1_918; \
	69243	float16x4_t __s2_918 = __p2_918; \
	69244	__ret_918 = vfmlslq_high_f16(__s0_918, __s1_918, (float16x8_t) {vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918), vget_lane_f16(__s2_918, __p3_918)}); \
	69245	__ret_918; \
	69246	})
	69247	#else
	69248	#define vfmlslq_lane_high_f16(__p0_919, __p1_919, __p2_919, __p3_919) __extension__ ({ \
	69249	float32x4_t __ret_919; \
	69250	float32x4_t __s0_919 = __p0_919; \
	69251	float16x8_t __s1_919 = __p1_919; \
	69252	float16x4_t __s2_919 = __p2_919; \
	69253	float32x4_t __rev0_919; __rev0_919 = __builtin_shufflevector(__s0_919, __s0_919, 3, 2, 1, 0); \
	69254	float16x8_t __rev1_919; __rev1_919 = __builtin_shufflevector(__s1_919, __s1_919, 7, 6, 5, 4, 3, 2, 1, 0); \
	69255	float16x4_t __rev2_919; __rev2_919 = __builtin_shufflevector(__s2_919, __s2_919, 3, 2, 1, 0); \
	69256	__ret_919 = __noswap_vfmlslq_high_f16(__rev0_919, __rev1_919, (float16x8_t) {__noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919), __noswap_vget_lane_f16(__rev2_919, __p3_919)}); \
	69257	__ret_919 = __builtin_shufflevector(__ret_919, __ret_919, 3, 2, 1, 0); \
	69258	__ret_919; \
	69259	})
	69260	#endif
	69261
	69262	#ifdef __LITTLE_ENDIAN__
	69263	#define vfmlsl_lane_high_f16(__p0_920, __p1_920, __p2_920, __p3_920) __extension__ ({ \
	69264	float32x2_t __ret_920; \
	69265	float32x2_t __s0_920 = __p0_920; \
	69266	float16x4_t __s1_920 = __p1_920; \
	69267	float16x4_t __s2_920 = __p2_920; \
	69268	__ret_920 = vfmlsl_high_f16(__s0_920, __s1_920, (float16x4_t) {vget_lane_f16(__s2_920, __p3_920), vget_lane_f16(__s2_920, __p3_920), vget_lane_f16(__s2_920, __p3_920), vget_lane_f16(__s2_920, __p3_920)}); \
	69269	__ret_920; \
	69270	})
	69271	#else
	69272	#define vfmlsl_lane_high_f16(__p0_921, __p1_921, __p2_921, __p3_921) __extension__ ({ \
	69273	float32x2_t __ret_921; \
	69274	float32x2_t __s0_921 = __p0_921; \
	69275	float16x4_t __s1_921 = __p1_921; \
	69276	float16x4_t __s2_921 = __p2_921; \
	69277	float32x2_t __rev0_921; __rev0_921 = __builtin_shufflevector(__s0_921, __s0_921, 1, 0); \
	69278	float16x4_t __rev1_921; __rev1_921 = __builtin_shufflevector(__s1_921, __s1_921, 3, 2, 1, 0); \
	69279	float16x4_t __rev2_921; __rev2_921 = __builtin_shufflevector(__s2_921, __s2_921, 3, 2, 1, 0); \
	69280	__ret_921 = __noswap_vfmlsl_high_f16(__rev0_921, __rev1_921, (float16x4_t) {__noswap_vget_lane_f16(__rev2_921, __p3_921), __noswap_vget_lane_f16(__rev2_921, __p3_921), __noswap_vget_lane_f16(__rev2_921, __p3_921), __noswap_vget_lane_f16(__rev2_921, __p3_921)}); \
	69281	__ret_921 = __builtin_shufflevector(__ret_921, __ret_921, 1, 0); \
	69282	__ret_921; \
	69283	})
	69284	#endif
	69285
	69286	#ifdef __LITTLE_ENDIAN__
	69287	#define vfmlslq_lane_low_f16(__p0_922, __p1_922, __p2_922, __p3_922) __extension__ ({ \
	69288	float32x4_t __ret_922; \
	69289	float32x4_t __s0_922 = __p0_922; \
	69290	float16x8_t __s1_922 = __p1_922; \
	69291	float16x4_t __s2_922 = __p2_922; \
	69292	__ret_922 = vfmlslq_low_f16(__s0_922, __s1_922, (float16x8_t) {vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922), vget_lane_f16(__s2_922, __p3_922)}); \
	69293	__ret_922; \
	69294	})
	69295	#else
	69296	#define vfmlslq_lane_low_f16(__p0_923, __p1_923, __p2_923, __p3_923) __extension__ ({ \
	69297	float32x4_t __ret_923; \
	69298	float32x4_t __s0_923 = __p0_923; \
	69299	float16x8_t __s1_923 = __p1_923; \
	69300	float16x4_t __s2_923 = __p2_923; \
	69301	float32x4_t __rev0_923; __rev0_923 = __builtin_shufflevector(__s0_923, __s0_923, 3, 2, 1, 0); \
	69302	float16x8_t __rev1_923; __rev1_923 = __builtin_shufflevector(__s1_923, __s1_923, 7, 6, 5, 4, 3, 2, 1, 0); \
	69303	float16x4_t __rev2_923; __rev2_923 = __builtin_shufflevector(__s2_923, __s2_923, 3, 2, 1, 0); \
	69304	__ret_923 = __noswap_vfmlslq_low_f16(__rev0_923, __rev1_923, (float16x8_t) {__noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923), __noswap_vget_lane_f16(__rev2_923, __p3_923)}); \
	69305	__ret_923 = __builtin_shufflevector(__ret_923, __ret_923, 3, 2, 1, 0); \
	69306	__ret_923; \
	69307	})
	69308	#endif
	69309
	69310	#ifdef __LITTLE_ENDIAN__
	69311	#define vfmlsl_lane_low_f16(__p0_924, __p1_924, __p2_924, __p3_924) __extension__ ({ \
	69312	float32x2_t __ret_924; \
	69313	float32x2_t __s0_924 = __p0_924; \
	69314	float16x4_t __s1_924 = __p1_924; \
	69315	float16x4_t __s2_924 = __p2_924; \
	69316	__ret_924 = vfmlsl_low_f16(__s0_924, __s1_924, (float16x4_t) {vget_lane_f16(__s2_924, __p3_924), vget_lane_f16(__s2_924, __p3_924), vget_lane_f16(__s2_924, __p3_924), vget_lane_f16(__s2_924, __p3_924)}); \
	69317	__ret_924; \
	69318	})
	69319	#else
	69320	#define vfmlsl_lane_low_f16(__p0_925, __p1_925, __p2_925, __p3_925) __extension__ ({ \
	69321	float32x2_t __ret_925; \
	69322	float32x2_t __s0_925 = __p0_925; \
	69323	float16x4_t __s1_925 = __p1_925; \
	69324	float16x4_t __s2_925 = __p2_925; \
	69325	float32x2_t __rev0_925; __rev0_925 = __builtin_shufflevector(__s0_925, __s0_925, 1, 0); \
	69326	float16x4_t __rev1_925; __rev1_925 = __builtin_shufflevector(__s1_925, __s1_925, 3, 2, 1, 0); \
	69327	float16x4_t __rev2_925; __rev2_925 = __builtin_shufflevector(__s2_925, __s2_925, 3, 2, 1, 0); \
	69328	__ret_925 = __noswap_vfmlsl_low_f16(__rev0_925, __rev1_925, (float16x4_t) {__noswap_vget_lane_f16(__rev2_925, __p3_925), __noswap_vget_lane_f16(__rev2_925, __p3_925), __noswap_vget_lane_f16(__rev2_925, __p3_925), __noswap_vget_lane_f16(__rev2_925, __p3_925)}); \
	69329	__ret_925 = __builtin_shufflevector(__ret_925, __ret_925, 1, 0); \
	69330	__ret_925; \
	69331	})
	69332	#endif
	69333
	69334	#ifdef __LITTLE_ENDIAN__
	69335	#define vfmlslq_laneq_high_f16(__p0_926, __p1_926, __p2_926, __p3_926) __extension__ ({ \
	69336	float32x4_t __ret_926; \
	69337	float32x4_t __s0_926 = __p0_926; \
	69338	float16x8_t __s1_926 = __p1_926; \
	69339	float16x8_t __s2_926 = __p2_926; \
	69340	__ret_926 = vfmlslq_high_f16(__s0_926, __s1_926, (float16x8_t) {vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926), vgetq_lane_f16(__s2_926, __p3_926)}); \
	69341	__ret_926; \
	69342	})
	69343	#else
	69344	#define vfmlslq_laneq_high_f16(__p0_927, __p1_927, __p2_927, __p3_927) __extension__ ({ \
	69345	float32x4_t __ret_927; \
	69346	float32x4_t __s0_927 = __p0_927; \
	69347	float16x8_t __s1_927 = __p1_927; \
	69348	float16x8_t __s2_927 = __p2_927; \
	69349	float32x4_t __rev0_927; __rev0_927 = __builtin_shufflevector(__s0_927, __s0_927, 3, 2, 1, 0); \
	69350	float16x8_t __rev1_927; __rev1_927 = __builtin_shufflevector(__s1_927, __s1_927, 7, 6, 5, 4, 3, 2, 1, 0); \
	69351	float16x8_t __rev2_927; __rev2_927 = __builtin_shufflevector(__s2_927, __s2_927, 7, 6, 5, 4, 3, 2, 1, 0); \
	69352	__ret_927 = __noswap_vfmlslq_high_f16(__rev0_927, __rev1_927, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927), __noswap_vgetq_lane_f16(__rev2_927, __p3_927)}); \
	69353	__ret_927 = __builtin_shufflevector(__ret_927, __ret_927, 3, 2, 1, 0); \
	69354	__ret_927; \
	69355	})
	69356	#endif
	69357
	69358	#ifdef __LITTLE_ENDIAN__
	69359	#define vfmlsl_laneq_high_f16(__p0_928, __p1_928, __p2_928, __p3_928) __extension__ ({ \
	69360	float32x2_t __ret_928; \
	69361	float32x2_t __s0_928 = __p0_928; \
	69362	float16x4_t __s1_928 = __p1_928; \
	69363	float16x8_t __s2_928 = __p2_928; \
	69364	__ret_928 = vfmlsl_high_f16(__s0_928, __s1_928, (float16x4_t) {vgetq_lane_f16(__s2_928, __p3_928), vgetq_lane_f16(__s2_928, __p3_928), vgetq_lane_f16(__s2_928, __p3_928), vgetq_lane_f16(__s2_928, __p3_928)}); \
	69365	__ret_928; \
	69366	})
	69367	#else
	69368	#define vfmlsl_laneq_high_f16(__p0_929, __p1_929, __p2_929, __p3_929) __extension__ ({ \
	69369	float32x2_t __ret_929; \
	69370	float32x2_t __s0_929 = __p0_929; \
	69371	float16x4_t __s1_929 = __p1_929; \
	69372	float16x8_t __s2_929 = __p2_929; \
	69373	float32x2_t __rev0_929; __rev0_929 = __builtin_shufflevector(__s0_929, __s0_929, 1, 0); \
	69374	float16x4_t __rev1_929; __rev1_929 = __builtin_shufflevector(__s1_929, __s1_929, 3, 2, 1, 0); \
	69375	float16x8_t __rev2_929; __rev2_929 = __builtin_shufflevector(__s2_929, __s2_929, 7, 6, 5, 4, 3, 2, 1, 0); \
	69376	__ret_929 = __noswap_vfmlsl_high_f16(__rev0_929, __rev1_929, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_929, __p3_929), __noswap_vgetq_lane_f16(__rev2_929, __p3_929), __noswap_vgetq_lane_f16(__rev2_929, __p3_929), __noswap_vgetq_lane_f16(__rev2_929, __p3_929)}); \
	69377	__ret_929 = __builtin_shufflevector(__ret_929, __ret_929, 1, 0); \
	69378	__ret_929; \
	69379	})
	69380	#endif
	69381
	69382	#ifdef __LITTLE_ENDIAN__
	69383	#define vfmlslq_laneq_low_f16(__p0_930, __p1_930, __p2_930, __p3_930) __extension__ ({ \
	69384	float32x4_t __ret_930; \
	69385	float32x4_t __s0_930 = __p0_930; \
	69386	float16x8_t __s1_930 = __p1_930; \
	69387	float16x8_t __s2_930 = __p2_930; \
	69388	__ret_930 = vfmlslq_low_f16(__s0_930, __s1_930, (float16x8_t) {vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930), vgetq_lane_f16(__s2_930, __p3_930)}); \
	69389	__ret_930; \
	69390	})
	69391	#else
	69392	#define vfmlslq_laneq_low_f16(__p0_931, __p1_931, __p2_931, __p3_931) __extension__ ({ \
	69393	float32x4_t __ret_931; \
	69394	float32x4_t __s0_931 = __p0_931; \
	69395	float16x8_t __s1_931 = __p1_931; \
	69396	float16x8_t __s2_931 = __p2_931; \
	69397	float32x4_t __rev0_931; __rev0_931 = __builtin_shufflevector(__s0_931, __s0_931, 3, 2, 1, 0); \
	69398	float16x8_t __rev1_931; __rev1_931 = __builtin_shufflevector(__s1_931, __s1_931, 7, 6, 5, 4, 3, 2, 1, 0); \
	69399	float16x8_t __rev2_931; __rev2_931 = __builtin_shufflevector(__s2_931, __s2_931, 7, 6, 5, 4, 3, 2, 1, 0); \
	69400	__ret_931 = __noswap_vfmlslq_low_f16(__rev0_931, __rev1_931, (float16x8_t) {__noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931), __noswap_vgetq_lane_f16(__rev2_931, __p3_931)}); \
	69401	__ret_931 = __builtin_shufflevector(__ret_931, __ret_931, 3, 2, 1, 0); \
	69402	__ret_931; \
	69403	})
	69404	#endif
	69405
	69406	#ifdef __LITTLE_ENDIAN__
	69407	#define vfmlsl_laneq_low_f16(__p0_932, __p1_932, __p2_932, __p3_932) __extension__ ({ \
	69408	float32x2_t __ret_932; \
	69409	float32x2_t __s0_932 = __p0_932; \
	69410	float16x4_t __s1_932 = __p1_932; \
	69411	float16x8_t __s2_932 = __p2_932; \
	69412	__ret_932 = vfmlsl_low_f16(__s0_932, __s1_932, (float16x4_t) {vgetq_lane_f16(__s2_932, __p3_932), vgetq_lane_f16(__s2_932, __p3_932), vgetq_lane_f16(__s2_932, __p3_932), vgetq_lane_f16(__s2_932, __p3_932)}); \
	69413	__ret_932; \
	69414	})
	69415	#else
	69416	#define vfmlsl_laneq_low_f16(__p0_933, __p1_933, __p2_933, __p3_933) __extension__ ({ \
	69417	float32x2_t __ret_933; \
	69418	float32x2_t __s0_933 = __p0_933; \
	69419	float16x4_t __s1_933 = __p1_933; \
	69420	float16x8_t __s2_933 = __p2_933; \
	69421	float32x2_t __rev0_933; __rev0_933 = __builtin_shufflevector(__s0_933, __s0_933, 1, 0); \
	69422	float16x4_t __rev1_933; __rev1_933 = __builtin_shufflevector(__s1_933, __s1_933, 3, 2, 1, 0); \
	69423	float16x8_t __rev2_933; __rev2_933 = __builtin_shufflevector(__s2_933, __s2_933, 7, 6, 5, 4, 3, 2, 1, 0); \
	69424	__ret_933 = __noswap_vfmlsl_low_f16(__rev0_933, __rev1_933, (float16x4_t) {__noswap_vgetq_lane_f16(__rev2_933, __p3_933), __noswap_vgetq_lane_f16(__rev2_933, __p3_933), __noswap_vgetq_lane_f16(__rev2_933, __p3_933), __noswap_vgetq_lane_f16(__rev2_933, __p3_933)}); \
	69425	__ret_933 = __builtin_shufflevector(__ret_933, __ret_933, 1, 0); \
	69426	__ret_933; \
	69427	})
	69428	#endif
	69429
	69430	#ifdef __LITTLE_ENDIAN__
	69431	#define vmulh_lane_f16(__p0_934, __p1_934, __p2_934) __extension__ ({ \
	69432	float16_t __ret_934; \
	69433	float16_t __s0_934 = __p0_934; \
	69434	float16x4_t __s1_934 = __p1_934; \
	69435	__ret_934 = __s0_934 * vget_lane_f16(__s1_934, __p2_934); \
	69436	__ret_934; \
	69437	})
	69438	#else
	69439	#define vmulh_lane_f16(__p0_935, __p1_935, __p2_935) __extension__ ({ \
	69440	float16_t __ret_935; \
	69441	float16_t __s0_935 = __p0_935; \
	69442	float16x4_t __s1_935 = __p1_935; \
	69443	float16x4_t __rev1_935; __rev1_935 = __builtin_shufflevector(__s1_935, __s1_935, 3, 2, 1, 0); \
	69444	__ret_935 = __s0_935 * __noswap_vget_lane_f16(__rev1_935, __p2_935); \
	69445	__ret_935; \
	69446	})
	69447	#endif
	69448
	69449	#ifdef __LITTLE_ENDIAN__
	69450	#define vmulh_laneq_f16(__p0_936, __p1_936, __p2_936) __extension__ ({ \
	69451	float16_t __ret_936; \
	69452	float16_t __s0_936 = __p0_936; \
	69453	float16x8_t __s1_936 = __p1_936; \
	69454	__ret_936 = __s0_936 * vgetq_lane_f16(__s1_936, __p2_936); \
	69455	__ret_936; \
	69456	})
	69457	#else
	69458	#define vmulh_laneq_f16(__p0_937, __p1_937, __p2_937) __extension__ ({ \
	69459	float16_t __ret_937; \
	69460	float16_t __s0_937 = __p0_937; \
	69461	float16x8_t __s1_937 = __p1_937; \
	69462	float16x8_t __rev1_937; __rev1_937 = __builtin_shufflevector(__s1_937, __s1_937, 7, 6, 5, 4, 3, 2, 1, 0); \
	69463	__ret_937 = __s0_937 * __noswap_vgetq_lane_f16(__rev1_937, __p2_937); \
	69464	__ret_937; \
69635	69465	})
69636	69466	#endif
69637	69467

lib/include/arm_neon_sve_bridge.h created+182

...	...	@@ -0,0 +1,182 @@
	1	/*===---- arm_neon_sve_bridge.h - ARM NEON SVE Bridge intrinsics -----------===
	2	*
	3	*
	4	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	5	* See https://llvm.org/LICENSE.txt for license information.
	6	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	7	*
	8	*===-----------------------------------------------------------------------===
	9	*/
	10
	11	#ifndef __ARM_NEON_SVE_BRIDGE_H
	12	#define __ARM_NEON_SVE_BRIDGE_H
	13
	14	#include <arm_neon.h>
	15	#include <arm_sve.h>
	16
	17	#ifdef __cplusplus
	18	extern "C" {
	19	#endif
	20
	21	/* Function attributes */
	22	#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
	23	#define __aio \
	24	static __inline__ \
	25	__attribute__((__always_inline__, __nodebug__, __overloadable__))
	26
	27	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s8)))
	28	svint8_t svset_neonq(svint8_t, int8x16_t);
	29	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s16)))
	30	svint16_t svset_neonq(svint16_t, int16x8_t);
	31	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s32)))
	32	svint32_t svset_neonq(svint32_t, int32x4_t);
	33	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s64)))
	34	svint64_t svset_neonq(svint64_t, int64x2_t);
	35	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u8)))
	36	svuint8_t svset_neonq(svuint8_t, uint8x16_t);
	37	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u16)))
	38	svuint16_t svset_neonq(svuint16_t, uint16x8_t);
	39	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u32)))
	40	svuint32_t svset_neonq(svuint32_t, uint32x4_t);
	41	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u64)))
	42	svuint64_t svset_neonq(svuint64_t, uint64x2_t);
	43	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f16)))
	44	svfloat16_t svset_neonq(svfloat16_t, float16x8_t);
	45	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f32)))
	46	svfloat32_t svset_neonq(svfloat32_t, float32x4_t);
	47	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f64)))
	48	svfloat64_t svset_neonq(svfloat64_t, float64x2_t);
	49	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s8)))
	50	svint8_t svset_neonq_s8(svint8_t, int8x16_t);
	51	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s16)))
	52	svint16_t svset_neonq_s16(svint16_t, int16x8_t);
	53	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s32)))
	54	svint32_t svset_neonq_s32(svint32_t, int32x4_t);
	55	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s64)))
	56	svint64_t svset_neonq_s64(svint64_t, int64x2_t);
	57	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u8)))
	58	svuint8_t svset_neonq_u8(svuint8_t, uint8x16_t);
	59	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u16)))
	60	svuint16_t svset_neonq_u16(svuint16_t, uint16x8_t);
	61	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u32)))
	62	svuint32_t svset_neonq_u32(svuint32_t, uint32x4_t);
	63	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u64)))
	64	svuint64_t svset_neonq_u64(svuint64_t, uint64x2_t);
	65	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f16)))
	66	svfloat16_t svset_neonq_f16(svfloat16_t, float16x8_t);
	67	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f32)))
	68	svfloat32_t svset_neonq_f32(svfloat32_t, float32x4_t);
	69	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f64)))
	70	svfloat64_t svset_neonq_f64(svfloat64_t, float64x2_t);
	71
	72	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s8)))
	73	int8x16_t svget_neonq(svint8_t);
	74	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s16)))
	75	int16x8_t svget_neonq(svint16_t);
	76	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s32)))
	77	int32x4_t svget_neonq(svint32_t);
	78	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s64)))
	79	int64x2_t svget_neonq(svint64_t);
	80	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u8)))
	81	uint8x16_t svget_neonq(svuint8_t);
	82	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u16)))
	83	uint16x8_t svget_neonq(svuint16_t);
	84	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u32)))
	85	uint32x4_t svget_neonq(svuint32_t);
	86	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u64)))
	87	uint64x2_t svget_neonq(svuint64_t);
	88	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f16)))
	89	float16x8_t svget_neonq(svfloat16_t);
	90	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f32)))
	91	float32x4_t svget_neonq(svfloat32_t);
	92	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f64)))
	93	float64x2_t svget_neonq(svfloat64_t);
	94	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s8)))
	95	int8x16_t svget_neonq_s8(svint8_t);
	96	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s16)))
	97	int16x8_t svget_neonq_s16(svint16_t);
	98	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s32)))
	99	int32x4_t svget_neonq_s32(svint32_t);
	100	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s64)))
	101	int64x2_t svget_neonq_s64(svint64_t);
	102	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u8)))
	103	uint8x16_t svget_neonq_u8(svuint8_t);
	104	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u16)))
	105	uint16x8_t svget_neonq_u16(svuint16_t);
	106	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u32)))
	107	uint32x4_t svget_neonq_u32(svuint32_t);
	108	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u64)))
	109	uint64x2_t svget_neonq_u64(svuint64_t);
	110	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f16)))
	111	float16x8_t svget_neonq_f16(svfloat16_t);
	112	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f32)))
	113	float32x4_t svget_neonq_f32(svfloat32_t);
	114	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f64)))
	115	float64x2_t svget_neonq_f64(svfloat64_t);
	116
	117	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s8)))
	118	svint8_t svdup_neonq(int8x16_t);
	119	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s16)))
	120	svint16_t svdup_neonq(int16x8_t);
	121	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s32)))
	122	svint32_t svdup_neonq(int32x4_t);
	123	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s64)))
	124	svint64_t svdup_neonq(int64x2_t);
	125	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u8)))
	126	svuint8_t svdup_neonq(uint8x16_t);
	127	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u16)))
	128	svuint16_t svdup_neonq(uint16x8_t);
	129	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u32)))
	130	svuint32_t svdup_neonq(uint32x4_t);
	131	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u64)))
	132	svuint64_t svdup_neonq(uint64x2_t);
	133	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f16)))
	134	svfloat16_t svdup_neonq(float16x8_t);
	135	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f32)))
	136	svfloat32_t svdup_neonq(float32x4_t);
	137	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f64)))
	138	svfloat64_t svdup_neonq(float64x2_t);
	139	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s8)))
	140	svint8_t svdup_neonq_s8(int8x16_t);
	141	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s16)))
	142	svint16_t svdup_neonq_s16(int16x8_t);
	143	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s32)))
	144	svint32_t svdup_neonq_s32(int32x4_t);
	145	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s64)))
	146	svint64_t svdup_neonq_s64(int64x2_t);
	147	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u8)))
	148	svuint8_t svdup_neonq_u8(uint8x16_t);
	149	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u16)))
	150	svuint16_t svdup_neonq_u16(uint16x8_t);
	151	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u32)))
	152	svuint32_t svdup_neonq_u32(uint32x4_t);
	153	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u64)))
	154	svuint64_t svdup_neonq_u64(uint64x2_t);
	155	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f16)))
	156	svfloat16_t svdup_neonq_f16(float16x8_t);
	157	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f32)))
	158	svfloat32_t svdup_neonq_f32(float32x4_t);
	159	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f64)))
	160	svfloat64_t svdup_neonq_f64(float64x2_t);
	161
	162	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_bf16)))
	163	svbfloat16_t svset_neonq(svbfloat16_t, bfloat16x8_t);
	164	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_bf16)))
	165	svbfloat16_t svset_neonq_bf16(svbfloat16_t, bfloat16x8_t);
	166	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_bf16)))
	167	bfloat16x8_t svget_neonq(svbfloat16_t);
	168	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_bf16)))
	169	bfloat16x8_t svget_neonq_bf16(svbfloat16_t);
	170	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_bf16)))
	171	svbfloat16_t svdup_neonq(bfloat16x8_t);
	172	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_bf16)))
	173	svbfloat16_t svdup_neonq_bf16(bfloat16x8_t);
	174
	175	#undef __ai
	176	#undef __aio
	177
	178	#ifdef __cplusplus
	179	} // extern "C"
	180	#endif
	181
	182	#endif //__ARM_NEON_SVE_BRIDGE_H

lib/include/arm_sve.h+7756-7908

...	...	@@ -11,10 +11,6 @@
11	11	#ifndef __ARM_SVE_H
12	12	#define __ARM_SVE_H
13	13
14		#if !defined(__ARM_FEATURE_SVE)
15		#error "SVE support not enabled"
16		#else
17
18	14	#if !defined(__LITTLE_ENDIAN__)
19	15	#error "Big endian is currently not supported for arm_sve.h"
20	16	#endif
...	...	@@ -39,19 +35,9 @@ typedef __SVUint32_t svuint32_t;
39	35	typedef __SVUint64_t svuint64_t;
40	36	typedef __SVFloat16_t svfloat16_t;
41	37
42		#if defined(__ARM_FEATURE_SVE_BF16) && !defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
43		#error "__ARM_FEATURE_BF16_SCALAR_ARITHMETIC must be defined when __ARM_FEATURE_SVE_BF16 is defined"
44		#endif
45
46		#if defined(__ARM_FEATURE_SVE_BF16)
47	38	typedef __SVBFloat16_t svbfloat16_t;
48		#endif
49
50		#if defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
51	39	#include <arm_bf16.h>
52	40	typedef __bf16 bfloat16_t;
53		#endif
54
55	41	typedef __SVFloat32_t svfloat32_t;
56	42	typedef __SVFloat64_t svfloat64_t;
57	43	typedef __clang_svint8x2_t svint8x2_t;
...	...	@@ -89,11 +75,9 @@ typedef __clang_svfloat32x4_t svfloat32x4_t;
89	75	typedef __clang_svfloat64x4_t svfloat64x4_t;
90	76	typedef __SVBool_t svbool_t;
91	77
92		#ifdef __ARM_FEATURE_SVE_BF16
93	78	typedef __clang_svbfloat16x2_t svbfloat16x2_t;
94	79	typedef __clang_svbfloat16x3_t svbfloat16x3_t;
95	80	typedef __clang_svbfloat16x4_t svbfloat16x4_t;
96		#endif
97	81	enum svpattern
98	82	{
99	83	SV_POW2 = 0,
...	...	@@ -145,9 +129,7 @@ enum svprfop
145	129	#define svreinterpret_s8_u32(...) __builtin_sve_reinterpret_s8_u32(__VA_ARGS__)
146	130	#define svreinterpret_s8_u64(...) __builtin_sve_reinterpret_s8_u64(__VA_ARGS__)
147	131	#define svreinterpret_s8_f16(...) __builtin_sve_reinterpret_s8_f16(__VA_ARGS__)
148		#if defined(__ARM_FEATURE_SVE_BF16)
149	132	#define svreinterpret_s8_bf16(...) __builtin_sve_reinterpret_s8_bf16(__VA_ARGS__)
150		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
151	133	#define svreinterpret_s8_f32(...) __builtin_sve_reinterpret_s8_f32(__VA_ARGS__)
152	134	#define svreinterpret_s8_f64(...) __builtin_sve_reinterpret_s8_f64(__VA_ARGS__)
153	135	#define svreinterpret_s16_s8(...) __builtin_sve_reinterpret_s16_s8(__VA_ARGS__)
...	...	@@ -159,9 +141,7 @@ enum svprfop
159	141	#define svreinterpret_s16_u32(...) __builtin_sve_reinterpret_s16_u32(__VA_ARGS__)
160	142	#define svreinterpret_s16_u64(...) __builtin_sve_reinterpret_s16_u64(__VA_ARGS__)
161	143	#define svreinterpret_s16_f16(...) __builtin_sve_reinterpret_s16_f16(__VA_ARGS__)
162		#if defined(__ARM_FEATURE_SVE_BF16)
163	144	#define svreinterpret_s16_bf16(...) __builtin_sve_reinterpret_s16_bf16(__VA_ARGS__)
164		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
165	145	#define svreinterpret_s16_f32(...) __builtin_sve_reinterpret_s16_f32(__VA_ARGS__)
166	146	#define svreinterpret_s16_f64(...) __builtin_sve_reinterpret_s16_f64(__VA_ARGS__)
167	147	#define svreinterpret_s32_s8(...) __builtin_sve_reinterpret_s32_s8(__VA_ARGS__)
...	...	@@ -173,9 +153,7 @@ enum svprfop
173	153	#define svreinterpret_s32_u32(...) __builtin_sve_reinterpret_s32_u32(__VA_ARGS__)
174	154	#define svreinterpret_s32_u64(...) __builtin_sve_reinterpret_s32_u64(__VA_ARGS__)
175	155	#define svreinterpret_s32_f16(...) __builtin_sve_reinterpret_s32_f16(__VA_ARGS__)
176		#if defined(__ARM_FEATURE_SVE_BF16)
177	156	#define svreinterpret_s32_bf16(...) __builtin_sve_reinterpret_s32_bf16(__VA_ARGS__)
178		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
179	157	#define svreinterpret_s32_f32(...) __builtin_sve_reinterpret_s32_f32(__VA_ARGS__)
180	158	#define svreinterpret_s32_f64(...) __builtin_sve_reinterpret_s32_f64(__VA_ARGS__)
181	159	#define svreinterpret_s64_s8(...) __builtin_sve_reinterpret_s64_s8(__VA_ARGS__)
...	...	@@ -187,9 +165,7 @@ enum svprfop
187	165	#define svreinterpret_s64_u32(...) __builtin_sve_reinterpret_s64_u32(__VA_ARGS__)
188	166	#define svreinterpret_s64_u64(...) __builtin_sve_reinterpret_s64_u64(__VA_ARGS__)
189	167	#define svreinterpret_s64_f16(...) __builtin_sve_reinterpret_s64_f16(__VA_ARGS__)
190		#if defined(__ARM_FEATURE_SVE_BF16)
191	168	#define svreinterpret_s64_bf16(...) __builtin_sve_reinterpret_s64_bf16(__VA_ARGS__)
192		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
193	169	#define svreinterpret_s64_f32(...) __builtin_sve_reinterpret_s64_f32(__VA_ARGS__)
194	170	#define svreinterpret_s64_f64(...) __builtin_sve_reinterpret_s64_f64(__VA_ARGS__)
195	171	#define svreinterpret_u8_s8(...) __builtin_sve_reinterpret_u8_s8(__VA_ARGS__)
...	...	@@ -201,9 +177,7 @@ enum svprfop
201	177	#define svreinterpret_u8_u32(...) __builtin_sve_reinterpret_u8_u32(__VA_ARGS__)
202	178	#define svreinterpret_u8_u64(...) __builtin_sve_reinterpret_u8_u64(__VA_ARGS__)
203	179	#define svreinterpret_u8_f16(...) __builtin_sve_reinterpret_u8_f16(__VA_ARGS__)
204		#if defined(__ARM_FEATURE_SVE_BF16)
205	180	#define svreinterpret_u8_bf16(...) __builtin_sve_reinterpret_u8_bf16(__VA_ARGS__)
206		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
207	181	#define svreinterpret_u8_f32(...) __builtin_sve_reinterpret_u8_f32(__VA_ARGS__)
208	182	#define svreinterpret_u8_f64(...) __builtin_sve_reinterpret_u8_f64(__VA_ARGS__)
209	183	#define svreinterpret_u16_s8(...) __builtin_sve_reinterpret_u16_s8(__VA_ARGS__)
...	...	@@ -215,9 +189,7 @@ enum svprfop
215	189	#define svreinterpret_u16_u32(...) __builtin_sve_reinterpret_u16_u32(__VA_ARGS__)
216	190	#define svreinterpret_u16_u64(...) __builtin_sve_reinterpret_u16_u64(__VA_ARGS__)
217	191	#define svreinterpret_u16_f16(...) __builtin_sve_reinterpret_u16_f16(__VA_ARGS__)
218		#if defined(__ARM_FEATURE_SVE_BF16)
219	192	#define svreinterpret_u16_bf16(...) __builtin_sve_reinterpret_u16_bf16(__VA_ARGS__)
220		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
221	193	#define svreinterpret_u16_f32(...) __builtin_sve_reinterpret_u16_f32(__VA_ARGS__)
222	194	#define svreinterpret_u16_f64(...) __builtin_sve_reinterpret_u16_f64(__VA_ARGS__)
223	195	#define svreinterpret_u32_s8(...) __builtin_sve_reinterpret_u32_s8(__VA_ARGS__)
...	...	@@ -229,9 +201,7 @@ enum svprfop
229	201	#define svreinterpret_u32_u32(...) __builtin_sve_reinterpret_u32_u32(__VA_ARGS__)
230	202	#define svreinterpret_u32_u64(...) __builtin_sve_reinterpret_u32_u64(__VA_ARGS__)
231	203	#define svreinterpret_u32_f16(...) __builtin_sve_reinterpret_u32_f16(__VA_ARGS__)
232		#if defined(__ARM_FEATURE_SVE_BF16)
233	204	#define svreinterpret_u32_bf16(...) __builtin_sve_reinterpret_u32_bf16(__VA_ARGS__)
234		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
235	205	#define svreinterpret_u32_f32(...) __builtin_sve_reinterpret_u32_f32(__VA_ARGS__)
236	206	#define svreinterpret_u32_f64(...) __builtin_sve_reinterpret_u32_f64(__VA_ARGS__)
237	207	#define svreinterpret_u64_s8(...) __builtin_sve_reinterpret_u64_s8(__VA_ARGS__)
...	...	@@ -243,9 +213,7 @@ enum svprfop
243	213	#define svreinterpret_u64_u32(...) __builtin_sve_reinterpret_u64_u32(__VA_ARGS__)
244	214	#define svreinterpret_u64_u64(...) __builtin_sve_reinterpret_u64_u64(__VA_ARGS__)
245	215	#define svreinterpret_u64_f16(...) __builtin_sve_reinterpret_u64_f16(__VA_ARGS__)
246		#if defined(__ARM_FEATURE_SVE_BF16)
247	216	#define svreinterpret_u64_bf16(...) __builtin_sve_reinterpret_u64_bf16(__VA_ARGS__)
248		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
249	217	#define svreinterpret_u64_f32(...) __builtin_sve_reinterpret_u64_f32(__VA_ARGS__)
250	218	#define svreinterpret_u64_f64(...) __builtin_sve_reinterpret_u64_f64(__VA_ARGS__)
251	219	#define svreinterpret_f16_s8(...) __builtin_sve_reinterpret_f16_s8(__VA_ARGS__)
...	...	@@ -257,47 +225,21 @@ enum svprfop
257	225	#define svreinterpret_f16_u32(...) __builtin_sve_reinterpret_f16_u32(__VA_ARGS__)
258	226	#define svreinterpret_f16_u64(...) __builtin_sve_reinterpret_f16_u64(__VA_ARGS__)
259	227	#define svreinterpret_f16_f16(...) __builtin_sve_reinterpret_f16_f16(__VA_ARGS__)
260		#if defined(__ARM_FEATURE_SVE_BF16)
261	228	#define svreinterpret_f16_bf16(...) __builtin_sve_reinterpret_f16_bf16(__VA_ARGS__)
262		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
263	229	#define svreinterpret_f16_f32(...) __builtin_sve_reinterpret_f16_f32(__VA_ARGS__)
264	230	#define svreinterpret_f16_f64(...) __builtin_sve_reinterpret_f16_f64(__VA_ARGS__)
265		#if defined(__ARM_FEATURE_SVE_BF16)
266	231	#define svreinterpret_bf16_s8(...) __builtin_sve_reinterpret_bf16_s8(__VA_ARGS__)
267		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
268		#if defined(__ARM_FEATURE_SVE_BF16)
269	232	#define svreinterpret_bf16_s16(...) __builtin_sve_reinterpret_bf16_s16(__VA_ARGS__)
270		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
271		#if defined(__ARM_FEATURE_SVE_BF16)
272	233	#define svreinterpret_bf16_s32(...) __builtin_sve_reinterpret_bf16_s32(__VA_ARGS__)
273		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
274		#if defined(__ARM_FEATURE_SVE_BF16)
275	234	#define svreinterpret_bf16_s64(...) __builtin_sve_reinterpret_bf16_s64(__VA_ARGS__)
276		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
277		#if defined(__ARM_FEATURE_SVE_BF16)
278	235	#define svreinterpret_bf16_u8(...) __builtin_sve_reinterpret_bf16_u8(__VA_ARGS__)
279		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
280		#if defined(__ARM_FEATURE_SVE_BF16)
281	236	#define svreinterpret_bf16_u16(...) __builtin_sve_reinterpret_bf16_u16(__VA_ARGS__)
282		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
283		#if defined(__ARM_FEATURE_SVE_BF16)
284	237	#define svreinterpret_bf16_u32(...) __builtin_sve_reinterpret_bf16_u32(__VA_ARGS__)
285		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
286		#if defined(__ARM_FEATURE_SVE_BF16)
287	238	#define svreinterpret_bf16_u64(...) __builtin_sve_reinterpret_bf16_u64(__VA_ARGS__)
288		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
289		#if defined(__ARM_FEATURE_SVE_BF16)
290	239	#define svreinterpret_bf16_f16(...) __builtin_sve_reinterpret_bf16_f16(__VA_ARGS__)
291		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
292		#if defined(__ARM_FEATURE_SVE_BF16)
293	240	#define svreinterpret_bf16_bf16(...) __builtin_sve_reinterpret_bf16_bf16(__VA_ARGS__)
294		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
295		#if defined(__ARM_FEATURE_SVE_BF16)
296	241	#define svreinterpret_bf16_f32(...) __builtin_sve_reinterpret_bf16_f32(__VA_ARGS__)
297		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
298		#if defined(__ARM_FEATURE_SVE_BF16)
299	242	#define svreinterpret_bf16_f64(...) __builtin_sve_reinterpret_bf16_f64(__VA_ARGS__)
300		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
301	243	#define svreinterpret_f32_s8(...) __builtin_sve_reinterpret_f32_s8(__VA_ARGS__)
302	244	#define svreinterpret_f32_s16(...) __builtin_sve_reinterpret_f32_s16(__VA_ARGS__)
303	245	#define svreinterpret_f32_s32(...) __builtin_sve_reinterpret_f32_s32(__VA_ARGS__)
...	...	@@ -307,9 +249,7 @@ enum svprfop
307	249	#define svreinterpret_f32_u32(...) __builtin_sve_reinterpret_f32_u32(__VA_ARGS__)
308	250	#define svreinterpret_f32_u64(...) __builtin_sve_reinterpret_f32_u64(__VA_ARGS__)
309	251	#define svreinterpret_f32_f16(...) __builtin_sve_reinterpret_f32_f16(__VA_ARGS__)
310		#if defined(__ARM_FEATURE_SVE_BF16)
311	252	#define svreinterpret_f32_bf16(...) __builtin_sve_reinterpret_f32_bf16(__VA_ARGS__)
312		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
313	253	#define svreinterpret_f32_f32(...) __builtin_sve_reinterpret_f32_f32(__VA_ARGS__)
314	254	#define svreinterpret_f32_f64(...) __builtin_sve_reinterpret_f32_f64(__VA_ARGS__)
315	255	#define svreinterpret_f64_s8(...) __builtin_sve_reinterpret_f64_s8(__VA_ARGS__)
...	...	@@ -321,630 +261,582 @@ enum svprfop
321	261	#define svreinterpret_f64_u32(...) __builtin_sve_reinterpret_f64_u32(__VA_ARGS__)
322	262	#define svreinterpret_f64_u64(...) __builtin_sve_reinterpret_f64_u64(__VA_ARGS__)
323	263	#define svreinterpret_f64_f16(...) __builtin_sve_reinterpret_f64_f16(__VA_ARGS__)
324		#if defined(__ARM_FEATURE_SVE_BF16)
325	264	#define svreinterpret_f64_bf16(...) __builtin_sve_reinterpret_f64_bf16(__VA_ARGS__)
326		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
327	265	#define svreinterpret_f64_f32(...) __builtin_sve_reinterpret_f64_f32(__VA_ARGS__)
328	266	#define svreinterpret_f64_f64(...) __builtin_sve_reinterpret_f64_f64(__VA_ARGS__)
329		__aio svint8_t svreinterpret_s8(svint8_t op) {
	267	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svint8_t op) {
330	268	return __builtin_sve_reinterpret_s8_s8(op);
331	269	}
332	270
333		__aio svint8_t svreinterpret_s8(svint16_t op) {
	271	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svint16_t op) {
334	272	return __builtin_sve_reinterpret_s8_s16(op);
335	273	}
336	274
337		__aio svint8_t svreinterpret_s8(svint32_t op) {
	275	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svint32_t op) {
338	276	return __builtin_sve_reinterpret_s8_s32(op);
339	277	}
340	278
341		__aio svint8_t svreinterpret_s8(svint64_t op) {
	279	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svint64_t op) {
342	280	return __builtin_sve_reinterpret_s8_s64(op);
343	281	}
344	282
345		__aio svint8_t svreinterpret_s8(svuint8_t op) {
	283	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svuint8_t op) {
346	284	return __builtin_sve_reinterpret_s8_u8(op);
347	285	}
348	286
349		__aio svint8_t svreinterpret_s8(svuint16_t op) {
	287	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svuint16_t op) {
350	288	return __builtin_sve_reinterpret_s8_u16(op);
351	289	}
352	290
353		__aio svint8_t svreinterpret_s8(svuint32_t op) {
	291	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svuint32_t op) {
354	292	return __builtin_sve_reinterpret_s8_u32(op);
355	293	}
356	294
357		__aio svint8_t svreinterpret_s8(svuint64_t op) {
	295	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svuint64_t op) {
358	296	return __builtin_sve_reinterpret_s8_u64(op);
359	297	}
360	298
361		__aio svint8_t svreinterpret_s8(svfloat16_t op) {
	299	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svfloat16_t op) {
362	300	return __builtin_sve_reinterpret_s8_f16(op);
363	301	}
364	302
365		#if defined(__ARM_FEATURE_SVE_BF16)
366		__aio svint8_t svreinterpret_s8(svbfloat16_t op) {
	303	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svbfloat16_t op) {
367	304	return __builtin_sve_reinterpret_s8_bf16(op);
368	305	}
369	306
370		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
371		__aio svint8_t svreinterpret_s8(svfloat32_t op) {
	307	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svfloat32_t op) {
372	308	return __builtin_sve_reinterpret_s8_f32(op);
373	309	}
374	310
375		__aio svint8_t svreinterpret_s8(svfloat64_t op) {
	311	__aio __attribute__((target("sve"))) svint8_t svreinterpret_s8(svfloat64_t op) {
376	312	return __builtin_sve_reinterpret_s8_f64(op);
377	313	}
378	314
379		__aio svint16_t svreinterpret_s16(svint8_t op) {
	315	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svint8_t op) {
380	316	return __builtin_sve_reinterpret_s16_s8(op);
381	317	}
382	318
383		__aio svint16_t svreinterpret_s16(svint16_t op) {
	319	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svint16_t op) {
384	320	return __builtin_sve_reinterpret_s16_s16(op);
385	321	}
386	322
387		__aio svint16_t svreinterpret_s16(svint32_t op) {
	323	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svint32_t op) {
388	324	return __builtin_sve_reinterpret_s16_s32(op);
389	325	}
390	326
391		__aio svint16_t svreinterpret_s16(svint64_t op) {
	327	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svint64_t op) {
392	328	return __builtin_sve_reinterpret_s16_s64(op);
393	329	}
394	330
395		__aio svint16_t svreinterpret_s16(svuint8_t op) {
	331	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svuint8_t op) {
396	332	return __builtin_sve_reinterpret_s16_u8(op);
397	333	}
398	334
399		__aio svint16_t svreinterpret_s16(svuint16_t op) {
	335	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svuint16_t op) {
400	336	return __builtin_sve_reinterpret_s16_u16(op);
401	337	}
402	338
403		__aio svint16_t svreinterpret_s16(svuint32_t op) {
	339	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svuint32_t op) {
404	340	return __builtin_sve_reinterpret_s16_u32(op);
405	341	}
406	342
407		__aio svint16_t svreinterpret_s16(svuint64_t op) {
	343	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svuint64_t op) {
408	344	return __builtin_sve_reinterpret_s16_u64(op);
409	345	}
410	346
411		__aio svint16_t svreinterpret_s16(svfloat16_t op) {
	347	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svfloat16_t op) {
412	348	return __builtin_sve_reinterpret_s16_f16(op);
413	349	}
414	350
415		#if defined(__ARM_FEATURE_SVE_BF16)
416		__aio svint16_t svreinterpret_s16(svbfloat16_t op) {
	351	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svbfloat16_t op) {
417	352	return __builtin_sve_reinterpret_s16_bf16(op);
418	353	}
419	354
420		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
421		__aio svint16_t svreinterpret_s16(svfloat32_t op) {
	355	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svfloat32_t op) {
422	356	return __builtin_sve_reinterpret_s16_f32(op);
423	357	}
424	358
425		__aio svint16_t svreinterpret_s16(svfloat64_t op) {
	359	__aio __attribute__((target("sve"))) svint16_t svreinterpret_s16(svfloat64_t op) {
426	360	return __builtin_sve_reinterpret_s16_f64(op);
427	361	}
428	362
429		__aio svint32_t svreinterpret_s32(svint8_t op) {
	363	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svint8_t op) {
430	364	return __builtin_sve_reinterpret_s32_s8(op);
431	365	}
432	366
433		__aio svint32_t svreinterpret_s32(svint16_t op) {
	367	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svint16_t op) {
434	368	return __builtin_sve_reinterpret_s32_s16(op);
435	369	}
436	370
437		__aio svint32_t svreinterpret_s32(svint32_t op) {
	371	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svint32_t op) {
438	372	return __builtin_sve_reinterpret_s32_s32(op);
439	373	}
440	374
441		__aio svint32_t svreinterpret_s32(svint64_t op) {
	375	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svint64_t op) {
442	376	return __builtin_sve_reinterpret_s32_s64(op);
443	377	}
444	378
445		__aio svint32_t svreinterpret_s32(svuint8_t op) {
	379	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svuint8_t op) {
446	380	return __builtin_sve_reinterpret_s32_u8(op);
447	381	}
448	382
449		__aio svint32_t svreinterpret_s32(svuint16_t op) {
	383	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svuint16_t op) {
450	384	return __builtin_sve_reinterpret_s32_u16(op);
451	385	}
452	386
453		__aio svint32_t svreinterpret_s32(svuint32_t op) {
	387	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svuint32_t op) {
454	388	return __builtin_sve_reinterpret_s32_u32(op);
455	389	}
456	390
457		__aio svint32_t svreinterpret_s32(svuint64_t op) {
	391	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svuint64_t op) {
458	392	return __builtin_sve_reinterpret_s32_u64(op);
459	393	}
460	394
461		__aio svint32_t svreinterpret_s32(svfloat16_t op) {
	395	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svfloat16_t op) {
462	396	return __builtin_sve_reinterpret_s32_f16(op);
463	397	}
464	398
465		#if defined(__ARM_FEATURE_SVE_BF16)
466		__aio svint32_t svreinterpret_s32(svbfloat16_t op) {
	399	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svbfloat16_t op) {
467	400	return __builtin_sve_reinterpret_s32_bf16(op);
468	401	}
469	402
470		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
471		__aio svint32_t svreinterpret_s32(svfloat32_t op) {
	403	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svfloat32_t op) {
472	404	return __builtin_sve_reinterpret_s32_f32(op);
473	405	}
474	406
475		__aio svint32_t svreinterpret_s32(svfloat64_t op) {
	407	__aio __attribute__((target("sve"))) svint32_t svreinterpret_s32(svfloat64_t op) {
476	408	return __builtin_sve_reinterpret_s32_f64(op);
477	409	}
478	410
479		__aio svint64_t svreinterpret_s64(svint8_t op) {
	411	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svint8_t op) {
480	412	return __builtin_sve_reinterpret_s64_s8(op);
481	413	}
482	414
483		__aio svint64_t svreinterpret_s64(svint16_t op) {
	415	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svint16_t op) {
484	416	return __builtin_sve_reinterpret_s64_s16(op);
485	417	}
486	418
487		__aio svint64_t svreinterpret_s64(svint32_t op) {
	419	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svint32_t op) {
488	420	return __builtin_sve_reinterpret_s64_s32(op);
489	421	}
490	422
491		__aio svint64_t svreinterpret_s64(svint64_t op) {
	423	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svint64_t op) {
492	424	return __builtin_sve_reinterpret_s64_s64(op);
493	425	}
494	426
495		__aio svint64_t svreinterpret_s64(svuint8_t op) {
	427	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svuint8_t op) {
496	428	return __builtin_sve_reinterpret_s64_u8(op);
497	429	}
498	430
499		__aio svint64_t svreinterpret_s64(svuint16_t op) {
	431	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svuint16_t op) {
500	432	return __builtin_sve_reinterpret_s64_u16(op);
501	433	}
502	434
503		__aio svint64_t svreinterpret_s64(svuint32_t op) {
	435	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svuint32_t op) {
504	436	return __builtin_sve_reinterpret_s64_u32(op);
505	437	}
506	438
507		__aio svint64_t svreinterpret_s64(svuint64_t op) {
	439	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svuint64_t op) {
508	440	return __builtin_sve_reinterpret_s64_u64(op);
509	441	}
510	442
511		__aio svint64_t svreinterpret_s64(svfloat16_t op) {
	443	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svfloat16_t op) {
512	444	return __builtin_sve_reinterpret_s64_f16(op);
513	445	}
514	446
515		#if defined(__ARM_FEATURE_SVE_BF16)
516		__aio svint64_t svreinterpret_s64(svbfloat16_t op) {
	447	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svbfloat16_t op) {
517	448	return __builtin_sve_reinterpret_s64_bf16(op);
518	449	}
519	450
520		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
521		__aio svint64_t svreinterpret_s64(svfloat32_t op) {
	451	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svfloat32_t op) {
522	452	return __builtin_sve_reinterpret_s64_f32(op);
523	453	}
524	454
525		__aio svint64_t svreinterpret_s64(svfloat64_t op) {
	455	__aio __attribute__((target("sve"))) svint64_t svreinterpret_s64(svfloat64_t op) {
526	456	return __builtin_sve_reinterpret_s64_f64(op);
527	457	}
528	458
529		__aio svuint8_t svreinterpret_u8(svint8_t op) {
	459	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svint8_t op) {
530	460	return __builtin_sve_reinterpret_u8_s8(op);
531	461	}
532	462
533		__aio svuint8_t svreinterpret_u8(svint16_t op) {
	463	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svint16_t op) {
534	464	return __builtin_sve_reinterpret_u8_s16(op);
535	465	}
536	466
537		__aio svuint8_t svreinterpret_u8(svint32_t op) {
	467	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svint32_t op) {
538	468	return __builtin_sve_reinterpret_u8_s32(op);
539	469	}
540	470
541		__aio svuint8_t svreinterpret_u8(svint64_t op) {
	471	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svint64_t op) {
542	472	return __builtin_sve_reinterpret_u8_s64(op);
543	473	}
544	474
545		__aio svuint8_t svreinterpret_u8(svuint8_t op) {
	475	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svuint8_t op) {
546	476	return __builtin_sve_reinterpret_u8_u8(op);
547	477	}
548	478
549		__aio svuint8_t svreinterpret_u8(svuint16_t op) {
	479	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svuint16_t op) {
550	480	return __builtin_sve_reinterpret_u8_u16(op);
551	481	}
552	482
553		__aio svuint8_t svreinterpret_u8(svuint32_t op) {
	483	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svuint32_t op) {
554	484	return __builtin_sve_reinterpret_u8_u32(op);
555	485	}
556	486
557		__aio svuint8_t svreinterpret_u8(svuint64_t op) {
	487	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svuint64_t op) {
558	488	return __builtin_sve_reinterpret_u8_u64(op);
559	489	}
560	490
561		__aio svuint8_t svreinterpret_u8(svfloat16_t op) {
	491	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svfloat16_t op) {
562	492	return __builtin_sve_reinterpret_u8_f16(op);
563	493	}
564	494
565		#if defined(__ARM_FEATURE_SVE_BF16)
566		__aio svuint8_t svreinterpret_u8(svbfloat16_t op) {
	495	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svbfloat16_t op) {
567	496	return __builtin_sve_reinterpret_u8_bf16(op);
568	497	}
569	498
570		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
571		__aio svuint8_t svreinterpret_u8(svfloat32_t op) {
	499	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svfloat32_t op) {
572	500	return __builtin_sve_reinterpret_u8_f32(op);
573	501	}
574	502
575		__aio svuint8_t svreinterpret_u8(svfloat64_t op) {
	503	__aio __attribute__((target("sve"))) svuint8_t svreinterpret_u8(svfloat64_t op) {
576	504	return __builtin_sve_reinterpret_u8_f64(op);
577	505	}
578	506
579		__aio svuint16_t svreinterpret_u16(svint8_t op) {
	507	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svint8_t op) {
580	508	return __builtin_sve_reinterpret_u16_s8(op);
581	509	}
582	510
583		__aio svuint16_t svreinterpret_u16(svint16_t op) {
	511	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svint16_t op) {
584	512	return __builtin_sve_reinterpret_u16_s16(op);
585	513	}
586	514
587		__aio svuint16_t svreinterpret_u16(svint32_t op) {
	515	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svint32_t op) {
588	516	return __builtin_sve_reinterpret_u16_s32(op);
589	517	}
590	518
591		__aio svuint16_t svreinterpret_u16(svint64_t op) {
	519	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svint64_t op) {
592	520	return __builtin_sve_reinterpret_u16_s64(op);
593	521	}
594	522
595		__aio svuint16_t svreinterpret_u16(svuint8_t op) {
	523	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svuint8_t op) {
596	524	return __builtin_sve_reinterpret_u16_u8(op);
597	525	}
598	526
599		__aio svuint16_t svreinterpret_u16(svuint16_t op) {
	527	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svuint16_t op) {
600	528	return __builtin_sve_reinterpret_u16_u16(op);
601	529	}
602	530
603		__aio svuint16_t svreinterpret_u16(svuint32_t op) {
	531	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svuint32_t op) {
604	532	return __builtin_sve_reinterpret_u16_u32(op);
605	533	}
606	534
607		__aio svuint16_t svreinterpret_u16(svuint64_t op) {
	535	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svuint64_t op) {
608	536	return __builtin_sve_reinterpret_u16_u64(op);
609	537	}
610	538
611		__aio svuint16_t svreinterpret_u16(svfloat16_t op) {
	539	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svfloat16_t op) {
612	540	return __builtin_sve_reinterpret_u16_f16(op);
613	541	}
614	542
615		#if defined(__ARM_FEATURE_SVE_BF16)
616		__aio svuint16_t svreinterpret_u16(svbfloat16_t op) {
	543	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svbfloat16_t op) {
617	544	return __builtin_sve_reinterpret_u16_bf16(op);
618	545	}
619	546
620		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
621		__aio svuint16_t svreinterpret_u16(svfloat32_t op) {
	547	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svfloat32_t op) {
622	548	return __builtin_sve_reinterpret_u16_f32(op);
623	549	}
624	550
625		__aio svuint16_t svreinterpret_u16(svfloat64_t op) {
	551	__aio __attribute__((target("sve"))) svuint16_t svreinterpret_u16(svfloat64_t op) {
626	552	return __builtin_sve_reinterpret_u16_f64(op);
627	553	}
628	554
629		__aio svuint32_t svreinterpret_u32(svint8_t op) {
	555	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svint8_t op) {
630	556	return __builtin_sve_reinterpret_u32_s8(op);
631	557	}
632	558
633		__aio svuint32_t svreinterpret_u32(svint16_t op) {
	559	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svint16_t op) {
634	560	return __builtin_sve_reinterpret_u32_s16(op);
635	561	}
636	562
637		__aio svuint32_t svreinterpret_u32(svint32_t op) {
	563	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svint32_t op) {
638	564	return __builtin_sve_reinterpret_u32_s32(op);
639	565	}
640	566
641		__aio svuint32_t svreinterpret_u32(svint64_t op) {
	567	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svint64_t op) {
642	568	return __builtin_sve_reinterpret_u32_s64(op);
643	569	}
644	570
645		__aio svuint32_t svreinterpret_u32(svuint8_t op) {
	571	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svuint8_t op) {
646	572	return __builtin_sve_reinterpret_u32_u8(op);
647	573	}
648	574
649		__aio svuint32_t svreinterpret_u32(svuint16_t op) {
	575	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svuint16_t op) {
650	576	return __builtin_sve_reinterpret_u32_u16(op);
651	577	}
652	578
653		__aio svuint32_t svreinterpret_u32(svuint32_t op) {
	579	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svuint32_t op) {
654	580	return __builtin_sve_reinterpret_u32_u32(op);
655	581	}
656	582
657		__aio svuint32_t svreinterpret_u32(svuint64_t op) {
	583	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svuint64_t op) {
658	584	return __builtin_sve_reinterpret_u32_u64(op);
659	585	}
660	586
661		__aio svuint32_t svreinterpret_u32(svfloat16_t op) {
	587	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svfloat16_t op) {
662	588	return __builtin_sve_reinterpret_u32_f16(op);
663	589	}
664	590
665		#if defined(__ARM_FEATURE_SVE_BF16)
666		__aio svuint32_t svreinterpret_u32(svbfloat16_t op) {
	591	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svbfloat16_t op) {
667	592	return __builtin_sve_reinterpret_u32_bf16(op);
668	593	}
669	594
670		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
671		__aio svuint32_t svreinterpret_u32(svfloat32_t op) {
	595	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svfloat32_t op) {
672	596	return __builtin_sve_reinterpret_u32_f32(op);
673	597	}
674	598
675		__aio svuint32_t svreinterpret_u32(svfloat64_t op) {
	599	__aio __attribute__((target("sve"))) svuint32_t svreinterpret_u32(svfloat64_t op) {
676	600	return __builtin_sve_reinterpret_u32_f64(op);
677	601	}
678	602
679		__aio svuint64_t svreinterpret_u64(svint8_t op) {
	603	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svint8_t op) {
680	604	return __builtin_sve_reinterpret_u64_s8(op);
681	605	}
682	606
683		__aio svuint64_t svreinterpret_u64(svint16_t op) {
	607	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svint16_t op) {
684	608	return __builtin_sve_reinterpret_u64_s16(op);
685	609	}
686	610
687		__aio svuint64_t svreinterpret_u64(svint32_t op) {
	611	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svint32_t op) {
688	612	return __builtin_sve_reinterpret_u64_s32(op);
689	613	}
690	614
691		__aio svuint64_t svreinterpret_u64(svint64_t op) {
	615	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svint64_t op) {
692	616	return __builtin_sve_reinterpret_u64_s64(op);
693	617	}
694	618
695		__aio svuint64_t svreinterpret_u64(svuint8_t op) {
	619	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svuint8_t op) {
696	620	return __builtin_sve_reinterpret_u64_u8(op);
697	621	}
698	622
699		__aio svuint64_t svreinterpret_u64(svuint16_t op) {
	623	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svuint16_t op) {
700	624	return __builtin_sve_reinterpret_u64_u16(op);
701	625	}
702	626
703		__aio svuint64_t svreinterpret_u64(svuint32_t op) {
	627	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svuint32_t op) {
704	628	return __builtin_sve_reinterpret_u64_u32(op);
705	629	}
706	630
707		__aio svuint64_t svreinterpret_u64(svuint64_t op) {
	631	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svuint64_t op) {
708	632	return __builtin_sve_reinterpret_u64_u64(op);
709	633	}
710	634
711		__aio svuint64_t svreinterpret_u64(svfloat16_t op) {
	635	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svfloat16_t op) {
712	636	return __builtin_sve_reinterpret_u64_f16(op);
713	637	}
714	638
715		#if defined(__ARM_FEATURE_SVE_BF16)
716		__aio svuint64_t svreinterpret_u64(svbfloat16_t op) {
	639	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svbfloat16_t op) {
717	640	return __builtin_sve_reinterpret_u64_bf16(op);
718	641	}
719	642
720		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
721		__aio svuint64_t svreinterpret_u64(svfloat32_t op) {
	643	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svfloat32_t op) {
722	644	return __builtin_sve_reinterpret_u64_f32(op);
723	645	}
724	646
725		__aio svuint64_t svreinterpret_u64(svfloat64_t op) {
	647	__aio __attribute__((target("sve"))) svuint64_t svreinterpret_u64(svfloat64_t op) {
726	648	return __builtin_sve_reinterpret_u64_f64(op);
727	649	}
728	650
729		__aio svfloat16_t svreinterpret_f16(svint8_t op) {
	651	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svint8_t op) {
730	652	return __builtin_sve_reinterpret_f16_s8(op);
731	653	}
732	654
733		__aio svfloat16_t svreinterpret_f16(svint16_t op) {
	655	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svint16_t op) {
734	656	return __builtin_sve_reinterpret_f16_s16(op);
735	657	}
736	658
737		__aio svfloat16_t svreinterpret_f16(svint32_t op) {
	659	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svint32_t op) {
738	660	return __builtin_sve_reinterpret_f16_s32(op);
739	661	}
740	662
741		__aio svfloat16_t svreinterpret_f16(svint64_t op) {
	663	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svint64_t op) {
742	664	return __builtin_sve_reinterpret_f16_s64(op);
743	665	}
744	666
745		__aio svfloat16_t svreinterpret_f16(svuint8_t op) {
	667	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svuint8_t op) {
746	668	return __builtin_sve_reinterpret_f16_u8(op);
747	669	}
748	670
749		__aio svfloat16_t svreinterpret_f16(svuint16_t op) {
	671	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svuint16_t op) {
750	672	return __builtin_sve_reinterpret_f16_u16(op);
751	673	}
752	674
753		__aio svfloat16_t svreinterpret_f16(svuint32_t op) {
	675	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svuint32_t op) {
754	676	return __builtin_sve_reinterpret_f16_u32(op);
755	677	}
756	678
757		__aio svfloat16_t svreinterpret_f16(svuint64_t op) {
	679	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svuint64_t op) {
758	680	return __builtin_sve_reinterpret_f16_u64(op);
759	681	}
760	682
761		__aio svfloat16_t svreinterpret_f16(svfloat16_t op) {
	683	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svfloat16_t op) {
762	684	return __builtin_sve_reinterpret_f16_f16(op);
763	685	}
764	686
765		#if defined(__ARM_FEATURE_SVE_BF16)
766		__aio svfloat16_t svreinterpret_f16(svbfloat16_t op) {
	687	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svbfloat16_t op) {
767	688	return __builtin_sve_reinterpret_f16_bf16(op);
768	689	}
769	690
770		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
771		__aio svfloat16_t svreinterpret_f16(svfloat32_t op) {
	691	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svfloat32_t op) {
772	692	return __builtin_sve_reinterpret_f16_f32(op);
773	693	}
774	694
775		__aio svfloat16_t svreinterpret_f16(svfloat64_t op) {
	695	__aio __attribute__((target("sve"))) svfloat16_t svreinterpret_f16(svfloat64_t op) {
776	696	return __builtin_sve_reinterpret_f16_f64(op);
777	697	}
778	698
779		#if defined(__ARM_FEATURE_SVE_BF16)
780		__aio svbfloat16_t svreinterpret_bf16(svint8_t op) {
	699	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svint8_t op) {
781	700	return __builtin_sve_reinterpret_bf16_s8(op);
782	701	}
783	702
784		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
785		#if defined(__ARM_FEATURE_SVE_BF16)
786		__aio svbfloat16_t svreinterpret_bf16(svint16_t op) {
	703	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svint16_t op) {
787	704	return __builtin_sve_reinterpret_bf16_s16(op);
788	705	}
789	706
790		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
791		#if defined(__ARM_FEATURE_SVE_BF16)
792		__aio svbfloat16_t svreinterpret_bf16(svint32_t op) {
	707	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svint32_t op) {
793	708	return __builtin_sve_reinterpret_bf16_s32(op);
794	709	}
795	710
796		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
797		#if defined(__ARM_FEATURE_SVE_BF16)
798		__aio svbfloat16_t svreinterpret_bf16(svint64_t op) {
	711	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svint64_t op) {
799	712	return __builtin_sve_reinterpret_bf16_s64(op);
800	713	}
801	714
802		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
803		#if defined(__ARM_FEATURE_SVE_BF16)
804		__aio svbfloat16_t svreinterpret_bf16(svuint8_t op) {
	715	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svuint8_t op) {
805	716	return __builtin_sve_reinterpret_bf16_u8(op);
806	717	}
807	718
808		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
809		#if defined(__ARM_FEATURE_SVE_BF16)
810		__aio svbfloat16_t svreinterpret_bf16(svuint16_t op) {
	719	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svuint16_t op) {
811	720	return __builtin_sve_reinterpret_bf16_u16(op);
812	721	}
813	722
814		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
815		#if defined(__ARM_FEATURE_SVE_BF16)
816		__aio svbfloat16_t svreinterpret_bf16(svuint32_t op) {
	723	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svuint32_t op) {
817	724	return __builtin_sve_reinterpret_bf16_u32(op);
818	725	}
819	726
820		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
821		#if defined(__ARM_FEATURE_SVE_BF16)
822		__aio svbfloat16_t svreinterpret_bf16(svuint64_t op) {
	727	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svuint64_t op) {
823	728	return __builtin_sve_reinterpret_bf16_u64(op);
824	729	}
825	730
826		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
827		#if defined(__ARM_FEATURE_SVE_BF16)
828		__aio svbfloat16_t svreinterpret_bf16(svfloat16_t op) {
	731	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svfloat16_t op) {
829	732	return __builtin_sve_reinterpret_bf16_f16(op);
830	733	}
831	734
832		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
833		#if defined(__ARM_FEATURE_SVE_BF16)
834		__aio svbfloat16_t svreinterpret_bf16(svbfloat16_t op) {
	735	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svbfloat16_t op) {
835	736	return __builtin_sve_reinterpret_bf16_bf16(op);
836	737	}
837	738
838		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
839		#if defined(__ARM_FEATURE_SVE_BF16)
840		__aio svbfloat16_t svreinterpret_bf16(svfloat32_t op) {
	739	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svfloat32_t op) {
841	740	return __builtin_sve_reinterpret_bf16_f32(op);
842	741	}
843	742
844		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
845		#if defined(__ARM_FEATURE_SVE_BF16)
846		__aio svbfloat16_t svreinterpret_bf16(svfloat64_t op) {
	743	__aio __attribute__((target("sve"))) svbfloat16_t svreinterpret_bf16(svfloat64_t op) {
847	744	return __builtin_sve_reinterpret_bf16_f64(op);
848	745	}
849	746
850		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
851		__aio svfloat32_t svreinterpret_f32(svint8_t op) {
	747	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svint8_t op) {
852	748	return __builtin_sve_reinterpret_f32_s8(op);
853	749	}
854	750
855		__aio svfloat32_t svreinterpret_f32(svint16_t op) {
	751	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svint16_t op) {
856	752	return __builtin_sve_reinterpret_f32_s16(op);
857	753	}
858	754
859		__aio svfloat32_t svreinterpret_f32(svint32_t op) {
	755	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svint32_t op) {
860	756	return __builtin_sve_reinterpret_f32_s32(op);
861	757	}
862	758
863		__aio svfloat32_t svreinterpret_f32(svint64_t op) {
	759	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svint64_t op) {
864	760	return __builtin_sve_reinterpret_f32_s64(op);
865	761	}
866	762
867		__aio svfloat32_t svreinterpret_f32(svuint8_t op) {
	763	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svuint8_t op) {
868	764	return __builtin_sve_reinterpret_f32_u8(op);
869	765	}
870	766
871		__aio svfloat32_t svreinterpret_f32(svuint16_t op) {
	767	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svuint16_t op) {
872	768	return __builtin_sve_reinterpret_f32_u16(op);
873	769	}
874	770
875		__aio svfloat32_t svreinterpret_f32(svuint32_t op) {
	771	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svuint32_t op) {
876	772	return __builtin_sve_reinterpret_f32_u32(op);
877	773	}
878	774
879		__aio svfloat32_t svreinterpret_f32(svuint64_t op) {
	775	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svuint64_t op) {
880	776	return __builtin_sve_reinterpret_f32_u64(op);
881	777	}
882	778
883		__aio svfloat32_t svreinterpret_f32(svfloat16_t op) {
	779	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svfloat16_t op) {
884	780	return __builtin_sve_reinterpret_f32_f16(op);
885	781	}
886	782
887		#if defined(__ARM_FEATURE_SVE_BF16)
888		__aio svfloat32_t svreinterpret_f32(svbfloat16_t op) {
	783	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svbfloat16_t op) {
889	784	return __builtin_sve_reinterpret_f32_bf16(op);
890	785	}
891	786
892		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
893		__aio svfloat32_t svreinterpret_f32(svfloat32_t op) {
	787	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svfloat32_t op) {
894	788	return __builtin_sve_reinterpret_f32_f32(op);
895	789	}
896	790
897		__aio svfloat32_t svreinterpret_f32(svfloat64_t op) {
	791	__aio __attribute__((target("sve"))) svfloat32_t svreinterpret_f32(svfloat64_t op) {
898	792	return __builtin_sve_reinterpret_f32_f64(op);
899	793	}
900	794
901		__aio svfloat64_t svreinterpret_f64(svint8_t op) {
	795	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svint8_t op) {
902	796	return __builtin_sve_reinterpret_f64_s8(op);
903	797	}
904	798
905		__aio svfloat64_t svreinterpret_f64(svint16_t op) {
	799	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svint16_t op) {
906	800	return __builtin_sve_reinterpret_f64_s16(op);
907	801	}
908	802
909		__aio svfloat64_t svreinterpret_f64(svint32_t op) {
	803	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svint32_t op) {
910	804	return __builtin_sve_reinterpret_f64_s32(op);
911	805	}
912	806
913		__aio svfloat64_t svreinterpret_f64(svint64_t op) {
	807	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svint64_t op) {
914	808	return __builtin_sve_reinterpret_f64_s64(op);
915	809	}
916	810
917		__aio svfloat64_t svreinterpret_f64(svuint8_t op) {
	811	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svuint8_t op) {
918	812	return __builtin_sve_reinterpret_f64_u8(op);
919	813	}
920	814
921		__aio svfloat64_t svreinterpret_f64(svuint16_t op) {
	815	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svuint16_t op) {
922	816	return __builtin_sve_reinterpret_f64_u16(op);
923	817	}
924	818
925		__aio svfloat64_t svreinterpret_f64(svuint32_t op) {
	819	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svuint32_t op) {
926	820	return __builtin_sve_reinterpret_f64_u32(op);
927	821	}
928	822
929		__aio svfloat64_t svreinterpret_f64(svuint64_t op) {
	823	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svuint64_t op) {
930	824	return __builtin_sve_reinterpret_f64_u64(op);
931	825	}
932	826
933		__aio svfloat64_t svreinterpret_f64(svfloat16_t op) {
	827	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svfloat16_t op) {
934	828	return __builtin_sve_reinterpret_f64_f16(op);
935	829	}
936	830
937		#if defined(__ARM_FEATURE_SVE_BF16)
938		__aio svfloat64_t svreinterpret_f64(svbfloat16_t op) {
	831	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svbfloat16_t op) {
939	832	return __builtin_sve_reinterpret_f64_bf16(op);
940	833	}
941	834
942		#endif /* #if defined(__ARM_FEATURE_SVE_BF16) */
943		__aio svfloat64_t svreinterpret_f64(svfloat32_t op) {
	835	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svfloat32_t op) {
944	836	return __builtin_sve_reinterpret_f64_f32(op);
945	837	}
946	838
947		__aio svfloat64_t svreinterpret_f64(svfloat64_t op) {
	839	__aio __attribute__((target("sve"))) svfloat64_t svreinterpret_f64(svfloat64_t op) {
948	840	return __builtin_sve_reinterpret_f64_f64(op);
949	841	}
950	842
...	...	@@ -15698,7617 +15590,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s64)))
15698	15590	svint64_t svzip2(svint64_t, svint64_t);
15699	15591	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s16)))
15700	15592	svint16_t svzip2(svint16_t, svint16_t);
15701
15702		#if defined (__ARM_FEATURE_SVE2_BITPERM)
15703		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u8)))
15704		svuint8_t svbdep_n_u8(svuint8_t, uint8_t);
15705		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u32)))
15706		svuint32_t svbdep_n_u32(svuint32_t, uint32_t);
15707		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u64)))
15708		svuint64_t svbdep_n_u64(svuint64_t, uint64_t);
15709		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u16)))
15710		svuint16_t svbdep_n_u16(svuint16_t, uint16_t);
15711		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u8)))
15712		svuint8_t svbdep_u8(svuint8_t, svuint8_t);
15713		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u32)))
15714		svuint32_t svbdep_u32(svuint32_t, svuint32_t);
15715		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u64)))
15716		svuint64_t svbdep_u64(svuint64_t, svuint64_t);
15717		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u16)))
15718		svuint16_t svbdep_u16(svuint16_t, svuint16_t);
15719		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u8)))
15720		svuint8_t svbext_n_u8(svuint8_t, uint8_t);
15721		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u32)))
15722		svuint32_t svbext_n_u32(svuint32_t, uint32_t);
15723		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u64)))
15724		svuint64_t svbext_n_u64(svuint64_t, uint64_t);
15725		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u16)))
15726		svuint16_t svbext_n_u16(svuint16_t, uint16_t);
15727		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u8)))
15728		svuint8_t svbext_u8(svuint8_t, svuint8_t);
15729		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u32)))
15730		svuint32_t svbext_u32(svuint32_t, svuint32_t);
15731		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u64)))
15732		svuint64_t svbext_u64(svuint64_t, svuint64_t);
15733		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u16)))
15734		svuint16_t svbext_u16(svuint16_t, svuint16_t);
15735		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u8)))
15736		svuint8_t svbgrp_n_u8(svuint8_t, uint8_t);
15737		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u32)))
15738		svuint32_t svbgrp_n_u32(svuint32_t, uint32_t);
15739		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u64)))
15740		svuint64_t svbgrp_n_u64(svuint64_t, uint64_t);
15741		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u16)))
15742		svuint16_t svbgrp_n_u16(svuint16_t, uint16_t);
15743		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u8)))
15744		svuint8_t svbgrp_u8(svuint8_t, svuint8_t);
15745		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u32)))
15746		svuint32_t svbgrp_u32(svuint32_t, svuint32_t);
15747		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u64)))
15748		svuint64_t svbgrp_u64(svuint64_t, svuint64_t);
15749		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u16)))
15750		svuint16_t svbgrp_u16(svuint16_t, svuint16_t);
15751		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u8)))
15752		svuint8_t svbdep(svuint8_t, uint8_t);
15753		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u32)))
15754		svuint32_t svbdep(svuint32_t, uint32_t);
15755		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u64)))
15756		svuint64_t svbdep(svuint64_t, uint64_t);
15757		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u16)))
15758		svuint16_t svbdep(svuint16_t, uint16_t);
15759		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u8)))
15760		svuint8_t svbdep(svuint8_t, svuint8_t);
15761		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u32)))
15762		svuint32_t svbdep(svuint32_t, svuint32_t);
15763		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u64)))
15764		svuint64_t svbdep(svuint64_t, svuint64_t);
15765		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u16)))
15766		svuint16_t svbdep(svuint16_t, svuint16_t);
15767		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u8)))
15768		svuint8_t svbext(svuint8_t, uint8_t);
15769		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u32)))
15770		svuint32_t svbext(svuint32_t, uint32_t);
15771		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u64)))
15772		svuint64_t svbext(svuint64_t, uint64_t);
15773		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u16)))
15774		svuint16_t svbext(svuint16_t, uint16_t);
15775		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u8)))
15776		svuint8_t svbext(svuint8_t, svuint8_t);
15777		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u32)))
15778		svuint32_t svbext(svuint32_t, svuint32_t);
15779		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u64)))
15780		svuint64_t svbext(svuint64_t, svuint64_t);
15781		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u16)))
15782		svuint16_t svbext(svuint16_t, svuint16_t);
15783		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u8)))
15784		svuint8_t svbgrp(svuint8_t, uint8_t);
15785		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u32)))
15786		svuint32_t svbgrp(svuint32_t, uint32_t);
15787		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u64)))
15788		svuint64_t svbgrp(svuint64_t, uint64_t);
15789		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u16)))
15790		svuint16_t svbgrp(svuint16_t, uint16_t);
15791		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u8)))
15792		svuint8_t svbgrp(svuint8_t, svuint8_t);
15793		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u32)))
15794		svuint32_t svbgrp(svuint32_t, svuint32_t);
15795		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u64)))
15796		svuint64_t svbgrp(svuint64_t, svuint64_t);
15797		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u16)))
15798		svuint16_t svbgrp(svuint16_t, svuint16_t);
15799		#endif //defined (__ARM_FEATURE_SVE2_BITPERM)
15800
15801		#if defined(__ARM_FEATURE_SVE2)
15802		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s8)))
15803		svint8_t svaba_n_s8(svint8_t, svint8_t, int8_t);
15804		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s32)))
15805		svint32_t svaba_n_s32(svint32_t, svint32_t, int32_t);
15806		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s64)))
15807		svint64_t svaba_n_s64(svint64_t, svint64_t, int64_t);
15808		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s16)))
15809		svint16_t svaba_n_s16(svint16_t, svint16_t, int16_t);
15810		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u8)))
15811		svuint8_t svaba_n_u8(svuint8_t, svuint8_t, uint8_t);
15812		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u32)))
15813		svuint32_t svaba_n_u32(svuint32_t, svuint32_t, uint32_t);
15814		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u64)))
15815		svuint64_t svaba_n_u64(svuint64_t, svuint64_t, uint64_t);
15816		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u16)))
15817		svuint16_t svaba_n_u16(svuint16_t, svuint16_t, uint16_t);
15818		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s8)))
15819		svint8_t svaba_s8(svint8_t, svint8_t, svint8_t);
15820		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s32)))
15821		svint32_t svaba_s32(svint32_t, svint32_t, svint32_t);
15822		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s64)))
15823		svint64_t svaba_s64(svint64_t, svint64_t, svint64_t);
15824		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s16)))
15825		svint16_t svaba_s16(svint16_t, svint16_t, svint16_t);
15826		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u8)))
15827		svuint8_t svaba_u8(svuint8_t, svuint8_t, svuint8_t);
15828		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u32)))
15829		svuint32_t svaba_u32(svuint32_t, svuint32_t, svuint32_t);
15830		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u64)))
15831		svuint64_t svaba_u64(svuint64_t, svuint64_t, svuint64_t);
15832		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u16)))
15833		svuint16_t svaba_u16(svuint16_t, svuint16_t, svuint16_t);
15834		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s32)))
15835		svint32_t svabalb_n_s32(svint32_t, svint16_t, int16_t);
15836		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s64)))
15837		svint64_t svabalb_n_s64(svint64_t, svint32_t, int32_t);
15838		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s16)))
15839		svint16_t svabalb_n_s16(svint16_t, svint8_t, int8_t);
15840		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u32)))
15841		svuint32_t svabalb_n_u32(svuint32_t, svuint16_t, uint16_t);
15842		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u64)))
15843		svuint64_t svabalb_n_u64(svuint64_t, svuint32_t, uint32_t);
15844		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u16)))
15845		svuint16_t svabalb_n_u16(svuint16_t, svuint8_t, uint8_t);
15846		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s32)))
15847		svint32_t svabalb_s32(svint32_t, svint16_t, svint16_t);
15848		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s64)))
15849		svint64_t svabalb_s64(svint64_t, svint32_t, svint32_t);
15850		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s16)))
15851		svint16_t svabalb_s16(svint16_t, svint8_t, svint8_t);
15852		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u32)))
15853		svuint32_t svabalb_u32(svuint32_t, svuint16_t, svuint16_t);
15854		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u64)))
15855		svuint64_t svabalb_u64(svuint64_t, svuint32_t, svuint32_t);
15856		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u16)))
15857		svuint16_t svabalb_u16(svuint16_t, svuint8_t, svuint8_t);
15858		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s32)))
15859		svint32_t svabalt_n_s32(svint32_t, svint16_t, int16_t);
15860		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s64)))
15861		svint64_t svabalt_n_s64(svint64_t, svint32_t, int32_t);
15862		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s16)))
15863		svint16_t svabalt_n_s16(svint16_t, svint8_t, int8_t);
15864		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u32)))
15865		svuint32_t svabalt_n_u32(svuint32_t, svuint16_t, uint16_t);
15866		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u64)))
15867		svuint64_t svabalt_n_u64(svuint64_t, svuint32_t, uint32_t);
15868		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u16)))
15869		svuint16_t svabalt_n_u16(svuint16_t, svuint8_t, uint8_t);
15870		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s32)))
15871		svint32_t svabalt_s32(svint32_t, svint16_t, svint16_t);
15872		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s64)))
15873		svint64_t svabalt_s64(svint64_t, svint32_t, svint32_t);
15874		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s16)))
15875		svint16_t svabalt_s16(svint16_t, svint8_t, svint8_t);
15876		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u32)))
15877		svuint32_t svabalt_u32(svuint32_t, svuint16_t, svuint16_t);
15878		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u64)))
15879		svuint64_t svabalt_u64(svuint64_t, svuint32_t, svuint32_t);
15880		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u16)))
15881		svuint16_t svabalt_u16(svuint16_t, svuint8_t, svuint8_t);
15882		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s32)))
15883		svint32_t svabdlb_n_s32(svint16_t, int16_t);
15884		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s64)))
15885		svint64_t svabdlb_n_s64(svint32_t, int32_t);
15886		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s16)))
15887		svint16_t svabdlb_n_s16(svint8_t, int8_t);
15888		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u32)))
15889		svuint32_t svabdlb_n_u32(svuint16_t, uint16_t);
15890		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u64)))
15891		svuint64_t svabdlb_n_u64(svuint32_t, uint32_t);
15892		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u16)))
15893		svuint16_t svabdlb_n_u16(svuint8_t, uint8_t);
15894		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s32)))
15895		svint32_t svabdlb_s32(svint16_t, svint16_t);
15896		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s64)))
15897		svint64_t svabdlb_s64(svint32_t, svint32_t);
15898		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s16)))
15899		svint16_t svabdlb_s16(svint8_t, svint8_t);
15900		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u32)))
15901		svuint32_t svabdlb_u32(svuint16_t, svuint16_t);
15902		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u64)))
15903		svuint64_t svabdlb_u64(svuint32_t, svuint32_t);
15904		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u16)))
15905		svuint16_t svabdlb_u16(svuint8_t, svuint8_t);
15906		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s32)))
15907		svint32_t svabdlt_n_s32(svint16_t, int16_t);
15908		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s64)))
15909		svint64_t svabdlt_n_s64(svint32_t, int32_t);
15910		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s16)))
15911		svint16_t svabdlt_n_s16(svint8_t, int8_t);
15912		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u32)))
15913		svuint32_t svabdlt_n_u32(svuint16_t, uint16_t);
15914		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u64)))
15915		svuint64_t svabdlt_n_u64(svuint32_t, uint32_t);
15916		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u16)))
15917		svuint16_t svabdlt_n_u16(svuint8_t, uint8_t);
15918		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s32)))
15919		svint32_t svabdlt_s32(svint16_t, svint16_t);
15920		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s64)))
15921		svint64_t svabdlt_s64(svint32_t, svint32_t);
15922		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s16)))
15923		svint16_t svabdlt_s16(svint8_t, svint8_t);
15924		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u32)))
15925		svuint32_t svabdlt_u32(svuint16_t, svuint16_t);
15926		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u64)))
15927		svuint64_t svabdlt_u64(svuint32_t, svuint32_t);
15928		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u16)))
15929		svuint16_t svabdlt_u16(svuint8_t, svuint8_t);
15930		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_m)))
15931		svint32_t svadalp_s32_m(svbool_t, svint32_t, svint16_t);
15932		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_m)))
15933		svint64_t svadalp_s64_m(svbool_t, svint64_t, svint32_t);
15934		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_m)))
15935		svint16_t svadalp_s16_m(svbool_t, svint16_t, svint8_t);
15936		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_x)))
15937		svint32_t svadalp_s32_x(svbool_t, svint32_t, svint16_t);
15938		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_x)))
15939		svint64_t svadalp_s64_x(svbool_t, svint64_t, svint32_t);
15940		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_x)))
15941		svint16_t svadalp_s16_x(svbool_t, svint16_t, svint8_t);
15942		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_z)))
15943		svint32_t svadalp_s32_z(svbool_t, svint32_t, svint16_t);
15944		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_z)))
15945		svint64_t svadalp_s64_z(svbool_t, svint64_t, svint32_t);
15946		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_z)))
15947		svint16_t svadalp_s16_z(svbool_t, svint16_t, svint8_t);
15948		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_m)))
15949		svuint32_t svadalp_u32_m(svbool_t, svuint32_t, svuint16_t);
15950		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_m)))
15951		svuint64_t svadalp_u64_m(svbool_t, svuint64_t, svuint32_t);
15952		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_m)))
15953		svuint16_t svadalp_u16_m(svbool_t, svuint16_t, svuint8_t);
15954		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_x)))
15955		svuint32_t svadalp_u32_x(svbool_t, svuint32_t, svuint16_t);
15956		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_x)))
15957		svuint64_t svadalp_u64_x(svbool_t, svuint64_t, svuint32_t);
15958		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_x)))
15959		svuint16_t svadalp_u16_x(svbool_t, svuint16_t, svuint8_t);
15960		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_z)))
15961		svuint32_t svadalp_u32_z(svbool_t, svuint32_t, svuint16_t);
15962		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_z)))
15963		svuint64_t svadalp_u64_z(svbool_t, svuint64_t, svuint32_t);
15964		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_z)))
15965		svuint16_t svadalp_u16_z(svbool_t, svuint16_t, svuint8_t);
15966		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u32)))
15967		svuint32_t svadclb_n_u32(svuint32_t, svuint32_t, uint32_t);
15968		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u64)))
15969		svuint64_t svadclb_n_u64(svuint64_t, svuint64_t, uint64_t);
15970		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u32)))
15971		svuint32_t svadclb_u32(svuint32_t, svuint32_t, svuint32_t);
15972		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u64)))
15973		svuint64_t svadclb_u64(svuint64_t, svuint64_t, svuint64_t);
15974		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u32)))
15975		svuint32_t svadclt_n_u32(svuint32_t, svuint32_t, uint32_t);
15976		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u64)))
15977		svuint64_t svadclt_n_u64(svuint64_t, svuint64_t, uint64_t);
15978		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u32)))
15979		svuint32_t svadclt_u32(svuint32_t, svuint32_t, svuint32_t);
15980		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u64)))
15981		svuint64_t svadclt_u64(svuint64_t, svuint64_t, svuint64_t);
15982		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u32)))
15983		svuint16_t svaddhnb_n_u32(svuint32_t, uint32_t);
15984		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u64)))
15985		svuint32_t svaddhnb_n_u64(svuint64_t, uint64_t);
15986		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u16)))
15987		svuint8_t svaddhnb_n_u16(svuint16_t, uint16_t);
15988		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s32)))
15989		svint16_t svaddhnb_n_s32(svint32_t, int32_t);
15990		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s64)))
15991		svint32_t svaddhnb_n_s64(svint64_t, int64_t);
15992		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s16)))
15993		svint8_t svaddhnb_n_s16(svint16_t, int16_t);
15994		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u32)))
15995		svuint16_t svaddhnb_u32(svuint32_t, svuint32_t);
15996		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u64)))
15997		svuint32_t svaddhnb_u64(svuint64_t, svuint64_t);
15998		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u16)))
15999		svuint8_t svaddhnb_u16(svuint16_t, svuint16_t);
16000		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s32)))
16001		svint16_t svaddhnb_s32(svint32_t, svint32_t);
16002		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s64)))
16003		svint32_t svaddhnb_s64(svint64_t, svint64_t);
16004		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s16)))
16005		svint8_t svaddhnb_s16(svint16_t, svint16_t);
16006		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u32)))
16007		svuint16_t svaddhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
16008		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u64)))
16009		svuint32_t svaddhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
16010		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u16)))
16011		svuint8_t svaddhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
16012		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s32)))
16013		svint16_t svaddhnt_n_s32(svint16_t, svint32_t, int32_t);
16014		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s64)))
16015		svint32_t svaddhnt_n_s64(svint32_t, svint64_t, int64_t);
16016		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s16)))
16017		svint8_t svaddhnt_n_s16(svint8_t, svint16_t, int16_t);
16018		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u32)))
16019		svuint16_t svaddhnt_u32(svuint16_t, svuint32_t, svuint32_t);
16020		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u64)))
16021		svuint32_t svaddhnt_u64(svuint32_t, svuint64_t, svuint64_t);
16022		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u16)))
16023		svuint8_t svaddhnt_u16(svuint8_t, svuint16_t, svuint16_t);
16024		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s32)))
16025		svint16_t svaddhnt_s32(svint16_t, svint32_t, svint32_t);
16026		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s64)))
16027		svint32_t svaddhnt_s64(svint32_t, svint64_t, svint64_t);
16028		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s16)))
16029		svint8_t svaddhnt_s16(svint8_t, svint16_t, svint16_t);
16030		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s32)))
16031		svint32_t svaddlb_n_s32(svint16_t, int16_t);
16032		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s64)))
16033		svint64_t svaddlb_n_s64(svint32_t, int32_t);
16034		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s16)))
16035		svint16_t svaddlb_n_s16(svint8_t, int8_t);
16036		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u32)))
16037		svuint32_t svaddlb_n_u32(svuint16_t, uint16_t);
16038		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u64)))
16039		svuint64_t svaddlb_n_u64(svuint32_t, uint32_t);
16040		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u16)))
16041		svuint16_t svaddlb_n_u16(svuint8_t, uint8_t);
16042		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s32)))
16043		svint32_t svaddlb_s32(svint16_t, svint16_t);
16044		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s64)))
16045		svint64_t svaddlb_s64(svint32_t, svint32_t);
16046		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s16)))
16047		svint16_t svaddlb_s16(svint8_t, svint8_t);
16048		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u32)))
16049		svuint32_t svaddlb_u32(svuint16_t, svuint16_t);
16050		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u64)))
16051		svuint64_t svaddlb_u64(svuint32_t, svuint32_t);
16052		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u16)))
16053		svuint16_t svaddlb_u16(svuint8_t, svuint8_t);
16054		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s32)))
16055		svint32_t svaddlbt_n_s32(svint16_t, int16_t);
16056		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s64)))
16057		svint64_t svaddlbt_n_s64(svint32_t, int32_t);
16058		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s16)))
16059		svint16_t svaddlbt_n_s16(svint8_t, int8_t);
16060		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s32)))
16061		svint32_t svaddlbt_s32(svint16_t, svint16_t);
16062		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s64)))
16063		svint64_t svaddlbt_s64(svint32_t, svint32_t);
16064		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s16)))
16065		svint16_t svaddlbt_s16(svint8_t, svint8_t);
16066		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s32)))
16067		svint32_t svaddlt_n_s32(svint16_t, int16_t);
16068		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s64)))
16069		svint64_t svaddlt_n_s64(svint32_t, int32_t);
16070		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s16)))
16071		svint16_t svaddlt_n_s16(svint8_t, int8_t);
16072		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u32)))
16073		svuint32_t svaddlt_n_u32(svuint16_t, uint16_t);
16074		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u64)))
16075		svuint64_t svaddlt_n_u64(svuint32_t, uint32_t);
16076		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u16)))
16077		svuint16_t svaddlt_n_u16(svuint8_t, uint8_t);
16078		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s32)))
16079		svint32_t svaddlt_s32(svint16_t, svint16_t);
16080		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s64)))
16081		svint64_t svaddlt_s64(svint32_t, svint32_t);
16082		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s16)))
16083		svint16_t svaddlt_s16(svint8_t, svint8_t);
16084		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u32)))
16085		svuint32_t svaddlt_u32(svuint16_t, svuint16_t);
16086		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u64)))
16087		svuint64_t svaddlt_u64(svuint32_t, svuint32_t);
16088		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u16)))
16089		svuint16_t svaddlt_u16(svuint8_t, svuint8_t);
16090		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_m)))
16091		svfloat64_t svaddp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
16092		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_m)))
16093		svfloat32_t svaddp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
16094		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_m)))
16095		svfloat16_t svaddp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
16096		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_x)))
16097		svfloat64_t svaddp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
16098		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_x)))
16099		svfloat32_t svaddp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
16100		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_x)))
16101		svfloat16_t svaddp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
16102		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_m)))
16103		svuint8_t svaddp_u8_m(svbool_t, svuint8_t, svuint8_t);
16104		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_m)))
16105		svuint32_t svaddp_u32_m(svbool_t, svuint32_t, svuint32_t);
16106		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_m)))
16107		svuint64_t svaddp_u64_m(svbool_t, svuint64_t, svuint64_t);
16108		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_m)))
16109		svuint16_t svaddp_u16_m(svbool_t, svuint16_t, svuint16_t);
16110		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_m)))
16111		svint8_t svaddp_s8_m(svbool_t, svint8_t, svint8_t);
16112		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_m)))
16113		svint32_t svaddp_s32_m(svbool_t, svint32_t, svint32_t);
16114		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_m)))
16115		svint64_t svaddp_s64_m(svbool_t, svint64_t, svint64_t);
16116		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_m)))
16117		svint16_t svaddp_s16_m(svbool_t, svint16_t, svint16_t);
16118		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_x)))
16119		svuint8_t svaddp_u8_x(svbool_t, svuint8_t, svuint8_t);
16120		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_x)))
16121		svuint32_t svaddp_u32_x(svbool_t, svuint32_t, svuint32_t);
16122		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_x)))
16123		svuint64_t svaddp_u64_x(svbool_t, svuint64_t, svuint64_t);
16124		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_x)))
16125		svuint16_t svaddp_u16_x(svbool_t, svuint16_t, svuint16_t);
16126		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_x)))
16127		svint8_t svaddp_s8_x(svbool_t, svint8_t, svint8_t);
16128		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_x)))
16129		svint32_t svaddp_s32_x(svbool_t, svint32_t, svint32_t);
16130		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_x)))
16131		svint64_t svaddp_s64_x(svbool_t, svint64_t, svint64_t);
16132		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_x)))
16133		svint16_t svaddp_s16_x(svbool_t, svint16_t, svint16_t);
16134		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s32)))
16135		svint32_t svaddwb_n_s32(svint32_t, int16_t);
16136		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s64)))
16137		svint64_t svaddwb_n_s64(svint64_t, int32_t);
16138		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s16)))
16139		svint16_t svaddwb_n_s16(svint16_t, int8_t);
16140		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u32)))
16141		svuint32_t svaddwb_n_u32(svuint32_t, uint16_t);
16142		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u64)))
16143		svuint64_t svaddwb_n_u64(svuint64_t, uint32_t);
16144		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u16)))
16145		svuint16_t svaddwb_n_u16(svuint16_t, uint8_t);
16146		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s32)))
16147		svint32_t svaddwb_s32(svint32_t, svint16_t);
16148		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s64)))
16149		svint64_t svaddwb_s64(svint64_t, svint32_t);
16150		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s16)))
16151		svint16_t svaddwb_s16(svint16_t, svint8_t);
16152		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u32)))
16153		svuint32_t svaddwb_u32(svuint32_t, svuint16_t);
16154		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u64)))
16155		svuint64_t svaddwb_u64(svuint64_t, svuint32_t);
16156		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u16)))
16157		svuint16_t svaddwb_u16(svuint16_t, svuint8_t);
16158		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s32)))
16159		svint32_t svaddwt_n_s32(svint32_t, int16_t);
16160		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s64)))
16161		svint64_t svaddwt_n_s64(svint64_t, int32_t);
16162		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s16)))
16163		svint16_t svaddwt_n_s16(svint16_t, int8_t);
16164		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u32)))
16165		svuint32_t svaddwt_n_u32(svuint32_t, uint16_t);
16166		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u64)))
16167		svuint64_t svaddwt_n_u64(svuint64_t, uint32_t);
16168		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u16)))
16169		svuint16_t svaddwt_n_u16(svuint16_t, uint8_t);
16170		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s32)))
16171		svint32_t svaddwt_s32(svint32_t, svint16_t);
16172		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s64)))
16173		svint64_t svaddwt_s64(svint64_t, svint32_t);
16174		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s16)))
16175		svint16_t svaddwt_s16(svint16_t, svint8_t);
16176		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u32)))
16177		svuint32_t svaddwt_u32(svuint32_t, svuint16_t);
16178		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u64)))
16179		svuint64_t svaddwt_u64(svuint64_t, svuint32_t);
16180		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u16)))
16181		svuint16_t svaddwt_u16(svuint16_t, svuint8_t);
16182		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u8)))
16183		svuint8_t svbcax_n_u8(svuint8_t, svuint8_t, uint8_t);
16184		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u32)))
16185		svuint32_t svbcax_n_u32(svuint32_t, svuint32_t, uint32_t);
16186		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u64)))
16187		svuint64_t svbcax_n_u64(svuint64_t, svuint64_t, uint64_t);
16188		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u16)))
16189		svuint16_t svbcax_n_u16(svuint16_t, svuint16_t, uint16_t);
16190		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s8)))
16191		svint8_t svbcax_n_s8(svint8_t, svint8_t, int8_t);
16192		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s32)))
16193		svint32_t svbcax_n_s32(svint32_t, svint32_t, int32_t);
16194		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s64)))
16195		svint64_t svbcax_n_s64(svint64_t, svint64_t, int64_t);
16196		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s16)))
16197		svint16_t svbcax_n_s16(svint16_t, svint16_t, int16_t);
16198		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u8)))
16199		svuint8_t svbcax_u8(svuint8_t, svuint8_t, svuint8_t);
16200		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u32)))
16201		svuint32_t svbcax_u32(svuint32_t, svuint32_t, svuint32_t);
16202		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u64)))
16203		svuint64_t svbcax_u64(svuint64_t, svuint64_t, svuint64_t);
16204		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u16)))
16205		svuint16_t svbcax_u16(svuint16_t, svuint16_t, svuint16_t);
16206		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s8)))
16207		svint8_t svbcax_s8(svint8_t, svint8_t, svint8_t);
16208		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s32)))
16209		svint32_t svbcax_s32(svint32_t, svint32_t, svint32_t);
16210		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s64)))
16211		svint64_t svbcax_s64(svint64_t, svint64_t, svint64_t);
16212		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s16)))
16213		svint16_t svbcax_s16(svint16_t, svint16_t, svint16_t);
16214		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u8)))
16215		svuint8_t svbsl1n_n_u8(svuint8_t, svuint8_t, uint8_t);
16216		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u32)))
16217		svuint32_t svbsl1n_n_u32(svuint32_t, svuint32_t, uint32_t);
16218		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u64)))
16219		svuint64_t svbsl1n_n_u64(svuint64_t, svuint64_t, uint64_t);
16220		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u16)))
16221		svuint16_t svbsl1n_n_u16(svuint16_t, svuint16_t, uint16_t);
16222		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s8)))
16223		svint8_t svbsl1n_n_s8(svint8_t, svint8_t, int8_t);
16224		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s32)))
16225		svint32_t svbsl1n_n_s32(svint32_t, svint32_t, int32_t);
16226		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s64)))
16227		svint64_t svbsl1n_n_s64(svint64_t, svint64_t, int64_t);
16228		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s16)))
16229		svint16_t svbsl1n_n_s16(svint16_t, svint16_t, int16_t);
16230		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u8)))
16231		svuint8_t svbsl1n_u8(svuint8_t, svuint8_t, svuint8_t);
16232		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u32)))
16233		svuint32_t svbsl1n_u32(svuint32_t, svuint32_t, svuint32_t);
16234		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u64)))
16235		svuint64_t svbsl1n_u64(svuint64_t, svuint64_t, svuint64_t);
16236		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u16)))
16237		svuint16_t svbsl1n_u16(svuint16_t, svuint16_t, svuint16_t);
16238		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s8)))
16239		svint8_t svbsl1n_s8(svint8_t, svint8_t, svint8_t);
16240		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s32)))
16241		svint32_t svbsl1n_s32(svint32_t, svint32_t, svint32_t);
16242		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s64)))
16243		svint64_t svbsl1n_s64(svint64_t, svint64_t, svint64_t);
16244		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s16)))
16245		svint16_t svbsl1n_s16(svint16_t, svint16_t, svint16_t);
16246		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u8)))
16247		svuint8_t svbsl2n_n_u8(svuint8_t, svuint8_t, uint8_t);
16248		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u32)))
16249		svuint32_t svbsl2n_n_u32(svuint32_t, svuint32_t, uint32_t);
16250		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u64)))
16251		svuint64_t svbsl2n_n_u64(svuint64_t, svuint64_t, uint64_t);
16252		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u16)))
16253		svuint16_t svbsl2n_n_u16(svuint16_t, svuint16_t, uint16_t);
16254		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s8)))
16255		svint8_t svbsl2n_n_s8(svint8_t, svint8_t, int8_t);
16256		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s32)))
16257		svint32_t svbsl2n_n_s32(svint32_t, svint32_t, int32_t);
16258		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s64)))
16259		svint64_t svbsl2n_n_s64(svint64_t, svint64_t, int64_t);
16260		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s16)))
16261		svint16_t svbsl2n_n_s16(svint16_t, svint16_t, int16_t);
16262		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u8)))
16263		svuint8_t svbsl2n_u8(svuint8_t, svuint8_t, svuint8_t);
16264		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u32)))
16265		svuint32_t svbsl2n_u32(svuint32_t, svuint32_t, svuint32_t);
16266		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u64)))
16267		svuint64_t svbsl2n_u64(svuint64_t, svuint64_t, svuint64_t);
16268		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u16)))
16269		svuint16_t svbsl2n_u16(svuint16_t, svuint16_t, svuint16_t);
16270		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s8)))
16271		svint8_t svbsl2n_s8(svint8_t, svint8_t, svint8_t);
16272		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s32)))
16273		svint32_t svbsl2n_s32(svint32_t, svint32_t, svint32_t);
16274		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s64)))
16275		svint64_t svbsl2n_s64(svint64_t, svint64_t, svint64_t);
16276		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s16)))
16277		svint16_t svbsl2n_s16(svint16_t, svint16_t, svint16_t);
16278		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u8)))
16279		svuint8_t svbsl_n_u8(svuint8_t, svuint8_t, uint8_t);
16280		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u32)))
16281		svuint32_t svbsl_n_u32(svuint32_t, svuint32_t, uint32_t);
16282		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u64)))
16283		svuint64_t svbsl_n_u64(svuint64_t, svuint64_t, uint64_t);
16284		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u16)))
16285		svuint16_t svbsl_n_u16(svuint16_t, svuint16_t, uint16_t);
16286		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s8)))
16287		svint8_t svbsl_n_s8(svint8_t, svint8_t, int8_t);
16288		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s32)))
16289		svint32_t svbsl_n_s32(svint32_t, svint32_t, int32_t);
16290		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s64)))
16291		svint64_t svbsl_n_s64(svint64_t, svint64_t, int64_t);
16292		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s16)))
16293		svint16_t svbsl_n_s16(svint16_t, svint16_t, int16_t);
16294		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u8)))
16295		svuint8_t svbsl_u8(svuint8_t, svuint8_t, svuint8_t);
16296		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u32)))
16297		svuint32_t svbsl_u32(svuint32_t, svuint32_t, svuint32_t);
16298		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u64)))
16299		svuint64_t svbsl_u64(svuint64_t, svuint64_t, svuint64_t);
16300		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u16)))
16301		svuint16_t svbsl_u16(svuint16_t, svuint16_t, svuint16_t);
16302		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s8)))
16303		svint8_t svbsl_s8(svint8_t, svint8_t, svint8_t);
16304		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s32)))
16305		svint32_t svbsl_s32(svint32_t, svint32_t, svint32_t);
16306		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s64)))
16307		svint64_t svbsl_s64(svint64_t, svint64_t, svint64_t);
16308		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s16)))
16309		svint16_t svbsl_s16(svint16_t, svint16_t, svint16_t);
16310		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u8)))
16311		svuint8_t svcadd_u8(svuint8_t, svuint8_t, uint64_t);
16312		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u32)))
16313		svuint32_t svcadd_u32(svuint32_t, svuint32_t, uint64_t);
16314		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u64)))
16315		svuint64_t svcadd_u64(svuint64_t, svuint64_t, uint64_t);
16316		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u16)))
16317		svuint16_t svcadd_u16(svuint16_t, svuint16_t, uint64_t);
16318		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s8)))
16319		svint8_t svcadd_s8(svint8_t, svint8_t, uint64_t);
16320		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s32)))
16321		svint32_t svcadd_s32(svint32_t, svint32_t, uint64_t);
16322		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s64)))
16323		svint64_t svcadd_s64(svint64_t, svint64_t, uint64_t);
16324		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s16)))
16325		svint16_t svcadd_s16(svint16_t, svint16_t, uint64_t);
16326		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s32)))
16327		svint32_t svcdot_s32(svint32_t, svint8_t, svint8_t, uint64_t);
16328		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s64)))
16329		svint64_t svcdot_s64(svint64_t, svint16_t, svint16_t, uint64_t);
16330		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s32)))
16331		svint32_t svcdot_lane_s32(svint32_t, svint8_t, svint8_t, uint64_t, uint64_t);
16332		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s64)))
16333		svint64_t svcdot_lane_s64(svint64_t, svint16_t, svint16_t, uint64_t, uint64_t);
16334		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u8)))
16335		svuint8_t svcmla_u8(svuint8_t, svuint8_t, svuint8_t, uint64_t);
16336		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u32)))
16337		svuint32_t svcmla_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t);
16338		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u64)))
16339		svuint64_t svcmla_u64(svuint64_t, svuint64_t, svuint64_t, uint64_t);
16340		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u16)))
16341		svuint16_t svcmla_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t);
16342		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s8)))
16343		svint8_t svcmla_s8(svint8_t, svint8_t, svint8_t, uint64_t);
16344		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s32)))
16345		svint32_t svcmla_s32(svint32_t, svint32_t, svint32_t, uint64_t);
16346		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s64)))
16347		svint64_t svcmla_s64(svint64_t, svint64_t, svint64_t, uint64_t);
16348		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s16)))
16349		svint16_t svcmla_s16(svint16_t, svint16_t, svint16_t, uint64_t);
16350		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u32)))
16351		svuint32_t svcmla_lane_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t, uint64_t);
16352		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u16)))
16353		svuint16_t svcmla_lane_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t, uint64_t);
16354		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s32)))
16355		svint32_t svcmla_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
16356		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s16)))
16357		svint16_t svcmla_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
16358		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_m)))
16359		svfloat32_t svcvtlt_f32_f16_m(svfloat32_t, svbool_t, svfloat16_t);
16360		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_x)))
16361		svfloat32_t svcvtlt_f32_f16_x(svbool_t, svfloat16_t);
16362		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_m)))
16363		svfloat64_t svcvtlt_f64_f32_m(svfloat64_t, svbool_t, svfloat32_t);
16364		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_x)))
16365		svfloat64_t svcvtlt_f64_f32_x(svbool_t, svfloat32_t);
16366		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f16_f32_m)))
16367		svfloat16_t svcvtnt_f16_f32_m(svfloat16_t, svbool_t, svfloat32_t);
16368		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f32_f64_m)))
16369		svfloat32_t svcvtnt_f32_f64_m(svfloat32_t, svbool_t, svfloat64_t);
16370		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_m)))
16371		svfloat32_t svcvtx_f32_f64_m(svfloat32_t, svbool_t, svfloat64_t);
16372		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_x)))
16373		svfloat32_t svcvtx_f32_f64_x(svbool_t, svfloat64_t);
16374		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_z)))
16375		svfloat32_t svcvtx_f32_f64_z(svbool_t, svfloat64_t);
16376		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtxnt_f32_f64_m)))
16377		svfloat32_t svcvtxnt_f32_f64_m(svfloat32_t, svbool_t, svfloat64_t);
16378		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u8)))
16379		svuint8_t sveor3_n_u8(svuint8_t, svuint8_t, uint8_t);
16380		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u32)))
16381		svuint32_t sveor3_n_u32(svuint32_t, svuint32_t, uint32_t);
16382		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u64)))
16383		svuint64_t sveor3_n_u64(svuint64_t, svuint64_t, uint64_t);
16384		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u16)))
16385		svuint16_t sveor3_n_u16(svuint16_t, svuint16_t, uint16_t);
16386		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s8)))
16387		svint8_t sveor3_n_s8(svint8_t, svint8_t, int8_t);
16388		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s32)))
16389		svint32_t sveor3_n_s32(svint32_t, svint32_t, int32_t);
16390		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s64)))
16391		svint64_t sveor3_n_s64(svint64_t, svint64_t, int64_t);
16392		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s16)))
16393		svint16_t sveor3_n_s16(svint16_t, svint16_t, int16_t);
16394		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u8)))
16395		svuint8_t sveor3_u8(svuint8_t, svuint8_t, svuint8_t);
16396		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u32)))
16397		svuint32_t sveor3_u32(svuint32_t, svuint32_t, svuint32_t);
16398		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u64)))
16399		svuint64_t sveor3_u64(svuint64_t, svuint64_t, svuint64_t);
16400		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u16)))
16401		svuint16_t sveor3_u16(svuint16_t, svuint16_t, svuint16_t);
16402		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s8)))
16403		svint8_t sveor3_s8(svint8_t, svint8_t, svint8_t);
16404		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s32)))
16405		svint32_t sveor3_s32(svint32_t, svint32_t, svint32_t);
16406		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s64)))
16407		svint64_t sveor3_s64(svint64_t, svint64_t, svint64_t);
16408		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s16)))
16409		svint16_t sveor3_s16(svint16_t, svint16_t, svint16_t);
16410		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u8)))
16411		svuint8_t sveorbt_n_u8(svuint8_t, svuint8_t, uint8_t);
16412		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u32)))
16413		svuint32_t sveorbt_n_u32(svuint32_t, svuint32_t, uint32_t);
16414		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u64)))
16415		svuint64_t sveorbt_n_u64(svuint64_t, svuint64_t, uint64_t);
16416		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u16)))
16417		svuint16_t sveorbt_n_u16(svuint16_t, svuint16_t, uint16_t);
16418		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s8)))
16419		svint8_t sveorbt_n_s8(svint8_t, svint8_t, int8_t);
16420		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s32)))
16421		svint32_t sveorbt_n_s32(svint32_t, svint32_t, int32_t);
16422		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s64)))
16423		svint64_t sveorbt_n_s64(svint64_t, svint64_t, int64_t);
16424		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s16)))
16425		svint16_t sveorbt_n_s16(svint16_t, svint16_t, int16_t);
16426		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u8)))
16427		svuint8_t sveorbt_u8(svuint8_t, svuint8_t, svuint8_t);
16428		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u32)))
16429		svuint32_t sveorbt_u32(svuint32_t, svuint32_t, svuint32_t);
16430		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u64)))
16431		svuint64_t sveorbt_u64(svuint64_t, svuint64_t, svuint64_t);
16432		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u16)))
16433		svuint16_t sveorbt_u16(svuint16_t, svuint16_t, svuint16_t);
16434		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s8)))
16435		svint8_t sveorbt_s8(svint8_t, svint8_t, svint8_t);
16436		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s32)))
16437		svint32_t sveorbt_s32(svint32_t, svint32_t, svint32_t);
16438		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s64)))
16439		svint64_t sveorbt_s64(svint64_t, svint64_t, svint64_t);
16440		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s16)))
16441		svint16_t sveorbt_s16(svint16_t, svint16_t, svint16_t);
16442		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u8)))
16443		svuint8_t sveortb_n_u8(svuint8_t, svuint8_t, uint8_t);
16444		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u32)))
16445		svuint32_t sveortb_n_u32(svuint32_t, svuint32_t, uint32_t);
16446		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u64)))
16447		svuint64_t sveortb_n_u64(svuint64_t, svuint64_t, uint64_t);
16448		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u16)))
16449		svuint16_t sveortb_n_u16(svuint16_t, svuint16_t, uint16_t);
16450		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s8)))
16451		svint8_t sveortb_n_s8(svint8_t, svint8_t, int8_t);
16452		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s32)))
16453		svint32_t sveortb_n_s32(svint32_t, svint32_t, int32_t);
16454		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s64)))
16455		svint64_t sveortb_n_s64(svint64_t, svint64_t, int64_t);
16456		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s16)))
16457		svint16_t sveortb_n_s16(svint16_t, svint16_t, int16_t);
16458		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u8)))
16459		svuint8_t sveortb_u8(svuint8_t, svuint8_t, svuint8_t);
16460		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u32)))
16461		svuint32_t sveortb_u32(svuint32_t, svuint32_t, svuint32_t);
16462		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u64)))
16463		svuint64_t sveortb_u64(svuint64_t, svuint64_t, svuint64_t);
16464		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u16)))
16465		svuint16_t sveortb_u16(svuint16_t, svuint16_t, svuint16_t);
16466		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s8)))
16467		svint8_t sveortb_s8(svint8_t, svint8_t, svint8_t);
16468		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s32)))
16469		svint32_t sveortb_s32(svint32_t, svint32_t, svint32_t);
16470		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s64)))
16471		svint64_t sveortb_s64(svint64_t, svint64_t, svint64_t);
16472		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s16)))
16473		svint16_t sveortb_s16(svint16_t, svint16_t, svint16_t);
16474		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_m)))
16475		svint8_t svhadd_n_s8_m(svbool_t, svint8_t, int8_t);
16476		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_m)))
16477		svint32_t svhadd_n_s32_m(svbool_t, svint32_t, int32_t);
16478		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_m)))
16479		svint64_t svhadd_n_s64_m(svbool_t, svint64_t, int64_t);
16480		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_m)))
16481		svint16_t svhadd_n_s16_m(svbool_t, svint16_t, int16_t);
16482		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_x)))
16483		svint8_t svhadd_n_s8_x(svbool_t, svint8_t, int8_t);
16484		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_x)))
16485		svint32_t svhadd_n_s32_x(svbool_t, svint32_t, int32_t);
16486		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_x)))
16487		svint64_t svhadd_n_s64_x(svbool_t, svint64_t, int64_t);
16488		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_x)))
16489		svint16_t svhadd_n_s16_x(svbool_t, svint16_t, int16_t);
16490		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_z)))
16491		svint8_t svhadd_n_s8_z(svbool_t, svint8_t, int8_t);
16492		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_z)))
16493		svint32_t svhadd_n_s32_z(svbool_t, svint32_t, int32_t);
16494		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_z)))
16495		svint64_t svhadd_n_s64_z(svbool_t, svint64_t, int64_t);
16496		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_z)))
16497		svint16_t svhadd_n_s16_z(svbool_t, svint16_t, int16_t);
16498		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_m)))
16499		svuint8_t svhadd_n_u8_m(svbool_t, svuint8_t, uint8_t);
16500		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_m)))
16501		svuint32_t svhadd_n_u32_m(svbool_t, svuint32_t, uint32_t);
16502		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_m)))
16503		svuint64_t svhadd_n_u64_m(svbool_t, svuint64_t, uint64_t);
16504		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_m)))
16505		svuint16_t svhadd_n_u16_m(svbool_t, svuint16_t, uint16_t);
16506		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_x)))
16507		svuint8_t svhadd_n_u8_x(svbool_t, svuint8_t, uint8_t);
16508		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_x)))
16509		svuint32_t svhadd_n_u32_x(svbool_t, svuint32_t, uint32_t);
16510		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_x)))
16511		svuint64_t svhadd_n_u64_x(svbool_t, svuint64_t, uint64_t);
16512		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_x)))
16513		svuint16_t svhadd_n_u16_x(svbool_t, svuint16_t, uint16_t);
16514		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_z)))
16515		svuint8_t svhadd_n_u8_z(svbool_t, svuint8_t, uint8_t);
16516		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_z)))
16517		svuint32_t svhadd_n_u32_z(svbool_t, svuint32_t, uint32_t);
16518		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_z)))
16519		svuint64_t svhadd_n_u64_z(svbool_t, svuint64_t, uint64_t);
16520		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_z)))
16521		svuint16_t svhadd_n_u16_z(svbool_t, svuint16_t, uint16_t);
16522		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_m)))
16523		svint8_t svhadd_s8_m(svbool_t, svint8_t, svint8_t);
16524		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_m)))
16525		svint32_t svhadd_s32_m(svbool_t, svint32_t, svint32_t);
16526		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_m)))
16527		svint64_t svhadd_s64_m(svbool_t, svint64_t, svint64_t);
16528		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_m)))
16529		svint16_t svhadd_s16_m(svbool_t, svint16_t, svint16_t);
16530		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_x)))
16531		svint8_t svhadd_s8_x(svbool_t, svint8_t, svint8_t);
16532		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_x)))
16533		svint32_t svhadd_s32_x(svbool_t, svint32_t, svint32_t);
16534		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_x)))
16535		svint64_t svhadd_s64_x(svbool_t, svint64_t, svint64_t);
16536		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_x)))
16537		svint16_t svhadd_s16_x(svbool_t, svint16_t, svint16_t);
16538		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_z)))
16539		svint8_t svhadd_s8_z(svbool_t, svint8_t, svint8_t);
16540		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_z)))
16541		svint32_t svhadd_s32_z(svbool_t, svint32_t, svint32_t);
16542		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_z)))
16543		svint64_t svhadd_s64_z(svbool_t, svint64_t, svint64_t);
16544		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_z)))
16545		svint16_t svhadd_s16_z(svbool_t, svint16_t, svint16_t);
16546		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_m)))
16547		svuint8_t svhadd_u8_m(svbool_t, svuint8_t, svuint8_t);
16548		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_m)))
16549		svuint32_t svhadd_u32_m(svbool_t, svuint32_t, svuint32_t);
16550		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_m)))
16551		svuint64_t svhadd_u64_m(svbool_t, svuint64_t, svuint64_t);
16552		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_m)))
16553		svuint16_t svhadd_u16_m(svbool_t, svuint16_t, svuint16_t);
16554		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_x)))
16555		svuint8_t svhadd_u8_x(svbool_t, svuint8_t, svuint8_t);
16556		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_x)))
16557		svuint32_t svhadd_u32_x(svbool_t, svuint32_t, svuint32_t);
16558		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_x)))
16559		svuint64_t svhadd_u64_x(svbool_t, svuint64_t, svuint64_t);
16560		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_x)))
16561		svuint16_t svhadd_u16_x(svbool_t, svuint16_t, svuint16_t);
16562		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_z)))
16563		svuint8_t svhadd_u8_z(svbool_t, svuint8_t, svuint8_t);
16564		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_z)))
16565		svuint32_t svhadd_u32_z(svbool_t, svuint32_t, svuint32_t);
16566		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_z)))
16567		svuint64_t svhadd_u64_z(svbool_t, svuint64_t, svuint64_t);
16568		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_z)))
16569		svuint16_t svhadd_u16_z(svbool_t, svuint16_t, svuint16_t);
16570		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u32_z)))
16571		svuint32_t svhistcnt_u32_z(svbool_t, svuint32_t, svuint32_t);
16572		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u64_z)))
16573		svuint64_t svhistcnt_u64_z(svbool_t, svuint64_t, svuint64_t);
16574		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s32_z)))
16575		svuint32_t svhistcnt_s32_z(svbool_t, svint32_t, svint32_t);
16576		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s64_z)))
16577		svuint64_t svhistcnt_s64_z(svbool_t, svint64_t, svint64_t);
16578		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_u8)))
16579		svuint8_t svhistseg_u8(svuint8_t, svuint8_t);
16580		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_s8)))
16581		svuint8_t svhistseg_s8(svint8_t, svint8_t);
16582		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_m)))
16583		svint8_t svhsub_n_s8_m(svbool_t, svint8_t, int8_t);
16584		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_m)))
16585		svint32_t svhsub_n_s32_m(svbool_t, svint32_t, int32_t);
16586		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_m)))
16587		svint64_t svhsub_n_s64_m(svbool_t, svint64_t, int64_t);
16588		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_m)))
16589		svint16_t svhsub_n_s16_m(svbool_t, svint16_t, int16_t);
16590		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_x)))
16591		svint8_t svhsub_n_s8_x(svbool_t, svint8_t, int8_t);
16592		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_x)))
16593		svint32_t svhsub_n_s32_x(svbool_t, svint32_t, int32_t);
16594		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_x)))
16595		svint64_t svhsub_n_s64_x(svbool_t, svint64_t, int64_t);
16596		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_x)))
16597		svint16_t svhsub_n_s16_x(svbool_t, svint16_t, int16_t);
16598		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_z)))
16599		svint8_t svhsub_n_s8_z(svbool_t, svint8_t, int8_t);
16600		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_z)))
16601		svint32_t svhsub_n_s32_z(svbool_t, svint32_t, int32_t);
16602		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_z)))
16603		svint64_t svhsub_n_s64_z(svbool_t, svint64_t, int64_t);
16604		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_z)))
16605		svint16_t svhsub_n_s16_z(svbool_t, svint16_t, int16_t);
16606		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_m)))
16607		svuint8_t svhsub_n_u8_m(svbool_t, svuint8_t, uint8_t);
16608		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_m)))
16609		svuint32_t svhsub_n_u32_m(svbool_t, svuint32_t, uint32_t);
16610		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_m)))
16611		svuint64_t svhsub_n_u64_m(svbool_t, svuint64_t, uint64_t);
16612		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_m)))
16613		svuint16_t svhsub_n_u16_m(svbool_t, svuint16_t, uint16_t);
16614		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_x)))
16615		svuint8_t svhsub_n_u8_x(svbool_t, svuint8_t, uint8_t);
16616		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_x)))
16617		svuint32_t svhsub_n_u32_x(svbool_t, svuint32_t, uint32_t);
16618		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_x)))
16619		svuint64_t svhsub_n_u64_x(svbool_t, svuint64_t, uint64_t);
16620		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_x)))
16621		svuint16_t svhsub_n_u16_x(svbool_t, svuint16_t, uint16_t);
16622		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_z)))
16623		svuint8_t svhsub_n_u8_z(svbool_t, svuint8_t, uint8_t);
16624		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_z)))
16625		svuint32_t svhsub_n_u32_z(svbool_t, svuint32_t, uint32_t);
16626		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_z)))
16627		svuint64_t svhsub_n_u64_z(svbool_t, svuint64_t, uint64_t);
16628		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_z)))
16629		svuint16_t svhsub_n_u16_z(svbool_t, svuint16_t, uint16_t);
16630		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_m)))
16631		svint8_t svhsub_s8_m(svbool_t, svint8_t, svint8_t);
16632		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_m)))
16633		svint32_t svhsub_s32_m(svbool_t, svint32_t, svint32_t);
16634		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_m)))
16635		svint64_t svhsub_s64_m(svbool_t, svint64_t, svint64_t);
16636		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_m)))
16637		svint16_t svhsub_s16_m(svbool_t, svint16_t, svint16_t);
16638		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_x)))
16639		svint8_t svhsub_s8_x(svbool_t, svint8_t, svint8_t);
16640		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_x)))
16641		svint32_t svhsub_s32_x(svbool_t, svint32_t, svint32_t);
16642		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_x)))
16643		svint64_t svhsub_s64_x(svbool_t, svint64_t, svint64_t);
16644		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_x)))
16645		svint16_t svhsub_s16_x(svbool_t, svint16_t, svint16_t);
16646		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_z)))
16647		svint8_t svhsub_s8_z(svbool_t, svint8_t, svint8_t);
16648		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_z)))
16649		svint32_t svhsub_s32_z(svbool_t, svint32_t, svint32_t);
16650		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_z)))
16651		svint64_t svhsub_s64_z(svbool_t, svint64_t, svint64_t);
16652		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_z)))
16653		svint16_t svhsub_s16_z(svbool_t, svint16_t, svint16_t);
16654		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_m)))
16655		svuint8_t svhsub_u8_m(svbool_t, svuint8_t, svuint8_t);
16656		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_m)))
16657		svuint32_t svhsub_u32_m(svbool_t, svuint32_t, svuint32_t);
16658		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_m)))
16659		svuint64_t svhsub_u64_m(svbool_t, svuint64_t, svuint64_t);
16660		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_m)))
16661		svuint16_t svhsub_u16_m(svbool_t, svuint16_t, svuint16_t);
16662		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_x)))
16663		svuint8_t svhsub_u8_x(svbool_t, svuint8_t, svuint8_t);
16664		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_x)))
16665		svuint32_t svhsub_u32_x(svbool_t, svuint32_t, svuint32_t);
16666		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_x)))
16667		svuint64_t svhsub_u64_x(svbool_t, svuint64_t, svuint64_t);
16668		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_x)))
16669		svuint16_t svhsub_u16_x(svbool_t, svuint16_t, svuint16_t);
16670		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_z)))
16671		svuint8_t svhsub_u8_z(svbool_t, svuint8_t, svuint8_t);
16672		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_z)))
16673		svuint32_t svhsub_u32_z(svbool_t, svuint32_t, svuint32_t);
16674		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_z)))
16675		svuint64_t svhsub_u64_z(svbool_t, svuint64_t, svuint64_t);
16676		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_z)))
16677		svuint16_t svhsub_u16_z(svbool_t, svuint16_t, svuint16_t);
16678		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_m)))
16679		svint8_t svhsubr_n_s8_m(svbool_t, svint8_t, int8_t);
16680		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_m)))
16681		svint32_t svhsubr_n_s32_m(svbool_t, svint32_t, int32_t);
16682		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_m)))
16683		svint64_t svhsubr_n_s64_m(svbool_t, svint64_t, int64_t);
16684		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_m)))
16685		svint16_t svhsubr_n_s16_m(svbool_t, svint16_t, int16_t);
16686		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_x)))
16687		svint8_t svhsubr_n_s8_x(svbool_t, svint8_t, int8_t);
16688		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_x)))
16689		svint32_t svhsubr_n_s32_x(svbool_t, svint32_t, int32_t);
16690		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_x)))
16691		svint64_t svhsubr_n_s64_x(svbool_t, svint64_t, int64_t);
16692		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_x)))
16693		svint16_t svhsubr_n_s16_x(svbool_t, svint16_t, int16_t);
16694		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_z)))
16695		svint8_t svhsubr_n_s8_z(svbool_t, svint8_t, int8_t);
16696		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_z)))
16697		svint32_t svhsubr_n_s32_z(svbool_t, svint32_t, int32_t);
16698		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_z)))
16699		svint64_t svhsubr_n_s64_z(svbool_t, svint64_t, int64_t);
16700		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_z)))
16701		svint16_t svhsubr_n_s16_z(svbool_t, svint16_t, int16_t);
16702		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_m)))
16703		svuint8_t svhsubr_n_u8_m(svbool_t, svuint8_t, uint8_t);
16704		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_m)))
16705		svuint32_t svhsubr_n_u32_m(svbool_t, svuint32_t, uint32_t);
16706		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_m)))
16707		svuint64_t svhsubr_n_u64_m(svbool_t, svuint64_t, uint64_t);
16708		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_m)))
16709		svuint16_t svhsubr_n_u16_m(svbool_t, svuint16_t, uint16_t);
16710		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_x)))
16711		svuint8_t svhsubr_n_u8_x(svbool_t, svuint8_t, uint8_t);
16712		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_x)))
16713		svuint32_t svhsubr_n_u32_x(svbool_t, svuint32_t, uint32_t);
16714		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_x)))
16715		svuint64_t svhsubr_n_u64_x(svbool_t, svuint64_t, uint64_t);
16716		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_x)))
16717		svuint16_t svhsubr_n_u16_x(svbool_t, svuint16_t, uint16_t);
16718		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_z)))
16719		svuint8_t svhsubr_n_u8_z(svbool_t, svuint8_t, uint8_t);
16720		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_z)))
16721		svuint32_t svhsubr_n_u32_z(svbool_t, svuint32_t, uint32_t);
16722		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_z)))
16723		svuint64_t svhsubr_n_u64_z(svbool_t, svuint64_t, uint64_t);
16724		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_z)))
16725		svuint16_t svhsubr_n_u16_z(svbool_t, svuint16_t, uint16_t);
16726		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_m)))
16727		svint8_t svhsubr_s8_m(svbool_t, svint8_t, svint8_t);
16728		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_m)))
16729		svint32_t svhsubr_s32_m(svbool_t, svint32_t, svint32_t);
16730		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_m)))
16731		svint64_t svhsubr_s64_m(svbool_t, svint64_t, svint64_t);
16732		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_m)))
16733		svint16_t svhsubr_s16_m(svbool_t, svint16_t, svint16_t);
16734		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_x)))
16735		svint8_t svhsubr_s8_x(svbool_t, svint8_t, svint8_t);
16736		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_x)))
16737		svint32_t svhsubr_s32_x(svbool_t, svint32_t, svint32_t);
16738		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_x)))
16739		svint64_t svhsubr_s64_x(svbool_t, svint64_t, svint64_t);
16740		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_x)))
16741		svint16_t svhsubr_s16_x(svbool_t, svint16_t, svint16_t);
16742		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_z)))
16743		svint8_t svhsubr_s8_z(svbool_t, svint8_t, svint8_t);
16744		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_z)))
16745		svint32_t svhsubr_s32_z(svbool_t, svint32_t, svint32_t);
16746		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_z)))
16747		svint64_t svhsubr_s64_z(svbool_t, svint64_t, svint64_t);
16748		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_z)))
16749		svint16_t svhsubr_s16_z(svbool_t, svint16_t, svint16_t);
16750		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_m)))
16751		svuint8_t svhsubr_u8_m(svbool_t, svuint8_t, svuint8_t);
16752		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_m)))
16753		svuint32_t svhsubr_u32_m(svbool_t, svuint32_t, svuint32_t);
16754		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_m)))
16755		svuint64_t svhsubr_u64_m(svbool_t, svuint64_t, svuint64_t);
16756		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_m)))
16757		svuint16_t svhsubr_u16_m(svbool_t, svuint16_t, svuint16_t);
16758		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_x)))
16759		svuint8_t svhsubr_u8_x(svbool_t, svuint8_t, svuint8_t);
16760		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_x)))
16761		svuint32_t svhsubr_u32_x(svbool_t, svuint32_t, svuint32_t);
16762		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_x)))
16763		svuint64_t svhsubr_u64_x(svbool_t, svuint64_t, svuint64_t);
16764		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_x)))
16765		svuint16_t svhsubr_u16_x(svbool_t, svuint16_t, svuint16_t);
16766		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_z)))
16767		svuint8_t svhsubr_u8_z(svbool_t, svuint8_t, svuint8_t);
16768		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_z)))
16769		svuint32_t svhsubr_u32_z(svbool_t, svuint32_t, svuint32_t);
16770		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_z)))
16771		svuint64_t svhsubr_u64_z(svbool_t, svuint64_t, svuint64_t);
16772		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_z)))
16773		svuint16_t svhsubr_u16_z(svbool_t, svuint16_t, svuint16_t);
16774		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_u32)))
16775		svuint32_t svldnt1_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
16776		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_u64)))
16777		svuint64_t svldnt1_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
16778		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_f64)))
16779		svfloat64_t svldnt1_gather_u64base_index_f64(svbool_t, svuint64_t, int64_t);
16780		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_f32)))
16781		svfloat32_t svldnt1_gather_u32base_index_f32(svbool_t, svuint32_t, int64_t);
16782		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_s32)))
16783		svint32_t svldnt1_gather_u32base_index_s32(svbool_t, svuint32_t, int64_t);
16784		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_s64)))
16785		svint64_t svldnt1_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
16786		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_u32)))
16787		svuint32_t svldnt1_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
16788		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_u64)))
16789		svuint64_t svldnt1_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
16790		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_f64)))
16791		svfloat64_t svldnt1_gather_u64base_offset_f64(svbool_t, svuint64_t, int64_t);
16792		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_f32)))
16793		svfloat32_t svldnt1_gather_u32base_offset_f32(svbool_t, svuint32_t, int64_t);
16794		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_s32)))
16795		svint32_t svldnt1_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
16796		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_s64)))
16797		svint64_t svldnt1_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
16798		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_u32)))
16799		svuint32_t svldnt1_gather_u32base_u32(svbool_t, svuint32_t);
16800		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_u64)))
16801		svuint64_t svldnt1_gather_u64base_u64(svbool_t, svuint64_t);
16802		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_f64)))
16803		svfloat64_t svldnt1_gather_u64base_f64(svbool_t, svuint64_t);
16804		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_f32)))
16805		svfloat32_t svldnt1_gather_u32base_f32(svbool_t, svuint32_t);
16806		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_s32)))
16807		svint32_t svldnt1_gather_u32base_s32(svbool_t, svuint32_t);
16808		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_s64)))
16809		svint64_t svldnt1_gather_u64base_s64(svbool_t, svuint64_t);
16810		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_u64)))
16811		svuint64_t svldnt1_gather_s64index_u64(svbool_t, uint64_t const *, svint64_t);
16812		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_f64)))
16813		svfloat64_t svldnt1_gather_s64index_f64(svbool_t, float64_t const *, svint64_t);
16814		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_s64)))
16815		svint64_t svldnt1_gather_s64index_s64(svbool_t, int64_t const *, svint64_t);
16816		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_u64)))
16817		svuint64_t svldnt1_gather_u64index_u64(svbool_t, uint64_t const *, svuint64_t);
16818		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_f64)))
16819		svfloat64_t svldnt1_gather_u64index_f64(svbool_t, float64_t const *, svuint64_t);
16820		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_s64)))
16821		svint64_t svldnt1_gather_u64index_s64(svbool_t, int64_t const *, svuint64_t);
16822		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_u32)))
16823		svuint32_t svldnt1_gather_u32offset_u32(svbool_t, uint32_t const *, svuint32_t);
16824		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_f32)))
16825		svfloat32_t svldnt1_gather_u32offset_f32(svbool_t, float32_t const *, svuint32_t);
16826		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_s32)))
16827		svint32_t svldnt1_gather_u32offset_s32(svbool_t, int32_t const *, svuint32_t);
16828		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_u64)))
16829		svuint64_t svldnt1_gather_s64offset_u64(svbool_t, uint64_t const *, svint64_t);
16830		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_f64)))
16831		svfloat64_t svldnt1_gather_s64offset_f64(svbool_t, float64_t const *, svint64_t);
16832		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_s64)))
16833		svint64_t svldnt1_gather_s64offset_s64(svbool_t, int64_t const *, svint64_t);
16834		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_u64)))
16835		svuint64_t svldnt1_gather_u64offset_u64(svbool_t, uint64_t const *, svuint64_t);
16836		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_f64)))
16837		svfloat64_t svldnt1_gather_u64offset_f64(svbool_t, float64_t const *, svuint64_t);
16838		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_s64)))
16839		svint64_t svldnt1_gather_u64offset_s64(svbool_t, int64_t const *, svuint64_t);
16840		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_u32)))
16841		svuint32_t svldnt1sb_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
16842		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_u64)))
16843		svuint64_t svldnt1sb_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
16844		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_s32)))
16845		svint32_t svldnt1sb_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
16846		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_s64)))
16847		svint64_t svldnt1sb_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
16848		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_u32)))
16849		svuint32_t svldnt1sb_gather_u32base_u32(svbool_t, svuint32_t);
16850		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_u64)))
16851		svuint64_t svldnt1sb_gather_u64base_u64(svbool_t, svuint64_t);
16852		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_s32)))
16853		svint32_t svldnt1sb_gather_u32base_s32(svbool_t, svuint32_t);
16854		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_s64)))
16855		svint64_t svldnt1sb_gather_u64base_s64(svbool_t, svuint64_t);
16856		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_u32)))
16857		svuint32_t svldnt1sb_gather_u32offset_u32(svbool_t, int8_t const *, svuint32_t);
16858		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_s32)))
16859		svint32_t svldnt1sb_gather_u32offset_s32(svbool_t, int8_t const *, svuint32_t);
16860		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_u64)))
16861		svuint64_t svldnt1sb_gather_s64offset_u64(svbool_t, int8_t const *, svint64_t);
16862		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_s64)))
16863		svint64_t svldnt1sb_gather_s64offset_s64(svbool_t, int8_t const *, svint64_t);
16864		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_u64)))
16865		svuint64_t svldnt1sb_gather_u64offset_u64(svbool_t, int8_t const *, svuint64_t);
16866		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_s64)))
16867		svint64_t svldnt1sb_gather_u64offset_s64(svbool_t, int8_t const *, svuint64_t);
16868		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_u32)))
16869		svuint32_t svldnt1sh_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
16870		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_u64)))
16871		svuint64_t svldnt1sh_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
16872		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_s32)))
16873		svint32_t svldnt1sh_gather_u32base_index_s32(svbool_t, svuint32_t, int64_t);
16874		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_s64)))
16875		svint64_t svldnt1sh_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
16876		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_u32)))
16877		svuint32_t svldnt1sh_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
16878		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_u64)))
16879		svuint64_t svldnt1sh_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
16880		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_s32)))
16881		svint32_t svldnt1sh_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
16882		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_s64)))
16883		svint64_t svldnt1sh_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
16884		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_u32)))
16885		svuint32_t svldnt1sh_gather_u32base_u32(svbool_t, svuint32_t);
16886		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_u64)))
16887		svuint64_t svldnt1sh_gather_u64base_u64(svbool_t, svuint64_t);
16888		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_s32)))
16889		svint32_t svldnt1sh_gather_u32base_s32(svbool_t, svuint32_t);
16890		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_s64)))
16891		svint64_t svldnt1sh_gather_u64base_s64(svbool_t, svuint64_t);
16892		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_u64)))
16893		svuint64_t svldnt1sh_gather_s64index_u64(svbool_t, int16_t const *, svint64_t);
16894		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_s64)))
16895		svint64_t svldnt1sh_gather_s64index_s64(svbool_t, int16_t const *, svint64_t);
16896		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_u64)))
16897		svuint64_t svldnt1sh_gather_u64index_u64(svbool_t, int16_t const *, svuint64_t);
16898		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_s64)))
16899		svint64_t svldnt1sh_gather_u64index_s64(svbool_t, int16_t const *, svuint64_t);
16900		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_u32)))
16901		svuint32_t svldnt1sh_gather_u32offset_u32(svbool_t, int16_t const *, svuint32_t);
16902		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_s32)))
16903		svint32_t svldnt1sh_gather_u32offset_s32(svbool_t, int16_t const *, svuint32_t);
16904		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_u64)))
16905		svuint64_t svldnt1sh_gather_s64offset_u64(svbool_t, int16_t const *, svint64_t);
16906		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_s64)))
16907		svint64_t svldnt1sh_gather_s64offset_s64(svbool_t, int16_t const *, svint64_t);
16908		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_u64)))
16909		svuint64_t svldnt1sh_gather_u64offset_u64(svbool_t, int16_t const *, svuint64_t);
16910		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_s64)))
16911		svint64_t svldnt1sh_gather_u64offset_s64(svbool_t, int16_t const *, svuint64_t);
16912		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_u64)))
16913		svuint64_t svldnt1sw_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
16914		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_s64)))
16915		svint64_t svldnt1sw_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
16916		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_u64)))
16917		svuint64_t svldnt1sw_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
16918		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_s64)))
16919		svint64_t svldnt1sw_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
16920		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_u64)))
16921		svuint64_t svldnt1sw_gather_u64base_u64(svbool_t, svuint64_t);
16922		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_s64)))
16923		svint64_t svldnt1sw_gather_u64base_s64(svbool_t, svuint64_t);
16924		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_u64)))
16925		svuint64_t svldnt1sw_gather_s64index_u64(svbool_t, int32_t const *, svint64_t);
16926		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_s64)))
16927		svint64_t svldnt1sw_gather_s64index_s64(svbool_t, int32_t const *, svint64_t);
16928		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_u64)))
16929		svuint64_t svldnt1sw_gather_u64index_u64(svbool_t, int32_t const *, svuint64_t);
16930		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_s64)))
16931		svint64_t svldnt1sw_gather_u64index_s64(svbool_t, int32_t const *, svuint64_t);
16932		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_u64)))
16933		svuint64_t svldnt1sw_gather_s64offset_u64(svbool_t, int32_t const *, svint64_t);
16934		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_s64)))
16935		svint64_t svldnt1sw_gather_s64offset_s64(svbool_t, int32_t const *, svint64_t);
16936		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_u64)))
16937		svuint64_t svldnt1sw_gather_u64offset_u64(svbool_t, int32_t const *, svuint64_t);
16938		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_s64)))
16939		svint64_t svldnt1sw_gather_u64offset_s64(svbool_t, int32_t const *, svuint64_t);
16940		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_u32)))
16941		svuint32_t svldnt1ub_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
16942		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_u64)))
16943		svuint64_t svldnt1ub_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
16944		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_s32)))
16945		svint32_t svldnt1ub_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
16946		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_s64)))
16947		svint64_t svldnt1ub_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
16948		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_u32)))
16949		svuint32_t svldnt1ub_gather_u32base_u32(svbool_t, svuint32_t);
16950		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_u64)))
16951		svuint64_t svldnt1ub_gather_u64base_u64(svbool_t, svuint64_t);
16952		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_s32)))
16953		svint32_t svldnt1ub_gather_u32base_s32(svbool_t, svuint32_t);
16954		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_s64)))
16955		svint64_t svldnt1ub_gather_u64base_s64(svbool_t, svuint64_t);
16956		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_u32)))
16957		svuint32_t svldnt1ub_gather_u32offset_u32(svbool_t, uint8_t const *, svuint32_t);
16958		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_s32)))
16959		svint32_t svldnt1ub_gather_u32offset_s32(svbool_t, uint8_t const *, svuint32_t);
16960		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_u64)))
16961		svuint64_t svldnt1ub_gather_s64offset_u64(svbool_t, uint8_t const *, svint64_t);
16962		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_s64)))
16963		svint64_t svldnt1ub_gather_s64offset_s64(svbool_t, uint8_t const *, svint64_t);
16964		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_u64)))
16965		svuint64_t svldnt1ub_gather_u64offset_u64(svbool_t, uint8_t const *, svuint64_t);
16966		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_s64)))
16967		svint64_t svldnt1ub_gather_u64offset_s64(svbool_t, uint8_t const *, svuint64_t);
16968		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_u32)))
16969		svuint32_t svldnt1uh_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
16970		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_u64)))
16971		svuint64_t svldnt1uh_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
16972		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_s32)))
16973		svint32_t svldnt1uh_gather_u32base_index_s32(svbool_t, svuint32_t, int64_t);
16974		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_s64)))
16975		svint64_t svldnt1uh_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
16976		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_u32)))
16977		svuint32_t svldnt1uh_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
16978		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_u64)))
16979		svuint64_t svldnt1uh_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
16980		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_s32)))
16981		svint32_t svldnt1uh_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
16982		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_s64)))
16983		svint64_t svldnt1uh_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
16984		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_u32)))
16985		svuint32_t svldnt1uh_gather_u32base_u32(svbool_t, svuint32_t);
16986		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_u64)))
16987		svuint64_t svldnt1uh_gather_u64base_u64(svbool_t, svuint64_t);
16988		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_s32)))
16989		svint32_t svldnt1uh_gather_u32base_s32(svbool_t, svuint32_t);
16990		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_s64)))
16991		svint64_t svldnt1uh_gather_u64base_s64(svbool_t, svuint64_t);
16992		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_u64)))
16993		svuint64_t svldnt1uh_gather_s64index_u64(svbool_t, uint16_t const *, svint64_t);
16994		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_s64)))
16995		svint64_t svldnt1uh_gather_s64index_s64(svbool_t, uint16_t const *, svint64_t);
16996		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_u64)))
16997		svuint64_t svldnt1uh_gather_u64index_u64(svbool_t, uint16_t const *, svuint64_t);
16998		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_s64)))
16999		svint64_t svldnt1uh_gather_u64index_s64(svbool_t, uint16_t const *, svuint64_t);
17000		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_u32)))
17001		svuint32_t svldnt1uh_gather_u32offset_u32(svbool_t, uint16_t const *, svuint32_t);
17002		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_s32)))
17003		svint32_t svldnt1uh_gather_u32offset_s32(svbool_t, uint16_t const *, svuint32_t);
17004		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_u64)))
17005		svuint64_t svldnt1uh_gather_s64offset_u64(svbool_t, uint16_t const *, svint64_t);
17006		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_s64)))
17007		svint64_t svldnt1uh_gather_s64offset_s64(svbool_t, uint16_t const *, svint64_t);
17008		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_u64)))
17009		svuint64_t svldnt1uh_gather_u64offset_u64(svbool_t, uint16_t const *, svuint64_t);
17010		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_s64)))
17011		svint64_t svldnt1uh_gather_u64offset_s64(svbool_t, uint16_t const *, svuint64_t);
17012		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_u64)))
17013		svuint64_t svldnt1uw_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
17014		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_s64)))
17015		svint64_t svldnt1uw_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
17016		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_u64)))
17017		svuint64_t svldnt1uw_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
17018		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_s64)))
17019		svint64_t svldnt1uw_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
17020		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_u64)))
17021		svuint64_t svldnt1uw_gather_u64base_u64(svbool_t, svuint64_t);
17022		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_s64)))
17023		svint64_t svldnt1uw_gather_u64base_s64(svbool_t, svuint64_t);
17024		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_u64)))
17025		svuint64_t svldnt1uw_gather_s64index_u64(svbool_t, uint32_t const *, svint64_t);
17026		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_s64)))
17027		svint64_t svldnt1uw_gather_s64index_s64(svbool_t, uint32_t const *, svint64_t);
17028		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_u64)))
17029		svuint64_t svldnt1uw_gather_u64index_u64(svbool_t, uint32_t const *, svuint64_t);
17030		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_s64)))
17031		svint64_t svldnt1uw_gather_u64index_s64(svbool_t, uint32_t const *, svuint64_t);
17032		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_u64)))
17033		svuint64_t svldnt1uw_gather_s64offset_u64(svbool_t, uint32_t const *, svint64_t);
17034		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_s64)))
17035		svint64_t svldnt1uw_gather_s64offset_s64(svbool_t, uint32_t const *, svint64_t);
17036		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_u64)))
17037		svuint64_t svldnt1uw_gather_u64offset_u64(svbool_t, uint32_t const *, svuint64_t);
17038		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_s64)))
17039		svint64_t svldnt1uw_gather_u64offset_s64(svbool_t, uint32_t const *, svuint64_t);
17040		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_m)))
17041		svint64_t svlogb_f64_m(svint64_t, svbool_t, svfloat64_t);
17042		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_m)))
17043		svint32_t svlogb_f32_m(svint32_t, svbool_t, svfloat32_t);
17044		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_m)))
17045		svint16_t svlogb_f16_m(svint16_t, svbool_t, svfloat16_t);
17046		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_x)))
17047		svint64_t svlogb_f64_x(svbool_t, svfloat64_t);
17048		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_x)))
17049		svint32_t svlogb_f32_x(svbool_t, svfloat32_t);
17050		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_x)))
17051		svint16_t svlogb_f16_x(svbool_t, svfloat16_t);
17052		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_z)))
17053		svint64_t svlogb_f64_z(svbool_t, svfloat64_t);
17054		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_z)))
17055		svint32_t svlogb_f32_z(svbool_t, svfloat32_t);
17056		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_z)))
17057		svint16_t svlogb_f16_z(svbool_t, svfloat16_t);
17058		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u8)))
17059		svbool_t svmatch_u8(svbool_t, svuint8_t, svuint8_t);
17060		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u16)))
17061		svbool_t svmatch_u16(svbool_t, svuint16_t, svuint16_t);
17062		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s8)))
17063		svbool_t svmatch_s8(svbool_t, svint8_t, svint8_t);
17064		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s16)))
17065		svbool_t svmatch_s16(svbool_t, svint16_t, svint16_t);
17066		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_m)))
17067		svfloat64_t svmaxnmp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
17068		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_m)))
17069		svfloat32_t svmaxnmp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
17070		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_m)))
17071		svfloat16_t svmaxnmp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
17072		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_x)))
17073		svfloat64_t svmaxnmp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
17074		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_x)))
17075		svfloat32_t svmaxnmp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
17076		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_x)))
17077		svfloat16_t svmaxnmp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
17078		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_m)))
17079		svfloat64_t svmaxp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
17080		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_m)))
17081		svfloat32_t svmaxp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
17082		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_m)))
17083		svfloat16_t svmaxp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
17084		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_x)))
17085		svfloat64_t svmaxp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
17086		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_x)))
17087		svfloat32_t svmaxp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
17088		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_x)))
17089		svfloat16_t svmaxp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
17090		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_m)))
17091		svint8_t svmaxp_s8_m(svbool_t, svint8_t, svint8_t);
17092		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_m)))
17093		svint32_t svmaxp_s32_m(svbool_t, svint32_t, svint32_t);
17094		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_m)))
17095		svint64_t svmaxp_s64_m(svbool_t, svint64_t, svint64_t);
17096		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_m)))
17097		svint16_t svmaxp_s16_m(svbool_t, svint16_t, svint16_t);
17098		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_x)))
17099		svint8_t svmaxp_s8_x(svbool_t, svint8_t, svint8_t);
17100		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_x)))
17101		svint32_t svmaxp_s32_x(svbool_t, svint32_t, svint32_t);
17102		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_x)))
17103		svint64_t svmaxp_s64_x(svbool_t, svint64_t, svint64_t);
17104		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_x)))
17105		svint16_t svmaxp_s16_x(svbool_t, svint16_t, svint16_t);
17106		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_m)))
17107		svuint8_t svmaxp_u8_m(svbool_t, svuint8_t, svuint8_t);
17108		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_m)))
17109		svuint32_t svmaxp_u32_m(svbool_t, svuint32_t, svuint32_t);
17110		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_m)))
17111		svuint64_t svmaxp_u64_m(svbool_t, svuint64_t, svuint64_t);
17112		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_m)))
17113		svuint16_t svmaxp_u16_m(svbool_t, svuint16_t, svuint16_t);
17114		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_x)))
17115		svuint8_t svmaxp_u8_x(svbool_t, svuint8_t, svuint8_t);
17116		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_x)))
17117		svuint32_t svmaxp_u32_x(svbool_t, svuint32_t, svuint32_t);
17118		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_x)))
17119		svuint64_t svmaxp_u64_x(svbool_t, svuint64_t, svuint64_t);
17120		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_x)))
17121		svuint16_t svmaxp_u16_x(svbool_t, svuint16_t, svuint16_t);
17122		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_m)))
17123		svfloat64_t svminnmp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
17124		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_m)))
17125		svfloat32_t svminnmp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
17126		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_m)))
17127		svfloat16_t svminnmp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
17128		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_x)))
17129		svfloat64_t svminnmp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
17130		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_x)))
17131		svfloat32_t svminnmp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
17132		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_x)))
17133		svfloat16_t svminnmp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
17134		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_m)))
17135		svfloat64_t svminp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
17136		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_m)))
17137		svfloat32_t svminp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
17138		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_m)))
17139		svfloat16_t svminp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
17140		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_x)))
17141		svfloat64_t svminp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
17142		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_x)))
17143		svfloat32_t svminp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
17144		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_x)))
17145		svfloat16_t svminp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
17146		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_m)))
17147		svint8_t svminp_s8_m(svbool_t, svint8_t, svint8_t);
17148		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_m)))
17149		svint32_t svminp_s32_m(svbool_t, svint32_t, svint32_t);
17150		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_m)))
17151		svint64_t svminp_s64_m(svbool_t, svint64_t, svint64_t);
17152		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_m)))
17153		svint16_t svminp_s16_m(svbool_t, svint16_t, svint16_t);
17154		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_x)))
17155		svint8_t svminp_s8_x(svbool_t, svint8_t, svint8_t);
17156		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_x)))
17157		svint32_t svminp_s32_x(svbool_t, svint32_t, svint32_t);
17158		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_x)))
17159		svint64_t svminp_s64_x(svbool_t, svint64_t, svint64_t);
17160		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_x)))
17161		svint16_t svminp_s16_x(svbool_t, svint16_t, svint16_t);
17162		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_m)))
17163		svuint8_t svminp_u8_m(svbool_t, svuint8_t, svuint8_t);
17164		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_m)))
17165		svuint32_t svminp_u32_m(svbool_t, svuint32_t, svuint32_t);
17166		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_m)))
17167		svuint64_t svminp_u64_m(svbool_t, svuint64_t, svuint64_t);
17168		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_m)))
17169		svuint16_t svminp_u16_m(svbool_t, svuint16_t, svuint16_t);
17170		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_x)))
17171		svuint8_t svminp_u8_x(svbool_t, svuint8_t, svuint8_t);
17172		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_x)))
17173		svuint32_t svminp_u32_x(svbool_t, svuint32_t, svuint32_t);
17174		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_x)))
17175		svuint64_t svminp_u64_x(svbool_t, svuint64_t, svuint64_t);
17176		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_x)))
17177		svuint16_t svminp_u16_x(svbool_t, svuint16_t, svuint16_t);
17178		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u32)))
17179		svuint32_t svmla_lane_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t);
17180		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u64)))
17181		svuint64_t svmla_lane_u64(svuint64_t, svuint64_t, svuint64_t, uint64_t);
17182		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u16)))
17183		svuint16_t svmla_lane_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t);
17184		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s32)))
17185		svint32_t svmla_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
17186		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s64)))
17187		svint64_t svmla_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
17188		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s16)))
17189		svint16_t svmla_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
17190		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_f32)))
17191		svfloat32_t svmlalb_n_f32(svfloat32_t, svfloat16_t, float16_t);
17192		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s32)))
17193		svint32_t svmlalb_n_s32(svint32_t, svint16_t, int16_t);
17194		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s64)))
17195		svint64_t svmlalb_n_s64(svint64_t, svint32_t, int32_t);
17196		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s16)))
17197		svint16_t svmlalb_n_s16(svint16_t, svint8_t, int8_t);
17198		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u32)))
17199		svuint32_t svmlalb_n_u32(svuint32_t, svuint16_t, uint16_t);
17200		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u64)))
17201		svuint64_t svmlalb_n_u64(svuint64_t, svuint32_t, uint32_t);
17202		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u16)))
17203		svuint16_t svmlalb_n_u16(svuint16_t, svuint8_t, uint8_t);
17204		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_f32)))
17205		svfloat32_t svmlalb_f32(svfloat32_t, svfloat16_t, svfloat16_t);
17206		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s32)))
17207		svint32_t svmlalb_s32(svint32_t, svint16_t, svint16_t);
17208		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s64)))
17209		svint64_t svmlalb_s64(svint64_t, svint32_t, svint32_t);
17210		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s16)))
17211		svint16_t svmlalb_s16(svint16_t, svint8_t, svint8_t);
17212		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u32)))
17213		svuint32_t svmlalb_u32(svuint32_t, svuint16_t, svuint16_t);
17214		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u64)))
17215		svuint64_t svmlalb_u64(svuint64_t, svuint32_t, svuint32_t);
17216		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u16)))
17217		svuint16_t svmlalb_u16(svuint16_t, svuint8_t, svuint8_t);
17218		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_f32)))
17219		svfloat32_t svmlalb_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
17220		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s32)))
17221		svint32_t svmlalb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17222		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s64)))
17223		svint64_t svmlalb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17224		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u32)))
17225		svuint32_t svmlalb_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
17226		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u64)))
17227		svuint64_t svmlalb_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
17228		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_f32)))
17229		svfloat32_t svmlalt_n_f32(svfloat32_t, svfloat16_t, float16_t);
17230		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s32)))
17231		svint32_t svmlalt_n_s32(svint32_t, svint16_t, int16_t);
17232		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s64)))
17233		svint64_t svmlalt_n_s64(svint64_t, svint32_t, int32_t);
17234		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s16)))
17235		svint16_t svmlalt_n_s16(svint16_t, svint8_t, int8_t);
17236		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u32)))
17237		svuint32_t svmlalt_n_u32(svuint32_t, svuint16_t, uint16_t);
17238		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u64)))
17239		svuint64_t svmlalt_n_u64(svuint64_t, svuint32_t, uint32_t);
17240		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u16)))
17241		svuint16_t svmlalt_n_u16(svuint16_t, svuint8_t, uint8_t);
17242		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_f32)))
17243		svfloat32_t svmlalt_f32(svfloat32_t, svfloat16_t, svfloat16_t);
17244		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s32)))
17245		svint32_t svmlalt_s32(svint32_t, svint16_t, svint16_t);
17246		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s64)))
17247		svint64_t svmlalt_s64(svint64_t, svint32_t, svint32_t);
17248		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s16)))
17249		svint16_t svmlalt_s16(svint16_t, svint8_t, svint8_t);
17250		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u32)))
17251		svuint32_t svmlalt_u32(svuint32_t, svuint16_t, svuint16_t);
17252		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u64)))
17253		svuint64_t svmlalt_u64(svuint64_t, svuint32_t, svuint32_t);
17254		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u16)))
17255		svuint16_t svmlalt_u16(svuint16_t, svuint8_t, svuint8_t);
17256		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_f32)))
17257		svfloat32_t svmlalt_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
17258		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s32)))
17259		svint32_t svmlalt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17260		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s64)))
17261		svint64_t svmlalt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17262		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u32)))
17263		svuint32_t svmlalt_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
17264		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u64)))
17265		svuint64_t svmlalt_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
17266		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u32)))
17267		svuint32_t svmls_lane_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t);
17268		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u64)))
17269		svuint64_t svmls_lane_u64(svuint64_t, svuint64_t, svuint64_t, uint64_t);
17270		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u16)))
17271		svuint16_t svmls_lane_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t);
17272		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s32)))
17273		svint32_t svmls_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
17274		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s64)))
17275		svint64_t svmls_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
17276		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s16)))
17277		svint16_t svmls_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
17278		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_f32)))
17279		svfloat32_t svmlslb_n_f32(svfloat32_t, svfloat16_t, float16_t);
17280		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s32)))
17281		svint32_t svmlslb_n_s32(svint32_t, svint16_t, int16_t);
17282		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s64)))
17283		svint64_t svmlslb_n_s64(svint64_t, svint32_t, int32_t);
17284		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s16)))
17285		svint16_t svmlslb_n_s16(svint16_t, svint8_t, int8_t);
17286		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u32)))
17287		svuint32_t svmlslb_n_u32(svuint32_t, svuint16_t, uint16_t);
17288		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u64)))
17289		svuint64_t svmlslb_n_u64(svuint64_t, svuint32_t, uint32_t);
17290		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u16)))
17291		svuint16_t svmlslb_n_u16(svuint16_t, svuint8_t, uint8_t);
17292		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_f32)))
17293		svfloat32_t svmlslb_f32(svfloat32_t, svfloat16_t, svfloat16_t);
17294		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s32)))
17295		svint32_t svmlslb_s32(svint32_t, svint16_t, svint16_t);
17296		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s64)))
17297		svint64_t svmlslb_s64(svint64_t, svint32_t, svint32_t);
17298		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s16)))
17299		svint16_t svmlslb_s16(svint16_t, svint8_t, svint8_t);
17300		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u32)))
17301		svuint32_t svmlslb_u32(svuint32_t, svuint16_t, svuint16_t);
17302		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u64)))
17303		svuint64_t svmlslb_u64(svuint64_t, svuint32_t, svuint32_t);
17304		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u16)))
17305		svuint16_t svmlslb_u16(svuint16_t, svuint8_t, svuint8_t);
17306		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_f32)))
17307		svfloat32_t svmlslb_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
17308		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s32)))
17309		svint32_t svmlslb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17310		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s64)))
17311		svint64_t svmlslb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17312		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u32)))
17313		svuint32_t svmlslb_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
17314		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u64)))
17315		svuint64_t svmlslb_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
17316		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_f32)))
17317		svfloat32_t svmlslt_n_f32(svfloat32_t, svfloat16_t, float16_t);
17318		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s32)))
17319		svint32_t svmlslt_n_s32(svint32_t, svint16_t, int16_t);
17320		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s64)))
17321		svint64_t svmlslt_n_s64(svint64_t, svint32_t, int32_t);
17322		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s16)))
17323		svint16_t svmlslt_n_s16(svint16_t, svint8_t, int8_t);
17324		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u32)))
17325		svuint32_t svmlslt_n_u32(svuint32_t, svuint16_t, uint16_t);
17326		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u64)))
17327		svuint64_t svmlslt_n_u64(svuint64_t, svuint32_t, uint32_t);
17328		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u16)))
17329		svuint16_t svmlslt_n_u16(svuint16_t, svuint8_t, uint8_t);
17330		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_f32)))
17331		svfloat32_t svmlslt_f32(svfloat32_t, svfloat16_t, svfloat16_t);
17332		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s32)))
17333		svint32_t svmlslt_s32(svint32_t, svint16_t, svint16_t);
17334		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s64)))
17335		svint64_t svmlslt_s64(svint64_t, svint32_t, svint32_t);
17336		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s16)))
17337		svint16_t svmlslt_s16(svint16_t, svint8_t, svint8_t);
17338		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u32)))
17339		svuint32_t svmlslt_u32(svuint32_t, svuint16_t, svuint16_t);
17340		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u64)))
17341		svuint64_t svmlslt_u64(svuint64_t, svuint32_t, svuint32_t);
17342		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u16)))
17343		svuint16_t svmlslt_u16(svuint16_t, svuint8_t, svuint8_t);
17344		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_f32)))
17345		svfloat32_t svmlslt_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
17346		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s32)))
17347		svint32_t svmlslt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17348		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s64)))
17349		svint64_t svmlslt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17350		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u32)))
17351		svuint32_t svmlslt_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
17352		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u64)))
17353		svuint64_t svmlslt_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
17354		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s32)))
17355		svint32_t svmovlb_s32(svint16_t);
17356		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s64)))
17357		svint64_t svmovlb_s64(svint32_t);
17358		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s16)))
17359		svint16_t svmovlb_s16(svint8_t);
17360		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u32)))
17361		svuint32_t svmovlb_u32(svuint16_t);
17362		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u64)))
17363		svuint64_t svmovlb_u64(svuint32_t);
17364		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u16)))
17365		svuint16_t svmovlb_u16(svuint8_t);
17366		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s32)))
17367		svint32_t svmovlt_s32(svint16_t);
17368		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s64)))
17369		svint64_t svmovlt_s64(svint32_t);
17370		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s16)))
17371		svint16_t svmovlt_s16(svint8_t);
17372		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u32)))
17373		svuint32_t svmovlt_u32(svuint16_t);
17374		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u64)))
17375		svuint64_t svmovlt_u64(svuint32_t);
17376		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u16)))
17377		svuint16_t svmovlt_u16(svuint8_t);
17378		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u32)))
17379		svuint32_t svmul_lane_u32(svuint32_t, svuint32_t, uint64_t);
17380		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u64)))
17381		svuint64_t svmul_lane_u64(svuint64_t, svuint64_t, uint64_t);
17382		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u16)))
17383		svuint16_t svmul_lane_u16(svuint16_t, svuint16_t, uint64_t);
17384		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s32)))
17385		svint32_t svmul_lane_s32(svint32_t, svint32_t, uint64_t);
17386		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s64)))
17387		svint64_t svmul_lane_s64(svint64_t, svint64_t, uint64_t);
17388		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s16)))
17389		svint16_t svmul_lane_s16(svint16_t, svint16_t, uint64_t);
17390		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s32)))
17391		svint32_t svmullb_n_s32(svint16_t, int16_t);
17392		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s64)))
17393		svint64_t svmullb_n_s64(svint32_t, int32_t);
17394		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s16)))
17395		svint16_t svmullb_n_s16(svint8_t, int8_t);
17396		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u32)))
17397		svuint32_t svmullb_n_u32(svuint16_t, uint16_t);
17398		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u64)))
17399		svuint64_t svmullb_n_u64(svuint32_t, uint32_t);
17400		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u16)))
17401		svuint16_t svmullb_n_u16(svuint8_t, uint8_t);
17402		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s32)))
17403		svint32_t svmullb_s32(svint16_t, svint16_t);
17404		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s64)))
17405		svint64_t svmullb_s64(svint32_t, svint32_t);
17406		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s16)))
17407		svint16_t svmullb_s16(svint8_t, svint8_t);
17408		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u32)))
17409		svuint32_t svmullb_u32(svuint16_t, svuint16_t);
17410		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u64)))
17411		svuint64_t svmullb_u64(svuint32_t, svuint32_t);
17412		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u16)))
17413		svuint16_t svmullb_u16(svuint8_t, svuint8_t);
17414		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s32)))
17415		svint32_t svmullb_lane_s32(svint16_t, svint16_t, uint64_t);
17416		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s64)))
17417		svint64_t svmullb_lane_s64(svint32_t, svint32_t, uint64_t);
17418		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u32)))
17419		svuint32_t svmullb_lane_u32(svuint16_t, svuint16_t, uint64_t);
17420		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u64)))
17421		svuint64_t svmullb_lane_u64(svuint32_t, svuint32_t, uint64_t);
17422		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s32)))
17423		svint32_t svmullt_n_s32(svint16_t, int16_t);
17424		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s64)))
17425		svint64_t svmullt_n_s64(svint32_t, int32_t);
17426		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s16)))
17427		svint16_t svmullt_n_s16(svint8_t, int8_t);
17428		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u32)))
17429		svuint32_t svmullt_n_u32(svuint16_t, uint16_t);
17430		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u64)))
17431		svuint64_t svmullt_n_u64(svuint32_t, uint32_t);
17432		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u16)))
17433		svuint16_t svmullt_n_u16(svuint8_t, uint8_t);
17434		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s32)))
17435		svint32_t svmullt_s32(svint16_t, svint16_t);
17436		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s64)))
17437		svint64_t svmullt_s64(svint32_t, svint32_t);
17438		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s16)))
17439		svint16_t svmullt_s16(svint8_t, svint8_t);
17440		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u32)))
17441		svuint32_t svmullt_u32(svuint16_t, svuint16_t);
17442		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u64)))
17443		svuint64_t svmullt_u64(svuint32_t, svuint32_t);
17444		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u16)))
17445		svuint16_t svmullt_u16(svuint8_t, svuint8_t);
17446		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s32)))
17447		svint32_t svmullt_lane_s32(svint16_t, svint16_t, uint64_t);
17448		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s64)))
17449		svint64_t svmullt_lane_s64(svint32_t, svint32_t, uint64_t);
17450		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u32)))
17451		svuint32_t svmullt_lane_u32(svuint16_t, svuint16_t, uint64_t);
17452		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u64)))
17453		svuint64_t svmullt_lane_u64(svuint32_t, svuint32_t, uint64_t);
17454		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u8)))
17455		svuint8_t svnbsl_n_u8(svuint8_t, svuint8_t, uint8_t);
17456		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u32)))
17457		svuint32_t svnbsl_n_u32(svuint32_t, svuint32_t, uint32_t);
17458		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u64)))
17459		svuint64_t svnbsl_n_u64(svuint64_t, svuint64_t, uint64_t);
17460		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u16)))
17461		svuint16_t svnbsl_n_u16(svuint16_t, svuint16_t, uint16_t);
17462		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s8)))
17463		svint8_t svnbsl_n_s8(svint8_t, svint8_t, int8_t);
17464		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s32)))
17465		svint32_t svnbsl_n_s32(svint32_t, svint32_t, int32_t);
17466		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s64)))
17467		svint64_t svnbsl_n_s64(svint64_t, svint64_t, int64_t);
17468		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s16)))
17469		svint16_t svnbsl_n_s16(svint16_t, svint16_t, int16_t);
17470		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u8)))
17471		svuint8_t svnbsl_u8(svuint8_t, svuint8_t, svuint8_t);
17472		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u32)))
17473		svuint32_t svnbsl_u32(svuint32_t, svuint32_t, svuint32_t);
17474		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u64)))
17475		svuint64_t svnbsl_u64(svuint64_t, svuint64_t, svuint64_t);
17476		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u16)))
17477		svuint16_t svnbsl_u16(svuint16_t, svuint16_t, svuint16_t);
17478		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s8)))
17479		svint8_t svnbsl_s8(svint8_t, svint8_t, svint8_t);
17480		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s32)))
17481		svint32_t svnbsl_s32(svint32_t, svint32_t, svint32_t);
17482		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s64)))
17483		svint64_t svnbsl_s64(svint64_t, svint64_t, svint64_t);
17484		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s16)))
17485		svint16_t svnbsl_s16(svint16_t, svint16_t, svint16_t);
17486		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u8)))
17487		svbool_t svnmatch_u8(svbool_t, svuint8_t, svuint8_t);
17488		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u16)))
17489		svbool_t svnmatch_u16(svbool_t, svuint16_t, svuint16_t);
17490		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s8)))
17491		svbool_t svnmatch_s8(svbool_t, svint8_t, svint8_t);
17492		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s16)))
17493		svbool_t svnmatch_s16(svbool_t, svint16_t, svint16_t);
17494		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_n_u8)))
17495		svuint8_t svpmul_n_u8(svuint8_t, uint8_t);
17496		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_u8)))
17497		svuint8_t svpmul_u8(svuint8_t, svuint8_t);
17498		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u64)))
17499		svuint64_t svpmullb_n_u64(svuint32_t, uint32_t);
17500		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u16)))
17501		svuint16_t svpmullb_n_u16(svuint8_t, uint8_t);
17502		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u64)))
17503		svuint64_t svpmullb_u64(svuint32_t, svuint32_t);
17504		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u16)))
17505		svuint16_t svpmullb_u16(svuint8_t, svuint8_t);
17506		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u8)))
17507		svuint8_t svpmullb_pair_n_u8(svuint8_t, uint8_t);
17508		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u32)))
17509		svuint32_t svpmullb_pair_n_u32(svuint32_t, uint32_t);
17510		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u8)))
17511		svuint8_t svpmullb_pair_u8(svuint8_t, svuint8_t);
17512		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u32)))
17513		svuint32_t svpmullb_pair_u32(svuint32_t, svuint32_t);
17514		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u64)))
17515		svuint64_t svpmullt_n_u64(svuint32_t, uint32_t);
17516		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u16)))
17517		svuint16_t svpmullt_n_u16(svuint8_t, uint8_t);
17518		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u64)))
17519		svuint64_t svpmullt_u64(svuint32_t, svuint32_t);
17520		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u16)))
17521		svuint16_t svpmullt_u16(svuint8_t, svuint8_t);
17522		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u8)))
17523		svuint8_t svpmullt_pair_n_u8(svuint8_t, uint8_t);
17524		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u32)))
17525		svuint32_t svpmullt_pair_n_u32(svuint32_t, uint32_t);
17526		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u8)))
17527		svuint8_t svpmullt_pair_u8(svuint8_t, svuint8_t);
17528		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u32)))
17529		svuint32_t svpmullt_pair_u32(svuint32_t, svuint32_t);
17530		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_m)))
17531		svint8_t svqabs_s8_m(svint8_t, svbool_t, svint8_t);
17532		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_m)))
17533		svint32_t svqabs_s32_m(svint32_t, svbool_t, svint32_t);
17534		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_m)))
17535		svint64_t svqabs_s64_m(svint64_t, svbool_t, svint64_t);
17536		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_m)))
17537		svint16_t svqabs_s16_m(svint16_t, svbool_t, svint16_t);
17538		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_x)))
17539		svint8_t svqabs_s8_x(svbool_t, svint8_t);
17540		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_x)))
17541		svint32_t svqabs_s32_x(svbool_t, svint32_t);
17542		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_x)))
17543		svint64_t svqabs_s64_x(svbool_t, svint64_t);
17544		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_x)))
17545		svint16_t svqabs_s16_x(svbool_t, svint16_t);
17546		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_z)))
17547		svint8_t svqabs_s8_z(svbool_t, svint8_t);
17548		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_z)))
17549		svint32_t svqabs_s32_z(svbool_t, svint32_t);
17550		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_z)))
17551		svint64_t svqabs_s64_z(svbool_t, svint64_t);
17552		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_z)))
17553		svint16_t svqabs_s16_z(svbool_t, svint16_t);
17554		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_m)))
17555		svint8_t svqadd_n_s8_m(svbool_t, svint8_t, int8_t);
17556		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_m)))
17557		svint32_t svqadd_n_s32_m(svbool_t, svint32_t, int32_t);
17558		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_m)))
17559		svint64_t svqadd_n_s64_m(svbool_t, svint64_t, int64_t);
17560		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_m)))
17561		svint16_t svqadd_n_s16_m(svbool_t, svint16_t, int16_t);
17562		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_x)))
17563		svint8_t svqadd_n_s8_x(svbool_t, svint8_t, int8_t);
17564		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_x)))
17565		svint32_t svqadd_n_s32_x(svbool_t, svint32_t, int32_t);
17566		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_x)))
17567		svint64_t svqadd_n_s64_x(svbool_t, svint64_t, int64_t);
17568		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_x)))
17569		svint16_t svqadd_n_s16_x(svbool_t, svint16_t, int16_t);
17570		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_z)))
17571		svint8_t svqadd_n_s8_z(svbool_t, svint8_t, int8_t);
17572		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_z)))
17573		svint32_t svqadd_n_s32_z(svbool_t, svint32_t, int32_t);
17574		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_z)))
17575		svint64_t svqadd_n_s64_z(svbool_t, svint64_t, int64_t);
17576		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_z)))
17577		svint16_t svqadd_n_s16_z(svbool_t, svint16_t, int16_t);
17578		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_m)))
17579		svuint8_t svqadd_n_u8_m(svbool_t, svuint8_t, uint8_t);
17580		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_m)))
17581		svuint32_t svqadd_n_u32_m(svbool_t, svuint32_t, uint32_t);
17582		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_m)))
17583		svuint64_t svqadd_n_u64_m(svbool_t, svuint64_t, uint64_t);
17584		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_m)))
17585		svuint16_t svqadd_n_u16_m(svbool_t, svuint16_t, uint16_t);
17586		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_x)))
17587		svuint8_t svqadd_n_u8_x(svbool_t, svuint8_t, uint8_t);
17588		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_x)))
17589		svuint32_t svqadd_n_u32_x(svbool_t, svuint32_t, uint32_t);
17590		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_x)))
17591		svuint64_t svqadd_n_u64_x(svbool_t, svuint64_t, uint64_t);
17592		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_x)))
17593		svuint16_t svqadd_n_u16_x(svbool_t, svuint16_t, uint16_t);
17594		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_z)))
17595		svuint8_t svqadd_n_u8_z(svbool_t, svuint8_t, uint8_t);
17596		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_z)))
17597		svuint32_t svqadd_n_u32_z(svbool_t, svuint32_t, uint32_t);
17598		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_z)))
17599		svuint64_t svqadd_n_u64_z(svbool_t, svuint64_t, uint64_t);
17600		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_z)))
17601		svuint16_t svqadd_n_u16_z(svbool_t, svuint16_t, uint16_t);
17602		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_m)))
17603		svint8_t svqadd_s8_m(svbool_t, svint8_t, svint8_t);
17604		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_m)))
17605		svint32_t svqadd_s32_m(svbool_t, svint32_t, svint32_t);
17606		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_m)))
17607		svint64_t svqadd_s64_m(svbool_t, svint64_t, svint64_t);
17608		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_m)))
17609		svint16_t svqadd_s16_m(svbool_t, svint16_t, svint16_t);
17610		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_x)))
17611		svint8_t svqadd_s8_x(svbool_t, svint8_t, svint8_t);
17612		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_x)))
17613		svint32_t svqadd_s32_x(svbool_t, svint32_t, svint32_t);
17614		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_x)))
17615		svint64_t svqadd_s64_x(svbool_t, svint64_t, svint64_t);
17616		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_x)))
17617		svint16_t svqadd_s16_x(svbool_t, svint16_t, svint16_t);
17618		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_z)))
17619		svint8_t svqadd_s8_z(svbool_t, svint8_t, svint8_t);
17620		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_z)))
17621		svint32_t svqadd_s32_z(svbool_t, svint32_t, svint32_t);
17622		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_z)))
17623		svint64_t svqadd_s64_z(svbool_t, svint64_t, svint64_t);
17624		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_z)))
17625		svint16_t svqadd_s16_z(svbool_t, svint16_t, svint16_t);
17626		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_m)))
17627		svuint8_t svqadd_u8_m(svbool_t, svuint8_t, svuint8_t);
17628		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_m)))
17629		svuint32_t svqadd_u32_m(svbool_t, svuint32_t, svuint32_t);
17630		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_m)))
17631		svuint64_t svqadd_u64_m(svbool_t, svuint64_t, svuint64_t);
17632		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_m)))
17633		svuint16_t svqadd_u16_m(svbool_t, svuint16_t, svuint16_t);
17634		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_x)))
17635		svuint8_t svqadd_u8_x(svbool_t, svuint8_t, svuint8_t);
17636		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_x)))
17637		svuint32_t svqadd_u32_x(svbool_t, svuint32_t, svuint32_t);
17638		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_x)))
17639		svuint64_t svqadd_u64_x(svbool_t, svuint64_t, svuint64_t);
17640		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_x)))
17641		svuint16_t svqadd_u16_x(svbool_t, svuint16_t, svuint16_t);
17642		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_z)))
17643		svuint8_t svqadd_u8_z(svbool_t, svuint8_t, svuint8_t);
17644		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_z)))
17645		svuint32_t svqadd_u32_z(svbool_t, svuint32_t, svuint32_t);
17646		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_z)))
17647		svuint64_t svqadd_u64_z(svbool_t, svuint64_t, svuint64_t);
17648		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_z)))
17649		svuint16_t svqadd_u16_z(svbool_t, svuint16_t, svuint16_t);
17650		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s8)))
17651		svint8_t svqcadd_s8(svint8_t, svint8_t, uint64_t);
17652		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s32)))
17653		svint32_t svqcadd_s32(svint32_t, svint32_t, uint64_t);
17654		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s64)))
17655		svint64_t svqcadd_s64(svint64_t, svint64_t, uint64_t);
17656		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s16)))
17657		svint16_t svqcadd_s16(svint16_t, svint16_t, uint64_t);
17658		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s32)))
17659		svint32_t svqdmlalb_n_s32(svint32_t, svint16_t, int16_t);
17660		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s64)))
17661		svint64_t svqdmlalb_n_s64(svint64_t, svint32_t, int32_t);
17662		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s16)))
17663		svint16_t svqdmlalb_n_s16(svint16_t, svint8_t, int8_t);
17664		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s32)))
17665		svint32_t svqdmlalb_s32(svint32_t, svint16_t, svint16_t);
17666		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s64)))
17667		svint64_t svqdmlalb_s64(svint64_t, svint32_t, svint32_t);
17668		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s16)))
17669		svint16_t svqdmlalb_s16(svint16_t, svint8_t, svint8_t);
17670		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s32)))
17671		svint32_t svqdmlalb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17672		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s64)))
17673		svint64_t svqdmlalb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17674		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s32)))
17675		svint32_t svqdmlalbt_n_s32(svint32_t, svint16_t, int16_t);
17676		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s64)))
17677		svint64_t svqdmlalbt_n_s64(svint64_t, svint32_t, int32_t);
17678		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s16)))
17679		svint16_t svqdmlalbt_n_s16(svint16_t, svint8_t, int8_t);
17680		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s32)))
17681		svint32_t svqdmlalbt_s32(svint32_t, svint16_t, svint16_t);
17682		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s64)))
17683		svint64_t svqdmlalbt_s64(svint64_t, svint32_t, svint32_t);
17684		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s16)))
17685		svint16_t svqdmlalbt_s16(svint16_t, svint8_t, svint8_t);
17686		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s32)))
17687		svint32_t svqdmlalt_n_s32(svint32_t, svint16_t, int16_t);
17688		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s64)))
17689		svint64_t svqdmlalt_n_s64(svint64_t, svint32_t, int32_t);
17690		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s16)))
17691		svint16_t svqdmlalt_n_s16(svint16_t, svint8_t, int8_t);
17692		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s32)))
17693		svint32_t svqdmlalt_s32(svint32_t, svint16_t, svint16_t);
17694		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s64)))
17695		svint64_t svqdmlalt_s64(svint64_t, svint32_t, svint32_t);
17696		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s16)))
17697		svint16_t svqdmlalt_s16(svint16_t, svint8_t, svint8_t);
17698		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s32)))
17699		svint32_t svqdmlalt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17700		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s64)))
17701		svint64_t svqdmlalt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17702		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s32)))
17703		svint32_t svqdmlslb_n_s32(svint32_t, svint16_t, int16_t);
17704		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s64)))
17705		svint64_t svqdmlslb_n_s64(svint64_t, svint32_t, int32_t);
17706		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s16)))
17707		svint16_t svqdmlslb_n_s16(svint16_t, svint8_t, int8_t);
17708		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s32)))
17709		svint32_t svqdmlslb_s32(svint32_t, svint16_t, svint16_t);
17710		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s64)))
17711		svint64_t svqdmlslb_s64(svint64_t, svint32_t, svint32_t);
17712		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s16)))
17713		svint16_t svqdmlslb_s16(svint16_t, svint8_t, svint8_t);
17714		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s32)))
17715		svint32_t svqdmlslb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17716		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s64)))
17717		svint64_t svqdmlslb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17718		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s32)))
17719		svint32_t svqdmlslbt_n_s32(svint32_t, svint16_t, int16_t);
17720		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s64)))
17721		svint64_t svqdmlslbt_n_s64(svint64_t, svint32_t, int32_t);
17722		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s16)))
17723		svint16_t svqdmlslbt_n_s16(svint16_t, svint8_t, int8_t);
17724		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s32)))
17725		svint32_t svqdmlslbt_s32(svint32_t, svint16_t, svint16_t);
17726		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s64)))
17727		svint64_t svqdmlslbt_s64(svint64_t, svint32_t, svint32_t);
17728		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s16)))
17729		svint16_t svqdmlslbt_s16(svint16_t, svint8_t, svint8_t);
17730		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s32)))
17731		svint32_t svqdmlslt_n_s32(svint32_t, svint16_t, int16_t);
17732		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s64)))
17733		svint64_t svqdmlslt_n_s64(svint64_t, svint32_t, int32_t);
17734		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s16)))
17735		svint16_t svqdmlslt_n_s16(svint16_t, svint8_t, int8_t);
17736		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s32)))
17737		svint32_t svqdmlslt_s32(svint32_t, svint16_t, svint16_t);
17738		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s64)))
17739		svint64_t svqdmlslt_s64(svint64_t, svint32_t, svint32_t);
17740		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s16)))
17741		svint16_t svqdmlslt_s16(svint16_t, svint8_t, svint8_t);
17742		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s32)))
17743		svint32_t svqdmlslt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
17744		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s64)))
17745		svint64_t svqdmlslt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
17746		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s8)))
17747		svint8_t svqdmulh_n_s8(svint8_t, int8_t);
17748		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s32)))
17749		svint32_t svqdmulh_n_s32(svint32_t, int32_t);
17750		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s64)))
17751		svint64_t svqdmulh_n_s64(svint64_t, int64_t);
17752		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s16)))
17753		svint16_t svqdmulh_n_s16(svint16_t, int16_t);
17754		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s8)))
17755		svint8_t svqdmulh_s8(svint8_t, svint8_t);
17756		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s32)))
17757		svint32_t svqdmulh_s32(svint32_t, svint32_t);
17758		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s64)))
17759		svint64_t svqdmulh_s64(svint64_t, svint64_t);
17760		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s16)))
17761		svint16_t svqdmulh_s16(svint16_t, svint16_t);
17762		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s32)))
17763		svint32_t svqdmulh_lane_s32(svint32_t, svint32_t, uint64_t);
17764		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s64)))
17765		svint64_t svqdmulh_lane_s64(svint64_t, svint64_t, uint64_t);
17766		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s16)))
17767		svint16_t svqdmulh_lane_s16(svint16_t, svint16_t, uint64_t);
17768		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s32)))
17769		svint32_t svqdmullb_n_s32(svint16_t, int16_t);
17770		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s64)))
17771		svint64_t svqdmullb_n_s64(svint32_t, int32_t);
17772		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s16)))
17773		svint16_t svqdmullb_n_s16(svint8_t, int8_t);
17774		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s32)))
17775		svint32_t svqdmullb_s32(svint16_t, svint16_t);
17776		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s64)))
17777		svint64_t svqdmullb_s64(svint32_t, svint32_t);
17778		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s16)))
17779		svint16_t svqdmullb_s16(svint8_t, svint8_t);
17780		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s32)))
17781		svint32_t svqdmullb_lane_s32(svint16_t, svint16_t, uint64_t);
17782		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s64)))
17783		svint64_t svqdmullb_lane_s64(svint32_t, svint32_t, uint64_t);
17784		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s32)))
17785		svint32_t svqdmullt_n_s32(svint16_t, int16_t);
17786		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s64)))
17787		svint64_t svqdmullt_n_s64(svint32_t, int32_t);
17788		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s16)))
17789		svint16_t svqdmullt_n_s16(svint8_t, int8_t);
17790		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s32)))
17791		svint32_t svqdmullt_s32(svint16_t, svint16_t);
17792		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s64)))
17793		svint64_t svqdmullt_s64(svint32_t, svint32_t);
17794		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s16)))
17795		svint16_t svqdmullt_s16(svint8_t, svint8_t);
17796		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s32)))
17797		svint32_t svqdmullt_lane_s32(svint16_t, svint16_t, uint64_t);
17798		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s64)))
17799		svint64_t svqdmullt_lane_s64(svint32_t, svint32_t, uint64_t);
17800		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_m)))
17801		svint8_t svqneg_s8_m(svint8_t, svbool_t, svint8_t);
17802		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_m)))
17803		svint32_t svqneg_s32_m(svint32_t, svbool_t, svint32_t);
17804		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_m)))
17805		svint64_t svqneg_s64_m(svint64_t, svbool_t, svint64_t);
17806		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_m)))
17807		svint16_t svqneg_s16_m(svint16_t, svbool_t, svint16_t);
17808		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_x)))
17809		svint8_t svqneg_s8_x(svbool_t, svint8_t);
17810		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_x)))
17811		svint32_t svqneg_s32_x(svbool_t, svint32_t);
17812		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_x)))
17813		svint64_t svqneg_s64_x(svbool_t, svint64_t);
17814		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_x)))
17815		svint16_t svqneg_s16_x(svbool_t, svint16_t);
17816		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_z)))
17817		svint8_t svqneg_s8_z(svbool_t, svint8_t);
17818		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_z)))
17819		svint32_t svqneg_s32_z(svbool_t, svint32_t);
17820		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_z)))
17821		svint64_t svqneg_s64_z(svbool_t, svint64_t);
17822		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_z)))
17823		svint16_t svqneg_s16_z(svbool_t, svint16_t);
17824		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s8)))
17825		svint8_t svqrdcmlah_s8(svint8_t, svint8_t, svint8_t, uint64_t);
17826		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s32)))
17827		svint32_t svqrdcmlah_s32(svint32_t, svint32_t, svint32_t, uint64_t);
17828		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s64)))
17829		svint64_t svqrdcmlah_s64(svint64_t, svint64_t, svint64_t, uint64_t);
17830		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s16)))
17831		svint16_t svqrdcmlah_s16(svint16_t, svint16_t, svint16_t, uint64_t);
17832		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s32)))
17833		svint32_t svqrdcmlah_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
17834		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s16)))
17835		svint16_t svqrdcmlah_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
17836		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s8)))
17837		svint8_t svqrdmlah_n_s8(svint8_t, svint8_t, int8_t);
17838		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s32)))
17839		svint32_t svqrdmlah_n_s32(svint32_t, svint32_t, int32_t);
17840		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s64)))
17841		svint64_t svqrdmlah_n_s64(svint64_t, svint64_t, int64_t);
17842		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s16)))
17843		svint16_t svqrdmlah_n_s16(svint16_t, svint16_t, int16_t);
17844		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s8)))
17845		svint8_t svqrdmlah_s8(svint8_t, svint8_t, svint8_t);
17846		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s32)))
17847		svint32_t svqrdmlah_s32(svint32_t, svint32_t, svint32_t);
17848		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s64)))
17849		svint64_t svqrdmlah_s64(svint64_t, svint64_t, svint64_t);
17850		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s16)))
17851		svint16_t svqrdmlah_s16(svint16_t, svint16_t, svint16_t);
17852		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s32)))
17853		svint32_t svqrdmlah_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
17854		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s64)))
17855		svint64_t svqrdmlah_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
17856		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s16)))
17857		svint16_t svqrdmlah_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
17858		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s8)))
17859		svint8_t svqrdmlsh_n_s8(svint8_t, svint8_t, int8_t);
17860		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s32)))
17861		svint32_t svqrdmlsh_n_s32(svint32_t, svint32_t, int32_t);
17862		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s64)))
17863		svint64_t svqrdmlsh_n_s64(svint64_t, svint64_t, int64_t);
17864		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s16)))
17865		svint16_t svqrdmlsh_n_s16(svint16_t, svint16_t, int16_t);
17866		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s8)))
17867		svint8_t svqrdmlsh_s8(svint8_t, svint8_t, svint8_t);
17868		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s32)))
17869		svint32_t svqrdmlsh_s32(svint32_t, svint32_t, svint32_t);
17870		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s64)))
17871		svint64_t svqrdmlsh_s64(svint64_t, svint64_t, svint64_t);
17872		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s16)))
17873		svint16_t svqrdmlsh_s16(svint16_t, svint16_t, svint16_t);
17874		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s32)))
17875		svint32_t svqrdmlsh_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
17876		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s64)))
17877		svint64_t svqrdmlsh_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
17878		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s16)))
17879		svint16_t svqrdmlsh_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
17880		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s8)))
17881		svint8_t svqrdmulh_n_s8(svint8_t, int8_t);
17882		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s32)))
17883		svint32_t svqrdmulh_n_s32(svint32_t, int32_t);
17884		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s64)))
17885		svint64_t svqrdmulh_n_s64(svint64_t, int64_t);
17886		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s16)))
17887		svint16_t svqrdmulh_n_s16(svint16_t, int16_t);
17888		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s8)))
17889		svint8_t svqrdmulh_s8(svint8_t, svint8_t);
17890		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s32)))
17891		svint32_t svqrdmulh_s32(svint32_t, svint32_t);
17892		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s64)))
17893		svint64_t svqrdmulh_s64(svint64_t, svint64_t);
17894		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s16)))
17895		svint16_t svqrdmulh_s16(svint16_t, svint16_t);
17896		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s32)))
17897		svint32_t svqrdmulh_lane_s32(svint32_t, svint32_t, uint64_t);
17898		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s64)))
17899		svint64_t svqrdmulh_lane_s64(svint64_t, svint64_t, uint64_t);
17900		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s16)))
17901		svint16_t svqrdmulh_lane_s16(svint16_t, svint16_t, uint64_t);
17902		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_m)))
17903		svint8_t svqrshl_n_s8_m(svbool_t, svint8_t, int8_t);
17904		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_m)))
17905		svint32_t svqrshl_n_s32_m(svbool_t, svint32_t, int32_t);
17906		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_m)))
17907		svint64_t svqrshl_n_s64_m(svbool_t, svint64_t, int64_t);
17908		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_m)))
17909		svint16_t svqrshl_n_s16_m(svbool_t, svint16_t, int16_t);
17910		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_x)))
17911		svint8_t svqrshl_n_s8_x(svbool_t, svint8_t, int8_t);
17912		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_x)))
17913		svint32_t svqrshl_n_s32_x(svbool_t, svint32_t, int32_t);
17914		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_x)))
17915		svint64_t svqrshl_n_s64_x(svbool_t, svint64_t, int64_t);
17916		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_x)))
17917		svint16_t svqrshl_n_s16_x(svbool_t, svint16_t, int16_t);
17918		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_z)))
17919		svint8_t svqrshl_n_s8_z(svbool_t, svint8_t, int8_t);
17920		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_z)))
17921		svint32_t svqrshl_n_s32_z(svbool_t, svint32_t, int32_t);
17922		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_z)))
17923		svint64_t svqrshl_n_s64_z(svbool_t, svint64_t, int64_t);
17924		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_z)))
17925		svint16_t svqrshl_n_s16_z(svbool_t, svint16_t, int16_t);
17926		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_m)))
17927		svuint8_t svqrshl_n_u8_m(svbool_t, svuint8_t, int8_t);
17928		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_m)))
17929		svuint32_t svqrshl_n_u32_m(svbool_t, svuint32_t, int32_t);
17930		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_m)))
17931		svuint64_t svqrshl_n_u64_m(svbool_t, svuint64_t, int64_t);
17932		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_m)))
17933		svuint16_t svqrshl_n_u16_m(svbool_t, svuint16_t, int16_t);
17934		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_x)))
17935		svuint8_t svqrshl_n_u8_x(svbool_t, svuint8_t, int8_t);
17936		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_x)))
17937		svuint32_t svqrshl_n_u32_x(svbool_t, svuint32_t, int32_t);
17938		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_x)))
17939		svuint64_t svqrshl_n_u64_x(svbool_t, svuint64_t, int64_t);
17940		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_x)))
17941		svuint16_t svqrshl_n_u16_x(svbool_t, svuint16_t, int16_t);
17942		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_z)))
17943		svuint8_t svqrshl_n_u8_z(svbool_t, svuint8_t, int8_t);
17944		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_z)))
17945		svuint32_t svqrshl_n_u32_z(svbool_t, svuint32_t, int32_t);
17946		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_z)))
17947		svuint64_t svqrshl_n_u64_z(svbool_t, svuint64_t, int64_t);
17948		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_z)))
17949		svuint16_t svqrshl_n_u16_z(svbool_t, svuint16_t, int16_t);
17950		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_m)))
17951		svint8_t svqrshl_s8_m(svbool_t, svint8_t, svint8_t);
17952		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_m)))
17953		svint32_t svqrshl_s32_m(svbool_t, svint32_t, svint32_t);
17954		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_m)))
17955		svint64_t svqrshl_s64_m(svbool_t, svint64_t, svint64_t);
17956		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_m)))
17957		svint16_t svqrshl_s16_m(svbool_t, svint16_t, svint16_t);
17958		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_x)))
17959		svint8_t svqrshl_s8_x(svbool_t, svint8_t, svint8_t);
17960		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_x)))
17961		svint32_t svqrshl_s32_x(svbool_t, svint32_t, svint32_t);
17962		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_x)))
17963		svint64_t svqrshl_s64_x(svbool_t, svint64_t, svint64_t);
17964		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_x)))
17965		svint16_t svqrshl_s16_x(svbool_t, svint16_t, svint16_t);
17966		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_z)))
17967		svint8_t svqrshl_s8_z(svbool_t, svint8_t, svint8_t);
17968		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_z)))
17969		svint32_t svqrshl_s32_z(svbool_t, svint32_t, svint32_t);
17970		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_z)))
17971		svint64_t svqrshl_s64_z(svbool_t, svint64_t, svint64_t);
17972		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_z)))
17973		svint16_t svqrshl_s16_z(svbool_t, svint16_t, svint16_t);
17974		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_m)))
17975		svuint8_t svqrshl_u8_m(svbool_t, svuint8_t, svint8_t);
17976		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_m)))
17977		svuint32_t svqrshl_u32_m(svbool_t, svuint32_t, svint32_t);
17978		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_m)))
17979		svuint64_t svqrshl_u64_m(svbool_t, svuint64_t, svint64_t);
17980		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_m)))
17981		svuint16_t svqrshl_u16_m(svbool_t, svuint16_t, svint16_t);
17982		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_x)))
17983		svuint8_t svqrshl_u8_x(svbool_t, svuint8_t, svint8_t);
17984		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_x)))
17985		svuint32_t svqrshl_u32_x(svbool_t, svuint32_t, svint32_t);
17986		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_x)))
17987		svuint64_t svqrshl_u64_x(svbool_t, svuint64_t, svint64_t);
17988		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_x)))
17989		svuint16_t svqrshl_u16_x(svbool_t, svuint16_t, svint16_t);
17990		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_z)))
17991		svuint8_t svqrshl_u8_z(svbool_t, svuint8_t, svint8_t);
17992		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_z)))
17993		svuint32_t svqrshl_u32_z(svbool_t, svuint32_t, svint32_t);
17994		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_z)))
17995		svuint64_t svqrshl_u64_z(svbool_t, svuint64_t, svint64_t);
17996		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_z)))
17997		svuint16_t svqrshl_u16_z(svbool_t, svuint16_t, svint16_t);
17998		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s32)))
17999		svint16_t svqrshrnb_n_s32(svint32_t, uint64_t);
18000		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s64)))
18001		svint32_t svqrshrnb_n_s64(svint64_t, uint64_t);
18002		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s16)))
18003		svint8_t svqrshrnb_n_s16(svint16_t, uint64_t);
18004		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u32)))
18005		svuint16_t svqrshrnb_n_u32(svuint32_t, uint64_t);
18006		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u64)))
18007		svuint32_t svqrshrnb_n_u64(svuint64_t, uint64_t);
18008		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u16)))
18009		svuint8_t svqrshrnb_n_u16(svuint16_t, uint64_t);
18010		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s32)))
18011		svint16_t svqrshrnt_n_s32(svint16_t, svint32_t, uint64_t);
18012		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s64)))
18013		svint32_t svqrshrnt_n_s64(svint32_t, svint64_t, uint64_t);
18014		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s16)))
18015		svint8_t svqrshrnt_n_s16(svint8_t, svint16_t, uint64_t);
18016		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u32)))
18017		svuint16_t svqrshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
18018		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u64)))
18019		svuint32_t svqrshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
18020		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u16)))
18021		svuint8_t svqrshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
18022		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s32)))
18023		svuint16_t svqrshrunb_n_s32(svint32_t, uint64_t);
18024		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s64)))
18025		svuint32_t svqrshrunb_n_s64(svint64_t, uint64_t);
18026		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s16)))
18027		svuint8_t svqrshrunb_n_s16(svint16_t, uint64_t);
18028		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s32)))
18029		svuint16_t svqrshrunt_n_s32(svuint16_t, svint32_t, uint64_t);
18030		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s64)))
18031		svuint32_t svqrshrunt_n_s64(svuint32_t, svint64_t, uint64_t);
18032		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s16)))
18033		svuint8_t svqrshrunt_n_s16(svuint8_t, svint16_t, uint64_t);
18034		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_m)))
18035		svint8_t svqshl_n_s8_m(svbool_t, svint8_t, int8_t);
18036		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_m)))
18037		svint32_t svqshl_n_s32_m(svbool_t, svint32_t, int32_t);
18038		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_m)))
18039		svint64_t svqshl_n_s64_m(svbool_t, svint64_t, int64_t);
18040		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_m)))
18041		svint16_t svqshl_n_s16_m(svbool_t, svint16_t, int16_t);
18042		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_x)))
18043		svint8_t svqshl_n_s8_x(svbool_t, svint8_t, int8_t);
18044		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_x)))
18045		svint32_t svqshl_n_s32_x(svbool_t, svint32_t, int32_t);
18046		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_x)))
18047		svint64_t svqshl_n_s64_x(svbool_t, svint64_t, int64_t);
18048		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_x)))
18049		svint16_t svqshl_n_s16_x(svbool_t, svint16_t, int16_t);
18050		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_z)))
18051		svint8_t svqshl_n_s8_z(svbool_t, svint8_t, int8_t);
18052		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_z)))
18053		svint32_t svqshl_n_s32_z(svbool_t, svint32_t, int32_t);
18054		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_z)))
18055		svint64_t svqshl_n_s64_z(svbool_t, svint64_t, int64_t);
18056		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_z)))
18057		svint16_t svqshl_n_s16_z(svbool_t, svint16_t, int16_t);
18058		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_m)))
18059		svuint8_t svqshl_n_u8_m(svbool_t, svuint8_t, int8_t);
18060		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_m)))
18061		svuint32_t svqshl_n_u32_m(svbool_t, svuint32_t, int32_t);
18062		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_m)))
18063		svuint64_t svqshl_n_u64_m(svbool_t, svuint64_t, int64_t);
18064		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_m)))
18065		svuint16_t svqshl_n_u16_m(svbool_t, svuint16_t, int16_t);
18066		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_x)))
18067		svuint8_t svqshl_n_u8_x(svbool_t, svuint8_t, int8_t);
18068		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_x)))
18069		svuint32_t svqshl_n_u32_x(svbool_t, svuint32_t, int32_t);
18070		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_x)))
18071		svuint64_t svqshl_n_u64_x(svbool_t, svuint64_t, int64_t);
18072		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_x)))
18073		svuint16_t svqshl_n_u16_x(svbool_t, svuint16_t, int16_t);
18074		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_z)))
18075		svuint8_t svqshl_n_u8_z(svbool_t, svuint8_t, int8_t);
18076		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_z)))
18077		svuint32_t svqshl_n_u32_z(svbool_t, svuint32_t, int32_t);
18078		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_z)))
18079		svuint64_t svqshl_n_u64_z(svbool_t, svuint64_t, int64_t);
18080		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_z)))
18081		svuint16_t svqshl_n_u16_z(svbool_t, svuint16_t, int16_t);
18082		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_m)))
18083		svint8_t svqshl_s8_m(svbool_t, svint8_t, svint8_t);
18084		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_m)))
18085		svint32_t svqshl_s32_m(svbool_t, svint32_t, svint32_t);
18086		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_m)))
18087		svint64_t svqshl_s64_m(svbool_t, svint64_t, svint64_t);
18088		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_m)))
18089		svint16_t svqshl_s16_m(svbool_t, svint16_t, svint16_t);
18090		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_x)))
18091		svint8_t svqshl_s8_x(svbool_t, svint8_t, svint8_t);
18092		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_x)))
18093		svint32_t svqshl_s32_x(svbool_t, svint32_t, svint32_t);
18094		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_x)))
18095		svint64_t svqshl_s64_x(svbool_t, svint64_t, svint64_t);
18096		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_x)))
18097		svint16_t svqshl_s16_x(svbool_t, svint16_t, svint16_t);
18098		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_z)))
18099		svint8_t svqshl_s8_z(svbool_t, svint8_t, svint8_t);
18100		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_z)))
18101		svint32_t svqshl_s32_z(svbool_t, svint32_t, svint32_t);
18102		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_z)))
18103		svint64_t svqshl_s64_z(svbool_t, svint64_t, svint64_t);
18104		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_z)))
18105		svint16_t svqshl_s16_z(svbool_t, svint16_t, svint16_t);
18106		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_m)))
18107		svuint8_t svqshl_u8_m(svbool_t, svuint8_t, svint8_t);
18108		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_m)))
18109		svuint32_t svqshl_u32_m(svbool_t, svuint32_t, svint32_t);
18110		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_m)))
18111		svuint64_t svqshl_u64_m(svbool_t, svuint64_t, svint64_t);
18112		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_m)))
18113		svuint16_t svqshl_u16_m(svbool_t, svuint16_t, svint16_t);
18114		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_x)))
18115		svuint8_t svqshl_u8_x(svbool_t, svuint8_t, svint8_t);
18116		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_x)))
18117		svuint32_t svqshl_u32_x(svbool_t, svuint32_t, svint32_t);
18118		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_x)))
18119		svuint64_t svqshl_u64_x(svbool_t, svuint64_t, svint64_t);
18120		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_x)))
18121		svuint16_t svqshl_u16_x(svbool_t, svuint16_t, svint16_t);
18122		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_z)))
18123		svuint8_t svqshl_u8_z(svbool_t, svuint8_t, svint8_t);
18124		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_z)))
18125		svuint32_t svqshl_u32_z(svbool_t, svuint32_t, svint32_t);
18126		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_z)))
18127		svuint64_t svqshl_u64_z(svbool_t, svuint64_t, svint64_t);
18128		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_z)))
18129		svuint16_t svqshl_u16_z(svbool_t, svuint16_t, svint16_t);
18130		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_m)))
18131		svuint8_t svqshlu_n_s8_m(svbool_t, svint8_t, uint64_t);
18132		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_m)))
18133		svuint32_t svqshlu_n_s32_m(svbool_t, svint32_t, uint64_t);
18134		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_m)))
18135		svuint64_t svqshlu_n_s64_m(svbool_t, svint64_t, uint64_t);
18136		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_m)))
18137		svuint16_t svqshlu_n_s16_m(svbool_t, svint16_t, uint64_t);
18138		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_x)))
18139		svuint8_t svqshlu_n_s8_x(svbool_t, svint8_t, uint64_t);
18140		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_x)))
18141		svuint32_t svqshlu_n_s32_x(svbool_t, svint32_t, uint64_t);
18142		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_x)))
18143		svuint64_t svqshlu_n_s64_x(svbool_t, svint64_t, uint64_t);
18144		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_x)))
18145		svuint16_t svqshlu_n_s16_x(svbool_t, svint16_t, uint64_t);
18146		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_z)))
18147		svuint8_t svqshlu_n_s8_z(svbool_t, svint8_t, uint64_t);
18148		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_z)))
18149		svuint32_t svqshlu_n_s32_z(svbool_t, svint32_t, uint64_t);
18150		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_z)))
18151		svuint64_t svqshlu_n_s64_z(svbool_t, svint64_t, uint64_t);
18152		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_z)))
18153		svuint16_t svqshlu_n_s16_z(svbool_t, svint16_t, uint64_t);
18154		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s32)))
18155		svint16_t svqshrnb_n_s32(svint32_t, uint64_t);
18156		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s64)))
18157		svint32_t svqshrnb_n_s64(svint64_t, uint64_t);
18158		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s16)))
18159		svint8_t svqshrnb_n_s16(svint16_t, uint64_t);
18160		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u32)))
18161		svuint16_t svqshrnb_n_u32(svuint32_t, uint64_t);
18162		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u64)))
18163		svuint32_t svqshrnb_n_u64(svuint64_t, uint64_t);
18164		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u16)))
18165		svuint8_t svqshrnb_n_u16(svuint16_t, uint64_t);
18166		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s32)))
18167		svint16_t svqshrnt_n_s32(svint16_t, svint32_t, uint64_t);
18168		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s64)))
18169		svint32_t svqshrnt_n_s64(svint32_t, svint64_t, uint64_t);
18170		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s16)))
18171		svint8_t svqshrnt_n_s16(svint8_t, svint16_t, uint64_t);
18172		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u32)))
18173		svuint16_t svqshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
18174		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u64)))
18175		svuint32_t svqshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
18176		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u16)))
18177		svuint8_t svqshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
18178		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s32)))
18179		svuint16_t svqshrunb_n_s32(svint32_t, uint64_t);
18180		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s64)))
18181		svuint32_t svqshrunb_n_s64(svint64_t, uint64_t);
18182		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s16)))
18183		svuint8_t svqshrunb_n_s16(svint16_t, uint64_t);
18184		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s32)))
18185		svuint16_t svqshrunt_n_s32(svuint16_t, svint32_t, uint64_t);
18186		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s64)))
18187		svuint32_t svqshrunt_n_s64(svuint32_t, svint64_t, uint64_t);
18188		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s16)))
18189		svuint8_t svqshrunt_n_s16(svuint8_t, svint16_t, uint64_t);
18190		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_m)))
18191		svint8_t svqsub_n_s8_m(svbool_t, svint8_t, int8_t);
18192		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_m)))
18193		svint32_t svqsub_n_s32_m(svbool_t, svint32_t, int32_t);
18194		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_m)))
18195		svint64_t svqsub_n_s64_m(svbool_t, svint64_t, int64_t);
18196		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_m)))
18197		svint16_t svqsub_n_s16_m(svbool_t, svint16_t, int16_t);
18198		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_x)))
18199		svint8_t svqsub_n_s8_x(svbool_t, svint8_t, int8_t);
18200		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_x)))
18201		svint32_t svqsub_n_s32_x(svbool_t, svint32_t, int32_t);
18202		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_x)))
18203		svint64_t svqsub_n_s64_x(svbool_t, svint64_t, int64_t);
18204		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_x)))
18205		svint16_t svqsub_n_s16_x(svbool_t, svint16_t, int16_t);
18206		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_z)))
18207		svint8_t svqsub_n_s8_z(svbool_t, svint8_t, int8_t);
18208		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_z)))
18209		svint32_t svqsub_n_s32_z(svbool_t, svint32_t, int32_t);
18210		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_z)))
18211		svint64_t svqsub_n_s64_z(svbool_t, svint64_t, int64_t);
18212		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_z)))
18213		svint16_t svqsub_n_s16_z(svbool_t, svint16_t, int16_t);
18214		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_m)))
18215		svuint8_t svqsub_n_u8_m(svbool_t, svuint8_t, uint8_t);
18216		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_m)))
18217		svuint32_t svqsub_n_u32_m(svbool_t, svuint32_t, uint32_t);
18218		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_m)))
18219		svuint64_t svqsub_n_u64_m(svbool_t, svuint64_t, uint64_t);
18220		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_m)))
18221		svuint16_t svqsub_n_u16_m(svbool_t, svuint16_t, uint16_t);
18222		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_x)))
18223		svuint8_t svqsub_n_u8_x(svbool_t, svuint8_t, uint8_t);
18224		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_x)))
18225		svuint32_t svqsub_n_u32_x(svbool_t, svuint32_t, uint32_t);
18226		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_x)))
18227		svuint64_t svqsub_n_u64_x(svbool_t, svuint64_t, uint64_t);
18228		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_x)))
18229		svuint16_t svqsub_n_u16_x(svbool_t, svuint16_t, uint16_t);
18230		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_z)))
18231		svuint8_t svqsub_n_u8_z(svbool_t, svuint8_t, uint8_t);
18232		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_z)))
18233		svuint32_t svqsub_n_u32_z(svbool_t, svuint32_t, uint32_t);
18234		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_z)))
18235		svuint64_t svqsub_n_u64_z(svbool_t, svuint64_t, uint64_t);
18236		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_z)))
18237		svuint16_t svqsub_n_u16_z(svbool_t, svuint16_t, uint16_t);
18238		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_m)))
18239		svint8_t svqsub_s8_m(svbool_t, svint8_t, svint8_t);
18240		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_m)))
18241		svint32_t svqsub_s32_m(svbool_t, svint32_t, svint32_t);
18242		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_m)))
18243		svint64_t svqsub_s64_m(svbool_t, svint64_t, svint64_t);
18244		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_m)))
18245		svint16_t svqsub_s16_m(svbool_t, svint16_t, svint16_t);
18246		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_x)))
18247		svint8_t svqsub_s8_x(svbool_t, svint8_t, svint8_t);
18248		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_x)))
18249		svint32_t svqsub_s32_x(svbool_t, svint32_t, svint32_t);
18250		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_x)))
18251		svint64_t svqsub_s64_x(svbool_t, svint64_t, svint64_t);
18252		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_x)))
18253		svint16_t svqsub_s16_x(svbool_t, svint16_t, svint16_t);
18254		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_z)))
18255		svint8_t svqsub_s8_z(svbool_t, svint8_t, svint8_t);
18256		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_z)))
18257		svint32_t svqsub_s32_z(svbool_t, svint32_t, svint32_t);
18258		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_z)))
18259		svint64_t svqsub_s64_z(svbool_t, svint64_t, svint64_t);
18260		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_z)))
18261		svint16_t svqsub_s16_z(svbool_t, svint16_t, svint16_t);
18262		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_m)))
18263		svuint8_t svqsub_u8_m(svbool_t, svuint8_t, svuint8_t);
18264		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_m)))
18265		svuint32_t svqsub_u32_m(svbool_t, svuint32_t, svuint32_t);
18266		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_m)))
18267		svuint64_t svqsub_u64_m(svbool_t, svuint64_t, svuint64_t);
18268		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_m)))
18269		svuint16_t svqsub_u16_m(svbool_t, svuint16_t, svuint16_t);
18270		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_x)))
18271		svuint8_t svqsub_u8_x(svbool_t, svuint8_t, svuint8_t);
18272		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_x)))
18273		svuint32_t svqsub_u32_x(svbool_t, svuint32_t, svuint32_t);
18274		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_x)))
18275		svuint64_t svqsub_u64_x(svbool_t, svuint64_t, svuint64_t);
18276		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_x)))
18277		svuint16_t svqsub_u16_x(svbool_t, svuint16_t, svuint16_t);
18278		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_z)))
18279		svuint8_t svqsub_u8_z(svbool_t, svuint8_t, svuint8_t);
18280		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_z)))
18281		svuint32_t svqsub_u32_z(svbool_t, svuint32_t, svuint32_t);
18282		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_z)))
18283		svuint64_t svqsub_u64_z(svbool_t, svuint64_t, svuint64_t);
18284		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_z)))
18285		svuint16_t svqsub_u16_z(svbool_t, svuint16_t, svuint16_t);
18286		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_m)))
18287		svint8_t svqsubr_n_s8_m(svbool_t, svint8_t, int8_t);
18288		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_m)))
18289		svint32_t svqsubr_n_s32_m(svbool_t, svint32_t, int32_t);
18290		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_m)))
18291		svint64_t svqsubr_n_s64_m(svbool_t, svint64_t, int64_t);
18292		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_m)))
18293		svint16_t svqsubr_n_s16_m(svbool_t, svint16_t, int16_t);
18294		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_x)))
18295		svint8_t svqsubr_n_s8_x(svbool_t, svint8_t, int8_t);
18296		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_x)))
18297		svint32_t svqsubr_n_s32_x(svbool_t, svint32_t, int32_t);
18298		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_x)))
18299		svint64_t svqsubr_n_s64_x(svbool_t, svint64_t, int64_t);
18300		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_x)))
18301		svint16_t svqsubr_n_s16_x(svbool_t, svint16_t, int16_t);
18302		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_z)))
18303		svint8_t svqsubr_n_s8_z(svbool_t, svint8_t, int8_t);
18304		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_z)))
18305		svint32_t svqsubr_n_s32_z(svbool_t, svint32_t, int32_t);
18306		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_z)))
18307		svint64_t svqsubr_n_s64_z(svbool_t, svint64_t, int64_t);
18308		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_z)))
18309		svint16_t svqsubr_n_s16_z(svbool_t, svint16_t, int16_t);
18310		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_m)))
18311		svuint8_t svqsubr_n_u8_m(svbool_t, svuint8_t, uint8_t);
18312		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_m)))
18313		svuint32_t svqsubr_n_u32_m(svbool_t, svuint32_t, uint32_t);
18314		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_m)))
18315		svuint64_t svqsubr_n_u64_m(svbool_t, svuint64_t, uint64_t);
18316		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_m)))
18317		svuint16_t svqsubr_n_u16_m(svbool_t, svuint16_t, uint16_t);
18318		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_x)))
18319		svuint8_t svqsubr_n_u8_x(svbool_t, svuint8_t, uint8_t);
18320		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_x)))
18321		svuint32_t svqsubr_n_u32_x(svbool_t, svuint32_t, uint32_t);
18322		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_x)))
18323		svuint64_t svqsubr_n_u64_x(svbool_t, svuint64_t, uint64_t);
18324		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_x)))
18325		svuint16_t svqsubr_n_u16_x(svbool_t, svuint16_t, uint16_t);
18326		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_z)))
18327		svuint8_t svqsubr_n_u8_z(svbool_t, svuint8_t, uint8_t);
18328		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_z)))
18329		svuint32_t svqsubr_n_u32_z(svbool_t, svuint32_t, uint32_t);
18330		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_z)))
18331		svuint64_t svqsubr_n_u64_z(svbool_t, svuint64_t, uint64_t);
18332		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_z)))
18333		svuint16_t svqsubr_n_u16_z(svbool_t, svuint16_t, uint16_t);
18334		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_m)))
18335		svint8_t svqsubr_s8_m(svbool_t, svint8_t, svint8_t);
18336		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_m)))
18337		svint32_t svqsubr_s32_m(svbool_t, svint32_t, svint32_t);
18338		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_m)))
18339		svint64_t svqsubr_s64_m(svbool_t, svint64_t, svint64_t);
18340		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_m)))
18341		svint16_t svqsubr_s16_m(svbool_t, svint16_t, svint16_t);
18342		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_x)))
18343		svint8_t svqsubr_s8_x(svbool_t, svint8_t, svint8_t);
18344		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_x)))
18345		svint32_t svqsubr_s32_x(svbool_t, svint32_t, svint32_t);
18346		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_x)))
18347		svint64_t svqsubr_s64_x(svbool_t, svint64_t, svint64_t);
18348		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_x)))
18349		svint16_t svqsubr_s16_x(svbool_t, svint16_t, svint16_t);
18350		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_z)))
18351		svint8_t svqsubr_s8_z(svbool_t, svint8_t, svint8_t);
18352		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_z)))
18353		svint32_t svqsubr_s32_z(svbool_t, svint32_t, svint32_t);
18354		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_z)))
18355		svint64_t svqsubr_s64_z(svbool_t, svint64_t, svint64_t);
18356		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_z)))
18357		svint16_t svqsubr_s16_z(svbool_t, svint16_t, svint16_t);
18358		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_m)))
18359		svuint8_t svqsubr_u8_m(svbool_t, svuint8_t, svuint8_t);
18360		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_m)))
18361		svuint32_t svqsubr_u32_m(svbool_t, svuint32_t, svuint32_t);
18362		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_m)))
18363		svuint64_t svqsubr_u64_m(svbool_t, svuint64_t, svuint64_t);
18364		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_m)))
18365		svuint16_t svqsubr_u16_m(svbool_t, svuint16_t, svuint16_t);
18366		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_x)))
18367		svuint8_t svqsubr_u8_x(svbool_t, svuint8_t, svuint8_t);
18368		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_x)))
18369		svuint32_t svqsubr_u32_x(svbool_t, svuint32_t, svuint32_t);
18370		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_x)))
18371		svuint64_t svqsubr_u64_x(svbool_t, svuint64_t, svuint64_t);
18372		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_x)))
18373		svuint16_t svqsubr_u16_x(svbool_t, svuint16_t, svuint16_t);
18374		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_z)))
18375		svuint8_t svqsubr_u8_z(svbool_t, svuint8_t, svuint8_t);
18376		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_z)))
18377		svuint32_t svqsubr_u32_z(svbool_t, svuint32_t, svuint32_t);
18378		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_z)))
18379		svuint64_t svqsubr_u64_z(svbool_t, svuint64_t, svuint64_t);
18380		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_z)))
18381		svuint16_t svqsubr_u16_z(svbool_t, svuint16_t, svuint16_t);
18382		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s32)))
18383		svint16_t svqxtnb_s32(svint32_t);
18384		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s64)))
18385		svint32_t svqxtnb_s64(svint64_t);
18386		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s16)))
18387		svint8_t svqxtnb_s16(svint16_t);
18388		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u32)))
18389		svuint16_t svqxtnb_u32(svuint32_t);
18390		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u64)))
18391		svuint32_t svqxtnb_u64(svuint64_t);
18392		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u16)))
18393		svuint8_t svqxtnb_u16(svuint16_t);
18394		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s32)))
18395		svint16_t svqxtnt_s32(svint16_t, svint32_t);
18396		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s64)))
18397		svint32_t svqxtnt_s64(svint32_t, svint64_t);
18398		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s16)))
18399		svint8_t svqxtnt_s16(svint8_t, svint16_t);
18400		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u32)))
18401		svuint16_t svqxtnt_u32(svuint16_t, svuint32_t);
18402		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u64)))
18403		svuint32_t svqxtnt_u64(svuint32_t, svuint64_t);
18404		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u16)))
18405		svuint8_t svqxtnt_u16(svuint8_t, svuint16_t);
18406		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s32)))
18407		svuint16_t svqxtunb_s32(svint32_t);
18408		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s64)))
18409		svuint32_t svqxtunb_s64(svint64_t);
18410		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s16)))
18411		svuint8_t svqxtunb_s16(svint16_t);
18412		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s32)))
18413		svuint16_t svqxtunt_s32(svuint16_t, svint32_t);
18414		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s64)))
18415		svuint32_t svqxtunt_s64(svuint32_t, svint64_t);
18416		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s16)))
18417		svuint8_t svqxtunt_s16(svuint8_t, svint16_t);
18418		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u32)))
18419		svuint16_t svraddhnb_n_u32(svuint32_t, uint32_t);
18420		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u64)))
18421		svuint32_t svraddhnb_n_u64(svuint64_t, uint64_t);
18422		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u16)))
18423		svuint8_t svraddhnb_n_u16(svuint16_t, uint16_t);
18424		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s32)))
18425		svint16_t svraddhnb_n_s32(svint32_t, int32_t);
18426		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s64)))
18427		svint32_t svraddhnb_n_s64(svint64_t, int64_t);
18428		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s16)))
18429		svint8_t svraddhnb_n_s16(svint16_t, int16_t);
18430		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u32)))
18431		svuint16_t svraddhnb_u32(svuint32_t, svuint32_t);
18432		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u64)))
18433		svuint32_t svraddhnb_u64(svuint64_t, svuint64_t);
18434		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u16)))
18435		svuint8_t svraddhnb_u16(svuint16_t, svuint16_t);
18436		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s32)))
18437		svint16_t svraddhnb_s32(svint32_t, svint32_t);
18438		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s64)))
18439		svint32_t svraddhnb_s64(svint64_t, svint64_t);
18440		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s16)))
18441		svint8_t svraddhnb_s16(svint16_t, svint16_t);
18442		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u32)))
18443		svuint16_t svraddhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
18444		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u64)))
18445		svuint32_t svraddhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
18446		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u16)))
18447		svuint8_t svraddhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
18448		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s32)))
18449		svint16_t svraddhnt_n_s32(svint16_t, svint32_t, int32_t);
18450		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s64)))
18451		svint32_t svraddhnt_n_s64(svint32_t, svint64_t, int64_t);
18452		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s16)))
18453		svint8_t svraddhnt_n_s16(svint8_t, svint16_t, int16_t);
18454		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u32)))
18455		svuint16_t svraddhnt_u32(svuint16_t, svuint32_t, svuint32_t);
18456		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u64)))
18457		svuint32_t svraddhnt_u64(svuint32_t, svuint64_t, svuint64_t);
18458		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u16)))
18459		svuint8_t svraddhnt_u16(svuint8_t, svuint16_t, svuint16_t);
18460		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s32)))
18461		svint16_t svraddhnt_s32(svint16_t, svint32_t, svint32_t);
18462		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s64)))
18463		svint32_t svraddhnt_s64(svint32_t, svint64_t, svint64_t);
18464		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s16)))
18465		svint8_t svraddhnt_s16(svint8_t, svint16_t, svint16_t);
18466		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_m)))
18467		svuint32_t svrecpe_u32_m(svuint32_t, svbool_t, svuint32_t);
18468		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_x)))
18469		svuint32_t svrecpe_u32_x(svbool_t, svuint32_t);
18470		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_z)))
18471		svuint32_t svrecpe_u32_z(svbool_t, svuint32_t);
18472		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_m)))
18473		svint8_t svrhadd_n_s8_m(svbool_t, svint8_t, int8_t);
18474		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_m)))
18475		svint32_t svrhadd_n_s32_m(svbool_t, svint32_t, int32_t);
18476		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_m)))
18477		svint64_t svrhadd_n_s64_m(svbool_t, svint64_t, int64_t);
18478		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_m)))
18479		svint16_t svrhadd_n_s16_m(svbool_t, svint16_t, int16_t);
18480		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_x)))
18481		svint8_t svrhadd_n_s8_x(svbool_t, svint8_t, int8_t);
18482		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_x)))
18483		svint32_t svrhadd_n_s32_x(svbool_t, svint32_t, int32_t);
18484		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_x)))
18485		svint64_t svrhadd_n_s64_x(svbool_t, svint64_t, int64_t);
18486		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_x)))
18487		svint16_t svrhadd_n_s16_x(svbool_t, svint16_t, int16_t);
18488		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_z)))
18489		svint8_t svrhadd_n_s8_z(svbool_t, svint8_t, int8_t);
18490		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_z)))
18491		svint32_t svrhadd_n_s32_z(svbool_t, svint32_t, int32_t);
18492		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_z)))
18493		svint64_t svrhadd_n_s64_z(svbool_t, svint64_t, int64_t);
18494		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_z)))
18495		svint16_t svrhadd_n_s16_z(svbool_t, svint16_t, int16_t);
18496		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_m)))
18497		svuint8_t svrhadd_n_u8_m(svbool_t, svuint8_t, uint8_t);
18498		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_m)))
18499		svuint32_t svrhadd_n_u32_m(svbool_t, svuint32_t, uint32_t);
18500		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_m)))
18501		svuint64_t svrhadd_n_u64_m(svbool_t, svuint64_t, uint64_t);
18502		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_m)))
18503		svuint16_t svrhadd_n_u16_m(svbool_t, svuint16_t, uint16_t);
18504		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_x)))
18505		svuint8_t svrhadd_n_u8_x(svbool_t, svuint8_t, uint8_t);
18506		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_x)))
18507		svuint32_t svrhadd_n_u32_x(svbool_t, svuint32_t, uint32_t);
18508		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_x)))
18509		svuint64_t svrhadd_n_u64_x(svbool_t, svuint64_t, uint64_t);
18510		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_x)))
18511		svuint16_t svrhadd_n_u16_x(svbool_t, svuint16_t, uint16_t);
18512		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_z)))
18513		svuint8_t svrhadd_n_u8_z(svbool_t, svuint8_t, uint8_t);
18514		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_z)))
18515		svuint32_t svrhadd_n_u32_z(svbool_t, svuint32_t, uint32_t);
18516		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_z)))
18517		svuint64_t svrhadd_n_u64_z(svbool_t, svuint64_t, uint64_t);
18518		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_z)))
18519		svuint16_t svrhadd_n_u16_z(svbool_t, svuint16_t, uint16_t);
18520		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_m)))
18521		svint8_t svrhadd_s8_m(svbool_t, svint8_t, svint8_t);
18522		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_m)))
18523		svint32_t svrhadd_s32_m(svbool_t, svint32_t, svint32_t);
18524		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_m)))
18525		svint64_t svrhadd_s64_m(svbool_t, svint64_t, svint64_t);
18526		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_m)))
18527		svint16_t svrhadd_s16_m(svbool_t, svint16_t, svint16_t);
18528		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_x)))
18529		svint8_t svrhadd_s8_x(svbool_t, svint8_t, svint8_t);
18530		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_x)))
18531		svint32_t svrhadd_s32_x(svbool_t, svint32_t, svint32_t);
18532		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_x)))
18533		svint64_t svrhadd_s64_x(svbool_t, svint64_t, svint64_t);
18534		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_x)))
18535		svint16_t svrhadd_s16_x(svbool_t, svint16_t, svint16_t);
18536		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_z)))
18537		svint8_t svrhadd_s8_z(svbool_t, svint8_t, svint8_t);
18538		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_z)))
18539		svint32_t svrhadd_s32_z(svbool_t, svint32_t, svint32_t);
18540		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_z)))
18541		svint64_t svrhadd_s64_z(svbool_t, svint64_t, svint64_t);
18542		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_z)))
18543		svint16_t svrhadd_s16_z(svbool_t, svint16_t, svint16_t);
18544		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_m)))
18545		svuint8_t svrhadd_u8_m(svbool_t, svuint8_t, svuint8_t);
18546		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_m)))
18547		svuint32_t svrhadd_u32_m(svbool_t, svuint32_t, svuint32_t);
18548		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_m)))
18549		svuint64_t svrhadd_u64_m(svbool_t, svuint64_t, svuint64_t);
18550		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_m)))
18551		svuint16_t svrhadd_u16_m(svbool_t, svuint16_t, svuint16_t);
18552		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_x)))
18553		svuint8_t svrhadd_u8_x(svbool_t, svuint8_t, svuint8_t);
18554		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_x)))
18555		svuint32_t svrhadd_u32_x(svbool_t, svuint32_t, svuint32_t);
18556		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_x)))
18557		svuint64_t svrhadd_u64_x(svbool_t, svuint64_t, svuint64_t);
18558		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_x)))
18559		svuint16_t svrhadd_u16_x(svbool_t, svuint16_t, svuint16_t);
18560		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_z)))
18561		svuint8_t svrhadd_u8_z(svbool_t, svuint8_t, svuint8_t);
18562		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_z)))
18563		svuint32_t svrhadd_u32_z(svbool_t, svuint32_t, svuint32_t);
18564		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_z)))
18565		svuint64_t svrhadd_u64_z(svbool_t, svuint64_t, svuint64_t);
18566		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_z)))
18567		svuint16_t svrhadd_u16_z(svbool_t, svuint16_t, svuint16_t);
18568		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_m)))
18569		svint8_t svrshl_n_s8_m(svbool_t, svint8_t, int8_t);
18570		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_m)))
18571		svint32_t svrshl_n_s32_m(svbool_t, svint32_t, int32_t);
18572		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_m)))
18573		svint64_t svrshl_n_s64_m(svbool_t, svint64_t, int64_t);
18574		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_m)))
18575		svint16_t svrshl_n_s16_m(svbool_t, svint16_t, int16_t);
18576		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_x)))
18577		svint8_t svrshl_n_s8_x(svbool_t, svint8_t, int8_t);
18578		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_x)))
18579		svint32_t svrshl_n_s32_x(svbool_t, svint32_t, int32_t);
18580		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_x)))
18581		svint64_t svrshl_n_s64_x(svbool_t, svint64_t, int64_t);
18582		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_x)))
18583		svint16_t svrshl_n_s16_x(svbool_t, svint16_t, int16_t);
18584		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_z)))
18585		svint8_t svrshl_n_s8_z(svbool_t, svint8_t, int8_t);
18586		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_z)))
18587		svint32_t svrshl_n_s32_z(svbool_t, svint32_t, int32_t);
18588		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_z)))
18589		svint64_t svrshl_n_s64_z(svbool_t, svint64_t, int64_t);
18590		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_z)))
18591		svint16_t svrshl_n_s16_z(svbool_t, svint16_t, int16_t);
18592		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_m)))
18593		svuint8_t svrshl_n_u8_m(svbool_t, svuint8_t, int8_t);
18594		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_m)))
18595		svuint32_t svrshl_n_u32_m(svbool_t, svuint32_t, int32_t);
18596		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_m)))
18597		svuint64_t svrshl_n_u64_m(svbool_t, svuint64_t, int64_t);
18598		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_m)))
18599		svuint16_t svrshl_n_u16_m(svbool_t, svuint16_t, int16_t);
18600		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_x)))
18601		svuint8_t svrshl_n_u8_x(svbool_t, svuint8_t, int8_t);
18602		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_x)))
18603		svuint32_t svrshl_n_u32_x(svbool_t, svuint32_t, int32_t);
18604		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_x)))
18605		svuint64_t svrshl_n_u64_x(svbool_t, svuint64_t, int64_t);
18606		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_x)))
18607		svuint16_t svrshl_n_u16_x(svbool_t, svuint16_t, int16_t);
18608		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_z)))
18609		svuint8_t svrshl_n_u8_z(svbool_t, svuint8_t, int8_t);
18610		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_z)))
18611		svuint32_t svrshl_n_u32_z(svbool_t, svuint32_t, int32_t);
18612		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_z)))
18613		svuint64_t svrshl_n_u64_z(svbool_t, svuint64_t, int64_t);
18614		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_z)))
18615		svuint16_t svrshl_n_u16_z(svbool_t, svuint16_t, int16_t);
18616		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_m)))
18617		svint8_t svrshl_s8_m(svbool_t, svint8_t, svint8_t);
18618		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_m)))
18619		svint32_t svrshl_s32_m(svbool_t, svint32_t, svint32_t);
18620		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_m)))
18621		svint64_t svrshl_s64_m(svbool_t, svint64_t, svint64_t);
18622		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_m)))
18623		svint16_t svrshl_s16_m(svbool_t, svint16_t, svint16_t);
18624		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_x)))
18625		svint8_t svrshl_s8_x(svbool_t, svint8_t, svint8_t);
18626		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_x)))
18627		svint32_t svrshl_s32_x(svbool_t, svint32_t, svint32_t);
18628		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_x)))
18629		svint64_t svrshl_s64_x(svbool_t, svint64_t, svint64_t);
18630		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_x)))
18631		svint16_t svrshl_s16_x(svbool_t, svint16_t, svint16_t);
18632		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_z)))
18633		svint8_t svrshl_s8_z(svbool_t, svint8_t, svint8_t);
18634		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_z)))
18635		svint32_t svrshl_s32_z(svbool_t, svint32_t, svint32_t);
18636		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_z)))
18637		svint64_t svrshl_s64_z(svbool_t, svint64_t, svint64_t);
18638		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_z)))
18639		svint16_t svrshl_s16_z(svbool_t, svint16_t, svint16_t);
18640		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_m)))
18641		svuint8_t svrshl_u8_m(svbool_t, svuint8_t, svint8_t);
18642		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_m)))
18643		svuint32_t svrshl_u32_m(svbool_t, svuint32_t, svint32_t);
18644		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_m)))
18645		svuint64_t svrshl_u64_m(svbool_t, svuint64_t, svint64_t);
18646		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_m)))
18647		svuint16_t svrshl_u16_m(svbool_t, svuint16_t, svint16_t);
18648		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_x)))
18649		svuint8_t svrshl_u8_x(svbool_t, svuint8_t, svint8_t);
18650		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_x)))
18651		svuint32_t svrshl_u32_x(svbool_t, svuint32_t, svint32_t);
18652		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_x)))
18653		svuint64_t svrshl_u64_x(svbool_t, svuint64_t, svint64_t);
18654		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_x)))
18655		svuint16_t svrshl_u16_x(svbool_t, svuint16_t, svint16_t);
18656		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_z)))
18657		svuint8_t svrshl_u8_z(svbool_t, svuint8_t, svint8_t);
18658		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_z)))
18659		svuint32_t svrshl_u32_z(svbool_t, svuint32_t, svint32_t);
18660		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_z)))
18661		svuint64_t svrshl_u64_z(svbool_t, svuint64_t, svint64_t);
18662		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_z)))
18663		svuint16_t svrshl_u16_z(svbool_t, svuint16_t, svint16_t);
18664		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_m)))
18665		svint8_t svrshr_n_s8_m(svbool_t, svint8_t, uint64_t);
18666		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_m)))
18667		svint32_t svrshr_n_s32_m(svbool_t, svint32_t, uint64_t);
18668		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_m)))
18669		svint64_t svrshr_n_s64_m(svbool_t, svint64_t, uint64_t);
18670		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_m)))
18671		svint16_t svrshr_n_s16_m(svbool_t, svint16_t, uint64_t);
18672		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_m)))
18673		svuint8_t svrshr_n_u8_m(svbool_t, svuint8_t, uint64_t);
18674		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_m)))
18675		svuint32_t svrshr_n_u32_m(svbool_t, svuint32_t, uint64_t);
18676		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_m)))
18677		svuint64_t svrshr_n_u64_m(svbool_t, svuint64_t, uint64_t);
18678		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_m)))
18679		svuint16_t svrshr_n_u16_m(svbool_t, svuint16_t, uint64_t);
18680		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_x)))
18681		svint8_t svrshr_n_s8_x(svbool_t, svint8_t, uint64_t);
18682		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_x)))
18683		svint32_t svrshr_n_s32_x(svbool_t, svint32_t, uint64_t);
18684		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_x)))
18685		svint64_t svrshr_n_s64_x(svbool_t, svint64_t, uint64_t);
18686		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_x)))
18687		svint16_t svrshr_n_s16_x(svbool_t, svint16_t, uint64_t);
18688		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_x)))
18689		svuint8_t svrshr_n_u8_x(svbool_t, svuint8_t, uint64_t);
18690		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_x)))
18691		svuint32_t svrshr_n_u32_x(svbool_t, svuint32_t, uint64_t);
18692		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_x)))
18693		svuint64_t svrshr_n_u64_x(svbool_t, svuint64_t, uint64_t);
18694		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_x)))
18695		svuint16_t svrshr_n_u16_x(svbool_t, svuint16_t, uint64_t);
18696		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_z)))
18697		svint8_t svrshr_n_s8_z(svbool_t, svint8_t, uint64_t);
18698		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_z)))
18699		svint32_t svrshr_n_s32_z(svbool_t, svint32_t, uint64_t);
18700		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_z)))
18701		svint64_t svrshr_n_s64_z(svbool_t, svint64_t, uint64_t);
18702		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_z)))
18703		svint16_t svrshr_n_s16_z(svbool_t, svint16_t, uint64_t);
18704		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_z)))
18705		svuint8_t svrshr_n_u8_z(svbool_t, svuint8_t, uint64_t);
18706		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_z)))
18707		svuint32_t svrshr_n_u32_z(svbool_t, svuint32_t, uint64_t);
18708		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_z)))
18709		svuint64_t svrshr_n_u64_z(svbool_t, svuint64_t, uint64_t);
18710		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_z)))
18711		svuint16_t svrshr_n_u16_z(svbool_t, svuint16_t, uint64_t);
18712		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u32)))
18713		svuint16_t svrshrnb_n_u32(svuint32_t, uint64_t);
18714		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u64)))
18715		svuint32_t svrshrnb_n_u64(svuint64_t, uint64_t);
18716		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u16)))
18717		svuint8_t svrshrnb_n_u16(svuint16_t, uint64_t);
18718		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s32)))
18719		svint16_t svrshrnb_n_s32(svint32_t, uint64_t);
18720		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s64)))
18721		svint32_t svrshrnb_n_s64(svint64_t, uint64_t);
18722		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s16)))
18723		svint8_t svrshrnb_n_s16(svint16_t, uint64_t);
18724		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u32)))
18725		svuint16_t svrshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
18726		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u64)))
18727		svuint32_t svrshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
18728		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u16)))
18729		svuint8_t svrshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
18730		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s32)))
18731		svint16_t svrshrnt_n_s32(svint16_t, svint32_t, uint64_t);
18732		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s64)))
18733		svint32_t svrshrnt_n_s64(svint32_t, svint64_t, uint64_t);
18734		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s16)))
18735		svint8_t svrshrnt_n_s16(svint8_t, svint16_t, uint64_t);
18736		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_m)))
18737		svuint32_t svrsqrte_u32_m(svuint32_t, svbool_t, svuint32_t);
18738		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_x)))
18739		svuint32_t svrsqrte_u32_x(svbool_t, svuint32_t);
18740		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_z)))
18741		svuint32_t svrsqrte_u32_z(svbool_t, svuint32_t);
18742		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s8)))
18743		svint8_t svrsra_n_s8(svint8_t, svint8_t, uint64_t);
18744		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s32)))
18745		svint32_t svrsra_n_s32(svint32_t, svint32_t, uint64_t);
18746		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s64)))
18747		svint64_t svrsra_n_s64(svint64_t, svint64_t, uint64_t);
18748		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s16)))
18749		svint16_t svrsra_n_s16(svint16_t, svint16_t, uint64_t);
18750		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u8)))
18751		svuint8_t svrsra_n_u8(svuint8_t, svuint8_t, uint64_t);
18752		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u32)))
18753		svuint32_t svrsra_n_u32(svuint32_t, svuint32_t, uint64_t);
18754		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u64)))
18755		svuint64_t svrsra_n_u64(svuint64_t, svuint64_t, uint64_t);
18756		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u16)))
18757		svuint16_t svrsra_n_u16(svuint16_t, svuint16_t, uint64_t);
18758		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u32)))
18759		svuint16_t svrsubhnb_n_u32(svuint32_t, uint32_t);
18760		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u64)))
18761		svuint32_t svrsubhnb_n_u64(svuint64_t, uint64_t);
18762		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u16)))
18763		svuint8_t svrsubhnb_n_u16(svuint16_t, uint16_t);
18764		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s32)))
18765		svint16_t svrsubhnb_n_s32(svint32_t, int32_t);
18766		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s64)))
18767		svint32_t svrsubhnb_n_s64(svint64_t, int64_t);
18768		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s16)))
18769		svint8_t svrsubhnb_n_s16(svint16_t, int16_t);
18770		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u32)))
18771		svuint16_t svrsubhnb_u32(svuint32_t, svuint32_t);
18772		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u64)))
18773		svuint32_t svrsubhnb_u64(svuint64_t, svuint64_t);
18774		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u16)))
18775		svuint8_t svrsubhnb_u16(svuint16_t, svuint16_t);
18776		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s32)))
18777		svint16_t svrsubhnb_s32(svint32_t, svint32_t);
18778		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s64)))
18779		svint32_t svrsubhnb_s64(svint64_t, svint64_t);
18780		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s16)))
18781		svint8_t svrsubhnb_s16(svint16_t, svint16_t);
18782		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u32)))
18783		svuint16_t svrsubhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
18784		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u64)))
18785		svuint32_t svrsubhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
18786		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u16)))
18787		svuint8_t svrsubhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
18788		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s32)))
18789		svint16_t svrsubhnt_n_s32(svint16_t, svint32_t, int32_t);
18790		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s64)))
18791		svint32_t svrsubhnt_n_s64(svint32_t, svint64_t, int64_t);
18792		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s16)))
18793		svint8_t svrsubhnt_n_s16(svint8_t, svint16_t, int16_t);
18794		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u32)))
18795		svuint16_t svrsubhnt_u32(svuint16_t, svuint32_t, svuint32_t);
18796		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u64)))
18797		svuint32_t svrsubhnt_u64(svuint32_t, svuint64_t, svuint64_t);
18798		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u16)))
18799		svuint8_t svrsubhnt_u16(svuint8_t, svuint16_t, svuint16_t);
18800		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s32)))
18801		svint16_t svrsubhnt_s32(svint16_t, svint32_t, svint32_t);
18802		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s64)))
18803		svint32_t svrsubhnt_s64(svint32_t, svint64_t, svint64_t);
18804		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s16)))
18805		svint8_t svrsubhnt_s16(svint8_t, svint16_t, svint16_t);
18806		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u32)))
18807		svuint32_t svsbclb_n_u32(svuint32_t, svuint32_t, uint32_t);
18808		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u64)))
18809		svuint64_t svsbclb_n_u64(svuint64_t, svuint64_t, uint64_t);
18810		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u32)))
18811		svuint32_t svsbclb_u32(svuint32_t, svuint32_t, svuint32_t);
18812		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u64)))
18813		svuint64_t svsbclb_u64(svuint64_t, svuint64_t, svuint64_t);
18814		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u32)))
18815		svuint32_t svsbclt_n_u32(svuint32_t, svuint32_t, uint32_t);
18816		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u64)))
18817		svuint64_t svsbclt_n_u64(svuint64_t, svuint64_t, uint64_t);
18818		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u32)))
18819		svuint32_t svsbclt_u32(svuint32_t, svuint32_t, svuint32_t);
18820		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u64)))
18821		svuint64_t svsbclt_u64(svuint64_t, svuint64_t, svuint64_t);
18822		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s32)))
18823		svint32_t svshllb_n_s32(svint16_t, uint64_t);
18824		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s64)))
18825		svint64_t svshllb_n_s64(svint32_t, uint64_t);
18826		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s16)))
18827		svint16_t svshllb_n_s16(svint8_t, uint64_t);
18828		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u32)))
18829		svuint32_t svshllb_n_u32(svuint16_t, uint64_t);
18830		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u64)))
18831		svuint64_t svshllb_n_u64(svuint32_t, uint64_t);
18832		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u16)))
18833		svuint16_t svshllb_n_u16(svuint8_t, uint64_t);
18834		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s32)))
18835		svint32_t svshllt_n_s32(svint16_t, uint64_t);
18836		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s64)))
18837		svint64_t svshllt_n_s64(svint32_t, uint64_t);
18838		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s16)))
18839		svint16_t svshllt_n_s16(svint8_t, uint64_t);
18840		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u32)))
18841		svuint32_t svshllt_n_u32(svuint16_t, uint64_t);
18842		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u64)))
18843		svuint64_t svshllt_n_u64(svuint32_t, uint64_t);
18844		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u16)))
18845		svuint16_t svshllt_n_u16(svuint8_t, uint64_t);
18846		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u32)))
18847		svuint16_t svshrnb_n_u32(svuint32_t, uint64_t);
18848		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u64)))
18849		svuint32_t svshrnb_n_u64(svuint64_t, uint64_t);
18850		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u16)))
18851		svuint8_t svshrnb_n_u16(svuint16_t, uint64_t);
18852		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s32)))
18853		svint16_t svshrnb_n_s32(svint32_t, uint64_t);
18854		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s64)))
18855		svint32_t svshrnb_n_s64(svint64_t, uint64_t);
18856		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s16)))
18857		svint8_t svshrnb_n_s16(svint16_t, uint64_t);
18858		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u32)))
18859		svuint16_t svshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
18860		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u64)))
18861		svuint32_t svshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
18862		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u16)))
18863		svuint8_t svshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
18864		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s32)))
18865		svint16_t svshrnt_n_s32(svint16_t, svint32_t, uint64_t);
18866		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s64)))
18867		svint32_t svshrnt_n_s64(svint32_t, svint64_t, uint64_t);
18868		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s16)))
18869		svint8_t svshrnt_n_s16(svint8_t, svint16_t, uint64_t);
18870		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u8)))
18871		svuint8_t svsli_n_u8(svuint8_t, svuint8_t, uint64_t);
18872		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u32)))
18873		svuint32_t svsli_n_u32(svuint32_t, svuint32_t, uint64_t);
18874		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u64)))
18875		svuint64_t svsli_n_u64(svuint64_t, svuint64_t, uint64_t);
18876		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u16)))
18877		svuint16_t svsli_n_u16(svuint16_t, svuint16_t, uint64_t);
18878		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s8)))
18879		svint8_t svsli_n_s8(svint8_t, svint8_t, uint64_t);
18880		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s32)))
18881		svint32_t svsli_n_s32(svint32_t, svint32_t, uint64_t);
18882		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s64)))
18883		svint64_t svsli_n_s64(svint64_t, svint64_t, uint64_t);
18884		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s16)))
18885		svint16_t svsli_n_s16(svint16_t, svint16_t, uint64_t);
18886		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_m)))
18887		svuint8_t svsqadd_n_u8_m(svbool_t, svuint8_t, int8_t);
18888		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_m)))
18889		svuint32_t svsqadd_n_u32_m(svbool_t, svuint32_t, int32_t);
18890		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_m)))
18891		svuint64_t svsqadd_n_u64_m(svbool_t, svuint64_t, int64_t);
18892		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_m)))
18893		svuint16_t svsqadd_n_u16_m(svbool_t, svuint16_t, int16_t);
18894		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_x)))
18895		svuint8_t svsqadd_n_u8_x(svbool_t, svuint8_t, int8_t);
18896		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_x)))
18897		svuint32_t svsqadd_n_u32_x(svbool_t, svuint32_t, int32_t);
18898		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_x)))
18899		svuint64_t svsqadd_n_u64_x(svbool_t, svuint64_t, int64_t);
18900		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_x)))
18901		svuint16_t svsqadd_n_u16_x(svbool_t, svuint16_t, int16_t);
18902		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_z)))
18903		svuint8_t svsqadd_n_u8_z(svbool_t, svuint8_t, int8_t);
18904		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_z)))
18905		svuint32_t svsqadd_n_u32_z(svbool_t, svuint32_t, int32_t);
18906		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_z)))
18907		svuint64_t svsqadd_n_u64_z(svbool_t, svuint64_t, int64_t);
18908		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_z)))
18909		svuint16_t svsqadd_n_u16_z(svbool_t, svuint16_t, int16_t);
18910		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_m)))
18911		svuint8_t svsqadd_u8_m(svbool_t, svuint8_t, svint8_t);
18912		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_m)))
18913		svuint32_t svsqadd_u32_m(svbool_t, svuint32_t, svint32_t);
18914		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_m)))
18915		svuint64_t svsqadd_u64_m(svbool_t, svuint64_t, svint64_t);
18916		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_m)))
18917		svuint16_t svsqadd_u16_m(svbool_t, svuint16_t, svint16_t);
18918		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_x)))
18919		svuint8_t svsqadd_u8_x(svbool_t, svuint8_t, svint8_t);
18920		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_x)))
18921		svuint32_t svsqadd_u32_x(svbool_t, svuint32_t, svint32_t);
18922		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_x)))
18923		svuint64_t svsqadd_u64_x(svbool_t, svuint64_t, svint64_t);
18924		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_x)))
18925		svuint16_t svsqadd_u16_x(svbool_t, svuint16_t, svint16_t);
18926		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_z)))
18927		svuint8_t svsqadd_u8_z(svbool_t, svuint8_t, svint8_t);
18928		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_z)))
18929		svuint32_t svsqadd_u32_z(svbool_t, svuint32_t, svint32_t);
18930		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_z)))
18931		svuint64_t svsqadd_u64_z(svbool_t, svuint64_t, svint64_t);
18932		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_z)))
18933		svuint16_t svsqadd_u16_z(svbool_t, svuint16_t, svint16_t);
18934		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s8)))
18935		svint8_t svsra_n_s8(svint8_t, svint8_t, uint64_t);
18936		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s32)))
18937		svint32_t svsra_n_s32(svint32_t, svint32_t, uint64_t);
18938		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s64)))
18939		svint64_t svsra_n_s64(svint64_t, svint64_t, uint64_t);
18940		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s16)))
18941		svint16_t svsra_n_s16(svint16_t, svint16_t, uint64_t);
18942		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u8)))
18943		svuint8_t svsra_n_u8(svuint8_t, svuint8_t, uint64_t);
18944		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u32)))
18945		svuint32_t svsra_n_u32(svuint32_t, svuint32_t, uint64_t);
18946		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u64)))
18947		svuint64_t svsra_n_u64(svuint64_t, svuint64_t, uint64_t);
18948		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u16)))
18949		svuint16_t svsra_n_u16(svuint16_t, svuint16_t, uint64_t);
18950		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u8)))
18951		svuint8_t svsri_n_u8(svuint8_t, svuint8_t, uint64_t);
18952		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u32)))
18953		svuint32_t svsri_n_u32(svuint32_t, svuint32_t, uint64_t);
18954		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u64)))
18955		svuint64_t svsri_n_u64(svuint64_t, svuint64_t, uint64_t);
18956		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u16)))
18957		svuint16_t svsri_n_u16(svuint16_t, svuint16_t, uint64_t);
18958		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s8)))
18959		svint8_t svsri_n_s8(svint8_t, svint8_t, uint64_t);
18960		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s32)))
18961		svint32_t svsri_n_s32(svint32_t, svint32_t, uint64_t);
18962		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s64)))
18963		svint64_t svsri_n_s64(svint64_t, svint64_t, uint64_t);
18964		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s16)))
18965		svint16_t svsri_n_s16(svint16_t, svint16_t, uint64_t);
18966		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_u32)))
18967		void svstnt1_scatter_u32base_index_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
18968		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_u64)))
18969		void svstnt1_scatter_u64base_index_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
18970		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_f64)))
18971		void svstnt1_scatter_u64base_index_f64(svbool_t, svuint64_t, int64_t, svfloat64_t);
18972		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_f32)))
18973		void svstnt1_scatter_u32base_index_f32(svbool_t, svuint32_t, int64_t, svfloat32_t);
18974		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_s32)))
18975		void svstnt1_scatter_u32base_index_s32(svbool_t, svuint32_t, int64_t, svint32_t);
18976		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_s64)))
18977		void svstnt1_scatter_u64base_index_s64(svbool_t, svuint64_t, int64_t, svint64_t);
18978		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_u32)))
18979		void svstnt1_scatter_u32base_offset_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
18980		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_u64)))
18981		void svstnt1_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
18982		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_f64)))
18983		void svstnt1_scatter_u64base_offset_f64(svbool_t, svuint64_t, int64_t, svfloat64_t);
18984		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_f32)))
18985		void svstnt1_scatter_u32base_offset_f32(svbool_t, svuint32_t, int64_t, svfloat32_t);
18986		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_s32)))
18987		void svstnt1_scatter_u32base_offset_s32(svbool_t, svuint32_t, int64_t, svint32_t);
18988		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_s64)))
18989		void svstnt1_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
18990		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_u32)))
18991		void svstnt1_scatter_u32base_u32(svbool_t, svuint32_t, svuint32_t);
18992		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_u64)))
18993		void svstnt1_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
18994		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_f64)))
18995		void svstnt1_scatter_u64base_f64(svbool_t, svuint64_t, svfloat64_t);
18996		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_f32)))
18997		void svstnt1_scatter_u32base_f32(svbool_t, svuint32_t, svfloat32_t);
18998		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_s32)))
18999		void svstnt1_scatter_u32base_s32(svbool_t, svuint32_t, svint32_t);
19000		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_s64)))
19001		void svstnt1_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
19002		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_u64)))
19003		void svstnt1_scatter_s64index_u64(svbool_t, uint64_t *, svint64_t, svuint64_t);
19004		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_f64)))
19005		void svstnt1_scatter_s64index_f64(svbool_t, float64_t *, svint64_t, svfloat64_t);
19006		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_s64)))
19007		void svstnt1_scatter_s64index_s64(svbool_t, int64_t *, svint64_t, svint64_t);
19008		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_u64)))
19009		void svstnt1_scatter_u64index_u64(svbool_t, uint64_t *, svuint64_t, svuint64_t);
19010		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_f64)))
19011		void svstnt1_scatter_u64index_f64(svbool_t, float64_t *, svuint64_t, svfloat64_t);
19012		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_s64)))
19013		void svstnt1_scatter_u64index_s64(svbool_t, int64_t *, svuint64_t, svint64_t);
19014		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_u32)))
19015		void svstnt1_scatter_u32offset_u32(svbool_t, uint32_t *, svuint32_t, svuint32_t);
19016		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_f32)))
19017		void svstnt1_scatter_u32offset_f32(svbool_t, float32_t *, svuint32_t, svfloat32_t);
19018		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_s32)))
19019		void svstnt1_scatter_u32offset_s32(svbool_t, int32_t *, svuint32_t, svint32_t);
19020		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_u64)))
19021		void svstnt1_scatter_s64offset_u64(svbool_t, uint64_t *, svint64_t, svuint64_t);
19022		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_f64)))
19023		void svstnt1_scatter_s64offset_f64(svbool_t, float64_t *, svint64_t, svfloat64_t);
19024		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_s64)))
19025		void svstnt1_scatter_s64offset_s64(svbool_t, int64_t *, svint64_t, svint64_t);
19026		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_u64)))
19027		void svstnt1_scatter_u64offset_u64(svbool_t, uint64_t *, svuint64_t, svuint64_t);
19028		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_f64)))
19029		void svstnt1_scatter_u64offset_f64(svbool_t, float64_t *, svuint64_t, svfloat64_t);
19030		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_s64)))
19031		void svstnt1_scatter_u64offset_s64(svbool_t, int64_t *, svuint64_t, svint64_t);
19032		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_u32)))
19033		void svstnt1b_scatter_u32base_offset_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
19034		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_u64)))
19035		void svstnt1b_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
19036		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_s32)))
19037		void svstnt1b_scatter_u32base_offset_s32(svbool_t, svuint32_t, int64_t, svint32_t);
19038		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_s64)))
19039		void svstnt1b_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
19040		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_u32)))
19041		void svstnt1b_scatter_u32base_u32(svbool_t, svuint32_t, svuint32_t);
19042		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_u64)))
19043		void svstnt1b_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
19044		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_s32)))
19045		void svstnt1b_scatter_u32base_s32(svbool_t, svuint32_t, svint32_t);
19046		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_s64)))
19047		void svstnt1b_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
19048		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_s32)))
19049		void svstnt1b_scatter_u32offset_s32(svbool_t, int8_t *, svuint32_t, svint32_t);
19050		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_u32)))
19051		void svstnt1b_scatter_u32offset_u32(svbool_t, uint8_t *, svuint32_t, svuint32_t);
19052		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_s64)))
19053		void svstnt1b_scatter_s64offset_s64(svbool_t, int8_t *, svint64_t, svint64_t);
19054		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_u64)))
19055		void svstnt1b_scatter_s64offset_u64(svbool_t, uint8_t *, svint64_t, svuint64_t);
19056		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_s64)))
19057		void svstnt1b_scatter_u64offset_s64(svbool_t, int8_t *, svuint64_t, svint64_t);
19058		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_u64)))
19059		void svstnt1b_scatter_u64offset_u64(svbool_t, uint8_t *, svuint64_t, svuint64_t);
19060		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_u32)))
19061		void svstnt1h_scatter_u32base_index_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
19062		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_u64)))
19063		void svstnt1h_scatter_u64base_index_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
19064		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_s32)))
19065		void svstnt1h_scatter_u32base_index_s32(svbool_t, svuint32_t, int64_t, svint32_t);
19066		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_s64)))
19067		void svstnt1h_scatter_u64base_index_s64(svbool_t, svuint64_t, int64_t, svint64_t);
19068		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_u32)))
19069		void svstnt1h_scatter_u32base_offset_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
19070		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_u64)))
19071		void svstnt1h_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
19072		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_s32)))
19073		void svstnt1h_scatter_u32base_offset_s32(svbool_t, svuint32_t, int64_t, svint32_t);
19074		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_s64)))
19075		void svstnt1h_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
19076		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_u32)))
19077		void svstnt1h_scatter_u32base_u32(svbool_t, svuint32_t, svuint32_t);
19078		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_u64)))
19079		void svstnt1h_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
19080		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_s32)))
19081		void svstnt1h_scatter_u32base_s32(svbool_t, svuint32_t, svint32_t);
19082		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_s64)))
19083		void svstnt1h_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
19084		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_s64)))
19085		void svstnt1h_scatter_s64index_s64(svbool_t, int16_t *, svint64_t, svint64_t);
19086		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_u64)))
19087		void svstnt1h_scatter_s64index_u64(svbool_t, uint16_t *, svint64_t, svuint64_t);
19088		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_s64)))
19089		void svstnt1h_scatter_u64index_s64(svbool_t, int16_t *, svuint64_t, svint64_t);
19090		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_u64)))
19091		void svstnt1h_scatter_u64index_u64(svbool_t, uint16_t *, svuint64_t, svuint64_t);
19092		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_s32)))
19093		void svstnt1h_scatter_u32offset_s32(svbool_t, int16_t *, svuint32_t, svint32_t);
19094		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_u32)))
19095		void svstnt1h_scatter_u32offset_u32(svbool_t, uint16_t *, svuint32_t, svuint32_t);
19096		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_s64)))
19097		void svstnt1h_scatter_s64offset_s64(svbool_t, int16_t *, svint64_t, svint64_t);
19098		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_u64)))
19099		void svstnt1h_scatter_s64offset_u64(svbool_t, uint16_t *, svint64_t, svuint64_t);
19100		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_s64)))
19101		void svstnt1h_scatter_u64offset_s64(svbool_t, int16_t *, svuint64_t, svint64_t);
19102		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_u64)))
19103		void svstnt1h_scatter_u64offset_u64(svbool_t, uint16_t *, svuint64_t, svuint64_t);
19104		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_u64)))
19105		void svstnt1w_scatter_u64base_index_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
19106		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_s64)))
19107		void svstnt1w_scatter_u64base_index_s64(svbool_t, svuint64_t, int64_t, svint64_t);
19108		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_u64)))
19109		void svstnt1w_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
19110		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_s64)))
19111		void svstnt1w_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
19112		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_u64)))
19113		void svstnt1w_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
19114		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_s64)))
19115		void svstnt1w_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
19116		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_s64)))
19117		void svstnt1w_scatter_s64index_s64(svbool_t, int32_t *, svint64_t, svint64_t);
19118		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_u64)))
19119		void svstnt1w_scatter_s64index_u64(svbool_t, uint32_t *, svint64_t, svuint64_t);
19120		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_s64)))
19121		void svstnt1w_scatter_u64index_s64(svbool_t, int32_t *, svuint64_t, svint64_t);
19122		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_u64)))
19123		void svstnt1w_scatter_u64index_u64(svbool_t, uint32_t *, svuint64_t, svuint64_t);
19124		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_s64)))
19125		void svstnt1w_scatter_s64offset_s64(svbool_t, int32_t *, svint64_t, svint64_t);
19126		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_u64)))
19127		void svstnt1w_scatter_s64offset_u64(svbool_t, uint32_t *, svint64_t, svuint64_t);
19128		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_s64)))
19129		void svstnt1w_scatter_u64offset_s64(svbool_t, int32_t *, svuint64_t, svint64_t);
19130		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_u64)))
19131		void svstnt1w_scatter_u64offset_u64(svbool_t, uint32_t *, svuint64_t, svuint64_t);
19132		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u32)))
19133		svuint16_t svsubhnb_n_u32(svuint32_t, uint32_t);
19134		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u64)))
19135		svuint32_t svsubhnb_n_u64(svuint64_t, uint64_t);
19136		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u16)))
19137		svuint8_t svsubhnb_n_u16(svuint16_t, uint16_t);
19138		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s32)))
19139		svint16_t svsubhnb_n_s32(svint32_t, int32_t);
19140		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s64)))
19141		svint32_t svsubhnb_n_s64(svint64_t, int64_t);
19142		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s16)))
19143		svint8_t svsubhnb_n_s16(svint16_t, int16_t);
19144		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u32)))
19145		svuint16_t svsubhnb_u32(svuint32_t, svuint32_t);
19146		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u64)))
19147		svuint32_t svsubhnb_u64(svuint64_t, svuint64_t);
19148		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u16)))
19149		svuint8_t svsubhnb_u16(svuint16_t, svuint16_t);
19150		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s32)))
19151		svint16_t svsubhnb_s32(svint32_t, svint32_t);
19152		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s64)))
19153		svint32_t svsubhnb_s64(svint64_t, svint64_t);
19154		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s16)))
19155		svint8_t svsubhnb_s16(svint16_t, svint16_t);
19156		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u32)))
19157		svuint16_t svsubhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
19158		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u64)))
19159		svuint32_t svsubhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
19160		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u16)))
19161		svuint8_t svsubhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
19162		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s32)))
19163		svint16_t svsubhnt_n_s32(svint16_t, svint32_t, int32_t);
19164		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s64)))
19165		svint32_t svsubhnt_n_s64(svint32_t, svint64_t, int64_t);
19166		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s16)))
19167		svint8_t svsubhnt_n_s16(svint8_t, svint16_t, int16_t);
19168		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u32)))
19169		svuint16_t svsubhnt_u32(svuint16_t, svuint32_t, svuint32_t);
19170		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u64)))
19171		svuint32_t svsubhnt_u64(svuint32_t, svuint64_t, svuint64_t);
19172		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u16)))
19173		svuint8_t svsubhnt_u16(svuint8_t, svuint16_t, svuint16_t);
19174		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s32)))
19175		svint16_t svsubhnt_s32(svint16_t, svint32_t, svint32_t);
19176		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s64)))
19177		svint32_t svsubhnt_s64(svint32_t, svint64_t, svint64_t);
19178		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s16)))
19179		svint8_t svsubhnt_s16(svint8_t, svint16_t, svint16_t);
19180		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s32)))
19181		svint32_t svsublb_n_s32(svint16_t, int16_t);
19182		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s64)))
19183		svint64_t svsublb_n_s64(svint32_t, int32_t);
19184		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s16)))
19185		svint16_t svsublb_n_s16(svint8_t, int8_t);
19186		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u32)))
19187		svuint32_t svsublb_n_u32(svuint16_t, uint16_t);
19188		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u64)))
19189		svuint64_t svsublb_n_u64(svuint32_t, uint32_t);
19190		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u16)))
19191		svuint16_t svsublb_n_u16(svuint8_t, uint8_t);
19192		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s32)))
19193		svint32_t svsublb_s32(svint16_t, svint16_t);
19194		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s64)))
19195		svint64_t svsublb_s64(svint32_t, svint32_t);
19196		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s16)))
19197		svint16_t svsublb_s16(svint8_t, svint8_t);
19198		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u32)))
19199		svuint32_t svsublb_u32(svuint16_t, svuint16_t);
19200		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u64)))
19201		svuint64_t svsublb_u64(svuint32_t, svuint32_t);
19202		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u16)))
19203		svuint16_t svsublb_u16(svuint8_t, svuint8_t);
19204		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s32)))
19205		svint32_t svsublbt_n_s32(svint16_t, int16_t);
19206		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s64)))
19207		svint64_t svsublbt_n_s64(svint32_t, int32_t);
19208		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s16)))
19209		svint16_t svsublbt_n_s16(svint8_t, int8_t);
19210		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s32)))
19211		svint32_t svsublbt_s32(svint16_t, svint16_t);
19212		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s64)))
19213		svint64_t svsublbt_s64(svint32_t, svint32_t);
19214		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s16)))
19215		svint16_t svsublbt_s16(svint8_t, svint8_t);
19216		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s32)))
19217		svint32_t svsublt_n_s32(svint16_t, int16_t);
19218		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s64)))
19219		svint64_t svsublt_n_s64(svint32_t, int32_t);
19220		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s16)))
19221		svint16_t svsublt_n_s16(svint8_t, int8_t);
19222		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u32)))
19223		svuint32_t svsublt_n_u32(svuint16_t, uint16_t);
19224		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u64)))
19225		svuint64_t svsublt_n_u64(svuint32_t, uint32_t);
19226		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u16)))
19227		svuint16_t svsublt_n_u16(svuint8_t, uint8_t);
19228		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s32)))
19229		svint32_t svsublt_s32(svint16_t, svint16_t);
19230		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s64)))
19231		svint64_t svsublt_s64(svint32_t, svint32_t);
19232		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s16)))
19233		svint16_t svsublt_s16(svint8_t, svint8_t);
19234		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u32)))
19235		svuint32_t svsublt_u32(svuint16_t, svuint16_t);
19236		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u64)))
19237		svuint64_t svsublt_u64(svuint32_t, svuint32_t);
19238		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u16)))
19239		svuint16_t svsublt_u16(svuint8_t, svuint8_t);
19240		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s32)))
19241		svint32_t svsubltb_n_s32(svint16_t, int16_t);
19242		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s64)))
19243		svint64_t svsubltb_n_s64(svint32_t, int32_t);
19244		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s16)))
19245		svint16_t svsubltb_n_s16(svint8_t, int8_t);
19246		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s32)))
19247		svint32_t svsubltb_s32(svint16_t, svint16_t);
19248		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s64)))
19249		svint64_t svsubltb_s64(svint32_t, svint32_t);
19250		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s16)))
19251		svint16_t svsubltb_s16(svint8_t, svint8_t);
19252		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s32)))
19253		svint32_t svsubwb_n_s32(svint32_t, int16_t);
19254		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s64)))
19255		svint64_t svsubwb_n_s64(svint64_t, int32_t);
19256		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s16)))
19257		svint16_t svsubwb_n_s16(svint16_t, int8_t);
19258		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u32)))
19259		svuint32_t svsubwb_n_u32(svuint32_t, uint16_t);
19260		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u64)))
19261		svuint64_t svsubwb_n_u64(svuint64_t, uint32_t);
19262		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u16)))
19263		svuint16_t svsubwb_n_u16(svuint16_t, uint8_t);
19264		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s32)))
19265		svint32_t svsubwb_s32(svint32_t, svint16_t);
19266		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s64)))
19267		svint64_t svsubwb_s64(svint64_t, svint32_t);
19268		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s16)))
19269		svint16_t svsubwb_s16(svint16_t, svint8_t);
19270		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u32)))
19271		svuint32_t svsubwb_u32(svuint32_t, svuint16_t);
19272		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u64)))
19273		svuint64_t svsubwb_u64(svuint64_t, svuint32_t);
19274		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u16)))
19275		svuint16_t svsubwb_u16(svuint16_t, svuint8_t);
19276		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s32)))
19277		svint32_t svsubwt_n_s32(svint32_t, int16_t);
19278		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s64)))
19279		svint64_t svsubwt_n_s64(svint64_t, int32_t);
19280		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s16)))
19281		svint16_t svsubwt_n_s16(svint16_t, int8_t);
19282		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u32)))
19283		svuint32_t svsubwt_n_u32(svuint32_t, uint16_t);
19284		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u64)))
19285		svuint64_t svsubwt_n_u64(svuint64_t, uint32_t);
19286		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u16)))
19287		svuint16_t svsubwt_n_u16(svuint16_t, uint8_t);
19288		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s32)))
19289		svint32_t svsubwt_s32(svint32_t, svint16_t);
19290		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s64)))
19291		svint64_t svsubwt_s64(svint64_t, svint32_t);
19292		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s16)))
19293		svint16_t svsubwt_s16(svint16_t, svint8_t);
19294		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u32)))
19295		svuint32_t svsubwt_u32(svuint32_t, svuint16_t);
19296		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u64)))
19297		svuint64_t svsubwt_u64(svuint64_t, svuint32_t);
19298		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u16)))
19299		svuint16_t svsubwt_u16(svuint16_t, svuint8_t);
19300		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u8)))
19301		svuint8_t svtbl2_u8(svuint8x2_t, svuint8_t);
19302		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u32)))
19303		svuint32_t svtbl2_u32(svuint32x2_t, svuint32_t);
19304		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
19305		svuint64_t svtbl2_u64(svuint64x2_t, svuint64_t);
19306		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
19307		svuint16_t svtbl2_u16(svuint16x2_t, svuint16_t);
19308		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
19309		svint8_t svtbl2_s8(svint8x2_t, svuint8_t);
19310		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
19311		svfloat64_t svtbl2_f64(svfloat64x2_t, svuint64_t);
19312		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f32)))
19313		svfloat32_t svtbl2_f32(svfloat32x2_t, svuint32_t);
19314		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f16)))
19315		svfloat16_t svtbl2_f16(svfloat16x2_t, svuint16_t);
19316		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s32)))
19317		svint32_t svtbl2_s32(svint32x2_t, svuint32_t);
19318		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s64)))
19319		svint64_t svtbl2_s64(svint64x2_t, svuint64_t);
19320		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s16)))
19321		svint16_t svtbl2_s16(svint16x2_t, svuint16_t);
19322		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u8)))
19323		svuint8_t svtbx_u8(svuint8_t, svuint8_t, svuint8_t);
19324		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u32)))
19325		svuint32_t svtbx_u32(svuint32_t, svuint32_t, svuint32_t);
19326		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
19327		svuint64_t svtbx_u64(svuint64_t, svuint64_t, svuint64_t);
19328		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
19329		svuint16_t svtbx_u16(svuint16_t, svuint16_t, svuint16_t);
19330		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
19331		svint8_t svtbx_s8(svint8_t, svint8_t, svuint8_t);
19332		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
19333		svfloat64_t svtbx_f64(svfloat64_t, svfloat64_t, svuint64_t);
19334		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f32)))
19335		svfloat32_t svtbx_f32(svfloat32_t, svfloat32_t, svuint32_t);
19336		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f16)))
19337		svfloat16_t svtbx_f16(svfloat16_t, svfloat16_t, svuint16_t);
19338		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s32)))
19339		svint32_t svtbx_s32(svint32_t, svint32_t, svuint32_t);
19340		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s64)))
19341		svint64_t svtbx_s64(svint64_t, svint64_t, svuint64_t);
19342		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s16)))
19343		svint16_t svtbx_s16(svint16_t, svint16_t, svuint16_t);
19344		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_m)))
19345		svint8_t svuqadd_n_s8_m(svbool_t, svint8_t, uint8_t);
19346		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_m)))
19347		svint32_t svuqadd_n_s32_m(svbool_t, svint32_t, uint32_t);
19348		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_m)))
19349		svint64_t svuqadd_n_s64_m(svbool_t, svint64_t, uint64_t);
19350		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_m)))
19351		svint16_t svuqadd_n_s16_m(svbool_t, svint16_t, uint16_t);
19352		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_x)))
19353		svint8_t svuqadd_n_s8_x(svbool_t, svint8_t, uint8_t);
19354		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_x)))
19355		svint32_t svuqadd_n_s32_x(svbool_t, svint32_t, uint32_t);
19356		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_x)))
19357		svint64_t svuqadd_n_s64_x(svbool_t, svint64_t, uint64_t);
19358		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_x)))
19359		svint16_t svuqadd_n_s16_x(svbool_t, svint16_t, uint16_t);
19360		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_z)))
19361		svint8_t svuqadd_n_s8_z(svbool_t, svint8_t, uint8_t);
19362		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_z)))
19363		svint32_t svuqadd_n_s32_z(svbool_t, svint32_t, uint32_t);
19364		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_z)))
19365		svint64_t svuqadd_n_s64_z(svbool_t, svint64_t, uint64_t);
19366		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_z)))
19367		svint16_t svuqadd_n_s16_z(svbool_t, svint16_t, uint16_t);
19368		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_m)))
19369		svint8_t svuqadd_s8_m(svbool_t, svint8_t, svuint8_t);
19370		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_m)))
19371		svint32_t svuqadd_s32_m(svbool_t, svint32_t, svuint32_t);
19372		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_m)))
19373		svint64_t svuqadd_s64_m(svbool_t, svint64_t, svuint64_t);
19374		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_m)))
19375		svint16_t svuqadd_s16_m(svbool_t, svint16_t, svuint16_t);
19376		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_x)))
19377		svint8_t svuqadd_s8_x(svbool_t, svint8_t, svuint8_t);
19378		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_x)))
19379		svint32_t svuqadd_s32_x(svbool_t, svint32_t, svuint32_t);
19380		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_x)))
19381		svint64_t svuqadd_s64_x(svbool_t, svint64_t, svuint64_t);
19382		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_x)))
19383		svint16_t svuqadd_s16_x(svbool_t, svint16_t, svuint16_t);
19384		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_z)))
19385		svint8_t svuqadd_s8_z(svbool_t, svint8_t, svuint8_t);
19386		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_z)))
19387		svint32_t svuqadd_s32_z(svbool_t, svint32_t, svuint32_t);
19388		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_z)))
19389		svint64_t svuqadd_s64_z(svbool_t, svint64_t, svuint64_t);
19390		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_z)))
19391		svint16_t svuqadd_s16_z(svbool_t, svint16_t, svuint16_t);
19392		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s32)))
19393		svbool_t svwhilege_b8_s32(int32_t, int32_t);
19394		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s32)))
19395		svbool_t svwhilege_b32_s32(int32_t, int32_t);
19396		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s32)))
19397		svbool_t svwhilege_b64_s32(int32_t, int32_t);
19398		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s32)))
19399		svbool_t svwhilege_b16_s32(int32_t, int32_t);
19400		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64)))
19401		svbool_t svwhilege_b8_s64(int64_t, int64_t);
19402		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64)))
19403		svbool_t svwhilege_b32_s64(int64_t, int64_t);
19404		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64)))
19405		svbool_t svwhilege_b64_s64(int64_t, int64_t);
19406		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64)))
19407		svbool_t svwhilege_b16_s64(int64_t, int64_t);
19408		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u32)))
19409		svbool_t svwhilege_b8_u32(uint32_t, uint32_t);
19410		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u32)))
19411		svbool_t svwhilege_b32_u32(uint32_t, uint32_t);
19412		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u32)))
19413		svbool_t svwhilege_b64_u32(uint32_t, uint32_t);
19414		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u32)))
19415		svbool_t svwhilege_b16_u32(uint32_t, uint32_t);
19416		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64)))
19417		svbool_t svwhilege_b8_u64(uint64_t, uint64_t);
19418		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64)))
19419		svbool_t svwhilege_b32_u64(uint64_t, uint64_t);
19420		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64)))
19421		svbool_t svwhilege_b64_u64(uint64_t, uint64_t);
19422		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64)))
19423		svbool_t svwhilege_b16_u64(uint64_t, uint64_t);
19424		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s32)))
19425		svbool_t svwhilegt_b8_s32(int32_t, int32_t);
19426		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s32)))
19427		svbool_t svwhilegt_b32_s32(int32_t, int32_t);
19428		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s32)))
19429		svbool_t svwhilegt_b64_s32(int32_t, int32_t);
19430		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s32)))
19431		svbool_t svwhilegt_b16_s32(int32_t, int32_t);
19432		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64)))
19433		svbool_t svwhilegt_b8_s64(int64_t, int64_t);
19434		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64)))
19435		svbool_t svwhilegt_b32_s64(int64_t, int64_t);
19436		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64)))
19437		svbool_t svwhilegt_b64_s64(int64_t, int64_t);
19438		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64)))
19439		svbool_t svwhilegt_b16_s64(int64_t, int64_t);
19440		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u32)))
19441		svbool_t svwhilegt_b8_u32(uint32_t, uint32_t);
19442		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u32)))
19443		svbool_t svwhilegt_b32_u32(uint32_t, uint32_t);
19444		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u32)))
19445		svbool_t svwhilegt_b64_u32(uint32_t, uint32_t);
19446		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u32)))
19447		svbool_t svwhilegt_b16_u32(uint32_t, uint32_t);
19448		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64)))
19449		svbool_t svwhilegt_b8_u64(uint64_t, uint64_t);
19450		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64)))
19451		svbool_t svwhilegt_b32_u64(uint64_t, uint64_t);
19452		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64)))
19453		svbool_t svwhilegt_b64_u64(uint64_t, uint64_t);
19454		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64)))
19455		svbool_t svwhilegt_b16_u64(uint64_t, uint64_t);
19456		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u8)))
19457		svbool_t svwhilerw_u8(uint8_t const , uint8_t const );
19458		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s8)))
19459		svbool_t svwhilerw_s8(int8_t const , int8_t const );
19460		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u64)))
19461		svbool_t svwhilerw_u64(uint64_t const , uint64_t const );
19462		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f64)))
19463		svbool_t svwhilerw_f64(float64_t const , float64_t const );
19464		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
19465		svbool_t svwhilerw_s64(int64_t const , int64_t const );
19466		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
19467		svbool_t svwhilerw_u16(uint16_t const , uint16_t const );
19468		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
19469		svbool_t svwhilerw_f16(float16_t const , float16_t const );
19470		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
19471		svbool_t svwhilerw_s16(int16_t const , int16_t const );
19472		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u32)))
19473		svbool_t svwhilerw_u32(uint32_t const , uint32_t const );
19474		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f32)))
19475		svbool_t svwhilerw_f32(float32_t const , float32_t const );
19476		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s32)))
19477		svbool_t svwhilerw_s32(int32_t const , int32_t const );
19478		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u8)))
19479		svbool_t svwhilewr_u8(uint8_t const , uint8_t const );
19480		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s8)))
19481		svbool_t svwhilewr_s8(int8_t const , int8_t const );
19482		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u64)))
19483		svbool_t svwhilewr_u64(uint64_t const , uint64_t const );
19484		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f64)))
19485		svbool_t svwhilewr_f64(float64_t const , float64_t const );
19486		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
19487		svbool_t svwhilewr_s64(int64_t const , int64_t const );
19488		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
19489		svbool_t svwhilewr_u16(uint16_t const , uint16_t const );
19490		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
19491		svbool_t svwhilewr_f16(float16_t const , float16_t const );
19492		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
19493		svbool_t svwhilewr_s16(int16_t const , int16_t const );
19494		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u32)))
19495		svbool_t svwhilewr_u32(uint32_t const , uint32_t const );
19496		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f32)))
19497		svbool_t svwhilewr_f32(float32_t const , float32_t const );
19498		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s32)))
19499		svbool_t svwhilewr_s32(int32_t const , int32_t const );
19500		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u8)))
19501		svuint8_t svxar_n_u8(svuint8_t, svuint8_t, uint64_t);
19502		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u32)))
19503		svuint32_t svxar_n_u32(svuint32_t, svuint32_t, uint64_t);
19504		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u64)))
19505		svuint64_t svxar_n_u64(svuint64_t, svuint64_t, uint64_t);
19506		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u16)))
19507		svuint16_t svxar_n_u16(svuint16_t, svuint16_t, uint64_t);
19508		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s8)))
19509		svint8_t svxar_n_s8(svint8_t, svint8_t, uint64_t);
19510		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s32)))
19511		svint32_t svxar_n_s32(svint32_t, svint32_t, uint64_t);
19512		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s64)))
19513		svint64_t svxar_n_s64(svint64_t, svint64_t, uint64_t);
19514		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s16)))
19515		svint16_t svxar_n_s16(svint16_t, svint16_t, uint64_t);
19516		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s8)))
19517		svint8_t svaba(svint8_t, svint8_t, int8_t);
19518		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s32)))
19519		svint32_t svaba(svint32_t, svint32_t, int32_t);
19520		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s64)))
19521		svint64_t svaba(svint64_t, svint64_t, int64_t);
19522		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s16)))
19523		svint16_t svaba(svint16_t, svint16_t, int16_t);
19524		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u8)))
19525		svuint8_t svaba(svuint8_t, svuint8_t, uint8_t);
19526		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u32)))
19527		svuint32_t svaba(svuint32_t, svuint32_t, uint32_t);
19528		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u64)))
19529		svuint64_t svaba(svuint64_t, svuint64_t, uint64_t);
19530		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u16)))
19531		svuint16_t svaba(svuint16_t, svuint16_t, uint16_t);
19532		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s8)))
19533		svint8_t svaba(svint8_t, svint8_t, svint8_t);
19534		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s32)))
19535		svint32_t svaba(svint32_t, svint32_t, svint32_t);
19536		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s64)))
19537		svint64_t svaba(svint64_t, svint64_t, svint64_t);
19538		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s16)))
19539		svint16_t svaba(svint16_t, svint16_t, svint16_t);
19540		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u8)))
19541		svuint8_t svaba(svuint8_t, svuint8_t, svuint8_t);
19542		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u32)))
19543		svuint32_t svaba(svuint32_t, svuint32_t, svuint32_t);
19544		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u64)))
19545		svuint64_t svaba(svuint64_t, svuint64_t, svuint64_t);
19546		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u16)))
19547		svuint16_t svaba(svuint16_t, svuint16_t, svuint16_t);
19548		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s32)))
19549		svint32_t svabalb(svint32_t, svint16_t, int16_t);
19550		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s64)))
19551		svint64_t svabalb(svint64_t, svint32_t, int32_t);
19552		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s16)))
19553		svint16_t svabalb(svint16_t, svint8_t, int8_t);
19554		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u32)))
19555		svuint32_t svabalb(svuint32_t, svuint16_t, uint16_t);
19556		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u64)))
19557		svuint64_t svabalb(svuint64_t, svuint32_t, uint32_t);
19558		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u16)))
19559		svuint16_t svabalb(svuint16_t, svuint8_t, uint8_t);
19560		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s32)))
19561		svint32_t svabalb(svint32_t, svint16_t, svint16_t);
19562		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s64)))
19563		svint64_t svabalb(svint64_t, svint32_t, svint32_t);
19564		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s16)))
19565		svint16_t svabalb(svint16_t, svint8_t, svint8_t);
19566		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u32)))
19567		svuint32_t svabalb(svuint32_t, svuint16_t, svuint16_t);
19568		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u64)))
19569		svuint64_t svabalb(svuint64_t, svuint32_t, svuint32_t);
19570		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u16)))
19571		svuint16_t svabalb(svuint16_t, svuint8_t, svuint8_t);
19572		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s32)))
19573		svint32_t svabalt(svint32_t, svint16_t, int16_t);
19574		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s64)))
19575		svint64_t svabalt(svint64_t, svint32_t, int32_t);
19576		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s16)))
19577		svint16_t svabalt(svint16_t, svint8_t, int8_t);
19578		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u32)))
19579		svuint32_t svabalt(svuint32_t, svuint16_t, uint16_t);
19580		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u64)))
19581		svuint64_t svabalt(svuint64_t, svuint32_t, uint32_t);
19582		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u16)))
19583		svuint16_t svabalt(svuint16_t, svuint8_t, uint8_t);
19584		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s32)))
19585		svint32_t svabalt(svint32_t, svint16_t, svint16_t);
19586		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s64)))
19587		svint64_t svabalt(svint64_t, svint32_t, svint32_t);
19588		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s16)))
19589		svint16_t svabalt(svint16_t, svint8_t, svint8_t);
19590		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u32)))
19591		svuint32_t svabalt(svuint32_t, svuint16_t, svuint16_t);
19592		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u64)))
19593		svuint64_t svabalt(svuint64_t, svuint32_t, svuint32_t);
19594		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u16)))
19595		svuint16_t svabalt(svuint16_t, svuint8_t, svuint8_t);
19596		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s32)))
19597		svint32_t svabdlb(svint16_t, int16_t);
19598		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s64)))
19599		svint64_t svabdlb(svint32_t, int32_t);
19600		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s16)))
19601		svint16_t svabdlb(svint8_t, int8_t);
19602		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u32)))
19603		svuint32_t svabdlb(svuint16_t, uint16_t);
19604		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u64)))
19605		svuint64_t svabdlb(svuint32_t, uint32_t);
19606		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u16)))
19607		svuint16_t svabdlb(svuint8_t, uint8_t);
19608		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s32)))
19609		svint32_t svabdlb(svint16_t, svint16_t);
19610		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s64)))
19611		svint64_t svabdlb(svint32_t, svint32_t);
19612		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s16)))
19613		svint16_t svabdlb(svint8_t, svint8_t);
19614		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u32)))
19615		svuint32_t svabdlb(svuint16_t, svuint16_t);
19616		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u64)))
19617		svuint64_t svabdlb(svuint32_t, svuint32_t);
19618		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u16)))
19619		svuint16_t svabdlb(svuint8_t, svuint8_t);
19620		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s32)))
19621		svint32_t svabdlt(svint16_t, int16_t);
19622		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s64)))
19623		svint64_t svabdlt(svint32_t, int32_t);
19624		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s16)))
19625		svint16_t svabdlt(svint8_t, int8_t);
19626		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u32)))
19627		svuint32_t svabdlt(svuint16_t, uint16_t);
19628		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u64)))
19629		svuint64_t svabdlt(svuint32_t, uint32_t);
19630		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u16)))
19631		svuint16_t svabdlt(svuint8_t, uint8_t);
19632		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s32)))
19633		svint32_t svabdlt(svint16_t, svint16_t);
19634		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s64)))
19635		svint64_t svabdlt(svint32_t, svint32_t);
19636		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s16)))
19637		svint16_t svabdlt(svint8_t, svint8_t);
19638		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u32)))
19639		svuint32_t svabdlt(svuint16_t, svuint16_t);
19640		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u64)))
19641		svuint64_t svabdlt(svuint32_t, svuint32_t);
19642		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u16)))
19643		svuint16_t svabdlt(svuint8_t, svuint8_t);
19644		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_m)))
19645		svint32_t svadalp_m(svbool_t, svint32_t, svint16_t);
19646		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_m)))
19647		svint64_t svadalp_m(svbool_t, svint64_t, svint32_t);
19648		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_m)))
19649		svint16_t svadalp_m(svbool_t, svint16_t, svint8_t);
19650		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_x)))
19651		svint32_t svadalp_x(svbool_t, svint32_t, svint16_t);
19652		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_x)))
19653		svint64_t svadalp_x(svbool_t, svint64_t, svint32_t);
19654		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_x)))
19655		svint16_t svadalp_x(svbool_t, svint16_t, svint8_t);
19656		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_z)))
19657		svint32_t svadalp_z(svbool_t, svint32_t, svint16_t);
19658		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_z)))
19659		svint64_t svadalp_z(svbool_t, svint64_t, svint32_t);
19660		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_z)))
19661		svint16_t svadalp_z(svbool_t, svint16_t, svint8_t);
19662		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_m)))
19663		svuint32_t svadalp_m(svbool_t, svuint32_t, svuint16_t);
19664		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_m)))
19665		svuint64_t svadalp_m(svbool_t, svuint64_t, svuint32_t);
19666		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_m)))
19667		svuint16_t svadalp_m(svbool_t, svuint16_t, svuint8_t);
19668		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_x)))
19669		svuint32_t svadalp_x(svbool_t, svuint32_t, svuint16_t);
19670		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_x)))
19671		svuint64_t svadalp_x(svbool_t, svuint64_t, svuint32_t);
19672		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_x)))
19673		svuint16_t svadalp_x(svbool_t, svuint16_t, svuint8_t);
19674		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_z)))
19675		svuint32_t svadalp_z(svbool_t, svuint32_t, svuint16_t);
19676		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_z)))
19677		svuint64_t svadalp_z(svbool_t, svuint64_t, svuint32_t);
19678		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_z)))
19679		svuint16_t svadalp_z(svbool_t, svuint16_t, svuint8_t);
19680		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u32)))
19681		svuint32_t svadclb(svuint32_t, svuint32_t, uint32_t);
19682		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u64)))
19683		svuint64_t svadclb(svuint64_t, svuint64_t, uint64_t);
19684		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u32)))
19685		svuint32_t svadclb(svuint32_t, svuint32_t, svuint32_t);
19686		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u64)))
19687		svuint64_t svadclb(svuint64_t, svuint64_t, svuint64_t);
19688		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u32)))
19689		svuint32_t svadclt(svuint32_t, svuint32_t, uint32_t);
19690		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u64)))
19691		svuint64_t svadclt(svuint64_t, svuint64_t, uint64_t);
19692		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u32)))
19693		svuint32_t svadclt(svuint32_t, svuint32_t, svuint32_t);
19694		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u64)))
19695		svuint64_t svadclt(svuint64_t, svuint64_t, svuint64_t);
19696		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u32)))
19697		svuint16_t svaddhnb(svuint32_t, uint32_t);
19698		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u64)))
19699		svuint32_t svaddhnb(svuint64_t, uint64_t);
19700		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u16)))
19701		svuint8_t svaddhnb(svuint16_t, uint16_t);
19702		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s32)))
19703		svint16_t svaddhnb(svint32_t, int32_t);
19704		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s64)))
19705		svint32_t svaddhnb(svint64_t, int64_t);
19706		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s16)))
19707		svint8_t svaddhnb(svint16_t, int16_t);
19708		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u32)))
19709		svuint16_t svaddhnb(svuint32_t, svuint32_t);
19710		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u64)))
19711		svuint32_t svaddhnb(svuint64_t, svuint64_t);
19712		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u16)))
19713		svuint8_t svaddhnb(svuint16_t, svuint16_t);
19714		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s32)))
19715		svint16_t svaddhnb(svint32_t, svint32_t);
19716		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s64)))
19717		svint32_t svaddhnb(svint64_t, svint64_t);
19718		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s16)))
19719		svint8_t svaddhnb(svint16_t, svint16_t);
19720		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u32)))
19721		svuint16_t svaddhnt(svuint16_t, svuint32_t, uint32_t);
19722		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u64)))
19723		svuint32_t svaddhnt(svuint32_t, svuint64_t, uint64_t);
19724		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u16)))
19725		svuint8_t svaddhnt(svuint8_t, svuint16_t, uint16_t);
19726		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s32)))
19727		svint16_t svaddhnt(svint16_t, svint32_t, int32_t);
19728		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s64)))
19729		svint32_t svaddhnt(svint32_t, svint64_t, int64_t);
19730		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s16)))
19731		svint8_t svaddhnt(svint8_t, svint16_t, int16_t);
19732		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u32)))
19733		svuint16_t svaddhnt(svuint16_t, svuint32_t, svuint32_t);
19734		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u64)))
19735		svuint32_t svaddhnt(svuint32_t, svuint64_t, svuint64_t);
19736		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u16)))
19737		svuint8_t svaddhnt(svuint8_t, svuint16_t, svuint16_t);
19738		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s32)))
19739		svint16_t svaddhnt(svint16_t, svint32_t, svint32_t);
19740		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s64)))
19741		svint32_t svaddhnt(svint32_t, svint64_t, svint64_t);
19742		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s16)))
19743		svint8_t svaddhnt(svint8_t, svint16_t, svint16_t);
19744		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s32)))
19745		svint32_t svaddlb(svint16_t, int16_t);
19746		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s64)))
19747		svint64_t svaddlb(svint32_t, int32_t);
19748		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s16)))
19749		svint16_t svaddlb(svint8_t, int8_t);
19750		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u32)))
19751		svuint32_t svaddlb(svuint16_t, uint16_t);
19752		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u64)))
19753		svuint64_t svaddlb(svuint32_t, uint32_t);
19754		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u16)))
19755		svuint16_t svaddlb(svuint8_t, uint8_t);
19756		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s32)))
19757		svint32_t svaddlb(svint16_t, svint16_t);
19758		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s64)))
19759		svint64_t svaddlb(svint32_t, svint32_t);
19760		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s16)))
19761		svint16_t svaddlb(svint8_t, svint8_t);
19762		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u32)))
19763		svuint32_t svaddlb(svuint16_t, svuint16_t);
19764		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u64)))
19765		svuint64_t svaddlb(svuint32_t, svuint32_t);
19766		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u16)))
19767		svuint16_t svaddlb(svuint8_t, svuint8_t);
19768		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s32)))
19769		svint32_t svaddlbt(svint16_t, int16_t);
19770		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s64)))
19771		svint64_t svaddlbt(svint32_t, int32_t);
19772		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s16)))
19773		svint16_t svaddlbt(svint8_t, int8_t);
19774		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s32)))
19775		svint32_t svaddlbt(svint16_t, svint16_t);
19776		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s64)))
19777		svint64_t svaddlbt(svint32_t, svint32_t);
19778		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s16)))
19779		svint16_t svaddlbt(svint8_t, svint8_t);
19780		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s32)))
19781		svint32_t svaddlt(svint16_t, int16_t);
19782		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s64)))
19783		svint64_t svaddlt(svint32_t, int32_t);
19784		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s16)))
19785		svint16_t svaddlt(svint8_t, int8_t);
19786		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u32)))
19787		svuint32_t svaddlt(svuint16_t, uint16_t);
19788		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u64)))
19789		svuint64_t svaddlt(svuint32_t, uint32_t);
19790		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u16)))
19791		svuint16_t svaddlt(svuint8_t, uint8_t);
19792		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s32)))
19793		svint32_t svaddlt(svint16_t, svint16_t);
19794		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s64)))
19795		svint64_t svaddlt(svint32_t, svint32_t);
19796		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s16)))
19797		svint16_t svaddlt(svint8_t, svint8_t);
19798		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u32)))
19799		svuint32_t svaddlt(svuint16_t, svuint16_t);
19800		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u64)))
19801		svuint64_t svaddlt(svuint32_t, svuint32_t);
19802		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u16)))
19803		svuint16_t svaddlt(svuint8_t, svuint8_t);
19804		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_m)))
19805		svfloat64_t svaddp_m(svbool_t, svfloat64_t, svfloat64_t);
19806		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_m)))
19807		svfloat32_t svaddp_m(svbool_t, svfloat32_t, svfloat32_t);
19808		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_m)))
19809		svfloat16_t svaddp_m(svbool_t, svfloat16_t, svfloat16_t);
19810		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_x)))
19811		svfloat64_t svaddp_x(svbool_t, svfloat64_t, svfloat64_t);
19812		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_x)))
19813		svfloat32_t svaddp_x(svbool_t, svfloat32_t, svfloat32_t);
19814		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_x)))
19815		svfloat16_t svaddp_x(svbool_t, svfloat16_t, svfloat16_t);
19816		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_m)))
19817		svuint8_t svaddp_m(svbool_t, svuint8_t, svuint8_t);
19818		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_m)))
19819		svuint32_t svaddp_m(svbool_t, svuint32_t, svuint32_t);
19820		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_m)))
19821		svuint64_t svaddp_m(svbool_t, svuint64_t, svuint64_t);
19822		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_m)))
19823		svuint16_t svaddp_m(svbool_t, svuint16_t, svuint16_t);
19824		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_m)))
19825		svint8_t svaddp_m(svbool_t, svint8_t, svint8_t);
19826		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_m)))
19827		svint32_t svaddp_m(svbool_t, svint32_t, svint32_t);
19828		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_m)))
19829		svint64_t svaddp_m(svbool_t, svint64_t, svint64_t);
19830		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_m)))
19831		svint16_t svaddp_m(svbool_t, svint16_t, svint16_t);
19832		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_x)))
19833		svuint8_t svaddp_x(svbool_t, svuint8_t, svuint8_t);
19834		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_x)))
19835		svuint32_t svaddp_x(svbool_t, svuint32_t, svuint32_t);
19836		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_x)))
19837		svuint64_t svaddp_x(svbool_t, svuint64_t, svuint64_t);
19838		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_x)))
19839		svuint16_t svaddp_x(svbool_t, svuint16_t, svuint16_t);
19840		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_x)))
19841		svint8_t svaddp_x(svbool_t, svint8_t, svint8_t);
19842		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_x)))
19843		svint32_t svaddp_x(svbool_t, svint32_t, svint32_t);
19844		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_x)))
19845		svint64_t svaddp_x(svbool_t, svint64_t, svint64_t);
19846		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_x)))
19847		svint16_t svaddp_x(svbool_t, svint16_t, svint16_t);
19848		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s32)))
19849		svint32_t svaddwb(svint32_t, int16_t);
19850		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s64)))
19851		svint64_t svaddwb(svint64_t, int32_t);
19852		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s16)))
19853		svint16_t svaddwb(svint16_t, int8_t);
19854		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u32)))
19855		svuint32_t svaddwb(svuint32_t, uint16_t);
19856		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u64)))
19857		svuint64_t svaddwb(svuint64_t, uint32_t);
19858		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u16)))
19859		svuint16_t svaddwb(svuint16_t, uint8_t);
19860		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s32)))
19861		svint32_t svaddwb(svint32_t, svint16_t);
19862		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s64)))
19863		svint64_t svaddwb(svint64_t, svint32_t);
19864		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s16)))
19865		svint16_t svaddwb(svint16_t, svint8_t);
19866		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u32)))
19867		svuint32_t svaddwb(svuint32_t, svuint16_t);
19868		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u64)))
19869		svuint64_t svaddwb(svuint64_t, svuint32_t);
19870		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u16)))
19871		svuint16_t svaddwb(svuint16_t, svuint8_t);
19872		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s32)))
19873		svint32_t svaddwt(svint32_t, int16_t);
19874		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s64)))
19875		svint64_t svaddwt(svint64_t, int32_t);
19876		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s16)))
19877		svint16_t svaddwt(svint16_t, int8_t);
19878		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u32)))
19879		svuint32_t svaddwt(svuint32_t, uint16_t);
19880		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u64)))
19881		svuint64_t svaddwt(svuint64_t, uint32_t);
19882		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u16)))
19883		svuint16_t svaddwt(svuint16_t, uint8_t);
19884		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s32)))
19885		svint32_t svaddwt(svint32_t, svint16_t);
19886		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s64)))
19887		svint64_t svaddwt(svint64_t, svint32_t);
19888		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s16)))
19889		svint16_t svaddwt(svint16_t, svint8_t);
19890		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u32)))
19891		svuint32_t svaddwt(svuint32_t, svuint16_t);
19892		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u64)))
19893		svuint64_t svaddwt(svuint64_t, svuint32_t);
19894		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u16)))
19895		svuint16_t svaddwt(svuint16_t, svuint8_t);
19896		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u8)))
19897		svuint8_t svbcax(svuint8_t, svuint8_t, uint8_t);
19898		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u32)))
19899		svuint32_t svbcax(svuint32_t, svuint32_t, uint32_t);
19900		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u64)))
19901		svuint64_t svbcax(svuint64_t, svuint64_t, uint64_t);
19902		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u16)))
19903		svuint16_t svbcax(svuint16_t, svuint16_t, uint16_t);
19904		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s8)))
19905		svint8_t svbcax(svint8_t, svint8_t, int8_t);
19906		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s32)))
19907		svint32_t svbcax(svint32_t, svint32_t, int32_t);
19908		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s64)))
19909		svint64_t svbcax(svint64_t, svint64_t, int64_t);
19910		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s16)))
19911		svint16_t svbcax(svint16_t, svint16_t, int16_t);
19912		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u8)))
19913		svuint8_t svbcax(svuint8_t, svuint8_t, svuint8_t);
19914		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u32)))
19915		svuint32_t svbcax(svuint32_t, svuint32_t, svuint32_t);
19916		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u64)))
19917		svuint64_t svbcax(svuint64_t, svuint64_t, svuint64_t);
19918		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u16)))
19919		svuint16_t svbcax(svuint16_t, svuint16_t, svuint16_t);
19920		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s8)))
19921		svint8_t svbcax(svint8_t, svint8_t, svint8_t);
19922		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s32)))
19923		svint32_t svbcax(svint32_t, svint32_t, svint32_t);
19924		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s64)))
19925		svint64_t svbcax(svint64_t, svint64_t, svint64_t);
19926		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s16)))
19927		svint16_t svbcax(svint16_t, svint16_t, svint16_t);
19928		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u8)))
19929		svuint8_t svbsl1n(svuint8_t, svuint8_t, uint8_t);
19930		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u32)))
19931		svuint32_t svbsl1n(svuint32_t, svuint32_t, uint32_t);
19932		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u64)))
19933		svuint64_t svbsl1n(svuint64_t, svuint64_t, uint64_t);
19934		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u16)))
19935		svuint16_t svbsl1n(svuint16_t, svuint16_t, uint16_t);
19936		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s8)))
19937		svint8_t svbsl1n(svint8_t, svint8_t, int8_t);
19938		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s32)))
19939		svint32_t svbsl1n(svint32_t, svint32_t, int32_t);
19940		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s64)))
19941		svint64_t svbsl1n(svint64_t, svint64_t, int64_t);
19942		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s16)))
19943		svint16_t svbsl1n(svint16_t, svint16_t, int16_t);
19944		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u8)))
19945		svuint8_t svbsl1n(svuint8_t, svuint8_t, svuint8_t);
19946		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u32)))
19947		svuint32_t svbsl1n(svuint32_t, svuint32_t, svuint32_t);
19948		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u64)))
19949		svuint64_t svbsl1n(svuint64_t, svuint64_t, svuint64_t);
19950		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u16)))
19951		svuint16_t svbsl1n(svuint16_t, svuint16_t, svuint16_t);
19952		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s8)))
19953		svint8_t svbsl1n(svint8_t, svint8_t, svint8_t);
19954		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s32)))
19955		svint32_t svbsl1n(svint32_t, svint32_t, svint32_t);
19956		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s64)))
19957		svint64_t svbsl1n(svint64_t, svint64_t, svint64_t);
19958		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s16)))
19959		svint16_t svbsl1n(svint16_t, svint16_t, svint16_t);
19960		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u8)))
19961		svuint8_t svbsl2n(svuint8_t, svuint8_t, uint8_t);
19962		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u32)))
19963		svuint32_t svbsl2n(svuint32_t, svuint32_t, uint32_t);
19964		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u64)))
19965		svuint64_t svbsl2n(svuint64_t, svuint64_t, uint64_t);
19966		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u16)))
19967		svuint16_t svbsl2n(svuint16_t, svuint16_t, uint16_t);
19968		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s8)))
19969		svint8_t svbsl2n(svint8_t, svint8_t, int8_t);
19970		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s32)))
19971		svint32_t svbsl2n(svint32_t, svint32_t, int32_t);
19972		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s64)))
19973		svint64_t svbsl2n(svint64_t, svint64_t, int64_t);
19974		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s16)))
19975		svint16_t svbsl2n(svint16_t, svint16_t, int16_t);
19976		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u8)))
19977		svuint8_t svbsl2n(svuint8_t, svuint8_t, svuint8_t);
19978		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u32)))
19979		svuint32_t svbsl2n(svuint32_t, svuint32_t, svuint32_t);
19980		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u64)))
19981		svuint64_t svbsl2n(svuint64_t, svuint64_t, svuint64_t);
19982		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u16)))
19983		svuint16_t svbsl2n(svuint16_t, svuint16_t, svuint16_t);
19984		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s8)))
19985		svint8_t svbsl2n(svint8_t, svint8_t, svint8_t);
19986		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s32)))
19987		svint32_t svbsl2n(svint32_t, svint32_t, svint32_t);
19988		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s64)))
19989		svint64_t svbsl2n(svint64_t, svint64_t, svint64_t);
19990		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s16)))
19991		svint16_t svbsl2n(svint16_t, svint16_t, svint16_t);
19992		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u8)))
19993		svuint8_t svbsl(svuint8_t, svuint8_t, uint8_t);
19994		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u32)))
19995		svuint32_t svbsl(svuint32_t, svuint32_t, uint32_t);
19996		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u64)))
19997		svuint64_t svbsl(svuint64_t, svuint64_t, uint64_t);
19998		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u16)))
19999		svuint16_t svbsl(svuint16_t, svuint16_t, uint16_t);
20000		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s8)))
20001		svint8_t svbsl(svint8_t, svint8_t, int8_t);
20002		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s32)))
20003		svint32_t svbsl(svint32_t, svint32_t, int32_t);
20004		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s64)))
20005		svint64_t svbsl(svint64_t, svint64_t, int64_t);
20006		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s16)))
20007		svint16_t svbsl(svint16_t, svint16_t, int16_t);
20008		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u8)))
20009		svuint8_t svbsl(svuint8_t, svuint8_t, svuint8_t);
20010		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u32)))
20011		svuint32_t svbsl(svuint32_t, svuint32_t, svuint32_t);
20012		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u64)))
20013		svuint64_t svbsl(svuint64_t, svuint64_t, svuint64_t);
20014		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u16)))
20015		svuint16_t svbsl(svuint16_t, svuint16_t, svuint16_t);
20016		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s8)))
20017		svint8_t svbsl(svint8_t, svint8_t, svint8_t);
20018		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s32)))
20019		svint32_t svbsl(svint32_t, svint32_t, svint32_t);
20020		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s64)))
20021		svint64_t svbsl(svint64_t, svint64_t, svint64_t);
20022		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s16)))
20023		svint16_t svbsl(svint16_t, svint16_t, svint16_t);
20024		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u8)))
20025		svuint8_t svcadd(svuint8_t, svuint8_t, uint64_t);
20026		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u32)))
20027		svuint32_t svcadd(svuint32_t, svuint32_t, uint64_t);
20028		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u64)))
20029		svuint64_t svcadd(svuint64_t, svuint64_t, uint64_t);
20030		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u16)))
20031		svuint16_t svcadd(svuint16_t, svuint16_t, uint64_t);
20032		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s8)))
20033		svint8_t svcadd(svint8_t, svint8_t, uint64_t);
20034		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s32)))
20035		svint32_t svcadd(svint32_t, svint32_t, uint64_t);
20036		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s64)))
20037		svint64_t svcadd(svint64_t, svint64_t, uint64_t);
20038		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s16)))
20039		svint16_t svcadd(svint16_t, svint16_t, uint64_t);
20040		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s32)))
20041		svint32_t svcdot(svint32_t, svint8_t, svint8_t, uint64_t);
20042		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s64)))
20043		svint64_t svcdot(svint64_t, svint16_t, svint16_t, uint64_t);
20044		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s32)))
20045		svint32_t svcdot_lane(svint32_t, svint8_t, svint8_t, uint64_t, uint64_t);
20046		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s64)))
20047		svint64_t svcdot_lane(svint64_t, svint16_t, svint16_t, uint64_t, uint64_t);
20048		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u8)))
20049		svuint8_t svcmla(svuint8_t, svuint8_t, svuint8_t, uint64_t);
20050		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u32)))
20051		svuint32_t svcmla(svuint32_t, svuint32_t, svuint32_t, uint64_t);
20052		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u64)))
20053		svuint64_t svcmla(svuint64_t, svuint64_t, svuint64_t, uint64_t);
20054		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u16)))
20055		svuint16_t svcmla(svuint16_t, svuint16_t, svuint16_t, uint64_t);
20056		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s8)))
20057		svint8_t svcmla(svint8_t, svint8_t, svint8_t, uint64_t);
20058		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s32)))
20059		svint32_t svcmla(svint32_t, svint32_t, svint32_t, uint64_t);
20060		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s64)))
20061		svint64_t svcmla(svint64_t, svint64_t, svint64_t, uint64_t);
20062		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s16)))
20063		svint16_t svcmla(svint16_t, svint16_t, svint16_t, uint64_t);
20064		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u32)))
20065		svuint32_t svcmla_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t, uint64_t);
20066		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u16)))
20067		svuint16_t svcmla_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t, uint64_t);
20068		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s32)))
20069		svint32_t svcmla_lane(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
20070		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s16)))
20071		svint16_t svcmla_lane(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
20072		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_m)))
20073		svfloat32_t svcvtlt_f32_m(svfloat32_t, svbool_t, svfloat16_t);
20074		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_x)))
20075		svfloat32_t svcvtlt_f32_x(svbool_t, svfloat16_t);
20076		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_m)))
20077		svfloat64_t svcvtlt_f64_m(svfloat64_t, svbool_t, svfloat32_t);
20078		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_x)))
20079		svfloat64_t svcvtlt_f64_x(svbool_t, svfloat32_t);
20080		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f16_f32_m)))
20081		svfloat16_t svcvtnt_f16_m(svfloat16_t, svbool_t, svfloat32_t);
20082		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f32_f64_m)))
20083		svfloat32_t svcvtnt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
20084		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_m)))
20085		svfloat32_t svcvtx_f32_m(svfloat32_t, svbool_t, svfloat64_t);
20086		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_x)))
20087		svfloat32_t svcvtx_f32_x(svbool_t, svfloat64_t);
20088		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_z)))
20089		svfloat32_t svcvtx_f32_z(svbool_t, svfloat64_t);
20090		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtxnt_f32_f64_m)))
20091		svfloat32_t svcvtxnt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
20092		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u8)))
20093		svuint8_t sveor3(svuint8_t, svuint8_t, uint8_t);
20094		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u32)))
20095		svuint32_t sveor3(svuint32_t, svuint32_t, uint32_t);
20096		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u64)))
20097		svuint64_t sveor3(svuint64_t, svuint64_t, uint64_t);
20098		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u16)))
20099		svuint16_t sveor3(svuint16_t, svuint16_t, uint16_t);
20100		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s8)))
20101		svint8_t sveor3(svint8_t, svint8_t, int8_t);
20102		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s32)))
20103		svint32_t sveor3(svint32_t, svint32_t, int32_t);
20104		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s64)))
20105		svint64_t sveor3(svint64_t, svint64_t, int64_t);
20106		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s16)))
20107		svint16_t sveor3(svint16_t, svint16_t, int16_t);
20108		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u8)))
20109		svuint8_t sveor3(svuint8_t, svuint8_t, svuint8_t);
20110		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u32)))
20111		svuint32_t sveor3(svuint32_t, svuint32_t, svuint32_t);
20112		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u64)))
20113		svuint64_t sveor3(svuint64_t, svuint64_t, svuint64_t);
20114		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u16)))
20115		svuint16_t sveor3(svuint16_t, svuint16_t, svuint16_t);
20116		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s8)))
20117		svint8_t sveor3(svint8_t, svint8_t, svint8_t);
20118		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s32)))
20119		svint32_t sveor3(svint32_t, svint32_t, svint32_t);
20120		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s64)))
20121		svint64_t sveor3(svint64_t, svint64_t, svint64_t);
20122		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s16)))
20123		svint16_t sveor3(svint16_t, svint16_t, svint16_t);
20124		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u8)))
20125		svuint8_t sveorbt(svuint8_t, svuint8_t, uint8_t);
20126		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u32)))
20127		svuint32_t sveorbt(svuint32_t, svuint32_t, uint32_t);
20128		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u64)))
20129		svuint64_t sveorbt(svuint64_t, svuint64_t, uint64_t);
20130		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u16)))
20131		svuint16_t sveorbt(svuint16_t, svuint16_t, uint16_t);
20132		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s8)))
20133		svint8_t sveorbt(svint8_t, svint8_t, int8_t);
20134		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s32)))
20135		svint32_t sveorbt(svint32_t, svint32_t, int32_t);
20136		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s64)))
20137		svint64_t sveorbt(svint64_t, svint64_t, int64_t);
20138		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s16)))
20139		svint16_t sveorbt(svint16_t, svint16_t, int16_t);
20140		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u8)))
20141		svuint8_t sveorbt(svuint8_t, svuint8_t, svuint8_t);
20142		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u32)))
20143		svuint32_t sveorbt(svuint32_t, svuint32_t, svuint32_t);
20144		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u64)))
20145		svuint64_t sveorbt(svuint64_t, svuint64_t, svuint64_t);
20146		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u16)))
20147		svuint16_t sveorbt(svuint16_t, svuint16_t, svuint16_t);
20148		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s8)))
20149		svint8_t sveorbt(svint8_t, svint8_t, svint8_t);
20150		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s32)))
20151		svint32_t sveorbt(svint32_t, svint32_t, svint32_t);
20152		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s64)))
20153		svint64_t sveorbt(svint64_t, svint64_t, svint64_t);
20154		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s16)))
20155		svint16_t sveorbt(svint16_t, svint16_t, svint16_t);
20156		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u8)))
20157		svuint8_t sveortb(svuint8_t, svuint8_t, uint8_t);
20158		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u32)))
20159		svuint32_t sveortb(svuint32_t, svuint32_t, uint32_t);
20160		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u64)))
20161		svuint64_t sveortb(svuint64_t, svuint64_t, uint64_t);
20162		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u16)))
20163		svuint16_t sveortb(svuint16_t, svuint16_t, uint16_t);
20164		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s8)))
20165		svint8_t sveortb(svint8_t, svint8_t, int8_t);
20166		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s32)))
20167		svint32_t sveortb(svint32_t, svint32_t, int32_t);
20168		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s64)))
20169		svint64_t sveortb(svint64_t, svint64_t, int64_t);
20170		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s16)))
20171		svint16_t sveortb(svint16_t, svint16_t, int16_t);
20172		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u8)))
20173		svuint8_t sveortb(svuint8_t, svuint8_t, svuint8_t);
20174		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u32)))
20175		svuint32_t sveortb(svuint32_t, svuint32_t, svuint32_t);
20176		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u64)))
20177		svuint64_t sveortb(svuint64_t, svuint64_t, svuint64_t);
20178		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u16)))
20179		svuint16_t sveortb(svuint16_t, svuint16_t, svuint16_t);
20180		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s8)))
20181		svint8_t sveortb(svint8_t, svint8_t, svint8_t);
20182		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s32)))
20183		svint32_t sveortb(svint32_t, svint32_t, svint32_t);
20184		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s64)))
20185		svint64_t sveortb(svint64_t, svint64_t, svint64_t);
20186		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s16)))
20187		svint16_t sveortb(svint16_t, svint16_t, svint16_t);
20188		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_m)))
20189		svint8_t svhadd_m(svbool_t, svint8_t, int8_t);
20190		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_m)))
20191		svint32_t svhadd_m(svbool_t, svint32_t, int32_t);
20192		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_m)))
20193		svint64_t svhadd_m(svbool_t, svint64_t, int64_t);
20194		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_m)))
20195		svint16_t svhadd_m(svbool_t, svint16_t, int16_t);
20196		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_x)))
20197		svint8_t svhadd_x(svbool_t, svint8_t, int8_t);
20198		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_x)))
20199		svint32_t svhadd_x(svbool_t, svint32_t, int32_t);
20200		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_x)))
20201		svint64_t svhadd_x(svbool_t, svint64_t, int64_t);
20202		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_x)))
20203		svint16_t svhadd_x(svbool_t, svint16_t, int16_t);
20204		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_z)))
20205		svint8_t svhadd_z(svbool_t, svint8_t, int8_t);
20206		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_z)))
20207		svint32_t svhadd_z(svbool_t, svint32_t, int32_t);
20208		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_z)))
20209		svint64_t svhadd_z(svbool_t, svint64_t, int64_t);
20210		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_z)))
20211		svint16_t svhadd_z(svbool_t, svint16_t, int16_t);
20212		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_m)))
20213		svuint8_t svhadd_m(svbool_t, svuint8_t, uint8_t);
20214		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_m)))
20215		svuint32_t svhadd_m(svbool_t, svuint32_t, uint32_t);
20216		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_m)))
20217		svuint64_t svhadd_m(svbool_t, svuint64_t, uint64_t);
20218		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_m)))
20219		svuint16_t svhadd_m(svbool_t, svuint16_t, uint16_t);
20220		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_x)))
20221		svuint8_t svhadd_x(svbool_t, svuint8_t, uint8_t);
20222		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_x)))
20223		svuint32_t svhadd_x(svbool_t, svuint32_t, uint32_t);
20224		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_x)))
20225		svuint64_t svhadd_x(svbool_t, svuint64_t, uint64_t);
20226		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_x)))
20227		svuint16_t svhadd_x(svbool_t, svuint16_t, uint16_t);
20228		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_z)))
20229		svuint8_t svhadd_z(svbool_t, svuint8_t, uint8_t);
20230		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_z)))
20231		svuint32_t svhadd_z(svbool_t, svuint32_t, uint32_t);
20232		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_z)))
20233		svuint64_t svhadd_z(svbool_t, svuint64_t, uint64_t);
20234		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_z)))
20235		svuint16_t svhadd_z(svbool_t, svuint16_t, uint16_t);
20236		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_m)))
20237		svint8_t svhadd_m(svbool_t, svint8_t, svint8_t);
20238		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_m)))
20239		svint32_t svhadd_m(svbool_t, svint32_t, svint32_t);
20240		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_m)))
20241		svint64_t svhadd_m(svbool_t, svint64_t, svint64_t);
20242		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_m)))
20243		svint16_t svhadd_m(svbool_t, svint16_t, svint16_t);
20244		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_x)))
20245		svint8_t svhadd_x(svbool_t, svint8_t, svint8_t);
20246		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_x)))
20247		svint32_t svhadd_x(svbool_t, svint32_t, svint32_t);
20248		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_x)))
20249		svint64_t svhadd_x(svbool_t, svint64_t, svint64_t);
20250		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_x)))
20251		svint16_t svhadd_x(svbool_t, svint16_t, svint16_t);
20252		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_z)))
20253		svint8_t svhadd_z(svbool_t, svint8_t, svint8_t);
20254		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_z)))
20255		svint32_t svhadd_z(svbool_t, svint32_t, svint32_t);
20256		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_z)))
20257		svint64_t svhadd_z(svbool_t, svint64_t, svint64_t);
20258		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_z)))
20259		svint16_t svhadd_z(svbool_t, svint16_t, svint16_t);
20260		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_m)))
20261		svuint8_t svhadd_m(svbool_t, svuint8_t, svuint8_t);
20262		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_m)))
20263		svuint32_t svhadd_m(svbool_t, svuint32_t, svuint32_t);
20264		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_m)))
20265		svuint64_t svhadd_m(svbool_t, svuint64_t, svuint64_t);
20266		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_m)))
20267		svuint16_t svhadd_m(svbool_t, svuint16_t, svuint16_t);
20268		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_x)))
20269		svuint8_t svhadd_x(svbool_t, svuint8_t, svuint8_t);
20270		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_x)))
20271		svuint32_t svhadd_x(svbool_t, svuint32_t, svuint32_t);
20272		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_x)))
20273		svuint64_t svhadd_x(svbool_t, svuint64_t, svuint64_t);
20274		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_x)))
20275		svuint16_t svhadd_x(svbool_t, svuint16_t, svuint16_t);
20276		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_z)))
20277		svuint8_t svhadd_z(svbool_t, svuint8_t, svuint8_t);
20278		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_z)))
20279		svuint32_t svhadd_z(svbool_t, svuint32_t, svuint32_t);
20280		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_z)))
20281		svuint64_t svhadd_z(svbool_t, svuint64_t, svuint64_t);
20282		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_z)))
20283		svuint16_t svhadd_z(svbool_t, svuint16_t, svuint16_t);
20284		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u32_z)))
20285		svuint32_t svhistcnt_z(svbool_t, svuint32_t, svuint32_t);
20286		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u64_z)))
20287		svuint64_t svhistcnt_z(svbool_t, svuint64_t, svuint64_t);
20288		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s32_z)))
20289		svuint32_t svhistcnt_z(svbool_t, svint32_t, svint32_t);
20290		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s64_z)))
20291		svuint64_t svhistcnt_z(svbool_t, svint64_t, svint64_t);
20292		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_u8)))
20293		svuint8_t svhistseg(svuint8_t, svuint8_t);
20294		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_s8)))
20295		svuint8_t svhistseg(svint8_t, svint8_t);
20296		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_m)))
20297		svint8_t svhsub_m(svbool_t, svint8_t, int8_t);
20298		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_m)))
20299		svint32_t svhsub_m(svbool_t, svint32_t, int32_t);
20300		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_m)))
20301		svint64_t svhsub_m(svbool_t, svint64_t, int64_t);
20302		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_m)))
20303		svint16_t svhsub_m(svbool_t, svint16_t, int16_t);
20304		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_x)))
20305		svint8_t svhsub_x(svbool_t, svint8_t, int8_t);
20306		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_x)))
20307		svint32_t svhsub_x(svbool_t, svint32_t, int32_t);
20308		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_x)))
20309		svint64_t svhsub_x(svbool_t, svint64_t, int64_t);
20310		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_x)))
20311		svint16_t svhsub_x(svbool_t, svint16_t, int16_t);
20312		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_z)))
20313		svint8_t svhsub_z(svbool_t, svint8_t, int8_t);
20314		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_z)))
20315		svint32_t svhsub_z(svbool_t, svint32_t, int32_t);
20316		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_z)))
20317		svint64_t svhsub_z(svbool_t, svint64_t, int64_t);
20318		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_z)))
20319		svint16_t svhsub_z(svbool_t, svint16_t, int16_t);
20320		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_m)))
20321		svuint8_t svhsub_m(svbool_t, svuint8_t, uint8_t);
20322		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_m)))
20323		svuint32_t svhsub_m(svbool_t, svuint32_t, uint32_t);
20324		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_m)))
20325		svuint64_t svhsub_m(svbool_t, svuint64_t, uint64_t);
20326		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_m)))
20327		svuint16_t svhsub_m(svbool_t, svuint16_t, uint16_t);
20328		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_x)))
20329		svuint8_t svhsub_x(svbool_t, svuint8_t, uint8_t);
20330		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_x)))
20331		svuint32_t svhsub_x(svbool_t, svuint32_t, uint32_t);
20332		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_x)))
20333		svuint64_t svhsub_x(svbool_t, svuint64_t, uint64_t);
20334		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_x)))
20335		svuint16_t svhsub_x(svbool_t, svuint16_t, uint16_t);
20336		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_z)))
20337		svuint8_t svhsub_z(svbool_t, svuint8_t, uint8_t);
20338		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_z)))
20339		svuint32_t svhsub_z(svbool_t, svuint32_t, uint32_t);
20340		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_z)))
20341		svuint64_t svhsub_z(svbool_t, svuint64_t, uint64_t);
20342		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_z)))
20343		svuint16_t svhsub_z(svbool_t, svuint16_t, uint16_t);
20344		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_m)))
20345		svint8_t svhsub_m(svbool_t, svint8_t, svint8_t);
20346		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_m)))
20347		svint32_t svhsub_m(svbool_t, svint32_t, svint32_t);
20348		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_m)))
20349		svint64_t svhsub_m(svbool_t, svint64_t, svint64_t);
20350		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_m)))
20351		svint16_t svhsub_m(svbool_t, svint16_t, svint16_t);
20352		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_x)))
20353		svint8_t svhsub_x(svbool_t, svint8_t, svint8_t);
20354		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_x)))
20355		svint32_t svhsub_x(svbool_t, svint32_t, svint32_t);
20356		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_x)))
20357		svint64_t svhsub_x(svbool_t, svint64_t, svint64_t);
20358		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_x)))
20359		svint16_t svhsub_x(svbool_t, svint16_t, svint16_t);
20360		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_z)))
20361		svint8_t svhsub_z(svbool_t, svint8_t, svint8_t);
20362		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_z)))
20363		svint32_t svhsub_z(svbool_t, svint32_t, svint32_t);
20364		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_z)))
20365		svint64_t svhsub_z(svbool_t, svint64_t, svint64_t);
20366		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_z)))
20367		svint16_t svhsub_z(svbool_t, svint16_t, svint16_t);
20368		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_m)))
20369		svuint8_t svhsub_m(svbool_t, svuint8_t, svuint8_t);
20370		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_m)))
20371		svuint32_t svhsub_m(svbool_t, svuint32_t, svuint32_t);
20372		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_m)))
20373		svuint64_t svhsub_m(svbool_t, svuint64_t, svuint64_t);
20374		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_m)))
20375		svuint16_t svhsub_m(svbool_t, svuint16_t, svuint16_t);
20376		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_x)))
20377		svuint8_t svhsub_x(svbool_t, svuint8_t, svuint8_t);
20378		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_x)))
20379		svuint32_t svhsub_x(svbool_t, svuint32_t, svuint32_t);
20380		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_x)))
20381		svuint64_t svhsub_x(svbool_t, svuint64_t, svuint64_t);
20382		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_x)))
20383		svuint16_t svhsub_x(svbool_t, svuint16_t, svuint16_t);
20384		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_z)))
20385		svuint8_t svhsub_z(svbool_t, svuint8_t, svuint8_t);
20386		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_z)))
20387		svuint32_t svhsub_z(svbool_t, svuint32_t, svuint32_t);
20388		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_z)))
20389		svuint64_t svhsub_z(svbool_t, svuint64_t, svuint64_t);
20390		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_z)))
20391		svuint16_t svhsub_z(svbool_t, svuint16_t, svuint16_t);
20392		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_m)))
20393		svint8_t svhsubr_m(svbool_t, svint8_t, int8_t);
20394		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_m)))
20395		svint32_t svhsubr_m(svbool_t, svint32_t, int32_t);
20396		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_m)))
20397		svint64_t svhsubr_m(svbool_t, svint64_t, int64_t);
20398		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_m)))
20399		svint16_t svhsubr_m(svbool_t, svint16_t, int16_t);
20400		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_x)))
20401		svint8_t svhsubr_x(svbool_t, svint8_t, int8_t);
20402		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_x)))
20403		svint32_t svhsubr_x(svbool_t, svint32_t, int32_t);
20404		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_x)))
20405		svint64_t svhsubr_x(svbool_t, svint64_t, int64_t);
20406		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_x)))
20407		svint16_t svhsubr_x(svbool_t, svint16_t, int16_t);
20408		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_z)))
20409		svint8_t svhsubr_z(svbool_t, svint8_t, int8_t);
20410		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_z)))
20411		svint32_t svhsubr_z(svbool_t, svint32_t, int32_t);
20412		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_z)))
20413		svint64_t svhsubr_z(svbool_t, svint64_t, int64_t);
20414		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_z)))
20415		svint16_t svhsubr_z(svbool_t, svint16_t, int16_t);
20416		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_m)))
20417		svuint8_t svhsubr_m(svbool_t, svuint8_t, uint8_t);
20418		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_m)))
20419		svuint32_t svhsubr_m(svbool_t, svuint32_t, uint32_t);
20420		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_m)))
20421		svuint64_t svhsubr_m(svbool_t, svuint64_t, uint64_t);
20422		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_m)))
20423		svuint16_t svhsubr_m(svbool_t, svuint16_t, uint16_t);
20424		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_x)))
20425		svuint8_t svhsubr_x(svbool_t, svuint8_t, uint8_t);
20426		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_x)))
20427		svuint32_t svhsubr_x(svbool_t, svuint32_t, uint32_t);
20428		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_x)))
20429		svuint64_t svhsubr_x(svbool_t, svuint64_t, uint64_t);
20430		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_x)))
20431		svuint16_t svhsubr_x(svbool_t, svuint16_t, uint16_t);
20432		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_z)))
20433		svuint8_t svhsubr_z(svbool_t, svuint8_t, uint8_t);
20434		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_z)))
20435		svuint32_t svhsubr_z(svbool_t, svuint32_t, uint32_t);
20436		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_z)))
20437		svuint64_t svhsubr_z(svbool_t, svuint64_t, uint64_t);
20438		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_z)))
20439		svuint16_t svhsubr_z(svbool_t, svuint16_t, uint16_t);
20440		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_m)))
20441		svint8_t svhsubr_m(svbool_t, svint8_t, svint8_t);
20442		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_m)))
20443		svint32_t svhsubr_m(svbool_t, svint32_t, svint32_t);
20444		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_m)))
20445		svint64_t svhsubr_m(svbool_t, svint64_t, svint64_t);
20446		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_m)))
20447		svint16_t svhsubr_m(svbool_t, svint16_t, svint16_t);
20448		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_x)))
20449		svint8_t svhsubr_x(svbool_t, svint8_t, svint8_t);
20450		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_x)))
20451		svint32_t svhsubr_x(svbool_t, svint32_t, svint32_t);
20452		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_x)))
20453		svint64_t svhsubr_x(svbool_t, svint64_t, svint64_t);
20454		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_x)))
20455		svint16_t svhsubr_x(svbool_t, svint16_t, svint16_t);
20456		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_z)))
20457		svint8_t svhsubr_z(svbool_t, svint8_t, svint8_t);
20458		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_z)))
20459		svint32_t svhsubr_z(svbool_t, svint32_t, svint32_t);
20460		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_z)))
20461		svint64_t svhsubr_z(svbool_t, svint64_t, svint64_t);
20462		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_z)))
20463		svint16_t svhsubr_z(svbool_t, svint16_t, svint16_t);
20464		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_m)))
20465		svuint8_t svhsubr_m(svbool_t, svuint8_t, svuint8_t);
20466		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_m)))
20467		svuint32_t svhsubr_m(svbool_t, svuint32_t, svuint32_t);
20468		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_m)))
20469		svuint64_t svhsubr_m(svbool_t, svuint64_t, svuint64_t);
20470		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_m)))
20471		svuint16_t svhsubr_m(svbool_t, svuint16_t, svuint16_t);
20472		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_x)))
20473		svuint8_t svhsubr_x(svbool_t, svuint8_t, svuint8_t);
20474		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_x)))
20475		svuint32_t svhsubr_x(svbool_t, svuint32_t, svuint32_t);
20476		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_x)))
20477		svuint64_t svhsubr_x(svbool_t, svuint64_t, svuint64_t);
20478		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_x)))
20479		svuint16_t svhsubr_x(svbool_t, svuint16_t, svuint16_t);
20480		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_z)))
20481		svuint8_t svhsubr_z(svbool_t, svuint8_t, svuint8_t);
20482		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_z)))
20483		svuint32_t svhsubr_z(svbool_t, svuint32_t, svuint32_t);
20484		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_z)))
20485		svuint64_t svhsubr_z(svbool_t, svuint64_t, svuint64_t);
20486		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_z)))
20487		svuint16_t svhsubr_z(svbool_t, svuint16_t, svuint16_t);
20488		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_u32)))
20489		svuint32_t svldnt1_gather_index_u32(svbool_t, svuint32_t, int64_t);
20490		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_u64)))
20491		svuint64_t svldnt1_gather_index_u64(svbool_t, svuint64_t, int64_t);
20492		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_f64)))
20493		svfloat64_t svldnt1_gather_index_f64(svbool_t, svuint64_t, int64_t);
20494		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_f32)))
20495		svfloat32_t svldnt1_gather_index_f32(svbool_t, svuint32_t, int64_t);
20496		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_s32)))
20497		svint32_t svldnt1_gather_index_s32(svbool_t, svuint32_t, int64_t);
20498		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_s64)))
20499		svint64_t svldnt1_gather_index_s64(svbool_t, svuint64_t, int64_t);
20500		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_u32)))
20501		svuint32_t svldnt1_gather_offset_u32(svbool_t, svuint32_t, int64_t);
20502		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_u64)))
20503		svuint64_t svldnt1_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20504		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_f64)))
20505		svfloat64_t svldnt1_gather_offset_f64(svbool_t, svuint64_t, int64_t);
20506		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_f32)))
20507		svfloat32_t svldnt1_gather_offset_f32(svbool_t, svuint32_t, int64_t);
20508		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_s32)))
20509		svint32_t svldnt1_gather_offset_s32(svbool_t, svuint32_t, int64_t);
20510		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_s64)))
20511		svint64_t svldnt1_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20512		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_u32)))
20513		svuint32_t svldnt1_gather_u32(svbool_t, svuint32_t);
20514		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_u64)))
20515		svuint64_t svldnt1_gather_u64(svbool_t, svuint64_t);
20516		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_f64)))
20517		svfloat64_t svldnt1_gather_f64(svbool_t, svuint64_t);
20518		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_f32)))
20519		svfloat32_t svldnt1_gather_f32(svbool_t, svuint32_t);
20520		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_s32)))
20521		svint32_t svldnt1_gather_s32(svbool_t, svuint32_t);
20522		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_s64)))
20523		svint64_t svldnt1_gather_s64(svbool_t, svuint64_t);
20524		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_u64)))
20525		svuint64_t svldnt1_gather_index(svbool_t, uint64_t const *, svint64_t);
20526		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_f64)))
20527		svfloat64_t svldnt1_gather_index(svbool_t, float64_t const *, svint64_t);
20528		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_s64)))
20529		svint64_t svldnt1_gather_index(svbool_t, int64_t const *, svint64_t);
20530		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_u64)))
20531		svuint64_t svldnt1_gather_index(svbool_t, uint64_t const *, svuint64_t);
20532		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_f64)))
20533		svfloat64_t svldnt1_gather_index(svbool_t, float64_t const *, svuint64_t);
20534		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_s64)))
20535		svint64_t svldnt1_gather_index(svbool_t, int64_t const *, svuint64_t);
20536		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_u32)))
20537		svuint32_t svldnt1_gather_offset(svbool_t, uint32_t const *, svuint32_t);
20538		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_f32)))
20539		svfloat32_t svldnt1_gather_offset(svbool_t, float32_t const *, svuint32_t);
20540		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_s32)))
20541		svint32_t svldnt1_gather_offset(svbool_t, int32_t const *, svuint32_t);
20542		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_u64)))
20543		svuint64_t svldnt1_gather_offset(svbool_t, uint64_t const *, svint64_t);
20544		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_f64)))
20545		svfloat64_t svldnt1_gather_offset(svbool_t, float64_t const *, svint64_t);
20546		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_s64)))
20547		svint64_t svldnt1_gather_offset(svbool_t, int64_t const *, svint64_t);
20548		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_u64)))
20549		svuint64_t svldnt1_gather_offset(svbool_t, uint64_t const *, svuint64_t);
20550		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_f64)))
20551		svfloat64_t svldnt1_gather_offset(svbool_t, float64_t const *, svuint64_t);
20552		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_s64)))
20553		svint64_t svldnt1_gather_offset(svbool_t, int64_t const *, svuint64_t);
20554		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_u32)))
20555		svuint32_t svldnt1sb_gather_offset_u32(svbool_t, svuint32_t, int64_t);
20556		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_u64)))
20557		svuint64_t svldnt1sb_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20558		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_s32)))
20559		svint32_t svldnt1sb_gather_offset_s32(svbool_t, svuint32_t, int64_t);
20560		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_s64)))
20561		svint64_t svldnt1sb_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20562		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_u32)))
20563		svuint32_t svldnt1sb_gather_u32(svbool_t, svuint32_t);
20564		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_u64)))
20565		svuint64_t svldnt1sb_gather_u64(svbool_t, svuint64_t);
20566		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_s32)))
20567		svint32_t svldnt1sb_gather_s32(svbool_t, svuint32_t);
20568		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_s64)))
20569		svint64_t svldnt1sb_gather_s64(svbool_t, svuint64_t);
20570		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_u32)))
20571		svuint32_t svldnt1sb_gather_offset_u32(svbool_t, int8_t const *, svuint32_t);
20572		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_s32)))
20573		svint32_t svldnt1sb_gather_offset_s32(svbool_t, int8_t const *, svuint32_t);
20574		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_u64)))
20575		svuint64_t svldnt1sb_gather_offset_u64(svbool_t, int8_t const *, svint64_t);
20576		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_s64)))
20577		svint64_t svldnt1sb_gather_offset_s64(svbool_t, int8_t const *, svint64_t);
20578		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_u64)))
20579		svuint64_t svldnt1sb_gather_offset_u64(svbool_t, int8_t const *, svuint64_t);
20580		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_s64)))
20581		svint64_t svldnt1sb_gather_offset_s64(svbool_t, int8_t const *, svuint64_t);
20582		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_u32)))
20583		svuint32_t svldnt1sh_gather_index_u32(svbool_t, svuint32_t, int64_t);
20584		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_u64)))
20585		svuint64_t svldnt1sh_gather_index_u64(svbool_t, svuint64_t, int64_t);
20586		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_s32)))
20587		svint32_t svldnt1sh_gather_index_s32(svbool_t, svuint32_t, int64_t);
20588		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_s64)))
20589		svint64_t svldnt1sh_gather_index_s64(svbool_t, svuint64_t, int64_t);
20590		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_u32)))
20591		svuint32_t svldnt1sh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
20592		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_u64)))
20593		svuint64_t svldnt1sh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20594		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_s32)))
20595		svint32_t svldnt1sh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
20596		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_s64)))
20597		svint64_t svldnt1sh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20598		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_u32)))
20599		svuint32_t svldnt1sh_gather_u32(svbool_t, svuint32_t);
20600		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_u64)))
20601		svuint64_t svldnt1sh_gather_u64(svbool_t, svuint64_t);
20602		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_s32)))
20603		svint32_t svldnt1sh_gather_s32(svbool_t, svuint32_t);
20604		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_s64)))
20605		svint64_t svldnt1sh_gather_s64(svbool_t, svuint64_t);
20606		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_u64)))
20607		svuint64_t svldnt1sh_gather_index_u64(svbool_t, int16_t const *, svint64_t);
20608		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_s64)))
20609		svint64_t svldnt1sh_gather_index_s64(svbool_t, int16_t const *, svint64_t);
20610		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_u64)))
20611		svuint64_t svldnt1sh_gather_index_u64(svbool_t, int16_t const *, svuint64_t);
20612		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_s64)))
20613		svint64_t svldnt1sh_gather_index_s64(svbool_t, int16_t const *, svuint64_t);
20614		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_u32)))
20615		svuint32_t svldnt1sh_gather_offset_u32(svbool_t, int16_t const *, svuint32_t);
20616		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_s32)))
20617		svint32_t svldnt1sh_gather_offset_s32(svbool_t, int16_t const *, svuint32_t);
20618		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_u64)))
20619		svuint64_t svldnt1sh_gather_offset_u64(svbool_t, int16_t const *, svint64_t);
20620		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_s64)))
20621		svint64_t svldnt1sh_gather_offset_s64(svbool_t, int16_t const *, svint64_t);
20622		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_u64)))
20623		svuint64_t svldnt1sh_gather_offset_u64(svbool_t, int16_t const *, svuint64_t);
20624		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_s64)))
20625		svint64_t svldnt1sh_gather_offset_s64(svbool_t, int16_t const *, svuint64_t);
20626		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_u64)))
20627		svuint64_t svldnt1sw_gather_index_u64(svbool_t, svuint64_t, int64_t);
20628		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_s64)))
20629		svint64_t svldnt1sw_gather_index_s64(svbool_t, svuint64_t, int64_t);
20630		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_u64)))
20631		svuint64_t svldnt1sw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20632		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_s64)))
20633		svint64_t svldnt1sw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20634		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_u64)))
20635		svuint64_t svldnt1sw_gather_u64(svbool_t, svuint64_t);
20636		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_s64)))
20637		svint64_t svldnt1sw_gather_s64(svbool_t, svuint64_t);
20638		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_u64)))
20639		svuint64_t svldnt1sw_gather_index_u64(svbool_t, int32_t const *, svint64_t);
20640		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_s64)))
20641		svint64_t svldnt1sw_gather_index_s64(svbool_t, int32_t const *, svint64_t);
20642		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_u64)))
20643		svuint64_t svldnt1sw_gather_index_u64(svbool_t, int32_t const *, svuint64_t);
20644		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_s64)))
20645		svint64_t svldnt1sw_gather_index_s64(svbool_t, int32_t const *, svuint64_t);
20646		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_u64)))
20647		svuint64_t svldnt1sw_gather_offset_u64(svbool_t, int32_t const *, svint64_t);
20648		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_s64)))
20649		svint64_t svldnt1sw_gather_offset_s64(svbool_t, int32_t const *, svint64_t);
20650		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_u64)))
20651		svuint64_t svldnt1sw_gather_offset_u64(svbool_t, int32_t const *, svuint64_t);
20652		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_s64)))
20653		svint64_t svldnt1sw_gather_offset_s64(svbool_t, int32_t const *, svuint64_t);
20654		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_u32)))
20655		svuint32_t svldnt1ub_gather_offset_u32(svbool_t, svuint32_t, int64_t);
20656		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_u64)))
20657		svuint64_t svldnt1ub_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20658		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_s32)))
20659		svint32_t svldnt1ub_gather_offset_s32(svbool_t, svuint32_t, int64_t);
20660		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_s64)))
20661		svint64_t svldnt1ub_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20662		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_u32)))
20663		svuint32_t svldnt1ub_gather_u32(svbool_t, svuint32_t);
20664		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_u64)))
20665		svuint64_t svldnt1ub_gather_u64(svbool_t, svuint64_t);
20666		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_s32)))
20667		svint32_t svldnt1ub_gather_s32(svbool_t, svuint32_t);
20668		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_s64)))
20669		svint64_t svldnt1ub_gather_s64(svbool_t, svuint64_t);
20670		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_u32)))
20671		svuint32_t svldnt1ub_gather_offset_u32(svbool_t, uint8_t const *, svuint32_t);
20672		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_s32)))
20673		svint32_t svldnt1ub_gather_offset_s32(svbool_t, uint8_t const *, svuint32_t);
20674		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_u64)))
20675		svuint64_t svldnt1ub_gather_offset_u64(svbool_t, uint8_t const *, svint64_t);
20676		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_s64)))
20677		svint64_t svldnt1ub_gather_offset_s64(svbool_t, uint8_t const *, svint64_t);
20678		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_u64)))
20679		svuint64_t svldnt1ub_gather_offset_u64(svbool_t, uint8_t const *, svuint64_t);
20680		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_s64)))
20681		svint64_t svldnt1ub_gather_offset_s64(svbool_t, uint8_t const *, svuint64_t);
20682		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_u32)))
20683		svuint32_t svldnt1uh_gather_index_u32(svbool_t, svuint32_t, int64_t);
20684		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_u64)))
20685		svuint64_t svldnt1uh_gather_index_u64(svbool_t, svuint64_t, int64_t);
20686		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_s32)))
20687		svint32_t svldnt1uh_gather_index_s32(svbool_t, svuint32_t, int64_t);
20688		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_s64)))
20689		svint64_t svldnt1uh_gather_index_s64(svbool_t, svuint64_t, int64_t);
20690		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_u32)))
20691		svuint32_t svldnt1uh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
20692		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_u64)))
20693		svuint64_t svldnt1uh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20694		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_s32)))
20695		svint32_t svldnt1uh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
20696		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_s64)))
20697		svint64_t svldnt1uh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20698		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_u32)))
20699		svuint32_t svldnt1uh_gather_u32(svbool_t, svuint32_t);
20700		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_u64)))
20701		svuint64_t svldnt1uh_gather_u64(svbool_t, svuint64_t);
20702		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_s32)))
20703		svint32_t svldnt1uh_gather_s32(svbool_t, svuint32_t);
20704		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_s64)))
20705		svint64_t svldnt1uh_gather_s64(svbool_t, svuint64_t);
20706		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_u64)))
20707		svuint64_t svldnt1uh_gather_index_u64(svbool_t, uint16_t const *, svint64_t);
20708		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_s64)))
20709		svint64_t svldnt1uh_gather_index_s64(svbool_t, uint16_t const *, svint64_t);
20710		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_u64)))
20711		svuint64_t svldnt1uh_gather_index_u64(svbool_t, uint16_t const *, svuint64_t);
20712		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_s64)))
20713		svint64_t svldnt1uh_gather_index_s64(svbool_t, uint16_t const *, svuint64_t);
20714		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_u32)))
20715		svuint32_t svldnt1uh_gather_offset_u32(svbool_t, uint16_t const *, svuint32_t);
20716		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_s32)))
20717		svint32_t svldnt1uh_gather_offset_s32(svbool_t, uint16_t const *, svuint32_t);
20718		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_u64)))
20719		svuint64_t svldnt1uh_gather_offset_u64(svbool_t, uint16_t const *, svint64_t);
20720		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_s64)))
20721		svint64_t svldnt1uh_gather_offset_s64(svbool_t, uint16_t const *, svint64_t);
20722		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_u64)))
20723		svuint64_t svldnt1uh_gather_offset_u64(svbool_t, uint16_t const *, svuint64_t);
20724		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_s64)))
20725		svint64_t svldnt1uh_gather_offset_s64(svbool_t, uint16_t const *, svuint64_t);
20726		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_u64)))
20727		svuint64_t svldnt1uw_gather_index_u64(svbool_t, svuint64_t, int64_t);
20728		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_s64)))
20729		svint64_t svldnt1uw_gather_index_s64(svbool_t, svuint64_t, int64_t);
20730		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_u64)))
20731		svuint64_t svldnt1uw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
20732		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_s64)))
20733		svint64_t svldnt1uw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
20734		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_u64)))
20735		svuint64_t svldnt1uw_gather_u64(svbool_t, svuint64_t);
20736		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_s64)))
20737		svint64_t svldnt1uw_gather_s64(svbool_t, svuint64_t);
20738		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_u64)))
20739		svuint64_t svldnt1uw_gather_index_u64(svbool_t, uint32_t const *, svint64_t);
20740		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_s64)))
20741		svint64_t svldnt1uw_gather_index_s64(svbool_t, uint32_t const *, svint64_t);
20742		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_u64)))
20743		svuint64_t svldnt1uw_gather_index_u64(svbool_t, uint32_t const *, svuint64_t);
20744		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_s64)))
20745		svint64_t svldnt1uw_gather_index_s64(svbool_t, uint32_t const *, svuint64_t);
20746		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_u64)))
20747		svuint64_t svldnt1uw_gather_offset_u64(svbool_t, uint32_t const *, svint64_t);
20748		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_s64)))
20749		svint64_t svldnt1uw_gather_offset_s64(svbool_t, uint32_t const *, svint64_t);
20750		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_u64)))
20751		svuint64_t svldnt1uw_gather_offset_u64(svbool_t, uint32_t const *, svuint64_t);
20752		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_s64)))
20753		svint64_t svldnt1uw_gather_offset_s64(svbool_t, uint32_t const *, svuint64_t);
20754		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_m)))
20755		svint64_t svlogb_m(svint64_t, svbool_t, svfloat64_t);
20756		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_m)))
20757		svint32_t svlogb_m(svint32_t, svbool_t, svfloat32_t);
20758		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_m)))
20759		svint16_t svlogb_m(svint16_t, svbool_t, svfloat16_t);
20760		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_x)))
20761		svint64_t svlogb_x(svbool_t, svfloat64_t);
20762		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_x)))
20763		svint32_t svlogb_x(svbool_t, svfloat32_t);
20764		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_x)))
20765		svint16_t svlogb_x(svbool_t, svfloat16_t);
20766		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_z)))
20767		svint64_t svlogb_z(svbool_t, svfloat64_t);
20768		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_z)))
20769		svint32_t svlogb_z(svbool_t, svfloat32_t);
20770		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_z)))
20771		svint16_t svlogb_z(svbool_t, svfloat16_t);
20772		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u8)))
20773		svbool_t svmatch(svbool_t, svuint8_t, svuint8_t);
20774		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u16)))
20775		svbool_t svmatch(svbool_t, svuint16_t, svuint16_t);
20776		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s8)))
20777		svbool_t svmatch(svbool_t, svint8_t, svint8_t);
20778		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s16)))
20779		svbool_t svmatch(svbool_t, svint16_t, svint16_t);
20780		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_m)))
20781		svfloat64_t svmaxnmp_m(svbool_t, svfloat64_t, svfloat64_t);
20782		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_m)))
20783		svfloat32_t svmaxnmp_m(svbool_t, svfloat32_t, svfloat32_t);
20784		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_m)))
20785		svfloat16_t svmaxnmp_m(svbool_t, svfloat16_t, svfloat16_t);
20786		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_x)))
20787		svfloat64_t svmaxnmp_x(svbool_t, svfloat64_t, svfloat64_t);
20788		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_x)))
20789		svfloat32_t svmaxnmp_x(svbool_t, svfloat32_t, svfloat32_t);
20790		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_x)))
20791		svfloat16_t svmaxnmp_x(svbool_t, svfloat16_t, svfloat16_t);
20792		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_m)))
20793		svfloat64_t svmaxp_m(svbool_t, svfloat64_t, svfloat64_t);
20794		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_m)))
20795		svfloat32_t svmaxp_m(svbool_t, svfloat32_t, svfloat32_t);
20796		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_m)))
20797		svfloat16_t svmaxp_m(svbool_t, svfloat16_t, svfloat16_t);
20798		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_x)))
20799		svfloat64_t svmaxp_x(svbool_t, svfloat64_t, svfloat64_t);
20800		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_x)))
20801		svfloat32_t svmaxp_x(svbool_t, svfloat32_t, svfloat32_t);
20802		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_x)))
20803		svfloat16_t svmaxp_x(svbool_t, svfloat16_t, svfloat16_t);
20804		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_m)))
20805		svint8_t svmaxp_m(svbool_t, svint8_t, svint8_t);
20806		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_m)))
20807		svint32_t svmaxp_m(svbool_t, svint32_t, svint32_t);
20808		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_m)))
20809		svint64_t svmaxp_m(svbool_t, svint64_t, svint64_t);
20810		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_m)))
20811		svint16_t svmaxp_m(svbool_t, svint16_t, svint16_t);
20812		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_x)))
20813		svint8_t svmaxp_x(svbool_t, svint8_t, svint8_t);
20814		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_x)))
20815		svint32_t svmaxp_x(svbool_t, svint32_t, svint32_t);
20816		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_x)))
20817		svint64_t svmaxp_x(svbool_t, svint64_t, svint64_t);
20818		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_x)))
20819		svint16_t svmaxp_x(svbool_t, svint16_t, svint16_t);
20820		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_m)))
20821		svuint8_t svmaxp_m(svbool_t, svuint8_t, svuint8_t);
20822		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_m)))
20823		svuint32_t svmaxp_m(svbool_t, svuint32_t, svuint32_t);
20824		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_m)))
20825		svuint64_t svmaxp_m(svbool_t, svuint64_t, svuint64_t);
20826		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_m)))
20827		svuint16_t svmaxp_m(svbool_t, svuint16_t, svuint16_t);
20828		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_x)))
20829		svuint8_t svmaxp_x(svbool_t, svuint8_t, svuint8_t);
20830		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_x)))
20831		svuint32_t svmaxp_x(svbool_t, svuint32_t, svuint32_t);
20832		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_x)))
20833		svuint64_t svmaxp_x(svbool_t, svuint64_t, svuint64_t);
20834		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_x)))
20835		svuint16_t svmaxp_x(svbool_t, svuint16_t, svuint16_t);
20836		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_m)))
20837		svfloat64_t svminnmp_m(svbool_t, svfloat64_t, svfloat64_t);
20838		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_m)))
20839		svfloat32_t svminnmp_m(svbool_t, svfloat32_t, svfloat32_t);
20840		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_m)))
20841		svfloat16_t svminnmp_m(svbool_t, svfloat16_t, svfloat16_t);
20842		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_x)))
20843		svfloat64_t svminnmp_x(svbool_t, svfloat64_t, svfloat64_t);
20844		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_x)))
20845		svfloat32_t svminnmp_x(svbool_t, svfloat32_t, svfloat32_t);
20846		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_x)))
20847		svfloat16_t svminnmp_x(svbool_t, svfloat16_t, svfloat16_t);
20848		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_m)))
20849		svfloat64_t svminp_m(svbool_t, svfloat64_t, svfloat64_t);
20850		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_m)))
20851		svfloat32_t svminp_m(svbool_t, svfloat32_t, svfloat32_t);
20852		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_m)))
20853		svfloat16_t svminp_m(svbool_t, svfloat16_t, svfloat16_t);
20854		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_x)))
20855		svfloat64_t svminp_x(svbool_t, svfloat64_t, svfloat64_t);
20856		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_x)))
20857		svfloat32_t svminp_x(svbool_t, svfloat32_t, svfloat32_t);
20858		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_x)))
20859		svfloat16_t svminp_x(svbool_t, svfloat16_t, svfloat16_t);
20860		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_m)))
20861		svint8_t svminp_m(svbool_t, svint8_t, svint8_t);
20862		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_m)))
20863		svint32_t svminp_m(svbool_t, svint32_t, svint32_t);
20864		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_m)))
20865		svint64_t svminp_m(svbool_t, svint64_t, svint64_t);
20866		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_m)))
20867		svint16_t svminp_m(svbool_t, svint16_t, svint16_t);
20868		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_x)))
20869		svint8_t svminp_x(svbool_t, svint8_t, svint8_t);
20870		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_x)))
20871		svint32_t svminp_x(svbool_t, svint32_t, svint32_t);
20872		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_x)))
20873		svint64_t svminp_x(svbool_t, svint64_t, svint64_t);
20874		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_x)))
20875		svint16_t svminp_x(svbool_t, svint16_t, svint16_t);
20876		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_m)))
20877		svuint8_t svminp_m(svbool_t, svuint8_t, svuint8_t);
20878		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_m)))
20879		svuint32_t svminp_m(svbool_t, svuint32_t, svuint32_t);
20880		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_m)))
20881		svuint64_t svminp_m(svbool_t, svuint64_t, svuint64_t);
20882		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_m)))
20883		svuint16_t svminp_m(svbool_t, svuint16_t, svuint16_t);
20884		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_x)))
20885		svuint8_t svminp_x(svbool_t, svuint8_t, svuint8_t);
20886		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_x)))
20887		svuint32_t svminp_x(svbool_t, svuint32_t, svuint32_t);
20888		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_x)))
20889		svuint64_t svminp_x(svbool_t, svuint64_t, svuint64_t);
20890		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_x)))
20891		svuint16_t svminp_x(svbool_t, svuint16_t, svuint16_t);
20892		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u32)))
20893		svuint32_t svmla_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t);
20894		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u64)))
20895		svuint64_t svmla_lane(svuint64_t, svuint64_t, svuint64_t, uint64_t);
20896		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u16)))
20897		svuint16_t svmla_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t);
20898		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s32)))
20899		svint32_t svmla_lane(svint32_t, svint32_t, svint32_t, uint64_t);
20900		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s64)))
20901		svint64_t svmla_lane(svint64_t, svint64_t, svint64_t, uint64_t);
20902		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s16)))
20903		svint16_t svmla_lane(svint16_t, svint16_t, svint16_t, uint64_t);
20904		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_f32)))
20905		svfloat32_t svmlalb(svfloat32_t, svfloat16_t, float16_t);
20906		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s32)))
20907		svint32_t svmlalb(svint32_t, svint16_t, int16_t);
20908		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s64)))
20909		svint64_t svmlalb(svint64_t, svint32_t, int32_t);
20910		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s16)))
20911		svint16_t svmlalb(svint16_t, svint8_t, int8_t);
20912		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u32)))
20913		svuint32_t svmlalb(svuint32_t, svuint16_t, uint16_t);
20914		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u64)))
20915		svuint64_t svmlalb(svuint64_t, svuint32_t, uint32_t);
20916		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u16)))
20917		svuint16_t svmlalb(svuint16_t, svuint8_t, uint8_t);
20918		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_f32)))
20919		svfloat32_t svmlalb(svfloat32_t, svfloat16_t, svfloat16_t);
20920		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s32)))
20921		svint32_t svmlalb(svint32_t, svint16_t, svint16_t);
20922		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s64)))
20923		svint64_t svmlalb(svint64_t, svint32_t, svint32_t);
20924		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s16)))
20925		svint16_t svmlalb(svint16_t, svint8_t, svint8_t);
20926		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u32)))
20927		svuint32_t svmlalb(svuint32_t, svuint16_t, svuint16_t);
20928		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u64)))
20929		svuint64_t svmlalb(svuint64_t, svuint32_t, svuint32_t);
20930		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u16)))
20931		svuint16_t svmlalb(svuint16_t, svuint8_t, svuint8_t);
20932		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_f32)))
20933		svfloat32_t svmlalb_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
20934		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s32)))
20935		svint32_t svmlalb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
20936		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s64)))
20937		svint64_t svmlalb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
20938		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u32)))
20939		svuint32_t svmlalb_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
20940		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u64)))
20941		svuint64_t svmlalb_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
20942		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_f32)))
20943		svfloat32_t svmlalt(svfloat32_t, svfloat16_t, float16_t);
20944		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s32)))
20945		svint32_t svmlalt(svint32_t, svint16_t, int16_t);
20946		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s64)))
20947		svint64_t svmlalt(svint64_t, svint32_t, int32_t);
20948		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s16)))
20949		svint16_t svmlalt(svint16_t, svint8_t, int8_t);
20950		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u32)))
20951		svuint32_t svmlalt(svuint32_t, svuint16_t, uint16_t);
20952		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u64)))
20953		svuint64_t svmlalt(svuint64_t, svuint32_t, uint32_t);
20954		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u16)))
20955		svuint16_t svmlalt(svuint16_t, svuint8_t, uint8_t);
20956		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_f32)))
20957		svfloat32_t svmlalt(svfloat32_t, svfloat16_t, svfloat16_t);
20958		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s32)))
20959		svint32_t svmlalt(svint32_t, svint16_t, svint16_t);
20960		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s64)))
20961		svint64_t svmlalt(svint64_t, svint32_t, svint32_t);
20962		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s16)))
20963		svint16_t svmlalt(svint16_t, svint8_t, svint8_t);
20964		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u32)))
20965		svuint32_t svmlalt(svuint32_t, svuint16_t, svuint16_t);
20966		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u64)))
20967		svuint64_t svmlalt(svuint64_t, svuint32_t, svuint32_t);
20968		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u16)))
20969		svuint16_t svmlalt(svuint16_t, svuint8_t, svuint8_t);
20970		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_f32)))
20971		svfloat32_t svmlalt_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
20972		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s32)))
20973		svint32_t svmlalt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
20974		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s64)))
20975		svint64_t svmlalt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
20976		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u32)))
20977		svuint32_t svmlalt_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
20978		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u64)))
20979		svuint64_t svmlalt_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
20980		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u32)))
20981		svuint32_t svmls_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t);
20982		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u64)))
20983		svuint64_t svmls_lane(svuint64_t, svuint64_t, svuint64_t, uint64_t);
20984		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u16)))
20985		svuint16_t svmls_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t);
20986		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s32)))
20987		svint32_t svmls_lane(svint32_t, svint32_t, svint32_t, uint64_t);
20988		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s64)))
20989		svint64_t svmls_lane(svint64_t, svint64_t, svint64_t, uint64_t);
20990		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s16)))
20991		svint16_t svmls_lane(svint16_t, svint16_t, svint16_t, uint64_t);
20992		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_f32)))
20993		svfloat32_t svmlslb(svfloat32_t, svfloat16_t, float16_t);
20994		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s32)))
20995		svint32_t svmlslb(svint32_t, svint16_t, int16_t);
20996		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s64)))
20997		svint64_t svmlslb(svint64_t, svint32_t, int32_t);
20998		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s16)))
20999		svint16_t svmlslb(svint16_t, svint8_t, int8_t);
21000		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u32)))
21001		svuint32_t svmlslb(svuint32_t, svuint16_t, uint16_t);
21002		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u64)))
21003		svuint64_t svmlslb(svuint64_t, svuint32_t, uint32_t);
21004		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u16)))
21005		svuint16_t svmlslb(svuint16_t, svuint8_t, uint8_t);
21006		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_f32)))
21007		svfloat32_t svmlslb(svfloat32_t, svfloat16_t, svfloat16_t);
21008		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s32)))
21009		svint32_t svmlslb(svint32_t, svint16_t, svint16_t);
21010		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s64)))
21011		svint64_t svmlslb(svint64_t, svint32_t, svint32_t);
21012		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s16)))
21013		svint16_t svmlslb(svint16_t, svint8_t, svint8_t);
21014		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u32)))
21015		svuint32_t svmlslb(svuint32_t, svuint16_t, svuint16_t);
21016		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u64)))
21017		svuint64_t svmlslb(svuint64_t, svuint32_t, svuint32_t);
21018		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u16)))
21019		svuint16_t svmlslb(svuint16_t, svuint8_t, svuint8_t);
21020		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_f32)))
21021		svfloat32_t svmlslb_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
21022		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s32)))
21023		svint32_t svmlslb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
21024		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s64)))
21025		svint64_t svmlslb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
21026		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u32)))
21027		svuint32_t svmlslb_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
21028		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u64)))
21029		svuint64_t svmlslb_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
21030		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_f32)))
21031		svfloat32_t svmlslt(svfloat32_t, svfloat16_t, float16_t);
21032		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s32)))
21033		svint32_t svmlslt(svint32_t, svint16_t, int16_t);
21034		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s64)))
21035		svint64_t svmlslt(svint64_t, svint32_t, int32_t);
21036		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s16)))
21037		svint16_t svmlslt(svint16_t, svint8_t, int8_t);
21038		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u32)))
21039		svuint32_t svmlslt(svuint32_t, svuint16_t, uint16_t);
21040		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u64)))
21041		svuint64_t svmlslt(svuint64_t, svuint32_t, uint32_t);
21042		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u16)))
21043		svuint16_t svmlslt(svuint16_t, svuint8_t, uint8_t);
21044		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_f32)))
21045		svfloat32_t svmlslt(svfloat32_t, svfloat16_t, svfloat16_t);
21046		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s32)))
21047		svint32_t svmlslt(svint32_t, svint16_t, svint16_t);
21048		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s64)))
21049		svint64_t svmlslt(svint64_t, svint32_t, svint32_t);
21050		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s16)))
21051		svint16_t svmlslt(svint16_t, svint8_t, svint8_t);
21052		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u32)))
21053		svuint32_t svmlslt(svuint32_t, svuint16_t, svuint16_t);
21054		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u64)))
21055		svuint64_t svmlslt(svuint64_t, svuint32_t, svuint32_t);
21056		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u16)))
21057		svuint16_t svmlslt(svuint16_t, svuint8_t, svuint8_t);
21058		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_f32)))
21059		svfloat32_t svmlslt_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
21060		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s32)))
21061		svint32_t svmlslt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
21062		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s64)))
21063		svint64_t svmlslt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
21064		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u32)))
21065		svuint32_t svmlslt_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
21066		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u64)))
21067		svuint64_t svmlslt_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
21068		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s32)))
21069		svint32_t svmovlb(svint16_t);
21070		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s64)))
21071		svint64_t svmovlb(svint32_t);
21072		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s16)))
21073		svint16_t svmovlb(svint8_t);
21074		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u32)))
21075		svuint32_t svmovlb(svuint16_t);
21076		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u64)))
21077		svuint64_t svmovlb(svuint32_t);
21078		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u16)))
21079		svuint16_t svmovlb(svuint8_t);
21080		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s32)))
21081		svint32_t svmovlt(svint16_t);
21082		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s64)))
21083		svint64_t svmovlt(svint32_t);
21084		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s16)))
21085		svint16_t svmovlt(svint8_t);
21086		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u32)))
21087		svuint32_t svmovlt(svuint16_t);
21088		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u64)))
21089		svuint64_t svmovlt(svuint32_t);
21090		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u16)))
21091		svuint16_t svmovlt(svuint8_t);
21092		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u32)))
21093		svuint32_t svmul_lane(svuint32_t, svuint32_t, uint64_t);
21094		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u64)))
21095		svuint64_t svmul_lane(svuint64_t, svuint64_t, uint64_t);
21096		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u16)))
21097		svuint16_t svmul_lane(svuint16_t, svuint16_t, uint64_t);
21098		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s32)))
21099		svint32_t svmul_lane(svint32_t, svint32_t, uint64_t);
21100		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s64)))
21101		svint64_t svmul_lane(svint64_t, svint64_t, uint64_t);
21102		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s16)))
21103		svint16_t svmul_lane(svint16_t, svint16_t, uint64_t);
21104		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s32)))
21105		svint32_t svmullb(svint16_t, int16_t);
21106		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s64)))
21107		svint64_t svmullb(svint32_t, int32_t);
21108		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s16)))
21109		svint16_t svmullb(svint8_t, int8_t);
21110		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u32)))
21111		svuint32_t svmullb(svuint16_t, uint16_t);
21112		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u64)))
21113		svuint64_t svmullb(svuint32_t, uint32_t);
21114		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u16)))
21115		svuint16_t svmullb(svuint8_t, uint8_t);
21116		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s32)))
21117		svint32_t svmullb(svint16_t, svint16_t);
21118		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s64)))
21119		svint64_t svmullb(svint32_t, svint32_t);
21120		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s16)))
21121		svint16_t svmullb(svint8_t, svint8_t);
21122		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u32)))
21123		svuint32_t svmullb(svuint16_t, svuint16_t);
21124		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u64)))
21125		svuint64_t svmullb(svuint32_t, svuint32_t);
21126		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u16)))
21127		svuint16_t svmullb(svuint8_t, svuint8_t);
21128		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s32)))
21129		svint32_t svmullb_lane(svint16_t, svint16_t, uint64_t);
21130		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s64)))
21131		svint64_t svmullb_lane(svint32_t, svint32_t, uint64_t);
21132		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u32)))
21133		svuint32_t svmullb_lane(svuint16_t, svuint16_t, uint64_t);
21134		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u64)))
21135		svuint64_t svmullb_lane(svuint32_t, svuint32_t, uint64_t);
21136		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s32)))
21137		svint32_t svmullt(svint16_t, int16_t);
21138		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s64)))
21139		svint64_t svmullt(svint32_t, int32_t);
21140		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s16)))
21141		svint16_t svmullt(svint8_t, int8_t);
21142		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u32)))
21143		svuint32_t svmullt(svuint16_t, uint16_t);
21144		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u64)))
21145		svuint64_t svmullt(svuint32_t, uint32_t);
21146		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u16)))
21147		svuint16_t svmullt(svuint8_t, uint8_t);
21148		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s32)))
21149		svint32_t svmullt(svint16_t, svint16_t);
21150		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s64)))
21151		svint64_t svmullt(svint32_t, svint32_t);
21152		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s16)))
21153		svint16_t svmullt(svint8_t, svint8_t);
21154		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u32)))
21155		svuint32_t svmullt(svuint16_t, svuint16_t);
21156		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u64)))
21157		svuint64_t svmullt(svuint32_t, svuint32_t);
21158		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u16)))
21159		svuint16_t svmullt(svuint8_t, svuint8_t);
21160		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s32)))
21161		svint32_t svmullt_lane(svint16_t, svint16_t, uint64_t);
21162		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s64)))
21163		svint64_t svmullt_lane(svint32_t, svint32_t, uint64_t);
21164		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u32)))
21165		svuint32_t svmullt_lane(svuint16_t, svuint16_t, uint64_t);
21166		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u64)))
21167		svuint64_t svmullt_lane(svuint32_t, svuint32_t, uint64_t);
21168		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u8)))
21169		svuint8_t svnbsl(svuint8_t, svuint8_t, uint8_t);
21170		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u32)))
21171		svuint32_t svnbsl(svuint32_t, svuint32_t, uint32_t);
21172		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u64)))
21173		svuint64_t svnbsl(svuint64_t, svuint64_t, uint64_t);
21174		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u16)))
21175		svuint16_t svnbsl(svuint16_t, svuint16_t, uint16_t);
21176		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s8)))
21177		svint8_t svnbsl(svint8_t, svint8_t, int8_t);
21178		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s32)))
21179		svint32_t svnbsl(svint32_t, svint32_t, int32_t);
21180		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s64)))
21181		svint64_t svnbsl(svint64_t, svint64_t, int64_t);
21182		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s16)))
21183		svint16_t svnbsl(svint16_t, svint16_t, int16_t);
21184		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u8)))
21185		svuint8_t svnbsl(svuint8_t, svuint8_t, svuint8_t);
21186		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u32)))
21187		svuint32_t svnbsl(svuint32_t, svuint32_t, svuint32_t);
21188		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u64)))
21189		svuint64_t svnbsl(svuint64_t, svuint64_t, svuint64_t);
21190		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u16)))
21191		svuint16_t svnbsl(svuint16_t, svuint16_t, svuint16_t);
21192		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s8)))
21193		svint8_t svnbsl(svint8_t, svint8_t, svint8_t);
21194		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s32)))
21195		svint32_t svnbsl(svint32_t, svint32_t, svint32_t);
21196		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s64)))
21197		svint64_t svnbsl(svint64_t, svint64_t, svint64_t);
21198		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s16)))
21199		svint16_t svnbsl(svint16_t, svint16_t, svint16_t);
21200		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u8)))
21201		svbool_t svnmatch(svbool_t, svuint8_t, svuint8_t);
21202		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u16)))
21203		svbool_t svnmatch(svbool_t, svuint16_t, svuint16_t);
21204		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s8)))
21205		svbool_t svnmatch(svbool_t, svint8_t, svint8_t);
21206		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s16)))
21207		svbool_t svnmatch(svbool_t, svint16_t, svint16_t);
21208		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_n_u8)))
21209		svuint8_t svpmul(svuint8_t, uint8_t);
21210		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_u8)))
21211		svuint8_t svpmul(svuint8_t, svuint8_t);
21212		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u64)))
21213		svuint64_t svpmullb(svuint32_t, uint32_t);
21214		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u16)))
21215		svuint16_t svpmullb(svuint8_t, uint8_t);
21216		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u64)))
21217		svuint64_t svpmullb(svuint32_t, svuint32_t);
21218		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u16)))
21219		svuint16_t svpmullb(svuint8_t, svuint8_t);
21220		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u8)))
21221		svuint8_t svpmullb_pair(svuint8_t, uint8_t);
21222		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u32)))
21223		svuint32_t svpmullb_pair(svuint32_t, uint32_t);
21224		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u8)))
21225		svuint8_t svpmullb_pair(svuint8_t, svuint8_t);
21226		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u32)))
21227		svuint32_t svpmullb_pair(svuint32_t, svuint32_t);
21228		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u64)))
21229		svuint64_t svpmullt(svuint32_t, uint32_t);
21230		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u16)))
21231		svuint16_t svpmullt(svuint8_t, uint8_t);
21232		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u64)))
21233		svuint64_t svpmullt(svuint32_t, svuint32_t);
21234		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u16)))
21235		svuint16_t svpmullt(svuint8_t, svuint8_t);
21236		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u8)))
21237		svuint8_t svpmullt_pair(svuint8_t, uint8_t);
21238		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u32)))
21239		svuint32_t svpmullt_pair(svuint32_t, uint32_t);
21240		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u8)))
21241		svuint8_t svpmullt_pair(svuint8_t, svuint8_t);
21242		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u32)))
21243		svuint32_t svpmullt_pair(svuint32_t, svuint32_t);
21244		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_m)))
21245		svint8_t svqabs_m(svint8_t, svbool_t, svint8_t);
21246		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_m)))
21247		svint32_t svqabs_m(svint32_t, svbool_t, svint32_t);
21248		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_m)))
21249		svint64_t svqabs_m(svint64_t, svbool_t, svint64_t);
21250		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_m)))
21251		svint16_t svqabs_m(svint16_t, svbool_t, svint16_t);
21252		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_x)))
21253		svint8_t svqabs_x(svbool_t, svint8_t);
21254		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_x)))
21255		svint32_t svqabs_x(svbool_t, svint32_t);
21256		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_x)))
21257		svint64_t svqabs_x(svbool_t, svint64_t);
21258		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_x)))
21259		svint16_t svqabs_x(svbool_t, svint16_t);
21260		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_z)))
21261		svint8_t svqabs_z(svbool_t, svint8_t);
21262		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_z)))
21263		svint32_t svqabs_z(svbool_t, svint32_t);
21264		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_z)))
21265		svint64_t svqabs_z(svbool_t, svint64_t);
21266		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_z)))
21267		svint16_t svqabs_z(svbool_t, svint16_t);
21268		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_m)))
21269		svint8_t svqadd_m(svbool_t, svint8_t, int8_t);
21270		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_m)))
21271		svint32_t svqadd_m(svbool_t, svint32_t, int32_t);
21272		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_m)))
21273		svint64_t svqadd_m(svbool_t, svint64_t, int64_t);
21274		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_m)))
21275		svint16_t svqadd_m(svbool_t, svint16_t, int16_t);
21276		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_x)))
21277		svint8_t svqadd_x(svbool_t, svint8_t, int8_t);
21278		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_x)))
21279		svint32_t svqadd_x(svbool_t, svint32_t, int32_t);
21280		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_x)))
21281		svint64_t svqadd_x(svbool_t, svint64_t, int64_t);
21282		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_x)))
21283		svint16_t svqadd_x(svbool_t, svint16_t, int16_t);
21284		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_z)))
21285		svint8_t svqadd_z(svbool_t, svint8_t, int8_t);
21286		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_z)))
21287		svint32_t svqadd_z(svbool_t, svint32_t, int32_t);
21288		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_z)))
21289		svint64_t svqadd_z(svbool_t, svint64_t, int64_t);
21290		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_z)))
21291		svint16_t svqadd_z(svbool_t, svint16_t, int16_t);
21292		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_m)))
21293		svuint8_t svqadd_m(svbool_t, svuint8_t, uint8_t);
21294		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_m)))
21295		svuint32_t svqadd_m(svbool_t, svuint32_t, uint32_t);
21296		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_m)))
21297		svuint64_t svqadd_m(svbool_t, svuint64_t, uint64_t);
21298		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_m)))
21299		svuint16_t svqadd_m(svbool_t, svuint16_t, uint16_t);
21300		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_x)))
21301		svuint8_t svqadd_x(svbool_t, svuint8_t, uint8_t);
21302		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_x)))
21303		svuint32_t svqadd_x(svbool_t, svuint32_t, uint32_t);
21304		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_x)))
21305		svuint64_t svqadd_x(svbool_t, svuint64_t, uint64_t);
21306		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_x)))
21307		svuint16_t svqadd_x(svbool_t, svuint16_t, uint16_t);
21308		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_z)))
21309		svuint8_t svqadd_z(svbool_t, svuint8_t, uint8_t);
21310		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_z)))
21311		svuint32_t svqadd_z(svbool_t, svuint32_t, uint32_t);
21312		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_z)))
21313		svuint64_t svqadd_z(svbool_t, svuint64_t, uint64_t);
21314		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_z)))
21315		svuint16_t svqadd_z(svbool_t, svuint16_t, uint16_t);
21316		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_m)))
21317		svint8_t svqadd_m(svbool_t, svint8_t, svint8_t);
21318		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_m)))
21319		svint32_t svqadd_m(svbool_t, svint32_t, svint32_t);
21320		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_m)))
21321		svint64_t svqadd_m(svbool_t, svint64_t, svint64_t);
21322		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_m)))
21323		svint16_t svqadd_m(svbool_t, svint16_t, svint16_t);
21324		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_x)))
21325		svint8_t svqadd_x(svbool_t, svint8_t, svint8_t);
21326		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_x)))
21327		svint32_t svqadd_x(svbool_t, svint32_t, svint32_t);
21328		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_x)))
21329		svint64_t svqadd_x(svbool_t, svint64_t, svint64_t);
21330		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_x)))
21331		svint16_t svqadd_x(svbool_t, svint16_t, svint16_t);
21332		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_z)))
21333		svint8_t svqadd_z(svbool_t, svint8_t, svint8_t);
21334		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_z)))
21335		svint32_t svqadd_z(svbool_t, svint32_t, svint32_t);
21336		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_z)))
21337		svint64_t svqadd_z(svbool_t, svint64_t, svint64_t);
21338		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_z)))
21339		svint16_t svqadd_z(svbool_t, svint16_t, svint16_t);
21340		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_m)))
21341		svuint8_t svqadd_m(svbool_t, svuint8_t, svuint8_t);
21342		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_m)))
21343		svuint32_t svqadd_m(svbool_t, svuint32_t, svuint32_t);
21344		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_m)))
21345		svuint64_t svqadd_m(svbool_t, svuint64_t, svuint64_t);
21346		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_m)))
21347		svuint16_t svqadd_m(svbool_t, svuint16_t, svuint16_t);
21348		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_x)))
21349		svuint8_t svqadd_x(svbool_t, svuint8_t, svuint8_t);
21350		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_x)))
21351		svuint32_t svqadd_x(svbool_t, svuint32_t, svuint32_t);
21352		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_x)))
21353		svuint64_t svqadd_x(svbool_t, svuint64_t, svuint64_t);
21354		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_x)))
21355		svuint16_t svqadd_x(svbool_t, svuint16_t, svuint16_t);
21356		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_z)))
21357		svuint8_t svqadd_z(svbool_t, svuint8_t, svuint8_t);
21358		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_z)))
21359		svuint32_t svqadd_z(svbool_t, svuint32_t, svuint32_t);
21360		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_z)))
21361		svuint64_t svqadd_z(svbool_t, svuint64_t, svuint64_t);
21362		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_z)))
21363		svuint16_t svqadd_z(svbool_t, svuint16_t, svuint16_t);
21364		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s8)))
21365		svint8_t svqcadd(svint8_t, svint8_t, uint64_t);
21366		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s32)))
21367		svint32_t svqcadd(svint32_t, svint32_t, uint64_t);
21368		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s64)))
21369		svint64_t svqcadd(svint64_t, svint64_t, uint64_t);
21370		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s16)))
21371		svint16_t svqcadd(svint16_t, svint16_t, uint64_t);
21372		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s32)))
21373		svint32_t svqdmlalb(svint32_t, svint16_t, int16_t);
21374		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s64)))
21375		svint64_t svqdmlalb(svint64_t, svint32_t, int32_t);
21376		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s16)))
21377		svint16_t svqdmlalb(svint16_t, svint8_t, int8_t);
21378		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s32)))
21379		svint32_t svqdmlalb(svint32_t, svint16_t, svint16_t);
21380		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s64)))
21381		svint64_t svqdmlalb(svint64_t, svint32_t, svint32_t);
21382		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s16)))
21383		svint16_t svqdmlalb(svint16_t, svint8_t, svint8_t);
21384		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s32)))
21385		svint32_t svqdmlalb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
21386		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s64)))
21387		svint64_t svqdmlalb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
21388		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s32)))
21389		svint32_t svqdmlalbt(svint32_t, svint16_t, int16_t);
21390		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s64)))
21391		svint64_t svqdmlalbt(svint64_t, svint32_t, int32_t);
21392		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s16)))
21393		svint16_t svqdmlalbt(svint16_t, svint8_t, int8_t);
21394		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s32)))
21395		svint32_t svqdmlalbt(svint32_t, svint16_t, svint16_t);
21396		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s64)))
21397		svint64_t svqdmlalbt(svint64_t, svint32_t, svint32_t);
21398		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s16)))
21399		svint16_t svqdmlalbt(svint16_t, svint8_t, svint8_t);
21400		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s32)))
21401		svint32_t svqdmlalt(svint32_t, svint16_t, int16_t);
21402		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s64)))
21403		svint64_t svqdmlalt(svint64_t, svint32_t, int32_t);
21404		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s16)))
21405		svint16_t svqdmlalt(svint16_t, svint8_t, int8_t);
21406		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s32)))
21407		svint32_t svqdmlalt(svint32_t, svint16_t, svint16_t);
21408		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s64)))
21409		svint64_t svqdmlalt(svint64_t, svint32_t, svint32_t);
21410		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s16)))
21411		svint16_t svqdmlalt(svint16_t, svint8_t, svint8_t);
21412		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s32)))
21413		svint32_t svqdmlalt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
21414		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s64)))
21415		svint64_t svqdmlalt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
21416		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s32)))
21417		svint32_t svqdmlslb(svint32_t, svint16_t, int16_t);
21418		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s64)))
21419		svint64_t svqdmlslb(svint64_t, svint32_t, int32_t);
21420		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s16)))
21421		svint16_t svqdmlslb(svint16_t, svint8_t, int8_t);
21422		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s32)))
21423		svint32_t svqdmlslb(svint32_t, svint16_t, svint16_t);
21424		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s64)))
21425		svint64_t svqdmlslb(svint64_t, svint32_t, svint32_t);
21426		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s16)))
21427		svint16_t svqdmlslb(svint16_t, svint8_t, svint8_t);
21428		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s32)))
21429		svint32_t svqdmlslb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
21430		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s64)))
21431		svint64_t svqdmlslb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
21432		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s32)))
21433		svint32_t svqdmlslbt(svint32_t, svint16_t, int16_t);
21434		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s64)))
21435		svint64_t svqdmlslbt(svint64_t, svint32_t, int32_t);
21436		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s16)))
21437		svint16_t svqdmlslbt(svint16_t, svint8_t, int8_t);
21438		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s32)))
21439		svint32_t svqdmlslbt(svint32_t, svint16_t, svint16_t);
21440		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s64)))
21441		svint64_t svqdmlslbt(svint64_t, svint32_t, svint32_t);
21442		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s16)))
21443		svint16_t svqdmlslbt(svint16_t, svint8_t, svint8_t);
21444		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s32)))
21445		svint32_t svqdmlslt(svint32_t, svint16_t, int16_t);
21446		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s64)))
21447		svint64_t svqdmlslt(svint64_t, svint32_t, int32_t);
21448		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s16)))
21449		svint16_t svqdmlslt(svint16_t, svint8_t, int8_t);
21450		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s32)))
21451		svint32_t svqdmlslt(svint32_t, svint16_t, svint16_t);
21452		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s64)))
21453		svint64_t svqdmlslt(svint64_t, svint32_t, svint32_t);
21454		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s16)))
21455		svint16_t svqdmlslt(svint16_t, svint8_t, svint8_t);
21456		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s32)))
21457		svint32_t svqdmlslt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
21458		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s64)))
21459		svint64_t svqdmlslt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
21460		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s8)))
21461		svint8_t svqdmulh(svint8_t, int8_t);
21462		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s32)))
21463		svint32_t svqdmulh(svint32_t, int32_t);
21464		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s64)))
21465		svint64_t svqdmulh(svint64_t, int64_t);
21466		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s16)))
21467		svint16_t svqdmulh(svint16_t, int16_t);
21468		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s8)))
21469		svint8_t svqdmulh(svint8_t, svint8_t);
21470		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s32)))
21471		svint32_t svqdmulh(svint32_t, svint32_t);
21472		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s64)))
21473		svint64_t svqdmulh(svint64_t, svint64_t);
21474		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s16)))
21475		svint16_t svqdmulh(svint16_t, svint16_t);
21476		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s32)))
21477		svint32_t svqdmulh_lane(svint32_t, svint32_t, uint64_t);
21478		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s64)))
21479		svint64_t svqdmulh_lane(svint64_t, svint64_t, uint64_t);
21480		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s16)))
21481		svint16_t svqdmulh_lane(svint16_t, svint16_t, uint64_t);
21482		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s32)))
21483		svint32_t svqdmullb(svint16_t, int16_t);
21484		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s64)))
21485		svint64_t svqdmullb(svint32_t, int32_t);
21486		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s16)))
21487		svint16_t svqdmullb(svint8_t, int8_t);
21488		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s32)))
21489		svint32_t svqdmullb(svint16_t, svint16_t);
21490		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s64)))
21491		svint64_t svqdmullb(svint32_t, svint32_t);
21492		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s16)))
21493		svint16_t svqdmullb(svint8_t, svint8_t);
21494		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s32)))
21495		svint32_t svqdmullb_lane(svint16_t, svint16_t, uint64_t);
21496		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s64)))
21497		svint64_t svqdmullb_lane(svint32_t, svint32_t, uint64_t);
21498		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s32)))
21499		svint32_t svqdmullt(svint16_t, int16_t);
21500		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s64)))
21501		svint64_t svqdmullt(svint32_t, int32_t);
21502		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s16)))
21503		svint16_t svqdmullt(svint8_t, int8_t);
21504		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s32)))
21505		svint32_t svqdmullt(svint16_t, svint16_t);
21506		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s64)))
21507		svint64_t svqdmullt(svint32_t, svint32_t);
21508		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s16)))
21509		svint16_t svqdmullt(svint8_t, svint8_t);
21510		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s32)))
21511		svint32_t svqdmullt_lane(svint16_t, svint16_t, uint64_t);
21512		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s64)))
21513		svint64_t svqdmullt_lane(svint32_t, svint32_t, uint64_t);
21514		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_m)))
21515		svint8_t svqneg_m(svint8_t, svbool_t, svint8_t);
21516		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_m)))
21517		svint32_t svqneg_m(svint32_t, svbool_t, svint32_t);
21518		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_m)))
21519		svint64_t svqneg_m(svint64_t, svbool_t, svint64_t);
21520		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_m)))
21521		svint16_t svqneg_m(svint16_t, svbool_t, svint16_t);
21522		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_x)))
21523		svint8_t svqneg_x(svbool_t, svint8_t);
21524		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_x)))
21525		svint32_t svqneg_x(svbool_t, svint32_t);
21526		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_x)))
21527		svint64_t svqneg_x(svbool_t, svint64_t);
21528		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_x)))
21529		svint16_t svqneg_x(svbool_t, svint16_t);
21530		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_z)))
21531		svint8_t svqneg_z(svbool_t, svint8_t);
21532		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_z)))
21533		svint32_t svqneg_z(svbool_t, svint32_t);
21534		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_z)))
21535		svint64_t svqneg_z(svbool_t, svint64_t);
21536		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_z)))
21537		svint16_t svqneg_z(svbool_t, svint16_t);
21538		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s8)))
21539		svint8_t svqrdcmlah(svint8_t, svint8_t, svint8_t, uint64_t);
21540		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s32)))
21541		svint32_t svqrdcmlah(svint32_t, svint32_t, svint32_t, uint64_t);
21542		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s64)))
21543		svint64_t svqrdcmlah(svint64_t, svint64_t, svint64_t, uint64_t);
21544		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s16)))
21545		svint16_t svqrdcmlah(svint16_t, svint16_t, svint16_t, uint64_t);
21546		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s32)))
21547		svint32_t svqrdcmlah_lane(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
21548		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s16)))
21549		svint16_t svqrdcmlah_lane(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
21550		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s8)))
21551		svint8_t svqrdmlah(svint8_t, svint8_t, int8_t);
21552		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s32)))
21553		svint32_t svqrdmlah(svint32_t, svint32_t, int32_t);
21554		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s64)))
21555		svint64_t svqrdmlah(svint64_t, svint64_t, int64_t);
21556		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s16)))
21557		svint16_t svqrdmlah(svint16_t, svint16_t, int16_t);
21558		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s8)))
21559		svint8_t svqrdmlah(svint8_t, svint8_t, svint8_t);
21560		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s32)))
21561		svint32_t svqrdmlah(svint32_t, svint32_t, svint32_t);
21562		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s64)))
21563		svint64_t svqrdmlah(svint64_t, svint64_t, svint64_t);
21564		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s16)))
21565		svint16_t svqrdmlah(svint16_t, svint16_t, svint16_t);
21566		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s32)))
21567		svint32_t svqrdmlah_lane(svint32_t, svint32_t, svint32_t, uint64_t);
21568		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s64)))
21569		svint64_t svqrdmlah_lane(svint64_t, svint64_t, svint64_t, uint64_t);
21570		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s16)))
21571		svint16_t svqrdmlah_lane(svint16_t, svint16_t, svint16_t, uint64_t);
21572		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s8)))
21573		svint8_t svqrdmlsh(svint8_t, svint8_t, int8_t);
21574		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s32)))
21575		svint32_t svqrdmlsh(svint32_t, svint32_t, int32_t);
21576		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s64)))
21577		svint64_t svqrdmlsh(svint64_t, svint64_t, int64_t);
21578		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s16)))
21579		svint16_t svqrdmlsh(svint16_t, svint16_t, int16_t);
21580		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s8)))
21581		svint8_t svqrdmlsh(svint8_t, svint8_t, svint8_t);
21582		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s32)))
21583		svint32_t svqrdmlsh(svint32_t, svint32_t, svint32_t);
21584		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s64)))
21585		svint64_t svqrdmlsh(svint64_t, svint64_t, svint64_t);
21586		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s16)))
21587		svint16_t svqrdmlsh(svint16_t, svint16_t, svint16_t);
21588		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s32)))
21589		svint32_t svqrdmlsh_lane(svint32_t, svint32_t, svint32_t, uint64_t);
21590		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s64)))
21591		svint64_t svqrdmlsh_lane(svint64_t, svint64_t, svint64_t, uint64_t);
21592		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s16)))
21593		svint16_t svqrdmlsh_lane(svint16_t, svint16_t, svint16_t, uint64_t);
21594		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s8)))
21595		svint8_t svqrdmulh(svint8_t, int8_t);
21596		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s32)))
21597		svint32_t svqrdmulh(svint32_t, int32_t);
21598		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s64)))
21599		svint64_t svqrdmulh(svint64_t, int64_t);
21600		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s16)))
21601		svint16_t svqrdmulh(svint16_t, int16_t);
21602		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s8)))
21603		svint8_t svqrdmulh(svint8_t, svint8_t);
21604		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s32)))
21605		svint32_t svqrdmulh(svint32_t, svint32_t);
21606		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s64)))
21607		svint64_t svqrdmulh(svint64_t, svint64_t);
21608		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s16)))
21609		svint16_t svqrdmulh(svint16_t, svint16_t);
21610		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s32)))
21611		svint32_t svqrdmulh_lane(svint32_t, svint32_t, uint64_t);
21612		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s64)))
21613		svint64_t svqrdmulh_lane(svint64_t, svint64_t, uint64_t);
21614		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s16)))
21615		svint16_t svqrdmulh_lane(svint16_t, svint16_t, uint64_t);
21616		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_m)))
21617		svint8_t svqrshl_m(svbool_t, svint8_t, int8_t);
21618		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_m)))
21619		svint32_t svqrshl_m(svbool_t, svint32_t, int32_t);
21620		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_m)))
21621		svint64_t svqrshl_m(svbool_t, svint64_t, int64_t);
21622		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_m)))
21623		svint16_t svqrshl_m(svbool_t, svint16_t, int16_t);
21624		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_x)))
21625		svint8_t svqrshl_x(svbool_t, svint8_t, int8_t);
21626		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_x)))
21627		svint32_t svqrshl_x(svbool_t, svint32_t, int32_t);
21628		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_x)))
21629		svint64_t svqrshl_x(svbool_t, svint64_t, int64_t);
21630		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_x)))
21631		svint16_t svqrshl_x(svbool_t, svint16_t, int16_t);
21632		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_z)))
21633		svint8_t svqrshl_z(svbool_t, svint8_t, int8_t);
21634		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_z)))
21635		svint32_t svqrshl_z(svbool_t, svint32_t, int32_t);
21636		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_z)))
21637		svint64_t svqrshl_z(svbool_t, svint64_t, int64_t);
21638		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_z)))
21639		svint16_t svqrshl_z(svbool_t, svint16_t, int16_t);
21640		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_m)))
21641		svuint8_t svqrshl_m(svbool_t, svuint8_t, int8_t);
21642		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_m)))
21643		svuint32_t svqrshl_m(svbool_t, svuint32_t, int32_t);
21644		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_m)))
21645		svuint64_t svqrshl_m(svbool_t, svuint64_t, int64_t);
21646		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_m)))
21647		svuint16_t svqrshl_m(svbool_t, svuint16_t, int16_t);
21648		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_x)))
21649		svuint8_t svqrshl_x(svbool_t, svuint8_t, int8_t);
21650		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_x)))
21651		svuint32_t svqrshl_x(svbool_t, svuint32_t, int32_t);
21652		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_x)))
21653		svuint64_t svqrshl_x(svbool_t, svuint64_t, int64_t);
21654		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_x)))
21655		svuint16_t svqrshl_x(svbool_t, svuint16_t, int16_t);
21656		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_z)))
21657		svuint8_t svqrshl_z(svbool_t, svuint8_t, int8_t);
21658		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_z)))
21659		svuint32_t svqrshl_z(svbool_t, svuint32_t, int32_t);
21660		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_z)))
21661		svuint64_t svqrshl_z(svbool_t, svuint64_t, int64_t);
21662		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_z)))
21663		svuint16_t svqrshl_z(svbool_t, svuint16_t, int16_t);
21664		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_m)))
21665		svint8_t svqrshl_m(svbool_t, svint8_t, svint8_t);
21666		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_m)))
21667		svint32_t svqrshl_m(svbool_t, svint32_t, svint32_t);
21668		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_m)))
21669		svint64_t svqrshl_m(svbool_t, svint64_t, svint64_t);
21670		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_m)))
21671		svint16_t svqrshl_m(svbool_t, svint16_t, svint16_t);
21672		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_x)))
21673		svint8_t svqrshl_x(svbool_t, svint8_t, svint8_t);
21674		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_x)))
21675		svint32_t svqrshl_x(svbool_t, svint32_t, svint32_t);
21676		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_x)))
21677		svint64_t svqrshl_x(svbool_t, svint64_t, svint64_t);
21678		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_x)))
21679		svint16_t svqrshl_x(svbool_t, svint16_t, svint16_t);
21680		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_z)))
21681		svint8_t svqrshl_z(svbool_t, svint8_t, svint8_t);
21682		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_z)))
21683		svint32_t svqrshl_z(svbool_t, svint32_t, svint32_t);
21684		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_z)))
21685		svint64_t svqrshl_z(svbool_t, svint64_t, svint64_t);
21686		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_z)))
21687		svint16_t svqrshl_z(svbool_t, svint16_t, svint16_t);
21688		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_m)))
21689		svuint8_t svqrshl_m(svbool_t, svuint8_t, svint8_t);
21690		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_m)))
21691		svuint32_t svqrshl_m(svbool_t, svuint32_t, svint32_t);
21692		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_m)))
21693		svuint64_t svqrshl_m(svbool_t, svuint64_t, svint64_t);
21694		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_m)))
21695		svuint16_t svqrshl_m(svbool_t, svuint16_t, svint16_t);
21696		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_x)))
21697		svuint8_t svqrshl_x(svbool_t, svuint8_t, svint8_t);
21698		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_x)))
21699		svuint32_t svqrshl_x(svbool_t, svuint32_t, svint32_t);
21700		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_x)))
21701		svuint64_t svqrshl_x(svbool_t, svuint64_t, svint64_t);
21702		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_x)))
21703		svuint16_t svqrshl_x(svbool_t, svuint16_t, svint16_t);
21704		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_z)))
21705		svuint8_t svqrshl_z(svbool_t, svuint8_t, svint8_t);
21706		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_z)))
21707		svuint32_t svqrshl_z(svbool_t, svuint32_t, svint32_t);
21708		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_z)))
21709		svuint64_t svqrshl_z(svbool_t, svuint64_t, svint64_t);
21710		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_z)))
21711		svuint16_t svqrshl_z(svbool_t, svuint16_t, svint16_t);
21712		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s32)))
21713		svint16_t svqrshrnb(svint32_t, uint64_t);
21714		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s64)))
21715		svint32_t svqrshrnb(svint64_t, uint64_t);
21716		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s16)))
21717		svint8_t svqrshrnb(svint16_t, uint64_t);
21718		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u32)))
21719		svuint16_t svqrshrnb(svuint32_t, uint64_t);
21720		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u64)))
21721		svuint32_t svqrshrnb(svuint64_t, uint64_t);
21722		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u16)))
21723		svuint8_t svqrshrnb(svuint16_t, uint64_t);
21724		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s32)))
21725		svint16_t svqrshrnt(svint16_t, svint32_t, uint64_t);
21726		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s64)))
21727		svint32_t svqrshrnt(svint32_t, svint64_t, uint64_t);
21728		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s16)))
21729		svint8_t svqrshrnt(svint8_t, svint16_t, uint64_t);
21730		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u32)))
21731		svuint16_t svqrshrnt(svuint16_t, svuint32_t, uint64_t);
21732		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u64)))
21733		svuint32_t svqrshrnt(svuint32_t, svuint64_t, uint64_t);
21734		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u16)))
21735		svuint8_t svqrshrnt(svuint8_t, svuint16_t, uint64_t);
21736		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s32)))
21737		svuint16_t svqrshrunb(svint32_t, uint64_t);
21738		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s64)))
21739		svuint32_t svqrshrunb(svint64_t, uint64_t);
21740		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s16)))
21741		svuint8_t svqrshrunb(svint16_t, uint64_t);
21742		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s32)))
21743		svuint16_t svqrshrunt(svuint16_t, svint32_t, uint64_t);
21744		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s64)))
21745		svuint32_t svqrshrunt(svuint32_t, svint64_t, uint64_t);
21746		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s16)))
21747		svuint8_t svqrshrunt(svuint8_t, svint16_t, uint64_t);
21748		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_m)))
21749		svint8_t svqshl_m(svbool_t, svint8_t, int8_t);
21750		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_m)))
21751		svint32_t svqshl_m(svbool_t, svint32_t, int32_t);
21752		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_m)))
21753		svint64_t svqshl_m(svbool_t, svint64_t, int64_t);
21754		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_m)))
21755		svint16_t svqshl_m(svbool_t, svint16_t, int16_t);
21756		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_x)))
21757		svint8_t svqshl_x(svbool_t, svint8_t, int8_t);
21758		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_x)))
21759		svint32_t svqshl_x(svbool_t, svint32_t, int32_t);
21760		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_x)))
21761		svint64_t svqshl_x(svbool_t, svint64_t, int64_t);
21762		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_x)))
21763		svint16_t svqshl_x(svbool_t, svint16_t, int16_t);
21764		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_z)))
21765		svint8_t svqshl_z(svbool_t, svint8_t, int8_t);
21766		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_z)))
21767		svint32_t svqshl_z(svbool_t, svint32_t, int32_t);
21768		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_z)))
21769		svint64_t svqshl_z(svbool_t, svint64_t, int64_t);
21770		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_z)))
21771		svint16_t svqshl_z(svbool_t, svint16_t, int16_t);
21772		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_m)))
21773		svuint8_t svqshl_m(svbool_t, svuint8_t, int8_t);
21774		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_m)))
21775		svuint32_t svqshl_m(svbool_t, svuint32_t, int32_t);
21776		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_m)))
21777		svuint64_t svqshl_m(svbool_t, svuint64_t, int64_t);
21778		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_m)))
21779		svuint16_t svqshl_m(svbool_t, svuint16_t, int16_t);
21780		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_x)))
21781		svuint8_t svqshl_x(svbool_t, svuint8_t, int8_t);
21782		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_x)))
21783		svuint32_t svqshl_x(svbool_t, svuint32_t, int32_t);
21784		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_x)))
21785		svuint64_t svqshl_x(svbool_t, svuint64_t, int64_t);
21786		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_x)))
21787		svuint16_t svqshl_x(svbool_t, svuint16_t, int16_t);
21788		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_z)))
21789		svuint8_t svqshl_z(svbool_t, svuint8_t, int8_t);
21790		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_z)))
21791		svuint32_t svqshl_z(svbool_t, svuint32_t, int32_t);
21792		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_z)))
21793		svuint64_t svqshl_z(svbool_t, svuint64_t, int64_t);
21794		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_z)))
21795		svuint16_t svqshl_z(svbool_t, svuint16_t, int16_t);
21796		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_m)))
21797		svint8_t svqshl_m(svbool_t, svint8_t, svint8_t);
21798		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_m)))
21799		svint32_t svqshl_m(svbool_t, svint32_t, svint32_t);
21800		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_m)))
21801		svint64_t svqshl_m(svbool_t, svint64_t, svint64_t);
21802		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_m)))
21803		svint16_t svqshl_m(svbool_t, svint16_t, svint16_t);
21804		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_x)))
21805		svint8_t svqshl_x(svbool_t, svint8_t, svint8_t);
21806		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_x)))
21807		svint32_t svqshl_x(svbool_t, svint32_t, svint32_t);
21808		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_x)))
21809		svint64_t svqshl_x(svbool_t, svint64_t, svint64_t);
21810		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_x)))
21811		svint16_t svqshl_x(svbool_t, svint16_t, svint16_t);
21812		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_z)))
21813		svint8_t svqshl_z(svbool_t, svint8_t, svint8_t);
21814		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_z)))
21815		svint32_t svqshl_z(svbool_t, svint32_t, svint32_t);
21816		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_z)))
21817		svint64_t svqshl_z(svbool_t, svint64_t, svint64_t);
21818		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_z)))
21819		svint16_t svqshl_z(svbool_t, svint16_t, svint16_t);
21820		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_m)))
21821		svuint8_t svqshl_m(svbool_t, svuint8_t, svint8_t);
21822		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_m)))
21823		svuint32_t svqshl_m(svbool_t, svuint32_t, svint32_t);
21824		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_m)))
21825		svuint64_t svqshl_m(svbool_t, svuint64_t, svint64_t);
21826		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_m)))
21827		svuint16_t svqshl_m(svbool_t, svuint16_t, svint16_t);
21828		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_x)))
21829		svuint8_t svqshl_x(svbool_t, svuint8_t, svint8_t);
21830		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_x)))
21831		svuint32_t svqshl_x(svbool_t, svuint32_t, svint32_t);
21832		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_x)))
21833		svuint64_t svqshl_x(svbool_t, svuint64_t, svint64_t);
21834		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_x)))
21835		svuint16_t svqshl_x(svbool_t, svuint16_t, svint16_t);
21836		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_z)))
21837		svuint8_t svqshl_z(svbool_t, svuint8_t, svint8_t);
21838		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_z)))
21839		svuint32_t svqshl_z(svbool_t, svuint32_t, svint32_t);
21840		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_z)))
21841		svuint64_t svqshl_z(svbool_t, svuint64_t, svint64_t);
21842		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_z)))
21843		svuint16_t svqshl_z(svbool_t, svuint16_t, svint16_t);
21844		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_m)))
21845		svuint8_t svqshlu_m(svbool_t, svint8_t, uint64_t);
21846		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_m)))
21847		svuint32_t svqshlu_m(svbool_t, svint32_t, uint64_t);
21848		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_m)))
21849		svuint64_t svqshlu_m(svbool_t, svint64_t, uint64_t);
21850		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_m)))
21851		svuint16_t svqshlu_m(svbool_t, svint16_t, uint64_t);
21852		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_x)))
21853		svuint8_t svqshlu_x(svbool_t, svint8_t, uint64_t);
21854		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_x)))
21855		svuint32_t svqshlu_x(svbool_t, svint32_t, uint64_t);
21856		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_x)))
21857		svuint64_t svqshlu_x(svbool_t, svint64_t, uint64_t);
21858		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_x)))
21859		svuint16_t svqshlu_x(svbool_t, svint16_t, uint64_t);
21860		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_z)))
21861		svuint8_t svqshlu_z(svbool_t, svint8_t, uint64_t);
21862		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_z)))
21863		svuint32_t svqshlu_z(svbool_t, svint32_t, uint64_t);
21864		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_z)))
21865		svuint64_t svqshlu_z(svbool_t, svint64_t, uint64_t);
21866		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_z)))
21867		svuint16_t svqshlu_z(svbool_t, svint16_t, uint64_t);
21868		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s32)))
21869		svint16_t svqshrnb(svint32_t, uint64_t);
21870		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s64)))
21871		svint32_t svqshrnb(svint64_t, uint64_t);
21872		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s16)))
21873		svint8_t svqshrnb(svint16_t, uint64_t);
21874		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u32)))
21875		svuint16_t svqshrnb(svuint32_t, uint64_t);
21876		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u64)))
21877		svuint32_t svqshrnb(svuint64_t, uint64_t);
21878		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u16)))
21879		svuint8_t svqshrnb(svuint16_t, uint64_t);
21880		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s32)))
21881		svint16_t svqshrnt(svint16_t, svint32_t, uint64_t);
21882		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s64)))
21883		svint32_t svqshrnt(svint32_t, svint64_t, uint64_t);
21884		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s16)))
21885		svint8_t svqshrnt(svint8_t, svint16_t, uint64_t);
21886		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u32)))
21887		svuint16_t svqshrnt(svuint16_t, svuint32_t, uint64_t);
21888		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u64)))
21889		svuint32_t svqshrnt(svuint32_t, svuint64_t, uint64_t);
21890		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u16)))
21891		svuint8_t svqshrnt(svuint8_t, svuint16_t, uint64_t);
21892		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s32)))
21893		svuint16_t svqshrunb(svint32_t, uint64_t);
21894		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s64)))
21895		svuint32_t svqshrunb(svint64_t, uint64_t);
21896		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s16)))
21897		svuint8_t svqshrunb(svint16_t, uint64_t);
21898		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s32)))
21899		svuint16_t svqshrunt(svuint16_t, svint32_t, uint64_t);
21900		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s64)))
21901		svuint32_t svqshrunt(svuint32_t, svint64_t, uint64_t);
21902		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s16)))
21903		svuint8_t svqshrunt(svuint8_t, svint16_t, uint64_t);
21904		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_m)))
21905		svint8_t svqsub_m(svbool_t, svint8_t, int8_t);
21906		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_m)))
21907		svint32_t svqsub_m(svbool_t, svint32_t, int32_t);
21908		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_m)))
21909		svint64_t svqsub_m(svbool_t, svint64_t, int64_t);
21910		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_m)))
21911		svint16_t svqsub_m(svbool_t, svint16_t, int16_t);
21912		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_x)))
21913		svint8_t svqsub_x(svbool_t, svint8_t, int8_t);
21914		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_x)))
21915		svint32_t svqsub_x(svbool_t, svint32_t, int32_t);
21916		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_x)))
21917		svint64_t svqsub_x(svbool_t, svint64_t, int64_t);
21918		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_x)))
21919		svint16_t svqsub_x(svbool_t, svint16_t, int16_t);
21920		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_z)))
21921		svint8_t svqsub_z(svbool_t, svint8_t, int8_t);
21922		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_z)))
21923		svint32_t svqsub_z(svbool_t, svint32_t, int32_t);
21924		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_z)))
21925		svint64_t svqsub_z(svbool_t, svint64_t, int64_t);
21926		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_z)))
21927		svint16_t svqsub_z(svbool_t, svint16_t, int16_t);
21928		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_m)))
21929		svuint8_t svqsub_m(svbool_t, svuint8_t, uint8_t);
21930		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_m)))
21931		svuint32_t svqsub_m(svbool_t, svuint32_t, uint32_t);
21932		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_m)))
21933		svuint64_t svqsub_m(svbool_t, svuint64_t, uint64_t);
21934		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_m)))
21935		svuint16_t svqsub_m(svbool_t, svuint16_t, uint16_t);
21936		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_x)))
21937		svuint8_t svqsub_x(svbool_t, svuint8_t, uint8_t);
21938		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_x)))
21939		svuint32_t svqsub_x(svbool_t, svuint32_t, uint32_t);
21940		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_x)))
21941		svuint64_t svqsub_x(svbool_t, svuint64_t, uint64_t);
21942		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_x)))
21943		svuint16_t svqsub_x(svbool_t, svuint16_t, uint16_t);
21944		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_z)))
21945		svuint8_t svqsub_z(svbool_t, svuint8_t, uint8_t);
21946		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_z)))
21947		svuint32_t svqsub_z(svbool_t, svuint32_t, uint32_t);
21948		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_z)))
21949		svuint64_t svqsub_z(svbool_t, svuint64_t, uint64_t);
21950		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_z)))
21951		svuint16_t svqsub_z(svbool_t, svuint16_t, uint16_t);
21952		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_m)))
21953		svint8_t svqsub_m(svbool_t, svint8_t, svint8_t);
21954		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_m)))
21955		svint32_t svqsub_m(svbool_t, svint32_t, svint32_t);
21956		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_m)))
21957		svint64_t svqsub_m(svbool_t, svint64_t, svint64_t);
21958		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_m)))
21959		svint16_t svqsub_m(svbool_t, svint16_t, svint16_t);
21960		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_x)))
21961		svint8_t svqsub_x(svbool_t, svint8_t, svint8_t);
21962		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_x)))
21963		svint32_t svqsub_x(svbool_t, svint32_t, svint32_t);
21964		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_x)))
21965		svint64_t svqsub_x(svbool_t, svint64_t, svint64_t);
21966		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_x)))
21967		svint16_t svqsub_x(svbool_t, svint16_t, svint16_t);
21968		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_z)))
21969		svint8_t svqsub_z(svbool_t, svint8_t, svint8_t);
21970		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_z)))
21971		svint32_t svqsub_z(svbool_t, svint32_t, svint32_t);
21972		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_z)))
21973		svint64_t svqsub_z(svbool_t, svint64_t, svint64_t);
21974		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_z)))
21975		svint16_t svqsub_z(svbool_t, svint16_t, svint16_t);
21976		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_m)))
21977		svuint8_t svqsub_m(svbool_t, svuint8_t, svuint8_t);
21978		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_m)))
21979		svuint32_t svqsub_m(svbool_t, svuint32_t, svuint32_t);
21980		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_m)))
21981		svuint64_t svqsub_m(svbool_t, svuint64_t, svuint64_t);
21982		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_m)))
21983		svuint16_t svqsub_m(svbool_t, svuint16_t, svuint16_t);
21984		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_x)))
21985		svuint8_t svqsub_x(svbool_t, svuint8_t, svuint8_t);
21986		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_x)))
21987		svuint32_t svqsub_x(svbool_t, svuint32_t, svuint32_t);
21988		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_x)))
21989		svuint64_t svqsub_x(svbool_t, svuint64_t, svuint64_t);
21990		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_x)))
21991		svuint16_t svqsub_x(svbool_t, svuint16_t, svuint16_t);
21992		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_z)))
21993		svuint8_t svqsub_z(svbool_t, svuint8_t, svuint8_t);
21994		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_z)))
21995		svuint32_t svqsub_z(svbool_t, svuint32_t, svuint32_t);
21996		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_z)))
21997		svuint64_t svqsub_z(svbool_t, svuint64_t, svuint64_t);
21998		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_z)))
21999		svuint16_t svqsub_z(svbool_t, svuint16_t, svuint16_t);
22000		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_m)))
22001		svint8_t svqsubr_m(svbool_t, svint8_t, int8_t);
22002		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_m)))
22003		svint32_t svqsubr_m(svbool_t, svint32_t, int32_t);
22004		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_m)))
22005		svint64_t svqsubr_m(svbool_t, svint64_t, int64_t);
22006		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_m)))
22007		svint16_t svqsubr_m(svbool_t, svint16_t, int16_t);
22008		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_x)))
22009		svint8_t svqsubr_x(svbool_t, svint8_t, int8_t);
22010		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_x)))
22011		svint32_t svqsubr_x(svbool_t, svint32_t, int32_t);
22012		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_x)))
22013		svint64_t svqsubr_x(svbool_t, svint64_t, int64_t);
22014		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_x)))
22015		svint16_t svqsubr_x(svbool_t, svint16_t, int16_t);
22016		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_z)))
22017		svint8_t svqsubr_z(svbool_t, svint8_t, int8_t);
22018		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_z)))
22019		svint32_t svqsubr_z(svbool_t, svint32_t, int32_t);
22020		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_z)))
22021		svint64_t svqsubr_z(svbool_t, svint64_t, int64_t);
22022		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_z)))
22023		svint16_t svqsubr_z(svbool_t, svint16_t, int16_t);
22024		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_m)))
22025		svuint8_t svqsubr_m(svbool_t, svuint8_t, uint8_t);
22026		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_m)))
22027		svuint32_t svqsubr_m(svbool_t, svuint32_t, uint32_t);
22028		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_m)))
22029		svuint64_t svqsubr_m(svbool_t, svuint64_t, uint64_t);
22030		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_m)))
22031		svuint16_t svqsubr_m(svbool_t, svuint16_t, uint16_t);
22032		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_x)))
22033		svuint8_t svqsubr_x(svbool_t, svuint8_t, uint8_t);
22034		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_x)))
22035		svuint32_t svqsubr_x(svbool_t, svuint32_t, uint32_t);
22036		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_x)))
22037		svuint64_t svqsubr_x(svbool_t, svuint64_t, uint64_t);
22038		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_x)))
22039		svuint16_t svqsubr_x(svbool_t, svuint16_t, uint16_t);
22040		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_z)))
22041		svuint8_t svqsubr_z(svbool_t, svuint8_t, uint8_t);
22042		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_z)))
22043		svuint32_t svqsubr_z(svbool_t, svuint32_t, uint32_t);
22044		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_z)))
22045		svuint64_t svqsubr_z(svbool_t, svuint64_t, uint64_t);
22046		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_z)))
22047		svuint16_t svqsubr_z(svbool_t, svuint16_t, uint16_t);
22048		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_m)))
22049		svint8_t svqsubr_m(svbool_t, svint8_t, svint8_t);
22050		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_m)))
22051		svint32_t svqsubr_m(svbool_t, svint32_t, svint32_t);
22052		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_m)))
22053		svint64_t svqsubr_m(svbool_t, svint64_t, svint64_t);
22054		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_m)))
22055		svint16_t svqsubr_m(svbool_t, svint16_t, svint16_t);
22056		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_x)))
22057		svint8_t svqsubr_x(svbool_t, svint8_t, svint8_t);
22058		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_x)))
22059		svint32_t svqsubr_x(svbool_t, svint32_t, svint32_t);
22060		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_x)))
22061		svint64_t svqsubr_x(svbool_t, svint64_t, svint64_t);
22062		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_x)))
22063		svint16_t svqsubr_x(svbool_t, svint16_t, svint16_t);
22064		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_z)))
22065		svint8_t svqsubr_z(svbool_t, svint8_t, svint8_t);
22066		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_z)))
22067		svint32_t svqsubr_z(svbool_t, svint32_t, svint32_t);
22068		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_z)))
22069		svint64_t svqsubr_z(svbool_t, svint64_t, svint64_t);
22070		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_z)))
22071		svint16_t svqsubr_z(svbool_t, svint16_t, svint16_t);
22072		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_m)))
22073		svuint8_t svqsubr_m(svbool_t, svuint8_t, svuint8_t);
22074		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_m)))
22075		svuint32_t svqsubr_m(svbool_t, svuint32_t, svuint32_t);
22076		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_m)))
22077		svuint64_t svqsubr_m(svbool_t, svuint64_t, svuint64_t);
22078		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_m)))
22079		svuint16_t svqsubr_m(svbool_t, svuint16_t, svuint16_t);
22080		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_x)))
22081		svuint8_t svqsubr_x(svbool_t, svuint8_t, svuint8_t);
22082		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_x)))
22083		svuint32_t svqsubr_x(svbool_t, svuint32_t, svuint32_t);
22084		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_x)))
22085		svuint64_t svqsubr_x(svbool_t, svuint64_t, svuint64_t);
22086		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_x)))
22087		svuint16_t svqsubr_x(svbool_t, svuint16_t, svuint16_t);
22088		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_z)))
22089		svuint8_t svqsubr_z(svbool_t, svuint8_t, svuint8_t);
22090		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_z)))
22091		svuint32_t svqsubr_z(svbool_t, svuint32_t, svuint32_t);
22092		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_z)))
22093		svuint64_t svqsubr_z(svbool_t, svuint64_t, svuint64_t);
22094		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_z)))
22095		svuint16_t svqsubr_z(svbool_t, svuint16_t, svuint16_t);
22096		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s32)))
22097		svint16_t svqxtnb(svint32_t);
22098		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s64)))
22099		svint32_t svqxtnb(svint64_t);
22100		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s16)))
22101		svint8_t svqxtnb(svint16_t);
22102		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u32)))
22103		svuint16_t svqxtnb(svuint32_t);
22104		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u64)))
22105		svuint32_t svqxtnb(svuint64_t);
22106		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u16)))
22107		svuint8_t svqxtnb(svuint16_t);
22108		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s32)))
22109		svint16_t svqxtnt(svint16_t, svint32_t);
22110		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s64)))
22111		svint32_t svqxtnt(svint32_t, svint64_t);
22112		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s16)))
22113		svint8_t svqxtnt(svint8_t, svint16_t);
22114		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u32)))
22115		svuint16_t svqxtnt(svuint16_t, svuint32_t);
22116		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u64)))
22117		svuint32_t svqxtnt(svuint32_t, svuint64_t);
22118		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u16)))
22119		svuint8_t svqxtnt(svuint8_t, svuint16_t);
22120		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s32)))
22121		svuint16_t svqxtunb(svint32_t);
22122		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s64)))
22123		svuint32_t svqxtunb(svint64_t);
22124		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s16)))
22125		svuint8_t svqxtunb(svint16_t);
22126		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s32)))
22127		svuint16_t svqxtunt(svuint16_t, svint32_t);
22128		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s64)))
22129		svuint32_t svqxtunt(svuint32_t, svint64_t);
22130		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s16)))
22131		svuint8_t svqxtunt(svuint8_t, svint16_t);
22132		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u32)))
22133		svuint16_t svraddhnb(svuint32_t, uint32_t);
22134		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u64)))
22135		svuint32_t svraddhnb(svuint64_t, uint64_t);
22136		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u16)))
22137		svuint8_t svraddhnb(svuint16_t, uint16_t);
22138		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s32)))
22139		svint16_t svraddhnb(svint32_t, int32_t);
22140		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s64)))
22141		svint32_t svraddhnb(svint64_t, int64_t);
22142		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s16)))
22143		svint8_t svraddhnb(svint16_t, int16_t);
22144		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u32)))
22145		svuint16_t svraddhnb(svuint32_t, svuint32_t);
22146		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u64)))
22147		svuint32_t svraddhnb(svuint64_t, svuint64_t);
22148		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u16)))
22149		svuint8_t svraddhnb(svuint16_t, svuint16_t);
22150		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s32)))
22151		svint16_t svraddhnb(svint32_t, svint32_t);
22152		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s64)))
22153		svint32_t svraddhnb(svint64_t, svint64_t);
22154		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s16)))
22155		svint8_t svraddhnb(svint16_t, svint16_t);
22156		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u32)))
22157		svuint16_t svraddhnt(svuint16_t, svuint32_t, uint32_t);
22158		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u64)))
22159		svuint32_t svraddhnt(svuint32_t, svuint64_t, uint64_t);
22160		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u16)))
22161		svuint8_t svraddhnt(svuint8_t, svuint16_t, uint16_t);
22162		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s32)))
22163		svint16_t svraddhnt(svint16_t, svint32_t, int32_t);
22164		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s64)))
22165		svint32_t svraddhnt(svint32_t, svint64_t, int64_t);
22166		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s16)))
22167		svint8_t svraddhnt(svint8_t, svint16_t, int16_t);
22168		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u32)))
22169		svuint16_t svraddhnt(svuint16_t, svuint32_t, svuint32_t);
22170		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u64)))
22171		svuint32_t svraddhnt(svuint32_t, svuint64_t, svuint64_t);
22172		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u16)))
22173		svuint8_t svraddhnt(svuint8_t, svuint16_t, svuint16_t);
22174		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s32)))
22175		svint16_t svraddhnt(svint16_t, svint32_t, svint32_t);
22176		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s64)))
22177		svint32_t svraddhnt(svint32_t, svint64_t, svint64_t);
22178		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s16)))
22179		svint8_t svraddhnt(svint8_t, svint16_t, svint16_t);
22180		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_m)))
22181		svuint32_t svrecpe_m(svuint32_t, svbool_t, svuint32_t);
22182		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_x)))
22183		svuint32_t svrecpe_x(svbool_t, svuint32_t);
22184		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_z)))
22185		svuint32_t svrecpe_z(svbool_t, svuint32_t);
22186		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_m)))
22187		svint8_t svrhadd_m(svbool_t, svint8_t, int8_t);
22188		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_m)))
22189		svint32_t svrhadd_m(svbool_t, svint32_t, int32_t);
22190		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_m)))
22191		svint64_t svrhadd_m(svbool_t, svint64_t, int64_t);
22192		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_m)))
22193		svint16_t svrhadd_m(svbool_t, svint16_t, int16_t);
22194		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_x)))
22195		svint8_t svrhadd_x(svbool_t, svint8_t, int8_t);
22196		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_x)))
22197		svint32_t svrhadd_x(svbool_t, svint32_t, int32_t);
22198		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_x)))
22199		svint64_t svrhadd_x(svbool_t, svint64_t, int64_t);
22200		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_x)))
22201		svint16_t svrhadd_x(svbool_t, svint16_t, int16_t);
22202		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_z)))
22203		svint8_t svrhadd_z(svbool_t, svint8_t, int8_t);
22204		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_z)))
22205		svint32_t svrhadd_z(svbool_t, svint32_t, int32_t);
22206		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_z)))
22207		svint64_t svrhadd_z(svbool_t, svint64_t, int64_t);
22208		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_z)))
22209		svint16_t svrhadd_z(svbool_t, svint16_t, int16_t);
22210		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_m)))
22211		svuint8_t svrhadd_m(svbool_t, svuint8_t, uint8_t);
22212		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_m)))
22213		svuint32_t svrhadd_m(svbool_t, svuint32_t, uint32_t);
22214		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_m)))
22215		svuint64_t svrhadd_m(svbool_t, svuint64_t, uint64_t);
22216		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_m)))
22217		svuint16_t svrhadd_m(svbool_t, svuint16_t, uint16_t);
22218		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_x)))
22219		svuint8_t svrhadd_x(svbool_t, svuint8_t, uint8_t);
22220		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_x)))
22221		svuint32_t svrhadd_x(svbool_t, svuint32_t, uint32_t);
22222		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_x)))
22223		svuint64_t svrhadd_x(svbool_t, svuint64_t, uint64_t);
22224		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_x)))
22225		svuint16_t svrhadd_x(svbool_t, svuint16_t, uint16_t);
22226		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_z)))
22227		svuint8_t svrhadd_z(svbool_t, svuint8_t, uint8_t);
22228		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_z)))
22229		svuint32_t svrhadd_z(svbool_t, svuint32_t, uint32_t);
22230		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_z)))
22231		svuint64_t svrhadd_z(svbool_t, svuint64_t, uint64_t);
22232		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_z)))
22233		svuint16_t svrhadd_z(svbool_t, svuint16_t, uint16_t);
22234		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_m)))
22235		svint8_t svrhadd_m(svbool_t, svint8_t, svint8_t);
22236		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_m)))
22237		svint32_t svrhadd_m(svbool_t, svint32_t, svint32_t);
22238		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_m)))
22239		svint64_t svrhadd_m(svbool_t, svint64_t, svint64_t);
22240		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_m)))
22241		svint16_t svrhadd_m(svbool_t, svint16_t, svint16_t);
22242		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_x)))
22243		svint8_t svrhadd_x(svbool_t, svint8_t, svint8_t);
22244		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_x)))
22245		svint32_t svrhadd_x(svbool_t, svint32_t, svint32_t);
22246		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_x)))
22247		svint64_t svrhadd_x(svbool_t, svint64_t, svint64_t);
22248		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_x)))
22249		svint16_t svrhadd_x(svbool_t, svint16_t, svint16_t);
22250		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_z)))
22251		svint8_t svrhadd_z(svbool_t, svint8_t, svint8_t);
22252		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_z)))
22253		svint32_t svrhadd_z(svbool_t, svint32_t, svint32_t);
22254		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_z)))
22255		svint64_t svrhadd_z(svbool_t, svint64_t, svint64_t);
22256		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_z)))
22257		svint16_t svrhadd_z(svbool_t, svint16_t, svint16_t);
22258		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_m)))
22259		svuint8_t svrhadd_m(svbool_t, svuint8_t, svuint8_t);
22260		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_m)))
22261		svuint32_t svrhadd_m(svbool_t, svuint32_t, svuint32_t);
22262		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_m)))
22263		svuint64_t svrhadd_m(svbool_t, svuint64_t, svuint64_t);
22264		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_m)))
22265		svuint16_t svrhadd_m(svbool_t, svuint16_t, svuint16_t);
22266		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_x)))
22267		svuint8_t svrhadd_x(svbool_t, svuint8_t, svuint8_t);
22268		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_x)))
22269		svuint32_t svrhadd_x(svbool_t, svuint32_t, svuint32_t);
22270		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_x)))
22271		svuint64_t svrhadd_x(svbool_t, svuint64_t, svuint64_t);
22272		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_x)))
22273		svuint16_t svrhadd_x(svbool_t, svuint16_t, svuint16_t);
22274		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_z)))
22275		svuint8_t svrhadd_z(svbool_t, svuint8_t, svuint8_t);
22276		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_z)))
22277		svuint32_t svrhadd_z(svbool_t, svuint32_t, svuint32_t);
22278		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_z)))
22279		svuint64_t svrhadd_z(svbool_t, svuint64_t, svuint64_t);
22280		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_z)))
22281		svuint16_t svrhadd_z(svbool_t, svuint16_t, svuint16_t);
22282		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_m)))
22283		svint8_t svrshl_m(svbool_t, svint8_t, int8_t);
22284		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_m)))
22285		svint32_t svrshl_m(svbool_t, svint32_t, int32_t);
22286		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_m)))
22287		svint64_t svrshl_m(svbool_t, svint64_t, int64_t);
22288		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_m)))
22289		svint16_t svrshl_m(svbool_t, svint16_t, int16_t);
22290		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_x)))
22291		svint8_t svrshl_x(svbool_t, svint8_t, int8_t);
22292		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_x)))
22293		svint32_t svrshl_x(svbool_t, svint32_t, int32_t);
22294		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_x)))
22295		svint64_t svrshl_x(svbool_t, svint64_t, int64_t);
22296		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_x)))
22297		svint16_t svrshl_x(svbool_t, svint16_t, int16_t);
22298		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_z)))
22299		svint8_t svrshl_z(svbool_t, svint8_t, int8_t);
22300		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_z)))
22301		svint32_t svrshl_z(svbool_t, svint32_t, int32_t);
22302		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_z)))
22303		svint64_t svrshl_z(svbool_t, svint64_t, int64_t);
22304		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_z)))
22305		svint16_t svrshl_z(svbool_t, svint16_t, int16_t);
22306		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_m)))
22307		svuint8_t svrshl_m(svbool_t, svuint8_t, int8_t);
22308		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_m)))
22309		svuint32_t svrshl_m(svbool_t, svuint32_t, int32_t);
22310		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_m)))
22311		svuint64_t svrshl_m(svbool_t, svuint64_t, int64_t);
22312		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_m)))
22313		svuint16_t svrshl_m(svbool_t, svuint16_t, int16_t);
22314		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_x)))
22315		svuint8_t svrshl_x(svbool_t, svuint8_t, int8_t);
22316		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_x)))
22317		svuint32_t svrshl_x(svbool_t, svuint32_t, int32_t);
22318		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_x)))
22319		svuint64_t svrshl_x(svbool_t, svuint64_t, int64_t);
22320		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_x)))
22321		svuint16_t svrshl_x(svbool_t, svuint16_t, int16_t);
22322		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_z)))
22323		svuint8_t svrshl_z(svbool_t, svuint8_t, int8_t);
22324		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_z)))
22325		svuint32_t svrshl_z(svbool_t, svuint32_t, int32_t);
22326		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_z)))
22327		svuint64_t svrshl_z(svbool_t, svuint64_t, int64_t);
22328		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_z)))
22329		svuint16_t svrshl_z(svbool_t, svuint16_t, int16_t);
22330		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_m)))
22331		svint8_t svrshl_m(svbool_t, svint8_t, svint8_t);
22332		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_m)))
22333		svint32_t svrshl_m(svbool_t, svint32_t, svint32_t);
22334		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_m)))
22335		svint64_t svrshl_m(svbool_t, svint64_t, svint64_t);
22336		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_m)))
22337		svint16_t svrshl_m(svbool_t, svint16_t, svint16_t);
22338		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_x)))
22339		svint8_t svrshl_x(svbool_t, svint8_t, svint8_t);
22340		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_x)))
22341		svint32_t svrshl_x(svbool_t, svint32_t, svint32_t);
22342		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_x)))
22343		svint64_t svrshl_x(svbool_t, svint64_t, svint64_t);
22344		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_x)))
22345		svint16_t svrshl_x(svbool_t, svint16_t, svint16_t);
22346		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_z)))
22347		svint8_t svrshl_z(svbool_t, svint8_t, svint8_t);
22348		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_z)))
22349		svint32_t svrshl_z(svbool_t, svint32_t, svint32_t);
22350		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_z)))
22351		svint64_t svrshl_z(svbool_t, svint64_t, svint64_t);
22352		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_z)))
22353		svint16_t svrshl_z(svbool_t, svint16_t, svint16_t);
22354		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_m)))
22355		svuint8_t svrshl_m(svbool_t, svuint8_t, svint8_t);
22356		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_m)))
22357		svuint32_t svrshl_m(svbool_t, svuint32_t, svint32_t);
22358		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_m)))
22359		svuint64_t svrshl_m(svbool_t, svuint64_t, svint64_t);
22360		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_m)))
22361		svuint16_t svrshl_m(svbool_t, svuint16_t, svint16_t);
22362		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_x)))
22363		svuint8_t svrshl_x(svbool_t, svuint8_t, svint8_t);
22364		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_x)))
22365		svuint32_t svrshl_x(svbool_t, svuint32_t, svint32_t);
22366		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_x)))
22367		svuint64_t svrshl_x(svbool_t, svuint64_t, svint64_t);
22368		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_x)))
22369		svuint16_t svrshl_x(svbool_t, svuint16_t, svint16_t);
22370		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_z)))
22371		svuint8_t svrshl_z(svbool_t, svuint8_t, svint8_t);
22372		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_z)))
22373		svuint32_t svrshl_z(svbool_t, svuint32_t, svint32_t);
22374		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_z)))
22375		svuint64_t svrshl_z(svbool_t, svuint64_t, svint64_t);
22376		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_z)))
22377		svuint16_t svrshl_z(svbool_t, svuint16_t, svint16_t);
22378		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_m)))
22379		svint8_t svrshr_m(svbool_t, svint8_t, uint64_t);
22380		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_m)))
22381		svint32_t svrshr_m(svbool_t, svint32_t, uint64_t);
22382		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_m)))
22383		svint64_t svrshr_m(svbool_t, svint64_t, uint64_t);
22384		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_m)))
22385		svint16_t svrshr_m(svbool_t, svint16_t, uint64_t);
22386		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_m)))
22387		svuint8_t svrshr_m(svbool_t, svuint8_t, uint64_t);
22388		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_m)))
22389		svuint32_t svrshr_m(svbool_t, svuint32_t, uint64_t);
22390		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_m)))
22391		svuint64_t svrshr_m(svbool_t, svuint64_t, uint64_t);
22392		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_m)))
22393		svuint16_t svrshr_m(svbool_t, svuint16_t, uint64_t);
22394		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_x)))
22395		svint8_t svrshr_x(svbool_t, svint8_t, uint64_t);
22396		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_x)))
22397		svint32_t svrshr_x(svbool_t, svint32_t, uint64_t);
22398		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_x)))
22399		svint64_t svrshr_x(svbool_t, svint64_t, uint64_t);
22400		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_x)))
22401		svint16_t svrshr_x(svbool_t, svint16_t, uint64_t);
22402		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_x)))
22403		svuint8_t svrshr_x(svbool_t, svuint8_t, uint64_t);
22404		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_x)))
22405		svuint32_t svrshr_x(svbool_t, svuint32_t, uint64_t);
22406		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_x)))
22407		svuint64_t svrshr_x(svbool_t, svuint64_t, uint64_t);
22408		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_x)))
22409		svuint16_t svrshr_x(svbool_t, svuint16_t, uint64_t);
22410		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_z)))
22411		svint8_t svrshr_z(svbool_t, svint8_t, uint64_t);
22412		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_z)))
22413		svint32_t svrshr_z(svbool_t, svint32_t, uint64_t);
22414		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_z)))
22415		svint64_t svrshr_z(svbool_t, svint64_t, uint64_t);
22416		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_z)))
22417		svint16_t svrshr_z(svbool_t, svint16_t, uint64_t);
22418		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_z)))
22419		svuint8_t svrshr_z(svbool_t, svuint8_t, uint64_t);
22420		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_z)))
22421		svuint32_t svrshr_z(svbool_t, svuint32_t, uint64_t);
22422		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_z)))
22423		svuint64_t svrshr_z(svbool_t, svuint64_t, uint64_t);
22424		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_z)))
22425		svuint16_t svrshr_z(svbool_t, svuint16_t, uint64_t);
22426		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u32)))
22427		svuint16_t svrshrnb(svuint32_t, uint64_t);
22428		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u64)))
22429		svuint32_t svrshrnb(svuint64_t, uint64_t);
22430		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u16)))
22431		svuint8_t svrshrnb(svuint16_t, uint64_t);
22432		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s32)))
22433		svint16_t svrshrnb(svint32_t, uint64_t);
22434		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s64)))
22435		svint32_t svrshrnb(svint64_t, uint64_t);
22436		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s16)))
22437		svint8_t svrshrnb(svint16_t, uint64_t);
22438		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u32)))
22439		svuint16_t svrshrnt(svuint16_t, svuint32_t, uint64_t);
22440		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u64)))
22441		svuint32_t svrshrnt(svuint32_t, svuint64_t, uint64_t);
22442		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u16)))
22443		svuint8_t svrshrnt(svuint8_t, svuint16_t, uint64_t);
22444		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s32)))
22445		svint16_t svrshrnt(svint16_t, svint32_t, uint64_t);
22446		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s64)))
22447		svint32_t svrshrnt(svint32_t, svint64_t, uint64_t);
22448		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s16)))
22449		svint8_t svrshrnt(svint8_t, svint16_t, uint64_t);
22450		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_m)))
22451		svuint32_t svrsqrte_m(svuint32_t, svbool_t, svuint32_t);
22452		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_x)))
22453		svuint32_t svrsqrte_x(svbool_t, svuint32_t);
22454		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_z)))
22455		svuint32_t svrsqrte_z(svbool_t, svuint32_t);
22456		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s8)))
22457		svint8_t svrsra(svint8_t, svint8_t, uint64_t);
22458		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s32)))
22459		svint32_t svrsra(svint32_t, svint32_t, uint64_t);
22460		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s64)))
22461		svint64_t svrsra(svint64_t, svint64_t, uint64_t);
22462		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s16)))
22463		svint16_t svrsra(svint16_t, svint16_t, uint64_t);
22464		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u8)))
22465		svuint8_t svrsra(svuint8_t, svuint8_t, uint64_t);
22466		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u32)))
22467		svuint32_t svrsra(svuint32_t, svuint32_t, uint64_t);
22468		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u64)))
22469		svuint64_t svrsra(svuint64_t, svuint64_t, uint64_t);
22470		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u16)))
22471		svuint16_t svrsra(svuint16_t, svuint16_t, uint64_t);
22472		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u32)))
22473		svuint16_t svrsubhnb(svuint32_t, uint32_t);
22474		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u64)))
22475		svuint32_t svrsubhnb(svuint64_t, uint64_t);
22476		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u16)))
22477		svuint8_t svrsubhnb(svuint16_t, uint16_t);
22478		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s32)))
22479		svint16_t svrsubhnb(svint32_t, int32_t);
22480		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s64)))
22481		svint32_t svrsubhnb(svint64_t, int64_t);
22482		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s16)))
22483		svint8_t svrsubhnb(svint16_t, int16_t);
22484		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u32)))
22485		svuint16_t svrsubhnb(svuint32_t, svuint32_t);
22486		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u64)))
22487		svuint32_t svrsubhnb(svuint64_t, svuint64_t);
22488		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u16)))
22489		svuint8_t svrsubhnb(svuint16_t, svuint16_t);
22490		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s32)))
22491		svint16_t svrsubhnb(svint32_t, svint32_t);
22492		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s64)))
22493		svint32_t svrsubhnb(svint64_t, svint64_t);
22494		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s16)))
22495		svint8_t svrsubhnb(svint16_t, svint16_t);
22496		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u32)))
22497		svuint16_t svrsubhnt(svuint16_t, svuint32_t, uint32_t);
22498		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u64)))
22499		svuint32_t svrsubhnt(svuint32_t, svuint64_t, uint64_t);
22500		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u16)))
22501		svuint8_t svrsubhnt(svuint8_t, svuint16_t, uint16_t);
22502		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s32)))
22503		svint16_t svrsubhnt(svint16_t, svint32_t, int32_t);
22504		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s64)))
22505		svint32_t svrsubhnt(svint32_t, svint64_t, int64_t);
22506		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s16)))
22507		svint8_t svrsubhnt(svint8_t, svint16_t, int16_t);
22508		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u32)))
22509		svuint16_t svrsubhnt(svuint16_t, svuint32_t, svuint32_t);
22510		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u64)))
22511		svuint32_t svrsubhnt(svuint32_t, svuint64_t, svuint64_t);
22512		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u16)))
22513		svuint8_t svrsubhnt(svuint8_t, svuint16_t, svuint16_t);
22514		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s32)))
22515		svint16_t svrsubhnt(svint16_t, svint32_t, svint32_t);
22516		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s64)))
22517		svint32_t svrsubhnt(svint32_t, svint64_t, svint64_t);
22518		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s16)))
22519		svint8_t svrsubhnt(svint8_t, svint16_t, svint16_t);
22520		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u32)))
22521		svuint32_t svsbclb(svuint32_t, svuint32_t, uint32_t);
22522		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u64)))
22523		svuint64_t svsbclb(svuint64_t, svuint64_t, uint64_t);
22524		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u32)))
22525		svuint32_t svsbclb(svuint32_t, svuint32_t, svuint32_t);
22526		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u64)))
22527		svuint64_t svsbclb(svuint64_t, svuint64_t, svuint64_t);
22528		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u32)))
22529		svuint32_t svsbclt(svuint32_t, svuint32_t, uint32_t);
22530		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u64)))
22531		svuint64_t svsbclt(svuint64_t, svuint64_t, uint64_t);
22532		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u32)))
22533		svuint32_t svsbclt(svuint32_t, svuint32_t, svuint32_t);
22534		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u64)))
22535		svuint64_t svsbclt(svuint64_t, svuint64_t, svuint64_t);
22536		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s32)))
22537		svint32_t svshllb(svint16_t, uint64_t);
22538		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s64)))
22539		svint64_t svshllb(svint32_t, uint64_t);
22540		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s16)))
22541		svint16_t svshllb(svint8_t, uint64_t);
22542		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u32)))
22543		svuint32_t svshllb(svuint16_t, uint64_t);
22544		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u64)))
22545		svuint64_t svshllb(svuint32_t, uint64_t);
22546		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u16)))
22547		svuint16_t svshllb(svuint8_t, uint64_t);
22548		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s32)))
22549		svint32_t svshllt(svint16_t, uint64_t);
22550		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s64)))
22551		svint64_t svshllt(svint32_t, uint64_t);
22552		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s16)))
22553		svint16_t svshllt(svint8_t, uint64_t);
22554		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u32)))
22555		svuint32_t svshllt(svuint16_t, uint64_t);
22556		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u64)))
22557		svuint64_t svshllt(svuint32_t, uint64_t);
22558		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u16)))
22559		svuint16_t svshllt(svuint8_t, uint64_t);
22560		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u32)))
22561		svuint16_t svshrnb(svuint32_t, uint64_t);
22562		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u64)))
22563		svuint32_t svshrnb(svuint64_t, uint64_t);
22564		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u16)))
22565		svuint8_t svshrnb(svuint16_t, uint64_t);
22566		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s32)))
22567		svint16_t svshrnb(svint32_t, uint64_t);
22568		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s64)))
22569		svint32_t svshrnb(svint64_t, uint64_t);
22570		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s16)))
22571		svint8_t svshrnb(svint16_t, uint64_t);
22572		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u32)))
22573		svuint16_t svshrnt(svuint16_t, svuint32_t, uint64_t);
22574		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u64)))
22575		svuint32_t svshrnt(svuint32_t, svuint64_t, uint64_t);
22576		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u16)))
22577		svuint8_t svshrnt(svuint8_t, svuint16_t, uint64_t);
22578		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s32)))
22579		svint16_t svshrnt(svint16_t, svint32_t, uint64_t);
22580		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s64)))
22581		svint32_t svshrnt(svint32_t, svint64_t, uint64_t);
22582		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s16)))
22583		svint8_t svshrnt(svint8_t, svint16_t, uint64_t);
22584		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u8)))
22585		svuint8_t svsli(svuint8_t, svuint8_t, uint64_t);
22586		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u32)))
22587		svuint32_t svsli(svuint32_t, svuint32_t, uint64_t);
22588		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u64)))
22589		svuint64_t svsli(svuint64_t, svuint64_t, uint64_t);
22590		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u16)))
22591		svuint16_t svsli(svuint16_t, svuint16_t, uint64_t);
22592		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s8)))
22593		svint8_t svsli(svint8_t, svint8_t, uint64_t);
22594		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s32)))
22595		svint32_t svsli(svint32_t, svint32_t, uint64_t);
22596		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s64)))
22597		svint64_t svsli(svint64_t, svint64_t, uint64_t);
22598		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s16)))
22599		svint16_t svsli(svint16_t, svint16_t, uint64_t);
22600		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_m)))
22601		svuint8_t svsqadd_m(svbool_t, svuint8_t, int8_t);
22602		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_m)))
22603		svuint32_t svsqadd_m(svbool_t, svuint32_t, int32_t);
22604		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_m)))
22605		svuint64_t svsqadd_m(svbool_t, svuint64_t, int64_t);
22606		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_m)))
22607		svuint16_t svsqadd_m(svbool_t, svuint16_t, int16_t);
22608		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_x)))
22609		svuint8_t svsqadd_x(svbool_t, svuint8_t, int8_t);
22610		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_x)))
22611		svuint32_t svsqadd_x(svbool_t, svuint32_t, int32_t);
22612		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_x)))
22613		svuint64_t svsqadd_x(svbool_t, svuint64_t, int64_t);
22614		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_x)))
22615		svuint16_t svsqadd_x(svbool_t, svuint16_t, int16_t);
22616		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_z)))
22617		svuint8_t svsqadd_z(svbool_t, svuint8_t, int8_t);
22618		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_z)))
22619		svuint32_t svsqadd_z(svbool_t, svuint32_t, int32_t);
22620		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_z)))
22621		svuint64_t svsqadd_z(svbool_t, svuint64_t, int64_t);
22622		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_z)))
22623		svuint16_t svsqadd_z(svbool_t, svuint16_t, int16_t);
22624		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_m)))
22625		svuint8_t svsqadd_m(svbool_t, svuint8_t, svint8_t);
22626		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_m)))
22627		svuint32_t svsqadd_m(svbool_t, svuint32_t, svint32_t);
22628		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_m)))
22629		svuint64_t svsqadd_m(svbool_t, svuint64_t, svint64_t);
22630		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_m)))
22631		svuint16_t svsqadd_m(svbool_t, svuint16_t, svint16_t);
22632		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_x)))
22633		svuint8_t svsqadd_x(svbool_t, svuint8_t, svint8_t);
22634		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_x)))
22635		svuint32_t svsqadd_x(svbool_t, svuint32_t, svint32_t);
22636		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_x)))
22637		svuint64_t svsqadd_x(svbool_t, svuint64_t, svint64_t);
22638		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_x)))
22639		svuint16_t svsqadd_x(svbool_t, svuint16_t, svint16_t);
22640		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_z)))
22641		svuint8_t svsqadd_z(svbool_t, svuint8_t, svint8_t);
22642		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_z)))
22643		svuint32_t svsqadd_z(svbool_t, svuint32_t, svint32_t);
22644		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_z)))
22645		svuint64_t svsqadd_z(svbool_t, svuint64_t, svint64_t);
22646		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_z)))
22647		svuint16_t svsqadd_z(svbool_t, svuint16_t, svint16_t);
22648		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s8)))
22649		svint8_t svsra(svint8_t, svint8_t, uint64_t);
22650		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s32)))
22651		svint32_t svsra(svint32_t, svint32_t, uint64_t);
22652		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s64)))
22653		svint64_t svsra(svint64_t, svint64_t, uint64_t);
22654		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s16)))
22655		svint16_t svsra(svint16_t, svint16_t, uint64_t);
22656		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u8)))
22657		svuint8_t svsra(svuint8_t, svuint8_t, uint64_t);
22658		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u32)))
22659		svuint32_t svsra(svuint32_t, svuint32_t, uint64_t);
22660		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u64)))
22661		svuint64_t svsra(svuint64_t, svuint64_t, uint64_t);
22662		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u16)))
22663		svuint16_t svsra(svuint16_t, svuint16_t, uint64_t);
22664		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u8)))
22665		svuint8_t svsri(svuint8_t, svuint8_t, uint64_t);
22666		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u32)))
22667		svuint32_t svsri(svuint32_t, svuint32_t, uint64_t);
22668		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u64)))
22669		svuint64_t svsri(svuint64_t, svuint64_t, uint64_t);
22670		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u16)))
22671		svuint16_t svsri(svuint16_t, svuint16_t, uint64_t);
22672		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s8)))
22673		svint8_t svsri(svint8_t, svint8_t, uint64_t);
22674		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s32)))
22675		svint32_t svsri(svint32_t, svint32_t, uint64_t);
22676		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s64)))
22677		svint64_t svsri(svint64_t, svint64_t, uint64_t);
22678		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s16)))
22679		svint16_t svsri(svint16_t, svint16_t, uint64_t);
22680		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_u32)))
22681		void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
22682		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_u64)))
22683		void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
22684		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_f64)))
22685		void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svfloat64_t);
22686		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_f32)))
22687		void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svfloat32_t);
22688		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_s32)))
22689		void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
22690		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_s64)))
22691		void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
22692		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_u32)))
22693		void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
22694		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_u64)))
22695		void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
22696		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_f64)))
22697		void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svfloat64_t);
22698		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_f32)))
22699		void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svfloat32_t);
22700		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_s32)))
22701		void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
22702		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_s64)))
22703		void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
22704		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_u32)))
22705		void svstnt1_scatter(svbool_t, svuint32_t, svuint32_t);
22706		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_u64)))
22707		void svstnt1_scatter(svbool_t, svuint64_t, svuint64_t);
22708		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_f64)))
22709		void svstnt1_scatter(svbool_t, svuint64_t, svfloat64_t);
22710		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_f32)))
22711		void svstnt1_scatter(svbool_t, svuint32_t, svfloat32_t);
22712		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_s32)))
22713		void svstnt1_scatter(svbool_t, svuint32_t, svint32_t);
22714		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_s64)))
22715		void svstnt1_scatter(svbool_t, svuint64_t, svint64_t);
22716		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_u64)))
22717		void svstnt1_scatter_index(svbool_t, uint64_t *, svint64_t, svuint64_t);
22718		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_f64)))
22719		void svstnt1_scatter_index(svbool_t, float64_t *, svint64_t, svfloat64_t);
22720		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_s64)))
22721		void svstnt1_scatter_index(svbool_t, int64_t *, svint64_t, svint64_t);
22722		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_u64)))
22723		void svstnt1_scatter_index(svbool_t, uint64_t *, svuint64_t, svuint64_t);
22724		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_f64)))
22725		void svstnt1_scatter_index(svbool_t, float64_t *, svuint64_t, svfloat64_t);
22726		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_s64)))
22727		void svstnt1_scatter_index(svbool_t, int64_t *, svuint64_t, svint64_t);
22728		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_u32)))
22729		void svstnt1_scatter_offset(svbool_t, uint32_t *, svuint32_t, svuint32_t);
22730		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_f32)))
22731		void svstnt1_scatter_offset(svbool_t, float32_t *, svuint32_t, svfloat32_t);
22732		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_s32)))
22733		void svstnt1_scatter_offset(svbool_t, int32_t *, svuint32_t, svint32_t);
22734		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_u64)))
22735		void svstnt1_scatter_offset(svbool_t, uint64_t *, svint64_t, svuint64_t);
22736		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_f64)))
22737		void svstnt1_scatter_offset(svbool_t, float64_t *, svint64_t, svfloat64_t);
22738		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_s64)))
22739		void svstnt1_scatter_offset(svbool_t, int64_t *, svint64_t, svint64_t);
22740		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_u64)))
22741		void svstnt1_scatter_offset(svbool_t, uint64_t *, svuint64_t, svuint64_t);
22742		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_f64)))
22743		void svstnt1_scatter_offset(svbool_t, float64_t *, svuint64_t, svfloat64_t);
22744		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_s64)))
22745		void svstnt1_scatter_offset(svbool_t, int64_t *, svuint64_t, svint64_t);
22746		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_u32)))
22747		void svstnt1b_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
22748		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_u64)))
22749		void svstnt1b_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
22750		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_s32)))
22751		void svstnt1b_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
22752		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_s64)))
22753		void svstnt1b_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
22754		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_u32)))
22755		void svstnt1b_scatter(svbool_t, svuint32_t, svuint32_t);
22756		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_u64)))
22757		void svstnt1b_scatter(svbool_t, svuint64_t, svuint64_t);
22758		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_s32)))
22759		void svstnt1b_scatter(svbool_t, svuint32_t, svint32_t);
22760		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_s64)))
22761		void svstnt1b_scatter(svbool_t, svuint64_t, svint64_t);
22762		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_s32)))
22763		void svstnt1b_scatter_offset(svbool_t, int8_t *, svuint32_t, svint32_t);
22764		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_u32)))
22765		void svstnt1b_scatter_offset(svbool_t, uint8_t *, svuint32_t, svuint32_t);
22766		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_s64)))
22767		void svstnt1b_scatter_offset(svbool_t, int8_t *, svint64_t, svint64_t);
22768		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_u64)))
22769		void svstnt1b_scatter_offset(svbool_t, uint8_t *, svint64_t, svuint64_t);
22770		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_s64)))
22771		void svstnt1b_scatter_offset(svbool_t, int8_t *, svuint64_t, svint64_t);
22772		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_u64)))
22773		void svstnt1b_scatter_offset(svbool_t, uint8_t *, svuint64_t, svuint64_t);
22774		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_u32)))
22775		void svstnt1h_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
22776		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_u64)))
22777		void svstnt1h_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
22778		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_s32)))
22779		void svstnt1h_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
22780		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_s64)))
22781		void svstnt1h_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
22782		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_u32)))
22783		void svstnt1h_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
22784		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_u64)))
22785		void svstnt1h_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
22786		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_s32)))
22787		void svstnt1h_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
22788		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_s64)))
22789		void svstnt1h_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
22790		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_u32)))
22791		void svstnt1h_scatter(svbool_t, svuint32_t, svuint32_t);
22792		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_u64)))
22793		void svstnt1h_scatter(svbool_t, svuint64_t, svuint64_t);
22794		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_s32)))
22795		void svstnt1h_scatter(svbool_t, svuint32_t, svint32_t);
22796		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_s64)))
22797		void svstnt1h_scatter(svbool_t, svuint64_t, svint64_t);
22798		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_s64)))
22799		void svstnt1h_scatter_index(svbool_t, int16_t *, svint64_t, svint64_t);
22800		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_u64)))
22801		void svstnt1h_scatter_index(svbool_t, uint16_t *, svint64_t, svuint64_t);
22802		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_s64)))
22803		void svstnt1h_scatter_index(svbool_t, int16_t *, svuint64_t, svint64_t);
22804		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_u64)))
22805		void svstnt1h_scatter_index(svbool_t, uint16_t *, svuint64_t, svuint64_t);
22806		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_s32)))
22807		void svstnt1h_scatter_offset(svbool_t, int16_t *, svuint32_t, svint32_t);
22808		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_u32)))
22809		void svstnt1h_scatter_offset(svbool_t, uint16_t *, svuint32_t, svuint32_t);
22810		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_s64)))
22811		void svstnt1h_scatter_offset(svbool_t, int16_t *, svint64_t, svint64_t);
22812		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_u64)))
22813		void svstnt1h_scatter_offset(svbool_t, uint16_t *, svint64_t, svuint64_t);
22814		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_s64)))
22815		void svstnt1h_scatter_offset(svbool_t, int16_t *, svuint64_t, svint64_t);
22816		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_u64)))
22817		void svstnt1h_scatter_offset(svbool_t, uint16_t *, svuint64_t, svuint64_t);
22818		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_u64)))
22819		void svstnt1w_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
22820		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_s64)))
22821		void svstnt1w_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
22822		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_u64)))
22823		void svstnt1w_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
22824		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_s64)))
22825		void svstnt1w_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
22826		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_u64)))
22827		void svstnt1w_scatter(svbool_t, svuint64_t, svuint64_t);
22828		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_s64)))
22829		void svstnt1w_scatter(svbool_t, svuint64_t, svint64_t);
22830		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_s64)))
22831		void svstnt1w_scatter_index(svbool_t, int32_t *, svint64_t, svint64_t);
22832		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_u64)))
22833		void svstnt1w_scatter_index(svbool_t, uint32_t *, svint64_t, svuint64_t);
22834		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_s64)))
22835		void svstnt1w_scatter_index(svbool_t, int32_t *, svuint64_t, svint64_t);
22836		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_u64)))
22837		void svstnt1w_scatter_index(svbool_t, uint32_t *, svuint64_t, svuint64_t);
22838		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_s64)))
22839		void svstnt1w_scatter_offset(svbool_t, int32_t *, svint64_t, svint64_t);
22840		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_u64)))
22841		void svstnt1w_scatter_offset(svbool_t, uint32_t *, svint64_t, svuint64_t);
22842		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_s64)))
22843		void svstnt1w_scatter_offset(svbool_t, int32_t *, svuint64_t, svint64_t);
22844		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_u64)))
22845		void svstnt1w_scatter_offset(svbool_t, uint32_t *, svuint64_t, svuint64_t);
22846		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u32)))
22847		svuint16_t svsubhnb(svuint32_t, uint32_t);
22848		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u64)))
22849		svuint32_t svsubhnb(svuint64_t, uint64_t);
22850		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u16)))
22851		svuint8_t svsubhnb(svuint16_t, uint16_t);
22852		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s32)))
22853		svint16_t svsubhnb(svint32_t, int32_t);
22854		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s64)))
22855		svint32_t svsubhnb(svint64_t, int64_t);
22856		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s16)))
22857		svint8_t svsubhnb(svint16_t, int16_t);
22858		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u32)))
22859		svuint16_t svsubhnb(svuint32_t, svuint32_t);
22860		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u64)))
22861		svuint32_t svsubhnb(svuint64_t, svuint64_t);
22862		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u16)))
22863		svuint8_t svsubhnb(svuint16_t, svuint16_t);
22864		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s32)))
22865		svint16_t svsubhnb(svint32_t, svint32_t);
22866		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s64)))
22867		svint32_t svsubhnb(svint64_t, svint64_t);
22868		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s16)))
22869		svint8_t svsubhnb(svint16_t, svint16_t);
22870		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u32)))
22871		svuint16_t svsubhnt(svuint16_t, svuint32_t, uint32_t);
22872		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u64)))
22873		svuint32_t svsubhnt(svuint32_t, svuint64_t, uint64_t);
22874		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u16)))
22875		svuint8_t svsubhnt(svuint8_t, svuint16_t, uint16_t);
22876		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s32)))
22877		svint16_t svsubhnt(svint16_t, svint32_t, int32_t);
22878		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s64)))
22879		svint32_t svsubhnt(svint32_t, svint64_t, int64_t);
22880		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s16)))
22881		svint8_t svsubhnt(svint8_t, svint16_t, int16_t);
22882		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u32)))
22883		svuint16_t svsubhnt(svuint16_t, svuint32_t, svuint32_t);
22884		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u64)))
22885		svuint32_t svsubhnt(svuint32_t, svuint64_t, svuint64_t);
22886		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u16)))
22887		svuint8_t svsubhnt(svuint8_t, svuint16_t, svuint16_t);
22888		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s32)))
22889		svint16_t svsubhnt(svint16_t, svint32_t, svint32_t);
22890		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s64)))
22891		svint32_t svsubhnt(svint32_t, svint64_t, svint64_t);
22892		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s16)))
22893		svint8_t svsubhnt(svint8_t, svint16_t, svint16_t);
22894		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s32)))
22895		svint32_t svsublb(svint16_t, int16_t);
22896		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s64)))
22897		svint64_t svsublb(svint32_t, int32_t);
22898		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s16)))
22899		svint16_t svsublb(svint8_t, int8_t);
22900		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u32)))
22901		svuint32_t svsublb(svuint16_t, uint16_t);
22902		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u64)))
22903		svuint64_t svsublb(svuint32_t, uint32_t);
22904		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u16)))
22905		svuint16_t svsublb(svuint8_t, uint8_t);
22906		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s32)))
22907		svint32_t svsublb(svint16_t, svint16_t);
22908		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s64)))
22909		svint64_t svsublb(svint32_t, svint32_t);
22910		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s16)))
22911		svint16_t svsublb(svint8_t, svint8_t);
22912		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u32)))
22913		svuint32_t svsublb(svuint16_t, svuint16_t);
22914		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u64)))
22915		svuint64_t svsublb(svuint32_t, svuint32_t);
22916		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u16)))
22917		svuint16_t svsublb(svuint8_t, svuint8_t);
22918		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s32)))
22919		svint32_t svsublbt(svint16_t, int16_t);
22920		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s64)))
22921		svint64_t svsublbt(svint32_t, int32_t);
22922		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s16)))
22923		svint16_t svsublbt(svint8_t, int8_t);
22924		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s32)))
22925		svint32_t svsublbt(svint16_t, svint16_t);
22926		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s64)))
22927		svint64_t svsublbt(svint32_t, svint32_t);
22928		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s16)))
22929		svint16_t svsublbt(svint8_t, svint8_t);
22930		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s32)))
22931		svint32_t svsublt(svint16_t, int16_t);
22932		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s64)))
22933		svint64_t svsublt(svint32_t, int32_t);
22934		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s16)))
22935		svint16_t svsublt(svint8_t, int8_t);
22936		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u32)))
22937		svuint32_t svsublt(svuint16_t, uint16_t);
22938		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u64)))
22939		svuint64_t svsublt(svuint32_t, uint32_t);
22940		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u16)))
22941		svuint16_t svsublt(svuint8_t, uint8_t);
22942		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s32)))
22943		svint32_t svsublt(svint16_t, svint16_t);
22944		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s64)))
22945		svint64_t svsublt(svint32_t, svint32_t);
22946		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s16)))
22947		svint16_t svsublt(svint8_t, svint8_t);
22948		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u32)))
22949		svuint32_t svsublt(svuint16_t, svuint16_t);
22950		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u64)))
22951		svuint64_t svsublt(svuint32_t, svuint32_t);
22952		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u16)))
22953		svuint16_t svsublt(svuint8_t, svuint8_t);
22954		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s32)))
22955		svint32_t svsubltb(svint16_t, int16_t);
22956		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s64)))
22957		svint64_t svsubltb(svint32_t, int32_t);
22958		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s16)))
22959		svint16_t svsubltb(svint8_t, int8_t);
22960		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s32)))
22961		svint32_t svsubltb(svint16_t, svint16_t);
22962		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s64)))
22963		svint64_t svsubltb(svint32_t, svint32_t);
22964		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s16)))
22965		svint16_t svsubltb(svint8_t, svint8_t);
22966		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s32)))
22967		svint32_t svsubwb(svint32_t, int16_t);
22968		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s64)))
22969		svint64_t svsubwb(svint64_t, int32_t);
22970		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s16)))
22971		svint16_t svsubwb(svint16_t, int8_t);
22972		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u32)))
22973		svuint32_t svsubwb(svuint32_t, uint16_t);
22974		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u64)))
22975		svuint64_t svsubwb(svuint64_t, uint32_t);
22976		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u16)))
22977		svuint16_t svsubwb(svuint16_t, uint8_t);
22978		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s32)))
22979		svint32_t svsubwb(svint32_t, svint16_t);
22980		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s64)))
22981		svint64_t svsubwb(svint64_t, svint32_t);
22982		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s16)))
22983		svint16_t svsubwb(svint16_t, svint8_t);
22984		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u32)))
22985		svuint32_t svsubwb(svuint32_t, svuint16_t);
22986		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u64)))
22987		svuint64_t svsubwb(svuint64_t, svuint32_t);
22988		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u16)))
22989		svuint16_t svsubwb(svuint16_t, svuint8_t);
22990		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s32)))
22991		svint32_t svsubwt(svint32_t, int16_t);
22992		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s64)))
22993		svint64_t svsubwt(svint64_t, int32_t);
22994		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s16)))
22995		svint16_t svsubwt(svint16_t, int8_t);
22996		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u32)))
22997		svuint32_t svsubwt(svuint32_t, uint16_t);
22998		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u64)))
22999		svuint64_t svsubwt(svuint64_t, uint32_t);
23000		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u16)))
23001		svuint16_t svsubwt(svuint16_t, uint8_t);
23002		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s32)))
23003		svint32_t svsubwt(svint32_t, svint16_t);
23004		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s64)))
23005		svint64_t svsubwt(svint64_t, svint32_t);
23006		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s16)))
23007		svint16_t svsubwt(svint16_t, svint8_t);
23008		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u32)))
23009		svuint32_t svsubwt(svuint32_t, svuint16_t);
23010		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u64)))
23011		svuint64_t svsubwt(svuint64_t, svuint32_t);
23012		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u16)))
23013		svuint16_t svsubwt(svuint16_t, svuint8_t);
23014		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u8)))
23015		svuint8_t svtbl2(svuint8x2_t, svuint8_t);
23016		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u32)))
23017		svuint32_t svtbl2(svuint32x2_t, svuint32_t);
23018		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
23019		svuint64_t svtbl2(svuint64x2_t, svuint64_t);
23020		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
23021		svuint16_t svtbl2(svuint16x2_t, svuint16_t);
23022		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
23023		svint8_t svtbl2(svint8x2_t, svuint8_t);
23024		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
23025		svfloat64_t svtbl2(svfloat64x2_t, svuint64_t);
23026		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f32)))
23027		svfloat32_t svtbl2(svfloat32x2_t, svuint32_t);
23028		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f16)))
23029		svfloat16_t svtbl2(svfloat16x2_t, svuint16_t);
23030		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s32)))
23031		svint32_t svtbl2(svint32x2_t, svuint32_t);
23032		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s64)))
23033		svint64_t svtbl2(svint64x2_t, svuint64_t);
23034		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s16)))
23035		svint16_t svtbl2(svint16x2_t, svuint16_t);
23036		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u8)))
23037		svuint8_t svtbx(svuint8_t, svuint8_t, svuint8_t);
23038		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u32)))
23039		svuint32_t svtbx(svuint32_t, svuint32_t, svuint32_t);
23040		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
23041		svuint64_t svtbx(svuint64_t, svuint64_t, svuint64_t);
23042		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
23043		svuint16_t svtbx(svuint16_t, svuint16_t, svuint16_t);
23044		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
23045		svint8_t svtbx(svint8_t, svint8_t, svuint8_t);
23046		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
23047		svfloat64_t svtbx(svfloat64_t, svfloat64_t, svuint64_t);
23048		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f32)))
23049		svfloat32_t svtbx(svfloat32_t, svfloat32_t, svuint32_t);
23050		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f16)))
23051		svfloat16_t svtbx(svfloat16_t, svfloat16_t, svuint16_t);
23052		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s32)))
23053		svint32_t svtbx(svint32_t, svint32_t, svuint32_t);
23054		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s64)))
23055		svint64_t svtbx(svint64_t, svint64_t, svuint64_t);
23056		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s16)))
23057		svint16_t svtbx(svint16_t, svint16_t, svuint16_t);
23058		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_m)))
23059		svint8_t svuqadd_m(svbool_t, svint8_t, uint8_t);
23060		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_m)))
23061		svint32_t svuqadd_m(svbool_t, svint32_t, uint32_t);
23062		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_m)))
23063		svint64_t svuqadd_m(svbool_t, svint64_t, uint64_t);
23064		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_m)))
23065		svint16_t svuqadd_m(svbool_t, svint16_t, uint16_t);
23066		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_x)))
23067		svint8_t svuqadd_x(svbool_t, svint8_t, uint8_t);
23068		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_x)))
23069		svint32_t svuqadd_x(svbool_t, svint32_t, uint32_t);
23070		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_x)))
23071		svint64_t svuqadd_x(svbool_t, svint64_t, uint64_t);
23072		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_x)))
23073		svint16_t svuqadd_x(svbool_t, svint16_t, uint16_t);
23074		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_z)))
23075		svint8_t svuqadd_z(svbool_t, svint8_t, uint8_t);
23076		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_z)))
23077		svint32_t svuqadd_z(svbool_t, svint32_t, uint32_t);
23078		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_z)))
23079		svint64_t svuqadd_z(svbool_t, svint64_t, uint64_t);
23080		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_z)))
23081		svint16_t svuqadd_z(svbool_t, svint16_t, uint16_t);
23082		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_m)))
23083		svint8_t svuqadd_m(svbool_t, svint8_t, svuint8_t);
23084		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_m)))
23085		svint32_t svuqadd_m(svbool_t, svint32_t, svuint32_t);
23086		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_m)))
23087		svint64_t svuqadd_m(svbool_t, svint64_t, svuint64_t);
23088		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_m)))
23089		svint16_t svuqadd_m(svbool_t, svint16_t, svuint16_t);
23090		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_x)))
23091		svint8_t svuqadd_x(svbool_t, svint8_t, svuint8_t);
23092		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_x)))
23093		svint32_t svuqadd_x(svbool_t, svint32_t, svuint32_t);
23094		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_x)))
23095		svint64_t svuqadd_x(svbool_t, svint64_t, svuint64_t);
23096		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_x)))
23097		svint16_t svuqadd_x(svbool_t, svint16_t, svuint16_t);
23098		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_z)))
23099		svint8_t svuqadd_z(svbool_t, svint8_t, svuint8_t);
23100		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_z)))
23101		svint32_t svuqadd_z(svbool_t, svint32_t, svuint32_t);
23102		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_z)))
23103		svint64_t svuqadd_z(svbool_t, svint64_t, svuint64_t);
23104		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_z)))
23105		svint16_t svuqadd_z(svbool_t, svint16_t, svuint16_t);
23106		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s32)))
23107		svbool_t svwhilege_b8(int32_t, int32_t);
23108		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s32)))
23109		svbool_t svwhilege_b32(int32_t, int32_t);
23110		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s32)))
23111		svbool_t svwhilege_b64(int32_t, int32_t);
23112		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s32)))
23113		svbool_t svwhilege_b16(int32_t, int32_t);
23114		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64)))
23115		svbool_t svwhilege_b8(int64_t, int64_t);
23116		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64)))
23117		svbool_t svwhilege_b32(int64_t, int64_t);
23118		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64)))
23119		svbool_t svwhilege_b64(int64_t, int64_t);
23120		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64)))
23121		svbool_t svwhilege_b16(int64_t, int64_t);
23122		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u32)))
23123		svbool_t svwhilege_b8(uint32_t, uint32_t);
23124		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u32)))
23125		svbool_t svwhilege_b32(uint32_t, uint32_t);
23126		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u32)))
23127		svbool_t svwhilege_b64(uint32_t, uint32_t);
23128		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u32)))
23129		svbool_t svwhilege_b16(uint32_t, uint32_t);
23130		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64)))
23131		svbool_t svwhilege_b8(uint64_t, uint64_t);
23132		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64)))
23133		svbool_t svwhilege_b32(uint64_t, uint64_t);
23134		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64)))
23135		svbool_t svwhilege_b64(uint64_t, uint64_t);
23136		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64)))
23137		svbool_t svwhilege_b16(uint64_t, uint64_t);
23138		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s32)))
23139		svbool_t svwhilegt_b8(int32_t, int32_t);
23140		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s32)))
23141		svbool_t svwhilegt_b32(int32_t, int32_t);
23142		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s32)))
23143		svbool_t svwhilegt_b64(int32_t, int32_t);
23144		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s32)))
23145		svbool_t svwhilegt_b16(int32_t, int32_t);
23146		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64)))
23147		svbool_t svwhilegt_b8(int64_t, int64_t);
23148		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64)))
23149		svbool_t svwhilegt_b32(int64_t, int64_t);
23150		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64)))
23151		svbool_t svwhilegt_b64(int64_t, int64_t);
23152		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64)))
23153		svbool_t svwhilegt_b16(int64_t, int64_t);
23154		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u32)))
23155		svbool_t svwhilegt_b8(uint32_t, uint32_t);
23156		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u32)))
23157		svbool_t svwhilegt_b32(uint32_t, uint32_t);
23158		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u32)))
23159		svbool_t svwhilegt_b64(uint32_t, uint32_t);
23160		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u32)))
23161		svbool_t svwhilegt_b16(uint32_t, uint32_t);
23162		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64)))
23163		svbool_t svwhilegt_b8(uint64_t, uint64_t);
23164		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64)))
23165		svbool_t svwhilegt_b32(uint64_t, uint64_t);
23166		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64)))
23167		svbool_t svwhilegt_b64(uint64_t, uint64_t);
23168		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64)))
23169		svbool_t svwhilegt_b16(uint64_t, uint64_t);
23170		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u8)))
23171		svbool_t svwhilerw(uint8_t const , uint8_t const );
23172		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s8)))
23173		svbool_t svwhilerw(int8_t const , int8_t const );
23174		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u64)))
23175		svbool_t svwhilerw(uint64_t const , uint64_t const );
23176		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f64)))
23177		svbool_t svwhilerw(float64_t const , float64_t const );
23178		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
23179		svbool_t svwhilerw(int64_t const , int64_t const );
23180		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
23181		svbool_t svwhilerw(uint16_t const , uint16_t const );
23182		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
23183		svbool_t svwhilerw(float16_t const , float16_t const );
23184		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
23185		svbool_t svwhilerw(int16_t const , int16_t const );
23186		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u32)))
23187		svbool_t svwhilerw(uint32_t const , uint32_t const );
23188		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f32)))
23189		svbool_t svwhilerw(float32_t const , float32_t const );
23190		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s32)))
23191		svbool_t svwhilerw(int32_t const , int32_t const );
23192		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u8)))
23193		svbool_t svwhilewr(uint8_t const , uint8_t const );
23194		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s8)))
23195		svbool_t svwhilewr(int8_t const , int8_t const );
23196		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u64)))
23197		svbool_t svwhilewr(uint64_t const , uint64_t const );
23198		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f64)))
23199		svbool_t svwhilewr(float64_t const , float64_t const );
23200		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
23201		svbool_t svwhilewr(int64_t const , int64_t const );
23202		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
23203		svbool_t svwhilewr(uint16_t const , uint16_t const );
23204		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
23205		svbool_t svwhilewr(float16_t const , float16_t const );
23206		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
23207		svbool_t svwhilewr(int16_t const , int16_t const );
23208		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u32)))
23209		svbool_t svwhilewr(uint32_t const , uint32_t const );
23210		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f32)))
23211		svbool_t svwhilewr(float32_t const , float32_t const );
23212		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s32)))
23213		svbool_t svwhilewr(int32_t const , int32_t const );
23214		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u8)))
23215		svuint8_t svxar(svuint8_t, svuint8_t, uint64_t);
23216		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u32)))
23217		svuint32_t svxar(svuint32_t, svuint32_t, uint64_t);
23218		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u64)))
23219		svuint64_t svxar(svuint64_t, svuint64_t, uint64_t);
23220		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u16)))
23221		svuint16_t svxar(svuint16_t, svuint16_t, uint64_t);
23222		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s8)))
23223		svint8_t svxar(svint8_t, svint8_t, uint64_t);
23224		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s32)))
23225		svint32_t svxar(svint32_t, svint32_t, uint64_t);
23226		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s64)))
23227		svint64_t svxar(svint64_t, svint64_t, uint64_t);
23228		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s16)))
23229		svint16_t svxar(svint16_t, svint16_t, uint64_t);
23230		#endif //defined(__ARM_FEATURE_SVE2)
23231
23232		#if defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
23233		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
23234		svbool_t svwhilerw_bf16(bfloat16_t const , bfloat16_t const );
23235		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
23236		svbool_t svwhilewr_bf16(bfloat16_t const , bfloat16_t const );
23237		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
23238		svbool_t svwhilerw(bfloat16_t const , bfloat16_t const );
23239		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
23240		svbool_t svwhilewr(bfloat16_t const , bfloat16_t const );
23241		#endif //defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_BF16_SCALAR_ARITHMETIC)
23242
23243		#if defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_SVE_BF16)
23244		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
23245		svbfloat16_t svtbl2_bf16(svbfloat16x2_t, svuint16_t);
23246		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
23247		svbfloat16_t svtbx_bf16(svbfloat16_t, svbfloat16_t, svuint16_t);
23248		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
23249		svbfloat16_t svtbl2(svbfloat16x2_t, svuint16_t);
23250		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
23251		svbfloat16_t svtbx(svbfloat16_t, svbfloat16_t, svuint16_t);
23252		#endif //defined(__ARM_FEATURE_SVE2) && defined(__ARM_FEATURE_SVE_BF16)
23253
23254		#if defined(__ARM_FEATURE_SVE2_AES)
23255		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesd_u8)))
23256		svuint8_t svaesd_u8(svuint8_t, svuint8_t);
23257		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaese_u8)))
23258		svuint8_t svaese_u8(svuint8_t, svuint8_t);
23259		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesimc_u8)))
23260		svuint8_t svaesimc_u8(svuint8_t);
23261		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesmc_u8)))
23262		svuint8_t svaesmc_u8(svuint8_t);
23263		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u64)))
23264		svuint64_t svpmullb_pair_n_u64(svuint64_t, uint64_t);
23265		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u64)))
23266		svuint64_t svpmullb_pair_u64(svuint64_t, svuint64_t);
23267		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u64)))
23268		svuint64_t svpmullt_pair_n_u64(svuint64_t, uint64_t);
23269		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u64)))
23270		svuint64_t svpmullt_pair_u64(svuint64_t, svuint64_t);
23271		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesd_u8)))
23272		svuint8_t svaesd(svuint8_t, svuint8_t);
23273		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaese_u8)))
23274		svuint8_t svaese(svuint8_t, svuint8_t);
23275		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesimc_u8)))
23276		svuint8_t svaesimc(svuint8_t);
23277		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesmc_u8)))
23278		svuint8_t svaesmc(svuint8_t);
23279		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u64)))
23280		svuint64_t svpmullb_pair(svuint64_t, uint64_t);
23281		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u64)))
23282		svuint64_t svpmullb_pair(svuint64_t, svuint64_t);
23283		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u64)))
23284		svuint64_t svpmullt_pair(svuint64_t, uint64_t);
23285		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u64)))
23286		svuint64_t svpmullt_pair(svuint64_t, svuint64_t);
23287		#endif //defined(__ARM_FEATURE_SVE2_AES)
23288
23289		#if defined(__ARM_FEATURE_SVE2_SHA3)
23290		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
23291		svuint64_t svrax1_u64(svuint64_t, svuint64_t);
23292		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
23293		svint64_t svrax1_s64(svint64_t, svint64_t);
23294		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
23295		svuint64_t svrax1(svuint64_t, svuint64_t);
23296		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
23297		svint64_t svrax1(svint64_t, svint64_t);
23298		#endif //defined(__ARM_FEATURE_SVE2_SHA3)
23299
23300		#if defined(__ARM_FEATURE_SVE2_SM4)
23301		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
23302		svuint32_t svsm4e_u32(svuint32_t, svuint32_t);
23303		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
23304		svuint32_t svsm4ekey_u32(svuint32_t, svuint32_t);
23305		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
23306		svuint32_t svsm4e(svuint32_t, svuint32_t);
23307		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
23308		svuint32_t svsm4ekey(svuint32_t, svuint32_t);
23309		#endif //defined(__ARM_FEATURE_SVE2_SM4)
23310
23311		#if defined(__ARM_FEATURE_SVE_BF16)
23312	15593	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_n_f32)))
23313	15594	svfloat32_t svbfdot_n_f32(svfloat32_t, svbfloat16_t, bfloat16_t);
23314	15595	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_f32)))
...	...	@@ -23625,16 +15906,34 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_bf16)))
23625	15906	svbfloat16_t svzip1(svbfloat16_t, svbfloat16_t);
23626	15907	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_bf16)))
23627	15908	svbfloat16_t svzip2(svbfloat16_t, svbfloat16_t);
23628		#endif //defined(__ARM_FEATURE_SVE_BF16)
23629
23630		#if defined(__ARM_FEATURE_SVE_MATMUL_FP32)
	15909	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
	15910	svbfloat16_t svtrn1q_bf16(svbfloat16_t, svbfloat16_t);
	15911	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
	15912	svbfloat16_t svtrn2q_bf16(svbfloat16_t, svbfloat16_t);
	15913	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
	15914	svbfloat16_t svuzp1q_bf16(svbfloat16_t, svbfloat16_t);
	15915	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
	15916	svbfloat16_t svuzp2q_bf16(svbfloat16_t, svbfloat16_t);
	15917	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
	15918	svbfloat16_t svzip1q_bf16(svbfloat16_t, svbfloat16_t);
	15919	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
	15920	svbfloat16_t svzip2q_bf16(svbfloat16_t, svbfloat16_t);
	15921	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
	15922	svbfloat16_t svtrn1q(svbfloat16_t, svbfloat16_t);
	15923	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
	15924	svbfloat16_t svtrn2q(svbfloat16_t, svbfloat16_t);
	15925	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
	15926	svbfloat16_t svuzp1q(svbfloat16_t, svbfloat16_t);
	15927	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
	15928	svbfloat16_t svuzp2q(svbfloat16_t, svbfloat16_t);
	15929	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
	15930	svbfloat16_t svzip1q(svbfloat16_t, svbfloat16_t);
	15931	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
	15932	svbfloat16_t svzip2q(svbfloat16_t, svbfloat16_t);
23631	15933	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f32)))
23632	15934	svfloat32_t svmmla_f32(svfloat32_t, svfloat32_t, svfloat32_t);
23633	15935	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f32)))
23634	15936	svfloat32_t svmmla(svfloat32_t, svfloat32_t, svfloat32_t);
23635		#endif //defined(__ARM_FEATURE_SVE_MATMUL_FP32)
23636
23637		#if defined(__ARM_FEATURE_SVE_MATMUL_FP64)
23638	15937	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u8)))
23639	15938	svuint8_t svld1ro_u8(svbool_t, uint8_t const *);
23640	15939	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u32)))
...	...	@@ -23947,40 +16246,10 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s64)))
23947	16246	svint64_t svzip2q(svint64_t, svint64_t);
23948	16247	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s16)))
23949	16248	svint16_t svzip2q(svint16_t, svint16_t);
23950		#endif //defined(__ARM_FEATURE_SVE_MATMUL_FP64)
23951
23952		#if defined(__ARM_FEATURE_SVE_MATMUL_FP64) && defined(__ARM_FEATURE_SVE_BF16)
23953	16249	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
23954	16250	svbfloat16_t svld1ro_bf16(svbool_t, bfloat16_t const *);
23955		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
23956		svbfloat16_t svtrn1q_bf16(svbfloat16_t, svbfloat16_t);
23957		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
23958		svbfloat16_t svtrn2q_bf16(svbfloat16_t, svbfloat16_t);
23959		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
23960		svbfloat16_t svuzp1q_bf16(svbfloat16_t, svbfloat16_t);
23961		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
23962		svbfloat16_t svuzp2q_bf16(svbfloat16_t, svbfloat16_t);
23963		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
23964		svbfloat16_t svzip1q_bf16(svbfloat16_t, svbfloat16_t);
23965		__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
23966		svbfloat16_t svzip2q_bf16(svbfloat16_t, svbfloat16_t);
23967	16251	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
23968	16252	svbfloat16_t svld1ro(svbool_t, bfloat16_t const *);
23969		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
23970		svbfloat16_t svtrn1q(svbfloat16_t, svbfloat16_t);
23971		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
23972		svbfloat16_t svtrn2q(svbfloat16_t, svbfloat16_t);
23973		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
23974		svbfloat16_t svuzp1q(svbfloat16_t, svbfloat16_t);
23975		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
23976		svbfloat16_t svuzp2q(svbfloat16_t, svbfloat16_t);
23977		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
23978		svbfloat16_t svzip1q(svbfloat16_t, svbfloat16_t);
23979		__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
23980		svbfloat16_t svzip2q(svbfloat16_t, svbfloat16_t);
23981		#endif //defined(__ARM_FEATURE_SVE_MATMUL_FP64) && defined(__ARM_FEATURE_SVE_BF16)
23982
23983		#if defined(__ARM_FEATURE_SVE_MATMUL_INT8)
23984	16253	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_s32)))
23985	16254	svint32_t svmmla_s32(svint32_t, svint8_t, svint8_t);
23986	16255	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_u32)))
...	...	@@ -24017,13 +16286,7596 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusdot_lane_s32)))
24017	16286	svint32_t svusdot_lane(svint32_t, svuint8_t, svint8_t, uint64_t);
24018	16287	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusmmla_s32)))
24019	16288	svint32_t svusmmla(svint32_t, svuint8_t, svint8_t);
24020		#endif //defined(__ARM_FEATURE_SVE_MATMUL_INT8)
24021		#if defined(__ARM_FEATURE_SVE_BF16)
	16289	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s8)))
	16290	svint8_t svaba_n_s8(svint8_t, svint8_t, int8_t);
	16291	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s32)))
	16292	svint32_t svaba_n_s32(svint32_t, svint32_t, int32_t);
	16293	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s64)))
	16294	svint64_t svaba_n_s64(svint64_t, svint64_t, int64_t);
	16295	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s16)))
	16296	svint16_t svaba_n_s16(svint16_t, svint16_t, int16_t);
	16297	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u8)))
	16298	svuint8_t svaba_n_u8(svuint8_t, svuint8_t, uint8_t);
	16299	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u32)))
	16300	svuint32_t svaba_n_u32(svuint32_t, svuint32_t, uint32_t);
	16301	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u64)))
	16302	svuint64_t svaba_n_u64(svuint64_t, svuint64_t, uint64_t);
	16303	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u16)))
	16304	svuint16_t svaba_n_u16(svuint16_t, svuint16_t, uint16_t);
	16305	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s8)))
	16306	svint8_t svaba_s8(svint8_t, svint8_t, svint8_t);
	16307	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s32)))
	16308	svint32_t svaba_s32(svint32_t, svint32_t, svint32_t);
	16309	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s64)))
	16310	svint64_t svaba_s64(svint64_t, svint64_t, svint64_t);
	16311	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s16)))
	16312	svint16_t svaba_s16(svint16_t, svint16_t, svint16_t);
	16313	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u8)))
	16314	svuint8_t svaba_u8(svuint8_t, svuint8_t, svuint8_t);
	16315	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u32)))
	16316	svuint32_t svaba_u32(svuint32_t, svuint32_t, svuint32_t);
	16317	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u64)))
	16318	svuint64_t svaba_u64(svuint64_t, svuint64_t, svuint64_t);
	16319	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u16)))
	16320	svuint16_t svaba_u16(svuint16_t, svuint16_t, svuint16_t);
	16321	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s32)))
	16322	svint32_t svabalb_n_s32(svint32_t, svint16_t, int16_t);
	16323	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s64)))
	16324	svint64_t svabalb_n_s64(svint64_t, svint32_t, int32_t);
	16325	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s16)))
	16326	svint16_t svabalb_n_s16(svint16_t, svint8_t, int8_t);
	16327	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u32)))
	16328	svuint32_t svabalb_n_u32(svuint32_t, svuint16_t, uint16_t);
	16329	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u64)))
	16330	svuint64_t svabalb_n_u64(svuint64_t, svuint32_t, uint32_t);
	16331	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u16)))
	16332	svuint16_t svabalb_n_u16(svuint16_t, svuint8_t, uint8_t);
	16333	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s32)))
	16334	svint32_t svabalb_s32(svint32_t, svint16_t, svint16_t);
	16335	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s64)))
	16336	svint64_t svabalb_s64(svint64_t, svint32_t, svint32_t);
	16337	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s16)))
	16338	svint16_t svabalb_s16(svint16_t, svint8_t, svint8_t);
	16339	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u32)))
	16340	svuint32_t svabalb_u32(svuint32_t, svuint16_t, svuint16_t);
	16341	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u64)))
	16342	svuint64_t svabalb_u64(svuint64_t, svuint32_t, svuint32_t);
	16343	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u16)))
	16344	svuint16_t svabalb_u16(svuint16_t, svuint8_t, svuint8_t);
	16345	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s32)))
	16346	svint32_t svabalt_n_s32(svint32_t, svint16_t, int16_t);
	16347	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s64)))
	16348	svint64_t svabalt_n_s64(svint64_t, svint32_t, int32_t);
	16349	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s16)))
	16350	svint16_t svabalt_n_s16(svint16_t, svint8_t, int8_t);
	16351	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u32)))
	16352	svuint32_t svabalt_n_u32(svuint32_t, svuint16_t, uint16_t);
	16353	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u64)))
	16354	svuint64_t svabalt_n_u64(svuint64_t, svuint32_t, uint32_t);
	16355	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u16)))
	16356	svuint16_t svabalt_n_u16(svuint16_t, svuint8_t, uint8_t);
	16357	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s32)))
	16358	svint32_t svabalt_s32(svint32_t, svint16_t, svint16_t);
	16359	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s64)))
	16360	svint64_t svabalt_s64(svint64_t, svint32_t, svint32_t);
	16361	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s16)))
	16362	svint16_t svabalt_s16(svint16_t, svint8_t, svint8_t);
	16363	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u32)))
	16364	svuint32_t svabalt_u32(svuint32_t, svuint16_t, svuint16_t);
	16365	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u64)))
	16366	svuint64_t svabalt_u64(svuint64_t, svuint32_t, svuint32_t);
	16367	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u16)))
	16368	svuint16_t svabalt_u16(svuint16_t, svuint8_t, svuint8_t);
	16369	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s32)))
	16370	svint32_t svabdlb_n_s32(svint16_t, int16_t);
	16371	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s64)))
	16372	svint64_t svabdlb_n_s64(svint32_t, int32_t);
	16373	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s16)))
	16374	svint16_t svabdlb_n_s16(svint8_t, int8_t);
	16375	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u32)))
	16376	svuint32_t svabdlb_n_u32(svuint16_t, uint16_t);
	16377	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u64)))
	16378	svuint64_t svabdlb_n_u64(svuint32_t, uint32_t);
	16379	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u16)))
	16380	svuint16_t svabdlb_n_u16(svuint8_t, uint8_t);
	16381	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s32)))
	16382	svint32_t svabdlb_s32(svint16_t, svint16_t);
	16383	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s64)))
	16384	svint64_t svabdlb_s64(svint32_t, svint32_t);
	16385	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s16)))
	16386	svint16_t svabdlb_s16(svint8_t, svint8_t);
	16387	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u32)))
	16388	svuint32_t svabdlb_u32(svuint16_t, svuint16_t);
	16389	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u64)))
	16390	svuint64_t svabdlb_u64(svuint32_t, svuint32_t);
	16391	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u16)))
	16392	svuint16_t svabdlb_u16(svuint8_t, svuint8_t);
	16393	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s32)))
	16394	svint32_t svabdlt_n_s32(svint16_t, int16_t);
	16395	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s64)))
	16396	svint64_t svabdlt_n_s64(svint32_t, int32_t);
	16397	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s16)))
	16398	svint16_t svabdlt_n_s16(svint8_t, int8_t);
	16399	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u32)))
	16400	svuint32_t svabdlt_n_u32(svuint16_t, uint16_t);
	16401	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u64)))
	16402	svuint64_t svabdlt_n_u64(svuint32_t, uint32_t);
	16403	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u16)))
	16404	svuint16_t svabdlt_n_u16(svuint8_t, uint8_t);
	16405	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s32)))
	16406	svint32_t svabdlt_s32(svint16_t, svint16_t);
	16407	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s64)))
	16408	svint64_t svabdlt_s64(svint32_t, svint32_t);
	16409	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s16)))
	16410	svint16_t svabdlt_s16(svint8_t, svint8_t);
	16411	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u32)))
	16412	svuint32_t svabdlt_u32(svuint16_t, svuint16_t);
	16413	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u64)))
	16414	svuint64_t svabdlt_u64(svuint32_t, svuint32_t);
	16415	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u16)))
	16416	svuint16_t svabdlt_u16(svuint8_t, svuint8_t);
	16417	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_m)))
	16418	svint32_t svadalp_s32_m(svbool_t, svint32_t, svint16_t);
	16419	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_m)))
	16420	svint64_t svadalp_s64_m(svbool_t, svint64_t, svint32_t);
	16421	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_m)))
	16422	svint16_t svadalp_s16_m(svbool_t, svint16_t, svint8_t);
	16423	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_x)))
	16424	svint32_t svadalp_s32_x(svbool_t, svint32_t, svint16_t);
	16425	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_x)))
	16426	svint64_t svadalp_s64_x(svbool_t, svint64_t, svint32_t);
	16427	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_x)))
	16428	svint16_t svadalp_s16_x(svbool_t, svint16_t, svint8_t);
	16429	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_z)))
	16430	svint32_t svadalp_s32_z(svbool_t, svint32_t, svint16_t);
	16431	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_z)))
	16432	svint64_t svadalp_s64_z(svbool_t, svint64_t, svint32_t);
	16433	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_z)))
	16434	svint16_t svadalp_s16_z(svbool_t, svint16_t, svint8_t);
	16435	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_m)))
	16436	svuint32_t svadalp_u32_m(svbool_t, svuint32_t, svuint16_t);
	16437	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_m)))
	16438	svuint64_t svadalp_u64_m(svbool_t, svuint64_t, svuint32_t);
	16439	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_m)))
	16440	svuint16_t svadalp_u16_m(svbool_t, svuint16_t, svuint8_t);
	16441	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_x)))
	16442	svuint32_t svadalp_u32_x(svbool_t, svuint32_t, svuint16_t);
	16443	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_x)))
	16444	svuint64_t svadalp_u64_x(svbool_t, svuint64_t, svuint32_t);
	16445	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_x)))
	16446	svuint16_t svadalp_u16_x(svbool_t, svuint16_t, svuint8_t);
	16447	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_z)))
	16448	svuint32_t svadalp_u32_z(svbool_t, svuint32_t, svuint16_t);
	16449	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_z)))
	16450	svuint64_t svadalp_u64_z(svbool_t, svuint64_t, svuint32_t);
	16451	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_z)))
	16452	svuint16_t svadalp_u16_z(svbool_t, svuint16_t, svuint8_t);
	16453	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u32)))
	16454	svuint32_t svadclb_n_u32(svuint32_t, svuint32_t, uint32_t);
	16455	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u64)))
	16456	svuint64_t svadclb_n_u64(svuint64_t, svuint64_t, uint64_t);
	16457	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u32)))
	16458	svuint32_t svadclb_u32(svuint32_t, svuint32_t, svuint32_t);
	16459	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u64)))
	16460	svuint64_t svadclb_u64(svuint64_t, svuint64_t, svuint64_t);
	16461	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u32)))
	16462	svuint32_t svadclt_n_u32(svuint32_t, svuint32_t, uint32_t);
	16463	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u64)))
	16464	svuint64_t svadclt_n_u64(svuint64_t, svuint64_t, uint64_t);
	16465	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u32)))
	16466	svuint32_t svadclt_u32(svuint32_t, svuint32_t, svuint32_t);
	16467	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u64)))
	16468	svuint64_t svadclt_u64(svuint64_t, svuint64_t, svuint64_t);
	16469	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u32)))
	16470	svuint16_t svaddhnb_n_u32(svuint32_t, uint32_t);
	16471	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u64)))
	16472	svuint32_t svaddhnb_n_u64(svuint64_t, uint64_t);
	16473	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u16)))
	16474	svuint8_t svaddhnb_n_u16(svuint16_t, uint16_t);
	16475	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s32)))
	16476	svint16_t svaddhnb_n_s32(svint32_t, int32_t);
	16477	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s64)))
	16478	svint32_t svaddhnb_n_s64(svint64_t, int64_t);
	16479	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s16)))
	16480	svint8_t svaddhnb_n_s16(svint16_t, int16_t);
	16481	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u32)))
	16482	svuint16_t svaddhnb_u32(svuint32_t, svuint32_t);
	16483	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u64)))
	16484	svuint32_t svaddhnb_u64(svuint64_t, svuint64_t);
	16485	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u16)))
	16486	svuint8_t svaddhnb_u16(svuint16_t, svuint16_t);
	16487	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s32)))
	16488	svint16_t svaddhnb_s32(svint32_t, svint32_t);
	16489	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s64)))
	16490	svint32_t svaddhnb_s64(svint64_t, svint64_t);
	16491	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s16)))
	16492	svint8_t svaddhnb_s16(svint16_t, svint16_t);
	16493	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u32)))
	16494	svuint16_t svaddhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
	16495	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u64)))
	16496	svuint32_t svaddhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	16497	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u16)))
	16498	svuint8_t svaddhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
	16499	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s32)))
	16500	svint16_t svaddhnt_n_s32(svint16_t, svint32_t, int32_t);
	16501	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s64)))
	16502	svint32_t svaddhnt_n_s64(svint32_t, svint64_t, int64_t);
	16503	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s16)))
	16504	svint8_t svaddhnt_n_s16(svint8_t, svint16_t, int16_t);
	16505	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u32)))
	16506	svuint16_t svaddhnt_u32(svuint16_t, svuint32_t, svuint32_t);
	16507	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u64)))
	16508	svuint32_t svaddhnt_u64(svuint32_t, svuint64_t, svuint64_t);
	16509	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u16)))
	16510	svuint8_t svaddhnt_u16(svuint8_t, svuint16_t, svuint16_t);
	16511	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s32)))
	16512	svint16_t svaddhnt_s32(svint16_t, svint32_t, svint32_t);
	16513	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s64)))
	16514	svint32_t svaddhnt_s64(svint32_t, svint64_t, svint64_t);
	16515	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s16)))
	16516	svint8_t svaddhnt_s16(svint8_t, svint16_t, svint16_t);
	16517	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s32)))
	16518	svint32_t svaddlb_n_s32(svint16_t, int16_t);
	16519	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s64)))
	16520	svint64_t svaddlb_n_s64(svint32_t, int32_t);
	16521	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s16)))
	16522	svint16_t svaddlb_n_s16(svint8_t, int8_t);
	16523	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u32)))
	16524	svuint32_t svaddlb_n_u32(svuint16_t, uint16_t);
	16525	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u64)))
	16526	svuint64_t svaddlb_n_u64(svuint32_t, uint32_t);
	16527	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u16)))
	16528	svuint16_t svaddlb_n_u16(svuint8_t, uint8_t);
	16529	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s32)))
	16530	svint32_t svaddlb_s32(svint16_t, svint16_t);
	16531	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s64)))
	16532	svint64_t svaddlb_s64(svint32_t, svint32_t);
	16533	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s16)))
	16534	svint16_t svaddlb_s16(svint8_t, svint8_t);
	16535	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u32)))
	16536	svuint32_t svaddlb_u32(svuint16_t, svuint16_t);
	16537	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u64)))
	16538	svuint64_t svaddlb_u64(svuint32_t, svuint32_t);
	16539	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u16)))
	16540	svuint16_t svaddlb_u16(svuint8_t, svuint8_t);
	16541	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s32)))
	16542	svint32_t svaddlbt_n_s32(svint16_t, int16_t);
	16543	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s64)))
	16544	svint64_t svaddlbt_n_s64(svint32_t, int32_t);
	16545	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s16)))
	16546	svint16_t svaddlbt_n_s16(svint8_t, int8_t);
	16547	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s32)))
	16548	svint32_t svaddlbt_s32(svint16_t, svint16_t);
	16549	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s64)))
	16550	svint64_t svaddlbt_s64(svint32_t, svint32_t);
	16551	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s16)))
	16552	svint16_t svaddlbt_s16(svint8_t, svint8_t);
	16553	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s32)))
	16554	svint32_t svaddlt_n_s32(svint16_t, int16_t);
	16555	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s64)))
	16556	svint64_t svaddlt_n_s64(svint32_t, int32_t);
	16557	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s16)))
	16558	svint16_t svaddlt_n_s16(svint8_t, int8_t);
	16559	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u32)))
	16560	svuint32_t svaddlt_n_u32(svuint16_t, uint16_t);
	16561	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u64)))
	16562	svuint64_t svaddlt_n_u64(svuint32_t, uint32_t);
	16563	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u16)))
	16564	svuint16_t svaddlt_n_u16(svuint8_t, uint8_t);
	16565	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s32)))
	16566	svint32_t svaddlt_s32(svint16_t, svint16_t);
	16567	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s64)))
	16568	svint64_t svaddlt_s64(svint32_t, svint32_t);
	16569	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s16)))
	16570	svint16_t svaddlt_s16(svint8_t, svint8_t);
	16571	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u32)))
	16572	svuint32_t svaddlt_u32(svuint16_t, svuint16_t);
	16573	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u64)))
	16574	svuint64_t svaddlt_u64(svuint32_t, svuint32_t);
	16575	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u16)))
	16576	svuint16_t svaddlt_u16(svuint8_t, svuint8_t);
	16577	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_m)))
	16578	svfloat64_t svaddp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
	16579	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_m)))
	16580	svfloat32_t svaddp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
	16581	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_m)))
	16582	svfloat16_t svaddp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
	16583	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_x)))
	16584	svfloat64_t svaddp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
	16585	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_x)))
	16586	svfloat32_t svaddp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
	16587	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_x)))
	16588	svfloat16_t svaddp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
	16589	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_m)))
	16590	svuint8_t svaddp_u8_m(svbool_t, svuint8_t, svuint8_t);
	16591	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_m)))
	16592	svuint32_t svaddp_u32_m(svbool_t, svuint32_t, svuint32_t);
	16593	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_m)))
	16594	svuint64_t svaddp_u64_m(svbool_t, svuint64_t, svuint64_t);
	16595	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_m)))
	16596	svuint16_t svaddp_u16_m(svbool_t, svuint16_t, svuint16_t);
	16597	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_m)))
	16598	svint8_t svaddp_s8_m(svbool_t, svint8_t, svint8_t);
	16599	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_m)))
	16600	svint32_t svaddp_s32_m(svbool_t, svint32_t, svint32_t);
	16601	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_m)))
	16602	svint64_t svaddp_s64_m(svbool_t, svint64_t, svint64_t);
	16603	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_m)))
	16604	svint16_t svaddp_s16_m(svbool_t, svint16_t, svint16_t);
	16605	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_x)))
	16606	svuint8_t svaddp_u8_x(svbool_t, svuint8_t, svuint8_t);
	16607	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_x)))
	16608	svuint32_t svaddp_u32_x(svbool_t, svuint32_t, svuint32_t);
	16609	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_x)))
	16610	svuint64_t svaddp_u64_x(svbool_t, svuint64_t, svuint64_t);
	16611	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_x)))
	16612	svuint16_t svaddp_u16_x(svbool_t, svuint16_t, svuint16_t);
	16613	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_x)))
	16614	svint8_t svaddp_s8_x(svbool_t, svint8_t, svint8_t);
	16615	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_x)))
	16616	svint32_t svaddp_s32_x(svbool_t, svint32_t, svint32_t);
	16617	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_x)))
	16618	svint64_t svaddp_s64_x(svbool_t, svint64_t, svint64_t);
	16619	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_x)))
	16620	svint16_t svaddp_s16_x(svbool_t, svint16_t, svint16_t);
	16621	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s32)))
	16622	svint32_t svaddwb_n_s32(svint32_t, int16_t);
	16623	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s64)))
	16624	svint64_t svaddwb_n_s64(svint64_t, int32_t);
	16625	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s16)))
	16626	svint16_t svaddwb_n_s16(svint16_t, int8_t);
	16627	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u32)))
	16628	svuint32_t svaddwb_n_u32(svuint32_t, uint16_t);
	16629	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u64)))
	16630	svuint64_t svaddwb_n_u64(svuint64_t, uint32_t);
	16631	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u16)))
	16632	svuint16_t svaddwb_n_u16(svuint16_t, uint8_t);
	16633	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s32)))
	16634	svint32_t svaddwb_s32(svint32_t, svint16_t);
	16635	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s64)))
	16636	svint64_t svaddwb_s64(svint64_t, svint32_t);
	16637	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s16)))
	16638	svint16_t svaddwb_s16(svint16_t, svint8_t);
	16639	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u32)))
	16640	svuint32_t svaddwb_u32(svuint32_t, svuint16_t);
	16641	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u64)))
	16642	svuint64_t svaddwb_u64(svuint64_t, svuint32_t);
	16643	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u16)))
	16644	svuint16_t svaddwb_u16(svuint16_t, svuint8_t);
	16645	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s32)))
	16646	svint32_t svaddwt_n_s32(svint32_t, int16_t);
	16647	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s64)))
	16648	svint64_t svaddwt_n_s64(svint64_t, int32_t);
	16649	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s16)))
	16650	svint16_t svaddwt_n_s16(svint16_t, int8_t);
	16651	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u32)))
	16652	svuint32_t svaddwt_n_u32(svuint32_t, uint16_t);
	16653	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u64)))
	16654	svuint64_t svaddwt_n_u64(svuint64_t, uint32_t);
	16655	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u16)))
	16656	svuint16_t svaddwt_n_u16(svuint16_t, uint8_t);
	16657	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s32)))
	16658	svint32_t svaddwt_s32(svint32_t, svint16_t);
	16659	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s64)))
	16660	svint64_t svaddwt_s64(svint64_t, svint32_t);
	16661	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s16)))
	16662	svint16_t svaddwt_s16(svint16_t, svint8_t);
	16663	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u32)))
	16664	svuint32_t svaddwt_u32(svuint32_t, svuint16_t);
	16665	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u64)))
	16666	svuint64_t svaddwt_u64(svuint64_t, svuint32_t);
	16667	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u16)))
	16668	svuint16_t svaddwt_u16(svuint16_t, svuint8_t);
	16669	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u8)))
	16670	svuint8_t svbcax_n_u8(svuint8_t, svuint8_t, uint8_t);
	16671	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u32)))
	16672	svuint32_t svbcax_n_u32(svuint32_t, svuint32_t, uint32_t);
	16673	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u64)))
	16674	svuint64_t svbcax_n_u64(svuint64_t, svuint64_t, uint64_t);
	16675	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u16)))
	16676	svuint16_t svbcax_n_u16(svuint16_t, svuint16_t, uint16_t);
	16677	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s8)))
	16678	svint8_t svbcax_n_s8(svint8_t, svint8_t, int8_t);
	16679	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s32)))
	16680	svint32_t svbcax_n_s32(svint32_t, svint32_t, int32_t);
	16681	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s64)))
	16682	svint64_t svbcax_n_s64(svint64_t, svint64_t, int64_t);
	16683	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s16)))
	16684	svint16_t svbcax_n_s16(svint16_t, svint16_t, int16_t);
	16685	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u8)))
	16686	svuint8_t svbcax_u8(svuint8_t, svuint8_t, svuint8_t);
	16687	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u32)))
	16688	svuint32_t svbcax_u32(svuint32_t, svuint32_t, svuint32_t);
	16689	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u64)))
	16690	svuint64_t svbcax_u64(svuint64_t, svuint64_t, svuint64_t);
	16691	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u16)))
	16692	svuint16_t svbcax_u16(svuint16_t, svuint16_t, svuint16_t);
	16693	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s8)))
	16694	svint8_t svbcax_s8(svint8_t, svint8_t, svint8_t);
	16695	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s32)))
	16696	svint32_t svbcax_s32(svint32_t, svint32_t, svint32_t);
	16697	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s64)))
	16698	svint64_t svbcax_s64(svint64_t, svint64_t, svint64_t);
	16699	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s16)))
	16700	svint16_t svbcax_s16(svint16_t, svint16_t, svint16_t);
	16701	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u8)))
	16702	svuint8_t svbsl1n_n_u8(svuint8_t, svuint8_t, uint8_t);
	16703	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u32)))
	16704	svuint32_t svbsl1n_n_u32(svuint32_t, svuint32_t, uint32_t);
	16705	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u64)))
	16706	svuint64_t svbsl1n_n_u64(svuint64_t, svuint64_t, uint64_t);
	16707	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u16)))
	16708	svuint16_t svbsl1n_n_u16(svuint16_t, svuint16_t, uint16_t);
	16709	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s8)))
	16710	svint8_t svbsl1n_n_s8(svint8_t, svint8_t, int8_t);
	16711	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s32)))
	16712	svint32_t svbsl1n_n_s32(svint32_t, svint32_t, int32_t);
	16713	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s64)))
	16714	svint64_t svbsl1n_n_s64(svint64_t, svint64_t, int64_t);
	16715	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s16)))
	16716	svint16_t svbsl1n_n_s16(svint16_t, svint16_t, int16_t);
	16717	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u8)))
	16718	svuint8_t svbsl1n_u8(svuint8_t, svuint8_t, svuint8_t);
	16719	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u32)))
	16720	svuint32_t svbsl1n_u32(svuint32_t, svuint32_t, svuint32_t);
	16721	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u64)))
	16722	svuint64_t svbsl1n_u64(svuint64_t, svuint64_t, svuint64_t);
	16723	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u16)))
	16724	svuint16_t svbsl1n_u16(svuint16_t, svuint16_t, svuint16_t);
	16725	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s8)))
	16726	svint8_t svbsl1n_s8(svint8_t, svint8_t, svint8_t);
	16727	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s32)))
	16728	svint32_t svbsl1n_s32(svint32_t, svint32_t, svint32_t);
	16729	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s64)))
	16730	svint64_t svbsl1n_s64(svint64_t, svint64_t, svint64_t);
	16731	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s16)))
	16732	svint16_t svbsl1n_s16(svint16_t, svint16_t, svint16_t);
	16733	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u8)))
	16734	svuint8_t svbsl2n_n_u8(svuint8_t, svuint8_t, uint8_t);
	16735	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u32)))
	16736	svuint32_t svbsl2n_n_u32(svuint32_t, svuint32_t, uint32_t);
	16737	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u64)))
	16738	svuint64_t svbsl2n_n_u64(svuint64_t, svuint64_t, uint64_t);
	16739	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u16)))
	16740	svuint16_t svbsl2n_n_u16(svuint16_t, svuint16_t, uint16_t);
	16741	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s8)))
	16742	svint8_t svbsl2n_n_s8(svint8_t, svint8_t, int8_t);
	16743	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s32)))
	16744	svint32_t svbsl2n_n_s32(svint32_t, svint32_t, int32_t);
	16745	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s64)))
	16746	svint64_t svbsl2n_n_s64(svint64_t, svint64_t, int64_t);
	16747	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s16)))
	16748	svint16_t svbsl2n_n_s16(svint16_t, svint16_t, int16_t);
	16749	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u8)))
	16750	svuint8_t svbsl2n_u8(svuint8_t, svuint8_t, svuint8_t);
	16751	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u32)))
	16752	svuint32_t svbsl2n_u32(svuint32_t, svuint32_t, svuint32_t);
	16753	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u64)))
	16754	svuint64_t svbsl2n_u64(svuint64_t, svuint64_t, svuint64_t);
	16755	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u16)))
	16756	svuint16_t svbsl2n_u16(svuint16_t, svuint16_t, svuint16_t);
	16757	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s8)))
	16758	svint8_t svbsl2n_s8(svint8_t, svint8_t, svint8_t);
	16759	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s32)))
	16760	svint32_t svbsl2n_s32(svint32_t, svint32_t, svint32_t);
	16761	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s64)))
	16762	svint64_t svbsl2n_s64(svint64_t, svint64_t, svint64_t);
	16763	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s16)))
	16764	svint16_t svbsl2n_s16(svint16_t, svint16_t, svint16_t);
	16765	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u8)))
	16766	svuint8_t svbsl_n_u8(svuint8_t, svuint8_t, uint8_t);
	16767	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u32)))
	16768	svuint32_t svbsl_n_u32(svuint32_t, svuint32_t, uint32_t);
	16769	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u64)))
	16770	svuint64_t svbsl_n_u64(svuint64_t, svuint64_t, uint64_t);
	16771	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u16)))
	16772	svuint16_t svbsl_n_u16(svuint16_t, svuint16_t, uint16_t);
	16773	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s8)))
	16774	svint8_t svbsl_n_s8(svint8_t, svint8_t, int8_t);
	16775	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s32)))
	16776	svint32_t svbsl_n_s32(svint32_t, svint32_t, int32_t);
	16777	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s64)))
	16778	svint64_t svbsl_n_s64(svint64_t, svint64_t, int64_t);
	16779	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s16)))
	16780	svint16_t svbsl_n_s16(svint16_t, svint16_t, int16_t);
	16781	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u8)))
	16782	svuint8_t svbsl_u8(svuint8_t, svuint8_t, svuint8_t);
	16783	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u32)))
	16784	svuint32_t svbsl_u32(svuint32_t, svuint32_t, svuint32_t);
	16785	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u64)))
	16786	svuint64_t svbsl_u64(svuint64_t, svuint64_t, svuint64_t);
	16787	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u16)))
	16788	svuint16_t svbsl_u16(svuint16_t, svuint16_t, svuint16_t);
	16789	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s8)))
	16790	svint8_t svbsl_s8(svint8_t, svint8_t, svint8_t);
	16791	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s32)))
	16792	svint32_t svbsl_s32(svint32_t, svint32_t, svint32_t);
	16793	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s64)))
	16794	svint64_t svbsl_s64(svint64_t, svint64_t, svint64_t);
	16795	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s16)))
	16796	svint16_t svbsl_s16(svint16_t, svint16_t, svint16_t);
	16797	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u8)))
	16798	svuint8_t svcadd_u8(svuint8_t, svuint8_t, uint64_t);
	16799	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u32)))
	16800	svuint32_t svcadd_u32(svuint32_t, svuint32_t, uint64_t);
	16801	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u64)))
	16802	svuint64_t svcadd_u64(svuint64_t, svuint64_t, uint64_t);
	16803	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u16)))
	16804	svuint16_t svcadd_u16(svuint16_t, svuint16_t, uint64_t);
	16805	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s8)))
	16806	svint8_t svcadd_s8(svint8_t, svint8_t, uint64_t);
	16807	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s32)))
	16808	svint32_t svcadd_s32(svint32_t, svint32_t, uint64_t);
	16809	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s64)))
	16810	svint64_t svcadd_s64(svint64_t, svint64_t, uint64_t);
	16811	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s16)))
	16812	svint16_t svcadd_s16(svint16_t, svint16_t, uint64_t);
	16813	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s32)))
	16814	svint32_t svcdot_s32(svint32_t, svint8_t, svint8_t, uint64_t);
	16815	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s64)))
	16816	svint64_t svcdot_s64(svint64_t, svint16_t, svint16_t, uint64_t);
	16817	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s32)))
	16818	svint32_t svcdot_lane_s32(svint32_t, svint8_t, svint8_t, uint64_t, uint64_t);
	16819	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s64)))
	16820	svint64_t svcdot_lane_s64(svint64_t, svint16_t, svint16_t, uint64_t, uint64_t);
	16821	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u8)))
	16822	svuint8_t svcmla_u8(svuint8_t, svuint8_t, svuint8_t, uint64_t);
	16823	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u32)))
	16824	svuint32_t svcmla_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t);
	16825	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u64)))
	16826	svuint64_t svcmla_u64(svuint64_t, svuint64_t, svuint64_t, uint64_t);
	16827	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u16)))
	16828	svuint16_t svcmla_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t);
	16829	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s8)))
	16830	svint8_t svcmla_s8(svint8_t, svint8_t, svint8_t, uint64_t);
	16831	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s32)))
	16832	svint32_t svcmla_s32(svint32_t, svint32_t, svint32_t, uint64_t);
	16833	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s64)))
	16834	svint64_t svcmla_s64(svint64_t, svint64_t, svint64_t, uint64_t);
	16835	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s16)))
	16836	svint16_t svcmla_s16(svint16_t, svint16_t, svint16_t, uint64_t);
	16837	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u32)))
	16838	svuint32_t svcmla_lane_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t, uint64_t);
	16839	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u16)))
	16840	svuint16_t svcmla_lane_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t, uint64_t);
	16841	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s32)))
	16842	svint32_t svcmla_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
	16843	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s16)))
	16844	svint16_t svcmla_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
	16845	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_m)))
	16846	svfloat32_t svcvtlt_f32_f16_m(svfloat32_t, svbool_t, svfloat16_t);
	16847	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_x)))
	16848	svfloat32_t svcvtlt_f32_f16_x(svbool_t, svfloat16_t);
	16849	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_m)))
	16850	svfloat64_t svcvtlt_f64_f32_m(svfloat64_t, svbool_t, svfloat32_t);
	16851	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_x)))
	16852	svfloat64_t svcvtlt_f64_f32_x(svbool_t, svfloat32_t);
	16853	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f16_f32_m)))
	16854	svfloat16_t svcvtnt_f16_f32_m(svfloat16_t, svbool_t, svfloat32_t);
	16855	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f32_f64_m)))
	16856	svfloat32_t svcvtnt_f32_f64_m(svfloat32_t, svbool_t, svfloat64_t);
	16857	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_m)))
	16858	svfloat32_t svcvtx_f32_f64_m(svfloat32_t, svbool_t, svfloat64_t);
	16859	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_x)))
	16860	svfloat32_t svcvtx_f32_f64_x(svbool_t, svfloat64_t);
	16861	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_z)))
	16862	svfloat32_t svcvtx_f32_f64_z(svbool_t, svfloat64_t);
	16863	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtxnt_f32_f64_m)))
	16864	svfloat32_t svcvtxnt_f32_f64_m(svfloat32_t, svbool_t, svfloat64_t);
	16865	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u8)))
	16866	svuint8_t sveor3_n_u8(svuint8_t, svuint8_t, uint8_t);
	16867	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u32)))
	16868	svuint32_t sveor3_n_u32(svuint32_t, svuint32_t, uint32_t);
	16869	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u64)))
	16870	svuint64_t sveor3_n_u64(svuint64_t, svuint64_t, uint64_t);
	16871	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u16)))
	16872	svuint16_t sveor3_n_u16(svuint16_t, svuint16_t, uint16_t);
	16873	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s8)))
	16874	svint8_t sveor3_n_s8(svint8_t, svint8_t, int8_t);
	16875	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s32)))
	16876	svint32_t sveor3_n_s32(svint32_t, svint32_t, int32_t);
	16877	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s64)))
	16878	svint64_t sveor3_n_s64(svint64_t, svint64_t, int64_t);
	16879	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s16)))
	16880	svint16_t sveor3_n_s16(svint16_t, svint16_t, int16_t);
	16881	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u8)))
	16882	svuint8_t sveor3_u8(svuint8_t, svuint8_t, svuint8_t);
	16883	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u32)))
	16884	svuint32_t sveor3_u32(svuint32_t, svuint32_t, svuint32_t);
	16885	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u64)))
	16886	svuint64_t sveor3_u64(svuint64_t, svuint64_t, svuint64_t);
	16887	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u16)))
	16888	svuint16_t sveor3_u16(svuint16_t, svuint16_t, svuint16_t);
	16889	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s8)))
	16890	svint8_t sveor3_s8(svint8_t, svint8_t, svint8_t);
	16891	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s32)))
	16892	svint32_t sveor3_s32(svint32_t, svint32_t, svint32_t);
	16893	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s64)))
	16894	svint64_t sveor3_s64(svint64_t, svint64_t, svint64_t);
	16895	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s16)))
	16896	svint16_t sveor3_s16(svint16_t, svint16_t, svint16_t);
	16897	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u8)))
	16898	svuint8_t sveorbt_n_u8(svuint8_t, svuint8_t, uint8_t);
	16899	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u32)))
	16900	svuint32_t sveorbt_n_u32(svuint32_t, svuint32_t, uint32_t);
	16901	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u64)))
	16902	svuint64_t sveorbt_n_u64(svuint64_t, svuint64_t, uint64_t);
	16903	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u16)))
	16904	svuint16_t sveorbt_n_u16(svuint16_t, svuint16_t, uint16_t);
	16905	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s8)))
	16906	svint8_t sveorbt_n_s8(svint8_t, svint8_t, int8_t);
	16907	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s32)))
	16908	svint32_t sveorbt_n_s32(svint32_t, svint32_t, int32_t);
	16909	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s64)))
	16910	svint64_t sveorbt_n_s64(svint64_t, svint64_t, int64_t);
	16911	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s16)))
	16912	svint16_t sveorbt_n_s16(svint16_t, svint16_t, int16_t);
	16913	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u8)))
	16914	svuint8_t sveorbt_u8(svuint8_t, svuint8_t, svuint8_t);
	16915	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u32)))
	16916	svuint32_t sveorbt_u32(svuint32_t, svuint32_t, svuint32_t);
	16917	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u64)))
	16918	svuint64_t sveorbt_u64(svuint64_t, svuint64_t, svuint64_t);
	16919	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u16)))
	16920	svuint16_t sveorbt_u16(svuint16_t, svuint16_t, svuint16_t);
	16921	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s8)))
	16922	svint8_t sveorbt_s8(svint8_t, svint8_t, svint8_t);
	16923	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s32)))
	16924	svint32_t sveorbt_s32(svint32_t, svint32_t, svint32_t);
	16925	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s64)))
	16926	svint64_t sveorbt_s64(svint64_t, svint64_t, svint64_t);
	16927	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s16)))
	16928	svint16_t sveorbt_s16(svint16_t, svint16_t, svint16_t);
	16929	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u8)))
	16930	svuint8_t sveortb_n_u8(svuint8_t, svuint8_t, uint8_t);
	16931	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u32)))
	16932	svuint32_t sveortb_n_u32(svuint32_t, svuint32_t, uint32_t);
	16933	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u64)))
	16934	svuint64_t sveortb_n_u64(svuint64_t, svuint64_t, uint64_t);
	16935	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u16)))
	16936	svuint16_t sveortb_n_u16(svuint16_t, svuint16_t, uint16_t);
	16937	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s8)))
	16938	svint8_t sveortb_n_s8(svint8_t, svint8_t, int8_t);
	16939	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s32)))
	16940	svint32_t sveortb_n_s32(svint32_t, svint32_t, int32_t);
	16941	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s64)))
	16942	svint64_t sveortb_n_s64(svint64_t, svint64_t, int64_t);
	16943	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s16)))
	16944	svint16_t sveortb_n_s16(svint16_t, svint16_t, int16_t);
	16945	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u8)))
	16946	svuint8_t sveortb_u8(svuint8_t, svuint8_t, svuint8_t);
	16947	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u32)))
	16948	svuint32_t sveortb_u32(svuint32_t, svuint32_t, svuint32_t);
	16949	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u64)))
	16950	svuint64_t sveortb_u64(svuint64_t, svuint64_t, svuint64_t);
	16951	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u16)))
	16952	svuint16_t sveortb_u16(svuint16_t, svuint16_t, svuint16_t);
	16953	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s8)))
	16954	svint8_t sveortb_s8(svint8_t, svint8_t, svint8_t);
	16955	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s32)))
	16956	svint32_t sveortb_s32(svint32_t, svint32_t, svint32_t);
	16957	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s64)))
	16958	svint64_t sveortb_s64(svint64_t, svint64_t, svint64_t);
	16959	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s16)))
	16960	svint16_t sveortb_s16(svint16_t, svint16_t, svint16_t);
	16961	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_m)))
	16962	svint8_t svhadd_n_s8_m(svbool_t, svint8_t, int8_t);
	16963	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_m)))
	16964	svint32_t svhadd_n_s32_m(svbool_t, svint32_t, int32_t);
	16965	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_m)))
	16966	svint64_t svhadd_n_s64_m(svbool_t, svint64_t, int64_t);
	16967	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_m)))
	16968	svint16_t svhadd_n_s16_m(svbool_t, svint16_t, int16_t);
	16969	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_x)))
	16970	svint8_t svhadd_n_s8_x(svbool_t, svint8_t, int8_t);
	16971	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_x)))
	16972	svint32_t svhadd_n_s32_x(svbool_t, svint32_t, int32_t);
	16973	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_x)))
	16974	svint64_t svhadd_n_s64_x(svbool_t, svint64_t, int64_t);
	16975	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_x)))
	16976	svint16_t svhadd_n_s16_x(svbool_t, svint16_t, int16_t);
	16977	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_z)))
	16978	svint8_t svhadd_n_s8_z(svbool_t, svint8_t, int8_t);
	16979	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_z)))
	16980	svint32_t svhadd_n_s32_z(svbool_t, svint32_t, int32_t);
	16981	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_z)))
	16982	svint64_t svhadd_n_s64_z(svbool_t, svint64_t, int64_t);
	16983	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_z)))
	16984	svint16_t svhadd_n_s16_z(svbool_t, svint16_t, int16_t);
	16985	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_m)))
	16986	svuint8_t svhadd_n_u8_m(svbool_t, svuint8_t, uint8_t);
	16987	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_m)))
	16988	svuint32_t svhadd_n_u32_m(svbool_t, svuint32_t, uint32_t);
	16989	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_m)))
	16990	svuint64_t svhadd_n_u64_m(svbool_t, svuint64_t, uint64_t);
	16991	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_m)))
	16992	svuint16_t svhadd_n_u16_m(svbool_t, svuint16_t, uint16_t);
	16993	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_x)))
	16994	svuint8_t svhadd_n_u8_x(svbool_t, svuint8_t, uint8_t);
	16995	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_x)))
	16996	svuint32_t svhadd_n_u32_x(svbool_t, svuint32_t, uint32_t);
	16997	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_x)))
	16998	svuint64_t svhadd_n_u64_x(svbool_t, svuint64_t, uint64_t);
	16999	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_x)))
	17000	svuint16_t svhadd_n_u16_x(svbool_t, svuint16_t, uint16_t);
	17001	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_z)))
	17002	svuint8_t svhadd_n_u8_z(svbool_t, svuint8_t, uint8_t);
	17003	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_z)))
	17004	svuint32_t svhadd_n_u32_z(svbool_t, svuint32_t, uint32_t);
	17005	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_z)))
	17006	svuint64_t svhadd_n_u64_z(svbool_t, svuint64_t, uint64_t);
	17007	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_z)))
	17008	svuint16_t svhadd_n_u16_z(svbool_t, svuint16_t, uint16_t);
	17009	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_m)))
	17010	svint8_t svhadd_s8_m(svbool_t, svint8_t, svint8_t);
	17011	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_m)))
	17012	svint32_t svhadd_s32_m(svbool_t, svint32_t, svint32_t);
	17013	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_m)))
	17014	svint64_t svhadd_s64_m(svbool_t, svint64_t, svint64_t);
	17015	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_m)))
	17016	svint16_t svhadd_s16_m(svbool_t, svint16_t, svint16_t);
	17017	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_x)))
	17018	svint8_t svhadd_s8_x(svbool_t, svint8_t, svint8_t);
	17019	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_x)))
	17020	svint32_t svhadd_s32_x(svbool_t, svint32_t, svint32_t);
	17021	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_x)))
	17022	svint64_t svhadd_s64_x(svbool_t, svint64_t, svint64_t);
	17023	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_x)))
	17024	svint16_t svhadd_s16_x(svbool_t, svint16_t, svint16_t);
	17025	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_z)))
	17026	svint8_t svhadd_s8_z(svbool_t, svint8_t, svint8_t);
	17027	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_z)))
	17028	svint32_t svhadd_s32_z(svbool_t, svint32_t, svint32_t);
	17029	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_z)))
	17030	svint64_t svhadd_s64_z(svbool_t, svint64_t, svint64_t);
	17031	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_z)))
	17032	svint16_t svhadd_s16_z(svbool_t, svint16_t, svint16_t);
	17033	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_m)))
	17034	svuint8_t svhadd_u8_m(svbool_t, svuint8_t, svuint8_t);
	17035	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_m)))
	17036	svuint32_t svhadd_u32_m(svbool_t, svuint32_t, svuint32_t);
	17037	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_m)))
	17038	svuint64_t svhadd_u64_m(svbool_t, svuint64_t, svuint64_t);
	17039	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_m)))
	17040	svuint16_t svhadd_u16_m(svbool_t, svuint16_t, svuint16_t);
	17041	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_x)))
	17042	svuint8_t svhadd_u8_x(svbool_t, svuint8_t, svuint8_t);
	17043	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_x)))
	17044	svuint32_t svhadd_u32_x(svbool_t, svuint32_t, svuint32_t);
	17045	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_x)))
	17046	svuint64_t svhadd_u64_x(svbool_t, svuint64_t, svuint64_t);
	17047	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_x)))
	17048	svuint16_t svhadd_u16_x(svbool_t, svuint16_t, svuint16_t);
	17049	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_z)))
	17050	svuint8_t svhadd_u8_z(svbool_t, svuint8_t, svuint8_t);
	17051	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_z)))
	17052	svuint32_t svhadd_u32_z(svbool_t, svuint32_t, svuint32_t);
	17053	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_z)))
	17054	svuint64_t svhadd_u64_z(svbool_t, svuint64_t, svuint64_t);
	17055	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_z)))
	17056	svuint16_t svhadd_u16_z(svbool_t, svuint16_t, svuint16_t);
	17057	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u32_z)))
	17058	svuint32_t svhistcnt_u32_z(svbool_t, svuint32_t, svuint32_t);
	17059	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u64_z)))
	17060	svuint64_t svhistcnt_u64_z(svbool_t, svuint64_t, svuint64_t);
	17061	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s32_z)))
	17062	svuint32_t svhistcnt_s32_z(svbool_t, svint32_t, svint32_t);
	17063	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s64_z)))
	17064	svuint64_t svhistcnt_s64_z(svbool_t, svint64_t, svint64_t);
	17065	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_u8)))
	17066	svuint8_t svhistseg_u8(svuint8_t, svuint8_t);
	17067	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_s8)))
	17068	svuint8_t svhistseg_s8(svint8_t, svint8_t);
	17069	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_m)))
	17070	svint8_t svhsub_n_s8_m(svbool_t, svint8_t, int8_t);
	17071	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_m)))
	17072	svint32_t svhsub_n_s32_m(svbool_t, svint32_t, int32_t);
	17073	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_m)))
	17074	svint64_t svhsub_n_s64_m(svbool_t, svint64_t, int64_t);
	17075	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_m)))
	17076	svint16_t svhsub_n_s16_m(svbool_t, svint16_t, int16_t);
	17077	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_x)))
	17078	svint8_t svhsub_n_s8_x(svbool_t, svint8_t, int8_t);
	17079	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_x)))
	17080	svint32_t svhsub_n_s32_x(svbool_t, svint32_t, int32_t);
	17081	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_x)))
	17082	svint64_t svhsub_n_s64_x(svbool_t, svint64_t, int64_t);
	17083	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_x)))
	17084	svint16_t svhsub_n_s16_x(svbool_t, svint16_t, int16_t);
	17085	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_z)))
	17086	svint8_t svhsub_n_s8_z(svbool_t, svint8_t, int8_t);
	17087	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_z)))
	17088	svint32_t svhsub_n_s32_z(svbool_t, svint32_t, int32_t);
	17089	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_z)))
	17090	svint64_t svhsub_n_s64_z(svbool_t, svint64_t, int64_t);
	17091	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_z)))
	17092	svint16_t svhsub_n_s16_z(svbool_t, svint16_t, int16_t);
	17093	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_m)))
	17094	svuint8_t svhsub_n_u8_m(svbool_t, svuint8_t, uint8_t);
	17095	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_m)))
	17096	svuint32_t svhsub_n_u32_m(svbool_t, svuint32_t, uint32_t);
	17097	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_m)))
	17098	svuint64_t svhsub_n_u64_m(svbool_t, svuint64_t, uint64_t);
	17099	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_m)))
	17100	svuint16_t svhsub_n_u16_m(svbool_t, svuint16_t, uint16_t);
	17101	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_x)))
	17102	svuint8_t svhsub_n_u8_x(svbool_t, svuint8_t, uint8_t);
	17103	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_x)))
	17104	svuint32_t svhsub_n_u32_x(svbool_t, svuint32_t, uint32_t);
	17105	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_x)))
	17106	svuint64_t svhsub_n_u64_x(svbool_t, svuint64_t, uint64_t);
	17107	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_x)))
	17108	svuint16_t svhsub_n_u16_x(svbool_t, svuint16_t, uint16_t);
	17109	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_z)))
	17110	svuint8_t svhsub_n_u8_z(svbool_t, svuint8_t, uint8_t);
	17111	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_z)))
	17112	svuint32_t svhsub_n_u32_z(svbool_t, svuint32_t, uint32_t);
	17113	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_z)))
	17114	svuint64_t svhsub_n_u64_z(svbool_t, svuint64_t, uint64_t);
	17115	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_z)))
	17116	svuint16_t svhsub_n_u16_z(svbool_t, svuint16_t, uint16_t);
	17117	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_m)))
	17118	svint8_t svhsub_s8_m(svbool_t, svint8_t, svint8_t);
	17119	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_m)))
	17120	svint32_t svhsub_s32_m(svbool_t, svint32_t, svint32_t);
	17121	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_m)))
	17122	svint64_t svhsub_s64_m(svbool_t, svint64_t, svint64_t);
	17123	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_m)))
	17124	svint16_t svhsub_s16_m(svbool_t, svint16_t, svint16_t);
	17125	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_x)))
	17126	svint8_t svhsub_s8_x(svbool_t, svint8_t, svint8_t);
	17127	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_x)))
	17128	svint32_t svhsub_s32_x(svbool_t, svint32_t, svint32_t);
	17129	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_x)))
	17130	svint64_t svhsub_s64_x(svbool_t, svint64_t, svint64_t);
	17131	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_x)))
	17132	svint16_t svhsub_s16_x(svbool_t, svint16_t, svint16_t);
	17133	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_z)))
	17134	svint8_t svhsub_s8_z(svbool_t, svint8_t, svint8_t);
	17135	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_z)))
	17136	svint32_t svhsub_s32_z(svbool_t, svint32_t, svint32_t);
	17137	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_z)))
	17138	svint64_t svhsub_s64_z(svbool_t, svint64_t, svint64_t);
	17139	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_z)))
	17140	svint16_t svhsub_s16_z(svbool_t, svint16_t, svint16_t);
	17141	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_m)))
	17142	svuint8_t svhsub_u8_m(svbool_t, svuint8_t, svuint8_t);
	17143	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_m)))
	17144	svuint32_t svhsub_u32_m(svbool_t, svuint32_t, svuint32_t);
	17145	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_m)))
	17146	svuint64_t svhsub_u64_m(svbool_t, svuint64_t, svuint64_t);
	17147	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_m)))
	17148	svuint16_t svhsub_u16_m(svbool_t, svuint16_t, svuint16_t);
	17149	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_x)))
	17150	svuint8_t svhsub_u8_x(svbool_t, svuint8_t, svuint8_t);
	17151	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_x)))
	17152	svuint32_t svhsub_u32_x(svbool_t, svuint32_t, svuint32_t);
	17153	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_x)))
	17154	svuint64_t svhsub_u64_x(svbool_t, svuint64_t, svuint64_t);
	17155	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_x)))
	17156	svuint16_t svhsub_u16_x(svbool_t, svuint16_t, svuint16_t);
	17157	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_z)))
	17158	svuint8_t svhsub_u8_z(svbool_t, svuint8_t, svuint8_t);
	17159	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_z)))
	17160	svuint32_t svhsub_u32_z(svbool_t, svuint32_t, svuint32_t);
	17161	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_z)))
	17162	svuint64_t svhsub_u64_z(svbool_t, svuint64_t, svuint64_t);
	17163	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_z)))
	17164	svuint16_t svhsub_u16_z(svbool_t, svuint16_t, svuint16_t);
	17165	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_m)))
	17166	svint8_t svhsubr_n_s8_m(svbool_t, svint8_t, int8_t);
	17167	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_m)))
	17168	svint32_t svhsubr_n_s32_m(svbool_t, svint32_t, int32_t);
	17169	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_m)))
	17170	svint64_t svhsubr_n_s64_m(svbool_t, svint64_t, int64_t);
	17171	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_m)))
	17172	svint16_t svhsubr_n_s16_m(svbool_t, svint16_t, int16_t);
	17173	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_x)))
	17174	svint8_t svhsubr_n_s8_x(svbool_t, svint8_t, int8_t);
	17175	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_x)))
	17176	svint32_t svhsubr_n_s32_x(svbool_t, svint32_t, int32_t);
	17177	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_x)))
	17178	svint64_t svhsubr_n_s64_x(svbool_t, svint64_t, int64_t);
	17179	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_x)))
	17180	svint16_t svhsubr_n_s16_x(svbool_t, svint16_t, int16_t);
	17181	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_z)))
	17182	svint8_t svhsubr_n_s8_z(svbool_t, svint8_t, int8_t);
	17183	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_z)))
	17184	svint32_t svhsubr_n_s32_z(svbool_t, svint32_t, int32_t);
	17185	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_z)))
	17186	svint64_t svhsubr_n_s64_z(svbool_t, svint64_t, int64_t);
	17187	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_z)))
	17188	svint16_t svhsubr_n_s16_z(svbool_t, svint16_t, int16_t);
	17189	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_m)))
	17190	svuint8_t svhsubr_n_u8_m(svbool_t, svuint8_t, uint8_t);
	17191	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_m)))
	17192	svuint32_t svhsubr_n_u32_m(svbool_t, svuint32_t, uint32_t);
	17193	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_m)))
	17194	svuint64_t svhsubr_n_u64_m(svbool_t, svuint64_t, uint64_t);
	17195	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_m)))
	17196	svuint16_t svhsubr_n_u16_m(svbool_t, svuint16_t, uint16_t);
	17197	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_x)))
	17198	svuint8_t svhsubr_n_u8_x(svbool_t, svuint8_t, uint8_t);
	17199	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_x)))
	17200	svuint32_t svhsubr_n_u32_x(svbool_t, svuint32_t, uint32_t);
	17201	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_x)))
	17202	svuint64_t svhsubr_n_u64_x(svbool_t, svuint64_t, uint64_t);
	17203	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_x)))
	17204	svuint16_t svhsubr_n_u16_x(svbool_t, svuint16_t, uint16_t);
	17205	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_z)))
	17206	svuint8_t svhsubr_n_u8_z(svbool_t, svuint8_t, uint8_t);
	17207	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_z)))
	17208	svuint32_t svhsubr_n_u32_z(svbool_t, svuint32_t, uint32_t);
	17209	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_z)))
	17210	svuint64_t svhsubr_n_u64_z(svbool_t, svuint64_t, uint64_t);
	17211	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_z)))
	17212	svuint16_t svhsubr_n_u16_z(svbool_t, svuint16_t, uint16_t);
	17213	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_m)))
	17214	svint8_t svhsubr_s8_m(svbool_t, svint8_t, svint8_t);
	17215	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_m)))
	17216	svint32_t svhsubr_s32_m(svbool_t, svint32_t, svint32_t);
	17217	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_m)))
	17218	svint64_t svhsubr_s64_m(svbool_t, svint64_t, svint64_t);
	17219	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_m)))
	17220	svint16_t svhsubr_s16_m(svbool_t, svint16_t, svint16_t);
	17221	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_x)))
	17222	svint8_t svhsubr_s8_x(svbool_t, svint8_t, svint8_t);
	17223	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_x)))
	17224	svint32_t svhsubr_s32_x(svbool_t, svint32_t, svint32_t);
	17225	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_x)))
	17226	svint64_t svhsubr_s64_x(svbool_t, svint64_t, svint64_t);
	17227	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_x)))
	17228	svint16_t svhsubr_s16_x(svbool_t, svint16_t, svint16_t);
	17229	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_z)))
	17230	svint8_t svhsubr_s8_z(svbool_t, svint8_t, svint8_t);
	17231	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_z)))
	17232	svint32_t svhsubr_s32_z(svbool_t, svint32_t, svint32_t);
	17233	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_z)))
	17234	svint64_t svhsubr_s64_z(svbool_t, svint64_t, svint64_t);
	17235	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_z)))
	17236	svint16_t svhsubr_s16_z(svbool_t, svint16_t, svint16_t);
	17237	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_m)))
	17238	svuint8_t svhsubr_u8_m(svbool_t, svuint8_t, svuint8_t);
	17239	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_m)))
	17240	svuint32_t svhsubr_u32_m(svbool_t, svuint32_t, svuint32_t);
	17241	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_m)))
	17242	svuint64_t svhsubr_u64_m(svbool_t, svuint64_t, svuint64_t);
	17243	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_m)))
	17244	svuint16_t svhsubr_u16_m(svbool_t, svuint16_t, svuint16_t);
	17245	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_x)))
	17246	svuint8_t svhsubr_u8_x(svbool_t, svuint8_t, svuint8_t);
	17247	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_x)))
	17248	svuint32_t svhsubr_u32_x(svbool_t, svuint32_t, svuint32_t);
	17249	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_x)))
	17250	svuint64_t svhsubr_u64_x(svbool_t, svuint64_t, svuint64_t);
	17251	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_x)))
	17252	svuint16_t svhsubr_u16_x(svbool_t, svuint16_t, svuint16_t);
	17253	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_z)))
	17254	svuint8_t svhsubr_u8_z(svbool_t, svuint8_t, svuint8_t);
	17255	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_z)))
	17256	svuint32_t svhsubr_u32_z(svbool_t, svuint32_t, svuint32_t);
	17257	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_z)))
	17258	svuint64_t svhsubr_u64_z(svbool_t, svuint64_t, svuint64_t);
	17259	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_z)))
	17260	svuint16_t svhsubr_u16_z(svbool_t, svuint16_t, svuint16_t);
	17261	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_u32)))
	17262	svuint32_t svldnt1_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
	17263	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_u64)))
	17264	svuint64_t svldnt1_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
	17265	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_f64)))
	17266	svfloat64_t svldnt1_gather_u64base_index_f64(svbool_t, svuint64_t, int64_t);
	17267	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_f32)))
	17268	svfloat32_t svldnt1_gather_u32base_index_f32(svbool_t, svuint32_t, int64_t);
	17269	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_s32)))
	17270	svint32_t svldnt1_gather_u32base_index_s32(svbool_t, svuint32_t, int64_t);
	17271	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_s64)))
	17272	svint64_t svldnt1_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
	17273	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_u32)))
	17274	svuint32_t svldnt1_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
	17275	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_u64)))
	17276	svuint64_t svldnt1_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17277	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_f64)))
	17278	svfloat64_t svldnt1_gather_u64base_offset_f64(svbool_t, svuint64_t, int64_t);
	17279	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_f32)))
	17280	svfloat32_t svldnt1_gather_u32base_offset_f32(svbool_t, svuint32_t, int64_t);
	17281	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_s32)))
	17282	svint32_t svldnt1_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
	17283	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_s64)))
	17284	svint64_t svldnt1_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17285	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_u32)))
	17286	svuint32_t svldnt1_gather_u32base_u32(svbool_t, svuint32_t);
	17287	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_u64)))
	17288	svuint64_t svldnt1_gather_u64base_u64(svbool_t, svuint64_t);
	17289	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_f64)))
	17290	svfloat64_t svldnt1_gather_u64base_f64(svbool_t, svuint64_t);
	17291	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_f32)))
	17292	svfloat32_t svldnt1_gather_u32base_f32(svbool_t, svuint32_t);
	17293	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_s32)))
	17294	svint32_t svldnt1_gather_u32base_s32(svbool_t, svuint32_t);
	17295	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_s64)))
	17296	svint64_t svldnt1_gather_u64base_s64(svbool_t, svuint64_t);
	17297	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_u64)))
	17298	svuint64_t svldnt1_gather_s64index_u64(svbool_t, uint64_t const *, svint64_t);
	17299	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_f64)))
	17300	svfloat64_t svldnt1_gather_s64index_f64(svbool_t, float64_t const *, svint64_t);
	17301	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_s64)))
	17302	svint64_t svldnt1_gather_s64index_s64(svbool_t, int64_t const *, svint64_t);
	17303	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_u64)))
	17304	svuint64_t svldnt1_gather_u64index_u64(svbool_t, uint64_t const *, svuint64_t);
	17305	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_f64)))
	17306	svfloat64_t svldnt1_gather_u64index_f64(svbool_t, float64_t const *, svuint64_t);
	17307	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_s64)))
	17308	svint64_t svldnt1_gather_u64index_s64(svbool_t, int64_t const *, svuint64_t);
	17309	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_u32)))
	17310	svuint32_t svldnt1_gather_u32offset_u32(svbool_t, uint32_t const *, svuint32_t);
	17311	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_f32)))
	17312	svfloat32_t svldnt1_gather_u32offset_f32(svbool_t, float32_t const *, svuint32_t);
	17313	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_s32)))
	17314	svint32_t svldnt1_gather_u32offset_s32(svbool_t, int32_t const *, svuint32_t);
	17315	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_u64)))
	17316	svuint64_t svldnt1_gather_s64offset_u64(svbool_t, uint64_t const *, svint64_t);
	17317	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_f64)))
	17318	svfloat64_t svldnt1_gather_s64offset_f64(svbool_t, float64_t const *, svint64_t);
	17319	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_s64)))
	17320	svint64_t svldnt1_gather_s64offset_s64(svbool_t, int64_t const *, svint64_t);
	17321	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_u64)))
	17322	svuint64_t svldnt1_gather_u64offset_u64(svbool_t, uint64_t const *, svuint64_t);
	17323	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_f64)))
	17324	svfloat64_t svldnt1_gather_u64offset_f64(svbool_t, float64_t const *, svuint64_t);
	17325	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_s64)))
	17326	svint64_t svldnt1_gather_u64offset_s64(svbool_t, int64_t const *, svuint64_t);
	17327	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_u32)))
	17328	svuint32_t svldnt1sb_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
	17329	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_u64)))
	17330	svuint64_t svldnt1sb_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17331	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_s32)))
	17332	svint32_t svldnt1sb_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
	17333	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_s64)))
	17334	svint64_t svldnt1sb_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17335	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_u32)))
	17336	svuint32_t svldnt1sb_gather_u32base_u32(svbool_t, svuint32_t);
	17337	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_u64)))
	17338	svuint64_t svldnt1sb_gather_u64base_u64(svbool_t, svuint64_t);
	17339	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_s32)))
	17340	svint32_t svldnt1sb_gather_u32base_s32(svbool_t, svuint32_t);
	17341	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_s64)))
	17342	svint64_t svldnt1sb_gather_u64base_s64(svbool_t, svuint64_t);
	17343	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_u32)))
	17344	svuint32_t svldnt1sb_gather_u32offset_u32(svbool_t, int8_t const *, svuint32_t);
	17345	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_s32)))
	17346	svint32_t svldnt1sb_gather_u32offset_s32(svbool_t, int8_t const *, svuint32_t);
	17347	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_u64)))
	17348	svuint64_t svldnt1sb_gather_s64offset_u64(svbool_t, int8_t const *, svint64_t);
	17349	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_s64)))
	17350	svint64_t svldnt1sb_gather_s64offset_s64(svbool_t, int8_t const *, svint64_t);
	17351	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_u64)))
	17352	svuint64_t svldnt1sb_gather_u64offset_u64(svbool_t, int8_t const *, svuint64_t);
	17353	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_s64)))
	17354	svint64_t svldnt1sb_gather_u64offset_s64(svbool_t, int8_t const *, svuint64_t);
	17355	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_u32)))
	17356	svuint32_t svldnt1sh_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
	17357	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_u64)))
	17358	svuint64_t svldnt1sh_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
	17359	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_s32)))
	17360	svint32_t svldnt1sh_gather_u32base_index_s32(svbool_t, svuint32_t, int64_t);
	17361	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_s64)))
	17362	svint64_t svldnt1sh_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
	17363	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_u32)))
	17364	svuint32_t svldnt1sh_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
	17365	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_u64)))
	17366	svuint64_t svldnt1sh_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17367	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_s32)))
	17368	svint32_t svldnt1sh_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
	17369	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_s64)))
	17370	svint64_t svldnt1sh_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17371	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_u32)))
	17372	svuint32_t svldnt1sh_gather_u32base_u32(svbool_t, svuint32_t);
	17373	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_u64)))
	17374	svuint64_t svldnt1sh_gather_u64base_u64(svbool_t, svuint64_t);
	17375	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_s32)))
	17376	svint32_t svldnt1sh_gather_u32base_s32(svbool_t, svuint32_t);
	17377	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_s64)))
	17378	svint64_t svldnt1sh_gather_u64base_s64(svbool_t, svuint64_t);
	17379	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_u64)))
	17380	svuint64_t svldnt1sh_gather_s64index_u64(svbool_t, int16_t const *, svint64_t);
	17381	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_s64)))
	17382	svint64_t svldnt1sh_gather_s64index_s64(svbool_t, int16_t const *, svint64_t);
	17383	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_u64)))
	17384	svuint64_t svldnt1sh_gather_u64index_u64(svbool_t, int16_t const *, svuint64_t);
	17385	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_s64)))
	17386	svint64_t svldnt1sh_gather_u64index_s64(svbool_t, int16_t const *, svuint64_t);
	17387	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_u32)))
	17388	svuint32_t svldnt1sh_gather_u32offset_u32(svbool_t, int16_t const *, svuint32_t);
	17389	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_s32)))
	17390	svint32_t svldnt1sh_gather_u32offset_s32(svbool_t, int16_t const *, svuint32_t);
	17391	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_u64)))
	17392	svuint64_t svldnt1sh_gather_s64offset_u64(svbool_t, int16_t const *, svint64_t);
	17393	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_s64)))
	17394	svint64_t svldnt1sh_gather_s64offset_s64(svbool_t, int16_t const *, svint64_t);
	17395	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_u64)))
	17396	svuint64_t svldnt1sh_gather_u64offset_u64(svbool_t, int16_t const *, svuint64_t);
	17397	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_s64)))
	17398	svint64_t svldnt1sh_gather_u64offset_s64(svbool_t, int16_t const *, svuint64_t);
	17399	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_u64)))
	17400	svuint64_t svldnt1sw_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
	17401	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_s64)))
	17402	svint64_t svldnt1sw_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
	17403	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_u64)))
	17404	svuint64_t svldnt1sw_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17405	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_s64)))
	17406	svint64_t svldnt1sw_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17407	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_u64)))
	17408	svuint64_t svldnt1sw_gather_u64base_u64(svbool_t, svuint64_t);
	17409	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_s64)))
	17410	svint64_t svldnt1sw_gather_u64base_s64(svbool_t, svuint64_t);
	17411	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_u64)))
	17412	svuint64_t svldnt1sw_gather_s64index_u64(svbool_t, int32_t const *, svint64_t);
	17413	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_s64)))
	17414	svint64_t svldnt1sw_gather_s64index_s64(svbool_t, int32_t const *, svint64_t);
	17415	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_u64)))
	17416	svuint64_t svldnt1sw_gather_u64index_u64(svbool_t, int32_t const *, svuint64_t);
	17417	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_s64)))
	17418	svint64_t svldnt1sw_gather_u64index_s64(svbool_t, int32_t const *, svuint64_t);
	17419	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_u64)))
	17420	svuint64_t svldnt1sw_gather_s64offset_u64(svbool_t, int32_t const *, svint64_t);
	17421	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_s64)))
	17422	svint64_t svldnt1sw_gather_s64offset_s64(svbool_t, int32_t const *, svint64_t);
	17423	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_u64)))
	17424	svuint64_t svldnt1sw_gather_u64offset_u64(svbool_t, int32_t const *, svuint64_t);
	17425	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_s64)))
	17426	svint64_t svldnt1sw_gather_u64offset_s64(svbool_t, int32_t const *, svuint64_t);
	17427	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_u32)))
	17428	svuint32_t svldnt1ub_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
	17429	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_u64)))
	17430	svuint64_t svldnt1ub_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17431	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_s32)))
	17432	svint32_t svldnt1ub_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
	17433	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_s64)))
	17434	svint64_t svldnt1ub_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17435	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_u32)))
	17436	svuint32_t svldnt1ub_gather_u32base_u32(svbool_t, svuint32_t);
	17437	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_u64)))
	17438	svuint64_t svldnt1ub_gather_u64base_u64(svbool_t, svuint64_t);
	17439	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_s32)))
	17440	svint32_t svldnt1ub_gather_u32base_s32(svbool_t, svuint32_t);
	17441	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_s64)))
	17442	svint64_t svldnt1ub_gather_u64base_s64(svbool_t, svuint64_t);
	17443	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_u32)))
	17444	svuint32_t svldnt1ub_gather_u32offset_u32(svbool_t, uint8_t const *, svuint32_t);
	17445	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_s32)))
	17446	svint32_t svldnt1ub_gather_u32offset_s32(svbool_t, uint8_t const *, svuint32_t);
	17447	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_u64)))
	17448	svuint64_t svldnt1ub_gather_s64offset_u64(svbool_t, uint8_t const *, svint64_t);
	17449	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_s64)))
	17450	svint64_t svldnt1ub_gather_s64offset_s64(svbool_t, uint8_t const *, svint64_t);
	17451	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_u64)))
	17452	svuint64_t svldnt1ub_gather_u64offset_u64(svbool_t, uint8_t const *, svuint64_t);
	17453	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_s64)))
	17454	svint64_t svldnt1ub_gather_u64offset_s64(svbool_t, uint8_t const *, svuint64_t);
	17455	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_u32)))
	17456	svuint32_t svldnt1uh_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
	17457	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_u64)))
	17458	svuint64_t svldnt1uh_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
	17459	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_s32)))
	17460	svint32_t svldnt1uh_gather_u32base_index_s32(svbool_t, svuint32_t, int64_t);
	17461	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_s64)))
	17462	svint64_t svldnt1uh_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
	17463	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_u32)))
	17464	svuint32_t svldnt1uh_gather_u32base_offset_u32(svbool_t, svuint32_t, int64_t);
	17465	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_u64)))
	17466	svuint64_t svldnt1uh_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17467	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_s32)))
	17468	svint32_t svldnt1uh_gather_u32base_offset_s32(svbool_t, svuint32_t, int64_t);
	17469	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_s64)))
	17470	svint64_t svldnt1uh_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17471	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_u32)))
	17472	svuint32_t svldnt1uh_gather_u32base_u32(svbool_t, svuint32_t);
	17473	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_u64)))
	17474	svuint64_t svldnt1uh_gather_u64base_u64(svbool_t, svuint64_t);
	17475	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_s32)))
	17476	svint32_t svldnt1uh_gather_u32base_s32(svbool_t, svuint32_t);
	17477	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_s64)))
	17478	svint64_t svldnt1uh_gather_u64base_s64(svbool_t, svuint64_t);
	17479	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_u64)))
	17480	svuint64_t svldnt1uh_gather_s64index_u64(svbool_t, uint16_t const *, svint64_t);
	17481	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_s64)))
	17482	svint64_t svldnt1uh_gather_s64index_s64(svbool_t, uint16_t const *, svint64_t);
	17483	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_u64)))
	17484	svuint64_t svldnt1uh_gather_u64index_u64(svbool_t, uint16_t const *, svuint64_t);
	17485	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_s64)))
	17486	svint64_t svldnt1uh_gather_u64index_s64(svbool_t, uint16_t const *, svuint64_t);
	17487	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_u32)))
	17488	svuint32_t svldnt1uh_gather_u32offset_u32(svbool_t, uint16_t const *, svuint32_t);
	17489	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_s32)))
	17490	svint32_t svldnt1uh_gather_u32offset_s32(svbool_t, uint16_t const *, svuint32_t);
	17491	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_u64)))
	17492	svuint64_t svldnt1uh_gather_s64offset_u64(svbool_t, uint16_t const *, svint64_t);
	17493	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_s64)))
	17494	svint64_t svldnt1uh_gather_s64offset_s64(svbool_t, uint16_t const *, svint64_t);
	17495	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_u64)))
	17496	svuint64_t svldnt1uh_gather_u64offset_u64(svbool_t, uint16_t const *, svuint64_t);
	17497	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_s64)))
	17498	svint64_t svldnt1uh_gather_u64offset_s64(svbool_t, uint16_t const *, svuint64_t);
	17499	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_u64)))
	17500	svuint64_t svldnt1uw_gather_u64base_index_u64(svbool_t, svuint64_t, int64_t);
	17501	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_s64)))
	17502	svint64_t svldnt1uw_gather_u64base_index_s64(svbool_t, svuint64_t, int64_t);
	17503	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_u64)))
	17504	svuint64_t svldnt1uw_gather_u64base_offset_u64(svbool_t, svuint64_t, int64_t);
	17505	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_s64)))
	17506	svint64_t svldnt1uw_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
	17507	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_u64)))
	17508	svuint64_t svldnt1uw_gather_u64base_u64(svbool_t, svuint64_t);
	17509	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_s64)))
	17510	svint64_t svldnt1uw_gather_u64base_s64(svbool_t, svuint64_t);
	17511	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_u64)))
	17512	svuint64_t svldnt1uw_gather_s64index_u64(svbool_t, uint32_t const *, svint64_t);
	17513	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_s64)))
	17514	svint64_t svldnt1uw_gather_s64index_s64(svbool_t, uint32_t const *, svint64_t);
	17515	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_u64)))
	17516	svuint64_t svldnt1uw_gather_u64index_u64(svbool_t, uint32_t const *, svuint64_t);
	17517	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_s64)))
	17518	svint64_t svldnt1uw_gather_u64index_s64(svbool_t, uint32_t const *, svuint64_t);
	17519	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_u64)))
	17520	svuint64_t svldnt1uw_gather_s64offset_u64(svbool_t, uint32_t const *, svint64_t);
	17521	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_s64)))
	17522	svint64_t svldnt1uw_gather_s64offset_s64(svbool_t, uint32_t const *, svint64_t);
	17523	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_u64)))
	17524	svuint64_t svldnt1uw_gather_u64offset_u64(svbool_t, uint32_t const *, svuint64_t);
	17525	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_s64)))
	17526	svint64_t svldnt1uw_gather_u64offset_s64(svbool_t, uint32_t const *, svuint64_t);
	17527	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_m)))
	17528	svint64_t svlogb_f64_m(svint64_t, svbool_t, svfloat64_t);
	17529	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_m)))
	17530	svint32_t svlogb_f32_m(svint32_t, svbool_t, svfloat32_t);
	17531	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_m)))
	17532	svint16_t svlogb_f16_m(svint16_t, svbool_t, svfloat16_t);
	17533	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_x)))
	17534	svint64_t svlogb_f64_x(svbool_t, svfloat64_t);
	17535	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_x)))
	17536	svint32_t svlogb_f32_x(svbool_t, svfloat32_t);
	17537	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_x)))
	17538	svint16_t svlogb_f16_x(svbool_t, svfloat16_t);
	17539	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_z)))
	17540	svint64_t svlogb_f64_z(svbool_t, svfloat64_t);
	17541	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_z)))
	17542	svint32_t svlogb_f32_z(svbool_t, svfloat32_t);
	17543	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_z)))
	17544	svint16_t svlogb_f16_z(svbool_t, svfloat16_t);
	17545	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u8)))
	17546	svbool_t svmatch_u8(svbool_t, svuint8_t, svuint8_t);
	17547	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u16)))
	17548	svbool_t svmatch_u16(svbool_t, svuint16_t, svuint16_t);
	17549	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s8)))
	17550	svbool_t svmatch_s8(svbool_t, svint8_t, svint8_t);
	17551	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s16)))
	17552	svbool_t svmatch_s16(svbool_t, svint16_t, svint16_t);
	17553	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_m)))
	17554	svfloat64_t svmaxnmp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
	17555	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_m)))
	17556	svfloat32_t svmaxnmp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
	17557	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_m)))
	17558	svfloat16_t svmaxnmp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
	17559	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_x)))
	17560	svfloat64_t svmaxnmp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
	17561	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_x)))
	17562	svfloat32_t svmaxnmp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
	17563	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_x)))
	17564	svfloat16_t svmaxnmp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
	17565	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_m)))
	17566	svfloat64_t svmaxp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
	17567	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_m)))
	17568	svfloat32_t svmaxp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
	17569	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_m)))
	17570	svfloat16_t svmaxp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
	17571	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_x)))
	17572	svfloat64_t svmaxp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
	17573	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_x)))
	17574	svfloat32_t svmaxp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
	17575	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_x)))
	17576	svfloat16_t svmaxp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
	17577	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_m)))
	17578	svint8_t svmaxp_s8_m(svbool_t, svint8_t, svint8_t);
	17579	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_m)))
	17580	svint32_t svmaxp_s32_m(svbool_t, svint32_t, svint32_t);
	17581	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_m)))
	17582	svint64_t svmaxp_s64_m(svbool_t, svint64_t, svint64_t);
	17583	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_m)))
	17584	svint16_t svmaxp_s16_m(svbool_t, svint16_t, svint16_t);
	17585	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_x)))
	17586	svint8_t svmaxp_s8_x(svbool_t, svint8_t, svint8_t);
	17587	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_x)))
	17588	svint32_t svmaxp_s32_x(svbool_t, svint32_t, svint32_t);
	17589	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_x)))
	17590	svint64_t svmaxp_s64_x(svbool_t, svint64_t, svint64_t);
	17591	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_x)))
	17592	svint16_t svmaxp_s16_x(svbool_t, svint16_t, svint16_t);
	17593	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_m)))
	17594	svuint8_t svmaxp_u8_m(svbool_t, svuint8_t, svuint8_t);
	17595	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_m)))
	17596	svuint32_t svmaxp_u32_m(svbool_t, svuint32_t, svuint32_t);
	17597	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_m)))
	17598	svuint64_t svmaxp_u64_m(svbool_t, svuint64_t, svuint64_t);
	17599	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_m)))
	17600	svuint16_t svmaxp_u16_m(svbool_t, svuint16_t, svuint16_t);
	17601	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_x)))
	17602	svuint8_t svmaxp_u8_x(svbool_t, svuint8_t, svuint8_t);
	17603	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_x)))
	17604	svuint32_t svmaxp_u32_x(svbool_t, svuint32_t, svuint32_t);
	17605	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_x)))
	17606	svuint64_t svmaxp_u64_x(svbool_t, svuint64_t, svuint64_t);
	17607	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_x)))
	17608	svuint16_t svmaxp_u16_x(svbool_t, svuint16_t, svuint16_t);
	17609	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_m)))
	17610	svfloat64_t svminnmp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
	17611	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_m)))
	17612	svfloat32_t svminnmp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
	17613	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_m)))
	17614	svfloat16_t svminnmp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
	17615	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_x)))
	17616	svfloat64_t svminnmp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
	17617	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_x)))
	17618	svfloat32_t svminnmp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
	17619	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_x)))
	17620	svfloat16_t svminnmp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
	17621	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_m)))
	17622	svfloat64_t svminp_f64_m(svbool_t, svfloat64_t, svfloat64_t);
	17623	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_m)))
	17624	svfloat32_t svminp_f32_m(svbool_t, svfloat32_t, svfloat32_t);
	17625	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_m)))
	17626	svfloat16_t svminp_f16_m(svbool_t, svfloat16_t, svfloat16_t);
	17627	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_x)))
	17628	svfloat64_t svminp_f64_x(svbool_t, svfloat64_t, svfloat64_t);
	17629	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_x)))
	17630	svfloat32_t svminp_f32_x(svbool_t, svfloat32_t, svfloat32_t);
	17631	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_x)))
	17632	svfloat16_t svminp_f16_x(svbool_t, svfloat16_t, svfloat16_t);
	17633	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_m)))
	17634	svint8_t svminp_s8_m(svbool_t, svint8_t, svint8_t);
	17635	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_m)))
	17636	svint32_t svminp_s32_m(svbool_t, svint32_t, svint32_t);
	17637	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_m)))
	17638	svint64_t svminp_s64_m(svbool_t, svint64_t, svint64_t);
	17639	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_m)))
	17640	svint16_t svminp_s16_m(svbool_t, svint16_t, svint16_t);
	17641	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_x)))
	17642	svint8_t svminp_s8_x(svbool_t, svint8_t, svint8_t);
	17643	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_x)))
	17644	svint32_t svminp_s32_x(svbool_t, svint32_t, svint32_t);
	17645	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_x)))
	17646	svint64_t svminp_s64_x(svbool_t, svint64_t, svint64_t);
	17647	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_x)))
	17648	svint16_t svminp_s16_x(svbool_t, svint16_t, svint16_t);
	17649	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_m)))
	17650	svuint8_t svminp_u8_m(svbool_t, svuint8_t, svuint8_t);
	17651	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_m)))
	17652	svuint32_t svminp_u32_m(svbool_t, svuint32_t, svuint32_t);
	17653	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_m)))
	17654	svuint64_t svminp_u64_m(svbool_t, svuint64_t, svuint64_t);
	17655	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_m)))
	17656	svuint16_t svminp_u16_m(svbool_t, svuint16_t, svuint16_t);
	17657	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_x)))
	17658	svuint8_t svminp_u8_x(svbool_t, svuint8_t, svuint8_t);
	17659	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_x)))
	17660	svuint32_t svminp_u32_x(svbool_t, svuint32_t, svuint32_t);
	17661	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_x)))
	17662	svuint64_t svminp_u64_x(svbool_t, svuint64_t, svuint64_t);
	17663	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_x)))
	17664	svuint16_t svminp_u16_x(svbool_t, svuint16_t, svuint16_t);
	17665	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u32)))
	17666	svuint32_t svmla_lane_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t);
	17667	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u64)))
	17668	svuint64_t svmla_lane_u64(svuint64_t, svuint64_t, svuint64_t, uint64_t);
	17669	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u16)))
	17670	svuint16_t svmla_lane_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t);
	17671	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s32)))
	17672	svint32_t svmla_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
	17673	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s64)))
	17674	svint64_t svmla_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
	17675	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s16)))
	17676	svint16_t svmla_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
	17677	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_f32)))
	17678	svfloat32_t svmlalb_n_f32(svfloat32_t, svfloat16_t, float16_t);
	17679	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s32)))
	17680	svint32_t svmlalb_n_s32(svint32_t, svint16_t, int16_t);
	17681	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s64)))
	17682	svint64_t svmlalb_n_s64(svint64_t, svint32_t, int32_t);
	17683	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s16)))
	17684	svint16_t svmlalb_n_s16(svint16_t, svint8_t, int8_t);
	17685	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u32)))
	17686	svuint32_t svmlalb_n_u32(svuint32_t, svuint16_t, uint16_t);
	17687	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u64)))
	17688	svuint64_t svmlalb_n_u64(svuint64_t, svuint32_t, uint32_t);
	17689	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u16)))
	17690	svuint16_t svmlalb_n_u16(svuint16_t, svuint8_t, uint8_t);
	17691	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_f32)))
	17692	svfloat32_t svmlalb_f32(svfloat32_t, svfloat16_t, svfloat16_t);
	17693	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s32)))
	17694	svint32_t svmlalb_s32(svint32_t, svint16_t, svint16_t);
	17695	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s64)))
	17696	svint64_t svmlalb_s64(svint64_t, svint32_t, svint32_t);
	17697	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s16)))
	17698	svint16_t svmlalb_s16(svint16_t, svint8_t, svint8_t);
	17699	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u32)))
	17700	svuint32_t svmlalb_u32(svuint32_t, svuint16_t, svuint16_t);
	17701	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u64)))
	17702	svuint64_t svmlalb_u64(svuint64_t, svuint32_t, svuint32_t);
	17703	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u16)))
	17704	svuint16_t svmlalb_u16(svuint16_t, svuint8_t, svuint8_t);
	17705	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_f32)))
	17706	svfloat32_t svmlalb_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	17707	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s32)))
	17708	svint32_t svmlalb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	17709	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s64)))
	17710	svint64_t svmlalb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	17711	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u32)))
	17712	svuint32_t svmlalb_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	17713	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u64)))
	17714	svuint64_t svmlalb_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	17715	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_f32)))
	17716	svfloat32_t svmlalt_n_f32(svfloat32_t, svfloat16_t, float16_t);
	17717	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s32)))
	17718	svint32_t svmlalt_n_s32(svint32_t, svint16_t, int16_t);
	17719	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s64)))
	17720	svint64_t svmlalt_n_s64(svint64_t, svint32_t, int32_t);
	17721	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s16)))
	17722	svint16_t svmlalt_n_s16(svint16_t, svint8_t, int8_t);
	17723	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u32)))
	17724	svuint32_t svmlalt_n_u32(svuint32_t, svuint16_t, uint16_t);
	17725	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u64)))
	17726	svuint64_t svmlalt_n_u64(svuint64_t, svuint32_t, uint32_t);
	17727	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u16)))
	17728	svuint16_t svmlalt_n_u16(svuint16_t, svuint8_t, uint8_t);
	17729	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_f32)))
	17730	svfloat32_t svmlalt_f32(svfloat32_t, svfloat16_t, svfloat16_t);
	17731	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s32)))
	17732	svint32_t svmlalt_s32(svint32_t, svint16_t, svint16_t);
	17733	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s64)))
	17734	svint64_t svmlalt_s64(svint64_t, svint32_t, svint32_t);
	17735	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s16)))
	17736	svint16_t svmlalt_s16(svint16_t, svint8_t, svint8_t);
	17737	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u32)))
	17738	svuint32_t svmlalt_u32(svuint32_t, svuint16_t, svuint16_t);
	17739	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u64)))
	17740	svuint64_t svmlalt_u64(svuint64_t, svuint32_t, svuint32_t);
	17741	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u16)))
	17742	svuint16_t svmlalt_u16(svuint16_t, svuint8_t, svuint8_t);
	17743	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_f32)))
	17744	svfloat32_t svmlalt_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	17745	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s32)))
	17746	svint32_t svmlalt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	17747	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s64)))
	17748	svint64_t svmlalt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	17749	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u32)))
	17750	svuint32_t svmlalt_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	17751	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u64)))
	17752	svuint64_t svmlalt_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	17753	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u32)))
	17754	svuint32_t svmls_lane_u32(svuint32_t, svuint32_t, svuint32_t, uint64_t);
	17755	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u64)))
	17756	svuint64_t svmls_lane_u64(svuint64_t, svuint64_t, svuint64_t, uint64_t);
	17757	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u16)))
	17758	svuint16_t svmls_lane_u16(svuint16_t, svuint16_t, svuint16_t, uint64_t);
	17759	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s32)))
	17760	svint32_t svmls_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
	17761	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s64)))
	17762	svint64_t svmls_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
	17763	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s16)))
	17764	svint16_t svmls_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
	17765	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_f32)))
	17766	svfloat32_t svmlslb_n_f32(svfloat32_t, svfloat16_t, float16_t);
	17767	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s32)))
	17768	svint32_t svmlslb_n_s32(svint32_t, svint16_t, int16_t);
	17769	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s64)))
	17770	svint64_t svmlslb_n_s64(svint64_t, svint32_t, int32_t);
	17771	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s16)))
	17772	svint16_t svmlslb_n_s16(svint16_t, svint8_t, int8_t);
	17773	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u32)))
	17774	svuint32_t svmlslb_n_u32(svuint32_t, svuint16_t, uint16_t);
	17775	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u64)))
	17776	svuint64_t svmlslb_n_u64(svuint64_t, svuint32_t, uint32_t);
	17777	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u16)))
	17778	svuint16_t svmlslb_n_u16(svuint16_t, svuint8_t, uint8_t);
	17779	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_f32)))
	17780	svfloat32_t svmlslb_f32(svfloat32_t, svfloat16_t, svfloat16_t);
	17781	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s32)))
	17782	svint32_t svmlslb_s32(svint32_t, svint16_t, svint16_t);
	17783	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s64)))
	17784	svint64_t svmlslb_s64(svint64_t, svint32_t, svint32_t);
	17785	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s16)))
	17786	svint16_t svmlslb_s16(svint16_t, svint8_t, svint8_t);
	17787	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u32)))
	17788	svuint32_t svmlslb_u32(svuint32_t, svuint16_t, svuint16_t);
	17789	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u64)))
	17790	svuint64_t svmlslb_u64(svuint64_t, svuint32_t, svuint32_t);
	17791	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u16)))
	17792	svuint16_t svmlslb_u16(svuint16_t, svuint8_t, svuint8_t);
	17793	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_f32)))
	17794	svfloat32_t svmlslb_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	17795	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s32)))
	17796	svint32_t svmlslb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	17797	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s64)))
	17798	svint64_t svmlslb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	17799	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u32)))
	17800	svuint32_t svmlslb_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	17801	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u64)))
	17802	svuint64_t svmlslb_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	17803	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_f32)))
	17804	svfloat32_t svmlslt_n_f32(svfloat32_t, svfloat16_t, float16_t);
	17805	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s32)))
	17806	svint32_t svmlslt_n_s32(svint32_t, svint16_t, int16_t);
	17807	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s64)))
	17808	svint64_t svmlslt_n_s64(svint64_t, svint32_t, int32_t);
	17809	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s16)))
	17810	svint16_t svmlslt_n_s16(svint16_t, svint8_t, int8_t);
	17811	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u32)))
	17812	svuint32_t svmlslt_n_u32(svuint32_t, svuint16_t, uint16_t);
	17813	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u64)))
	17814	svuint64_t svmlslt_n_u64(svuint64_t, svuint32_t, uint32_t);
	17815	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u16)))
	17816	svuint16_t svmlslt_n_u16(svuint16_t, svuint8_t, uint8_t);
	17817	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_f32)))
	17818	svfloat32_t svmlslt_f32(svfloat32_t, svfloat16_t, svfloat16_t);
	17819	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s32)))
	17820	svint32_t svmlslt_s32(svint32_t, svint16_t, svint16_t);
	17821	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s64)))
	17822	svint64_t svmlslt_s64(svint64_t, svint32_t, svint32_t);
	17823	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s16)))
	17824	svint16_t svmlslt_s16(svint16_t, svint8_t, svint8_t);
	17825	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u32)))
	17826	svuint32_t svmlslt_u32(svuint32_t, svuint16_t, svuint16_t);
	17827	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u64)))
	17828	svuint64_t svmlslt_u64(svuint64_t, svuint32_t, svuint32_t);
	17829	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u16)))
	17830	svuint16_t svmlslt_u16(svuint16_t, svuint8_t, svuint8_t);
	17831	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_f32)))
	17832	svfloat32_t svmlslt_lane_f32(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	17833	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s32)))
	17834	svint32_t svmlslt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	17835	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s64)))
	17836	svint64_t svmlslt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	17837	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u32)))
	17838	svuint32_t svmlslt_lane_u32(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	17839	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u64)))
	17840	svuint64_t svmlslt_lane_u64(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	17841	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s32)))
	17842	svint32_t svmovlb_s32(svint16_t);
	17843	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s64)))
	17844	svint64_t svmovlb_s64(svint32_t);
	17845	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s16)))
	17846	svint16_t svmovlb_s16(svint8_t);
	17847	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u32)))
	17848	svuint32_t svmovlb_u32(svuint16_t);
	17849	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u64)))
	17850	svuint64_t svmovlb_u64(svuint32_t);
	17851	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u16)))
	17852	svuint16_t svmovlb_u16(svuint8_t);
	17853	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s32)))
	17854	svint32_t svmovlt_s32(svint16_t);
	17855	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s64)))
	17856	svint64_t svmovlt_s64(svint32_t);
	17857	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s16)))
	17858	svint16_t svmovlt_s16(svint8_t);
	17859	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u32)))
	17860	svuint32_t svmovlt_u32(svuint16_t);
	17861	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u64)))
	17862	svuint64_t svmovlt_u64(svuint32_t);
	17863	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u16)))
	17864	svuint16_t svmovlt_u16(svuint8_t);
	17865	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u32)))
	17866	svuint32_t svmul_lane_u32(svuint32_t, svuint32_t, uint64_t);
	17867	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u64)))
	17868	svuint64_t svmul_lane_u64(svuint64_t, svuint64_t, uint64_t);
	17869	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u16)))
	17870	svuint16_t svmul_lane_u16(svuint16_t, svuint16_t, uint64_t);
	17871	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s32)))
	17872	svint32_t svmul_lane_s32(svint32_t, svint32_t, uint64_t);
	17873	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s64)))
	17874	svint64_t svmul_lane_s64(svint64_t, svint64_t, uint64_t);
	17875	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s16)))
	17876	svint16_t svmul_lane_s16(svint16_t, svint16_t, uint64_t);
	17877	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s32)))
	17878	svint32_t svmullb_n_s32(svint16_t, int16_t);
	17879	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s64)))
	17880	svint64_t svmullb_n_s64(svint32_t, int32_t);
	17881	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s16)))
	17882	svint16_t svmullb_n_s16(svint8_t, int8_t);
	17883	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u32)))
	17884	svuint32_t svmullb_n_u32(svuint16_t, uint16_t);
	17885	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u64)))
	17886	svuint64_t svmullb_n_u64(svuint32_t, uint32_t);
	17887	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u16)))
	17888	svuint16_t svmullb_n_u16(svuint8_t, uint8_t);
	17889	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s32)))
	17890	svint32_t svmullb_s32(svint16_t, svint16_t);
	17891	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s64)))
	17892	svint64_t svmullb_s64(svint32_t, svint32_t);
	17893	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s16)))
	17894	svint16_t svmullb_s16(svint8_t, svint8_t);
	17895	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u32)))
	17896	svuint32_t svmullb_u32(svuint16_t, svuint16_t);
	17897	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u64)))
	17898	svuint64_t svmullb_u64(svuint32_t, svuint32_t);
	17899	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u16)))
	17900	svuint16_t svmullb_u16(svuint8_t, svuint8_t);
	17901	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s32)))
	17902	svint32_t svmullb_lane_s32(svint16_t, svint16_t, uint64_t);
	17903	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s64)))
	17904	svint64_t svmullb_lane_s64(svint32_t, svint32_t, uint64_t);
	17905	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u32)))
	17906	svuint32_t svmullb_lane_u32(svuint16_t, svuint16_t, uint64_t);
	17907	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u64)))
	17908	svuint64_t svmullb_lane_u64(svuint32_t, svuint32_t, uint64_t);
	17909	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s32)))
	17910	svint32_t svmullt_n_s32(svint16_t, int16_t);
	17911	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s64)))
	17912	svint64_t svmullt_n_s64(svint32_t, int32_t);
	17913	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s16)))
	17914	svint16_t svmullt_n_s16(svint8_t, int8_t);
	17915	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u32)))
	17916	svuint32_t svmullt_n_u32(svuint16_t, uint16_t);
	17917	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u64)))
	17918	svuint64_t svmullt_n_u64(svuint32_t, uint32_t);
	17919	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u16)))
	17920	svuint16_t svmullt_n_u16(svuint8_t, uint8_t);
	17921	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s32)))
	17922	svint32_t svmullt_s32(svint16_t, svint16_t);
	17923	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s64)))
	17924	svint64_t svmullt_s64(svint32_t, svint32_t);
	17925	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s16)))
	17926	svint16_t svmullt_s16(svint8_t, svint8_t);
	17927	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u32)))
	17928	svuint32_t svmullt_u32(svuint16_t, svuint16_t);
	17929	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u64)))
	17930	svuint64_t svmullt_u64(svuint32_t, svuint32_t);
	17931	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u16)))
	17932	svuint16_t svmullt_u16(svuint8_t, svuint8_t);
	17933	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s32)))
	17934	svint32_t svmullt_lane_s32(svint16_t, svint16_t, uint64_t);
	17935	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s64)))
	17936	svint64_t svmullt_lane_s64(svint32_t, svint32_t, uint64_t);
	17937	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u32)))
	17938	svuint32_t svmullt_lane_u32(svuint16_t, svuint16_t, uint64_t);
	17939	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u64)))
	17940	svuint64_t svmullt_lane_u64(svuint32_t, svuint32_t, uint64_t);
	17941	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u8)))
	17942	svuint8_t svnbsl_n_u8(svuint8_t, svuint8_t, uint8_t);
	17943	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u32)))
	17944	svuint32_t svnbsl_n_u32(svuint32_t, svuint32_t, uint32_t);
	17945	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u64)))
	17946	svuint64_t svnbsl_n_u64(svuint64_t, svuint64_t, uint64_t);
	17947	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u16)))
	17948	svuint16_t svnbsl_n_u16(svuint16_t, svuint16_t, uint16_t);
	17949	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s8)))
	17950	svint8_t svnbsl_n_s8(svint8_t, svint8_t, int8_t);
	17951	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s32)))
	17952	svint32_t svnbsl_n_s32(svint32_t, svint32_t, int32_t);
	17953	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s64)))
	17954	svint64_t svnbsl_n_s64(svint64_t, svint64_t, int64_t);
	17955	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s16)))
	17956	svint16_t svnbsl_n_s16(svint16_t, svint16_t, int16_t);
	17957	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u8)))
	17958	svuint8_t svnbsl_u8(svuint8_t, svuint8_t, svuint8_t);
	17959	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u32)))
	17960	svuint32_t svnbsl_u32(svuint32_t, svuint32_t, svuint32_t);
	17961	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u64)))
	17962	svuint64_t svnbsl_u64(svuint64_t, svuint64_t, svuint64_t);
	17963	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u16)))
	17964	svuint16_t svnbsl_u16(svuint16_t, svuint16_t, svuint16_t);
	17965	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s8)))
	17966	svint8_t svnbsl_s8(svint8_t, svint8_t, svint8_t);
	17967	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s32)))
	17968	svint32_t svnbsl_s32(svint32_t, svint32_t, svint32_t);
	17969	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s64)))
	17970	svint64_t svnbsl_s64(svint64_t, svint64_t, svint64_t);
	17971	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s16)))
	17972	svint16_t svnbsl_s16(svint16_t, svint16_t, svint16_t);
	17973	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u8)))
	17974	svbool_t svnmatch_u8(svbool_t, svuint8_t, svuint8_t);
	17975	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u16)))
	17976	svbool_t svnmatch_u16(svbool_t, svuint16_t, svuint16_t);
	17977	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s8)))
	17978	svbool_t svnmatch_s8(svbool_t, svint8_t, svint8_t);
	17979	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s16)))
	17980	svbool_t svnmatch_s16(svbool_t, svint16_t, svint16_t);
	17981	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_n_u8)))
	17982	svuint8_t svpmul_n_u8(svuint8_t, uint8_t);
	17983	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_u8)))
	17984	svuint8_t svpmul_u8(svuint8_t, svuint8_t);
	17985	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u64)))
	17986	svuint64_t svpmullb_n_u64(svuint32_t, uint32_t);
	17987	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u16)))
	17988	svuint16_t svpmullb_n_u16(svuint8_t, uint8_t);
	17989	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u64)))
	17990	svuint64_t svpmullb_u64(svuint32_t, svuint32_t);
	17991	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u16)))
	17992	svuint16_t svpmullb_u16(svuint8_t, svuint8_t);
	17993	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u8)))
	17994	svuint8_t svpmullb_pair_n_u8(svuint8_t, uint8_t);
	17995	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u32)))
	17996	svuint32_t svpmullb_pair_n_u32(svuint32_t, uint32_t);
	17997	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u8)))
	17998	svuint8_t svpmullb_pair_u8(svuint8_t, svuint8_t);
	17999	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u32)))
	18000	svuint32_t svpmullb_pair_u32(svuint32_t, svuint32_t);
	18001	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u64)))
	18002	svuint64_t svpmullt_n_u64(svuint32_t, uint32_t);
	18003	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u16)))
	18004	svuint16_t svpmullt_n_u16(svuint8_t, uint8_t);
	18005	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u64)))
	18006	svuint64_t svpmullt_u64(svuint32_t, svuint32_t);
	18007	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u16)))
	18008	svuint16_t svpmullt_u16(svuint8_t, svuint8_t);
	18009	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u8)))
	18010	svuint8_t svpmullt_pair_n_u8(svuint8_t, uint8_t);
	18011	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u32)))
	18012	svuint32_t svpmullt_pair_n_u32(svuint32_t, uint32_t);
	18013	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u8)))
	18014	svuint8_t svpmullt_pair_u8(svuint8_t, svuint8_t);
	18015	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u32)))
	18016	svuint32_t svpmullt_pair_u32(svuint32_t, svuint32_t);
	18017	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_m)))
	18018	svint8_t svqabs_s8_m(svint8_t, svbool_t, svint8_t);
	18019	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_m)))
	18020	svint32_t svqabs_s32_m(svint32_t, svbool_t, svint32_t);
	18021	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_m)))
	18022	svint64_t svqabs_s64_m(svint64_t, svbool_t, svint64_t);
	18023	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_m)))
	18024	svint16_t svqabs_s16_m(svint16_t, svbool_t, svint16_t);
	18025	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_x)))
	18026	svint8_t svqabs_s8_x(svbool_t, svint8_t);
	18027	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_x)))
	18028	svint32_t svqabs_s32_x(svbool_t, svint32_t);
	18029	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_x)))
	18030	svint64_t svqabs_s64_x(svbool_t, svint64_t);
	18031	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_x)))
	18032	svint16_t svqabs_s16_x(svbool_t, svint16_t);
	18033	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_z)))
	18034	svint8_t svqabs_s8_z(svbool_t, svint8_t);
	18035	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_z)))
	18036	svint32_t svqabs_s32_z(svbool_t, svint32_t);
	18037	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_z)))
	18038	svint64_t svqabs_s64_z(svbool_t, svint64_t);
	18039	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_z)))
	18040	svint16_t svqabs_s16_z(svbool_t, svint16_t);
	18041	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_m)))
	18042	svint8_t svqadd_n_s8_m(svbool_t, svint8_t, int8_t);
	18043	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_m)))
	18044	svint32_t svqadd_n_s32_m(svbool_t, svint32_t, int32_t);
	18045	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_m)))
	18046	svint64_t svqadd_n_s64_m(svbool_t, svint64_t, int64_t);
	18047	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_m)))
	18048	svint16_t svqadd_n_s16_m(svbool_t, svint16_t, int16_t);
	18049	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_x)))
	18050	svint8_t svqadd_n_s8_x(svbool_t, svint8_t, int8_t);
	18051	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_x)))
	18052	svint32_t svqadd_n_s32_x(svbool_t, svint32_t, int32_t);
	18053	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_x)))
	18054	svint64_t svqadd_n_s64_x(svbool_t, svint64_t, int64_t);
	18055	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_x)))
	18056	svint16_t svqadd_n_s16_x(svbool_t, svint16_t, int16_t);
	18057	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_z)))
	18058	svint8_t svqadd_n_s8_z(svbool_t, svint8_t, int8_t);
	18059	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_z)))
	18060	svint32_t svqadd_n_s32_z(svbool_t, svint32_t, int32_t);
	18061	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_z)))
	18062	svint64_t svqadd_n_s64_z(svbool_t, svint64_t, int64_t);
	18063	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_z)))
	18064	svint16_t svqadd_n_s16_z(svbool_t, svint16_t, int16_t);
	18065	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_m)))
	18066	svuint8_t svqadd_n_u8_m(svbool_t, svuint8_t, uint8_t);
	18067	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_m)))
	18068	svuint32_t svqadd_n_u32_m(svbool_t, svuint32_t, uint32_t);
	18069	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_m)))
	18070	svuint64_t svqadd_n_u64_m(svbool_t, svuint64_t, uint64_t);
	18071	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_m)))
	18072	svuint16_t svqadd_n_u16_m(svbool_t, svuint16_t, uint16_t);
	18073	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_x)))
	18074	svuint8_t svqadd_n_u8_x(svbool_t, svuint8_t, uint8_t);
	18075	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_x)))
	18076	svuint32_t svqadd_n_u32_x(svbool_t, svuint32_t, uint32_t);
	18077	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_x)))
	18078	svuint64_t svqadd_n_u64_x(svbool_t, svuint64_t, uint64_t);
	18079	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_x)))
	18080	svuint16_t svqadd_n_u16_x(svbool_t, svuint16_t, uint16_t);
	18081	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_z)))
	18082	svuint8_t svqadd_n_u8_z(svbool_t, svuint8_t, uint8_t);
	18083	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_z)))
	18084	svuint32_t svqadd_n_u32_z(svbool_t, svuint32_t, uint32_t);
	18085	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_z)))
	18086	svuint64_t svqadd_n_u64_z(svbool_t, svuint64_t, uint64_t);
	18087	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_z)))
	18088	svuint16_t svqadd_n_u16_z(svbool_t, svuint16_t, uint16_t);
	18089	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_m)))
	18090	svint8_t svqadd_s8_m(svbool_t, svint8_t, svint8_t);
	18091	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_m)))
	18092	svint32_t svqadd_s32_m(svbool_t, svint32_t, svint32_t);
	18093	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_m)))
	18094	svint64_t svqadd_s64_m(svbool_t, svint64_t, svint64_t);
	18095	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_m)))
	18096	svint16_t svqadd_s16_m(svbool_t, svint16_t, svint16_t);
	18097	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_x)))
	18098	svint8_t svqadd_s8_x(svbool_t, svint8_t, svint8_t);
	18099	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_x)))
	18100	svint32_t svqadd_s32_x(svbool_t, svint32_t, svint32_t);
	18101	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_x)))
	18102	svint64_t svqadd_s64_x(svbool_t, svint64_t, svint64_t);
	18103	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_x)))
	18104	svint16_t svqadd_s16_x(svbool_t, svint16_t, svint16_t);
	18105	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_z)))
	18106	svint8_t svqadd_s8_z(svbool_t, svint8_t, svint8_t);
	18107	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_z)))
	18108	svint32_t svqadd_s32_z(svbool_t, svint32_t, svint32_t);
	18109	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_z)))
	18110	svint64_t svqadd_s64_z(svbool_t, svint64_t, svint64_t);
	18111	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_z)))
	18112	svint16_t svqadd_s16_z(svbool_t, svint16_t, svint16_t);
	18113	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_m)))
	18114	svuint8_t svqadd_u8_m(svbool_t, svuint8_t, svuint8_t);
	18115	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_m)))
	18116	svuint32_t svqadd_u32_m(svbool_t, svuint32_t, svuint32_t);
	18117	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_m)))
	18118	svuint64_t svqadd_u64_m(svbool_t, svuint64_t, svuint64_t);
	18119	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_m)))
	18120	svuint16_t svqadd_u16_m(svbool_t, svuint16_t, svuint16_t);
	18121	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_x)))
	18122	svuint8_t svqadd_u8_x(svbool_t, svuint8_t, svuint8_t);
	18123	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_x)))
	18124	svuint32_t svqadd_u32_x(svbool_t, svuint32_t, svuint32_t);
	18125	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_x)))
	18126	svuint64_t svqadd_u64_x(svbool_t, svuint64_t, svuint64_t);
	18127	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_x)))
	18128	svuint16_t svqadd_u16_x(svbool_t, svuint16_t, svuint16_t);
	18129	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_z)))
	18130	svuint8_t svqadd_u8_z(svbool_t, svuint8_t, svuint8_t);
	18131	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_z)))
	18132	svuint32_t svqadd_u32_z(svbool_t, svuint32_t, svuint32_t);
	18133	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_z)))
	18134	svuint64_t svqadd_u64_z(svbool_t, svuint64_t, svuint64_t);
	18135	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_z)))
	18136	svuint16_t svqadd_u16_z(svbool_t, svuint16_t, svuint16_t);
	18137	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s8)))
	18138	svint8_t svqcadd_s8(svint8_t, svint8_t, uint64_t);
	18139	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s32)))
	18140	svint32_t svqcadd_s32(svint32_t, svint32_t, uint64_t);
	18141	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s64)))
	18142	svint64_t svqcadd_s64(svint64_t, svint64_t, uint64_t);
	18143	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s16)))
	18144	svint16_t svqcadd_s16(svint16_t, svint16_t, uint64_t);
	18145	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s32)))
	18146	svint32_t svqdmlalb_n_s32(svint32_t, svint16_t, int16_t);
	18147	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s64)))
	18148	svint64_t svqdmlalb_n_s64(svint64_t, svint32_t, int32_t);
	18149	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s16)))
	18150	svint16_t svqdmlalb_n_s16(svint16_t, svint8_t, int8_t);
	18151	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s32)))
	18152	svint32_t svqdmlalb_s32(svint32_t, svint16_t, svint16_t);
	18153	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s64)))
	18154	svint64_t svqdmlalb_s64(svint64_t, svint32_t, svint32_t);
	18155	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s16)))
	18156	svint16_t svqdmlalb_s16(svint16_t, svint8_t, svint8_t);
	18157	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s32)))
	18158	svint32_t svqdmlalb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	18159	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s64)))
	18160	svint64_t svqdmlalb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	18161	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s32)))
	18162	svint32_t svqdmlalbt_n_s32(svint32_t, svint16_t, int16_t);
	18163	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s64)))
	18164	svint64_t svqdmlalbt_n_s64(svint64_t, svint32_t, int32_t);
	18165	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s16)))
	18166	svint16_t svqdmlalbt_n_s16(svint16_t, svint8_t, int8_t);
	18167	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s32)))
	18168	svint32_t svqdmlalbt_s32(svint32_t, svint16_t, svint16_t);
	18169	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s64)))
	18170	svint64_t svqdmlalbt_s64(svint64_t, svint32_t, svint32_t);
	18171	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s16)))
	18172	svint16_t svqdmlalbt_s16(svint16_t, svint8_t, svint8_t);
	18173	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s32)))
	18174	svint32_t svqdmlalt_n_s32(svint32_t, svint16_t, int16_t);
	18175	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s64)))
	18176	svint64_t svqdmlalt_n_s64(svint64_t, svint32_t, int32_t);
	18177	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s16)))
	18178	svint16_t svqdmlalt_n_s16(svint16_t, svint8_t, int8_t);
	18179	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s32)))
	18180	svint32_t svqdmlalt_s32(svint32_t, svint16_t, svint16_t);
	18181	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s64)))
	18182	svint64_t svqdmlalt_s64(svint64_t, svint32_t, svint32_t);
	18183	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s16)))
	18184	svint16_t svqdmlalt_s16(svint16_t, svint8_t, svint8_t);
	18185	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s32)))
	18186	svint32_t svqdmlalt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	18187	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s64)))
	18188	svint64_t svqdmlalt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	18189	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s32)))
	18190	svint32_t svqdmlslb_n_s32(svint32_t, svint16_t, int16_t);
	18191	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s64)))
	18192	svint64_t svqdmlslb_n_s64(svint64_t, svint32_t, int32_t);
	18193	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s16)))
	18194	svint16_t svqdmlslb_n_s16(svint16_t, svint8_t, int8_t);
	18195	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s32)))
	18196	svint32_t svqdmlslb_s32(svint32_t, svint16_t, svint16_t);
	18197	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s64)))
	18198	svint64_t svqdmlslb_s64(svint64_t, svint32_t, svint32_t);
	18199	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s16)))
	18200	svint16_t svqdmlslb_s16(svint16_t, svint8_t, svint8_t);
	18201	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s32)))
	18202	svint32_t svqdmlslb_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	18203	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s64)))
	18204	svint64_t svqdmlslb_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	18205	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s32)))
	18206	svint32_t svqdmlslbt_n_s32(svint32_t, svint16_t, int16_t);
	18207	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s64)))
	18208	svint64_t svqdmlslbt_n_s64(svint64_t, svint32_t, int32_t);
	18209	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s16)))
	18210	svint16_t svqdmlslbt_n_s16(svint16_t, svint8_t, int8_t);
	18211	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s32)))
	18212	svint32_t svqdmlslbt_s32(svint32_t, svint16_t, svint16_t);
	18213	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s64)))
	18214	svint64_t svqdmlslbt_s64(svint64_t, svint32_t, svint32_t);
	18215	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s16)))
	18216	svint16_t svqdmlslbt_s16(svint16_t, svint8_t, svint8_t);
	18217	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s32)))
	18218	svint32_t svqdmlslt_n_s32(svint32_t, svint16_t, int16_t);
	18219	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s64)))
	18220	svint64_t svqdmlslt_n_s64(svint64_t, svint32_t, int32_t);
	18221	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s16)))
	18222	svint16_t svqdmlslt_n_s16(svint16_t, svint8_t, int8_t);
	18223	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s32)))
	18224	svint32_t svqdmlslt_s32(svint32_t, svint16_t, svint16_t);
	18225	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s64)))
	18226	svint64_t svqdmlslt_s64(svint64_t, svint32_t, svint32_t);
	18227	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s16)))
	18228	svint16_t svqdmlslt_s16(svint16_t, svint8_t, svint8_t);
	18229	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s32)))
	18230	svint32_t svqdmlslt_lane_s32(svint32_t, svint16_t, svint16_t, uint64_t);
	18231	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s64)))
	18232	svint64_t svqdmlslt_lane_s64(svint64_t, svint32_t, svint32_t, uint64_t);
	18233	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s8)))
	18234	svint8_t svqdmulh_n_s8(svint8_t, int8_t);
	18235	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s32)))
	18236	svint32_t svqdmulh_n_s32(svint32_t, int32_t);
	18237	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s64)))
	18238	svint64_t svqdmulh_n_s64(svint64_t, int64_t);
	18239	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s16)))
	18240	svint16_t svqdmulh_n_s16(svint16_t, int16_t);
	18241	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s8)))
	18242	svint8_t svqdmulh_s8(svint8_t, svint8_t);
	18243	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s32)))
	18244	svint32_t svqdmulh_s32(svint32_t, svint32_t);
	18245	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s64)))
	18246	svint64_t svqdmulh_s64(svint64_t, svint64_t);
	18247	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s16)))
	18248	svint16_t svqdmulh_s16(svint16_t, svint16_t);
	18249	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s32)))
	18250	svint32_t svqdmulh_lane_s32(svint32_t, svint32_t, uint64_t);
	18251	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s64)))
	18252	svint64_t svqdmulh_lane_s64(svint64_t, svint64_t, uint64_t);
	18253	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s16)))
	18254	svint16_t svqdmulh_lane_s16(svint16_t, svint16_t, uint64_t);
	18255	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s32)))
	18256	svint32_t svqdmullb_n_s32(svint16_t, int16_t);
	18257	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s64)))
	18258	svint64_t svqdmullb_n_s64(svint32_t, int32_t);
	18259	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s16)))
	18260	svint16_t svqdmullb_n_s16(svint8_t, int8_t);
	18261	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s32)))
	18262	svint32_t svqdmullb_s32(svint16_t, svint16_t);
	18263	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s64)))
	18264	svint64_t svqdmullb_s64(svint32_t, svint32_t);
	18265	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s16)))
	18266	svint16_t svqdmullb_s16(svint8_t, svint8_t);
	18267	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s32)))
	18268	svint32_t svqdmullb_lane_s32(svint16_t, svint16_t, uint64_t);
	18269	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s64)))
	18270	svint64_t svqdmullb_lane_s64(svint32_t, svint32_t, uint64_t);
	18271	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s32)))
	18272	svint32_t svqdmullt_n_s32(svint16_t, int16_t);
	18273	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s64)))
	18274	svint64_t svqdmullt_n_s64(svint32_t, int32_t);
	18275	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s16)))
	18276	svint16_t svqdmullt_n_s16(svint8_t, int8_t);
	18277	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s32)))
	18278	svint32_t svqdmullt_s32(svint16_t, svint16_t);
	18279	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s64)))
	18280	svint64_t svqdmullt_s64(svint32_t, svint32_t);
	18281	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s16)))
	18282	svint16_t svqdmullt_s16(svint8_t, svint8_t);
	18283	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s32)))
	18284	svint32_t svqdmullt_lane_s32(svint16_t, svint16_t, uint64_t);
	18285	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s64)))
	18286	svint64_t svqdmullt_lane_s64(svint32_t, svint32_t, uint64_t);
	18287	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_m)))
	18288	svint8_t svqneg_s8_m(svint8_t, svbool_t, svint8_t);
	18289	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_m)))
	18290	svint32_t svqneg_s32_m(svint32_t, svbool_t, svint32_t);
	18291	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_m)))
	18292	svint64_t svqneg_s64_m(svint64_t, svbool_t, svint64_t);
	18293	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_m)))
	18294	svint16_t svqneg_s16_m(svint16_t, svbool_t, svint16_t);
	18295	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_x)))
	18296	svint8_t svqneg_s8_x(svbool_t, svint8_t);
	18297	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_x)))
	18298	svint32_t svqneg_s32_x(svbool_t, svint32_t);
	18299	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_x)))
	18300	svint64_t svqneg_s64_x(svbool_t, svint64_t);
	18301	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_x)))
	18302	svint16_t svqneg_s16_x(svbool_t, svint16_t);
	18303	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_z)))
	18304	svint8_t svqneg_s8_z(svbool_t, svint8_t);
	18305	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_z)))
	18306	svint32_t svqneg_s32_z(svbool_t, svint32_t);
	18307	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_z)))
	18308	svint64_t svqneg_s64_z(svbool_t, svint64_t);
	18309	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_z)))
	18310	svint16_t svqneg_s16_z(svbool_t, svint16_t);
	18311	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s8)))
	18312	svint8_t svqrdcmlah_s8(svint8_t, svint8_t, svint8_t, uint64_t);
	18313	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s32)))
	18314	svint32_t svqrdcmlah_s32(svint32_t, svint32_t, svint32_t, uint64_t);
	18315	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s64)))
	18316	svint64_t svqrdcmlah_s64(svint64_t, svint64_t, svint64_t, uint64_t);
	18317	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s16)))
	18318	svint16_t svqrdcmlah_s16(svint16_t, svint16_t, svint16_t, uint64_t);
	18319	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s32)))
	18320	svint32_t svqrdcmlah_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
	18321	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s16)))
	18322	svint16_t svqrdcmlah_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
	18323	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s8)))
	18324	svint8_t svqrdmlah_n_s8(svint8_t, svint8_t, int8_t);
	18325	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s32)))
	18326	svint32_t svqrdmlah_n_s32(svint32_t, svint32_t, int32_t);
	18327	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s64)))
	18328	svint64_t svqrdmlah_n_s64(svint64_t, svint64_t, int64_t);
	18329	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s16)))
	18330	svint16_t svqrdmlah_n_s16(svint16_t, svint16_t, int16_t);
	18331	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s8)))
	18332	svint8_t svqrdmlah_s8(svint8_t, svint8_t, svint8_t);
	18333	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s32)))
	18334	svint32_t svqrdmlah_s32(svint32_t, svint32_t, svint32_t);
	18335	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s64)))
	18336	svint64_t svqrdmlah_s64(svint64_t, svint64_t, svint64_t);
	18337	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s16)))
	18338	svint16_t svqrdmlah_s16(svint16_t, svint16_t, svint16_t);
	18339	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s32)))
	18340	svint32_t svqrdmlah_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
	18341	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s64)))
	18342	svint64_t svqrdmlah_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
	18343	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s16)))
	18344	svint16_t svqrdmlah_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
	18345	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s8)))
	18346	svint8_t svqrdmlsh_n_s8(svint8_t, svint8_t, int8_t);
	18347	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s32)))
	18348	svint32_t svqrdmlsh_n_s32(svint32_t, svint32_t, int32_t);
	18349	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s64)))
	18350	svint64_t svqrdmlsh_n_s64(svint64_t, svint64_t, int64_t);
	18351	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s16)))
	18352	svint16_t svqrdmlsh_n_s16(svint16_t, svint16_t, int16_t);
	18353	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s8)))
	18354	svint8_t svqrdmlsh_s8(svint8_t, svint8_t, svint8_t);
	18355	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s32)))
	18356	svint32_t svqrdmlsh_s32(svint32_t, svint32_t, svint32_t);
	18357	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s64)))
	18358	svint64_t svqrdmlsh_s64(svint64_t, svint64_t, svint64_t);
	18359	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s16)))
	18360	svint16_t svqrdmlsh_s16(svint16_t, svint16_t, svint16_t);
	18361	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s32)))
	18362	svint32_t svqrdmlsh_lane_s32(svint32_t, svint32_t, svint32_t, uint64_t);
	18363	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s64)))
	18364	svint64_t svqrdmlsh_lane_s64(svint64_t, svint64_t, svint64_t, uint64_t);
	18365	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s16)))
	18366	svint16_t svqrdmlsh_lane_s16(svint16_t, svint16_t, svint16_t, uint64_t);
	18367	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s8)))
	18368	svint8_t svqrdmulh_n_s8(svint8_t, int8_t);
	18369	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s32)))
	18370	svint32_t svqrdmulh_n_s32(svint32_t, int32_t);
	18371	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s64)))
	18372	svint64_t svqrdmulh_n_s64(svint64_t, int64_t);
	18373	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s16)))
	18374	svint16_t svqrdmulh_n_s16(svint16_t, int16_t);
	18375	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s8)))
	18376	svint8_t svqrdmulh_s8(svint8_t, svint8_t);
	18377	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s32)))
	18378	svint32_t svqrdmulh_s32(svint32_t, svint32_t);
	18379	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s64)))
	18380	svint64_t svqrdmulh_s64(svint64_t, svint64_t);
	18381	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s16)))
	18382	svint16_t svqrdmulh_s16(svint16_t, svint16_t);
	18383	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s32)))
	18384	svint32_t svqrdmulh_lane_s32(svint32_t, svint32_t, uint64_t);
	18385	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s64)))
	18386	svint64_t svqrdmulh_lane_s64(svint64_t, svint64_t, uint64_t);
	18387	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s16)))
	18388	svint16_t svqrdmulh_lane_s16(svint16_t, svint16_t, uint64_t);
	18389	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_m)))
	18390	svint8_t svqrshl_n_s8_m(svbool_t, svint8_t, int8_t);
	18391	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_m)))
	18392	svint32_t svqrshl_n_s32_m(svbool_t, svint32_t, int32_t);
	18393	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_m)))
	18394	svint64_t svqrshl_n_s64_m(svbool_t, svint64_t, int64_t);
	18395	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_m)))
	18396	svint16_t svqrshl_n_s16_m(svbool_t, svint16_t, int16_t);
	18397	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_x)))
	18398	svint8_t svqrshl_n_s8_x(svbool_t, svint8_t, int8_t);
	18399	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_x)))
	18400	svint32_t svqrshl_n_s32_x(svbool_t, svint32_t, int32_t);
	18401	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_x)))
	18402	svint64_t svqrshl_n_s64_x(svbool_t, svint64_t, int64_t);
	18403	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_x)))
	18404	svint16_t svqrshl_n_s16_x(svbool_t, svint16_t, int16_t);
	18405	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_z)))
	18406	svint8_t svqrshl_n_s8_z(svbool_t, svint8_t, int8_t);
	18407	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_z)))
	18408	svint32_t svqrshl_n_s32_z(svbool_t, svint32_t, int32_t);
	18409	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_z)))
	18410	svint64_t svqrshl_n_s64_z(svbool_t, svint64_t, int64_t);
	18411	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_z)))
	18412	svint16_t svqrshl_n_s16_z(svbool_t, svint16_t, int16_t);
	18413	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_m)))
	18414	svuint8_t svqrshl_n_u8_m(svbool_t, svuint8_t, int8_t);
	18415	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_m)))
	18416	svuint32_t svqrshl_n_u32_m(svbool_t, svuint32_t, int32_t);
	18417	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_m)))
	18418	svuint64_t svqrshl_n_u64_m(svbool_t, svuint64_t, int64_t);
	18419	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_m)))
	18420	svuint16_t svqrshl_n_u16_m(svbool_t, svuint16_t, int16_t);
	18421	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_x)))
	18422	svuint8_t svqrshl_n_u8_x(svbool_t, svuint8_t, int8_t);
	18423	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_x)))
	18424	svuint32_t svqrshl_n_u32_x(svbool_t, svuint32_t, int32_t);
	18425	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_x)))
	18426	svuint64_t svqrshl_n_u64_x(svbool_t, svuint64_t, int64_t);
	18427	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_x)))
	18428	svuint16_t svqrshl_n_u16_x(svbool_t, svuint16_t, int16_t);
	18429	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_z)))
	18430	svuint8_t svqrshl_n_u8_z(svbool_t, svuint8_t, int8_t);
	18431	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_z)))
	18432	svuint32_t svqrshl_n_u32_z(svbool_t, svuint32_t, int32_t);
	18433	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_z)))
	18434	svuint64_t svqrshl_n_u64_z(svbool_t, svuint64_t, int64_t);
	18435	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_z)))
	18436	svuint16_t svqrshl_n_u16_z(svbool_t, svuint16_t, int16_t);
	18437	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_m)))
	18438	svint8_t svqrshl_s8_m(svbool_t, svint8_t, svint8_t);
	18439	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_m)))
	18440	svint32_t svqrshl_s32_m(svbool_t, svint32_t, svint32_t);
	18441	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_m)))
	18442	svint64_t svqrshl_s64_m(svbool_t, svint64_t, svint64_t);
	18443	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_m)))
	18444	svint16_t svqrshl_s16_m(svbool_t, svint16_t, svint16_t);
	18445	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_x)))
	18446	svint8_t svqrshl_s8_x(svbool_t, svint8_t, svint8_t);
	18447	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_x)))
	18448	svint32_t svqrshl_s32_x(svbool_t, svint32_t, svint32_t);
	18449	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_x)))
	18450	svint64_t svqrshl_s64_x(svbool_t, svint64_t, svint64_t);
	18451	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_x)))
	18452	svint16_t svqrshl_s16_x(svbool_t, svint16_t, svint16_t);
	18453	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_z)))
	18454	svint8_t svqrshl_s8_z(svbool_t, svint8_t, svint8_t);
	18455	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_z)))
	18456	svint32_t svqrshl_s32_z(svbool_t, svint32_t, svint32_t);
	18457	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_z)))
	18458	svint64_t svqrshl_s64_z(svbool_t, svint64_t, svint64_t);
	18459	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_z)))
	18460	svint16_t svqrshl_s16_z(svbool_t, svint16_t, svint16_t);
	18461	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_m)))
	18462	svuint8_t svqrshl_u8_m(svbool_t, svuint8_t, svint8_t);
	18463	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_m)))
	18464	svuint32_t svqrshl_u32_m(svbool_t, svuint32_t, svint32_t);
	18465	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_m)))
	18466	svuint64_t svqrshl_u64_m(svbool_t, svuint64_t, svint64_t);
	18467	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_m)))
	18468	svuint16_t svqrshl_u16_m(svbool_t, svuint16_t, svint16_t);
	18469	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_x)))
	18470	svuint8_t svqrshl_u8_x(svbool_t, svuint8_t, svint8_t);
	18471	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_x)))
	18472	svuint32_t svqrshl_u32_x(svbool_t, svuint32_t, svint32_t);
	18473	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_x)))
	18474	svuint64_t svqrshl_u64_x(svbool_t, svuint64_t, svint64_t);
	18475	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_x)))
	18476	svuint16_t svqrshl_u16_x(svbool_t, svuint16_t, svint16_t);
	18477	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_z)))
	18478	svuint8_t svqrshl_u8_z(svbool_t, svuint8_t, svint8_t);
	18479	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_z)))
	18480	svuint32_t svqrshl_u32_z(svbool_t, svuint32_t, svint32_t);
	18481	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_z)))
	18482	svuint64_t svqrshl_u64_z(svbool_t, svuint64_t, svint64_t);
	18483	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_z)))
	18484	svuint16_t svqrshl_u16_z(svbool_t, svuint16_t, svint16_t);
	18485	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s32)))
	18486	svint16_t svqrshrnb_n_s32(svint32_t, uint64_t);
	18487	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s64)))
	18488	svint32_t svqrshrnb_n_s64(svint64_t, uint64_t);
	18489	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s16)))
	18490	svint8_t svqrshrnb_n_s16(svint16_t, uint64_t);
	18491	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u32)))
	18492	svuint16_t svqrshrnb_n_u32(svuint32_t, uint64_t);
	18493	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u64)))
	18494	svuint32_t svqrshrnb_n_u64(svuint64_t, uint64_t);
	18495	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u16)))
	18496	svuint8_t svqrshrnb_n_u16(svuint16_t, uint64_t);
	18497	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s32)))
	18498	svint16_t svqrshrnt_n_s32(svint16_t, svint32_t, uint64_t);
	18499	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s64)))
	18500	svint32_t svqrshrnt_n_s64(svint32_t, svint64_t, uint64_t);
	18501	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s16)))
	18502	svint8_t svqrshrnt_n_s16(svint8_t, svint16_t, uint64_t);
	18503	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u32)))
	18504	svuint16_t svqrshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
	18505	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u64)))
	18506	svuint32_t svqrshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	18507	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u16)))
	18508	svuint8_t svqrshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
	18509	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s32)))
	18510	svuint16_t svqrshrunb_n_s32(svint32_t, uint64_t);
	18511	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s64)))
	18512	svuint32_t svqrshrunb_n_s64(svint64_t, uint64_t);
	18513	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s16)))
	18514	svuint8_t svqrshrunb_n_s16(svint16_t, uint64_t);
	18515	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s32)))
	18516	svuint16_t svqrshrunt_n_s32(svuint16_t, svint32_t, uint64_t);
	18517	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s64)))
	18518	svuint32_t svqrshrunt_n_s64(svuint32_t, svint64_t, uint64_t);
	18519	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s16)))
	18520	svuint8_t svqrshrunt_n_s16(svuint8_t, svint16_t, uint64_t);
	18521	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_m)))
	18522	svint8_t svqshl_n_s8_m(svbool_t, svint8_t, int8_t);
	18523	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_m)))
	18524	svint32_t svqshl_n_s32_m(svbool_t, svint32_t, int32_t);
	18525	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_m)))
	18526	svint64_t svqshl_n_s64_m(svbool_t, svint64_t, int64_t);
	18527	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_m)))
	18528	svint16_t svqshl_n_s16_m(svbool_t, svint16_t, int16_t);
	18529	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_x)))
	18530	svint8_t svqshl_n_s8_x(svbool_t, svint8_t, int8_t);
	18531	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_x)))
	18532	svint32_t svqshl_n_s32_x(svbool_t, svint32_t, int32_t);
	18533	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_x)))
	18534	svint64_t svqshl_n_s64_x(svbool_t, svint64_t, int64_t);
	18535	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_x)))
	18536	svint16_t svqshl_n_s16_x(svbool_t, svint16_t, int16_t);
	18537	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_z)))
	18538	svint8_t svqshl_n_s8_z(svbool_t, svint8_t, int8_t);
	18539	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_z)))
	18540	svint32_t svqshl_n_s32_z(svbool_t, svint32_t, int32_t);
	18541	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_z)))
	18542	svint64_t svqshl_n_s64_z(svbool_t, svint64_t, int64_t);
	18543	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_z)))
	18544	svint16_t svqshl_n_s16_z(svbool_t, svint16_t, int16_t);
	18545	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_m)))
	18546	svuint8_t svqshl_n_u8_m(svbool_t, svuint8_t, int8_t);
	18547	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_m)))
	18548	svuint32_t svqshl_n_u32_m(svbool_t, svuint32_t, int32_t);
	18549	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_m)))
	18550	svuint64_t svqshl_n_u64_m(svbool_t, svuint64_t, int64_t);
	18551	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_m)))
	18552	svuint16_t svqshl_n_u16_m(svbool_t, svuint16_t, int16_t);
	18553	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_x)))
	18554	svuint8_t svqshl_n_u8_x(svbool_t, svuint8_t, int8_t);
	18555	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_x)))
	18556	svuint32_t svqshl_n_u32_x(svbool_t, svuint32_t, int32_t);
	18557	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_x)))
	18558	svuint64_t svqshl_n_u64_x(svbool_t, svuint64_t, int64_t);
	18559	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_x)))
	18560	svuint16_t svqshl_n_u16_x(svbool_t, svuint16_t, int16_t);
	18561	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_z)))
	18562	svuint8_t svqshl_n_u8_z(svbool_t, svuint8_t, int8_t);
	18563	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_z)))
	18564	svuint32_t svqshl_n_u32_z(svbool_t, svuint32_t, int32_t);
	18565	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_z)))
	18566	svuint64_t svqshl_n_u64_z(svbool_t, svuint64_t, int64_t);
	18567	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_z)))
	18568	svuint16_t svqshl_n_u16_z(svbool_t, svuint16_t, int16_t);
	18569	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_m)))
	18570	svint8_t svqshl_s8_m(svbool_t, svint8_t, svint8_t);
	18571	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_m)))
	18572	svint32_t svqshl_s32_m(svbool_t, svint32_t, svint32_t);
	18573	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_m)))
	18574	svint64_t svqshl_s64_m(svbool_t, svint64_t, svint64_t);
	18575	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_m)))
	18576	svint16_t svqshl_s16_m(svbool_t, svint16_t, svint16_t);
	18577	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_x)))
	18578	svint8_t svqshl_s8_x(svbool_t, svint8_t, svint8_t);
	18579	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_x)))
	18580	svint32_t svqshl_s32_x(svbool_t, svint32_t, svint32_t);
	18581	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_x)))
	18582	svint64_t svqshl_s64_x(svbool_t, svint64_t, svint64_t);
	18583	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_x)))
	18584	svint16_t svqshl_s16_x(svbool_t, svint16_t, svint16_t);
	18585	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_z)))
	18586	svint8_t svqshl_s8_z(svbool_t, svint8_t, svint8_t);
	18587	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_z)))
	18588	svint32_t svqshl_s32_z(svbool_t, svint32_t, svint32_t);
	18589	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_z)))
	18590	svint64_t svqshl_s64_z(svbool_t, svint64_t, svint64_t);
	18591	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_z)))
	18592	svint16_t svqshl_s16_z(svbool_t, svint16_t, svint16_t);
	18593	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_m)))
	18594	svuint8_t svqshl_u8_m(svbool_t, svuint8_t, svint8_t);
	18595	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_m)))
	18596	svuint32_t svqshl_u32_m(svbool_t, svuint32_t, svint32_t);
	18597	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_m)))
	18598	svuint64_t svqshl_u64_m(svbool_t, svuint64_t, svint64_t);
	18599	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_m)))
	18600	svuint16_t svqshl_u16_m(svbool_t, svuint16_t, svint16_t);
	18601	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_x)))
	18602	svuint8_t svqshl_u8_x(svbool_t, svuint8_t, svint8_t);
	18603	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_x)))
	18604	svuint32_t svqshl_u32_x(svbool_t, svuint32_t, svint32_t);
	18605	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_x)))
	18606	svuint64_t svqshl_u64_x(svbool_t, svuint64_t, svint64_t);
	18607	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_x)))
	18608	svuint16_t svqshl_u16_x(svbool_t, svuint16_t, svint16_t);
	18609	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_z)))
	18610	svuint8_t svqshl_u8_z(svbool_t, svuint8_t, svint8_t);
	18611	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_z)))
	18612	svuint32_t svqshl_u32_z(svbool_t, svuint32_t, svint32_t);
	18613	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_z)))
	18614	svuint64_t svqshl_u64_z(svbool_t, svuint64_t, svint64_t);
	18615	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_z)))
	18616	svuint16_t svqshl_u16_z(svbool_t, svuint16_t, svint16_t);
	18617	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_m)))
	18618	svuint8_t svqshlu_n_s8_m(svbool_t, svint8_t, uint64_t);
	18619	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_m)))
	18620	svuint32_t svqshlu_n_s32_m(svbool_t, svint32_t, uint64_t);
	18621	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_m)))
	18622	svuint64_t svqshlu_n_s64_m(svbool_t, svint64_t, uint64_t);
	18623	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_m)))
	18624	svuint16_t svqshlu_n_s16_m(svbool_t, svint16_t, uint64_t);
	18625	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_x)))
	18626	svuint8_t svqshlu_n_s8_x(svbool_t, svint8_t, uint64_t);
	18627	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_x)))
	18628	svuint32_t svqshlu_n_s32_x(svbool_t, svint32_t, uint64_t);
	18629	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_x)))
	18630	svuint64_t svqshlu_n_s64_x(svbool_t, svint64_t, uint64_t);
	18631	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_x)))
	18632	svuint16_t svqshlu_n_s16_x(svbool_t, svint16_t, uint64_t);
	18633	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_z)))
	18634	svuint8_t svqshlu_n_s8_z(svbool_t, svint8_t, uint64_t);
	18635	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_z)))
	18636	svuint32_t svqshlu_n_s32_z(svbool_t, svint32_t, uint64_t);
	18637	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_z)))
	18638	svuint64_t svqshlu_n_s64_z(svbool_t, svint64_t, uint64_t);
	18639	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_z)))
	18640	svuint16_t svqshlu_n_s16_z(svbool_t, svint16_t, uint64_t);
	18641	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s32)))
	18642	svint16_t svqshrnb_n_s32(svint32_t, uint64_t);
	18643	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s64)))
	18644	svint32_t svqshrnb_n_s64(svint64_t, uint64_t);
	18645	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s16)))
	18646	svint8_t svqshrnb_n_s16(svint16_t, uint64_t);
	18647	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u32)))
	18648	svuint16_t svqshrnb_n_u32(svuint32_t, uint64_t);
	18649	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u64)))
	18650	svuint32_t svqshrnb_n_u64(svuint64_t, uint64_t);
	18651	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u16)))
	18652	svuint8_t svqshrnb_n_u16(svuint16_t, uint64_t);
	18653	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s32)))
	18654	svint16_t svqshrnt_n_s32(svint16_t, svint32_t, uint64_t);
	18655	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s64)))
	18656	svint32_t svqshrnt_n_s64(svint32_t, svint64_t, uint64_t);
	18657	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s16)))
	18658	svint8_t svqshrnt_n_s16(svint8_t, svint16_t, uint64_t);
	18659	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u32)))
	18660	svuint16_t svqshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
	18661	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u64)))
	18662	svuint32_t svqshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	18663	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u16)))
	18664	svuint8_t svqshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
	18665	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s32)))
	18666	svuint16_t svqshrunb_n_s32(svint32_t, uint64_t);
	18667	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s64)))
	18668	svuint32_t svqshrunb_n_s64(svint64_t, uint64_t);
	18669	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s16)))
	18670	svuint8_t svqshrunb_n_s16(svint16_t, uint64_t);
	18671	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s32)))
	18672	svuint16_t svqshrunt_n_s32(svuint16_t, svint32_t, uint64_t);
	18673	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s64)))
	18674	svuint32_t svqshrunt_n_s64(svuint32_t, svint64_t, uint64_t);
	18675	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s16)))
	18676	svuint8_t svqshrunt_n_s16(svuint8_t, svint16_t, uint64_t);
	18677	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_m)))
	18678	svint8_t svqsub_n_s8_m(svbool_t, svint8_t, int8_t);
	18679	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_m)))
	18680	svint32_t svqsub_n_s32_m(svbool_t, svint32_t, int32_t);
	18681	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_m)))
	18682	svint64_t svqsub_n_s64_m(svbool_t, svint64_t, int64_t);
	18683	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_m)))
	18684	svint16_t svqsub_n_s16_m(svbool_t, svint16_t, int16_t);
	18685	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_x)))
	18686	svint8_t svqsub_n_s8_x(svbool_t, svint8_t, int8_t);
	18687	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_x)))
	18688	svint32_t svqsub_n_s32_x(svbool_t, svint32_t, int32_t);
	18689	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_x)))
	18690	svint64_t svqsub_n_s64_x(svbool_t, svint64_t, int64_t);
	18691	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_x)))
	18692	svint16_t svqsub_n_s16_x(svbool_t, svint16_t, int16_t);
	18693	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_z)))
	18694	svint8_t svqsub_n_s8_z(svbool_t, svint8_t, int8_t);
	18695	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_z)))
	18696	svint32_t svqsub_n_s32_z(svbool_t, svint32_t, int32_t);
	18697	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_z)))
	18698	svint64_t svqsub_n_s64_z(svbool_t, svint64_t, int64_t);
	18699	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_z)))
	18700	svint16_t svqsub_n_s16_z(svbool_t, svint16_t, int16_t);
	18701	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_m)))
	18702	svuint8_t svqsub_n_u8_m(svbool_t, svuint8_t, uint8_t);
	18703	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_m)))
	18704	svuint32_t svqsub_n_u32_m(svbool_t, svuint32_t, uint32_t);
	18705	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_m)))
	18706	svuint64_t svqsub_n_u64_m(svbool_t, svuint64_t, uint64_t);
	18707	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_m)))
	18708	svuint16_t svqsub_n_u16_m(svbool_t, svuint16_t, uint16_t);
	18709	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_x)))
	18710	svuint8_t svqsub_n_u8_x(svbool_t, svuint8_t, uint8_t);
	18711	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_x)))
	18712	svuint32_t svqsub_n_u32_x(svbool_t, svuint32_t, uint32_t);
	18713	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_x)))
	18714	svuint64_t svqsub_n_u64_x(svbool_t, svuint64_t, uint64_t);
	18715	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_x)))
	18716	svuint16_t svqsub_n_u16_x(svbool_t, svuint16_t, uint16_t);
	18717	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_z)))
	18718	svuint8_t svqsub_n_u8_z(svbool_t, svuint8_t, uint8_t);
	18719	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_z)))
	18720	svuint32_t svqsub_n_u32_z(svbool_t, svuint32_t, uint32_t);
	18721	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_z)))
	18722	svuint64_t svqsub_n_u64_z(svbool_t, svuint64_t, uint64_t);
	18723	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_z)))
	18724	svuint16_t svqsub_n_u16_z(svbool_t, svuint16_t, uint16_t);
	18725	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_m)))
	18726	svint8_t svqsub_s8_m(svbool_t, svint8_t, svint8_t);
	18727	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_m)))
	18728	svint32_t svqsub_s32_m(svbool_t, svint32_t, svint32_t);
	18729	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_m)))
	18730	svint64_t svqsub_s64_m(svbool_t, svint64_t, svint64_t);
	18731	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_m)))
	18732	svint16_t svqsub_s16_m(svbool_t, svint16_t, svint16_t);
	18733	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_x)))
	18734	svint8_t svqsub_s8_x(svbool_t, svint8_t, svint8_t);
	18735	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_x)))
	18736	svint32_t svqsub_s32_x(svbool_t, svint32_t, svint32_t);
	18737	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_x)))
	18738	svint64_t svqsub_s64_x(svbool_t, svint64_t, svint64_t);
	18739	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_x)))
	18740	svint16_t svqsub_s16_x(svbool_t, svint16_t, svint16_t);
	18741	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_z)))
	18742	svint8_t svqsub_s8_z(svbool_t, svint8_t, svint8_t);
	18743	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_z)))
	18744	svint32_t svqsub_s32_z(svbool_t, svint32_t, svint32_t);
	18745	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_z)))
	18746	svint64_t svqsub_s64_z(svbool_t, svint64_t, svint64_t);
	18747	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_z)))
	18748	svint16_t svqsub_s16_z(svbool_t, svint16_t, svint16_t);
	18749	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_m)))
	18750	svuint8_t svqsub_u8_m(svbool_t, svuint8_t, svuint8_t);
	18751	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_m)))
	18752	svuint32_t svqsub_u32_m(svbool_t, svuint32_t, svuint32_t);
	18753	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_m)))
	18754	svuint64_t svqsub_u64_m(svbool_t, svuint64_t, svuint64_t);
	18755	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_m)))
	18756	svuint16_t svqsub_u16_m(svbool_t, svuint16_t, svuint16_t);
	18757	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_x)))
	18758	svuint8_t svqsub_u8_x(svbool_t, svuint8_t, svuint8_t);
	18759	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_x)))
	18760	svuint32_t svqsub_u32_x(svbool_t, svuint32_t, svuint32_t);
	18761	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_x)))
	18762	svuint64_t svqsub_u64_x(svbool_t, svuint64_t, svuint64_t);
	18763	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_x)))
	18764	svuint16_t svqsub_u16_x(svbool_t, svuint16_t, svuint16_t);
	18765	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_z)))
	18766	svuint8_t svqsub_u8_z(svbool_t, svuint8_t, svuint8_t);
	18767	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_z)))
	18768	svuint32_t svqsub_u32_z(svbool_t, svuint32_t, svuint32_t);
	18769	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_z)))
	18770	svuint64_t svqsub_u64_z(svbool_t, svuint64_t, svuint64_t);
	18771	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_z)))
	18772	svuint16_t svqsub_u16_z(svbool_t, svuint16_t, svuint16_t);
	18773	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_m)))
	18774	svint8_t svqsubr_n_s8_m(svbool_t, svint8_t, int8_t);
	18775	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_m)))
	18776	svint32_t svqsubr_n_s32_m(svbool_t, svint32_t, int32_t);
	18777	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_m)))
	18778	svint64_t svqsubr_n_s64_m(svbool_t, svint64_t, int64_t);
	18779	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_m)))
	18780	svint16_t svqsubr_n_s16_m(svbool_t, svint16_t, int16_t);
	18781	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_x)))
	18782	svint8_t svqsubr_n_s8_x(svbool_t, svint8_t, int8_t);
	18783	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_x)))
	18784	svint32_t svqsubr_n_s32_x(svbool_t, svint32_t, int32_t);
	18785	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_x)))
	18786	svint64_t svqsubr_n_s64_x(svbool_t, svint64_t, int64_t);
	18787	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_x)))
	18788	svint16_t svqsubr_n_s16_x(svbool_t, svint16_t, int16_t);
	18789	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_z)))
	18790	svint8_t svqsubr_n_s8_z(svbool_t, svint8_t, int8_t);
	18791	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_z)))
	18792	svint32_t svqsubr_n_s32_z(svbool_t, svint32_t, int32_t);
	18793	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_z)))
	18794	svint64_t svqsubr_n_s64_z(svbool_t, svint64_t, int64_t);
	18795	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_z)))
	18796	svint16_t svqsubr_n_s16_z(svbool_t, svint16_t, int16_t);
	18797	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_m)))
	18798	svuint8_t svqsubr_n_u8_m(svbool_t, svuint8_t, uint8_t);
	18799	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_m)))
	18800	svuint32_t svqsubr_n_u32_m(svbool_t, svuint32_t, uint32_t);
	18801	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_m)))
	18802	svuint64_t svqsubr_n_u64_m(svbool_t, svuint64_t, uint64_t);
	18803	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_m)))
	18804	svuint16_t svqsubr_n_u16_m(svbool_t, svuint16_t, uint16_t);
	18805	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_x)))
	18806	svuint8_t svqsubr_n_u8_x(svbool_t, svuint8_t, uint8_t);
	18807	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_x)))
	18808	svuint32_t svqsubr_n_u32_x(svbool_t, svuint32_t, uint32_t);
	18809	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_x)))
	18810	svuint64_t svqsubr_n_u64_x(svbool_t, svuint64_t, uint64_t);
	18811	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_x)))
	18812	svuint16_t svqsubr_n_u16_x(svbool_t, svuint16_t, uint16_t);
	18813	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_z)))
	18814	svuint8_t svqsubr_n_u8_z(svbool_t, svuint8_t, uint8_t);
	18815	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_z)))
	18816	svuint32_t svqsubr_n_u32_z(svbool_t, svuint32_t, uint32_t);
	18817	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_z)))
	18818	svuint64_t svqsubr_n_u64_z(svbool_t, svuint64_t, uint64_t);
	18819	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_z)))
	18820	svuint16_t svqsubr_n_u16_z(svbool_t, svuint16_t, uint16_t);
	18821	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_m)))
	18822	svint8_t svqsubr_s8_m(svbool_t, svint8_t, svint8_t);
	18823	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_m)))
	18824	svint32_t svqsubr_s32_m(svbool_t, svint32_t, svint32_t);
	18825	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_m)))
	18826	svint64_t svqsubr_s64_m(svbool_t, svint64_t, svint64_t);
	18827	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_m)))
	18828	svint16_t svqsubr_s16_m(svbool_t, svint16_t, svint16_t);
	18829	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_x)))
	18830	svint8_t svqsubr_s8_x(svbool_t, svint8_t, svint8_t);
	18831	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_x)))
	18832	svint32_t svqsubr_s32_x(svbool_t, svint32_t, svint32_t);
	18833	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_x)))
	18834	svint64_t svqsubr_s64_x(svbool_t, svint64_t, svint64_t);
	18835	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_x)))
	18836	svint16_t svqsubr_s16_x(svbool_t, svint16_t, svint16_t);
	18837	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_z)))
	18838	svint8_t svqsubr_s8_z(svbool_t, svint8_t, svint8_t);
	18839	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_z)))
	18840	svint32_t svqsubr_s32_z(svbool_t, svint32_t, svint32_t);
	18841	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_z)))
	18842	svint64_t svqsubr_s64_z(svbool_t, svint64_t, svint64_t);
	18843	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_z)))
	18844	svint16_t svqsubr_s16_z(svbool_t, svint16_t, svint16_t);
	18845	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_m)))
	18846	svuint8_t svqsubr_u8_m(svbool_t, svuint8_t, svuint8_t);
	18847	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_m)))
	18848	svuint32_t svqsubr_u32_m(svbool_t, svuint32_t, svuint32_t);
	18849	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_m)))
	18850	svuint64_t svqsubr_u64_m(svbool_t, svuint64_t, svuint64_t);
	18851	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_m)))
	18852	svuint16_t svqsubr_u16_m(svbool_t, svuint16_t, svuint16_t);
	18853	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_x)))
	18854	svuint8_t svqsubr_u8_x(svbool_t, svuint8_t, svuint8_t);
	18855	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_x)))
	18856	svuint32_t svqsubr_u32_x(svbool_t, svuint32_t, svuint32_t);
	18857	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_x)))
	18858	svuint64_t svqsubr_u64_x(svbool_t, svuint64_t, svuint64_t);
	18859	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_x)))
	18860	svuint16_t svqsubr_u16_x(svbool_t, svuint16_t, svuint16_t);
	18861	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_z)))
	18862	svuint8_t svqsubr_u8_z(svbool_t, svuint8_t, svuint8_t);
	18863	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_z)))
	18864	svuint32_t svqsubr_u32_z(svbool_t, svuint32_t, svuint32_t);
	18865	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_z)))
	18866	svuint64_t svqsubr_u64_z(svbool_t, svuint64_t, svuint64_t);
	18867	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_z)))
	18868	svuint16_t svqsubr_u16_z(svbool_t, svuint16_t, svuint16_t);
	18869	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s32)))
	18870	svint16_t svqxtnb_s32(svint32_t);
	18871	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s64)))
	18872	svint32_t svqxtnb_s64(svint64_t);
	18873	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s16)))
	18874	svint8_t svqxtnb_s16(svint16_t);
	18875	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u32)))
	18876	svuint16_t svqxtnb_u32(svuint32_t);
	18877	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u64)))
	18878	svuint32_t svqxtnb_u64(svuint64_t);
	18879	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u16)))
	18880	svuint8_t svqxtnb_u16(svuint16_t);
	18881	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s32)))
	18882	svint16_t svqxtnt_s32(svint16_t, svint32_t);
	18883	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s64)))
	18884	svint32_t svqxtnt_s64(svint32_t, svint64_t);
	18885	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s16)))
	18886	svint8_t svqxtnt_s16(svint8_t, svint16_t);
	18887	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u32)))
	18888	svuint16_t svqxtnt_u32(svuint16_t, svuint32_t);
	18889	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u64)))
	18890	svuint32_t svqxtnt_u64(svuint32_t, svuint64_t);
	18891	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u16)))
	18892	svuint8_t svqxtnt_u16(svuint8_t, svuint16_t);
	18893	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s32)))
	18894	svuint16_t svqxtunb_s32(svint32_t);
	18895	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s64)))
	18896	svuint32_t svqxtunb_s64(svint64_t);
	18897	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s16)))
	18898	svuint8_t svqxtunb_s16(svint16_t);
	18899	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s32)))
	18900	svuint16_t svqxtunt_s32(svuint16_t, svint32_t);
	18901	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s64)))
	18902	svuint32_t svqxtunt_s64(svuint32_t, svint64_t);
	18903	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s16)))
	18904	svuint8_t svqxtunt_s16(svuint8_t, svint16_t);
	18905	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u32)))
	18906	svuint16_t svraddhnb_n_u32(svuint32_t, uint32_t);
	18907	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u64)))
	18908	svuint32_t svraddhnb_n_u64(svuint64_t, uint64_t);
	18909	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u16)))
	18910	svuint8_t svraddhnb_n_u16(svuint16_t, uint16_t);
	18911	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s32)))
	18912	svint16_t svraddhnb_n_s32(svint32_t, int32_t);
	18913	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s64)))
	18914	svint32_t svraddhnb_n_s64(svint64_t, int64_t);
	18915	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s16)))
	18916	svint8_t svraddhnb_n_s16(svint16_t, int16_t);
	18917	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u32)))
	18918	svuint16_t svraddhnb_u32(svuint32_t, svuint32_t);
	18919	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u64)))
	18920	svuint32_t svraddhnb_u64(svuint64_t, svuint64_t);
	18921	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u16)))
	18922	svuint8_t svraddhnb_u16(svuint16_t, svuint16_t);
	18923	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s32)))
	18924	svint16_t svraddhnb_s32(svint32_t, svint32_t);
	18925	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s64)))
	18926	svint32_t svraddhnb_s64(svint64_t, svint64_t);
	18927	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s16)))
	18928	svint8_t svraddhnb_s16(svint16_t, svint16_t);
	18929	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u32)))
	18930	svuint16_t svraddhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
	18931	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u64)))
	18932	svuint32_t svraddhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	18933	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u16)))
	18934	svuint8_t svraddhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
	18935	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s32)))
	18936	svint16_t svraddhnt_n_s32(svint16_t, svint32_t, int32_t);
	18937	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s64)))
	18938	svint32_t svraddhnt_n_s64(svint32_t, svint64_t, int64_t);
	18939	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s16)))
	18940	svint8_t svraddhnt_n_s16(svint8_t, svint16_t, int16_t);
	18941	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u32)))
	18942	svuint16_t svraddhnt_u32(svuint16_t, svuint32_t, svuint32_t);
	18943	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u64)))
	18944	svuint32_t svraddhnt_u64(svuint32_t, svuint64_t, svuint64_t);
	18945	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u16)))
	18946	svuint8_t svraddhnt_u16(svuint8_t, svuint16_t, svuint16_t);
	18947	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s32)))
	18948	svint16_t svraddhnt_s32(svint16_t, svint32_t, svint32_t);
	18949	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s64)))
	18950	svint32_t svraddhnt_s64(svint32_t, svint64_t, svint64_t);
	18951	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s16)))
	18952	svint8_t svraddhnt_s16(svint8_t, svint16_t, svint16_t);
	18953	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_m)))
	18954	svuint32_t svrecpe_u32_m(svuint32_t, svbool_t, svuint32_t);
	18955	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_x)))
	18956	svuint32_t svrecpe_u32_x(svbool_t, svuint32_t);
	18957	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_z)))
	18958	svuint32_t svrecpe_u32_z(svbool_t, svuint32_t);
	18959	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_m)))
	18960	svint8_t svrhadd_n_s8_m(svbool_t, svint8_t, int8_t);
	18961	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_m)))
	18962	svint32_t svrhadd_n_s32_m(svbool_t, svint32_t, int32_t);
	18963	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_m)))
	18964	svint64_t svrhadd_n_s64_m(svbool_t, svint64_t, int64_t);
	18965	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_m)))
	18966	svint16_t svrhadd_n_s16_m(svbool_t, svint16_t, int16_t);
	18967	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_x)))
	18968	svint8_t svrhadd_n_s8_x(svbool_t, svint8_t, int8_t);
	18969	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_x)))
	18970	svint32_t svrhadd_n_s32_x(svbool_t, svint32_t, int32_t);
	18971	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_x)))
	18972	svint64_t svrhadd_n_s64_x(svbool_t, svint64_t, int64_t);
	18973	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_x)))
	18974	svint16_t svrhadd_n_s16_x(svbool_t, svint16_t, int16_t);
	18975	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_z)))
	18976	svint8_t svrhadd_n_s8_z(svbool_t, svint8_t, int8_t);
	18977	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_z)))
	18978	svint32_t svrhadd_n_s32_z(svbool_t, svint32_t, int32_t);
	18979	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_z)))
	18980	svint64_t svrhadd_n_s64_z(svbool_t, svint64_t, int64_t);
	18981	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_z)))
	18982	svint16_t svrhadd_n_s16_z(svbool_t, svint16_t, int16_t);
	18983	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_m)))
	18984	svuint8_t svrhadd_n_u8_m(svbool_t, svuint8_t, uint8_t);
	18985	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_m)))
	18986	svuint32_t svrhadd_n_u32_m(svbool_t, svuint32_t, uint32_t);
	18987	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_m)))
	18988	svuint64_t svrhadd_n_u64_m(svbool_t, svuint64_t, uint64_t);
	18989	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_m)))
	18990	svuint16_t svrhadd_n_u16_m(svbool_t, svuint16_t, uint16_t);
	18991	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_x)))
	18992	svuint8_t svrhadd_n_u8_x(svbool_t, svuint8_t, uint8_t);
	18993	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_x)))
	18994	svuint32_t svrhadd_n_u32_x(svbool_t, svuint32_t, uint32_t);
	18995	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_x)))
	18996	svuint64_t svrhadd_n_u64_x(svbool_t, svuint64_t, uint64_t);
	18997	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_x)))
	18998	svuint16_t svrhadd_n_u16_x(svbool_t, svuint16_t, uint16_t);
	18999	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_z)))
	19000	svuint8_t svrhadd_n_u8_z(svbool_t, svuint8_t, uint8_t);
	19001	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_z)))
	19002	svuint32_t svrhadd_n_u32_z(svbool_t, svuint32_t, uint32_t);
	19003	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_z)))
	19004	svuint64_t svrhadd_n_u64_z(svbool_t, svuint64_t, uint64_t);
	19005	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_z)))
	19006	svuint16_t svrhadd_n_u16_z(svbool_t, svuint16_t, uint16_t);
	19007	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_m)))
	19008	svint8_t svrhadd_s8_m(svbool_t, svint8_t, svint8_t);
	19009	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_m)))
	19010	svint32_t svrhadd_s32_m(svbool_t, svint32_t, svint32_t);
	19011	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_m)))
	19012	svint64_t svrhadd_s64_m(svbool_t, svint64_t, svint64_t);
	19013	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_m)))
	19014	svint16_t svrhadd_s16_m(svbool_t, svint16_t, svint16_t);
	19015	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_x)))
	19016	svint8_t svrhadd_s8_x(svbool_t, svint8_t, svint8_t);
	19017	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_x)))
	19018	svint32_t svrhadd_s32_x(svbool_t, svint32_t, svint32_t);
	19019	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_x)))
	19020	svint64_t svrhadd_s64_x(svbool_t, svint64_t, svint64_t);
	19021	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_x)))
	19022	svint16_t svrhadd_s16_x(svbool_t, svint16_t, svint16_t);
	19023	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_z)))
	19024	svint8_t svrhadd_s8_z(svbool_t, svint8_t, svint8_t);
	19025	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_z)))
	19026	svint32_t svrhadd_s32_z(svbool_t, svint32_t, svint32_t);
	19027	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_z)))
	19028	svint64_t svrhadd_s64_z(svbool_t, svint64_t, svint64_t);
	19029	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_z)))
	19030	svint16_t svrhadd_s16_z(svbool_t, svint16_t, svint16_t);
	19031	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_m)))
	19032	svuint8_t svrhadd_u8_m(svbool_t, svuint8_t, svuint8_t);
	19033	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_m)))
	19034	svuint32_t svrhadd_u32_m(svbool_t, svuint32_t, svuint32_t);
	19035	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_m)))
	19036	svuint64_t svrhadd_u64_m(svbool_t, svuint64_t, svuint64_t);
	19037	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_m)))
	19038	svuint16_t svrhadd_u16_m(svbool_t, svuint16_t, svuint16_t);
	19039	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_x)))
	19040	svuint8_t svrhadd_u8_x(svbool_t, svuint8_t, svuint8_t);
	19041	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_x)))
	19042	svuint32_t svrhadd_u32_x(svbool_t, svuint32_t, svuint32_t);
	19043	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_x)))
	19044	svuint64_t svrhadd_u64_x(svbool_t, svuint64_t, svuint64_t);
	19045	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_x)))
	19046	svuint16_t svrhadd_u16_x(svbool_t, svuint16_t, svuint16_t);
	19047	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_z)))
	19048	svuint8_t svrhadd_u8_z(svbool_t, svuint8_t, svuint8_t);
	19049	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_z)))
	19050	svuint32_t svrhadd_u32_z(svbool_t, svuint32_t, svuint32_t);
	19051	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_z)))
	19052	svuint64_t svrhadd_u64_z(svbool_t, svuint64_t, svuint64_t);
	19053	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_z)))
	19054	svuint16_t svrhadd_u16_z(svbool_t, svuint16_t, svuint16_t);
	19055	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_m)))
	19056	svint8_t svrshl_n_s8_m(svbool_t, svint8_t, int8_t);
	19057	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_m)))
	19058	svint32_t svrshl_n_s32_m(svbool_t, svint32_t, int32_t);
	19059	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_m)))
	19060	svint64_t svrshl_n_s64_m(svbool_t, svint64_t, int64_t);
	19061	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_m)))
	19062	svint16_t svrshl_n_s16_m(svbool_t, svint16_t, int16_t);
	19063	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_x)))
	19064	svint8_t svrshl_n_s8_x(svbool_t, svint8_t, int8_t);
	19065	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_x)))
	19066	svint32_t svrshl_n_s32_x(svbool_t, svint32_t, int32_t);
	19067	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_x)))
	19068	svint64_t svrshl_n_s64_x(svbool_t, svint64_t, int64_t);
	19069	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_x)))
	19070	svint16_t svrshl_n_s16_x(svbool_t, svint16_t, int16_t);
	19071	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_z)))
	19072	svint8_t svrshl_n_s8_z(svbool_t, svint8_t, int8_t);
	19073	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_z)))
	19074	svint32_t svrshl_n_s32_z(svbool_t, svint32_t, int32_t);
	19075	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_z)))
	19076	svint64_t svrshl_n_s64_z(svbool_t, svint64_t, int64_t);
	19077	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_z)))
	19078	svint16_t svrshl_n_s16_z(svbool_t, svint16_t, int16_t);
	19079	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_m)))
	19080	svuint8_t svrshl_n_u8_m(svbool_t, svuint8_t, int8_t);
	19081	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_m)))
	19082	svuint32_t svrshl_n_u32_m(svbool_t, svuint32_t, int32_t);
	19083	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_m)))
	19084	svuint64_t svrshl_n_u64_m(svbool_t, svuint64_t, int64_t);
	19085	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_m)))
	19086	svuint16_t svrshl_n_u16_m(svbool_t, svuint16_t, int16_t);
	19087	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_x)))
	19088	svuint8_t svrshl_n_u8_x(svbool_t, svuint8_t, int8_t);
	19089	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_x)))
	19090	svuint32_t svrshl_n_u32_x(svbool_t, svuint32_t, int32_t);
	19091	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_x)))
	19092	svuint64_t svrshl_n_u64_x(svbool_t, svuint64_t, int64_t);
	19093	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_x)))
	19094	svuint16_t svrshl_n_u16_x(svbool_t, svuint16_t, int16_t);
	19095	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_z)))
	19096	svuint8_t svrshl_n_u8_z(svbool_t, svuint8_t, int8_t);
	19097	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_z)))
	19098	svuint32_t svrshl_n_u32_z(svbool_t, svuint32_t, int32_t);
	19099	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_z)))
	19100	svuint64_t svrshl_n_u64_z(svbool_t, svuint64_t, int64_t);
	19101	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_z)))
	19102	svuint16_t svrshl_n_u16_z(svbool_t, svuint16_t, int16_t);
	19103	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_m)))
	19104	svint8_t svrshl_s8_m(svbool_t, svint8_t, svint8_t);
	19105	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_m)))
	19106	svint32_t svrshl_s32_m(svbool_t, svint32_t, svint32_t);
	19107	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_m)))
	19108	svint64_t svrshl_s64_m(svbool_t, svint64_t, svint64_t);
	19109	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_m)))
	19110	svint16_t svrshl_s16_m(svbool_t, svint16_t, svint16_t);
	19111	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_x)))
	19112	svint8_t svrshl_s8_x(svbool_t, svint8_t, svint8_t);
	19113	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_x)))
	19114	svint32_t svrshl_s32_x(svbool_t, svint32_t, svint32_t);
	19115	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_x)))
	19116	svint64_t svrshl_s64_x(svbool_t, svint64_t, svint64_t);
	19117	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_x)))
	19118	svint16_t svrshl_s16_x(svbool_t, svint16_t, svint16_t);
	19119	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_z)))
	19120	svint8_t svrshl_s8_z(svbool_t, svint8_t, svint8_t);
	19121	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_z)))
	19122	svint32_t svrshl_s32_z(svbool_t, svint32_t, svint32_t);
	19123	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_z)))
	19124	svint64_t svrshl_s64_z(svbool_t, svint64_t, svint64_t);
	19125	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_z)))
	19126	svint16_t svrshl_s16_z(svbool_t, svint16_t, svint16_t);
	19127	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_m)))
	19128	svuint8_t svrshl_u8_m(svbool_t, svuint8_t, svint8_t);
	19129	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_m)))
	19130	svuint32_t svrshl_u32_m(svbool_t, svuint32_t, svint32_t);
	19131	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_m)))
	19132	svuint64_t svrshl_u64_m(svbool_t, svuint64_t, svint64_t);
	19133	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_m)))
	19134	svuint16_t svrshl_u16_m(svbool_t, svuint16_t, svint16_t);
	19135	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_x)))
	19136	svuint8_t svrshl_u8_x(svbool_t, svuint8_t, svint8_t);
	19137	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_x)))
	19138	svuint32_t svrshl_u32_x(svbool_t, svuint32_t, svint32_t);
	19139	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_x)))
	19140	svuint64_t svrshl_u64_x(svbool_t, svuint64_t, svint64_t);
	19141	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_x)))
	19142	svuint16_t svrshl_u16_x(svbool_t, svuint16_t, svint16_t);
	19143	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_z)))
	19144	svuint8_t svrshl_u8_z(svbool_t, svuint8_t, svint8_t);
	19145	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_z)))
	19146	svuint32_t svrshl_u32_z(svbool_t, svuint32_t, svint32_t);
	19147	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_z)))
	19148	svuint64_t svrshl_u64_z(svbool_t, svuint64_t, svint64_t);
	19149	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_z)))
	19150	svuint16_t svrshl_u16_z(svbool_t, svuint16_t, svint16_t);
	19151	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_m)))
	19152	svint8_t svrshr_n_s8_m(svbool_t, svint8_t, uint64_t);
	19153	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_m)))
	19154	svint32_t svrshr_n_s32_m(svbool_t, svint32_t, uint64_t);
	19155	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_m)))
	19156	svint64_t svrshr_n_s64_m(svbool_t, svint64_t, uint64_t);
	19157	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_m)))
	19158	svint16_t svrshr_n_s16_m(svbool_t, svint16_t, uint64_t);
	19159	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_m)))
	19160	svuint8_t svrshr_n_u8_m(svbool_t, svuint8_t, uint64_t);
	19161	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_m)))
	19162	svuint32_t svrshr_n_u32_m(svbool_t, svuint32_t, uint64_t);
	19163	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_m)))
	19164	svuint64_t svrshr_n_u64_m(svbool_t, svuint64_t, uint64_t);
	19165	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_m)))
	19166	svuint16_t svrshr_n_u16_m(svbool_t, svuint16_t, uint64_t);
	19167	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_x)))
	19168	svint8_t svrshr_n_s8_x(svbool_t, svint8_t, uint64_t);
	19169	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_x)))
	19170	svint32_t svrshr_n_s32_x(svbool_t, svint32_t, uint64_t);
	19171	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_x)))
	19172	svint64_t svrshr_n_s64_x(svbool_t, svint64_t, uint64_t);
	19173	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_x)))
	19174	svint16_t svrshr_n_s16_x(svbool_t, svint16_t, uint64_t);
	19175	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_x)))
	19176	svuint8_t svrshr_n_u8_x(svbool_t, svuint8_t, uint64_t);
	19177	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_x)))
	19178	svuint32_t svrshr_n_u32_x(svbool_t, svuint32_t, uint64_t);
	19179	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_x)))
	19180	svuint64_t svrshr_n_u64_x(svbool_t, svuint64_t, uint64_t);
	19181	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_x)))
	19182	svuint16_t svrshr_n_u16_x(svbool_t, svuint16_t, uint64_t);
	19183	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_z)))
	19184	svint8_t svrshr_n_s8_z(svbool_t, svint8_t, uint64_t);
	19185	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_z)))
	19186	svint32_t svrshr_n_s32_z(svbool_t, svint32_t, uint64_t);
	19187	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_z)))
	19188	svint64_t svrshr_n_s64_z(svbool_t, svint64_t, uint64_t);
	19189	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_z)))
	19190	svint16_t svrshr_n_s16_z(svbool_t, svint16_t, uint64_t);
	19191	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_z)))
	19192	svuint8_t svrshr_n_u8_z(svbool_t, svuint8_t, uint64_t);
	19193	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_z)))
	19194	svuint32_t svrshr_n_u32_z(svbool_t, svuint32_t, uint64_t);
	19195	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_z)))
	19196	svuint64_t svrshr_n_u64_z(svbool_t, svuint64_t, uint64_t);
	19197	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_z)))
	19198	svuint16_t svrshr_n_u16_z(svbool_t, svuint16_t, uint64_t);
	19199	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u32)))
	19200	svuint16_t svrshrnb_n_u32(svuint32_t, uint64_t);
	19201	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u64)))
	19202	svuint32_t svrshrnb_n_u64(svuint64_t, uint64_t);
	19203	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u16)))
	19204	svuint8_t svrshrnb_n_u16(svuint16_t, uint64_t);
	19205	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s32)))
	19206	svint16_t svrshrnb_n_s32(svint32_t, uint64_t);
	19207	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s64)))
	19208	svint32_t svrshrnb_n_s64(svint64_t, uint64_t);
	19209	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s16)))
	19210	svint8_t svrshrnb_n_s16(svint16_t, uint64_t);
	19211	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u32)))
	19212	svuint16_t svrshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
	19213	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u64)))
	19214	svuint32_t svrshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	19215	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u16)))
	19216	svuint8_t svrshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
	19217	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s32)))
	19218	svint16_t svrshrnt_n_s32(svint16_t, svint32_t, uint64_t);
	19219	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s64)))
	19220	svint32_t svrshrnt_n_s64(svint32_t, svint64_t, uint64_t);
	19221	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s16)))
	19222	svint8_t svrshrnt_n_s16(svint8_t, svint16_t, uint64_t);
	19223	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_m)))
	19224	svuint32_t svrsqrte_u32_m(svuint32_t, svbool_t, svuint32_t);
	19225	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_x)))
	19226	svuint32_t svrsqrte_u32_x(svbool_t, svuint32_t);
	19227	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_z)))
	19228	svuint32_t svrsqrte_u32_z(svbool_t, svuint32_t);
	19229	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s8)))
	19230	svint8_t svrsra_n_s8(svint8_t, svint8_t, uint64_t);
	19231	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s32)))
	19232	svint32_t svrsra_n_s32(svint32_t, svint32_t, uint64_t);
	19233	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s64)))
	19234	svint64_t svrsra_n_s64(svint64_t, svint64_t, uint64_t);
	19235	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s16)))
	19236	svint16_t svrsra_n_s16(svint16_t, svint16_t, uint64_t);
	19237	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u8)))
	19238	svuint8_t svrsra_n_u8(svuint8_t, svuint8_t, uint64_t);
	19239	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u32)))
	19240	svuint32_t svrsra_n_u32(svuint32_t, svuint32_t, uint64_t);
	19241	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u64)))
	19242	svuint64_t svrsra_n_u64(svuint64_t, svuint64_t, uint64_t);
	19243	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u16)))
	19244	svuint16_t svrsra_n_u16(svuint16_t, svuint16_t, uint64_t);
	19245	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u32)))
	19246	svuint16_t svrsubhnb_n_u32(svuint32_t, uint32_t);
	19247	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u64)))
	19248	svuint32_t svrsubhnb_n_u64(svuint64_t, uint64_t);
	19249	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u16)))
	19250	svuint8_t svrsubhnb_n_u16(svuint16_t, uint16_t);
	19251	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s32)))
	19252	svint16_t svrsubhnb_n_s32(svint32_t, int32_t);
	19253	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s64)))
	19254	svint32_t svrsubhnb_n_s64(svint64_t, int64_t);
	19255	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s16)))
	19256	svint8_t svrsubhnb_n_s16(svint16_t, int16_t);
	19257	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u32)))
	19258	svuint16_t svrsubhnb_u32(svuint32_t, svuint32_t);
	19259	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u64)))
	19260	svuint32_t svrsubhnb_u64(svuint64_t, svuint64_t);
	19261	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u16)))
	19262	svuint8_t svrsubhnb_u16(svuint16_t, svuint16_t);
	19263	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s32)))
	19264	svint16_t svrsubhnb_s32(svint32_t, svint32_t);
	19265	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s64)))
	19266	svint32_t svrsubhnb_s64(svint64_t, svint64_t);
	19267	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s16)))
	19268	svint8_t svrsubhnb_s16(svint16_t, svint16_t);
	19269	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u32)))
	19270	svuint16_t svrsubhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
	19271	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u64)))
	19272	svuint32_t svrsubhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	19273	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u16)))
	19274	svuint8_t svrsubhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
	19275	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s32)))
	19276	svint16_t svrsubhnt_n_s32(svint16_t, svint32_t, int32_t);
	19277	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s64)))
	19278	svint32_t svrsubhnt_n_s64(svint32_t, svint64_t, int64_t);
	19279	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s16)))
	19280	svint8_t svrsubhnt_n_s16(svint8_t, svint16_t, int16_t);
	19281	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u32)))
	19282	svuint16_t svrsubhnt_u32(svuint16_t, svuint32_t, svuint32_t);
	19283	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u64)))
	19284	svuint32_t svrsubhnt_u64(svuint32_t, svuint64_t, svuint64_t);
	19285	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u16)))
	19286	svuint8_t svrsubhnt_u16(svuint8_t, svuint16_t, svuint16_t);
	19287	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s32)))
	19288	svint16_t svrsubhnt_s32(svint16_t, svint32_t, svint32_t);
	19289	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s64)))
	19290	svint32_t svrsubhnt_s64(svint32_t, svint64_t, svint64_t);
	19291	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s16)))
	19292	svint8_t svrsubhnt_s16(svint8_t, svint16_t, svint16_t);
	19293	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u32)))
	19294	svuint32_t svsbclb_n_u32(svuint32_t, svuint32_t, uint32_t);
	19295	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u64)))
	19296	svuint64_t svsbclb_n_u64(svuint64_t, svuint64_t, uint64_t);
	19297	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u32)))
	19298	svuint32_t svsbclb_u32(svuint32_t, svuint32_t, svuint32_t);
	19299	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u64)))
	19300	svuint64_t svsbclb_u64(svuint64_t, svuint64_t, svuint64_t);
	19301	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u32)))
	19302	svuint32_t svsbclt_n_u32(svuint32_t, svuint32_t, uint32_t);
	19303	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u64)))
	19304	svuint64_t svsbclt_n_u64(svuint64_t, svuint64_t, uint64_t);
	19305	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u32)))
	19306	svuint32_t svsbclt_u32(svuint32_t, svuint32_t, svuint32_t);
	19307	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u64)))
	19308	svuint64_t svsbclt_u64(svuint64_t, svuint64_t, svuint64_t);
	19309	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s32)))
	19310	svint32_t svshllb_n_s32(svint16_t, uint64_t);
	19311	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s64)))
	19312	svint64_t svshllb_n_s64(svint32_t, uint64_t);
	19313	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s16)))
	19314	svint16_t svshllb_n_s16(svint8_t, uint64_t);
	19315	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u32)))
	19316	svuint32_t svshllb_n_u32(svuint16_t, uint64_t);
	19317	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u64)))
	19318	svuint64_t svshllb_n_u64(svuint32_t, uint64_t);
	19319	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u16)))
	19320	svuint16_t svshllb_n_u16(svuint8_t, uint64_t);
	19321	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s32)))
	19322	svint32_t svshllt_n_s32(svint16_t, uint64_t);
	19323	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s64)))
	19324	svint64_t svshllt_n_s64(svint32_t, uint64_t);
	19325	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s16)))
	19326	svint16_t svshllt_n_s16(svint8_t, uint64_t);
	19327	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u32)))
	19328	svuint32_t svshllt_n_u32(svuint16_t, uint64_t);
	19329	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u64)))
	19330	svuint64_t svshllt_n_u64(svuint32_t, uint64_t);
	19331	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u16)))
	19332	svuint16_t svshllt_n_u16(svuint8_t, uint64_t);
	19333	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u32)))
	19334	svuint16_t svshrnb_n_u32(svuint32_t, uint64_t);
	19335	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u64)))
	19336	svuint32_t svshrnb_n_u64(svuint64_t, uint64_t);
	19337	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u16)))
	19338	svuint8_t svshrnb_n_u16(svuint16_t, uint64_t);
	19339	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s32)))
	19340	svint16_t svshrnb_n_s32(svint32_t, uint64_t);
	19341	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s64)))
	19342	svint32_t svshrnb_n_s64(svint64_t, uint64_t);
	19343	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s16)))
	19344	svint8_t svshrnb_n_s16(svint16_t, uint64_t);
	19345	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u32)))
	19346	svuint16_t svshrnt_n_u32(svuint16_t, svuint32_t, uint64_t);
	19347	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u64)))
	19348	svuint32_t svshrnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	19349	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u16)))
	19350	svuint8_t svshrnt_n_u16(svuint8_t, svuint16_t, uint64_t);
	19351	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s32)))
	19352	svint16_t svshrnt_n_s32(svint16_t, svint32_t, uint64_t);
	19353	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s64)))
	19354	svint32_t svshrnt_n_s64(svint32_t, svint64_t, uint64_t);
	19355	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s16)))
	19356	svint8_t svshrnt_n_s16(svint8_t, svint16_t, uint64_t);
	19357	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u8)))
	19358	svuint8_t svsli_n_u8(svuint8_t, svuint8_t, uint64_t);
	19359	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u32)))
	19360	svuint32_t svsli_n_u32(svuint32_t, svuint32_t, uint64_t);
	19361	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u64)))
	19362	svuint64_t svsli_n_u64(svuint64_t, svuint64_t, uint64_t);
	19363	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u16)))
	19364	svuint16_t svsli_n_u16(svuint16_t, svuint16_t, uint64_t);
	19365	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s8)))
	19366	svint8_t svsli_n_s8(svint8_t, svint8_t, uint64_t);
	19367	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s32)))
	19368	svint32_t svsli_n_s32(svint32_t, svint32_t, uint64_t);
	19369	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s64)))
	19370	svint64_t svsli_n_s64(svint64_t, svint64_t, uint64_t);
	19371	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s16)))
	19372	svint16_t svsli_n_s16(svint16_t, svint16_t, uint64_t);
	19373	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_m)))
	19374	svuint8_t svsqadd_n_u8_m(svbool_t, svuint8_t, int8_t);
	19375	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_m)))
	19376	svuint32_t svsqadd_n_u32_m(svbool_t, svuint32_t, int32_t);
	19377	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_m)))
	19378	svuint64_t svsqadd_n_u64_m(svbool_t, svuint64_t, int64_t);
	19379	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_m)))
	19380	svuint16_t svsqadd_n_u16_m(svbool_t, svuint16_t, int16_t);
	19381	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_x)))
	19382	svuint8_t svsqadd_n_u8_x(svbool_t, svuint8_t, int8_t);
	19383	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_x)))
	19384	svuint32_t svsqadd_n_u32_x(svbool_t, svuint32_t, int32_t);
	19385	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_x)))
	19386	svuint64_t svsqadd_n_u64_x(svbool_t, svuint64_t, int64_t);
	19387	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_x)))
	19388	svuint16_t svsqadd_n_u16_x(svbool_t, svuint16_t, int16_t);
	19389	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_z)))
	19390	svuint8_t svsqadd_n_u8_z(svbool_t, svuint8_t, int8_t);
	19391	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_z)))
	19392	svuint32_t svsqadd_n_u32_z(svbool_t, svuint32_t, int32_t);
	19393	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_z)))
	19394	svuint64_t svsqadd_n_u64_z(svbool_t, svuint64_t, int64_t);
	19395	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_z)))
	19396	svuint16_t svsqadd_n_u16_z(svbool_t, svuint16_t, int16_t);
	19397	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_m)))
	19398	svuint8_t svsqadd_u8_m(svbool_t, svuint8_t, svint8_t);
	19399	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_m)))
	19400	svuint32_t svsqadd_u32_m(svbool_t, svuint32_t, svint32_t);
	19401	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_m)))
	19402	svuint64_t svsqadd_u64_m(svbool_t, svuint64_t, svint64_t);
	19403	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_m)))
	19404	svuint16_t svsqadd_u16_m(svbool_t, svuint16_t, svint16_t);
	19405	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_x)))
	19406	svuint8_t svsqadd_u8_x(svbool_t, svuint8_t, svint8_t);
	19407	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_x)))
	19408	svuint32_t svsqadd_u32_x(svbool_t, svuint32_t, svint32_t);
	19409	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_x)))
	19410	svuint64_t svsqadd_u64_x(svbool_t, svuint64_t, svint64_t);
	19411	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_x)))
	19412	svuint16_t svsqadd_u16_x(svbool_t, svuint16_t, svint16_t);
	19413	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_z)))
	19414	svuint8_t svsqadd_u8_z(svbool_t, svuint8_t, svint8_t);
	19415	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_z)))
	19416	svuint32_t svsqadd_u32_z(svbool_t, svuint32_t, svint32_t);
	19417	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_z)))
	19418	svuint64_t svsqadd_u64_z(svbool_t, svuint64_t, svint64_t);
	19419	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_z)))
	19420	svuint16_t svsqadd_u16_z(svbool_t, svuint16_t, svint16_t);
	19421	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s8)))
	19422	svint8_t svsra_n_s8(svint8_t, svint8_t, uint64_t);
	19423	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s32)))
	19424	svint32_t svsra_n_s32(svint32_t, svint32_t, uint64_t);
	19425	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s64)))
	19426	svint64_t svsra_n_s64(svint64_t, svint64_t, uint64_t);
	19427	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s16)))
	19428	svint16_t svsra_n_s16(svint16_t, svint16_t, uint64_t);
	19429	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u8)))
	19430	svuint8_t svsra_n_u8(svuint8_t, svuint8_t, uint64_t);
	19431	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u32)))
	19432	svuint32_t svsra_n_u32(svuint32_t, svuint32_t, uint64_t);
	19433	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u64)))
	19434	svuint64_t svsra_n_u64(svuint64_t, svuint64_t, uint64_t);
	19435	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u16)))
	19436	svuint16_t svsra_n_u16(svuint16_t, svuint16_t, uint64_t);
	19437	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u8)))
	19438	svuint8_t svsri_n_u8(svuint8_t, svuint8_t, uint64_t);
	19439	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u32)))
	19440	svuint32_t svsri_n_u32(svuint32_t, svuint32_t, uint64_t);
	19441	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u64)))
	19442	svuint64_t svsri_n_u64(svuint64_t, svuint64_t, uint64_t);
	19443	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u16)))
	19444	svuint16_t svsri_n_u16(svuint16_t, svuint16_t, uint64_t);
	19445	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s8)))
	19446	svint8_t svsri_n_s8(svint8_t, svint8_t, uint64_t);
	19447	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s32)))
	19448	svint32_t svsri_n_s32(svint32_t, svint32_t, uint64_t);
	19449	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s64)))
	19450	svint64_t svsri_n_s64(svint64_t, svint64_t, uint64_t);
	19451	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s16)))
	19452	svint16_t svsri_n_s16(svint16_t, svint16_t, uint64_t);
	19453	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_u32)))
	19454	void svstnt1_scatter_u32base_index_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
	19455	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_u64)))
	19456	void svstnt1_scatter_u64base_index_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19457	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_f64)))
	19458	void svstnt1_scatter_u64base_index_f64(svbool_t, svuint64_t, int64_t, svfloat64_t);
	19459	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_f32)))
	19460	void svstnt1_scatter_u32base_index_f32(svbool_t, svuint32_t, int64_t, svfloat32_t);
	19461	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_s32)))
	19462	void svstnt1_scatter_u32base_index_s32(svbool_t, svuint32_t, int64_t, svint32_t);
	19463	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_s64)))
	19464	void svstnt1_scatter_u64base_index_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19465	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_u32)))
	19466	void svstnt1_scatter_u32base_offset_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
	19467	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_u64)))
	19468	void svstnt1_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19469	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_f64)))
	19470	void svstnt1_scatter_u64base_offset_f64(svbool_t, svuint64_t, int64_t, svfloat64_t);
	19471	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_f32)))
	19472	void svstnt1_scatter_u32base_offset_f32(svbool_t, svuint32_t, int64_t, svfloat32_t);
	19473	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_s32)))
	19474	void svstnt1_scatter_u32base_offset_s32(svbool_t, svuint32_t, int64_t, svint32_t);
	19475	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_s64)))
	19476	void svstnt1_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19477	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_u32)))
	19478	void svstnt1_scatter_u32base_u32(svbool_t, svuint32_t, svuint32_t);
	19479	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_u64)))
	19480	void svstnt1_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
	19481	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_f64)))
	19482	void svstnt1_scatter_u64base_f64(svbool_t, svuint64_t, svfloat64_t);
	19483	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_f32)))
	19484	void svstnt1_scatter_u32base_f32(svbool_t, svuint32_t, svfloat32_t);
	19485	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_s32)))
	19486	void svstnt1_scatter_u32base_s32(svbool_t, svuint32_t, svint32_t);
	19487	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_s64)))
	19488	void svstnt1_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
	19489	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_u64)))
	19490	void svstnt1_scatter_s64index_u64(svbool_t, uint64_t *, svint64_t, svuint64_t);
	19491	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_f64)))
	19492	void svstnt1_scatter_s64index_f64(svbool_t, float64_t *, svint64_t, svfloat64_t);
	19493	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_s64)))
	19494	void svstnt1_scatter_s64index_s64(svbool_t, int64_t *, svint64_t, svint64_t);
	19495	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_u64)))
	19496	void svstnt1_scatter_u64index_u64(svbool_t, uint64_t *, svuint64_t, svuint64_t);
	19497	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_f64)))
	19498	void svstnt1_scatter_u64index_f64(svbool_t, float64_t *, svuint64_t, svfloat64_t);
	19499	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_s64)))
	19500	void svstnt1_scatter_u64index_s64(svbool_t, int64_t *, svuint64_t, svint64_t);
	19501	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_u32)))
	19502	void svstnt1_scatter_u32offset_u32(svbool_t, uint32_t *, svuint32_t, svuint32_t);
	19503	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_f32)))
	19504	void svstnt1_scatter_u32offset_f32(svbool_t, float32_t *, svuint32_t, svfloat32_t);
	19505	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_s32)))
	19506	void svstnt1_scatter_u32offset_s32(svbool_t, int32_t *, svuint32_t, svint32_t);
	19507	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_u64)))
	19508	void svstnt1_scatter_s64offset_u64(svbool_t, uint64_t *, svint64_t, svuint64_t);
	19509	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_f64)))
	19510	void svstnt1_scatter_s64offset_f64(svbool_t, float64_t *, svint64_t, svfloat64_t);
	19511	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_s64)))
	19512	void svstnt1_scatter_s64offset_s64(svbool_t, int64_t *, svint64_t, svint64_t);
	19513	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_u64)))
	19514	void svstnt1_scatter_u64offset_u64(svbool_t, uint64_t *, svuint64_t, svuint64_t);
	19515	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_f64)))
	19516	void svstnt1_scatter_u64offset_f64(svbool_t, float64_t *, svuint64_t, svfloat64_t);
	19517	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_s64)))
	19518	void svstnt1_scatter_u64offset_s64(svbool_t, int64_t *, svuint64_t, svint64_t);
	19519	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_u32)))
	19520	void svstnt1b_scatter_u32base_offset_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
	19521	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_u64)))
	19522	void svstnt1b_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19523	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_s32)))
	19524	void svstnt1b_scatter_u32base_offset_s32(svbool_t, svuint32_t, int64_t, svint32_t);
	19525	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_s64)))
	19526	void svstnt1b_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19527	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_u32)))
	19528	void svstnt1b_scatter_u32base_u32(svbool_t, svuint32_t, svuint32_t);
	19529	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_u64)))
	19530	void svstnt1b_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
	19531	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_s32)))
	19532	void svstnt1b_scatter_u32base_s32(svbool_t, svuint32_t, svint32_t);
	19533	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_s64)))
	19534	void svstnt1b_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
	19535	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_s32)))
	19536	void svstnt1b_scatter_u32offset_s32(svbool_t, int8_t *, svuint32_t, svint32_t);
	19537	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_u32)))
	19538	void svstnt1b_scatter_u32offset_u32(svbool_t, uint8_t *, svuint32_t, svuint32_t);
	19539	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_s64)))
	19540	void svstnt1b_scatter_s64offset_s64(svbool_t, int8_t *, svint64_t, svint64_t);
	19541	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_u64)))
	19542	void svstnt1b_scatter_s64offset_u64(svbool_t, uint8_t *, svint64_t, svuint64_t);
	19543	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_s64)))
	19544	void svstnt1b_scatter_u64offset_s64(svbool_t, int8_t *, svuint64_t, svint64_t);
	19545	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_u64)))
	19546	void svstnt1b_scatter_u64offset_u64(svbool_t, uint8_t *, svuint64_t, svuint64_t);
	19547	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_u32)))
	19548	void svstnt1h_scatter_u32base_index_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
	19549	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_u64)))
	19550	void svstnt1h_scatter_u64base_index_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19551	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_s32)))
	19552	void svstnt1h_scatter_u32base_index_s32(svbool_t, svuint32_t, int64_t, svint32_t);
	19553	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_s64)))
	19554	void svstnt1h_scatter_u64base_index_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19555	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_u32)))
	19556	void svstnt1h_scatter_u32base_offset_u32(svbool_t, svuint32_t, int64_t, svuint32_t);
	19557	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_u64)))
	19558	void svstnt1h_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19559	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_s32)))
	19560	void svstnt1h_scatter_u32base_offset_s32(svbool_t, svuint32_t, int64_t, svint32_t);
	19561	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_s64)))
	19562	void svstnt1h_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19563	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_u32)))
	19564	void svstnt1h_scatter_u32base_u32(svbool_t, svuint32_t, svuint32_t);
	19565	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_u64)))
	19566	void svstnt1h_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
	19567	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_s32)))
	19568	void svstnt1h_scatter_u32base_s32(svbool_t, svuint32_t, svint32_t);
	19569	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_s64)))
	19570	void svstnt1h_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
	19571	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_s64)))
	19572	void svstnt1h_scatter_s64index_s64(svbool_t, int16_t *, svint64_t, svint64_t);
	19573	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_u64)))
	19574	void svstnt1h_scatter_s64index_u64(svbool_t, uint16_t *, svint64_t, svuint64_t);
	19575	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_s64)))
	19576	void svstnt1h_scatter_u64index_s64(svbool_t, int16_t *, svuint64_t, svint64_t);
	19577	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_u64)))
	19578	void svstnt1h_scatter_u64index_u64(svbool_t, uint16_t *, svuint64_t, svuint64_t);
	19579	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_s32)))
	19580	void svstnt1h_scatter_u32offset_s32(svbool_t, int16_t *, svuint32_t, svint32_t);
	19581	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_u32)))
	19582	void svstnt1h_scatter_u32offset_u32(svbool_t, uint16_t *, svuint32_t, svuint32_t);
	19583	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_s64)))
	19584	void svstnt1h_scatter_s64offset_s64(svbool_t, int16_t *, svint64_t, svint64_t);
	19585	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_u64)))
	19586	void svstnt1h_scatter_s64offset_u64(svbool_t, uint16_t *, svint64_t, svuint64_t);
	19587	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_s64)))
	19588	void svstnt1h_scatter_u64offset_s64(svbool_t, int16_t *, svuint64_t, svint64_t);
	19589	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_u64)))
	19590	void svstnt1h_scatter_u64offset_u64(svbool_t, uint16_t *, svuint64_t, svuint64_t);
	19591	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_u64)))
	19592	void svstnt1w_scatter_u64base_index_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19593	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_s64)))
	19594	void svstnt1w_scatter_u64base_index_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19595	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_u64)))
	19596	void svstnt1w_scatter_u64base_offset_u64(svbool_t, svuint64_t, int64_t, svuint64_t);
	19597	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_s64)))
	19598	void svstnt1w_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
	19599	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_u64)))
	19600	void svstnt1w_scatter_u64base_u64(svbool_t, svuint64_t, svuint64_t);
	19601	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_s64)))
	19602	void svstnt1w_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
	19603	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_s64)))
	19604	void svstnt1w_scatter_s64index_s64(svbool_t, int32_t *, svint64_t, svint64_t);
	19605	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_u64)))
	19606	void svstnt1w_scatter_s64index_u64(svbool_t, uint32_t *, svint64_t, svuint64_t);
	19607	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_s64)))
	19608	void svstnt1w_scatter_u64index_s64(svbool_t, int32_t *, svuint64_t, svint64_t);
	19609	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_u64)))
	19610	void svstnt1w_scatter_u64index_u64(svbool_t, uint32_t *, svuint64_t, svuint64_t);
	19611	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_s64)))
	19612	void svstnt1w_scatter_s64offset_s64(svbool_t, int32_t *, svint64_t, svint64_t);
	19613	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_u64)))
	19614	void svstnt1w_scatter_s64offset_u64(svbool_t, uint32_t *, svint64_t, svuint64_t);
	19615	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_s64)))
	19616	void svstnt1w_scatter_u64offset_s64(svbool_t, int32_t *, svuint64_t, svint64_t);
	19617	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_u64)))
	19618	void svstnt1w_scatter_u64offset_u64(svbool_t, uint32_t *, svuint64_t, svuint64_t);
	19619	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u32)))
	19620	svuint16_t svsubhnb_n_u32(svuint32_t, uint32_t);
	19621	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u64)))
	19622	svuint32_t svsubhnb_n_u64(svuint64_t, uint64_t);
	19623	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u16)))
	19624	svuint8_t svsubhnb_n_u16(svuint16_t, uint16_t);
	19625	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s32)))
	19626	svint16_t svsubhnb_n_s32(svint32_t, int32_t);
	19627	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s64)))
	19628	svint32_t svsubhnb_n_s64(svint64_t, int64_t);
	19629	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s16)))
	19630	svint8_t svsubhnb_n_s16(svint16_t, int16_t);
	19631	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u32)))
	19632	svuint16_t svsubhnb_u32(svuint32_t, svuint32_t);
	19633	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u64)))
	19634	svuint32_t svsubhnb_u64(svuint64_t, svuint64_t);
	19635	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u16)))
	19636	svuint8_t svsubhnb_u16(svuint16_t, svuint16_t);
	19637	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s32)))
	19638	svint16_t svsubhnb_s32(svint32_t, svint32_t);
	19639	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s64)))
	19640	svint32_t svsubhnb_s64(svint64_t, svint64_t);
	19641	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s16)))
	19642	svint8_t svsubhnb_s16(svint16_t, svint16_t);
	19643	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u32)))
	19644	svuint16_t svsubhnt_n_u32(svuint16_t, svuint32_t, uint32_t);
	19645	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u64)))
	19646	svuint32_t svsubhnt_n_u64(svuint32_t, svuint64_t, uint64_t);
	19647	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u16)))
	19648	svuint8_t svsubhnt_n_u16(svuint8_t, svuint16_t, uint16_t);
	19649	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s32)))
	19650	svint16_t svsubhnt_n_s32(svint16_t, svint32_t, int32_t);
	19651	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s64)))
	19652	svint32_t svsubhnt_n_s64(svint32_t, svint64_t, int64_t);
	19653	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s16)))
	19654	svint8_t svsubhnt_n_s16(svint8_t, svint16_t, int16_t);
	19655	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u32)))
	19656	svuint16_t svsubhnt_u32(svuint16_t, svuint32_t, svuint32_t);
	19657	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u64)))
	19658	svuint32_t svsubhnt_u64(svuint32_t, svuint64_t, svuint64_t);
	19659	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u16)))
	19660	svuint8_t svsubhnt_u16(svuint8_t, svuint16_t, svuint16_t);
	19661	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s32)))
	19662	svint16_t svsubhnt_s32(svint16_t, svint32_t, svint32_t);
	19663	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s64)))
	19664	svint32_t svsubhnt_s64(svint32_t, svint64_t, svint64_t);
	19665	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s16)))
	19666	svint8_t svsubhnt_s16(svint8_t, svint16_t, svint16_t);
	19667	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s32)))
	19668	svint32_t svsublb_n_s32(svint16_t, int16_t);
	19669	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s64)))
	19670	svint64_t svsublb_n_s64(svint32_t, int32_t);
	19671	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s16)))
	19672	svint16_t svsublb_n_s16(svint8_t, int8_t);
	19673	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u32)))
	19674	svuint32_t svsublb_n_u32(svuint16_t, uint16_t);
	19675	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u64)))
	19676	svuint64_t svsublb_n_u64(svuint32_t, uint32_t);
	19677	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u16)))
	19678	svuint16_t svsublb_n_u16(svuint8_t, uint8_t);
	19679	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s32)))
	19680	svint32_t svsublb_s32(svint16_t, svint16_t);
	19681	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s64)))
	19682	svint64_t svsublb_s64(svint32_t, svint32_t);
	19683	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s16)))
	19684	svint16_t svsublb_s16(svint8_t, svint8_t);
	19685	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u32)))
	19686	svuint32_t svsublb_u32(svuint16_t, svuint16_t);
	19687	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u64)))
	19688	svuint64_t svsublb_u64(svuint32_t, svuint32_t);
	19689	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u16)))
	19690	svuint16_t svsublb_u16(svuint8_t, svuint8_t);
	19691	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s32)))
	19692	svint32_t svsublbt_n_s32(svint16_t, int16_t);
	19693	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s64)))
	19694	svint64_t svsublbt_n_s64(svint32_t, int32_t);
	19695	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s16)))
	19696	svint16_t svsublbt_n_s16(svint8_t, int8_t);
	19697	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s32)))
	19698	svint32_t svsublbt_s32(svint16_t, svint16_t);
	19699	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s64)))
	19700	svint64_t svsublbt_s64(svint32_t, svint32_t);
	19701	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s16)))
	19702	svint16_t svsublbt_s16(svint8_t, svint8_t);
	19703	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s32)))
	19704	svint32_t svsublt_n_s32(svint16_t, int16_t);
	19705	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s64)))
	19706	svint64_t svsublt_n_s64(svint32_t, int32_t);
	19707	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s16)))
	19708	svint16_t svsublt_n_s16(svint8_t, int8_t);
	19709	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u32)))
	19710	svuint32_t svsublt_n_u32(svuint16_t, uint16_t);
	19711	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u64)))
	19712	svuint64_t svsublt_n_u64(svuint32_t, uint32_t);
	19713	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u16)))
	19714	svuint16_t svsublt_n_u16(svuint8_t, uint8_t);
	19715	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s32)))
	19716	svint32_t svsublt_s32(svint16_t, svint16_t);
	19717	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s64)))
	19718	svint64_t svsublt_s64(svint32_t, svint32_t);
	19719	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s16)))
	19720	svint16_t svsublt_s16(svint8_t, svint8_t);
	19721	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u32)))
	19722	svuint32_t svsublt_u32(svuint16_t, svuint16_t);
	19723	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u64)))
	19724	svuint64_t svsublt_u64(svuint32_t, svuint32_t);
	19725	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u16)))
	19726	svuint16_t svsublt_u16(svuint8_t, svuint8_t);
	19727	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s32)))
	19728	svint32_t svsubltb_n_s32(svint16_t, int16_t);
	19729	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s64)))
	19730	svint64_t svsubltb_n_s64(svint32_t, int32_t);
	19731	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s16)))
	19732	svint16_t svsubltb_n_s16(svint8_t, int8_t);
	19733	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s32)))
	19734	svint32_t svsubltb_s32(svint16_t, svint16_t);
	19735	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s64)))
	19736	svint64_t svsubltb_s64(svint32_t, svint32_t);
	19737	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s16)))
	19738	svint16_t svsubltb_s16(svint8_t, svint8_t);
	19739	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s32)))
	19740	svint32_t svsubwb_n_s32(svint32_t, int16_t);
	19741	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s64)))
	19742	svint64_t svsubwb_n_s64(svint64_t, int32_t);
	19743	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s16)))
	19744	svint16_t svsubwb_n_s16(svint16_t, int8_t);
	19745	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u32)))
	19746	svuint32_t svsubwb_n_u32(svuint32_t, uint16_t);
	19747	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u64)))
	19748	svuint64_t svsubwb_n_u64(svuint64_t, uint32_t);
	19749	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u16)))
	19750	svuint16_t svsubwb_n_u16(svuint16_t, uint8_t);
	19751	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s32)))
	19752	svint32_t svsubwb_s32(svint32_t, svint16_t);
	19753	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s64)))
	19754	svint64_t svsubwb_s64(svint64_t, svint32_t);
	19755	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s16)))
	19756	svint16_t svsubwb_s16(svint16_t, svint8_t);
	19757	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u32)))
	19758	svuint32_t svsubwb_u32(svuint32_t, svuint16_t);
	19759	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u64)))
	19760	svuint64_t svsubwb_u64(svuint64_t, svuint32_t);
	19761	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u16)))
	19762	svuint16_t svsubwb_u16(svuint16_t, svuint8_t);
	19763	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s32)))
	19764	svint32_t svsubwt_n_s32(svint32_t, int16_t);
	19765	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s64)))
	19766	svint64_t svsubwt_n_s64(svint64_t, int32_t);
	19767	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s16)))
	19768	svint16_t svsubwt_n_s16(svint16_t, int8_t);
	19769	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u32)))
	19770	svuint32_t svsubwt_n_u32(svuint32_t, uint16_t);
	19771	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u64)))
	19772	svuint64_t svsubwt_n_u64(svuint64_t, uint32_t);
	19773	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u16)))
	19774	svuint16_t svsubwt_n_u16(svuint16_t, uint8_t);
	19775	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s32)))
	19776	svint32_t svsubwt_s32(svint32_t, svint16_t);
	19777	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s64)))
	19778	svint64_t svsubwt_s64(svint64_t, svint32_t);
	19779	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s16)))
	19780	svint16_t svsubwt_s16(svint16_t, svint8_t);
	19781	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u32)))
	19782	svuint32_t svsubwt_u32(svuint32_t, svuint16_t);
	19783	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u64)))
	19784	svuint64_t svsubwt_u64(svuint64_t, svuint32_t);
	19785	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u16)))
	19786	svuint16_t svsubwt_u16(svuint16_t, svuint8_t);
	19787	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u8)))
	19788	svuint8_t svtbl2_u8(svuint8x2_t, svuint8_t);
	19789	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u32)))
	19790	svuint32_t svtbl2_u32(svuint32x2_t, svuint32_t);
	19791	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
	19792	svuint64_t svtbl2_u64(svuint64x2_t, svuint64_t);
	19793	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
	19794	svuint16_t svtbl2_u16(svuint16x2_t, svuint16_t);
	19795	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
	19796	svint8_t svtbl2_s8(svint8x2_t, svuint8_t);
	19797	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
	19798	svfloat64_t svtbl2_f64(svfloat64x2_t, svuint64_t);
	19799	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f32)))
	19800	svfloat32_t svtbl2_f32(svfloat32x2_t, svuint32_t);
	19801	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f16)))
	19802	svfloat16_t svtbl2_f16(svfloat16x2_t, svuint16_t);
	19803	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s32)))
	19804	svint32_t svtbl2_s32(svint32x2_t, svuint32_t);
	19805	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s64)))
	19806	svint64_t svtbl2_s64(svint64x2_t, svuint64_t);
	19807	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s16)))
	19808	svint16_t svtbl2_s16(svint16x2_t, svuint16_t);
	19809	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u8)))
	19810	svuint8_t svtbx_u8(svuint8_t, svuint8_t, svuint8_t);
	19811	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u32)))
	19812	svuint32_t svtbx_u32(svuint32_t, svuint32_t, svuint32_t);
	19813	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
	19814	svuint64_t svtbx_u64(svuint64_t, svuint64_t, svuint64_t);
	19815	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
	19816	svuint16_t svtbx_u16(svuint16_t, svuint16_t, svuint16_t);
	19817	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
	19818	svint8_t svtbx_s8(svint8_t, svint8_t, svuint8_t);
	19819	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
	19820	svfloat64_t svtbx_f64(svfloat64_t, svfloat64_t, svuint64_t);
	19821	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f32)))
	19822	svfloat32_t svtbx_f32(svfloat32_t, svfloat32_t, svuint32_t);
	19823	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f16)))
	19824	svfloat16_t svtbx_f16(svfloat16_t, svfloat16_t, svuint16_t);
	19825	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s32)))
	19826	svint32_t svtbx_s32(svint32_t, svint32_t, svuint32_t);
	19827	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s64)))
	19828	svint64_t svtbx_s64(svint64_t, svint64_t, svuint64_t);
	19829	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s16)))
	19830	svint16_t svtbx_s16(svint16_t, svint16_t, svuint16_t);
	19831	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_m)))
	19832	svint8_t svuqadd_n_s8_m(svbool_t, svint8_t, uint8_t);
	19833	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_m)))
	19834	svint32_t svuqadd_n_s32_m(svbool_t, svint32_t, uint32_t);
	19835	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_m)))
	19836	svint64_t svuqadd_n_s64_m(svbool_t, svint64_t, uint64_t);
	19837	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_m)))
	19838	svint16_t svuqadd_n_s16_m(svbool_t, svint16_t, uint16_t);
	19839	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_x)))
	19840	svint8_t svuqadd_n_s8_x(svbool_t, svint8_t, uint8_t);
	19841	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_x)))
	19842	svint32_t svuqadd_n_s32_x(svbool_t, svint32_t, uint32_t);
	19843	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_x)))
	19844	svint64_t svuqadd_n_s64_x(svbool_t, svint64_t, uint64_t);
	19845	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_x)))
	19846	svint16_t svuqadd_n_s16_x(svbool_t, svint16_t, uint16_t);
	19847	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_z)))
	19848	svint8_t svuqadd_n_s8_z(svbool_t, svint8_t, uint8_t);
	19849	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_z)))
	19850	svint32_t svuqadd_n_s32_z(svbool_t, svint32_t, uint32_t);
	19851	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_z)))
	19852	svint64_t svuqadd_n_s64_z(svbool_t, svint64_t, uint64_t);
	19853	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_z)))
	19854	svint16_t svuqadd_n_s16_z(svbool_t, svint16_t, uint16_t);
	19855	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_m)))
	19856	svint8_t svuqadd_s8_m(svbool_t, svint8_t, svuint8_t);
	19857	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_m)))
	19858	svint32_t svuqadd_s32_m(svbool_t, svint32_t, svuint32_t);
	19859	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_m)))
	19860	svint64_t svuqadd_s64_m(svbool_t, svint64_t, svuint64_t);
	19861	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_m)))
	19862	svint16_t svuqadd_s16_m(svbool_t, svint16_t, svuint16_t);
	19863	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_x)))
	19864	svint8_t svuqadd_s8_x(svbool_t, svint8_t, svuint8_t);
	19865	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_x)))
	19866	svint32_t svuqadd_s32_x(svbool_t, svint32_t, svuint32_t);
	19867	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_x)))
	19868	svint64_t svuqadd_s64_x(svbool_t, svint64_t, svuint64_t);
	19869	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_x)))
	19870	svint16_t svuqadd_s16_x(svbool_t, svint16_t, svuint16_t);
	19871	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_z)))
	19872	svint8_t svuqadd_s8_z(svbool_t, svint8_t, svuint8_t);
	19873	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_z)))
	19874	svint32_t svuqadd_s32_z(svbool_t, svint32_t, svuint32_t);
	19875	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_z)))
	19876	svint64_t svuqadd_s64_z(svbool_t, svint64_t, svuint64_t);
	19877	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_z)))
	19878	svint16_t svuqadd_s16_z(svbool_t, svint16_t, svuint16_t);
	19879	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s32)))
	19880	svbool_t svwhilege_b8_s32(int32_t, int32_t);
	19881	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s32)))
	19882	svbool_t svwhilege_b32_s32(int32_t, int32_t);
	19883	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s32)))
	19884	svbool_t svwhilege_b64_s32(int32_t, int32_t);
	19885	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s32)))
	19886	svbool_t svwhilege_b16_s32(int32_t, int32_t);
	19887	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64)))
	19888	svbool_t svwhilege_b8_s64(int64_t, int64_t);
	19889	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64)))
	19890	svbool_t svwhilege_b32_s64(int64_t, int64_t);
	19891	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64)))
	19892	svbool_t svwhilege_b64_s64(int64_t, int64_t);
	19893	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64)))
	19894	svbool_t svwhilege_b16_s64(int64_t, int64_t);
	19895	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u32)))
	19896	svbool_t svwhilege_b8_u32(uint32_t, uint32_t);
	19897	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u32)))
	19898	svbool_t svwhilege_b32_u32(uint32_t, uint32_t);
	19899	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u32)))
	19900	svbool_t svwhilege_b64_u32(uint32_t, uint32_t);
	19901	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u32)))
	19902	svbool_t svwhilege_b16_u32(uint32_t, uint32_t);
	19903	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64)))
	19904	svbool_t svwhilege_b8_u64(uint64_t, uint64_t);
	19905	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64)))
	19906	svbool_t svwhilege_b32_u64(uint64_t, uint64_t);
	19907	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64)))
	19908	svbool_t svwhilege_b64_u64(uint64_t, uint64_t);
	19909	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64)))
	19910	svbool_t svwhilege_b16_u64(uint64_t, uint64_t);
	19911	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s32)))
	19912	svbool_t svwhilegt_b8_s32(int32_t, int32_t);
	19913	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s32)))
	19914	svbool_t svwhilegt_b32_s32(int32_t, int32_t);
	19915	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s32)))
	19916	svbool_t svwhilegt_b64_s32(int32_t, int32_t);
	19917	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s32)))
	19918	svbool_t svwhilegt_b16_s32(int32_t, int32_t);
	19919	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64)))
	19920	svbool_t svwhilegt_b8_s64(int64_t, int64_t);
	19921	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64)))
	19922	svbool_t svwhilegt_b32_s64(int64_t, int64_t);
	19923	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64)))
	19924	svbool_t svwhilegt_b64_s64(int64_t, int64_t);
	19925	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64)))
	19926	svbool_t svwhilegt_b16_s64(int64_t, int64_t);
	19927	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u32)))
	19928	svbool_t svwhilegt_b8_u32(uint32_t, uint32_t);
	19929	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u32)))
	19930	svbool_t svwhilegt_b32_u32(uint32_t, uint32_t);
	19931	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u32)))
	19932	svbool_t svwhilegt_b64_u32(uint32_t, uint32_t);
	19933	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u32)))
	19934	svbool_t svwhilegt_b16_u32(uint32_t, uint32_t);
	19935	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64)))
	19936	svbool_t svwhilegt_b8_u64(uint64_t, uint64_t);
	19937	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64)))
	19938	svbool_t svwhilegt_b32_u64(uint64_t, uint64_t);
	19939	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64)))
	19940	svbool_t svwhilegt_b64_u64(uint64_t, uint64_t);
	19941	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64)))
	19942	svbool_t svwhilegt_b16_u64(uint64_t, uint64_t);
	19943	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u8)))
	19944	svbool_t svwhilerw_u8(uint8_t const , uint8_t const );
	19945	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s8)))
	19946	svbool_t svwhilerw_s8(int8_t const , int8_t const );
	19947	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u64)))
	19948	svbool_t svwhilerw_u64(uint64_t const , uint64_t const );
	19949	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f64)))
	19950	svbool_t svwhilerw_f64(float64_t const , float64_t const );
	19951	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
	19952	svbool_t svwhilerw_s64(int64_t const , int64_t const );
	19953	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
	19954	svbool_t svwhilerw_u16(uint16_t const , uint16_t const );
	19955	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
	19956	svbool_t svwhilerw_f16(float16_t const , float16_t const );
	19957	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
	19958	svbool_t svwhilerw_s16(int16_t const , int16_t const );
	19959	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u32)))
	19960	svbool_t svwhilerw_u32(uint32_t const , uint32_t const );
	19961	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f32)))
	19962	svbool_t svwhilerw_f32(float32_t const , float32_t const );
	19963	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s32)))
	19964	svbool_t svwhilerw_s32(int32_t const , int32_t const );
	19965	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u8)))
	19966	svbool_t svwhilewr_u8(uint8_t const , uint8_t const );
	19967	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s8)))
	19968	svbool_t svwhilewr_s8(int8_t const , int8_t const );
	19969	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u64)))
	19970	svbool_t svwhilewr_u64(uint64_t const , uint64_t const );
	19971	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f64)))
	19972	svbool_t svwhilewr_f64(float64_t const , float64_t const );
	19973	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
	19974	svbool_t svwhilewr_s64(int64_t const , int64_t const );
	19975	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
	19976	svbool_t svwhilewr_u16(uint16_t const , uint16_t const );
	19977	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
	19978	svbool_t svwhilewr_f16(float16_t const , float16_t const );
	19979	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
	19980	svbool_t svwhilewr_s16(int16_t const , int16_t const );
	19981	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u32)))
	19982	svbool_t svwhilewr_u32(uint32_t const , uint32_t const );
	19983	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f32)))
	19984	svbool_t svwhilewr_f32(float32_t const , float32_t const );
	19985	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s32)))
	19986	svbool_t svwhilewr_s32(int32_t const , int32_t const );
	19987	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u8)))
	19988	svuint8_t svxar_n_u8(svuint8_t, svuint8_t, uint64_t);
	19989	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u32)))
	19990	svuint32_t svxar_n_u32(svuint32_t, svuint32_t, uint64_t);
	19991	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u64)))
	19992	svuint64_t svxar_n_u64(svuint64_t, svuint64_t, uint64_t);
	19993	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u16)))
	19994	svuint16_t svxar_n_u16(svuint16_t, svuint16_t, uint64_t);
	19995	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s8)))
	19996	svint8_t svxar_n_s8(svint8_t, svint8_t, uint64_t);
	19997	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s32)))
	19998	svint32_t svxar_n_s32(svint32_t, svint32_t, uint64_t);
	19999	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s64)))
	20000	svint64_t svxar_n_s64(svint64_t, svint64_t, uint64_t);
	20001	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s16)))
	20002	svint16_t svxar_n_s16(svint16_t, svint16_t, uint64_t);
	20003	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s8)))
	20004	svint8_t svaba(svint8_t, svint8_t, int8_t);
	20005	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s32)))
	20006	svint32_t svaba(svint32_t, svint32_t, int32_t);
	20007	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s64)))
	20008	svint64_t svaba(svint64_t, svint64_t, int64_t);
	20009	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s16)))
	20010	svint16_t svaba(svint16_t, svint16_t, int16_t);
	20011	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u8)))
	20012	svuint8_t svaba(svuint8_t, svuint8_t, uint8_t);
	20013	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u32)))
	20014	svuint32_t svaba(svuint32_t, svuint32_t, uint32_t);
	20015	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u64)))
	20016	svuint64_t svaba(svuint64_t, svuint64_t, uint64_t);
	20017	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_u16)))
	20018	svuint16_t svaba(svuint16_t, svuint16_t, uint16_t);
	20019	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s8)))
	20020	svint8_t svaba(svint8_t, svint8_t, svint8_t);
	20021	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s32)))
	20022	svint32_t svaba(svint32_t, svint32_t, svint32_t);
	20023	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s64)))
	20024	svint64_t svaba(svint64_t, svint64_t, svint64_t);
	20025	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_s16)))
	20026	svint16_t svaba(svint16_t, svint16_t, svint16_t);
	20027	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u8)))
	20028	svuint8_t svaba(svuint8_t, svuint8_t, svuint8_t);
	20029	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u32)))
	20030	svuint32_t svaba(svuint32_t, svuint32_t, svuint32_t);
	20031	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u64)))
	20032	svuint64_t svaba(svuint64_t, svuint64_t, svuint64_t);
	20033	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_u16)))
	20034	svuint16_t svaba(svuint16_t, svuint16_t, svuint16_t);
	20035	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s32)))
	20036	svint32_t svabalb(svint32_t, svint16_t, int16_t);
	20037	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s64)))
	20038	svint64_t svabalb(svint64_t, svint32_t, int32_t);
	20039	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_s16)))
	20040	svint16_t svabalb(svint16_t, svint8_t, int8_t);
	20041	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u32)))
	20042	svuint32_t svabalb(svuint32_t, svuint16_t, uint16_t);
	20043	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u64)))
	20044	svuint64_t svabalb(svuint64_t, svuint32_t, uint32_t);
	20045	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_n_u16)))
	20046	svuint16_t svabalb(svuint16_t, svuint8_t, uint8_t);
	20047	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s32)))
	20048	svint32_t svabalb(svint32_t, svint16_t, svint16_t);
	20049	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s64)))
	20050	svint64_t svabalb(svint64_t, svint32_t, svint32_t);
	20051	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_s16)))
	20052	svint16_t svabalb(svint16_t, svint8_t, svint8_t);
	20053	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u32)))
	20054	svuint32_t svabalb(svuint32_t, svuint16_t, svuint16_t);
	20055	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u64)))
	20056	svuint64_t svabalb(svuint64_t, svuint32_t, svuint32_t);
	20057	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalb_u16)))
	20058	svuint16_t svabalb(svuint16_t, svuint8_t, svuint8_t);
	20059	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s32)))
	20060	svint32_t svabalt(svint32_t, svint16_t, int16_t);
	20061	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s64)))
	20062	svint64_t svabalt(svint64_t, svint32_t, int32_t);
	20063	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_s16)))
	20064	svint16_t svabalt(svint16_t, svint8_t, int8_t);
	20065	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u32)))
	20066	svuint32_t svabalt(svuint32_t, svuint16_t, uint16_t);
	20067	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u64)))
	20068	svuint64_t svabalt(svuint64_t, svuint32_t, uint32_t);
	20069	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_n_u16)))
	20070	svuint16_t svabalt(svuint16_t, svuint8_t, uint8_t);
	20071	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s32)))
	20072	svint32_t svabalt(svint32_t, svint16_t, svint16_t);
	20073	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s64)))
	20074	svint64_t svabalt(svint64_t, svint32_t, svint32_t);
	20075	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_s16)))
	20076	svint16_t svabalt(svint16_t, svint8_t, svint8_t);
	20077	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u32)))
	20078	svuint32_t svabalt(svuint32_t, svuint16_t, svuint16_t);
	20079	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u64)))
	20080	svuint64_t svabalt(svuint64_t, svuint32_t, svuint32_t);
	20081	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabalt_u16)))
	20082	svuint16_t svabalt(svuint16_t, svuint8_t, svuint8_t);
	20083	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s32)))
	20084	svint32_t svabdlb(svint16_t, int16_t);
	20085	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s64)))
	20086	svint64_t svabdlb(svint32_t, int32_t);
	20087	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_s16)))
	20088	svint16_t svabdlb(svint8_t, int8_t);
	20089	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u32)))
	20090	svuint32_t svabdlb(svuint16_t, uint16_t);
	20091	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u64)))
	20092	svuint64_t svabdlb(svuint32_t, uint32_t);
	20093	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_n_u16)))
	20094	svuint16_t svabdlb(svuint8_t, uint8_t);
	20095	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s32)))
	20096	svint32_t svabdlb(svint16_t, svint16_t);
	20097	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s64)))
	20098	svint64_t svabdlb(svint32_t, svint32_t);
	20099	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_s16)))
	20100	svint16_t svabdlb(svint8_t, svint8_t);
	20101	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u32)))
	20102	svuint32_t svabdlb(svuint16_t, svuint16_t);
	20103	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u64)))
	20104	svuint64_t svabdlb(svuint32_t, svuint32_t);
	20105	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlb_u16)))
	20106	svuint16_t svabdlb(svuint8_t, svuint8_t);
	20107	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s32)))
	20108	svint32_t svabdlt(svint16_t, int16_t);
	20109	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s64)))
	20110	svint64_t svabdlt(svint32_t, int32_t);
	20111	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_s16)))
	20112	svint16_t svabdlt(svint8_t, int8_t);
	20113	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u32)))
	20114	svuint32_t svabdlt(svuint16_t, uint16_t);
	20115	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u64)))
	20116	svuint64_t svabdlt(svuint32_t, uint32_t);
	20117	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_n_u16)))
	20118	svuint16_t svabdlt(svuint8_t, uint8_t);
	20119	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s32)))
	20120	svint32_t svabdlt(svint16_t, svint16_t);
	20121	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s64)))
	20122	svint64_t svabdlt(svint32_t, svint32_t);
	20123	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_s16)))
	20124	svint16_t svabdlt(svint8_t, svint8_t);
	20125	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u32)))
	20126	svuint32_t svabdlt(svuint16_t, svuint16_t);
	20127	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u64)))
	20128	svuint64_t svabdlt(svuint32_t, svuint32_t);
	20129	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svabdlt_u16)))
	20130	svuint16_t svabdlt(svuint8_t, svuint8_t);
	20131	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_m)))
	20132	svint32_t svadalp_m(svbool_t, svint32_t, svint16_t);
	20133	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_m)))
	20134	svint64_t svadalp_m(svbool_t, svint64_t, svint32_t);
	20135	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_m)))
	20136	svint16_t svadalp_m(svbool_t, svint16_t, svint8_t);
	20137	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_x)))
	20138	svint32_t svadalp_x(svbool_t, svint32_t, svint16_t);
	20139	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_x)))
	20140	svint64_t svadalp_x(svbool_t, svint64_t, svint32_t);
	20141	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_x)))
	20142	svint16_t svadalp_x(svbool_t, svint16_t, svint8_t);
	20143	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s32_z)))
	20144	svint32_t svadalp_z(svbool_t, svint32_t, svint16_t);
	20145	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s64_z)))
	20146	svint64_t svadalp_z(svbool_t, svint64_t, svint32_t);
	20147	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_s16_z)))
	20148	svint16_t svadalp_z(svbool_t, svint16_t, svint8_t);
	20149	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_m)))
	20150	svuint32_t svadalp_m(svbool_t, svuint32_t, svuint16_t);
	20151	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_m)))
	20152	svuint64_t svadalp_m(svbool_t, svuint64_t, svuint32_t);
	20153	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_m)))
	20154	svuint16_t svadalp_m(svbool_t, svuint16_t, svuint8_t);
	20155	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_x)))
	20156	svuint32_t svadalp_x(svbool_t, svuint32_t, svuint16_t);
	20157	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_x)))
	20158	svuint64_t svadalp_x(svbool_t, svuint64_t, svuint32_t);
	20159	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_x)))
	20160	svuint16_t svadalp_x(svbool_t, svuint16_t, svuint8_t);
	20161	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u32_z)))
	20162	svuint32_t svadalp_z(svbool_t, svuint32_t, svuint16_t);
	20163	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u64_z)))
	20164	svuint64_t svadalp_z(svbool_t, svuint64_t, svuint32_t);
	20165	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadalp_u16_z)))
	20166	svuint16_t svadalp_z(svbool_t, svuint16_t, svuint8_t);
	20167	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u32)))
	20168	svuint32_t svadclb(svuint32_t, svuint32_t, uint32_t);
	20169	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_n_u64)))
	20170	svuint64_t svadclb(svuint64_t, svuint64_t, uint64_t);
	20171	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u32)))
	20172	svuint32_t svadclb(svuint32_t, svuint32_t, svuint32_t);
	20173	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclb_u64)))
	20174	svuint64_t svadclb(svuint64_t, svuint64_t, svuint64_t);
	20175	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u32)))
	20176	svuint32_t svadclt(svuint32_t, svuint32_t, uint32_t);
	20177	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_n_u64)))
	20178	svuint64_t svadclt(svuint64_t, svuint64_t, uint64_t);
	20179	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u32)))
	20180	svuint32_t svadclt(svuint32_t, svuint32_t, svuint32_t);
	20181	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadclt_u64)))
	20182	svuint64_t svadclt(svuint64_t, svuint64_t, svuint64_t);
	20183	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u32)))
	20184	svuint16_t svaddhnb(svuint32_t, uint32_t);
	20185	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u64)))
	20186	svuint32_t svaddhnb(svuint64_t, uint64_t);
	20187	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_u16)))
	20188	svuint8_t svaddhnb(svuint16_t, uint16_t);
	20189	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s32)))
	20190	svint16_t svaddhnb(svint32_t, int32_t);
	20191	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s64)))
	20192	svint32_t svaddhnb(svint64_t, int64_t);
	20193	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_n_s16)))
	20194	svint8_t svaddhnb(svint16_t, int16_t);
	20195	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u32)))
	20196	svuint16_t svaddhnb(svuint32_t, svuint32_t);
	20197	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u64)))
	20198	svuint32_t svaddhnb(svuint64_t, svuint64_t);
	20199	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_u16)))
	20200	svuint8_t svaddhnb(svuint16_t, svuint16_t);
	20201	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s32)))
	20202	svint16_t svaddhnb(svint32_t, svint32_t);
	20203	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s64)))
	20204	svint32_t svaddhnb(svint64_t, svint64_t);
	20205	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnb_s16)))
	20206	svint8_t svaddhnb(svint16_t, svint16_t);
	20207	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u32)))
	20208	svuint16_t svaddhnt(svuint16_t, svuint32_t, uint32_t);
	20209	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u64)))
	20210	svuint32_t svaddhnt(svuint32_t, svuint64_t, uint64_t);
	20211	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_u16)))
	20212	svuint8_t svaddhnt(svuint8_t, svuint16_t, uint16_t);
	20213	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s32)))
	20214	svint16_t svaddhnt(svint16_t, svint32_t, int32_t);
	20215	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s64)))
	20216	svint32_t svaddhnt(svint32_t, svint64_t, int64_t);
	20217	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_n_s16)))
	20218	svint8_t svaddhnt(svint8_t, svint16_t, int16_t);
	20219	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u32)))
	20220	svuint16_t svaddhnt(svuint16_t, svuint32_t, svuint32_t);
	20221	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u64)))
	20222	svuint32_t svaddhnt(svuint32_t, svuint64_t, svuint64_t);
	20223	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_u16)))
	20224	svuint8_t svaddhnt(svuint8_t, svuint16_t, svuint16_t);
	20225	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s32)))
	20226	svint16_t svaddhnt(svint16_t, svint32_t, svint32_t);
	20227	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s64)))
	20228	svint32_t svaddhnt(svint32_t, svint64_t, svint64_t);
	20229	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddhnt_s16)))
	20230	svint8_t svaddhnt(svint8_t, svint16_t, svint16_t);
	20231	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s32)))
	20232	svint32_t svaddlb(svint16_t, int16_t);
	20233	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s64)))
	20234	svint64_t svaddlb(svint32_t, int32_t);
	20235	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_s16)))
	20236	svint16_t svaddlb(svint8_t, int8_t);
	20237	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u32)))
	20238	svuint32_t svaddlb(svuint16_t, uint16_t);
	20239	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u64)))
	20240	svuint64_t svaddlb(svuint32_t, uint32_t);
	20241	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_n_u16)))
	20242	svuint16_t svaddlb(svuint8_t, uint8_t);
	20243	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s32)))
	20244	svint32_t svaddlb(svint16_t, svint16_t);
	20245	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s64)))
	20246	svint64_t svaddlb(svint32_t, svint32_t);
	20247	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_s16)))
	20248	svint16_t svaddlb(svint8_t, svint8_t);
	20249	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u32)))
	20250	svuint32_t svaddlb(svuint16_t, svuint16_t);
	20251	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u64)))
	20252	svuint64_t svaddlb(svuint32_t, svuint32_t);
	20253	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlb_u16)))
	20254	svuint16_t svaddlb(svuint8_t, svuint8_t);
	20255	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s32)))
	20256	svint32_t svaddlbt(svint16_t, int16_t);
	20257	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s64)))
	20258	svint64_t svaddlbt(svint32_t, int32_t);
	20259	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_n_s16)))
	20260	svint16_t svaddlbt(svint8_t, int8_t);
	20261	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s32)))
	20262	svint32_t svaddlbt(svint16_t, svint16_t);
	20263	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s64)))
	20264	svint64_t svaddlbt(svint32_t, svint32_t);
	20265	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlbt_s16)))
	20266	svint16_t svaddlbt(svint8_t, svint8_t);
	20267	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s32)))
	20268	svint32_t svaddlt(svint16_t, int16_t);
	20269	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s64)))
	20270	svint64_t svaddlt(svint32_t, int32_t);
	20271	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_s16)))
	20272	svint16_t svaddlt(svint8_t, int8_t);
	20273	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u32)))
	20274	svuint32_t svaddlt(svuint16_t, uint16_t);
	20275	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u64)))
	20276	svuint64_t svaddlt(svuint32_t, uint32_t);
	20277	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_n_u16)))
	20278	svuint16_t svaddlt(svuint8_t, uint8_t);
	20279	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s32)))
	20280	svint32_t svaddlt(svint16_t, svint16_t);
	20281	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s64)))
	20282	svint64_t svaddlt(svint32_t, svint32_t);
	20283	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_s16)))
	20284	svint16_t svaddlt(svint8_t, svint8_t);
	20285	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u32)))
	20286	svuint32_t svaddlt(svuint16_t, svuint16_t);
	20287	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u64)))
	20288	svuint64_t svaddlt(svuint32_t, svuint32_t);
	20289	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddlt_u16)))
	20290	svuint16_t svaddlt(svuint8_t, svuint8_t);
	20291	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_m)))
	20292	svfloat64_t svaddp_m(svbool_t, svfloat64_t, svfloat64_t);
	20293	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_m)))
	20294	svfloat32_t svaddp_m(svbool_t, svfloat32_t, svfloat32_t);
	20295	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_m)))
	20296	svfloat16_t svaddp_m(svbool_t, svfloat16_t, svfloat16_t);
	20297	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f64_x)))
	20298	svfloat64_t svaddp_x(svbool_t, svfloat64_t, svfloat64_t);
	20299	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f32_x)))
	20300	svfloat32_t svaddp_x(svbool_t, svfloat32_t, svfloat32_t);
	20301	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_f16_x)))
	20302	svfloat16_t svaddp_x(svbool_t, svfloat16_t, svfloat16_t);
	20303	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_m)))
	20304	svuint8_t svaddp_m(svbool_t, svuint8_t, svuint8_t);
	20305	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_m)))
	20306	svuint32_t svaddp_m(svbool_t, svuint32_t, svuint32_t);
	20307	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_m)))
	20308	svuint64_t svaddp_m(svbool_t, svuint64_t, svuint64_t);
	20309	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_m)))
	20310	svuint16_t svaddp_m(svbool_t, svuint16_t, svuint16_t);
	20311	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_m)))
	20312	svint8_t svaddp_m(svbool_t, svint8_t, svint8_t);
	20313	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_m)))
	20314	svint32_t svaddp_m(svbool_t, svint32_t, svint32_t);
	20315	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_m)))
	20316	svint64_t svaddp_m(svbool_t, svint64_t, svint64_t);
	20317	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_m)))
	20318	svint16_t svaddp_m(svbool_t, svint16_t, svint16_t);
	20319	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u8_x)))
	20320	svuint8_t svaddp_x(svbool_t, svuint8_t, svuint8_t);
	20321	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u32_x)))
	20322	svuint32_t svaddp_x(svbool_t, svuint32_t, svuint32_t);
	20323	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u64_x)))
	20324	svuint64_t svaddp_x(svbool_t, svuint64_t, svuint64_t);
	20325	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_u16_x)))
	20326	svuint16_t svaddp_x(svbool_t, svuint16_t, svuint16_t);
	20327	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s8_x)))
	20328	svint8_t svaddp_x(svbool_t, svint8_t, svint8_t);
	20329	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s32_x)))
	20330	svint32_t svaddp_x(svbool_t, svint32_t, svint32_t);
	20331	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s64_x)))
	20332	svint64_t svaddp_x(svbool_t, svint64_t, svint64_t);
	20333	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddp_s16_x)))
	20334	svint16_t svaddp_x(svbool_t, svint16_t, svint16_t);
	20335	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s32)))
	20336	svint32_t svaddwb(svint32_t, int16_t);
	20337	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s64)))
	20338	svint64_t svaddwb(svint64_t, int32_t);
	20339	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_s16)))
	20340	svint16_t svaddwb(svint16_t, int8_t);
	20341	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u32)))
	20342	svuint32_t svaddwb(svuint32_t, uint16_t);
	20343	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u64)))
	20344	svuint64_t svaddwb(svuint64_t, uint32_t);
	20345	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_n_u16)))
	20346	svuint16_t svaddwb(svuint16_t, uint8_t);
	20347	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s32)))
	20348	svint32_t svaddwb(svint32_t, svint16_t);
	20349	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s64)))
	20350	svint64_t svaddwb(svint64_t, svint32_t);
	20351	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_s16)))
	20352	svint16_t svaddwb(svint16_t, svint8_t);
	20353	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u32)))
	20354	svuint32_t svaddwb(svuint32_t, svuint16_t);
	20355	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u64)))
	20356	svuint64_t svaddwb(svuint64_t, svuint32_t);
	20357	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwb_u16)))
	20358	svuint16_t svaddwb(svuint16_t, svuint8_t);
	20359	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s32)))
	20360	svint32_t svaddwt(svint32_t, int16_t);
	20361	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s64)))
	20362	svint64_t svaddwt(svint64_t, int32_t);
	20363	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_s16)))
	20364	svint16_t svaddwt(svint16_t, int8_t);
	20365	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u32)))
	20366	svuint32_t svaddwt(svuint32_t, uint16_t);
	20367	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u64)))
	20368	svuint64_t svaddwt(svuint64_t, uint32_t);
	20369	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_n_u16)))
	20370	svuint16_t svaddwt(svuint16_t, uint8_t);
	20371	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s32)))
	20372	svint32_t svaddwt(svint32_t, svint16_t);
	20373	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s64)))
	20374	svint64_t svaddwt(svint64_t, svint32_t);
	20375	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_s16)))
	20376	svint16_t svaddwt(svint16_t, svint8_t);
	20377	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u32)))
	20378	svuint32_t svaddwt(svuint32_t, svuint16_t);
	20379	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u64)))
	20380	svuint64_t svaddwt(svuint64_t, svuint32_t);
	20381	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddwt_u16)))
	20382	svuint16_t svaddwt(svuint16_t, svuint8_t);
	20383	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u8)))
	20384	svuint8_t svbcax(svuint8_t, svuint8_t, uint8_t);
	20385	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u32)))
	20386	svuint32_t svbcax(svuint32_t, svuint32_t, uint32_t);
	20387	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u64)))
	20388	svuint64_t svbcax(svuint64_t, svuint64_t, uint64_t);
	20389	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_u16)))
	20390	svuint16_t svbcax(svuint16_t, svuint16_t, uint16_t);
	20391	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s8)))
	20392	svint8_t svbcax(svint8_t, svint8_t, int8_t);
	20393	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s32)))
	20394	svint32_t svbcax(svint32_t, svint32_t, int32_t);
	20395	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s64)))
	20396	svint64_t svbcax(svint64_t, svint64_t, int64_t);
	20397	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_n_s16)))
	20398	svint16_t svbcax(svint16_t, svint16_t, int16_t);
	20399	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u8)))
	20400	svuint8_t svbcax(svuint8_t, svuint8_t, svuint8_t);
	20401	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u32)))
	20402	svuint32_t svbcax(svuint32_t, svuint32_t, svuint32_t);
	20403	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u64)))
	20404	svuint64_t svbcax(svuint64_t, svuint64_t, svuint64_t);
	20405	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_u16)))
	20406	svuint16_t svbcax(svuint16_t, svuint16_t, svuint16_t);
	20407	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s8)))
	20408	svint8_t svbcax(svint8_t, svint8_t, svint8_t);
	20409	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s32)))
	20410	svint32_t svbcax(svint32_t, svint32_t, svint32_t);
	20411	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s64)))
	20412	svint64_t svbcax(svint64_t, svint64_t, svint64_t);
	20413	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbcax_s16)))
	20414	svint16_t svbcax(svint16_t, svint16_t, svint16_t);
	20415	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u8)))
	20416	svuint8_t svbsl1n(svuint8_t, svuint8_t, uint8_t);
	20417	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u32)))
	20418	svuint32_t svbsl1n(svuint32_t, svuint32_t, uint32_t);
	20419	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u64)))
	20420	svuint64_t svbsl1n(svuint64_t, svuint64_t, uint64_t);
	20421	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_u16)))
	20422	svuint16_t svbsl1n(svuint16_t, svuint16_t, uint16_t);
	20423	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s8)))
	20424	svint8_t svbsl1n(svint8_t, svint8_t, int8_t);
	20425	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s32)))
	20426	svint32_t svbsl1n(svint32_t, svint32_t, int32_t);
	20427	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s64)))
	20428	svint64_t svbsl1n(svint64_t, svint64_t, int64_t);
	20429	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_n_s16)))
	20430	svint16_t svbsl1n(svint16_t, svint16_t, int16_t);
	20431	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u8)))
	20432	svuint8_t svbsl1n(svuint8_t, svuint8_t, svuint8_t);
	20433	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u32)))
	20434	svuint32_t svbsl1n(svuint32_t, svuint32_t, svuint32_t);
	20435	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u64)))
	20436	svuint64_t svbsl1n(svuint64_t, svuint64_t, svuint64_t);
	20437	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_u16)))
	20438	svuint16_t svbsl1n(svuint16_t, svuint16_t, svuint16_t);
	20439	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s8)))
	20440	svint8_t svbsl1n(svint8_t, svint8_t, svint8_t);
	20441	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s32)))
	20442	svint32_t svbsl1n(svint32_t, svint32_t, svint32_t);
	20443	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s64)))
	20444	svint64_t svbsl1n(svint64_t, svint64_t, svint64_t);
	20445	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl1n_s16)))
	20446	svint16_t svbsl1n(svint16_t, svint16_t, svint16_t);
	20447	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u8)))
	20448	svuint8_t svbsl2n(svuint8_t, svuint8_t, uint8_t);
	20449	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u32)))
	20450	svuint32_t svbsl2n(svuint32_t, svuint32_t, uint32_t);
	20451	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u64)))
	20452	svuint64_t svbsl2n(svuint64_t, svuint64_t, uint64_t);
	20453	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_u16)))
	20454	svuint16_t svbsl2n(svuint16_t, svuint16_t, uint16_t);
	20455	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s8)))
	20456	svint8_t svbsl2n(svint8_t, svint8_t, int8_t);
	20457	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s32)))
	20458	svint32_t svbsl2n(svint32_t, svint32_t, int32_t);
	20459	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s64)))
	20460	svint64_t svbsl2n(svint64_t, svint64_t, int64_t);
	20461	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_n_s16)))
	20462	svint16_t svbsl2n(svint16_t, svint16_t, int16_t);
	20463	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u8)))
	20464	svuint8_t svbsl2n(svuint8_t, svuint8_t, svuint8_t);
	20465	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u32)))
	20466	svuint32_t svbsl2n(svuint32_t, svuint32_t, svuint32_t);
	20467	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u64)))
	20468	svuint64_t svbsl2n(svuint64_t, svuint64_t, svuint64_t);
	20469	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_u16)))
	20470	svuint16_t svbsl2n(svuint16_t, svuint16_t, svuint16_t);
	20471	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s8)))
	20472	svint8_t svbsl2n(svint8_t, svint8_t, svint8_t);
	20473	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s32)))
	20474	svint32_t svbsl2n(svint32_t, svint32_t, svint32_t);
	20475	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s64)))
	20476	svint64_t svbsl2n(svint64_t, svint64_t, svint64_t);
	20477	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl2n_s16)))
	20478	svint16_t svbsl2n(svint16_t, svint16_t, svint16_t);
	20479	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u8)))
	20480	svuint8_t svbsl(svuint8_t, svuint8_t, uint8_t);
	20481	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u32)))
	20482	svuint32_t svbsl(svuint32_t, svuint32_t, uint32_t);
	20483	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u64)))
	20484	svuint64_t svbsl(svuint64_t, svuint64_t, uint64_t);
	20485	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_u16)))
	20486	svuint16_t svbsl(svuint16_t, svuint16_t, uint16_t);
	20487	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s8)))
	20488	svint8_t svbsl(svint8_t, svint8_t, int8_t);
	20489	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s32)))
	20490	svint32_t svbsl(svint32_t, svint32_t, int32_t);
	20491	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s64)))
	20492	svint64_t svbsl(svint64_t, svint64_t, int64_t);
	20493	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_n_s16)))
	20494	svint16_t svbsl(svint16_t, svint16_t, int16_t);
	20495	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u8)))
	20496	svuint8_t svbsl(svuint8_t, svuint8_t, svuint8_t);
	20497	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u32)))
	20498	svuint32_t svbsl(svuint32_t, svuint32_t, svuint32_t);
	20499	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u64)))
	20500	svuint64_t svbsl(svuint64_t, svuint64_t, svuint64_t);
	20501	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_u16)))
	20502	svuint16_t svbsl(svuint16_t, svuint16_t, svuint16_t);
	20503	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s8)))
	20504	svint8_t svbsl(svint8_t, svint8_t, svint8_t);
	20505	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s32)))
	20506	svint32_t svbsl(svint32_t, svint32_t, svint32_t);
	20507	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s64)))
	20508	svint64_t svbsl(svint64_t, svint64_t, svint64_t);
	20509	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbsl_s16)))
	20510	svint16_t svbsl(svint16_t, svint16_t, svint16_t);
	20511	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u8)))
	20512	svuint8_t svcadd(svuint8_t, svuint8_t, uint64_t);
	20513	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u32)))
	20514	svuint32_t svcadd(svuint32_t, svuint32_t, uint64_t);
	20515	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u64)))
	20516	svuint64_t svcadd(svuint64_t, svuint64_t, uint64_t);
	20517	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_u16)))
	20518	svuint16_t svcadd(svuint16_t, svuint16_t, uint64_t);
	20519	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s8)))
	20520	svint8_t svcadd(svint8_t, svint8_t, uint64_t);
	20521	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s32)))
	20522	svint32_t svcadd(svint32_t, svint32_t, uint64_t);
	20523	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s64)))
	20524	svint64_t svcadd(svint64_t, svint64_t, uint64_t);
	20525	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcadd_s16)))
	20526	svint16_t svcadd(svint16_t, svint16_t, uint64_t);
	20527	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s32)))
	20528	svint32_t svcdot(svint32_t, svint8_t, svint8_t, uint64_t);
	20529	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_s64)))
	20530	svint64_t svcdot(svint64_t, svint16_t, svint16_t, uint64_t);
	20531	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s32)))
	20532	svint32_t svcdot_lane(svint32_t, svint8_t, svint8_t, uint64_t, uint64_t);
	20533	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcdot_lane_s64)))
	20534	svint64_t svcdot_lane(svint64_t, svint16_t, svint16_t, uint64_t, uint64_t);
	20535	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u8)))
	20536	svuint8_t svcmla(svuint8_t, svuint8_t, svuint8_t, uint64_t);
	20537	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u32)))
	20538	svuint32_t svcmla(svuint32_t, svuint32_t, svuint32_t, uint64_t);
	20539	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u64)))
	20540	svuint64_t svcmla(svuint64_t, svuint64_t, svuint64_t, uint64_t);
	20541	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_u16)))
	20542	svuint16_t svcmla(svuint16_t, svuint16_t, svuint16_t, uint64_t);
	20543	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s8)))
	20544	svint8_t svcmla(svint8_t, svint8_t, svint8_t, uint64_t);
	20545	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s32)))
	20546	svint32_t svcmla(svint32_t, svint32_t, svint32_t, uint64_t);
	20547	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s64)))
	20548	svint64_t svcmla(svint64_t, svint64_t, svint64_t, uint64_t);
	20549	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_s16)))
	20550	svint16_t svcmla(svint16_t, svint16_t, svint16_t, uint64_t);
	20551	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u32)))
	20552	svuint32_t svcmla_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t, uint64_t);
	20553	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_u16)))
	20554	svuint16_t svcmla_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t, uint64_t);
	20555	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s32)))
	20556	svint32_t svcmla_lane(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
	20557	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcmla_lane_s16)))
	20558	svint16_t svcmla_lane(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
	20559	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_m)))
	20560	svfloat32_t svcvtlt_f32_m(svfloat32_t, svbool_t, svfloat16_t);
	20561	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f32_f16_x)))
	20562	svfloat32_t svcvtlt_f32_x(svbool_t, svfloat16_t);
	20563	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_m)))
	20564	svfloat64_t svcvtlt_f64_m(svfloat64_t, svbool_t, svfloat32_t);
	20565	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtlt_f64_f32_x)))
	20566	svfloat64_t svcvtlt_f64_x(svbool_t, svfloat32_t);
	20567	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f16_f32_m)))
	20568	svfloat16_t svcvtnt_f16_m(svfloat16_t, svbool_t, svfloat32_t);
	20569	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_f32_f64_m)))
	20570	svfloat32_t svcvtnt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
	20571	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_m)))
	20572	svfloat32_t svcvtx_f32_m(svfloat32_t, svbool_t, svfloat64_t);
	20573	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_x)))
	20574	svfloat32_t svcvtx_f32_x(svbool_t, svfloat64_t);
	20575	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtx_f32_f64_z)))
	20576	svfloat32_t svcvtx_f32_z(svbool_t, svfloat64_t);
	20577	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtxnt_f32_f64_m)))
	20578	svfloat32_t svcvtxnt_f32_m(svfloat32_t, svbool_t, svfloat64_t);
	20579	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u8)))
	20580	svuint8_t sveor3(svuint8_t, svuint8_t, uint8_t);
	20581	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u32)))
	20582	svuint32_t sveor3(svuint32_t, svuint32_t, uint32_t);
	20583	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u64)))
	20584	svuint64_t sveor3(svuint64_t, svuint64_t, uint64_t);
	20585	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_u16)))
	20586	svuint16_t sveor3(svuint16_t, svuint16_t, uint16_t);
	20587	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s8)))
	20588	svint8_t sveor3(svint8_t, svint8_t, int8_t);
	20589	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s32)))
	20590	svint32_t sveor3(svint32_t, svint32_t, int32_t);
	20591	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s64)))
	20592	svint64_t sveor3(svint64_t, svint64_t, int64_t);
	20593	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_n_s16)))
	20594	svint16_t sveor3(svint16_t, svint16_t, int16_t);
	20595	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u8)))
	20596	svuint8_t sveor3(svuint8_t, svuint8_t, svuint8_t);
	20597	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u32)))
	20598	svuint32_t sveor3(svuint32_t, svuint32_t, svuint32_t);
	20599	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u64)))
	20600	svuint64_t sveor3(svuint64_t, svuint64_t, svuint64_t);
	20601	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_u16)))
	20602	svuint16_t sveor3(svuint16_t, svuint16_t, svuint16_t);
	20603	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s8)))
	20604	svint8_t sveor3(svint8_t, svint8_t, svint8_t);
	20605	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s32)))
	20606	svint32_t sveor3(svint32_t, svint32_t, svint32_t);
	20607	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s64)))
	20608	svint64_t sveor3(svint64_t, svint64_t, svint64_t);
	20609	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveor3_s16)))
	20610	svint16_t sveor3(svint16_t, svint16_t, svint16_t);
	20611	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u8)))
	20612	svuint8_t sveorbt(svuint8_t, svuint8_t, uint8_t);
	20613	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u32)))
	20614	svuint32_t sveorbt(svuint32_t, svuint32_t, uint32_t);
	20615	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u64)))
	20616	svuint64_t sveorbt(svuint64_t, svuint64_t, uint64_t);
	20617	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_u16)))
	20618	svuint16_t sveorbt(svuint16_t, svuint16_t, uint16_t);
	20619	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s8)))
	20620	svint8_t sveorbt(svint8_t, svint8_t, int8_t);
	20621	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s32)))
	20622	svint32_t sveorbt(svint32_t, svint32_t, int32_t);
	20623	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s64)))
	20624	svint64_t sveorbt(svint64_t, svint64_t, int64_t);
	20625	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_n_s16)))
	20626	svint16_t sveorbt(svint16_t, svint16_t, int16_t);
	20627	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u8)))
	20628	svuint8_t sveorbt(svuint8_t, svuint8_t, svuint8_t);
	20629	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u32)))
	20630	svuint32_t sveorbt(svuint32_t, svuint32_t, svuint32_t);
	20631	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u64)))
	20632	svuint64_t sveorbt(svuint64_t, svuint64_t, svuint64_t);
	20633	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_u16)))
	20634	svuint16_t sveorbt(svuint16_t, svuint16_t, svuint16_t);
	20635	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s8)))
	20636	svint8_t sveorbt(svint8_t, svint8_t, svint8_t);
	20637	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s32)))
	20638	svint32_t sveorbt(svint32_t, svint32_t, svint32_t);
	20639	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s64)))
	20640	svint64_t sveorbt(svint64_t, svint64_t, svint64_t);
	20641	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorbt_s16)))
	20642	svint16_t sveorbt(svint16_t, svint16_t, svint16_t);
	20643	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u8)))
	20644	svuint8_t sveortb(svuint8_t, svuint8_t, uint8_t);
	20645	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u32)))
	20646	svuint32_t sveortb(svuint32_t, svuint32_t, uint32_t);
	20647	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u64)))
	20648	svuint64_t sveortb(svuint64_t, svuint64_t, uint64_t);
	20649	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_u16)))
	20650	svuint16_t sveortb(svuint16_t, svuint16_t, uint16_t);
	20651	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s8)))
	20652	svint8_t sveortb(svint8_t, svint8_t, int8_t);
	20653	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s32)))
	20654	svint32_t sveortb(svint32_t, svint32_t, int32_t);
	20655	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s64)))
	20656	svint64_t sveortb(svint64_t, svint64_t, int64_t);
	20657	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_n_s16)))
	20658	svint16_t sveortb(svint16_t, svint16_t, int16_t);
	20659	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u8)))
	20660	svuint8_t sveortb(svuint8_t, svuint8_t, svuint8_t);
	20661	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u32)))
	20662	svuint32_t sveortb(svuint32_t, svuint32_t, svuint32_t);
	20663	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u64)))
	20664	svuint64_t sveortb(svuint64_t, svuint64_t, svuint64_t);
	20665	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_u16)))
	20666	svuint16_t sveortb(svuint16_t, svuint16_t, svuint16_t);
	20667	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s8)))
	20668	svint8_t sveortb(svint8_t, svint8_t, svint8_t);
	20669	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s32)))
	20670	svint32_t sveortb(svint32_t, svint32_t, svint32_t);
	20671	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s64)))
	20672	svint64_t sveortb(svint64_t, svint64_t, svint64_t);
	20673	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveortb_s16)))
	20674	svint16_t sveortb(svint16_t, svint16_t, svint16_t);
	20675	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_m)))
	20676	svint8_t svhadd_m(svbool_t, svint8_t, int8_t);
	20677	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_m)))
	20678	svint32_t svhadd_m(svbool_t, svint32_t, int32_t);
	20679	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_m)))
	20680	svint64_t svhadd_m(svbool_t, svint64_t, int64_t);
	20681	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_m)))
	20682	svint16_t svhadd_m(svbool_t, svint16_t, int16_t);
	20683	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_x)))
	20684	svint8_t svhadd_x(svbool_t, svint8_t, int8_t);
	20685	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_x)))
	20686	svint32_t svhadd_x(svbool_t, svint32_t, int32_t);
	20687	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_x)))
	20688	svint64_t svhadd_x(svbool_t, svint64_t, int64_t);
	20689	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_x)))
	20690	svint16_t svhadd_x(svbool_t, svint16_t, int16_t);
	20691	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s8_z)))
	20692	svint8_t svhadd_z(svbool_t, svint8_t, int8_t);
	20693	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s32_z)))
	20694	svint32_t svhadd_z(svbool_t, svint32_t, int32_t);
	20695	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s64_z)))
	20696	svint64_t svhadd_z(svbool_t, svint64_t, int64_t);
	20697	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_s16_z)))
	20698	svint16_t svhadd_z(svbool_t, svint16_t, int16_t);
	20699	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_m)))
	20700	svuint8_t svhadd_m(svbool_t, svuint8_t, uint8_t);
	20701	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_m)))
	20702	svuint32_t svhadd_m(svbool_t, svuint32_t, uint32_t);
	20703	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_m)))
	20704	svuint64_t svhadd_m(svbool_t, svuint64_t, uint64_t);
	20705	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_m)))
	20706	svuint16_t svhadd_m(svbool_t, svuint16_t, uint16_t);
	20707	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_x)))
	20708	svuint8_t svhadd_x(svbool_t, svuint8_t, uint8_t);
	20709	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_x)))
	20710	svuint32_t svhadd_x(svbool_t, svuint32_t, uint32_t);
	20711	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_x)))
	20712	svuint64_t svhadd_x(svbool_t, svuint64_t, uint64_t);
	20713	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_x)))
	20714	svuint16_t svhadd_x(svbool_t, svuint16_t, uint16_t);
	20715	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u8_z)))
	20716	svuint8_t svhadd_z(svbool_t, svuint8_t, uint8_t);
	20717	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u32_z)))
	20718	svuint32_t svhadd_z(svbool_t, svuint32_t, uint32_t);
	20719	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u64_z)))
	20720	svuint64_t svhadd_z(svbool_t, svuint64_t, uint64_t);
	20721	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_n_u16_z)))
	20722	svuint16_t svhadd_z(svbool_t, svuint16_t, uint16_t);
	20723	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_m)))
	20724	svint8_t svhadd_m(svbool_t, svint8_t, svint8_t);
	20725	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_m)))
	20726	svint32_t svhadd_m(svbool_t, svint32_t, svint32_t);
	20727	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_m)))
	20728	svint64_t svhadd_m(svbool_t, svint64_t, svint64_t);
	20729	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_m)))
	20730	svint16_t svhadd_m(svbool_t, svint16_t, svint16_t);
	20731	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_x)))
	20732	svint8_t svhadd_x(svbool_t, svint8_t, svint8_t);
	20733	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_x)))
	20734	svint32_t svhadd_x(svbool_t, svint32_t, svint32_t);
	20735	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_x)))
	20736	svint64_t svhadd_x(svbool_t, svint64_t, svint64_t);
	20737	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_x)))
	20738	svint16_t svhadd_x(svbool_t, svint16_t, svint16_t);
	20739	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s8_z)))
	20740	svint8_t svhadd_z(svbool_t, svint8_t, svint8_t);
	20741	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s32_z)))
	20742	svint32_t svhadd_z(svbool_t, svint32_t, svint32_t);
	20743	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s64_z)))
	20744	svint64_t svhadd_z(svbool_t, svint64_t, svint64_t);
	20745	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_s16_z)))
	20746	svint16_t svhadd_z(svbool_t, svint16_t, svint16_t);
	20747	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_m)))
	20748	svuint8_t svhadd_m(svbool_t, svuint8_t, svuint8_t);
	20749	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_m)))
	20750	svuint32_t svhadd_m(svbool_t, svuint32_t, svuint32_t);
	20751	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_m)))
	20752	svuint64_t svhadd_m(svbool_t, svuint64_t, svuint64_t);
	20753	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_m)))
	20754	svuint16_t svhadd_m(svbool_t, svuint16_t, svuint16_t);
	20755	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_x)))
	20756	svuint8_t svhadd_x(svbool_t, svuint8_t, svuint8_t);
	20757	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_x)))
	20758	svuint32_t svhadd_x(svbool_t, svuint32_t, svuint32_t);
	20759	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_x)))
	20760	svuint64_t svhadd_x(svbool_t, svuint64_t, svuint64_t);
	20761	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_x)))
	20762	svuint16_t svhadd_x(svbool_t, svuint16_t, svuint16_t);
	20763	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u8_z)))
	20764	svuint8_t svhadd_z(svbool_t, svuint8_t, svuint8_t);
	20765	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u32_z)))
	20766	svuint32_t svhadd_z(svbool_t, svuint32_t, svuint32_t);
	20767	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u64_z)))
	20768	svuint64_t svhadd_z(svbool_t, svuint64_t, svuint64_t);
	20769	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhadd_u16_z)))
	20770	svuint16_t svhadd_z(svbool_t, svuint16_t, svuint16_t);
	20771	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u32_z)))
	20772	svuint32_t svhistcnt_z(svbool_t, svuint32_t, svuint32_t);
	20773	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u64_z)))
	20774	svuint64_t svhistcnt_z(svbool_t, svuint64_t, svuint64_t);
	20775	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s32_z)))
	20776	svuint32_t svhistcnt_z(svbool_t, svint32_t, svint32_t);
	20777	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_s64_z)))
	20778	svuint64_t svhistcnt_z(svbool_t, svint64_t, svint64_t);
	20779	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_u8)))
	20780	svuint8_t svhistseg(svuint8_t, svuint8_t);
	20781	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistseg_s8)))
	20782	svuint8_t svhistseg(svint8_t, svint8_t);
	20783	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_m)))
	20784	svint8_t svhsub_m(svbool_t, svint8_t, int8_t);
	20785	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_m)))
	20786	svint32_t svhsub_m(svbool_t, svint32_t, int32_t);
	20787	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_m)))
	20788	svint64_t svhsub_m(svbool_t, svint64_t, int64_t);
	20789	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_m)))
	20790	svint16_t svhsub_m(svbool_t, svint16_t, int16_t);
	20791	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_x)))
	20792	svint8_t svhsub_x(svbool_t, svint8_t, int8_t);
	20793	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_x)))
	20794	svint32_t svhsub_x(svbool_t, svint32_t, int32_t);
	20795	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_x)))
	20796	svint64_t svhsub_x(svbool_t, svint64_t, int64_t);
	20797	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_x)))
	20798	svint16_t svhsub_x(svbool_t, svint16_t, int16_t);
	20799	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s8_z)))
	20800	svint8_t svhsub_z(svbool_t, svint8_t, int8_t);
	20801	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s32_z)))
	20802	svint32_t svhsub_z(svbool_t, svint32_t, int32_t);
	20803	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s64_z)))
	20804	svint64_t svhsub_z(svbool_t, svint64_t, int64_t);
	20805	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_s16_z)))
	20806	svint16_t svhsub_z(svbool_t, svint16_t, int16_t);
	20807	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_m)))
	20808	svuint8_t svhsub_m(svbool_t, svuint8_t, uint8_t);
	20809	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_m)))
	20810	svuint32_t svhsub_m(svbool_t, svuint32_t, uint32_t);
	20811	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_m)))
	20812	svuint64_t svhsub_m(svbool_t, svuint64_t, uint64_t);
	20813	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_m)))
	20814	svuint16_t svhsub_m(svbool_t, svuint16_t, uint16_t);
	20815	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_x)))
	20816	svuint8_t svhsub_x(svbool_t, svuint8_t, uint8_t);
	20817	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_x)))
	20818	svuint32_t svhsub_x(svbool_t, svuint32_t, uint32_t);
	20819	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_x)))
	20820	svuint64_t svhsub_x(svbool_t, svuint64_t, uint64_t);
	20821	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_x)))
	20822	svuint16_t svhsub_x(svbool_t, svuint16_t, uint16_t);
	20823	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u8_z)))
	20824	svuint8_t svhsub_z(svbool_t, svuint8_t, uint8_t);
	20825	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u32_z)))
	20826	svuint32_t svhsub_z(svbool_t, svuint32_t, uint32_t);
	20827	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u64_z)))
	20828	svuint64_t svhsub_z(svbool_t, svuint64_t, uint64_t);
	20829	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_n_u16_z)))
	20830	svuint16_t svhsub_z(svbool_t, svuint16_t, uint16_t);
	20831	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_m)))
	20832	svint8_t svhsub_m(svbool_t, svint8_t, svint8_t);
	20833	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_m)))
	20834	svint32_t svhsub_m(svbool_t, svint32_t, svint32_t);
	20835	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_m)))
	20836	svint64_t svhsub_m(svbool_t, svint64_t, svint64_t);
	20837	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_m)))
	20838	svint16_t svhsub_m(svbool_t, svint16_t, svint16_t);
	20839	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_x)))
	20840	svint8_t svhsub_x(svbool_t, svint8_t, svint8_t);
	20841	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_x)))
	20842	svint32_t svhsub_x(svbool_t, svint32_t, svint32_t);
	20843	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_x)))
	20844	svint64_t svhsub_x(svbool_t, svint64_t, svint64_t);
	20845	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_x)))
	20846	svint16_t svhsub_x(svbool_t, svint16_t, svint16_t);
	20847	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s8_z)))
	20848	svint8_t svhsub_z(svbool_t, svint8_t, svint8_t);
	20849	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s32_z)))
	20850	svint32_t svhsub_z(svbool_t, svint32_t, svint32_t);
	20851	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s64_z)))
	20852	svint64_t svhsub_z(svbool_t, svint64_t, svint64_t);
	20853	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_s16_z)))
	20854	svint16_t svhsub_z(svbool_t, svint16_t, svint16_t);
	20855	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_m)))
	20856	svuint8_t svhsub_m(svbool_t, svuint8_t, svuint8_t);
	20857	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_m)))
	20858	svuint32_t svhsub_m(svbool_t, svuint32_t, svuint32_t);
	20859	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_m)))
	20860	svuint64_t svhsub_m(svbool_t, svuint64_t, svuint64_t);
	20861	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_m)))
	20862	svuint16_t svhsub_m(svbool_t, svuint16_t, svuint16_t);
	20863	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_x)))
	20864	svuint8_t svhsub_x(svbool_t, svuint8_t, svuint8_t);
	20865	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_x)))
	20866	svuint32_t svhsub_x(svbool_t, svuint32_t, svuint32_t);
	20867	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_x)))
	20868	svuint64_t svhsub_x(svbool_t, svuint64_t, svuint64_t);
	20869	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_x)))
	20870	svuint16_t svhsub_x(svbool_t, svuint16_t, svuint16_t);
	20871	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u8_z)))
	20872	svuint8_t svhsub_z(svbool_t, svuint8_t, svuint8_t);
	20873	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u32_z)))
	20874	svuint32_t svhsub_z(svbool_t, svuint32_t, svuint32_t);
	20875	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u64_z)))
	20876	svuint64_t svhsub_z(svbool_t, svuint64_t, svuint64_t);
	20877	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsub_u16_z)))
	20878	svuint16_t svhsub_z(svbool_t, svuint16_t, svuint16_t);
	20879	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_m)))
	20880	svint8_t svhsubr_m(svbool_t, svint8_t, int8_t);
	20881	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_m)))
	20882	svint32_t svhsubr_m(svbool_t, svint32_t, int32_t);
	20883	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_m)))
	20884	svint64_t svhsubr_m(svbool_t, svint64_t, int64_t);
	20885	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_m)))
	20886	svint16_t svhsubr_m(svbool_t, svint16_t, int16_t);
	20887	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_x)))
	20888	svint8_t svhsubr_x(svbool_t, svint8_t, int8_t);
	20889	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_x)))
	20890	svint32_t svhsubr_x(svbool_t, svint32_t, int32_t);
	20891	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_x)))
	20892	svint64_t svhsubr_x(svbool_t, svint64_t, int64_t);
	20893	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_x)))
	20894	svint16_t svhsubr_x(svbool_t, svint16_t, int16_t);
	20895	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s8_z)))
	20896	svint8_t svhsubr_z(svbool_t, svint8_t, int8_t);
	20897	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s32_z)))
	20898	svint32_t svhsubr_z(svbool_t, svint32_t, int32_t);
	20899	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s64_z)))
	20900	svint64_t svhsubr_z(svbool_t, svint64_t, int64_t);
	20901	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_s16_z)))
	20902	svint16_t svhsubr_z(svbool_t, svint16_t, int16_t);
	20903	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_m)))
	20904	svuint8_t svhsubr_m(svbool_t, svuint8_t, uint8_t);
	20905	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_m)))
	20906	svuint32_t svhsubr_m(svbool_t, svuint32_t, uint32_t);
	20907	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_m)))
	20908	svuint64_t svhsubr_m(svbool_t, svuint64_t, uint64_t);
	20909	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_m)))
	20910	svuint16_t svhsubr_m(svbool_t, svuint16_t, uint16_t);
	20911	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_x)))
	20912	svuint8_t svhsubr_x(svbool_t, svuint8_t, uint8_t);
	20913	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_x)))
	20914	svuint32_t svhsubr_x(svbool_t, svuint32_t, uint32_t);
	20915	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_x)))
	20916	svuint64_t svhsubr_x(svbool_t, svuint64_t, uint64_t);
	20917	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_x)))
	20918	svuint16_t svhsubr_x(svbool_t, svuint16_t, uint16_t);
	20919	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u8_z)))
	20920	svuint8_t svhsubr_z(svbool_t, svuint8_t, uint8_t);
	20921	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u32_z)))
	20922	svuint32_t svhsubr_z(svbool_t, svuint32_t, uint32_t);
	20923	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u64_z)))
	20924	svuint64_t svhsubr_z(svbool_t, svuint64_t, uint64_t);
	20925	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_n_u16_z)))
	20926	svuint16_t svhsubr_z(svbool_t, svuint16_t, uint16_t);
	20927	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_m)))
	20928	svint8_t svhsubr_m(svbool_t, svint8_t, svint8_t);
	20929	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_m)))
	20930	svint32_t svhsubr_m(svbool_t, svint32_t, svint32_t);
	20931	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_m)))
	20932	svint64_t svhsubr_m(svbool_t, svint64_t, svint64_t);
	20933	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_m)))
	20934	svint16_t svhsubr_m(svbool_t, svint16_t, svint16_t);
	20935	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_x)))
	20936	svint8_t svhsubr_x(svbool_t, svint8_t, svint8_t);
	20937	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_x)))
	20938	svint32_t svhsubr_x(svbool_t, svint32_t, svint32_t);
	20939	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_x)))
	20940	svint64_t svhsubr_x(svbool_t, svint64_t, svint64_t);
	20941	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_x)))
	20942	svint16_t svhsubr_x(svbool_t, svint16_t, svint16_t);
	20943	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s8_z)))
	20944	svint8_t svhsubr_z(svbool_t, svint8_t, svint8_t);
	20945	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s32_z)))
	20946	svint32_t svhsubr_z(svbool_t, svint32_t, svint32_t);
	20947	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s64_z)))
	20948	svint64_t svhsubr_z(svbool_t, svint64_t, svint64_t);
	20949	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_s16_z)))
	20950	svint16_t svhsubr_z(svbool_t, svint16_t, svint16_t);
	20951	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_m)))
	20952	svuint8_t svhsubr_m(svbool_t, svuint8_t, svuint8_t);
	20953	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_m)))
	20954	svuint32_t svhsubr_m(svbool_t, svuint32_t, svuint32_t);
	20955	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_m)))
	20956	svuint64_t svhsubr_m(svbool_t, svuint64_t, svuint64_t);
	20957	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_m)))
	20958	svuint16_t svhsubr_m(svbool_t, svuint16_t, svuint16_t);
	20959	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_x)))
	20960	svuint8_t svhsubr_x(svbool_t, svuint8_t, svuint8_t);
	20961	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_x)))
	20962	svuint32_t svhsubr_x(svbool_t, svuint32_t, svuint32_t);
	20963	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_x)))
	20964	svuint64_t svhsubr_x(svbool_t, svuint64_t, svuint64_t);
	20965	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_x)))
	20966	svuint16_t svhsubr_x(svbool_t, svuint16_t, svuint16_t);
	20967	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u8_z)))
	20968	svuint8_t svhsubr_z(svbool_t, svuint8_t, svuint8_t);
	20969	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u32_z)))
	20970	svuint32_t svhsubr_z(svbool_t, svuint32_t, svuint32_t);
	20971	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u64_z)))
	20972	svuint64_t svhsubr_z(svbool_t, svuint64_t, svuint64_t);
	20973	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhsubr_u16_z)))
	20974	svuint16_t svhsubr_z(svbool_t, svuint16_t, svuint16_t);
	20975	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_u32)))
	20976	svuint32_t svldnt1_gather_index_u32(svbool_t, svuint32_t, int64_t);
	20977	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_u64)))
	20978	svuint64_t svldnt1_gather_index_u64(svbool_t, svuint64_t, int64_t);
	20979	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_f64)))
	20980	svfloat64_t svldnt1_gather_index_f64(svbool_t, svuint64_t, int64_t);
	20981	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_f32)))
	20982	svfloat32_t svldnt1_gather_index_f32(svbool_t, svuint32_t, int64_t);
	20983	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_index_s32)))
	20984	svint32_t svldnt1_gather_index_s32(svbool_t, svuint32_t, int64_t);
	20985	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_index_s64)))
	20986	svint64_t svldnt1_gather_index_s64(svbool_t, svuint64_t, int64_t);
	20987	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_u32)))
	20988	svuint32_t svldnt1_gather_offset_u32(svbool_t, svuint32_t, int64_t);
	20989	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_u64)))
	20990	svuint64_t svldnt1_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	20991	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_f64)))
	20992	svfloat64_t svldnt1_gather_offset_f64(svbool_t, svuint64_t, int64_t);
	20993	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_f32)))
	20994	svfloat32_t svldnt1_gather_offset_f32(svbool_t, svuint32_t, int64_t);
	20995	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_offset_s32)))
	20996	svint32_t svldnt1_gather_offset_s32(svbool_t, svuint32_t, int64_t);
	20997	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_offset_s64)))
	20998	svint64_t svldnt1_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	20999	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_u32)))
	21000	svuint32_t svldnt1_gather_u32(svbool_t, svuint32_t);
	21001	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_u64)))
	21002	svuint64_t svldnt1_gather_u64(svbool_t, svuint64_t);
	21003	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_f64)))
	21004	svfloat64_t svldnt1_gather_f64(svbool_t, svuint64_t);
	21005	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_f32)))
	21006	svfloat32_t svldnt1_gather_f32(svbool_t, svuint32_t);
	21007	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32base_s32)))
	21008	svint32_t svldnt1_gather_s32(svbool_t, svuint32_t);
	21009	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64base_s64)))
	21010	svint64_t svldnt1_gather_s64(svbool_t, svuint64_t);
	21011	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_u64)))
	21012	svuint64_t svldnt1_gather_index(svbool_t, uint64_t const *, svint64_t);
	21013	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_f64)))
	21014	svfloat64_t svldnt1_gather_index(svbool_t, float64_t const *, svint64_t);
	21015	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64index_s64)))
	21016	svint64_t svldnt1_gather_index(svbool_t, int64_t const *, svint64_t);
	21017	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_u64)))
	21018	svuint64_t svldnt1_gather_index(svbool_t, uint64_t const *, svuint64_t);
	21019	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_f64)))
	21020	svfloat64_t svldnt1_gather_index(svbool_t, float64_t const *, svuint64_t);
	21021	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64index_s64)))
	21022	svint64_t svldnt1_gather_index(svbool_t, int64_t const *, svuint64_t);
	21023	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_u32)))
	21024	svuint32_t svldnt1_gather_offset(svbool_t, uint32_t const *, svuint32_t);
	21025	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_f32)))
	21026	svfloat32_t svldnt1_gather_offset(svbool_t, float32_t const *, svuint32_t);
	21027	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u32offset_s32)))
	21028	svint32_t svldnt1_gather_offset(svbool_t, int32_t const *, svuint32_t);
	21029	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_u64)))
	21030	svuint64_t svldnt1_gather_offset(svbool_t, uint64_t const *, svint64_t);
	21031	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_f64)))
	21032	svfloat64_t svldnt1_gather_offset(svbool_t, float64_t const *, svint64_t);
	21033	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_s64offset_s64)))
	21034	svint64_t svldnt1_gather_offset(svbool_t, int64_t const *, svint64_t);
	21035	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_u64)))
	21036	svuint64_t svldnt1_gather_offset(svbool_t, uint64_t const *, svuint64_t);
	21037	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_f64)))
	21038	svfloat64_t svldnt1_gather_offset(svbool_t, float64_t const *, svuint64_t);
	21039	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_gather_u64offset_s64)))
	21040	svint64_t svldnt1_gather_offset(svbool_t, int64_t const *, svuint64_t);
	21041	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_u32)))
	21042	svuint32_t svldnt1sb_gather_offset_u32(svbool_t, svuint32_t, int64_t);
	21043	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_u64)))
	21044	svuint64_t svldnt1sb_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	21045	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_offset_s32)))
	21046	svint32_t svldnt1sb_gather_offset_s32(svbool_t, svuint32_t, int64_t);
	21047	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_offset_s64)))
	21048	svint64_t svldnt1sb_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	21049	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_u32)))
	21050	svuint32_t svldnt1sb_gather_u32(svbool_t, svuint32_t);
	21051	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_u64)))
	21052	svuint64_t svldnt1sb_gather_u64(svbool_t, svuint64_t);
	21053	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32base_s32)))
	21054	svint32_t svldnt1sb_gather_s32(svbool_t, svuint32_t);
	21055	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64base_s64)))
	21056	svint64_t svldnt1sb_gather_s64(svbool_t, svuint64_t);
	21057	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_u32)))
	21058	svuint32_t svldnt1sb_gather_offset_u32(svbool_t, int8_t const *, svuint32_t);
	21059	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u32offset_s32)))
	21060	svint32_t svldnt1sb_gather_offset_s32(svbool_t, int8_t const *, svuint32_t);
	21061	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_u64)))
	21062	svuint64_t svldnt1sb_gather_offset_u64(svbool_t, int8_t const *, svint64_t);
	21063	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_s64offset_s64)))
	21064	svint64_t svldnt1sb_gather_offset_s64(svbool_t, int8_t const *, svint64_t);
	21065	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_u64)))
	21066	svuint64_t svldnt1sb_gather_offset_u64(svbool_t, int8_t const *, svuint64_t);
	21067	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sb_gather_u64offset_s64)))
	21068	svint64_t svldnt1sb_gather_offset_s64(svbool_t, int8_t const *, svuint64_t);
	21069	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_u32)))
	21070	svuint32_t svldnt1sh_gather_index_u32(svbool_t, svuint32_t, int64_t);
	21071	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_u64)))
	21072	svuint64_t svldnt1sh_gather_index_u64(svbool_t, svuint64_t, int64_t);
	21073	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_index_s32)))
	21074	svint32_t svldnt1sh_gather_index_s32(svbool_t, svuint32_t, int64_t);
	21075	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_index_s64)))
	21076	svint64_t svldnt1sh_gather_index_s64(svbool_t, svuint64_t, int64_t);
	21077	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_u32)))
	21078	svuint32_t svldnt1sh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
	21079	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_u64)))
	21080	svuint64_t svldnt1sh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	21081	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_offset_s32)))
	21082	svint32_t svldnt1sh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
	21083	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_offset_s64)))
	21084	svint64_t svldnt1sh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	21085	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_u32)))
	21086	svuint32_t svldnt1sh_gather_u32(svbool_t, svuint32_t);
	21087	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_u64)))
	21088	svuint64_t svldnt1sh_gather_u64(svbool_t, svuint64_t);
	21089	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32base_s32)))
	21090	svint32_t svldnt1sh_gather_s32(svbool_t, svuint32_t);
	21091	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64base_s64)))
	21092	svint64_t svldnt1sh_gather_s64(svbool_t, svuint64_t);
	21093	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_u64)))
	21094	svuint64_t svldnt1sh_gather_index_u64(svbool_t, int16_t const *, svint64_t);
	21095	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64index_s64)))
	21096	svint64_t svldnt1sh_gather_index_s64(svbool_t, int16_t const *, svint64_t);
	21097	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_u64)))
	21098	svuint64_t svldnt1sh_gather_index_u64(svbool_t, int16_t const *, svuint64_t);
	21099	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64index_s64)))
	21100	svint64_t svldnt1sh_gather_index_s64(svbool_t, int16_t const *, svuint64_t);
	21101	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_u32)))
	21102	svuint32_t svldnt1sh_gather_offset_u32(svbool_t, int16_t const *, svuint32_t);
	21103	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u32offset_s32)))
	21104	svint32_t svldnt1sh_gather_offset_s32(svbool_t, int16_t const *, svuint32_t);
	21105	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_u64)))
	21106	svuint64_t svldnt1sh_gather_offset_u64(svbool_t, int16_t const *, svint64_t);
	21107	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_s64offset_s64)))
	21108	svint64_t svldnt1sh_gather_offset_s64(svbool_t, int16_t const *, svint64_t);
	21109	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_u64)))
	21110	svuint64_t svldnt1sh_gather_offset_u64(svbool_t, int16_t const *, svuint64_t);
	21111	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sh_gather_u64offset_s64)))
	21112	svint64_t svldnt1sh_gather_offset_s64(svbool_t, int16_t const *, svuint64_t);
	21113	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_u64)))
	21114	svuint64_t svldnt1sw_gather_index_u64(svbool_t, svuint64_t, int64_t);
	21115	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_index_s64)))
	21116	svint64_t svldnt1sw_gather_index_s64(svbool_t, svuint64_t, int64_t);
	21117	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_u64)))
	21118	svuint64_t svldnt1sw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	21119	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_offset_s64)))
	21120	svint64_t svldnt1sw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	21121	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_u64)))
	21122	svuint64_t svldnt1sw_gather_u64(svbool_t, svuint64_t);
	21123	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64base_s64)))
	21124	svint64_t svldnt1sw_gather_s64(svbool_t, svuint64_t);
	21125	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_u64)))
	21126	svuint64_t svldnt1sw_gather_index_u64(svbool_t, int32_t const *, svint64_t);
	21127	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64index_s64)))
	21128	svint64_t svldnt1sw_gather_index_s64(svbool_t, int32_t const *, svint64_t);
	21129	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_u64)))
	21130	svuint64_t svldnt1sw_gather_index_u64(svbool_t, int32_t const *, svuint64_t);
	21131	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64index_s64)))
	21132	svint64_t svldnt1sw_gather_index_s64(svbool_t, int32_t const *, svuint64_t);
	21133	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_u64)))
	21134	svuint64_t svldnt1sw_gather_offset_u64(svbool_t, int32_t const *, svint64_t);
	21135	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_s64offset_s64)))
	21136	svint64_t svldnt1sw_gather_offset_s64(svbool_t, int32_t const *, svint64_t);
	21137	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_u64)))
	21138	svuint64_t svldnt1sw_gather_offset_u64(svbool_t, int32_t const *, svuint64_t);
	21139	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1sw_gather_u64offset_s64)))
	21140	svint64_t svldnt1sw_gather_offset_s64(svbool_t, int32_t const *, svuint64_t);
	21141	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_u32)))
	21142	svuint32_t svldnt1ub_gather_offset_u32(svbool_t, svuint32_t, int64_t);
	21143	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_u64)))
	21144	svuint64_t svldnt1ub_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	21145	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_offset_s32)))
	21146	svint32_t svldnt1ub_gather_offset_s32(svbool_t, svuint32_t, int64_t);
	21147	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_offset_s64)))
	21148	svint64_t svldnt1ub_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	21149	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_u32)))
	21150	svuint32_t svldnt1ub_gather_u32(svbool_t, svuint32_t);
	21151	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_u64)))
	21152	svuint64_t svldnt1ub_gather_u64(svbool_t, svuint64_t);
	21153	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32base_s32)))
	21154	svint32_t svldnt1ub_gather_s32(svbool_t, svuint32_t);
	21155	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64base_s64)))
	21156	svint64_t svldnt1ub_gather_s64(svbool_t, svuint64_t);
	21157	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_u32)))
	21158	svuint32_t svldnt1ub_gather_offset_u32(svbool_t, uint8_t const *, svuint32_t);
	21159	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u32offset_s32)))
	21160	svint32_t svldnt1ub_gather_offset_s32(svbool_t, uint8_t const *, svuint32_t);
	21161	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_u64)))
	21162	svuint64_t svldnt1ub_gather_offset_u64(svbool_t, uint8_t const *, svint64_t);
	21163	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_s64offset_s64)))
	21164	svint64_t svldnt1ub_gather_offset_s64(svbool_t, uint8_t const *, svint64_t);
	21165	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_u64)))
	21166	svuint64_t svldnt1ub_gather_offset_u64(svbool_t, uint8_t const *, svuint64_t);
	21167	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1ub_gather_u64offset_s64)))
	21168	svint64_t svldnt1ub_gather_offset_s64(svbool_t, uint8_t const *, svuint64_t);
	21169	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_u32)))
	21170	svuint32_t svldnt1uh_gather_index_u32(svbool_t, svuint32_t, int64_t);
	21171	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_u64)))
	21172	svuint64_t svldnt1uh_gather_index_u64(svbool_t, svuint64_t, int64_t);
	21173	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_index_s32)))
	21174	svint32_t svldnt1uh_gather_index_s32(svbool_t, svuint32_t, int64_t);
	21175	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_index_s64)))
	21176	svint64_t svldnt1uh_gather_index_s64(svbool_t, svuint64_t, int64_t);
	21177	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_u32)))
	21178	svuint32_t svldnt1uh_gather_offset_u32(svbool_t, svuint32_t, int64_t);
	21179	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_u64)))
	21180	svuint64_t svldnt1uh_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	21181	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_offset_s32)))
	21182	svint32_t svldnt1uh_gather_offset_s32(svbool_t, svuint32_t, int64_t);
	21183	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_offset_s64)))
	21184	svint64_t svldnt1uh_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	21185	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_u32)))
	21186	svuint32_t svldnt1uh_gather_u32(svbool_t, svuint32_t);
	21187	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_u64)))
	21188	svuint64_t svldnt1uh_gather_u64(svbool_t, svuint64_t);
	21189	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32base_s32)))
	21190	svint32_t svldnt1uh_gather_s32(svbool_t, svuint32_t);
	21191	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64base_s64)))
	21192	svint64_t svldnt1uh_gather_s64(svbool_t, svuint64_t);
	21193	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_u64)))
	21194	svuint64_t svldnt1uh_gather_index_u64(svbool_t, uint16_t const *, svint64_t);
	21195	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64index_s64)))
	21196	svint64_t svldnt1uh_gather_index_s64(svbool_t, uint16_t const *, svint64_t);
	21197	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_u64)))
	21198	svuint64_t svldnt1uh_gather_index_u64(svbool_t, uint16_t const *, svuint64_t);
	21199	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64index_s64)))
	21200	svint64_t svldnt1uh_gather_index_s64(svbool_t, uint16_t const *, svuint64_t);
	21201	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_u32)))
	21202	svuint32_t svldnt1uh_gather_offset_u32(svbool_t, uint16_t const *, svuint32_t);
	21203	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u32offset_s32)))
	21204	svint32_t svldnt1uh_gather_offset_s32(svbool_t, uint16_t const *, svuint32_t);
	21205	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_u64)))
	21206	svuint64_t svldnt1uh_gather_offset_u64(svbool_t, uint16_t const *, svint64_t);
	21207	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_s64offset_s64)))
	21208	svint64_t svldnt1uh_gather_offset_s64(svbool_t, uint16_t const *, svint64_t);
	21209	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_u64)))
	21210	svuint64_t svldnt1uh_gather_offset_u64(svbool_t, uint16_t const *, svuint64_t);
	21211	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uh_gather_u64offset_s64)))
	21212	svint64_t svldnt1uh_gather_offset_s64(svbool_t, uint16_t const *, svuint64_t);
	21213	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_u64)))
	21214	svuint64_t svldnt1uw_gather_index_u64(svbool_t, svuint64_t, int64_t);
	21215	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_index_s64)))
	21216	svint64_t svldnt1uw_gather_index_s64(svbool_t, svuint64_t, int64_t);
	21217	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_u64)))
	21218	svuint64_t svldnt1uw_gather_offset_u64(svbool_t, svuint64_t, int64_t);
	21219	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_offset_s64)))
	21220	svint64_t svldnt1uw_gather_offset_s64(svbool_t, svuint64_t, int64_t);
	21221	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_u64)))
	21222	svuint64_t svldnt1uw_gather_u64(svbool_t, svuint64_t);
	21223	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64base_s64)))
	21224	svint64_t svldnt1uw_gather_s64(svbool_t, svuint64_t);
	21225	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_u64)))
	21226	svuint64_t svldnt1uw_gather_index_u64(svbool_t, uint32_t const *, svint64_t);
	21227	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64index_s64)))
	21228	svint64_t svldnt1uw_gather_index_s64(svbool_t, uint32_t const *, svint64_t);
	21229	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_u64)))
	21230	svuint64_t svldnt1uw_gather_index_u64(svbool_t, uint32_t const *, svuint64_t);
	21231	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64index_s64)))
	21232	svint64_t svldnt1uw_gather_index_s64(svbool_t, uint32_t const *, svuint64_t);
	21233	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_u64)))
	21234	svuint64_t svldnt1uw_gather_offset_u64(svbool_t, uint32_t const *, svint64_t);
	21235	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_s64offset_s64)))
	21236	svint64_t svldnt1uw_gather_offset_s64(svbool_t, uint32_t const *, svint64_t);
	21237	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_u64)))
	21238	svuint64_t svldnt1uw_gather_offset_u64(svbool_t, uint32_t const *, svuint64_t);
	21239	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1uw_gather_u64offset_s64)))
	21240	svint64_t svldnt1uw_gather_offset_s64(svbool_t, uint32_t const *, svuint64_t);
	21241	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_m)))
	21242	svint64_t svlogb_m(svint64_t, svbool_t, svfloat64_t);
	21243	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_m)))
	21244	svint32_t svlogb_m(svint32_t, svbool_t, svfloat32_t);
	21245	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_m)))
	21246	svint16_t svlogb_m(svint16_t, svbool_t, svfloat16_t);
	21247	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_x)))
	21248	svint64_t svlogb_x(svbool_t, svfloat64_t);
	21249	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_x)))
	21250	svint32_t svlogb_x(svbool_t, svfloat32_t);
	21251	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_x)))
	21252	svint16_t svlogb_x(svbool_t, svfloat16_t);
	21253	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f64_z)))
	21254	svint64_t svlogb_z(svbool_t, svfloat64_t);
	21255	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f32_z)))
	21256	svint32_t svlogb_z(svbool_t, svfloat32_t);
	21257	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlogb_f16_z)))
	21258	svint16_t svlogb_z(svbool_t, svfloat16_t);
	21259	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u8)))
	21260	svbool_t svmatch(svbool_t, svuint8_t, svuint8_t);
	21261	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_u16)))
	21262	svbool_t svmatch(svbool_t, svuint16_t, svuint16_t);
	21263	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s8)))
	21264	svbool_t svmatch(svbool_t, svint8_t, svint8_t);
	21265	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmatch_s16)))
	21266	svbool_t svmatch(svbool_t, svint16_t, svint16_t);
	21267	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_m)))
	21268	svfloat64_t svmaxnmp_m(svbool_t, svfloat64_t, svfloat64_t);
	21269	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_m)))
	21270	svfloat32_t svmaxnmp_m(svbool_t, svfloat32_t, svfloat32_t);
	21271	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_m)))
	21272	svfloat16_t svmaxnmp_m(svbool_t, svfloat16_t, svfloat16_t);
	21273	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f64_x)))
	21274	svfloat64_t svmaxnmp_x(svbool_t, svfloat64_t, svfloat64_t);
	21275	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f32_x)))
	21276	svfloat32_t svmaxnmp_x(svbool_t, svfloat32_t, svfloat32_t);
	21277	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmp_f16_x)))
	21278	svfloat16_t svmaxnmp_x(svbool_t, svfloat16_t, svfloat16_t);
	21279	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_m)))
	21280	svfloat64_t svmaxp_m(svbool_t, svfloat64_t, svfloat64_t);
	21281	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_m)))
	21282	svfloat32_t svmaxp_m(svbool_t, svfloat32_t, svfloat32_t);
	21283	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_m)))
	21284	svfloat16_t svmaxp_m(svbool_t, svfloat16_t, svfloat16_t);
	21285	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f64_x)))
	21286	svfloat64_t svmaxp_x(svbool_t, svfloat64_t, svfloat64_t);
	21287	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f32_x)))
	21288	svfloat32_t svmaxp_x(svbool_t, svfloat32_t, svfloat32_t);
	21289	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_f16_x)))
	21290	svfloat16_t svmaxp_x(svbool_t, svfloat16_t, svfloat16_t);
	21291	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_m)))
	21292	svint8_t svmaxp_m(svbool_t, svint8_t, svint8_t);
	21293	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_m)))
	21294	svint32_t svmaxp_m(svbool_t, svint32_t, svint32_t);
	21295	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_m)))
	21296	svint64_t svmaxp_m(svbool_t, svint64_t, svint64_t);
	21297	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_m)))
	21298	svint16_t svmaxp_m(svbool_t, svint16_t, svint16_t);
	21299	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s8_x)))
	21300	svint8_t svmaxp_x(svbool_t, svint8_t, svint8_t);
	21301	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s32_x)))
	21302	svint32_t svmaxp_x(svbool_t, svint32_t, svint32_t);
	21303	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s64_x)))
	21304	svint64_t svmaxp_x(svbool_t, svint64_t, svint64_t);
	21305	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_s16_x)))
	21306	svint16_t svmaxp_x(svbool_t, svint16_t, svint16_t);
	21307	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_m)))
	21308	svuint8_t svmaxp_m(svbool_t, svuint8_t, svuint8_t);
	21309	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_m)))
	21310	svuint32_t svmaxp_m(svbool_t, svuint32_t, svuint32_t);
	21311	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_m)))
	21312	svuint64_t svmaxp_m(svbool_t, svuint64_t, svuint64_t);
	21313	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_m)))
	21314	svuint16_t svmaxp_m(svbool_t, svuint16_t, svuint16_t);
	21315	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u8_x)))
	21316	svuint8_t svmaxp_x(svbool_t, svuint8_t, svuint8_t);
	21317	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u32_x)))
	21318	svuint32_t svmaxp_x(svbool_t, svuint32_t, svuint32_t);
	21319	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u64_x)))
	21320	svuint64_t svmaxp_x(svbool_t, svuint64_t, svuint64_t);
	21321	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxp_u16_x)))
	21322	svuint16_t svmaxp_x(svbool_t, svuint16_t, svuint16_t);
	21323	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_m)))
	21324	svfloat64_t svminnmp_m(svbool_t, svfloat64_t, svfloat64_t);
	21325	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_m)))
	21326	svfloat32_t svminnmp_m(svbool_t, svfloat32_t, svfloat32_t);
	21327	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_m)))
	21328	svfloat16_t svminnmp_m(svbool_t, svfloat16_t, svfloat16_t);
	21329	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f64_x)))
	21330	svfloat64_t svminnmp_x(svbool_t, svfloat64_t, svfloat64_t);
	21331	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f32_x)))
	21332	svfloat32_t svminnmp_x(svbool_t, svfloat32_t, svfloat32_t);
	21333	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmp_f16_x)))
	21334	svfloat16_t svminnmp_x(svbool_t, svfloat16_t, svfloat16_t);
	21335	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_m)))
	21336	svfloat64_t svminp_m(svbool_t, svfloat64_t, svfloat64_t);
	21337	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_m)))
	21338	svfloat32_t svminp_m(svbool_t, svfloat32_t, svfloat32_t);
	21339	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_m)))
	21340	svfloat16_t svminp_m(svbool_t, svfloat16_t, svfloat16_t);
	21341	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f64_x)))
	21342	svfloat64_t svminp_x(svbool_t, svfloat64_t, svfloat64_t);
	21343	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f32_x)))
	21344	svfloat32_t svminp_x(svbool_t, svfloat32_t, svfloat32_t);
	21345	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_f16_x)))
	21346	svfloat16_t svminp_x(svbool_t, svfloat16_t, svfloat16_t);
	21347	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_m)))
	21348	svint8_t svminp_m(svbool_t, svint8_t, svint8_t);
	21349	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_m)))
	21350	svint32_t svminp_m(svbool_t, svint32_t, svint32_t);
	21351	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_m)))
	21352	svint64_t svminp_m(svbool_t, svint64_t, svint64_t);
	21353	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_m)))
	21354	svint16_t svminp_m(svbool_t, svint16_t, svint16_t);
	21355	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s8_x)))
	21356	svint8_t svminp_x(svbool_t, svint8_t, svint8_t);
	21357	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s32_x)))
	21358	svint32_t svminp_x(svbool_t, svint32_t, svint32_t);
	21359	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s64_x)))
	21360	svint64_t svminp_x(svbool_t, svint64_t, svint64_t);
	21361	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_s16_x)))
	21362	svint16_t svminp_x(svbool_t, svint16_t, svint16_t);
	21363	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_m)))
	21364	svuint8_t svminp_m(svbool_t, svuint8_t, svuint8_t);
	21365	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_m)))
	21366	svuint32_t svminp_m(svbool_t, svuint32_t, svuint32_t);
	21367	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_m)))
	21368	svuint64_t svminp_m(svbool_t, svuint64_t, svuint64_t);
	21369	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_m)))
	21370	svuint16_t svminp_m(svbool_t, svuint16_t, svuint16_t);
	21371	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u8_x)))
	21372	svuint8_t svminp_x(svbool_t, svuint8_t, svuint8_t);
	21373	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u32_x)))
	21374	svuint32_t svminp_x(svbool_t, svuint32_t, svuint32_t);
	21375	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u64_x)))
	21376	svuint64_t svminp_x(svbool_t, svuint64_t, svuint64_t);
	21377	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminp_u16_x)))
	21378	svuint16_t svminp_x(svbool_t, svuint16_t, svuint16_t);
	21379	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u32)))
	21380	svuint32_t svmla_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t);
	21381	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u64)))
	21382	svuint64_t svmla_lane(svuint64_t, svuint64_t, svuint64_t, uint64_t);
	21383	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_u16)))
	21384	svuint16_t svmla_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t);
	21385	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s32)))
	21386	svint32_t svmla_lane(svint32_t, svint32_t, svint32_t, uint64_t);
	21387	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s64)))
	21388	svint64_t svmla_lane(svint64_t, svint64_t, svint64_t, uint64_t);
	21389	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmla_lane_s16)))
	21390	svint16_t svmla_lane(svint16_t, svint16_t, svint16_t, uint64_t);
	21391	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_f32)))
	21392	svfloat32_t svmlalb(svfloat32_t, svfloat16_t, float16_t);
	21393	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s32)))
	21394	svint32_t svmlalb(svint32_t, svint16_t, int16_t);
	21395	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s64)))
	21396	svint64_t svmlalb(svint64_t, svint32_t, int32_t);
	21397	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_s16)))
	21398	svint16_t svmlalb(svint16_t, svint8_t, int8_t);
	21399	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u32)))
	21400	svuint32_t svmlalb(svuint32_t, svuint16_t, uint16_t);
	21401	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u64)))
	21402	svuint64_t svmlalb(svuint64_t, svuint32_t, uint32_t);
	21403	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_n_u16)))
	21404	svuint16_t svmlalb(svuint16_t, svuint8_t, uint8_t);
	21405	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_f32)))
	21406	svfloat32_t svmlalb(svfloat32_t, svfloat16_t, svfloat16_t);
	21407	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s32)))
	21408	svint32_t svmlalb(svint32_t, svint16_t, svint16_t);
	21409	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s64)))
	21410	svint64_t svmlalb(svint64_t, svint32_t, svint32_t);
	21411	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_s16)))
	21412	svint16_t svmlalb(svint16_t, svint8_t, svint8_t);
	21413	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u32)))
	21414	svuint32_t svmlalb(svuint32_t, svuint16_t, svuint16_t);
	21415	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u64)))
	21416	svuint64_t svmlalb(svuint64_t, svuint32_t, svuint32_t);
	21417	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_u16)))
	21418	svuint16_t svmlalb(svuint16_t, svuint8_t, svuint8_t);
	21419	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_f32)))
	21420	svfloat32_t svmlalb_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	21421	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s32)))
	21422	svint32_t svmlalb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21423	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_s64)))
	21424	svint64_t svmlalb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21425	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u32)))
	21426	svuint32_t svmlalb_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	21427	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalb_lane_u64)))
	21428	svuint64_t svmlalb_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	21429	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_f32)))
	21430	svfloat32_t svmlalt(svfloat32_t, svfloat16_t, float16_t);
	21431	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s32)))
	21432	svint32_t svmlalt(svint32_t, svint16_t, int16_t);
	21433	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s64)))
	21434	svint64_t svmlalt(svint64_t, svint32_t, int32_t);
	21435	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_s16)))
	21436	svint16_t svmlalt(svint16_t, svint8_t, int8_t);
	21437	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u32)))
	21438	svuint32_t svmlalt(svuint32_t, svuint16_t, uint16_t);
	21439	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u64)))
	21440	svuint64_t svmlalt(svuint64_t, svuint32_t, uint32_t);
	21441	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_n_u16)))
	21442	svuint16_t svmlalt(svuint16_t, svuint8_t, uint8_t);
	21443	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_f32)))
	21444	svfloat32_t svmlalt(svfloat32_t, svfloat16_t, svfloat16_t);
	21445	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s32)))
	21446	svint32_t svmlalt(svint32_t, svint16_t, svint16_t);
	21447	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s64)))
	21448	svint64_t svmlalt(svint64_t, svint32_t, svint32_t);
	21449	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_s16)))
	21450	svint16_t svmlalt(svint16_t, svint8_t, svint8_t);
	21451	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u32)))
	21452	svuint32_t svmlalt(svuint32_t, svuint16_t, svuint16_t);
	21453	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u64)))
	21454	svuint64_t svmlalt(svuint64_t, svuint32_t, svuint32_t);
	21455	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_u16)))
	21456	svuint16_t svmlalt(svuint16_t, svuint8_t, svuint8_t);
	21457	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_f32)))
	21458	svfloat32_t svmlalt_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	21459	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s32)))
	21460	svint32_t svmlalt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21461	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_s64)))
	21462	svint64_t svmlalt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21463	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u32)))
	21464	svuint32_t svmlalt_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	21465	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlalt_lane_u64)))
	21466	svuint64_t svmlalt_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	21467	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u32)))
	21468	svuint32_t svmls_lane(svuint32_t, svuint32_t, svuint32_t, uint64_t);
	21469	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u64)))
	21470	svuint64_t svmls_lane(svuint64_t, svuint64_t, svuint64_t, uint64_t);
	21471	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_u16)))
	21472	svuint16_t svmls_lane(svuint16_t, svuint16_t, svuint16_t, uint64_t);
	21473	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s32)))
	21474	svint32_t svmls_lane(svint32_t, svint32_t, svint32_t, uint64_t);
	21475	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s64)))
	21476	svint64_t svmls_lane(svint64_t, svint64_t, svint64_t, uint64_t);
	21477	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmls_lane_s16)))
	21478	svint16_t svmls_lane(svint16_t, svint16_t, svint16_t, uint64_t);
	21479	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_f32)))
	21480	svfloat32_t svmlslb(svfloat32_t, svfloat16_t, float16_t);
	21481	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s32)))
	21482	svint32_t svmlslb(svint32_t, svint16_t, int16_t);
	21483	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s64)))
	21484	svint64_t svmlslb(svint64_t, svint32_t, int32_t);
	21485	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_s16)))
	21486	svint16_t svmlslb(svint16_t, svint8_t, int8_t);
	21487	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u32)))
	21488	svuint32_t svmlslb(svuint32_t, svuint16_t, uint16_t);
	21489	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u64)))
	21490	svuint64_t svmlslb(svuint64_t, svuint32_t, uint32_t);
	21491	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_n_u16)))
	21492	svuint16_t svmlslb(svuint16_t, svuint8_t, uint8_t);
	21493	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_f32)))
	21494	svfloat32_t svmlslb(svfloat32_t, svfloat16_t, svfloat16_t);
	21495	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s32)))
	21496	svint32_t svmlslb(svint32_t, svint16_t, svint16_t);
	21497	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s64)))
	21498	svint64_t svmlslb(svint64_t, svint32_t, svint32_t);
	21499	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_s16)))
	21500	svint16_t svmlslb(svint16_t, svint8_t, svint8_t);
	21501	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u32)))
	21502	svuint32_t svmlslb(svuint32_t, svuint16_t, svuint16_t);
	21503	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u64)))
	21504	svuint64_t svmlslb(svuint64_t, svuint32_t, svuint32_t);
	21505	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_u16)))
	21506	svuint16_t svmlslb(svuint16_t, svuint8_t, svuint8_t);
	21507	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_f32)))
	21508	svfloat32_t svmlslb_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	21509	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s32)))
	21510	svint32_t svmlslb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21511	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_s64)))
	21512	svint64_t svmlslb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21513	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u32)))
	21514	svuint32_t svmlslb_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	21515	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslb_lane_u64)))
	21516	svuint64_t svmlslb_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	21517	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_f32)))
	21518	svfloat32_t svmlslt(svfloat32_t, svfloat16_t, float16_t);
	21519	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s32)))
	21520	svint32_t svmlslt(svint32_t, svint16_t, int16_t);
	21521	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s64)))
	21522	svint64_t svmlslt(svint64_t, svint32_t, int32_t);
	21523	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_s16)))
	21524	svint16_t svmlslt(svint16_t, svint8_t, int8_t);
	21525	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u32)))
	21526	svuint32_t svmlslt(svuint32_t, svuint16_t, uint16_t);
	21527	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u64)))
	21528	svuint64_t svmlslt(svuint64_t, svuint32_t, uint32_t);
	21529	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_n_u16)))
	21530	svuint16_t svmlslt(svuint16_t, svuint8_t, uint8_t);
	21531	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_f32)))
	21532	svfloat32_t svmlslt(svfloat32_t, svfloat16_t, svfloat16_t);
	21533	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s32)))
	21534	svint32_t svmlslt(svint32_t, svint16_t, svint16_t);
	21535	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s64)))
	21536	svint64_t svmlslt(svint64_t, svint32_t, svint32_t);
	21537	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_s16)))
	21538	svint16_t svmlslt(svint16_t, svint8_t, svint8_t);
	21539	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u32)))
	21540	svuint32_t svmlslt(svuint32_t, svuint16_t, svuint16_t);
	21541	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u64)))
	21542	svuint64_t svmlslt(svuint64_t, svuint32_t, svuint32_t);
	21543	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_u16)))
	21544	svuint16_t svmlslt(svuint16_t, svuint8_t, svuint8_t);
	21545	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_f32)))
	21546	svfloat32_t svmlslt_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
	21547	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s32)))
	21548	svint32_t svmlslt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21549	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_s64)))
	21550	svint64_t svmlslt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21551	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u32)))
	21552	svuint32_t svmlslt_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
	21553	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmlslt_lane_u64)))
	21554	svuint64_t svmlslt_lane(svuint64_t, svuint32_t, svuint32_t, uint64_t);
	21555	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s32)))
	21556	svint32_t svmovlb(svint16_t);
	21557	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s64)))
	21558	svint64_t svmovlb(svint32_t);
	21559	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_s16)))
	21560	svint16_t svmovlb(svint8_t);
	21561	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u32)))
	21562	svuint32_t svmovlb(svuint16_t);
	21563	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u64)))
	21564	svuint64_t svmovlb(svuint32_t);
	21565	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlb_u16)))
	21566	svuint16_t svmovlb(svuint8_t);
	21567	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s32)))
	21568	svint32_t svmovlt(svint16_t);
	21569	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s64)))
	21570	svint64_t svmovlt(svint32_t);
	21571	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_s16)))
	21572	svint16_t svmovlt(svint8_t);
	21573	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u32)))
	21574	svuint32_t svmovlt(svuint16_t);
	21575	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u64)))
	21576	svuint64_t svmovlt(svuint32_t);
	21577	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmovlt_u16)))
	21578	svuint16_t svmovlt(svuint8_t);
	21579	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u32)))
	21580	svuint32_t svmul_lane(svuint32_t, svuint32_t, uint64_t);
	21581	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u64)))
	21582	svuint64_t svmul_lane(svuint64_t, svuint64_t, uint64_t);
	21583	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_u16)))
	21584	svuint16_t svmul_lane(svuint16_t, svuint16_t, uint64_t);
	21585	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s32)))
	21586	svint32_t svmul_lane(svint32_t, svint32_t, uint64_t);
	21587	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s64)))
	21588	svint64_t svmul_lane(svint64_t, svint64_t, uint64_t);
	21589	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmul_lane_s16)))
	21590	svint16_t svmul_lane(svint16_t, svint16_t, uint64_t);
	21591	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s32)))
	21592	svint32_t svmullb(svint16_t, int16_t);
	21593	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s64)))
	21594	svint64_t svmullb(svint32_t, int32_t);
	21595	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_s16)))
	21596	svint16_t svmullb(svint8_t, int8_t);
	21597	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u32)))
	21598	svuint32_t svmullb(svuint16_t, uint16_t);
	21599	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u64)))
	21600	svuint64_t svmullb(svuint32_t, uint32_t);
	21601	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_n_u16)))
	21602	svuint16_t svmullb(svuint8_t, uint8_t);
	21603	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s32)))
	21604	svint32_t svmullb(svint16_t, svint16_t);
	21605	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s64)))
	21606	svint64_t svmullb(svint32_t, svint32_t);
	21607	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_s16)))
	21608	svint16_t svmullb(svint8_t, svint8_t);
	21609	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u32)))
	21610	svuint32_t svmullb(svuint16_t, svuint16_t);
	21611	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u64)))
	21612	svuint64_t svmullb(svuint32_t, svuint32_t);
	21613	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_u16)))
	21614	svuint16_t svmullb(svuint8_t, svuint8_t);
	21615	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s32)))
	21616	svint32_t svmullb_lane(svint16_t, svint16_t, uint64_t);
	21617	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_s64)))
	21618	svint64_t svmullb_lane(svint32_t, svint32_t, uint64_t);
	21619	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u32)))
	21620	svuint32_t svmullb_lane(svuint16_t, svuint16_t, uint64_t);
	21621	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullb_lane_u64)))
	21622	svuint64_t svmullb_lane(svuint32_t, svuint32_t, uint64_t);
	21623	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s32)))
	21624	svint32_t svmullt(svint16_t, int16_t);
	21625	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s64)))
	21626	svint64_t svmullt(svint32_t, int32_t);
	21627	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_s16)))
	21628	svint16_t svmullt(svint8_t, int8_t);
	21629	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u32)))
	21630	svuint32_t svmullt(svuint16_t, uint16_t);
	21631	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u64)))
	21632	svuint64_t svmullt(svuint32_t, uint32_t);
	21633	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_n_u16)))
	21634	svuint16_t svmullt(svuint8_t, uint8_t);
	21635	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s32)))
	21636	svint32_t svmullt(svint16_t, svint16_t);
	21637	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s64)))
	21638	svint64_t svmullt(svint32_t, svint32_t);
	21639	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_s16)))
	21640	svint16_t svmullt(svint8_t, svint8_t);
	21641	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u32)))
	21642	svuint32_t svmullt(svuint16_t, svuint16_t);
	21643	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u64)))
	21644	svuint64_t svmullt(svuint32_t, svuint32_t);
	21645	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_u16)))
	21646	svuint16_t svmullt(svuint8_t, svuint8_t);
	21647	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s32)))
	21648	svint32_t svmullt_lane(svint16_t, svint16_t, uint64_t);
	21649	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_s64)))
	21650	svint64_t svmullt_lane(svint32_t, svint32_t, uint64_t);
	21651	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u32)))
	21652	svuint32_t svmullt_lane(svuint16_t, svuint16_t, uint64_t);
	21653	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmullt_lane_u64)))
	21654	svuint64_t svmullt_lane(svuint32_t, svuint32_t, uint64_t);
	21655	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u8)))
	21656	svuint8_t svnbsl(svuint8_t, svuint8_t, uint8_t);
	21657	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u32)))
	21658	svuint32_t svnbsl(svuint32_t, svuint32_t, uint32_t);
	21659	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u64)))
	21660	svuint64_t svnbsl(svuint64_t, svuint64_t, uint64_t);
	21661	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_u16)))
	21662	svuint16_t svnbsl(svuint16_t, svuint16_t, uint16_t);
	21663	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s8)))
	21664	svint8_t svnbsl(svint8_t, svint8_t, int8_t);
	21665	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s32)))
	21666	svint32_t svnbsl(svint32_t, svint32_t, int32_t);
	21667	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s64)))
	21668	svint64_t svnbsl(svint64_t, svint64_t, int64_t);
	21669	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_n_s16)))
	21670	svint16_t svnbsl(svint16_t, svint16_t, int16_t);
	21671	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u8)))
	21672	svuint8_t svnbsl(svuint8_t, svuint8_t, svuint8_t);
	21673	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u32)))
	21674	svuint32_t svnbsl(svuint32_t, svuint32_t, svuint32_t);
	21675	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u64)))
	21676	svuint64_t svnbsl(svuint64_t, svuint64_t, svuint64_t);
	21677	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_u16)))
	21678	svuint16_t svnbsl(svuint16_t, svuint16_t, svuint16_t);
	21679	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s8)))
	21680	svint8_t svnbsl(svint8_t, svint8_t, svint8_t);
	21681	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s32)))
	21682	svint32_t svnbsl(svint32_t, svint32_t, svint32_t);
	21683	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s64)))
	21684	svint64_t svnbsl(svint64_t, svint64_t, svint64_t);
	21685	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnbsl_s16)))
	21686	svint16_t svnbsl(svint16_t, svint16_t, svint16_t);
	21687	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u8)))
	21688	svbool_t svnmatch(svbool_t, svuint8_t, svuint8_t);
	21689	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_u16)))
	21690	svbool_t svnmatch(svbool_t, svuint16_t, svuint16_t);
	21691	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s8)))
	21692	svbool_t svnmatch(svbool_t, svint8_t, svint8_t);
	21693	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svnmatch_s16)))
	21694	svbool_t svnmatch(svbool_t, svint16_t, svint16_t);
	21695	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_n_u8)))
	21696	svuint8_t svpmul(svuint8_t, uint8_t);
	21697	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmul_u8)))
	21698	svuint8_t svpmul(svuint8_t, svuint8_t);
	21699	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u64)))
	21700	svuint64_t svpmullb(svuint32_t, uint32_t);
	21701	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_n_u16)))
	21702	svuint16_t svpmullb(svuint8_t, uint8_t);
	21703	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u64)))
	21704	svuint64_t svpmullb(svuint32_t, svuint32_t);
	21705	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_u16)))
	21706	svuint16_t svpmullb(svuint8_t, svuint8_t);
	21707	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u8)))
	21708	svuint8_t svpmullb_pair(svuint8_t, uint8_t);
	21709	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u32)))
	21710	svuint32_t svpmullb_pair(svuint32_t, uint32_t);
	21711	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u8)))
	21712	svuint8_t svpmullb_pair(svuint8_t, svuint8_t);
	21713	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u32)))
	21714	svuint32_t svpmullb_pair(svuint32_t, svuint32_t);
	21715	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u64)))
	21716	svuint64_t svpmullt(svuint32_t, uint32_t);
	21717	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_n_u16)))
	21718	svuint16_t svpmullt(svuint8_t, uint8_t);
	21719	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u64)))
	21720	svuint64_t svpmullt(svuint32_t, svuint32_t);
	21721	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_u16)))
	21722	svuint16_t svpmullt(svuint8_t, svuint8_t);
	21723	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u8)))
	21724	svuint8_t svpmullt_pair(svuint8_t, uint8_t);
	21725	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u32)))
	21726	svuint32_t svpmullt_pair(svuint32_t, uint32_t);
	21727	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u8)))
	21728	svuint8_t svpmullt_pair(svuint8_t, svuint8_t);
	21729	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u32)))
	21730	svuint32_t svpmullt_pair(svuint32_t, svuint32_t);
	21731	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_m)))
	21732	svint8_t svqabs_m(svint8_t, svbool_t, svint8_t);
	21733	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_m)))
	21734	svint32_t svqabs_m(svint32_t, svbool_t, svint32_t);
	21735	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_m)))
	21736	svint64_t svqabs_m(svint64_t, svbool_t, svint64_t);
	21737	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_m)))
	21738	svint16_t svqabs_m(svint16_t, svbool_t, svint16_t);
	21739	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_x)))
	21740	svint8_t svqabs_x(svbool_t, svint8_t);
	21741	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_x)))
	21742	svint32_t svqabs_x(svbool_t, svint32_t);
	21743	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_x)))
	21744	svint64_t svqabs_x(svbool_t, svint64_t);
	21745	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_x)))
	21746	svint16_t svqabs_x(svbool_t, svint16_t);
	21747	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s8_z)))
	21748	svint8_t svqabs_z(svbool_t, svint8_t);
	21749	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s32_z)))
	21750	svint32_t svqabs_z(svbool_t, svint32_t);
	21751	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s64_z)))
	21752	svint64_t svqabs_z(svbool_t, svint64_t);
	21753	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqabs_s16_z)))
	21754	svint16_t svqabs_z(svbool_t, svint16_t);
	21755	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_m)))
	21756	svint8_t svqadd_m(svbool_t, svint8_t, int8_t);
	21757	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_m)))
	21758	svint32_t svqadd_m(svbool_t, svint32_t, int32_t);
	21759	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_m)))
	21760	svint64_t svqadd_m(svbool_t, svint64_t, int64_t);
	21761	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_m)))
	21762	svint16_t svqadd_m(svbool_t, svint16_t, int16_t);
	21763	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_x)))
	21764	svint8_t svqadd_x(svbool_t, svint8_t, int8_t);
	21765	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_x)))
	21766	svint32_t svqadd_x(svbool_t, svint32_t, int32_t);
	21767	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_x)))
	21768	svint64_t svqadd_x(svbool_t, svint64_t, int64_t);
	21769	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_x)))
	21770	svint16_t svqadd_x(svbool_t, svint16_t, int16_t);
	21771	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s8_z)))
	21772	svint8_t svqadd_z(svbool_t, svint8_t, int8_t);
	21773	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s32_z)))
	21774	svint32_t svqadd_z(svbool_t, svint32_t, int32_t);
	21775	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s64_z)))
	21776	svint64_t svqadd_z(svbool_t, svint64_t, int64_t);
	21777	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_s16_z)))
	21778	svint16_t svqadd_z(svbool_t, svint16_t, int16_t);
	21779	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_m)))
	21780	svuint8_t svqadd_m(svbool_t, svuint8_t, uint8_t);
	21781	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_m)))
	21782	svuint32_t svqadd_m(svbool_t, svuint32_t, uint32_t);
	21783	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_m)))
	21784	svuint64_t svqadd_m(svbool_t, svuint64_t, uint64_t);
	21785	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_m)))
	21786	svuint16_t svqadd_m(svbool_t, svuint16_t, uint16_t);
	21787	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_x)))
	21788	svuint8_t svqadd_x(svbool_t, svuint8_t, uint8_t);
	21789	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_x)))
	21790	svuint32_t svqadd_x(svbool_t, svuint32_t, uint32_t);
	21791	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_x)))
	21792	svuint64_t svqadd_x(svbool_t, svuint64_t, uint64_t);
	21793	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_x)))
	21794	svuint16_t svqadd_x(svbool_t, svuint16_t, uint16_t);
	21795	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u8_z)))
	21796	svuint8_t svqadd_z(svbool_t, svuint8_t, uint8_t);
	21797	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u32_z)))
	21798	svuint32_t svqadd_z(svbool_t, svuint32_t, uint32_t);
	21799	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u64_z)))
	21800	svuint64_t svqadd_z(svbool_t, svuint64_t, uint64_t);
	21801	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_n_u16_z)))
	21802	svuint16_t svqadd_z(svbool_t, svuint16_t, uint16_t);
	21803	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_m)))
	21804	svint8_t svqadd_m(svbool_t, svint8_t, svint8_t);
	21805	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_m)))
	21806	svint32_t svqadd_m(svbool_t, svint32_t, svint32_t);
	21807	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_m)))
	21808	svint64_t svqadd_m(svbool_t, svint64_t, svint64_t);
	21809	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_m)))
	21810	svint16_t svqadd_m(svbool_t, svint16_t, svint16_t);
	21811	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_x)))
	21812	svint8_t svqadd_x(svbool_t, svint8_t, svint8_t);
	21813	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_x)))
	21814	svint32_t svqadd_x(svbool_t, svint32_t, svint32_t);
	21815	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_x)))
	21816	svint64_t svqadd_x(svbool_t, svint64_t, svint64_t);
	21817	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_x)))
	21818	svint16_t svqadd_x(svbool_t, svint16_t, svint16_t);
	21819	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s8_z)))
	21820	svint8_t svqadd_z(svbool_t, svint8_t, svint8_t);
	21821	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s32_z)))
	21822	svint32_t svqadd_z(svbool_t, svint32_t, svint32_t);
	21823	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s64_z)))
	21824	svint64_t svqadd_z(svbool_t, svint64_t, svint64_t);
	21825	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_s16_z)))
	21826	svint16_t svqadd_z(svbool_t, svint16_t, svint16_t);
	21827	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_m)))
	21828	svuint8_t svqadd_m(svbool_t, svuint8_t, svuint8_t);
	21829	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_m)))
	21830	svuint32_t svqadd_m(svbool_t, svuint32_t, svuint32_t);
	21831	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_m)))
	21832	svuint64_t svqadd_m(svbool_t, svuint64_t, svuint64_t);
	21833	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_m)))
	21834	svuint16_t svqadd_m(svbool_t, svuint16_t, svuint16_t);
	21835	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_x)))
	21836	svuint8_t svqadd_x(svbool_t, svuint8_t, svuint8_t);
	21837	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_x)))
	21838	svuint32_t svqadd_x(svbool_t, svuint32_t, svuint32_t);
	21839	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_x)))
	21840	svuint64_t svqadd_x(svbool_t, svuint64_t, svuint64_t);
	21841	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_x)))
	21842	svuint16_t svqadd_x(svbool_t, svuint16_t, svuint16_t);
	21843	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u8_z)))
	21844	svuint8_t svqadd_z(svbool_t, svuint8_t, svuint8_t);
	21845	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u32_z)))
	21846	svuint32_t svqadd_z(svbool_t, svuint32_t, svuint32_t);
	21847	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u64_z)))
	21848	svuint64_t svqadd_z(svbool_t, svuint64_t, svuint64_t);
	21849	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqadd_u16_z)))
	21850	svuint16_t svqadd_z(svbool_t, svuint16_t, svuint16_t);
	21851	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s8)))
	21852	svint8_t svqcadd(svint8_t, svint8_t, uint64_t);
	21853	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s32)))
	21854	svint32_t svqcadd(svint32_t, svint32_t, uint64_t);
	21855	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s64)))
	21856	svint64_t svqcadd(svint64_t, svint64_t, uint64_t);
	21857	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcadd_s16)))
	21858	svint16_t svqcadd(svint16_t, svint16_t, uint64_t);
	21859	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s32)))
	21860	svint32_t svqdmlalb(svint32_t, svint16_t, int16_t);
	21861	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s64)))
	21862	svint64_t svqdmlalb(svint64_t, svint32_t, int32_t);
	21863	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_n_s16)))
	21864	svint16_t svqdmlalb(svint16_t, svint8_t, int8_t);
	21865	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s32)))
	21866	svint32_t svqdmlalb(svint32_t, svint16_t, svint16_t);
	21867	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s64)))
	21868	svint64_t svqdmlalb(svint64_t, svint32_t, svint32_t);
	21869	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_s16)))
	21870	svint16_t svqdmlalb(svint16_t, svint8_t, svint8_t);
	21871	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s32)))
	21872	svint32_t svqdmlalb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21873	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalb_lane_s64)))
	21874	svint64_t svqdmlalb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21875	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s32)))
	21876	svint32_t svqdmlalbt(svint32_t, svint16_t, int16_t);
	21877	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s64)))
	21878	svint64_t svqdmlalbt(svint64_t, svint32_t, int32_t);
	21879	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_n_s16)))
	21880	svint16_t svqdmlalbt(svint16_t, svint8_t, int8_t);
	21881	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s32)))
	21882	svint32_t svqdmlalbt(svint32_t, svint16_t, svint16_t);
	21883	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s64)))
	21884	svint64_t svqdmlalbt(svint64_t, svint32_t, svint32_t);
	21885	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalbt_s16)))
	21886	svint16_t svqdmlalbt(svint16_t, svint8_t, svint8_t);
	21887	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s32)))
	21888	svint32_t svqdmlalt(svint32_t, svint16_t, int16_t);
	21889	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s64)))
	21890	svint64_t svqdmlalt(svint64_t, svint32_t, int32_t);
	21891	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_n_s16)))
	21892	svint16_t svqdmlalt(svint16_t, svint8_t, int8_t);
	21893	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s32)))
	21894	svint32_t svqdmlalt(svint32_t, svint16_t, svint16_t);
	21895	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s64)))
	21896	svint64_t svqdmlalt(svint64_t, svint32_t, svint32_t);
	21897	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_s16)))
	21898	svint16_t svqdmlalt(svint16_t, svint8_t, svint8_t);
	21899	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s32)))
	21900	svint32_t svqdmlalt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21901	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlalt_lane_s64)))
	21902	svint64_t svqdmlalt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21903	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s32)))
	21904	svint32_t svqdmlslb(svint32_t, svint16_t, int16_t);
	21905	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s64)))
	21906	svint64_t svqdmlslb(svint64_t, svint32_t, int32_t);
	21907	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_n_s16)))
	21908	svint16_t svqdmlslb(svint16_t, svint8_t, int8_t);
	21909	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s32)))
	21910	svint32_t svqdmlslb(svint32_t, svint16_t, svint16_t);
	21911	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s64)))
	21912	svint64_t svqdmlslb(svint64_t, svint32_t, svint32_t);
	21913	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_s16)))
	21914	svint16_t svqdmlslb(svint16_t, svint8_t, svint8_t);
	21915	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s32)))
	21916	svint32_t svqdmlslb_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21917	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslb_lane_s64)))
	21918	svint64_t svqdmlslb_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21919	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s32)))
	21920	svint32_t svqdmlslbt(svint32_t, svint16_t, int16_t);
	21921	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s64)))
	21922	svint64_t svqdmlslbt(svint64_t, svint32_t, int32_t);
	21923	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_n_s16)))
	21924	svint16_t svqdmlslbt(svint16_t, svint8_t, int8_t);
	21925	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s32)))
	21926	svint32_t svqdmlslbt(svint32_t, svint16_t, svint16_t);
	21927	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s64)))
	21928	svint64_t svqdmlslbt(svint64_t, svint32_t, svint32_t);
	21929	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslbt_s16)))
	21930	svint16_t svqdmlslbt(svint16_t, svint8_t, svint8_t);
	21931	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s32)))
	21932	svint32_t svqdmlslt(svint32_t, svint16_t, int16_t);
	21933	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s64)))
	21934	svint64_t svqdmlslt(svint64_t, svint32_t, int32_t);
	21935	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_n_s16)))
	21936	svint16_t svqdmlslt(svint16_t, svint8_t, int8_t);
	21937	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s32)))
	21938	svint32_t svqdmlslt(svint32_t, svint16_t, svint16_t);
	21939	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s64)))
	21940	svint64_t svqdmlslt(svint64_t, svint32_t, svint32_t);
	21941	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_s16)))
	21942	svint16_t svqdmlslt(svint16_t, svint8_t, svint8_t);
	21943	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s32)))
	21944	svint32_t svqdmlslt_lane(svint32_t, svint16_t, svint16_t, uint64_t);
	21945	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmlslt_lane_s64)))
	21946	svint64_t svqdmlslt_lane(svint64_t, svint32_t, svint32_t, uint64_t);
	21947	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s8)))
	21948	svint8_t svqdmulh(svint8_t, int8_t);
	21949	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s32)))
	21950	svint32_t svqdmulh(svint32_t, int32_t);
	21951	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s64)))
	21952	svint64_t svqdmulh(svint64_t, int64_t);
	21953	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_n_s16)))
	21954	svint16_t svqdmulh(svint16_t, int16_t);
	21955	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s8)))
	21956	svint8_t svqdmulh(svint8_t, svint8_t);
	21957	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s32)))
	21958	svint32_t svqdmulh(svint32_t, svint32_t);
	21959	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s64)))
	21960	svint64_t svqdmulh(svint64_t, svint64_t);
	21961	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_s16)))
	21962	svint16_t svqdmulh(svint16_t, svint16_t);
	21963	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s32)))
	21964	svint32_t svqdmulh_lane(svint32_t, svint32_t, uint64_t);
	21965	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s64)))
	21966	svint64_t svqdmulh_lane(svint64_t, svint64_t, uint64_t);
	21967	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmulh_lane_s16)))
	21968	svint16_t svqdmulh_lane(svint16_t, svint16_t, uint64_t);
	21969	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s32)))
	21970	svint32_t svqdmullb(svint16_t, int16_t);
	21971	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s64)))
	21972	svint64_t svqdmullb(svint32_t, int32_t);
	21973	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_n_s16)))
	21974	svint16_t svqdmullb(svint8_t, int8_t);
	21975	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s32)))
	21976	svint32_t svqdmullb(svint16_t, svint16_t);
	21977	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s64)))
	21978	svint64_t svqdmullb(svint32_t, svint32_t);
	21979	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_s16)))
	21980	svint16_t svqdmullb(svint8_t, svint8_t);
	21981	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s32)))
	21982	svint32_t svqdmullb_lane(svint16_t, svint16_t, uint64_t);
	21983	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullb_lane_s64)))
	21984	svint64_t svqdmullb_lane(svint32_t, svint32_t, uint64_t);
	21985	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s32)))
	21986	svint32_t svqdmullt(svint16_t, int16_t);
	21987	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s64)))
	21988	svint64_t svqdmullt(svint32_t, int32_t);
	21989	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_n_s16)))
	21990	svint16_t svqdmullt(svint8_t, int8_t);
	21991	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s32)))
	21992	svint32_t svqdmullt(svint16_t, svint16_t);
	21993	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s64)))
	21994	svint64_t svqdmullt(svint32_t, svint32_t);
	21995	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_s16)))
	21996	svint16_t svqdmullt(svint8_t, svint8_t);
	21997	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s32)))
	21998	svint32_t svqdmullt_lane(svint16_t, svint16_t, uint64_t);
	21999	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqdmullt_lane_s64)))
	22000	svint64_t svqdmullt_lane(svint32_t, svint32_t, uint64_t);
	22001	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_m)))
	22002	svint8_t svqneg_m(svint8_t, svbool_t, svint8_t);
	22003	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_m)))
	22004	svint32_t svqneg_m(svint32_t, svbool_t, svint32_t);
	22005	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_m)))
	22006	svint64_t svqneg_m(svint64_t, svbool_t, svint64_t);
	22007	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_m)))
	22008	svint16_t svqneg_m(svint16_t, svbool_t, svint16_t);
	22009	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_x)))
	22010	svint8_t svqneg_x(svbool_t, svint8_t);
	22011	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_x)))
	22012	svint32_t svqneg_x(svbool_t, svint32_t);
	22013	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_x)))
	22014	svint64_t svqneg_x(svbool_t, svint64_t);
	22015	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_x)))
	22016	svint16_t svqneg_x(svbool_t, svint16_t);
	22017	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s8_z)))
	22018	svint8_t svqneg_z(svbool_t, svint8_t);
	22019	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s32_z)))
	22020	svint32_t svqneg_z(svbool_t, svint32_t);
	22021	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s64_z)))
	22022	svint64_t svqneg_z(svbool_t, svint64_t);
	22023	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqneg_s16_z)))
	22024	svint16_t svqneg_z(svbool_t, svint16_t);
	22025	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s8)))
	22026	svint8_t svqrdcmlah(svint8_t, svint8_t, svint8_t, uint64_t);
	22027	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s32)))
	22028	svint32_t svqrdcmlah(svint32_t, svint32_t, svint32_t, uint64_t);
	22029	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s64)))
	22030	svint64_t svqrdcmlah(svint64_t, svint64_t, svint64_t, uint64_t);
	22031	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_s16)))
	22032	svint16_t svqrdcmlah(svint16_t, svint16_t, svint16_t, uint64_t);
	22033	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s32)))
	22034	svint32_t svqrdcmlah_lane(svint32_t, svint32_t, svint32_t, uint64_t, uint64_t);
	22035	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdcmlah_lane_s16)))
	22036	svint16_t svqrdcmlah_lane(svint16_t, svint16_t, svint16_t, uint64_t, uint64_t);
	22037	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s8)))
	22038	svint8_t svqrdmlah(svint8_t, svint8_t, int8_t);
	22039	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s32)))
	22040	svint32_t svqrdmlah(svint32_t, svint32_t, int32_t);
	22041	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s64)))
	22042	svint64_t svqrdmlah(svint64_t, svint64_t, int64_t);
	22043	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_n_s16)))
	22044	svint16_t svqrdmlah(svint16_t, svint16_t, int16_t);
	22045	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s8)))
	22046	svint8_t svqrdmlah(svint8_t, svint8_t, svint8_t);
	22047	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s32)))
	22048	svint32_t svqrdmlah(svint32_t, svint32_t, svint32_t);
	22049	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s64)))
	22050	svint64_t svqrdmlah(svint64_t, svint64_t, svint64_t);
	22051	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_s16)))
	22052	svint16_t svqrdmlah(svint16_t, svint16_t, svint16_t);
	22053	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s32)))
	22054	svint32_t svqrdmlah_lane(svint32_t, svint32_t, svint32_t, uint64_t);
	22055	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s64)))
	22056	svint64_t svqrdmlah_lane(svint64_t, svint64_t, svint64_t, uint64_t);
	22057	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlah_lane_s16)))
	22058	svint16_t svqrdmlah_lane(svint16_t, svint16_t, svint16_t, uint64_t);
	22059	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s8)))
	22060	svint8_t svqrdmlsh(svint8_t, svint8_t, int8_t);
	22061	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s32)))
	22062	svint32_t svqrdmlsh(svint32_t, svint32_t, int32_t);
	22063	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s64)))
	22064	svint64_t svqrdmlsh(svint64_t, svint64_t, int64_t);
	22065	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_n_s16)))
	22066	svint16_t svqrdmlsh(svint16_t, svint16_t, int16_t);
	22067	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s8)))
	22068	svint8_t svqrdmlsh(svint8_t, svint8_t, svint8_t);
	22069	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s32)))
	22070	svint32_t svqrdmlsh(svint32_t, svint32_t, svint32_t);
	22071	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s64)))
	22072	svint64_t svqrdmlsh(svint64_t, svint64_t, svint64_t);
	22073	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_s16)))
	22074	svint16_t svqrdmlsh(svint16_t, svint16_t, svint16_t);
	22075	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s32)))
	22076	svint32_t svqrdmlsh_lane(svint32_t, svint32_t, svint32_t, uint64_t);
	22077	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s64)))
	22078	svint64_t svqrdmlsh_lane(svint64_t, svint64_t, svint64_t, uint64_t);
	22079	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmlsh_lane_s16)))
	22080	svint16_t svqrdmlsh_lane(svint16_t, svint16_t, svint16_t, uint64_t);
	22081	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s8)))
	22082	svint8_t svqrdmulh(svint8_t, int8_t);
	22083	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s32)))
	22084	svint32_t svqrdmulh(svint32_t, int32_t);
	22085	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s64)))
	22086	svint64_t svqrdmulh(svint64_t, int64_t);
	22087	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_n_s16)))
	22088	svint16_t svqrdmulh(svint16_t, int16_t);
	22089	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s8)))
	22090	svint8_t svqrdmulh(svint8_t, svint8_t);
	22091	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s32)))
	22092	svint32_t svqrdmulh(svint32_t, svint32_t);
	22093	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s64)))
	22094	svint64_t svqrdmulh(svint64_t, svint64_t);
	22095	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_s16)))
	22096	svint16_t svqrdmulh(svint16_t, svint16_t);
	22097	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s32)))
	22098	svint32_t svqrdmulh_lane(svint32_t, svint32_t, uint64_t);
	22099	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s64)))
	22100	svint64_t svqrdmulh_lane(svint64_t, svint64_t, uint64_t);
	22101	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrdmulh_lane_s16)))
	22102	svint16_t svqrdmulh_lane(svint16_t, svint16_t, uint64_t);
	22103	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_m)))
	22104	svint8_t svqrshl_m(svbool_t, svint8_t, int8_t);
	22105	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_m)))
	22106	svint32_t svqrshl_m(svbool_t, svint32_t, int32_t);
	22107	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_m)))
	22108	svint64_t svqrshl_m(svbool_t, svint64_t, int64_t);
	22109	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_m)))
	22110	svint16_t svqrshl_m(svbool_t, svint16_t, int16_t);
	22111	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_x)))
	22112	svint8_t svqrshl_x(svbool_t, svint8_t, int8_t);
	22113	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_x)))
	22114	svint32_t svqrshl_x(svbool_t, svint32_t, int32_t);
	22115	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_x)))
	22116	svint64_t svqrshl_x(svbool_t, svint64_t, int64_t);
	22117	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_x)))
	22118	svint16_t svqrshl_x(svbool_t, svint16_t, int16_t);
	22119	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s8_z)))
	22120	svint8_t svqrshl_z(svbool_t, svint8_t, int8_t);
	22121	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s32_z)))
	22122	svint32_t svqrshl_z(svbool_t, svint32_t, int32_t);
	22123	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s64_z)))
	22124	svint64_t svqrshl_z(svbool_t, svint64_t, int64_t);
	22125	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_s16_z)))
	22126	svint16_t svqrshl_z(svbool_t, svint16_t, int16_t);
	22127	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_m)))
	22128	svuint8_t svqrshl_m(svbool_t, svuint8_t, int8_t);
	22129	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_m)))
	22130	svuint32_t svqrshl_m(svbool_t, svuint32_t, int32_t);
	22131	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_m)))
	22132	svuint64_t svqrshl_m(svbool_t, svuint64_t, int64_t);
	22133	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_m)))
	22134	svuint16_t svqrshl_m(svbool_t, svuint16_t, int16_t);
	22135	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_x)))
	22136	svuint8_t svqrshl_x(svbool_t, svuint8_t, int8_t);
	22137	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_x)))
	22138	svuint32_t svqrshl_x(svbool_t, svuint32_t, int32_t);
	22139	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_x)))
	22140	svuint64_t svqrshl_x(svbool_t, svuint64_t, int64_t);
	22141	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_x)))
	22142	svuint16_t svqrshl_x(svbool_t, svuint16_t, int16_t);
	22143	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u8_z)))
	22144	svuint8_t svqrshl_z(svbool_t, svuint8_t, int8_t);
	22145	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u32_z)))
	22146	svuint32_t svqrshl_z(svbool_t, svuint32_t, int32_t);
	22147	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u64_z)))
	22148	svuint64_t svqrshl_z(svbool_t, svuint64_t, int64_t);
	22149	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_n_u16_z)))
	22150	svuint16_t svqrshl_z(svbool_t, svuint16_t, int16_t);
	22151	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_m)))
	22152	svint8_t svqrshl_m(svbool_t, svint8_t, svint8_t);
	22153	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_m)))
	22154	svint32_t svqrshl_m(svbool_t, svint32_t, svint32_t);
	22155	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_m)))
	22156	svint64_t svqrshl_m(svbool_t, svint64_t, svint64_t);
	22157	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_m)))
	22158	svint16_t svqrshl_m(svbool_t, svint16_t, svint16_t);
	22159	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_x)))
	22160	svint8_t svqrshl_x(svbool_t, svint8_t, svint8_t);
	22161	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_x)))
	22162	svint32_t svqrshl_x(svbool_t, svint32_t, svint32_t);
	22163	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_x)))
	22164	svint64_t svqrshl_x(svbool_t, svint64_t, svint64_t);
	22165	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_x)))
	22166	svint16_t svqrshl_x(svbool_t, svint16_t, svint16_t);
	22167	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s8_z)))
	22168	svint8_t svqrshl_z(svbool_t, svint8_t, svint8_t);
	22169	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s32_z)))
	22170	svint32_t svqrshl_z(svbool_t, svint32_t, svint32_t);
	22171	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s64_z)))
	22172	svint64_t svqrshl_z(svbool_t, svint64_t, svint64_t);
	22173	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_s16_z)))
	22174	svint16_t svqrshl_z(svbool_t, svint16_t, svint16_t);
	22175	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_m)))
	22176	svuint8_t svqrshl_m(svbool_t, svuint8_t, svint8_t);
	22177	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_m)))
	22178	svuint32_t svqrshl_m(svbool_t, svuint32_t, svint32_t);
	22179	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_m)))
	22180	svuint64_t svqrshl_m(svbool_t, svuint64_t, svint64_t);
	22181	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_m)))
	22182	svuint16_t svqrshl_m(svbool_t, svuint16_t, svint16_t);
	22183	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_x)))
	22184	svuint8_t svqrshl_x(svbool_t, svuint8_t, svint8_t);
	22185	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_x)))
	22186	svuint32_t svqrshl_x(svbool_t, svuint32_t, svint32_t);
	22187	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_x)))
	22188	svuint64_t svqrshl_x(svbool_t, svuint64_t, svint64_t);
	22189	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_x)))
	22190	svuint16_t svqrshl_x(svbool_t, svuint16_t, svint16_t);
	22191	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u8_z)))
	22192	svuint8_t svqrshl_z(svbool_t, svuint8_t, svint8_t);
	22193	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u32_z)))
	22194	svuint32_t svqrshl_z(svbool_t, svuint32_t, svint32_t);
	22195	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u64_z)))
	22196	svuint64_t svqrshl_z(svbool_t, svuint64_t, svint64_t);
	22197	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshl_u16_z)))
	22198	svuint16_t svqrshl_z(svbool_t, svuint16_t, svint16_t);
	22199	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s32)))
	22200	svint16_t svqrshrnb(svint32_t, uint64_t);
	22201	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s64)))
	22202	svint32_t svqrshrnb(svint64_t, uint64_t);
	22203	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_s16)))
	22204	svint8_t svqrshrnb(svint16_t, uint64_t);
	22205	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u32)))
	22206	svuint16_t svqrshrnb(svuint32_t, uint64_t);
	22207	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u64)))
	22208	svuint32_t svqrshrnb(svuint64_t, uint64_t);
	22209	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnb_n_u16)))
	22210	svuint8_t svqrshrnb(svuint16_t, uint64_t);
	22211	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s32)))
	22212	svint16_t svqrshrnt(svint16_t, svint32_t, uint64_t);
	22213	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s64)))
	22214	svint32_t svqrshrnt(svint32_t, svint64_t, uint64_t);
	22215	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_s16)))
	22216	svint8_t svqrshrnt(svint8_t, svint16_t, uint64_t);
	22217	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u32)))
	22218	svuint16_t svqrshrnt(svuint16_t, svuint32_t, uint64_t);
	22219	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u64)))
	22220	svuint32_t svqrshrnt(svuint32_t, svuint64_t, uint64_t);
	22221	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrnt_n_u16)))
	22222	svuint8_t svqrshrnt(svuint8_t, svuint16_t, uint64_t);
	22223	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s32)))
	22224	svuint16_t svqrshrunb(svint32_t, uint64_t);
	22225	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s64)))
	22226	svuint32_t svqrshrunb(svint64_t, uint64_t);
	22227	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunb_n_s16)))
	22228	svuint8_t svqrshrunb(svint16_t, uint64_t);
	22229	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s32)))
	22230	svuint16_t svqrshrunt(svuint16_t, svint32_t, uint64_t);
	22231	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s64)))
	22232	svuint32_t svqrshrunt(svuint32_t, svint64_t, uint64_t);
	22233	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrunt_n_s16)))
	22234	svuint8_t svqrshrunt(svuint8_t, svint16_t, uint64_t);
	22235	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_m)))
	22236	svint8_t svqshl_m(svbool_t, svint8_t, int8_t);
	22237	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_m)))
	22238	svint32_t svqshl_m(svbool_t, svint32_t, int32_t);
	22239	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_m)))
	22240	svint64_t svqshl_m(svbool_t, svint64_t, int64_t);
	22241	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_m)))
	22242	svint16_t svqshl_m(svbool_t, svint16_t, int16_t);
	22243	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_x)))
	22244	svint8_t svqshl_x(svbool_t, svint8_t, int8_t);
	22245	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_x)))
	22246	svint32_t svqshl_x(svbool_t, svint32_t, int32_t);
	22247	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_x)))
	22248	svint64_t svqshl_x(svbool_t, svint64_t, int64_t);
	22249	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_x)))
	22250	svint16_t svqshl_x(svbool_t, svint16_t, int16_t);
	22251	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s8_z)))
	22252	svint8_t svqshl_z(svbool_t, svint8_t, int8_t);
	22253	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s32_z)))
	22254	svint32_t svqshl_z(svbool_t, svint32_t, int32_t);
	22255	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s64_z)))
	22256	svint64_t svqshl_z(svbool_t, svint64_t, int64_t);
	22257	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_s16_z)))
	22258	svint16_t svqshl_z(svbool_t, svint16_t, int16_t);
	22259	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_m)))
	22260	svuint8_t svqshl_m(svbool_t, svuint8_t, int8_t);
	22261	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_m)))
	22262	svuint32_t svqshl_m(svbool_t, svuint32_t, int32_t);
	22263	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_m)))
	22264	svuint64_t svqshl_m(svbool_t, svuint64_t, int64_t);
	22265	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_m)))
	22266	svuint16_t svqshl_m(svbool_t, svuint16_t, int16_t);
	22267	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_x)))
	22268	svuint8_t svqshl_x(svbool_t, svuint8_t, int8_t);
	22269	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_x)))
	22270	svuint32_t svqshl_x(svbool_t, svuint32_t, int32_t);
	22271	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_x)))
	22272	svuint64_t svqshl_x(svbool_t, svuint64_t, int64_t);
	22273	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_x)))
	22274	svuint16_t svqshl_x(svbool_t, svuint16_t, int16_t);
	22275	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u8_z)))
	22276	svuint8_t svqshl_z(svbool_t, svuint8_t, int8_t);
	22277	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u32_z)))
	22278	svuint32_t svqshl_z(svbool_t, svuint32_t, int32_t);
	22279	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u64_z)))
	22280	svuint64_t svqshl_z(svbool_t, svuint64_t, int64_t);
	22281	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_n_u16_z)))
	22282	svuint16_t svqshl_z(svbool_t, svuint16_t, int16_t);
	22283	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_m)))
	22284	svint8_t svqshl_m(svbool_t, svint8_t, svint8_t);
	22285	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_m)))
	22286	svint32_t svqshl_m(svbool_t, svint32_t, svint32_t);
	22287	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_m)))
	22288	svint64_t svqshl_m(svbool_t, svint64_t, svint64_t);
	22289	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_m)))
	22290	svint16_t svqshl_m(svbool_t, svint16_t, svint16_t);
	22291	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_x)))
	22292	svint8_t svqshl_x(svbool_t, svint8_t, svint8_t);
	22293	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_x)))
	22294	svint32_t svqshl_x(svbool_t, svint32_t, svint32_t);
	22295	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_x)))
	22296	svint64_t svqshl_x(svbool_t, svint64_t, svint64_t);
	22297	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_x)))
	22298	svint16_t svqshl_x(svbool_t, svint16_t, svint16_t);
	22299	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s8_z)))
	22300	svint8_t svqshl_z(svbool_t, svint8_t, svint8_t);
	22301	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s32_z)))
	22302	svint32_t svqshl_z(svbool_t, svint32_t, svint32_t);
	22303	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s64_z)))
	22304	svint64_t svqshl_z(svbool_t, svint64_t, svint64_t);
	22305	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_s16_z)))
	22306	svint16_t svqshl_z(svbool_t, svint16_t, svint16_t);
	22307	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_m)))
	22308	svuint8_t svqshl_m(svbool_t, svuint8_t, svint8_t);
	22309	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_m)))
	22310	svuint32_t svqshl_m(svbool_t, svuint32_t, svint32_t);
	22311	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_m)))
	22312	svuint64_t svqshl_m(svbool_t, svuint64_t, svint64_t);
	22313	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_m)))
	22314	svuint16_t svqshl_m(svbool_t, svuint16_t, svint16_t);
	22315	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_x)))
	22316	svuint8_t svqshl_x(svbool_t, svuint8_t, svint8_t);
	22317	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_x)))
	22318	svuint32_t svqshl_x(svbool_t, svuint32_t, svint32_t);
	22319	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_x)))
	22320	svuint64_t svqshl_x(svbool_t, svuint64_t, svint64_t);
	22321	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_x)))
	22322	svuint16_t svqshl_x(svbool_t, svuint16_t, svint16_t);
	22323	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u8_z)))
	22324	svuint8_t svqshl_z(svbool_t, svuint8_t, svint8_t);
	22325	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u32_z)))
	22326	svuint32_t svqshl_z(svbool_t, svuint32_t, svint32_t);
	22327	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u64_z)))
	22328	svuint64_t svqshl_z(svbool_t, svuint64_t, svint64_t);
	22329	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshl_u16_z)))
	22330	svuint16_t svqshl_z(svbool_t, svuint16_t, svint16_t);
	22331	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_m)))
	22332	svuint8_t svqshlu_m(svbool_t, svint8_t, uint64_t);
	22333	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_m)))
	22334	svuint32_t svqshlu_m(svbool_t, svint32_t, uint64_t);
	22335	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_m)))
	22336	svuint64_t svqshlu_m(svbool_t, svint64_t, uint64_t);
	22337	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_m)))
	22338	svuint16_t svqshlu_m(svbool_t, svint16_t, uint64_t);
	22339	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_x)))
	22340	svuint8_t svqshlu_x(svbool_t, svint8_t, uint64_t);
	22341	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_x)))
	22342	svuint32_t svqshlu_x(svbool_t, svint32_t, uint64_t);
	22343	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_x)))
	22344	svuint64_t svqshlu_x(svbool_t, svint64_t, uint64_t);
	22345	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_x)))
	22346	svuint16_t svqshlu_x(svbool_t, svint16_t, uint64_t);
	22347	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s8_z)))
	22348	svuint8_t svqshlu_z(svbool_t, svint8_t, uint64_t);
	22349	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s32_z)))
	22350	svuint32_t svqshlu_z(svbool_t, svint32_t, uint64_t);
	22351	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s64_z)))
	22352	svuint64_t svqshlu_z(svbool_t, svint64_t, uint64_t);
	22353	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshlu_n_s16_z)))
	22354	svuint16_t svqshlu_z(svbool_t, svint16_t, uint64_t);
	22355	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s32)))
	22356	svint16_t svqshrnb(svint32_t, uint64_t);
	22357	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s64)))
	22358	svint32_t svqshrnb(svint64_t, uint64_t);
	22359	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_s16)))
	22360	svint8_t svqshrnb(svint16_t, uint64_t);
	22361	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u32)))
	22362	svuint16_t svqshrnb(svuint32_t, uint64_t);
	22363	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u64)))
	22364	svuint32_t svqshrnb(svuint64_t, uint64_t);
	22365	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnb_n_u16)))
	22366	svuint8_t svqshrnb(svuint16_t, uint64_t);
	22367	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s32)))
	22368	svint16_t svqshrnt(svint16_t, svint32_t, uint64_t);
	22369	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s64)))
	22370	svint32_t svqshrnt(svint32_t, svint64_t, uint64_t);
	22371	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_s16)))
	22372	svint8_t svqshrnt(svint8_t, svint16_t, uint64_t);
	22373	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u32)))
	22374	svuint16_t svqshrnt(svuint16_t, svuint32_t, uint64_t);
	22375	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u64)))
	22376	svuint32_t svqshrnt(svuint32_t, svuint64_t, uint64_t);
	22377	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrnt_n_u16)))
	22378	svuint8_t svqshrnt(svuint8_t, svuint16_t, uint64_t);
	22379	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s32)))
	22380	svuint16_t svqshrunb(svint32_t, uint64_t);
	22381	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s64)))
	22382	svuint32_t svqshrunb(svint64_t, uint64_t);
	22383	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunb_n_s16)))
	22384	svuint8_t svqshrunb(svint16_t, uint64_t);
	22385	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s32)))
	22386	svuint16_t svqshrunt(svuint16_t, svint32_t, uint64_t);
	22387	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s64)))
	22388	svuint32_t svqshrunt(svuint32_t, svint64_t, uint64_t);
	22389	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqshrunt_n_s16)))
	22390	svuint8_t svqshrunt(svuint8_t, svint16_t, uint64_t);
	22391	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_m)))
	22392	svint8_t svqsub_m(svbool_t, svint8_t, int8_t);
	22393	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_m)))
	22394	svint32_t svqsub_m(svbool_t, svint32_t, int32_t);
	22395	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_m)))
	22396	svint64_t svqsub_m(svbool_t, svint64_t, int64_t);
	22397	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_m)))
	22398	svint16_t svqsub_m(svbool_t, svint16_t, int16_t);
	22399	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_x)))
	22400	svint8_t svqsub_x(svbool_t, svint8_t, int8_t);
	22401	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_x)))
	22402	svint32_t svqsub_x(svbool_t, svint32_t, int32_t);
	22403	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_x)))
	22404	svint64_t svqsub_x(svbool_t, svint64_t, int64_t);
	22405	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_x)))
	22406	svint16_t svqsub_x(svbool_t, svint16_t, int16_t);
	22407	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s8_z)))
	22408	svint8_t svqsub_z(svbool_t, svint8_t, int8_t);
	22409	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s32_z)))
	22410	svint32_t svqsub_z(svbool_t, svint32_t, int32_t);
	22411	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s64_z)))
	22412	svint64_t svqsub_z(svbool_t, svint64_t, int64_t);
	22413	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_s16_z)))
	22414	svint16_t svqsub_z(svbool_t, svint16_t, int16_t);
	22415	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_m)))
	22416	svuint8_t svqsub_m(svbool_t, svuint8_t, uint8_t);
	22417	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_m)))
	22418	svuint32_t svqsub_m(svbool_t, svuint32_t, uint32_t);
	22419	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_m)))
	22420	svuint64_t svqsub_m(svbool_t, svuint64_t, uint64_t);
	22421	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_m)))
	22422	svuint16_t svqsub_m(svbool_t, svuint16_t, uint16_t);
	22423	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_x)))
	22424	svuint8_t svqsub_x(svbool_t, svuint8_t, uint8_t);
	22425	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_x)))
	22426	svuint32_t svqsub_x(svbool_t, svuint32_t, uint32_t);
	22427	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_x)))
	22428	svuint64_t svqsub_x(svbool_t, svuint64_t, uint64_t);
	22429	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_x)))
	22430	svuint16_t svqsub_x(svbool_t, svuint16_t, uint16_t);
	22431	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u8_z)))
	22432	svuint8_t svqsub_z(svbool_t, svuint8_t, uint8_t);
	22433	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u32_z)))
	22434	svuint32_t svqsub_z(svbool_t, svuint32_t, uint32_t);
	22435	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u64_z)))
	22436	svuint64_t svqsub_z(svbool_t, svuint64_t, uint64_t);
	22437	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_n_u16_z)))
	22438	svuint16_t svqsub_z(svbool_t, svuint16_t, uint16_t);
	22439	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_m)))
	22440	svint8_t svqsub_m(svbool_t, svint8_t, svint8_t);
	22441	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_m)))
	22442	svint32_t svqsub_m(svbool_t, svint32_t, svint32_t);
	22443	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_m)))
	22444	svint64_t svqsub_m(svbool_t, svint64_t, svint64_t);
	22445	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_m)))
	22446	svint16_t svqsub_m(svbool_t, svint16_t, svint16_t);
	22447	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_x)))
	22448	svint8_t svqsub_x(svbool_t, svint8_t, svint8_t);
	22449	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_x)))
	22450	svint32_t svqsub_x(svbool_t, svint32_t, svint32_t);
	22451	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_x)))
	22452	svint64_t svqsub_x(svbool_t, svint64_t, svint64_t);
	22453	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_x)))
	22454	svint16_t svqsub_x(svbool_t, svint16_t, svint16_t);
	22455	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s8_z)))
	22456	svint8_t svqsub_z(svbool_t, svint8_t, svint8_t);
	22457	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s32_z)))
	22458	svint32_t svqsub_z(svbool_t, svint32_t, svint32_t);
	22459	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s64_z)))
	22460	svint64_t svqsub_z(svbool_t, svint64_t, svint64_t);
	22461	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_s16_z)))
	22462	svint16_t svqsub_z(svbool_t, svint16_t, svint16_t);
	22463	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_m)))
	22464	svuint8_t svqsub_m(svbool_t, svuint8_t, svuint8_t);
	22465	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_m)))
	22466	svuint32_t svqsub_m(svbool_t, svuint32_t, svuint32_t);
	22467	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_m)))
	22468	svuint64_t svqsub_m(svbool_t, svuint64_t, svuint64_t);
	22469	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_m)))
	22470	svuint16_t svqsub_m(svbool_t, svuint16_t, svuint16_t);
	22471	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_x)))
	22472	svuint8_t svqsub_x(svbool_t, svuint8_t, svuint8_t);
	22473	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_x)))
	22474	svuint32_t svqsub_x(svbool_t, svuint32_t, svuint32_t);
	22475	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_x)))
	22476	svuint64_t svqsub_x(svbool_t, svuint64_t, svuint64_t);
	22477	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_x)))
	22478	svuint16_t svqsub_x(svbool_t, svuint16_t, svuint16_t);
	22479	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u8_z)))
	22480	svuint8_t svqsub_z(svbool_t, svuint8_t, svuint8_t);
	22481	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u32_z)))
	22482	svuint32_t svqsub_z(svbool_t, svuint32_t, svuint32_t);
	22483	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u64_z)))
	22484	svuint64_t svqsub_z(svbool_t, svuint64_t, svuint64_t);
	22485	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsub_u16_z)))
	22486	svuint16_t svqsub_z(svbool_t, svuint16_t, svuint16_t);
	22487	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_m)))
	22488	svint8_t svqsubr_m(svbool_t, svint8_t, int8_t);
	22489	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_m)))
	22490	svint32_t svqsubr_m(svbool_t, svint32_t, int32_t);
	22491	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_m)))
	22492	svint64_t svqsubr_m(svbool_t, svint64_t, int64_t);
	22493	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_m)))
	22494	svint16_t svqsubr_m(svbool_t, svint16_t, int16_t);
	22495	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_x)))
	22496	svint8_t svqsubr_x(svbool_t, svint8_t, int8_t);
	22497	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_x)))
	22498	svint32_t svqsubr_x(svbool_t, svint32_t, int32_t);
	22499	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_x)))
	22500	svint64_t svqsubr_x(svbool_t, svint64_t, int64_t);
	22501	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_x)))
	22502	svint16_t svqsubr_x(svbool_t, svint16_t, int16_t);
	22503	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s8_z)))
	22504	svint8_t svqsubr_z(svbool_t, svint8_t, int8_t);
	22505	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s32_z)))
	22506	svint32_t svqsubr_z(svbool_t, svint32_t, int32_t);
	22507	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s64_z)))
	22508	svint64_t svqsubr_z(svbool_t, svint64_t, int64_t);
	22509	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_s16_z)))
	22510	svint16_t svqsubr_z(svbool_t, svint16_t, int16_t);
	22511	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_m)))
	22512	svuint8_t svqsubr_m(svbool_t, svuint8_t, uint8_t);
	22513	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_m)))
	22514	svuint32_t svqsubr_m(svbool_t, svuint32_t, uint32_t);
	22515	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_m)))
	22516	svuint64_t svqsubr_m(svbool_t, svuint64_t, uint64_t);
	22517	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_m)))
	22518	svuint16_t svqsubr_m(svbool_t, svuint16_t, uint16_t);
	22519	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_x)))
	22520	svuint8_t svqsubr_x(svbool_t, svuint8_t, uint8_t);
	22521	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_x)))
	22522	svuint32_t svqsubr_x(svbool_t, svuint32_t, uint32_t);
	22523	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_x)))
	22524	svuint64_t svqsubr_x(svbool_t, svuint64_t, uint64_t);
	22525	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_x)))
	22526	svuint16_t svqsubr_x(svbool_t, svuint16_t, uint16_t);
	22527	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u8_z)))
	22528	svuint8_t svqsubr_z(svbool_t, svuint8_t, uint8_t);
	22529	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u32_z)))
	22530	svuint32_t svqsubr_z(svbool_t, svuint32_t, uint32_t);
	22531	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u64_z)))
	22532	svuint64_t svqsubr_z(svbool_t, svuint64_t, uint64_t);
	22533	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_n_u16_z)))
	22534	svuint16_t svqsubr_z(svbool_t, svuint16_t, uint16_t);
	22535	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_m)))
	22536	svint8_t svqsubr_m(svbool_t, svint8_t, svint8_t);
	22537	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_m)))
	22538	svint32_t svqsubr_m(svbool_t, svint32_t, svint32_t);
	22539	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_m)))
	22540	svint64_t svqsubr_m(svbool_t, svint64_t, svint64_t);
	22541	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_m)))
	22542	svint16_t svqsubr_m(svbool_t, svint16_t, svint16_t);
	22543	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_x)))
	22544	svint8_t svqsubr_x(svbool_t, svint8_t, svint8_t);
	22545	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_x)))
	22546	svint32_t svqsubr_x(svbool_t, svint32_t, svint32_t);
	22547	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_x)))
	22548	svint64_t svqsubr_x(svbool_t, svint64_t, svint64_t);
	22549	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_x)))
	22550	svint16_t svqsubr_x(svbool_t, svint16_t, svint16_t);
	22551	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s8_z)))
	22552	svint8_t svqsubr_z(svbool_t, svint8_t, svint8_t);
	22553	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s32_z)))
	22554	svint32_t svqsubr_z(svbool_t, svint32_t, svint32_t);
	22555	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s64_z)))
	22556	svint64_t svqsubr_z(svbool_t, svint64_t, svint64_t);
	22557	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_s16_z)))
	22558	svint16_t svqsubr_z(svbool_t, svint16_t, svint16_t);
	22559	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_m)))
	22560	svuint8_t svqsubr_m(svbool_t, svuint8_t, svuint8_t);
	22561	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_m)))
	22562	svuint32_t svqsubr_m(svbool_t, svuint32_t, svuint32_t);
	22563	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_m)))
	22564	svuint64_t svqsubr_m(svbool_t, svuint64_t, svuint64_t);
	22565	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_m)))
	22566	svuint16_t svqsubr_m(svbool_t, svuint16_t, svuint16_t);
	22567	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_x)))
	22568	svuint8_t svqsubr_x(svbool_t, svuint8_t, svuint8_t);
	22569	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_x)))
	22570	svuint32_t svqsubr_x(svbool_t, svuint32_t, svuint32_t);
	22571	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_x)))
	22572	svuint64_t svqsubr_x(svbool_t, svuint64_t, svuint64_t);
	22573	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_x)))
	22574	svuint16_t svqsubr_x(svbool_t, svuint16_t, svuint16_t);
	22575	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u8_z)))
	22576	svuint8_t svqsubr_z(svbool_t, svuint8_t, svuint8_t);
	22577	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u32_z)))
	22578	svuint32_t svqsubr_z(svbool_t, svuint32_t, svuint32_t);
	22579	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u64_z)))
	22580	svuint64_t svqsubr_z(svbool_t, svuint64_t, svuint64_t);
	22581	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqsubr_u16_z)))
	22582	svuint16_t svqsubr_z(svbool_t, svuint16_t, svuint16_t);
	22583	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s32)))
	22584	svint16_t svqxtnb(svint32_t);
	22585	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s64)))
	22586	svint32_t svqxtnb(svint64_t);
	22587	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_s16)))
	22588	svint8_t svqxtnb(svint16_t);
	22589	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u32)))
	22590	svuint16_t svqxtnb(svuint32_t);
	22591	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u64)))
	22592	svuint32_t svqxtnb(svuint64_t);
	22593	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnb_u16)))
	22594	svuint8_t svqxtnb(svuint16_t);
	22595	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s32)))
	22596	svint16_t svqxtnt(svint16_t, svint32_t);
	22597	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s64)))
	22598	svint32_t svqxtnt(svint32_t, svint64_t);
	22599	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_s16)))
	22600	svint8_t svqxtnt(svint8_t, svint16_t);
	22601	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u32)))
	22602	svuint16_t svqxtnt(svuint16_t, svuint32_t);
	22603	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u64)))
	22604	svuint32_t svqxtnt(svuint32_t, svuint64_t);
	22605	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtnt_u16)))
	22606	svuint8_t svqxtnt(svuint8_t, svuint16_t);
	22607	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s32)))
	22608	svuint16_t svqxtunb(svint32_t);
	22609	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s64)))
	22610	svuint32_t svqxtunb(svint64_t);
	22611	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunb_s16)))
	22612	svuint8_t svqxtunb(svint16_t);
	22613	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s32)))
	22614	svuint16_t svqxtunt(svuint16_t, svint32_t);
	22615	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s64)))
	22616	svuint32_t svqxtunt(svuint32_t, svint64_t);
	22617	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqxtunt_s16)))
	22618	svuint8_t svqxtunt(svuint8_t, svint16_t);
	22619	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u32)))
	22620	svuint16_t svraddhnb(svuint32_t, uint32_t);
	22621	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u64)))
	22622	svuint32_t svraddhnb(svuint64_t, uint64_t);
	22623	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_u16)))
	22624	svuint8_t svraddhnb(svuint16_t, uint16_t);
	22625	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s32)))
	22626	svint16_t svraddhnb(svint32_t, int32_t);
	22627	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s64)))
	22628	svint32_t svraddhnb(svint64_t, int64_t);
	22629	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_n_s16)))
	22630	svint8_t svraddhnb(svint16_t, int16_t);
	22631	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u32)))
	22632	svuint16_t svraddhnb(svuint32_t, svuint32_t);
	22633	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u64)))
	22634	svuint32_t svraddhnb(svuint64_t, svuint64_t);
	22635	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_u16)))
	22636	svuint8_t svraddhnb(svuint16_t, svuint16_t);
	22637	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s32)))
	22638	svint16_t svraddhnb(svint32_t, svint32_t);
	22639	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s64)))
	22640	svint32_t svraddhnb(svint64_t, svint64_t);
	22641	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnb_s16)))
	22642	svint8_t svraddhnb(svint16_t, svint16_t);
	22643	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u32)))
	22644	svuint16_t svraddhnt(svuint16_t, svuint32_t, uint32_t);
	22645	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u64)))
	22646	svuint32_t svraddhnt(svuint32_t, svuint64_t, uint64_t);
	22647	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_u16)))
	22648	svuint8_t svraddhnt(svuint8_t, svuint16_t, uint16_t);
	22649	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s32)))
	22650	svint16_t svraddhnt(svint16_t, svint32_t, int32_t);
	22651	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s64)))
	22652	svint32_t svraddhnt(svint32_t, svint64_t, int64_t);
	22653	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_n_s16)))
	22654	svint8_t svraddhnt(svint8_t, svint16_t, int16_t);
	22655	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u32)))
	22656	svuint16_t svraddhnt(svuint16_t, svuint32_t, svuint32_t);
	22657	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u64)))
	22658	svuint32_t svraddhnt(svuint32_t, svuint64_t, svuint64_t);
	22659	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_u16)))
	22660	svuint8_t svraddhnt(svuint8_t, svuint16_t, svuint16_t);
	22661	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s32)))
	22662	svint16_t svraddhnt(svint16_t, svint32_t, svint32_t);
	22663	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s64)))
	22664	svint32_t svraddhnt(svint32_t, svint64_t, svint64_t);
	22665	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svraddhnt_s16)))
	22666	svint8_t svraddhnt(svint8_t, svint16_t, svint16_t);
	22667	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_m)))
	22668	svuint32_t svrecpe_m(svuint32_t, svbool_t, svuint32_t);
	22669	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_x)))
	22670	svuint32_t svrecpe_x(svbool_t, svuint32_t);
	22671	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrecpe_u32_z)))
	22672	svuint32_t svrecpe_z(svbool_t, svuint32_t);
	22673	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_m)))
	22674	svint8_t svrhadd_m(svbool_t, svint8_t, int8_t);
	22675	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_m)))
	22676	svint32_t svrhadd_m(svbool_t, svint32_t, int32_t);
	22677	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_m)))
	22678	svint64_t svrhadd_m(svbool_t, svint64_t, int64_t);
	22679	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_m)))
	22680	svint16_t svrhadd_m(svbool_t, svint16_t, int16_t);
	22681	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_x)))
	22682	svint8_t svrhadd_x(svbool_t, svint8_t, int8_t);
	22683	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_x)))
	22684	svint32_t svrhadd_x(svbool_t, svint32_t, int32_t);
	22685	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_x)))
	22686	svint64_t svrhadd_x(svbool_t, svint64_t, int64_t);
	22687	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_x)))
	22688	svint16_t svrhadd_x(svbool_t, svint16_t, int16_t);
	22689	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s8_z)))
	22690	svint8_t svrhadd_z(svbool_t, svint8_t, int8_t);
	22691	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s32_z)))
	22692	svint32_t svrhadd_z(svbool_t, svint32_t, int32_t);
	22693	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s64_z)))
	22694	svint64_t svrhadd_z(svbool_t, svint64_t, int64_t);
	22695	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_s16_z)))
	22696	svint16_t svrhadd_z(svbool_t, svint16_t, int16_t);
	22697	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_m)))
	22698	svuint8_t svrhadd_m(svbool_t, svuint8_t, uint8_t);
	22699	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_m)))
	22700	svuint32_t svrhadd_m(svbool_t, svuint32_t, uint32_t);
	22701	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_m)))
	22702	svuint64_t svrhadd_m(svbool_t, svuint64_t, uint64_t);
	22703	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_m)))
	22704	svuint16_t svrhadd_m(svbool_t, svuint16_t, uint16_t);
	22705	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_x)))
	22706	svuint8_t svrhadd_x(svbool_t, svuint8_t, uint8_t);
	22707	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_x)))
	22708	svuint32_t svrhadd_x(svbool_t, svuint32_t, uint32_t);
	22709	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_x)))
	22710	svuint64_t svrhadd_x(svbool_t, svuint64_t, uint64_t);
	22711	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_x)))
	22712	svuint16_t svrhadd_x(svbool_t, svuint16_t, uint16_t);
	22713	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u8_z)))
	22714	svuint8_t svrhadd_z(svbool_t, svuint8_t, uint8_t);
	22715	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u32_z)))
	22716	svuint32_t svrhadd_z(svbool_t, svuint32_t, uint32_t);
	22717	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u64_z)))
	22718	svuint64_t svrhadd_z(svbool_t, svuint64_t, uint64_t);
	22719	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_n_u16_z)))
	22720	svuint16_t svrhadd_z(svbool_t, svuint16_t, uint16_t);
	22721	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_m)))
	22722	svint8_t svrhadd_m(svbool_t, svint8_t, svint8_t);
	22723	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_m)))
	22724	svint32_t svrhadd_m(svbool_t, svint32_t, svint32_t);
	22725	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_m)))
	22726	svint64_t svrhadd_m(svbool_t, svint64_t, svint64_t);
	22727	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_m)))
	22728	svint16_t svrhadd_m(svbool_t, svint16_t, svint16_t);
	22729	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_x)))
	22730	svint8_t svrhadd_x(svbool_t, svint8_t, svint8_t);
	22731	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_x)))
	22732	svint32_t svrhadd_x(svbool_t, svint32_t, svint32_t);
	22733	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_x)))
	22734	svint64_t svrhadd_x(svbool_t, svint64_t, svint64_t);
	22735	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_x)))
	22736	svint16_t svrhadd_x(svbool_t, svint16_t, svint16_t);
	22737	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s8_z)))
	22738	svint8_t svrhadd_z(svbool_t, svint8_t, svint8_t);
	22739	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s32_z)))
	22740	svint32_t svrhadd_z(svbool_t, svint32_t, svint32_t);
	22741	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s64_z)))
	22742	svint64_t svrhadd_z(svbool_t, svint64_t, svint64_t);
	22743	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_s16_z)))
	22744	svint16_t svrhadd_z(svbool_t, svint16_t, svint16_t);
	22745	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_m)))
	22746	svuint8_t svrhadd_m(svbool_t, svuint8_t, svuint8_t);
	22747	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_m)))
	22748	svuint32_t svrhadd_m(svbool_t, svuint32_t, svuint32_t);
	22749	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_m)))
	22750	svuint64_t svrhadd_m(svbool_t, svuint64_t, svuint64_t);
	22751	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_m)))
	22752	svuint16_t svrhadd_m(svbool_t, svuint16_t, svuint16_t);
	22753	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_x)))
	22754	svuint8_t svrhadd_x(svbool_t, svuint8_t, svuint8_t);
	22755	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_x)))
	22756	svuint32_t svrhadd_x(svbool_t, svuint32_t, svuint32_t);
	22757	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_x)))
	22758	svuint64_t svrhadd_x(svbool_t, svuint64_t, svuint64_t);
	22759	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_x)))
	22760	svuint16_t svrhadd_x(svbool_t, svuint16_t, svuint16_t);
	22761	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u8_z)))
	22762	svuint8_t svrhadd_z(svbool_t, svuint8_t, svuint8_t);
	22763	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u32_z)))
	22764	svuint32_t svrhadd_z(svbool_t, svuint32_t, svuint32_t);
	22765	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u64_z)))
	22766	svuint64_t svrhadd_z(svbool_t, svuint64_t, svuint64_t);
	22767	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrhadd_u16_z)))
	22768	svuint16_t svrhadd_z(svbool_t, svuint16_t, svuint16_t);
	22769	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_m)))
	22770	svint8_t svrshl_m(svbool_t, svint8_t, int8_t);
	22771	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_m)))
	22772	svint32_t svrshl_m(svbool_t, svint32_t, int32_t);
	22773	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_m)))
	22774	svint64_t svrshl_m(svbool_t, svint64_t, int64_t);
	22775	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_m)))
	22776	svint16_t svrshl_m(svbool_t, svint16_t, int16_t);
	22777	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_x)))
	22778	svint8_t svrshl_x(svbool_t, svint8_t, int8_t);
	22779	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_x)))
	22780	svint32_t svrshl_x(svbool_t, svint32_t, int32_t);
	22781	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_x)))
	22782	svint64_t svrshl_x(svbool_t, svint64_t, int64_t);
	22783	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_x)))
	22784	svint16_t svrshl_x(svbool_t, svint16_t, int16_t);
	22785	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s8_z)))
	22786	svint8_t svrshl_z(svbool_t, svint8_t, int8_t);
	22787	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s32_z)))
	22788	svint32_t svrshl_z(svbool_t, svint32_t, int32_t);
	22789	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s64_z)))
	22790	svint64_t svrshl_z(svbool_t, svint64_t, int64_t);
	22791	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_s16_z)))
	22792	svint16_t svrshl_z(svbool_t, svint16_t, int16_t);
	22793	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_m)))
	22794	svuint8_t svrshl_m(svbool_t, svuint8_t, int8_t);
	22795	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_m)))
	22796	svuint32_t svrshl_m(svbool_t, svuint32_t, int32_t);
	22797	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_m)))
	22798	svuint64_t svrshl_m(svbool_t, svuint64_t, int64_t);
	22799	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_m)))
	22800	svuint16_t svrshl_m(svbool_t, svuint16_t, int16_t);
	22801	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_x)))
	22802	svuint8_t svrshl_x(svbool_t, svuint8_t, int8_t);
	22803	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_x)))
	22804	svuint32_t svrshl_x(svbool_t, svuint32_t, int32_t);
	22805	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_x)))
	22806	svuint64_t svrshl_x(svbool_t, svuint64_t, int64_t);
	22807	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_x)))
	22808	svuint16_t svrshl_x(svbool_t, svuint16_t, int16_t);
	22809	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u8_z)))
	22810	svuint8_t svrshl_z(svbool_t, svuint8_t, int8_t);
	22811	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u32_z)))
	22812	svuint32_t svrshl_z(svbool_t, svuint32_t, int32_t);
	22813	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u64_z)))
	22814	svuint64_t svrshl_z(svbool_t, svuint64_t, int64_t);
	22815	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_n_u16_z)))
	22816	svuint16_t svrshl_z(svbool_t, svuint16_t, int16_t);
	22817	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_m)))
	22818	svint8_t svrshl_m(svbool_t, svint8_t, svint8_t);
	22819	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_m)))
	22820	svint32_t svrshl_m(svbool_t, svint32_t, svint32_t);
	22821	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_m)))
	22822	svint64_t svrshl_m(svbool_t, svint64_t, svint64_t);
	22823	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_m)))
	22824	svint16_t svrshl_m(svbool_t, svint16_t, svint16_t);
	22825	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_x)))
	22826	svint8_t svrshl_x(svbool_t, svint8_t, svint8_t);
	22827	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_x)))
	22828	svint32_t svrshl_x(svbool_t, svint32_t, svint32_t);
	22829	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_x)))
	22830	svint64_t svrshl_x(svbool_t, svint64_t, svint64_t);
	22831	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_x)))
	22832	svint16_t svrshl_x(svbool_t, svint16_t, svint16_t);
	22833	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s8_z)))
	22834	svint8_t svrshl_z(svbool_t, svint8_t, svint8_t);
	22835	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s32_z)))
	22836	svint32_t svrshl_z(svbool_t, svint32_t, svint32_t);
	22837	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s64_z)))
	22838	svint64_t svrshl_z(svbool_t, svint64_t, svint64_t);
	22839	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_s16_z)))
	22840	svint16_t svrshl_z(svbool_t, svint16_t, svint16_t);
	22841	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_m)))
	22842	svuint8_t svrshl_m(svbool_t, svuint8_t, svint8_t);
	22843	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_m)))
	22844	svuint32_t svrshl_m(svbool_t, svuint32_t, svint32_t);
	22845	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_m)))
	22846	svuint64_t svrshl_m(svbool_t, svuint64_t, svint64_t);
	22847	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_m)))
	22848	svuint16_t svrshl_m(svbool_t, svuint16_t, svint16_t);
	22849	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_x)))
	22850	svuint8_t svrshl_x(svbool_t, svuint8_t, svint8_t);
	22851	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_x)))
	22852	svuint32_t svrshl_x(svbool_t, svuint32_t, svint32_t);
	22853	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_x)))
	22854	svuint64_t svrshl_x(svbool_t, svuint64_t, svint64_t);
	22855	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_x)))
	22856	svuint16_t svrshl_x(svbool_t, svuint16_t, svint16_t);
	22857	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u8_z)))
	22858	svuint8_t svrshl_z(svbool_t, svuint8_t, svint8_t);
	22859	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u32_z)))
	22860	svuint32_t svrshl_z(svbool_t, svuint32_t, svint32_t);
	22861	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u64_z)))
	22862	svuint64_t svrshl_z(svbool_t, svuint64_t, svint64_t);
	22863	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshl_u16_z)))
	22864	svuint16_t svrshl_z(svbool_t, svuint16_t, svint16_t);
	22865	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_m)))
	22866	svint8_t svrshr_m(svbool_t, svint8_t, uint64_t);
	22867	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_m)))
	22868	svint32_t svrshr_m(svbool_t, svint32_t, uint64_t);
	22869	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_m)))
	22870	svint64_t svrshr_m(svbool_t, svint64_t, uint64_t);
	22871	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_m)))
	22872	svint16_t svrshr_m(svbool_t, svint16_t, uint64_t);
	22873	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_m)))
	22874	svuint8_t svrshr_m(svbool_t, svuint8_t, uint64_t);
	22875	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_m)))
	22876	svuint32_t svrshr_m(svbool_t, svuint32_t, uint64_t);
	22877	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_m)))
	22878	svuint64_t svrshr_m(svbool_t, svuint64_t, uint64_t);
	22879	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_m)))
	22880	svuint16_t svrshr_m(svbool_t, svuint16_t, uint64_t);
	22881	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_x)))
	22882	svint8_t svrshr_x(svbool_t, svint8_t, uint64_t);
	22883	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_x)))
	22884	svint32_t svrshr_x(svbool_t, svint32_t, uint64_t);
	22885	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_x)))
	22886	svint64_t svrshr_x(svbool_t, svint64_t, uint64_t);
	22887	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_x)))
	22888	svint16_t svrshr_x(svbool_t, svint16_t, uint64_t);
	22889	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_x)))
	22890	svuint8_t svrshr_x(svbool_t, svuint8_t, uint64_t);
	22891	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_x)))
	22892	svuint32_t svrshr_x(svbool_t, svuint32_t, uint64_t);
	22893	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_x)))
	22894	svuint64_t svrshr_x(svbool_t, svuint64_t, uint64_t);
	22895	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_x)))
	22896	svuint16_t svrshr_x(svbool_t, svuint16_t, uint64_t);
	22897	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s8_z)))
	22898	svint8_t svrshr_z(svbool_t, svint8_t, uint64_t);
	22899	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s32_z)))
	22900	svint32_t svrshr_z(svbool_t, svint32_t, uint64_t);
	22901	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s64_z)))
	22902	svint64_t svrshr_z(svbool_t, svint64_t, uint64_t);
	22903	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_s16_z)))
	22904	svint16_t svrshr_z(svbool_t, svint16_t, uint64_t);
	22905	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u8_z)))
	22906	svuint8_t svrshr_z(svbool_t, svuint8_t, uint64_t);
	22907	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u32_z)))
	22908	svuint32_t svrshr_z(svbool_t, svuint32_t, uint64_t);
	22909	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u64_z)))
	22910	svuint64_t svrshr_z(svbool_t, svuint64_t, uint64_t);
	22911	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshr_n_u16_z)))
	22912	svuint16_t svrshr_z(svbool_t, svuint16_t, uint64_t);
	22913	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u32)))
	22914	svuint16_t svrshrnb(svuint32_t, uint64_t);
	22915	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u64)))
	22916	svuint32_t svrshrnb(svuint64_t, uint64_t);
	22917	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_u16)))
	22918	svuint8_t svrshrnb(svuint16_t, uint64_t);
	22919	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s32)))
	22920	svint16_t svrshrnb(svint32_t, uint64_t);
	22921	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s64)))
	22922	svint32_t svrshrnb(svint64_t, uint64_t);
	22923	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnb_n_s16)))
	22924	svint8_t svrshrnb(svint16_t, uint64_t);
	22925	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u32)))
	22926	svuint16_t svrshrnt(svuint16_t, svuint32_t, uint64_t);
	22927	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u64)))
	22928	svuint32_t svrshrnt(svuint32_t, svuint64_t, uint64_t);
	22929	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_u16)))
	22930	svuint8_t svrshrnt(svuint8_t, svuint16_t, uint64_t);
	22931	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s32)))
	22932	svint16_t svrshrnt(svint16_t, svint32_t, uint64_t);
	22933	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s64)))
	22934	svint32_t svrshrnt(svint32_t, svint64_t, uint64_t);
	22935	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrshrnt_n_s16)))
	22936	svint8_t svrshrnt(svint8_t, svint16_t, uint64_t);
	22937	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_m)))
	22938	svuint32_t svrsqrte_m(svuint32_t, svbool_t, svuint32_t);
	22939	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_x)))
	22940	svuint32_t svrsqrte_x(svbool_t, svuint32_t);
	22941	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsqrte_u32_z)))
	22942	svuint32_t svrsqrte_z(svbool_t, svuint32_t);
	22943	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s8)))
	22944	svint8_t svrsra(svint8_t, svint8_t, uint64_t);
	22945	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s32)))
	22946	svint32_t svrsra(svint32_t, svint32_t, uint64_t);
	22947	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s64)))
	22948	svint64_t svrsra(svint64_t, svint64_t, uint64_t);
	22949	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_s16)))
	22950	svint16_t svrsra(svint16_t, svint16_t, uint64_t);
	22951	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u8)))
	22952	svuint8_t svrsra(svuint8_t, svuint8_t, uint64_t);
	22953	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u32)))
	22954	svuint32_t svrsra(svuint32_t, svuint32_t, uint64_t);
	22955	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u64)))
	22956	svuint64_t svrsra(svuint64_t, svuint64_t, uint64_t);
	22957	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsra_n_u16)))
	22958	svuint16_t svrsra(svuint16_t, svuint16_t, uint64_t);
	22959	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u32)))
	22960	svuint16_t svrsubhnb(svuint32_t, uint32_t);
	22961	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u64)))
	22962	svuint32_t svrsubhnb(svuint64_t, uint64_t);
	22963	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_u16)))
	22964	svuint8_t svrsubhnb(svuint16_t, uint16_t);
	22965	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s32)))
	22966	svint16_t svrsubhnb(svint32_t, int32_t);
	22967	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s64)))
	22968	svint32_t svrsubhnb(svint64_t, int64_t);
	22969	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_n_s16)))
	22970	svint8_t svrsubhnb(svint16_t, int16_t);
	22971	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u32)))
	22972	svuint16_t svrsubhnb(svuint32_t, svuint32_t);
	22973	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u64)))
	22974	svuint32_t svrsubhnb(svuint64_t, svuint64_t);
	22975	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_u16)))
	22976	svuint8_t svrsubhnb(svuint16_t, svuint16_t);
	22977	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s32)))
	22978	svint16_t svrsubhnb(svint32_t, svint32_t);
	22979	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s64)))
	22980	svint32_t svrsubhnb(svint64_t, svint64_t);
	22981	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnb_s16)))
	22982	svint8_t svrsubhnb(svint16_t, svint16_t);
	22983	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u32)))
	22984	svuint16_t svrsubhnt(svuint16_t, svuint32_t, uint32_t);
	22985	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u64)))
	22986	svuint32_t svrsubhnt(svuint32_t, svuint64_t, uint64_t);
	22987	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_u16)))
	22988	svuint8_t svrsubhnt(svuint8_t, svuint16_t, uint16_t);
	22989	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s32)))
	22990	svint16_t svrsubhnt(svint16_t, svint32_t, int32_t);
	22991	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s64)))
	22992	svint32_t svrsubhnt(svint32_t, svint64_t, int64_t);
	22993	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_n_s16)))
	22994	svint8_t svrsubhnt(svint8_t, svint16_t, int16_t);
	22995	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u32)))
	22996	svuint16_t svrsubhnt(svuint16_t, svuint32_t, svuint32_t);
	22997	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u64)))
	22998	svuint32_t svrsubhnt(svuint32_t, svuint64_t, svuint64_t);
	22999	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_u16)))
	23000	svuint8_t svrsubhnt(svuint8_t, svuint16_t, svuint16_t);
	23001	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s32)))
	23002	svint16_t svrsubhnt(svint16_t, svint32_t, svint32_t);
	23003	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s64)))
	23004	svint32_t svrsubhnt(svint32_t, svint64_t, svint64_t);
	23005	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrsubhnt_s16)))
	23006	svint8_t svrsubhnt(svint8_t, svint16_t, svint16_t);
	23007	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u32)))
	23008	svuint32_t svsbclb(svuint32_t, svuint32_t, uint32_t);
	23009	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_n_u64)))
	23010	svuint64_t svsbclb(svuint64_t, svuint64_t, uint64_t);
	23011	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u32)))
	23012	svuint32_t svsbclb(svuint32_t, svuint32_t, svuint32_t);
	23013	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclb_u64)))
	23014	svuint64_t svsbclb(svuint64_t, svuint64_t, svuint64_t);
	23015	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u32)))
	23016	svuint32_t svsbclt(svuint32_t, svuint32_t, uint32_t);
	23017	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_n_u64)))
	23018	svuint64_t svsbclt(svuint64_t, svuint64_t, uint64_t);
	23019	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u32)))
	23020	svuint32_t svsbclt(svuint32_t, svuint32_t, svuint32_t);
	23021	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsbclt_u64)))
	23022	svuint64_t svsbclt(svuint64_t, svuint64_t, svuint64_t);
	23023	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s32)))
	23024	svint32_t svshllb(svint16_t, uint64_t);
	23025	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s64)))
	23026	svint64_t svshllb(svint32_t, uint64_t);
	23027	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_s16)))
	23028	svint16_t svshllb(svint8_t, uint64_t);
	23029	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u32)))
	23030	svuint32_t svshllb(svuint16_t, uint64_t);
	23031	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u64)))
	23032	svuint64_t svshllb(svuint32_t, uint64_t);
	23033	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllb_n_u16)))
	23034	svuint16_t svshllb(svuint8_t, uint64_t);
	23035	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s32)))
	23036	svint32_t svshllt(svint16_t, uint64_t);
	23037	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s64)))
	23038	svint64_t svshllt(svint32_t, uint64_t);
	23039	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_s16)))
	23040	svint16_t svshllt(svint8_t, uint64_t);
	23041	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u32)))
	23042	svuint32_t svshllt(svuint16_t, uint64_t);
	23043	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u64)))
	23044	svuint64_t svshllt(svuint32_t, uint64_t);
	23045	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshllt_n_u16)))
	23046	svuint16_t svshllt(svuint8_t, uint64_t);
	23047	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u32)))
	23048	svuint16_t svshrnb(svuint32_t, uint64_t);
	23049	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u64)))
	23050	svuint32_t svshrnb(svuint64_t, uint64_t);
	23051	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_u16)))
	23052	svuint8_t svshrnb(svuint16_t, uint64_t);
	23053	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s32)))
	23054	svint16_t svshrnb(svint32_t, uint64_t);
	23055	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s64)))
	23056	svint32_t svshrnb(svint64_t, uint64_t);
	23057	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnb_n_s16)))
	23058	svint8_t svshrnb(svint16_t, uint64_t);
	23059	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u32)))
	23060	svuint16_t svshrnt(svuint16_t, svuint32_t, uint64_t);
	23061	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u64)))
	23062	svuint32_t svshrnt(svuint32_t, svuint64_t, uint64_t);
	23063	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_u16)))
	23064	svuint8_t svshrnt(svuint8_t, svuint16_t, uint64_t);
	23065	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s32)))
	23066	svint16_t svshrnt(svint16_t, svint32_t, uint64_t);
	23067	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s64)))
	23068	svint32_t svshrnt(svint32_t, svint64_t, uint64_t);
	23069	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svshrnt_n_s16)))
	23070	svint8_t svshrnt(svint8_t, svint16_t, uint64_t);
	23071	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u8)))
	23072	svuint8_t svsli(svuint8_t, svuint8_t, uint64_t);
	23073	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u32)))
	23074	svuint32_t svsli(svuint32_t, svuint32_t, uint64_t);
	23075	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u64)))
	23076	svuint64_t svsli(svuint64_t, svuint64_t, uint64_t);
	23077	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_u16)))
	23078	svuint16_t svsli(svuint16_t, svuint16_t, uint64_t);
	23079	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s8)))
	23080	svint8_t svsli(svint8_t, svint8_t, uint64_t);
	23081	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s32)))
	23082	svint32_t svsli(svint32_t, svint32_t, uint64_t);
	23083	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s64)))
	23084	svint64_t svsli(svint64_t, svint64_t, uint64_t);
	23085	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsli_n_s16)))
	23086	svint16_t svsli(svint16_t, svint16_t, uint64_t);
	23087	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_m)))
	23088	svuint8_t svsqadd_m(svbool_t, svuint8_t, int8_t);
	23089	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_m)))
	23090	svuint32_t svsqadd_m(svbool_t, svuint32_t, int32_t);
	23091	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_m)))
	23092	svuint64_t svsqadd_m(svbool_t, svuint64_t, int64_t);
	23093	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_m)))
	23094	svuint16_t svsqadd_m(svbool_t, svuint16_t, int16_t);
	23095	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_x)))
	23096	svuint8_t svsqadd_x(svbool_t, svuint8_t, int8_t);
	23097	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_x)))
	23098	svuint32_t svsqadd_x(svbool_t, svuint32_t, int32_t);
	23099	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_x)))
	23100	svuint64_t svsqadd_x(svbool_t, svuint64_t, int64_t);
	23101	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_x)))
	23102	svuint16_t svsqadd_x(svbool_t, svuint16_t, int16_t);
	23103	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u8_z)))
	23104	svuint8_t svsqadd_z(svbool_t, svuint8_t, int8_t);
	23105	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u32_z)))
	23106	svuint32_t svsqadd_z(svbool_t, svuint32_t, int32_t);
	23107	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u64_z)))
	23108	svuint64_t svsqadd_z(svbool_t, svuint64_t, int64_t);
	23109	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_n_u16_z)))
	23110	svuint16_t svsqadd_z(svbool_t, svuint16_t, int16_t);
	23111	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_m)))
	23112	svuint8_t svsqadd_m(svbool_t, svuint8_t, svint8_t);
	23113	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_m)))
	23114	svuint32_t svsqadd_m(svbool_t, svuint32_t, svint32_t);
	23115	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_m)))
	23116	svuint64_t svsqadd_m(svbool_t, svuint64_t, svint64_t);
	23117	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_m)))
	23118	svuint16_t svsqadd_m(svbool_t, svuint16_t, svint16_t);
	23119	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_x)))
	23120	svuint8_t svsqadd_x(svbool_t, svuint8_t, svint8_t);
	23121	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_x)))
	23122	svuint32_t svsqadd_x(svbool_t, svuint32_t, svint32_t);
	23123	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_x)))
	23124	svuint64_t svsqadd_x(svbool_t, svuint64_t, svint64_t);
	23125	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_x)))
	23126	svuint16_t svsqadd_x(svbool_t, svuint16_t, svint16_t);
	23127	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u8_z)))
	23128	svuint8_t svsqadd_z(svbool_t, svuint8_t, svint8_t);
	23129	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u32_z)))
	23130	svuint32_t svsqadd_z(svbool_t, svuint32_t, svint32_t);
	23131	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u64_z)))
	23132	svuint64_t svsqadd_z(svbool_t, svuint64_t, svint64_t);
	23133	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsqadd_u16_z)))
	23134	svuint16_t svsqadd_z(svbool_t, svuint16_t, svint16_t);
	23135	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s8)))
	23136	svint8_t svsra(svint8_t, svint8_t, uint64_t);
	23137	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s32)))
	23138	svint32_t svsra(svint32_t, svint32_t, uint64_t);
	23139	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s64)))
	23140	svint64_t svsra(svint64_t, svint64_t, uint64_t);
	23141	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_s16)))
	23142	svint16_t svsra(svint16_t, svint16_t, uint64_t);
	23143	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u8)))
	23144	svuint8_t svsra(svuint8_t, svuint8_t, uint64_t);
	23145	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u32)))
	23146	svuint32_t svsra(svuint32_t, svuint32_t, uint64_t);
	23147	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u64)))
	23148	svuint64_t svsra(svuint64_t, svuint64_t, uint64_t);
	23149	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsra_n_u16)))
	23150	svuint16_t svsra(svuint16_t, svuint16_t, uint64_t);
	23151	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u8)))
	23152	svuint8_t svsri(svuint8_t, svuint8_t, uint64_t);
	23153	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u32)))
	23154	svuint32_t svsri(svuint32_t, svuint32_t, uint64_t);
	23155	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u64)))
	23156	svuint64_t svsri(svuint64_t, svuint64_t, uint64_t);
	23157	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_u16)))
	23158	svuint16_t svsri(svuint16_t, svuint16_t, uint64_t);
	23159	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s8)))
	23160	svint8_t svsri(svint8_t, svint8_t, uint64_t);
	23161	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s32)))
	23162	svint32_t svsri(svint32_t, svint32_t, uint64_t);
	23163	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s64)))
	23164	svint64_t svsri(svint64_t, svint64_t, uint64_t);
	23165	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsri_n_s16)))
	23166	svint16_t svsri(svint16_t, svint16_t, uint64_t);
	23167	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_u32)))
	23168	void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
	23169	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_u64)))
	23170	void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
	23171	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_f64)))
	23172	void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svfloat64_t);
	23173	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_f32)))
	23174	void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svfloat32_t);
	23175	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_index_s32)))
	23176	void svstnt1_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
	23177	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_index_s64)))
	23178	void svstnt1_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
	23179	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_u32)))
	23180	void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
	23181	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_u64)))
	23182	void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
	23183	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_f64)))
	23184	void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svfloat64_t);
	23185	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_f32)))
	23186	void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svfloat32_t);
	23187	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_offset_s32)))
	23188	void svstnt1_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
	23189	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_offset_s64)))
	23190	void svstnt1_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
	23191	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_u32)))
	23192	void svstnt1_scatter(svbool_t, svuint32_t, svuint32_t);
	23193	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_u64)))
	23194	void svstnt1_scatter(svbool_t, svuint64_t, svuint64_t);
	23195	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_f64)))
	23196	void svstnt1_scatter(svbool_t, svuint64_t, svfloat64_t);
	23197	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_f32)))
	23198	void svstnt1_scatter(svbool_t, svuint32_t, svfloat32_t);
	23199	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32base_s32)))
	23200	void svstnt1_scatter(svbool_t, svuint32_t, svint32_t);
	23201	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64base_s64)))
	23202	void svstnt1_scatter(svbool_t, svuint64_t, svint64_t);
	23203	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_u64)))
	23204	void svstnt1_scatter_index(svbool_t, uint64_t *, svint64_t, svuint64_t);
	23205	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_f64)))
	23206	void svstnt1_scatter_index(svbool_t, float64_t *, svint64_t, svfloat64_t);
	23207	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64index_s64)))
	23208	void svstnt1_scatter_index(svbool_t, int64_t *, svint64_t, svint64_t);
	23209	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_u64)))
	23210	void svstnt1_scatter_index(svbool_t, uint64_t *, svuint64_t, svuint64_t);
	23211	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_f64)))
	23212	void svstnt1_scatter_index(svbool_t, float64_t *, svuint64_t, svfloat64_t);
	23213	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64index_s64)))
	23214	void svstnt1_scatter_index(svbool_t, int64_t *, svuint64_t, svint64_t);
	23215	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_u32)))
	23216	void svstnt1_scatter_offset(svbool_t, uint32_t *, svuint32_t, svuint32_t);
	23217	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_f32)))
	23218	void svstnt1_scatter_offset(svbool_t, float32_t *, svuint32_t, svfloat32_t);
	23219	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u32offset_s32)))
	23220	void svstnt1_scatter_offset(svbool_t, int32_t *, svuint32_t, svint32_t);
	23221	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_u64)))
	23222	void svstnt1_scatter_offset(svbool_t, uint64_t *, svint64_t, svuint64_t);
	23223	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_f64)))
	23224	void svstnt1_scatter_offset(svbool_t, float64_t *, svint64_t, svfloat64_t);
	23225	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_s64offset_s64)))
	23226	void svstnt1_scatter_offset(svbool_t, int64_t *, svint64_t, svint64_t);
	23227	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_u64)))
	23228	void svstnt1_scatter_offset(svbool_t, uint64_t *, svuint64_t, svuint64_t);
	23229	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_f64)))
	23230	void svstnt1_scatter_offset(svbool_t, float64_t *, svuint64_t, svfloat64_t);
	23231	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_scatter_u64offset_s64)))
	23232	void svstnt1_scatter_offset(svbool_t, int64_t *, svuint64_t, svint64_t);
	23233	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_u32)))
	23234	void svstnt1b_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
	23235	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_u64)))
	23236	void svstnt1b_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
	23237	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_offset_s32)))
	23238	void svstnt1b_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
	23239	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_offset_s64)))
	23240	void svstnt1b_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
	23241	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_u32)))
	23242	void svstnt1b_scatter(svbool_t, svuint32_t, svuint32_t);
	23243	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_u64)))
	23244	void svstnt1b_scatter(svbool_t, svuint64_t, svuint64_t);
	23245	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32base_s32)))
	23246	void svstnt1b_scatter(svbool_t, svuint32_t, svint32_t);
	23247	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64base_s64)))
	23248	void svstnt1b_scatter(svbool_t, svuint64_t, svint64_t);
	23249	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_s32)))
	23250	void svstnt1b_scatter_offset(svbool_t, int8_t *, svuint32_t, svint32_t);
	23251	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u32offset_u32)))
	23252	void svstnt1b_scatter_offset(svbool_t, uint8_t *, svuint32_t, svuint32_t);
	23253	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_s64)))
	23254	void svstnt1b_scatter_offset(svbool_t, int8_t *, svint64_t, svint64_t);
	23255	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_s64offset_u64)))
	23256	void svstnt1b_scatter_offset(svbool_t, uint8_t *, svint64_t, svuint64_t);
	23257	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_s64)))
	23258	void svstnt1b_scatter_offset(svbool_t, int8_t *, svuint64_t, svint64_t);
	23259	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1b_scatter_u64offset_u64)))
	23260	void svstnt1b_scatter_offset(svbool_t, uint8_t *, svuint64_t, svuint64_t);
	23261	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_u32)))
	23262	void svstnt1h_scatter_index(svbool_t, svuint32_t, int64_t, svuint32_t);
	23263	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_u64)))
	23264	void svstnt1h_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
	23265	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_index_s32)))
	23266	void svstnt1h_scatter_index(svbool_t, svuint32_t, int64_t, svint32_t);
	23267	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_index_s64)))
	23268	void svstnt1h_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
	23269	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_u32)))
	23270	void svstnt1h_scatter_offset(svbool_t, svuint32_t, int64_t, svuint32_t);
	23271	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_u64)))
	23272	void svstnt1h_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
	23273	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_offset_s32)))
	23274	void svstnt1h_scatter_offset(svbool_t, svuint32_t, int64_t, svint32_t);
	23275	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_offset_s64)))
	23276	void svstnt1h_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
	23277	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_u32)))
	23278	void svstnt1h_scatter(svbool_t, svuint32_t, svuint32_t);
	23279	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_u64)))
	23280	void svstnt1h_scatter(svbool_t, svuint64_t, svuint64_t);
	23281	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32base_s32)))
	23282	void svstnt1h_scatter(svbool_t, svuint32_t, svint32_t);
	23283	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64base_s64)))
	23284	void svstnt1h_scatter(svbool_t, svuint64_t, svint64_t);
	23285	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_s64)))
	23286	void svstnt1h_scatter_index(svbool_t, int16_t *, svint64_t, svint64_t);
	23287	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64index_u64)))
	23288	void svstnt1h_scatter_index(svbool_t, uint16_t *, svint64_t, svuint64_t);
	23289	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_s64)))
	23290	void svstnt1h_scatter_index(svbool_t, int16_t *, svuint64_t, svint64_t);
	23291	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64index_u64)))
	23292	void svstnt1h_scatter_index(svbool_t, uint16_t *, svuint64_t, svuint64_t);
	23293	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_s32)))
	23294	void svstnt1h_scatter_offset(svbool_t, int16_t *, svuint32_t, svint32_t);
	23295	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u32offset_u32)))
	23296	void svstnt1h_scatter_offset(svbool_t, uint16_t *, svuint32_t, svuint32_t);
	23297	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_s64)))
	23298	void svstnt1h_scatter_offset(svbool_t, int16_t *, svint64_t, svint64_t);
	23299	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_s64offset_u64)))
	23300	void svstnt1h_scatter_offset(svbool_t, uint16_t *, svint64_t, svuint64_t);
	23301	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_s64)))
	23302	void svstnt1h_scatter_offset(svbool_t, int16_t *, svuint64_t, svint64_t);
	23303	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1h_scatter_u64offset_u64)))
	23304	void svstnt1h_scatter_offset(svbool_t, uint16_t *, svuint64_t, svuint64_t);
	23305	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_u64)))
	23306	void svstnt1w_scatter_index(svbool_t, svuint64_t, int64_t, svuint64_t);
	23307	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_index_s64)))
	23308	void svstnt1w_scatter_index(svbool_t, svuint64_t, int64_t, svint64_t);
	23309	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_u64)))
	23310	void svstnt1w_scatter_offset(svbool_t, svuint64_t, int64_t, svuint64_t);
	23311	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_offset_s64)))
	23312	void svstnt1w_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
	23313	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_u64)))
	23314	void svstnt1w_scatter(svbool_t, svuint64_t, svuint64_t);
	23315	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64base_s64)))
	23316	void svstnt1w_scatter(svbool_t, svuint64_t, svint64_t);
	23317	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_s64)))
	23318	void svstnt1w_scatter_index(svbool_t, int32_t *, svint64_t, svint64_t);
	23319	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64index_u64)))
	23320	void svstnt1w_scatter_index(svbool_t, uint32_t *, svint64_t, svuint64_t);
	23321	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_s64)))
	23322	void svstnt1w_scatter_index(svbool_t, int32_t *, svuint64_t, svint64_t);
	23323	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64index_u64)))
	23324	void svstnt1w_scatter_index(svbool_t, uint32_t *, svuint64_t, svuint64_t);
	23325	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_s64)))
	23326	void svstnt1w_scatter_offset(svbool_t, int32_t *, svint64_t, svint64_t);
	23327	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_s64offset_u64)))
	23328	void svstnt1w_scatter_offset(svbool_t, uint32_t *, svint64_t, svuint64_t);
	23329	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_s64)))
	23330	void svstnt1w_scatter_offset(svbool_t, int32_t *, svuint64_t, svint64_t);
	23331	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_u64)))
	23332	void svstnt1w_scatter_offset(svbool_t, uint32_t *, svuint64_t, svuint64_t);
	23333	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u32)))
	23334	svuint16_t svsubhnb(svuint32_t, uint32_t);
	23335	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u64)))
	23336	svuint32_t svsubhnb(svuint64_t, uint64_t);
	23337	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_u16)))
	23338	svuint8_t svsubhnb(svuint16_t, uint16_t);
	23339	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s32)))
	23340	svint16_t svsubhnb(svint32_t, int32_t);
	23341	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s64)))
	23342	svint32_t svsubhnb(svint64_t, int64_t);
	23343	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_n_s16)))
	23344	svint8_t svsubhnb(svint16_t, int16_t);
	23345	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u32)))
	23346	svuint16_t svsubhnb(svuint32_t, svuint32_t);
	23347	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u64)))
	23348	svuint32_t svsubhnb(svuint64_t, svuint64_t);
	23349	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_u16)))
	23350	svuint8_t svsubhnb(svuint16_t, svuint16_t);
	23351	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s32)))
	23352	svint16_t svsubhnb(svint32_t, svint32_t);
	23353	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s64)))
	23354	svint32_t svsubhnb(svint64_t, svint64_t);
	23355	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnb_s16)))
	23356	svint8_t svsubhnb(svint16_t, svint16_t);
	23357	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u32)))
	23358	svuint16_t svsubhnt(svuint16_t, svuint32_t, uint32_t);
	23359	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u64)))
	23360	svuint32_t svsubhnt(svuint32_t, svuint64_t, uint64_t);
	23361	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_u16)))
	23362	svuint8_t svsubhnt(svuint8_t, svuint16_t, uint16_t);
	23363	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s32)))
	23364	svint16_t svsubhnt(svint16_t, svint32_t, int32_t);
	23365	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s64)))
	23366	svint32_t svsubhnt(svint32_t, svint64_t, int64_t);
	23367	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_n_s16)))
	23368	svint8_t svsubhnt(svint8_t, svint16_t, int16_t);
	23369	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u32)))
	23370	svuint16_t svsubhnt(svuint16_t, svuint32_t, svuint32_t);
	23371	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u64)))
	23372	svuint32_t svsubhnt(svuint32_t, svuint64_t, svuint64_t);
	23373	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_u16)))
	23374	svuint8_t svsubhnt(svuint8_t, svuint16_t, svuint16_t);
	23375	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s32)))
	23376	svint16_t svsubhnt(svint16_t, svint32_t, svint32_t);
	23377	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s64)))
	23378	svint32_t svsubhnt(svint32_t, svint64_t, svint64_t);
	23379	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubhnt_s16)))
	23380	svint8_t svsubhnt(svint8_t, svint16_t, svint16_t);
	23381	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s32)))
	23382	svint32_t svsublb(svint16_t, int16_t);
	23383	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s64)))
	23384	svint64_t svsublb(svint32_t, int32_t);
	23385	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_s16)))
	23386	svint16_t svsublb(svint8_t, int8_t);
	23387	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u32)))
	23388	svuint32_t svsublb(svuint16_t, uint16_t);
	23389	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u64)))
	23390	svuint64_t svsublb(svuint32_t, uint32_t);
	23391	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_n_u16)))
	23392	svuint16_t svsublb(svuint8_t, uint8_t);
	23393	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s32)))
	23394	svint32_t svsublb(svint16_t, svint16_t);
	23395	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s64)))
	23396	svint64_t svsublb(svint32_t, svint32_t);
	23397	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_s16)))
	23398	svint16_t svsublb(svint8_t, svint8_t);
	23399	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u32)))
	23400	svuint32_t svsublb(svuint16_t, svuint16_t);
	23401	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u64)))
	23402	svuint64_t svsublb(svuint32_t, svuint32_t);
	23403	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublb_u16)))
	23404	svuint16_t svsublb(svuint8_t, svuint8_t);
	23405	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s32)))
	23406	svint32_t svsublbt(svint16_t, int16_t);
	23407	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s64)))
	23408	svint64_t svsublbt(svint32_t, int32_t);
	23409	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_n_s16)))
	23410	svint16_t svsublbt(svint8_t, int8_t);
	23411	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s32)))
	23412	svint32_t svsublbt(svint16_t, svint16_t);
	23413	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s64)))
	23414	svint64_t svsublbt(svint32_t, svint32_t);
	23415	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublbt_s16)))
	23416	svint16_t svsublbt(svint8_t, svint8_t);
	23417	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s32)))
	23418	svint32_t svsublt(svint16_t, int16_t);
	23419	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s64)))
	23420	svint64_t svsublt(svint32_t, int32_t);
	23421	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_s16)))
	23422	svint16_t svsublt(svint8_t, int8_t);
	23423	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u32)))
	23424	svuint32_t svsublt(svuint16_t, uint16_t);
	23425	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u64)))
	23426	svuint64_t svsublt(svuint32_t, uint32_t);
	23427	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_n_u16)))
	23428	svuint16_t svsublt(svuint8_t, uint8_t);
	23429	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s32)))
	23430	svint32_t svsublt(svint16_t, svint16_t);
	23431	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s64)))
	23432	svint64_t svsublt(svint32_t, svint32_t);
	23433	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_s16)))
	23434	svint16_t svsublt(svint8_t, svint8_t);
	23435	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u32)))
	23436	svuint32_t svsublt(svuint16_t, svuint16_t);
	23437	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u64)))
	23438	svuint64_t svsublt(svuint32_t, svuint32_t);
	23439	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsublt_u16)))
	23440	svuint16_t svsublt(svuint8_t, svuint8_t);
	23441	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s32)))
	23442	svint32_t svsubltb(svint16_t, int16_t);
	23443	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s64)))
	23444	svint64_t svsubltb(svint32_t, int32_t);
	23445	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_n_s16)))
	23446	svint16_t svsubltb(svint8_t, int8_t);
	23447	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s32)))
	23448	svint32_t svsubltb(svint16_t, svint16_t);
	23449	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s64)))
	23450	svint64_t svsubltb(svint32_t, svint32_t);
	23451	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubltb_s16)))
	23452	svint16_t svsubltb(svint8_t, svint8_t);
	23453	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s32)))
	23454	svint32_t svsubwb(svint32_t, int16_t);
	23455	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s64)))
	23456	svint64_t svsubwb(svint64_t, int32_t);
	23457	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_s16)))
	23458	svint16_t svsubwb(svint16_t, int8_t);
	23459	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u32)))
	23460	svuint32_t svsubwb(svuint32_t, uint16_t);
	23461	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u64)))
	23462	svuint64_t svsubwb(svuint64_t, uint32_t);
	23463	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_n_u16)))
	23464	svuint16_t svsubwb(svuint16_t, uint8_t);
	23465	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s32)))
	23466	svint32_t svsubwb(svint32_t, svint16_t);
	23467	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s64)))
	23468	svint64_t svsubwb(svint64_t, svint32_t);
	23469	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_s16)))
	23470	svint16_t svsubwb(svint16_t, svint8_t);
	23471	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u32)))
	23472	svuint32_t svsubwb(svuint32_t, svuint16_t);
	23473	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u64)))
	23474	svuint64_t svsubwb(svuint64_t, svuint32_t);
	23475	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwb_u16)))
	23476	svuint16_t svsubwb(svuint16_t, svuint8_t);
	23477	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s32)))
	23478	svint32_t svsubwt(svint32_t, int16_t);
	23479	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s64)))
	23480	svint64_t svsubwt(svint64_t, int32_t);
	23481	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_s16)))
	23482	svint16_t svsubwt(svint16_t, int8_t);
	23483	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u32)))
	23484	svuint32_t svsubwt(svuint32_t, uint16_t);
	23485	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u64)))
	23486	svuint64_t svsubwt(svuint64_t, uint32_t);
	23487	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_n_u16)))
	23488	svuint16_t svsubwt(svuint16_t, uint8_t);
	23489	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s32)))
	23490	svint32_t svsubwt(svint32_t, svint16_t);
	23491	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s64)))
	23492	svint64_t svsubwt(svint64_t, svint32_t);
	23493	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_s16)))
	23494	svint16_t svsubwt(svint16_t, svint8_t);
	23495	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u32)))
	23496	svuint32_t svsubwt(svuint32_t, svuint16_t);
	23497	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u64)))
	23498	svuint64_t svsubwt(svuint64_t, svuint32_t);
	23499	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsubwt_u16)))
	23500	svuint16_t svsubwt(svuint16_t, svuint8_t);
	23501	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u8)))
	23502	svuint8_t svtbl2(svuint8x2_t, svuint8_t);
	23503	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u32)))
	23504	svuint32_t svtbl2(svuint32x2_t, svuint32_t);
	23505	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
	23506	svuint64_t svtbl2(svuint64x2_t, svuint64_t);
	23507	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
	23508	svuint16_t svtbl2(svuint16x2_t, svuint16_t);
	23509	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
	23510	svint8_t svtbl2(svint8x2_t, svuint8_t);
	23511	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
	23512	svfloat64_t svtbl2(svfloat64x2_t, svuint64_t);
	23513	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f32)))
	23514	svfloat32_t svtbl2(svfloat32x2_t, svuint32_t);
	23515	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f16)))
	23516	svfloat16_t svtbl2(svfloat16x2_t, svuint16_t);
	23517	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s32)))
	23518	svint32_t svtbl2(svint32x2_t, svuint32_t);
	23519	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s64)))
	23520	svint64_t svtbl2(svint64x2_t, svuint64_t);
	23521	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s16)))
	23522	svint16_t svtbl2(svint16x2_t, svuint16_t);
	23523	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u8)))
	23524	svuint8_t svtbx(svuint8_t, svuint8_t, svuint8_t);
	23525	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u32)))
	23526	svuint32_t svtbx(svuint32_t, svuint32_t, svuint32_t);
	23527	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
	23528	svuint64_t svtbx(svuint64_t, svuint64_t, svuint64_t);
	23529	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
	23530	svuint16_t svtbx(svuint16_t, svuint16_t, svuint16_t);
	23531	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
	23532	svint8_t svtbx(svint8_t, svint8_t, svuint8_t);
	23533	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
	23534	svfloat64_t svtbx(svfloat64_t, svfloat64_t, svuint64_t);
	23535	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f32)))
	23536	svfloat32_t svtbx(svfloat32_t, svfloat32_t, svuint32_t);
	23537	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f16)))
	23538	svfloat16_t svtbx(svfloat16_t, svfloat16_t, svuint16_t);
	23539	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s32)))
	23540	svint32_t svtbx(svint32_t, svint32_t, svuint32_t);
	23541	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s64)))
	23542	svint64_t svtbx(svint64_t, svint64_t, svuint64_t);
	23543	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s16)))
	23544	svint16_t svtbx(svint16_t, svint16_t, svuint16_t);
	23545	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_m)))
	23546	svint8_t svuqadd_m(svbool_t, svint8_t, uint8_t);
	23547	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_m)))
	23548	svint32_t svuqadd_m(svbool_t, svint32_t, uint32_t);
	23549	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_m)))
	23550	svint64_t svuqadd_m(svbool_t, svint64_t, uint64_t);
	23551	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_m)))
	23552	svint16_t svuqadd_m(svbool_t, svint16_t, uint16_t);
	23553	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_x)))
	23554	svint8_t svuqadd_x(svbool_t, svint8_t, uint8_t);
	23555	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_x)))
	23556	svint32_t svuqadd_x(svbool_t, svint32_t, uint32_t);
	23557	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_x)))
	23558	svint64_t svuqadd_x(svbool_t, svint64_t, uint64_t);
	23559	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_x)))
	23560	svint16_t svuqadd_x(svbool_t, svint16_t, uint16_t);
	23561	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s8_z)))
	23562	svint8_t svuqadd_z(svbool_t, svint8_t, uint8_t);
	23563	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s32_z)))
	23564	svint32_t svuqadd_z(svbool_t, svint32_t, uint32_t);
	23565	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s64_z)))
	23566	svint64_t svuqadd_z(svbool_t, svint64_t, uint64_t);
	23567	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_n_s16_z)))
	23568	svint16_t svuqadd_z(svbool_t, svint16_t, uint16_t);
	23569	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_m)))
	23570	svint8_t svuqadd_m(svbool_t, svint8_t, svuint8_t);
	23571	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_m)))
	23572	svint32_t svuqadd_m(svbool_t, svint32_t, svuint32_t);
	23573	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_m)))
	23574	svint64_t svuqadd_m(svbool_t, svint64_t, svuint64_t);
	23575	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_m)))
	23576	svint16_t svuqadd_m(svbool_t, svint16_t, svuint16_t);
	23577	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_x)))
	23578	svint8_t svuqadd_x(svbool_t, svint8_t, svuint8_t);
	23579	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_x)))
	23580	svint32_t svuqadd_x(svbool_t, svint32_t, svuint32_t);
	23581	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_x)))
	23582	svint64_t svuqadd_x(svbool_t, svint64_t, svuint64_t);
	23583	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_x)))
	23584	svint16_t svuqadd_x(svbool_t, svint16_t, svuint16_t);
	23585	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s8_z)))
	23586	svint8_t svuqadd_z(svbool_t, svint8_t, svuint8_t);
	23587	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s32_z)))
	23588	svint32_t svuqadd_z(svbool_t, svint32_t, svuint32_t);
	23589	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s64_z)))
	23590	svint64_t svuqadd_z(svbool_t, svint64_t, svuint64_t);
	23591	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuqadd_s16_z)))
	23592	svint16_t svuqadd_z(svbool_t, svint16_t, svuint16_t);
	23593	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s32)))
	23594	svbool_t svwhilege_b8(int32_t, int32_t);
	23595	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s32)))
	23596	svbool_t svwhilege_b32(int32_t, int32_t);
	23597	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s32)))
	23598	svbool_t svwhilege_b64(int32_t, int32_t);
	23599	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s32)))
	23600	svbool_t svwhilege_b16(int32_t, int32_t);
	23601	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64)))
	23602	svbool_t svwhilege_b8(int64_t, int64_t);
	23603	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64)))
	23604	svbool_t svwhilege_b32(int64_t, int64_t);
	23605	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64)))
	23606	svbool_t svwhilege_b64(int64_t, int64_t);
	23607	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64)))
	23608	svbool_t svwhilege_b16(int64_t, int64_t);
	23609	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u32)))
	23610	svbool_t svwhilege_b8(uint32_t, uint32_t);
	23611	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u32)))
	23612	svbool_t svwhilege_b32(uint32_t, uint32_t);
	23613	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u32)))
	23614	svbool_t svwhilege_b64(uint32_t, uint32_t);
	23615	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u32)))
	23616	svbool_t svwhilege_b16(uint32_t, uint32_t);
	23617	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64)))
	23618	svbool_t svwhilege_b8(uint64_t, uint64_t);
	23619	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64)))
	23620	svbool_t svwhilege_b32(uint64_t, uint64_t);
	23621	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64)))
	23622	svbool_t svwhilege_b64(uint64_t, uint64_t);
	23623	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64)))
	23624	svbool_t svwhilege_b16(uint64_t, uint64_t);
	23625	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s32)))
	23626	svbool_t svwhilegt_b8(int32_t, int32_t);
	23627	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s32)))
	23628	svbool_t svwhilegt_b32(int32_t, int32_t);
	23629	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s32)))
	23630	svbool_t svwhilegt_b64(int32_t, int32_t);
	23631	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s32)))
	23632	svbool_t svwhilegt_b16(int32_t, int32_t);
	23633	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64)))
	23634	svbool_t svwhilegt_b8(int64_t, int64_t);
	23635	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64)))
	23636	svbool_t svwhilegt_b32(int64_t, int64_t);
	23637	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64)))
	23638	svbool_t svwhilegt_b64(int64_t, int64_t);
	23639	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64)))
	23640	svbool_t svwhilegt_b16(int64_t, int64_t);
	23641	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u32)))
	23642	svbool_t svwhilegt_b8(uint32_t, uint32_t);
	23643	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u32)))
	23644	svbool_t svwhilegt_b32(uint32_t, uint32_t);
	23645	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u32)))
	23646	svbool_t svwhilegt_b64(uint32_t, uint32_t);
	23647	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u32)))
	23648	svbool_t svwhilegt_b16(uint32_t, uint32_t);
	23649	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64)))
	23650	svbool_t svwhilegt_b8(uint64_t, uint64_t);
	23651	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64)))
	23652	svbool_t svwhilegt_b32(uint64_t, uint64_t);
	23653	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64)))
	23654	svbool_t svwhilegt_b64(uint64_t, uint64_t);
	23655	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64)))
	23656	svbool_t svwhilegt_b16(uint64_t, uint64_t);
	23657	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u8)))
	23658	svbool_t svwhilerw(uint8_t const , uint8_t const );
	23659	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s8)))
	23660	svbool_t svwhilerw(int8_t const , int8_t const );
	23661	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u64)))
	23662	svbool_t svwhilerw(uint64_t const , uint64_t const );
	23663	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f64)))
	23664	svbool_t svwhilerw(float64_t const , float64_t const );
	23665	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
	23666	svbool_t svwhilerw(int64_t const , int64_t const );
	23667	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
	23668	svbool_t svwhilerw(uint16_t const , uint16_t const );
	23669	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
	23670	svbool_t svwhilerw(float16_t const , float16_t const );
	23671	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
	23672	svbool_t svwhilerw(int16_t const , int16_t const );
	23673	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u32)))
	23674	svbool_t svwhilerw(uint32_t const , uint32_t const );
	23675	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f32)))
	23676	svbool_t svwhilerw(float32_t const , float32_t const );
	23677	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s32)))
	23678	svbool_t svwhilerw(int32_t const , int32_t const );
	23679	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u8)))
	23680	svbool_t svwhilewr(uint8_t const , uint8_t const );
	23681	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s8)))
	23682	svbool_t svwhilewr(int8_t const , int8_t const );
	23683	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u64)))
	23684	svbool_t svwhilewr(uint64_t const , uint64_t const );
	23685	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f64)))
	23686	svbool_t svwhilewr(float64_t const , float64_t const );
	23687	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
	23688	svbool_t svwhilewr(int64_t const , int64_t const );
	23689	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
	23690	svbool_t svwhilewr(uint16_t const , uint16_t const );
	23691	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
	23692	svbool_t svwhilewr(float16_t const , float16_t const );
	23693	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
	23694	svbool_t svwhilewr(int16_t const , int16_t const );
	23695	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u32)))
	23696	svbool_t svwhilewr(uint32_t const , uint32_t const );
	23697	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f32)))
	23698	svbool_t svwhilewr(float32_t const , float32_t const );
	23699	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s32)))
	23700	svbool_t svwhilewr(int32_t const , int32_t const );
	23701	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u8)))
	23702	svuint8_t svxar(svuint8_t, svuint8_t, uint64_t);
	23703	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u32)))
	23704	svuint32_t svxar(svuint32_t, svuint32_t, uint64_t);
	23705	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u64)))
	23706	svuint64_t svxar(svuint64_t, svuint64_t, uint64_t);
	23707	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_u16)))
	23708	svuint16_t svxar(svuint16_t, svuint16_t, uint64_t);
	23709	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s8)))
	23710	svint8_t svxar(svint8_t, svint8_t, uint64_t);
	23711	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s32)))
	23712	svint32_t svxar(svint32_t, svint32_t, uint64_t);
	23713	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s64)))
	23714	svint64_t svxar(svint64_t, svint64_t, uint64_t);
	23715	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svxar_n_s16)))
	23716	svint16_t svxar(svint16_t, svint16_t, uint64_t);
	23717	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
	23718	svbfloat16_t svtbl2_bf16(svbfloat16x2_t, svuint16_t);
	23719	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
	23720	svbfloat16_t svtbx_bf16(svbfloat16_t, svbfloat16_t, svuint16_t);
	23721	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
	23722	svbool_t svwhilerw_bf16(bfloat16_t const , bfloat16_t const );
	23723	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
	23724	svbool_t svwhilewr_bf16(bfloat16_t const , bfloat16_t const );
	23725	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
	23726	svbfloat16_t svtbl2(svbfloat16x2_t, svuint16_t);
	23727	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
	23728	svbfloat16_t svtbx(svbfloat16_t, svbfloat16_t, svuint16_t);
	23729	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
	23730	svbool_t svwhilerw(bfloat16_t const , bfloat16_t const );
	23731	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
	23732	svbool_t svwhilewr(bfloat16_t const , bfloat16_t const );
	23733	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesd_u8)))
	23734	svuint8_t svaesd_u8(svuint8_t, svuint8_t);
	23735	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaese_u8)))
	23736	svuint8_t svaese_u8(svuint8_t, svuint8_t);
	23737	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesimc_u8)))
	23738	svuint8_t svaesimc_u8(svuint8_t);
	23739	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesmc_u8)))
	23740	svuint8_t svaesmc_u8(svuint8_t);
	23741	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u64)))
	23742	svuint64_t svpmullb_pair_n_u64(svuint64_t, uint64_t);
	23743	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u64)))
	23744	svuint64_t svpmullb_pair_u64(svuint64_t, svuint64_t);
	23745	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u64)))
	23746	svuint64_t svpmullt_pair_n_u64(svuint64_t, uint64_t);
	23747	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u64)))
	23748	svuint64_t svpmullt_pair_u64(svuint64_t, svuint64_t);
	23749	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesd_u8)))
	23750	svuint8_t svaesd(svuint8_t, svuint8_t);
	23751	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaese_u8)))
	23752	svuint8_t svaese(svuint8_t, svuint8_t);
	23753	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesimc_u8)))
	23754	svuint8_t svaesimc(svuint8_t);
	23755	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaesmc_u8)))
	23756	svuint8_t svaesmc(svuint8_t);
	23757	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_n_u64)))
	23758	svuint64_t svpmullb_pair(svuint64_t, uint64_t);
	23759	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullb_pair_u64)))
	23760	svuint64_t svpmullb_pair(svuint64_t, svuint64_t);
	23761	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_n_u64)))
	23762	svuint64_t svpmullt_pair(svuint64_t, uint64_t);
	23763	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmullt_pair_u64)))
	23764	svuint64_t svpmullt_pair(svuint64_t, svuint64_t);
	23765	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u8)))
	23766	svuint8_t svbdep_n_u8(svuint8_t, uint8_t);
	23767	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u32)))
	23768	svuint32_t svbdep_n_u32(svuint32_t, uint32_t);
	23769	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u64)))
	23770	svuint64_t svbdep_n_u64(svuint64_t, uint64_t);
	23771	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u16)))
	23772	svuint16_t svbdep_n_u16(svuint16_t, uint16_t);
	23773	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u8)))
	23774	svuint8_t svbdep_u8(svuint8_t, svuint8_t);
	23775	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u32)))
	23776	svuint32_t svbdep_u32(svuint32_t, svuint32_t);
	23777	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u64)))
	23778	svuint64_t svbdep_u64(svuint64_t, svuint64_t);
	23779	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u16)))
	23780	svuint16_t svbdep_u16(svuint16_t, svuint16_t);
	23781	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u8)))
	23782	svuint8_t svbext_n_u8(svuint8_t, uint8_t);
	23783	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u32)))
	23784	svuint32_t svbext_n_u32(svuint32_t, uint32_t);
	23785	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u64)))
	23786	svuint64_t svbext_n_u64(svuint64_t, uint64_t);
	23787	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u16)))
	23788	svuint16_t svbext_n_u16(svuint16_t, uint16_t);
	23789	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u8)))
	23790	svuint8_t svbext_u8(svuint8_t, svuint8_t);
	23791	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u32)))
	23792	svuint32_t svbext_u32(svuint32_t, svuint32_t);
	23793	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u64)))
	23794	svuint64_t svbext_u64(svuint64_t, svuint64_t);
	23795	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u16)))
	23796	svuint16_t svbext_u16(svuint16_t, svuint16_t);
	23797	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u8)))
	23798	svuint8_t svbgrp_n_u8(svuint8_t, uint8_t);
	23799	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u32)))
	23800	svuint32_t svbgrp_n_u32(svuint32_t, uint32_t);
	23801	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u64)))
	23802	svuint64_t svbgrp_n_u64(svuint64_t, uint64_t);
	23803	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u16)))
	23804	svuint16_t svbgrp_n_u16(svuint16_t, uint16_t);
	23805	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u8)))
	23806	svuint8_t svbgrp_u8(svuint8_t, svuint8_t);
	23807	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u32)))
	23808	svuint32_t svbgrp_u32(svuint32_t, svuint32_t);
	23809	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u64)))
	23810	svuint64_t svbgrp_u64(svuint64_t, svuint64_t);
	23811	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u16)))
	23812	svuint16_t svbgrp_u16(svuint16_t, svuint16_t);
	23813	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u8)))
	23814	svuint8_t svbdep(svuint8_t, uint8_t);
	23815	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u32)))
	23816	svuint32_t svbdep(svuint32_t, uint32_t);
	23817	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u64)))
	23818	svuint64_t svbdep(svuint64_t, uint64_t);
	23819	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_n_u16)))
	23820	svuint16_t svbdep(svuint16_t, uint16_t);
	23821	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u8)))
	23822	svuint8_t svbdep(svuint8_t, svuint8_t);
	23823	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u32)))
	23824	svuint32_t svbdep(svuint32_t, svuint32_t);
	23825	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u64)))
	23826	svuint64_t svbdep(svuint64_t, svuint64_t);
	23827	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbdep_u16)))
	23828	svuint16_t svbdep(svuint16_t, svuint16_t);
	23829	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u8)))
	23830	svuint8_t svbext(svuint8_t, uint8_t);
	23831	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u32)))
	23832	svuint32_t svbext(svuint32_t, uint32_t);
	23833	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u64)))
	23834	svuint64_t svbext(svuint64_t, uint64_t);
	23835	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_n_u16)))
	23836	svuint16_t svbext(svuint16_t, uint16_t);
	23837	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u8)))
	23838	svuint8_t svbext(svuint8_t, svuint8_t);
	23839	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u32)))
	23840	svuint32_t svbext(svuint32_t, svuint32_t);
	23841	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u64)))
	23842	svuint64_t svbext(svuint64_t, svuint64_t);
	23843	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbext_u16)))
	23844	svuint16_t svbext(svuint16_t, svuint16_t);
	23845	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u8)))
	23846	svuint8_t svbgrp(svuint8_t, uint8_t);
	23847	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u32)))
	23848	svuint32_t svbgrp(svuint32_t, uint32_t);
	23849	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u64)))
	23850	svuint64_t svbgrp(svuint64_t, uint64_t);
	23851	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_n_u16)))
	23852	svuint16_t svbgrp(svuint16_t, uint16_t);
	23853	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u8)))
	23854	svuint8_t svbgrp(svuint8_t, svuint8_t);
	23855	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u32)))
	23856	svuint32_t svbgrp(svuint32_t, svuint32_t);
	23857	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u64)))
	23858	svuint64_t svbgrp(svuint64_t, svuint64_t);
	23859	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u16)))
	23860	svuint16_t svbgrp(svuint16_t, svuint16_t);
	23861	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
	23862	svuint64_t svrax1_u64(svuint64_t, svuint64_t);
	23863	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
	23864	svint64_t svrax1_s64(svint64_t, svint64_t);
	23865	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
	23866	svuint64_t svrax1(svuint64_t, svuint64_t);
	23867	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
	23868	svint64_t svrax1(svint64_t, svint64_t);
	23869	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
	23870	svuint32_t svsm4e_u32(svuint32_t, svuint32_t);
	23871	__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
	23872	svuint32_t svsm4ekey_u32(svuint32_t, svuint32_t);
	23873	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
	23874	svuint32_t svsm4e(svuint32_t, svuint32_t);
	23875	__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
	23876	svuint32_t svsm4ekey(svuint32_t, svuint32_t);
24022	23877	#define svcvtnt_bf16_x svcvtnt_bf16_m
24023	23878	#define svcvtnt_bf16_f32_x svcvtnt_bf16_f32_m
24024		#endif /__ARM_FEATURE_SVE_BF16 /
24025
24026		#if defined(__ARM_FEATURE_SVE2)
24027	23879	#define svcvtnt_f16_x svcvtnt_f16_m
24028	23880	#define svcvtnt_f16_f32_x svcvtnt_f16_f32_m
24029	23881	#define svcvtnt_f32_x svcvtnt_f32_m
...	...	@@ -24032,8 +23884,6 @@ svint32_t svusmmla(svint32_t, svuint8_t, svint8_t);
24032	23884	#define svcvtxnt_f32_x svcvtxnt_f32_m
24033	23885	#define svcvtxnt_f32_f64_x svcvtxnt_f32_f64_m
24034	23886
24035		#endif /__ARM_FEATURE_SVE2 /
24036
24037	23887	#ifdef __cplusplus
24038	23888	} // extern "C"
24039	23889	#endif
...	...	@@ -24042,6 +23892,4 @@ svint32_t svusmmla(svint32_t, svuint8_t, svint8_t);
24042	23892
24043	23893	#undef __aio
24044	23894
24045		#endif /__ARM_FEATURE_SVE /
24046
24047	23895	#endif /* __ARM_SVE_H */

lib/include/avx512bf16intrin.h+18-15

...	...	@@ -10,12 +10,14 @@
10	10	#error "Never use <avx512bf16intrin.h> directly; include <immintrin.h> instead."
11	11	#endif
12	12
	13	#ifdef __SSE2__
	14
13	15	#ifndef __AVX512BF16INTRIN_H
14	16	#define __AVX512BF16INTRIN_H
15	17
16		typedef short __m512bh __attribute__((__vector_size__(64), __aligned__(64)));
17		typedef short __m256bh __attribute__((__vector_size__(32), __aligned__(32)));
18		typedef unsigned short __bfloat16;
	18	typedef __bf16 __v32bf __attribute__((__vector_size__(64), __aligned__(64)));
	19	typedef __bf16 __m512bh __attribute__((__vector_size__(64), __aligned__(64)));
	20	typedef __bf16 __bfloat16 __attribute__((deprecated("use __bf16 instead")));
19	21
20	22	#define __DEFAULT_FN_ATTRS512 \
21	23	__attribute__((__always_inline__, __nodebug__, __target__("avx512bf16"), \
...	...	@@ -33,7 +35,7 @@ typedef unsigned short __bfloat16;
33	35	/// A bfloat data.
34	36	/// \returns A float data whose sign field and exponent field keep unchanged,
35	37	/// and fraction field is extended to 23 bits.
36		static __inline__ float __DEFAULT_FN_ATTRS _mm_cvtsbh_ss(__bfloat16 __A) {
	38	static __inline__ float __DEFAULT_FN_ATTRS _mm_cvtsbh_ss(__bf16 __A) {
37	39	return __builtin_ia32_cvtsbf162ss_32(__A);
38	40	}
39	41
...	...	@@ -74,9 +76,9 @@ _mm512_cvtne2ps_pbh(__m512 __A, __m512 __B) {
74	76	/// conversion of __B, and higher 256 bits come from conversion of __A.
75	77	static __inline__ __m512bh __DEFAULT_FN_ATTRS512
76	78	_mm512_mask_cvtne2ps_pbh(__m512bh __W, __mmask32 __U, __m512 __A, __m512 __B) {
77		return (__m512bh)__builtin_ia32_selectw_512((__mmask32)__U,
78		(__v32hi)_mm512_cvtne2ps_pbh(__A, __B),
79		(__v32hi)__W);
	79	return (__m512bh)__builtin_ia32_selectpbf_512((__mmask32)__U,
	80	(__v32bf)_mm512_cvtne2ps_pbh(__A, __B),
	81	(__v32bf)__W);
80	82	}
81	83
82	84	/// Convert Two Packed Single Data to One Packed BF16 Data.
...	...	@@ -96,9 +98,9 @@ _mm512_mask_cvtne2ps_pbh(__m512bh __W, __mmask32 __U, __m512 __A, __m512 __B) {
96	98	/// conversion of __B, and higher 256 bits come from conversion of __A.
97	99	static __inline__ __m512bh __DEFAULT_FN_ATTRS512
98	100	_mm512_maskz_cvtne2ps_pbh(__mmask32 __U, __m512 __A, __m512 __B) {
99		return (__m512bh)__builtin_ia32_selectw_512((__mmask32)__U,
100		(__v32hi)_mm512_cvtne2ps_pbh(__A, __B),
101		(__v32hi)_mm512_setzero_si512());
	101	return (__m512bh)__builtin_ia32_selectpbf_512((__mmask32)__U,
	102	(__v32bf)_mm512_cvtne2ps_pbh(__A, __B),
	103	(__v32bf)_mm512_setzero_si512());
102	104	}
103	105
104	106	/// Convert Packed Single Data to Packed BF16 Data.
...	...	@@ -113,7 +115,7 @@ _mm512_maskz_cvtne2ps_pbh(__mmask32 __U, __m512 __A, __m512 __B) {
113	115	static __inline__ __m256bh __DEFAULT_FN_ATTRS512
114	116	_mm512_cvtneps_pbh(__m512 __A) {
115	117	return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
116		(__v16hi)_mm256_undefined_si256(),
	118	(__v16bf)_mm256_undefined_si256(),
117	119	(__mmask16)-1);
118	120	}
119	121
...	...	@@ -134,7 +136,7 @@ _mm512_cvtneps_pbh(__m512 __A) {
134	136	static __inline__ __m256bh __DEFAULT_FN_ATTRS512
135	137	_mm512_mask_cvtneps_pbh(__m256bh __W, __mmask16 __U, __m512 __A) {
136	138	return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
137		(__v16hi)__W,
	139	(__v16bf)__W,
138	140	(__mmask16)__U);
139	141	}
140	142
...	...	@@ -153,7 +155,7 @@ _mm512_mask_cvtneps_pbh(__m256bh __W, __mmask16 __U, __m512 __A) {
153	155	static __inline__ __m256bh __DEFAULT_FN_ATTRS512
154	156	_mm512_maskz_cvtneps_pbh(__mmask16 __U, __m512 __A) {
155	157	return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
156		(__v16hi)_mm256_setzero_si256(),
	158	(__v16bf)_mm256_setzero_si256(),
157	159	(__mmask16)__U);
158	160	}
159	161
...	...	@@ -174,8 +176,8 @@ _mm512_maskz_cvtneps_pbh(__mmask16 __U, __m512 __A) {
174	176	static __inline__ __m512 __DEFAULT_FN_ATTRS512
175	177	_mm512_dpbf16_ps(__m512 __D, __m512bh __A, __m512bh __B) {
176	178	return (__m512)__builtin_ia32_dpbf16ps_512((__v16sf) __D,
177		(__v16si) __A,
178		(__v16si) __B);
	179	(__v32bf) __A,
	180	(__v32bf) __B);
179	181	}
180	182
181	183	/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
...	...	@@ -277,3 +279,4 @@ _mm512_mask_cvtpbh_ps(__m512 __S, __mmask16 __U, __m256bh __A) {
277	279	#undef __DEFAULT_FN_ATTRS512
278	280
279	281	#endif
	282	#endif

lib/include/avx512fintrin.h+2-2

...	...	@@ -256,8 +256,8 @@ _mm512_maskz_broadcastq_epi64 (__mmask8 __M, __m128i __A)
256	256	static __inline __m512 __DEFAULT_FN_ATTRS512
257	257	_mm512_setzero_ps(void)
258	258	{
259		return __extension__ (__m512){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
260		0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
	259	return __extension__ (__m512){ 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
	260	0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f };
261	261	}
262	262
263	263	#define _mm512_setzero _mm512_setzero_ps

lib/include/avx512fp16intrin.h+6-9

...	...	@@ -10,6 +10,8 @@
10	10	#error "Never use <avx512fp16intrin.h> directly; include <immintrin.h> instead."
11	11	#endif
12	12
	13	#ifdef __SSE2__
	14
13	15	#ifndef __AVX512FP16INTRIN_H
14	16	#define __AVX512FP16INTRIN_H
15	17
...	...	@@ -17,12 +19,6 @@
17	19	typedef _Float16 __v32hf __attribute__((__vector_size__(64), __aligned__(64)));
18	20	typedef _Float16 __m512h __attribute__((__vector_size__(64), __aligned__(64)));
19	21	typedef _Float16 __m512h_u __attribute__((__vector_size__(64), __aligned__(1)));
20		typedef _Float16 __v8hf __attribute__((__vector_size__(16), __aligned__(16)));
21		typedef _Float16 __m128h __attribute__((__vector_size__(16), __aligned__(16)));
22		typedef _Float16 __m128h_u __attribute__((__vector_size__(16), __aligned__(1)));
23		typedef _Float16 __v16hf __attribute__((__vector_size__(32), __aligned__(32)));
24		typedef _Float16 __m256h __attribute__((__vector_size__(32), __aligned__(32)));
25		typedef _Float16 __m256h_u __attribute__((__vector_size__(32), __aligned__(1)));
26	22
27	23	/* Define the default attributes for the functions in this file. */
28	24	#define __DEFAULT_FN_ATTRS512 \
...	...	@@ -829,7 +825,7 @@ static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_load_sh(void const *__dp) {
829	825	struct __mm_load_sh_struct {
830	826	_Float16 __u;
831	827	} __attribute__((__packed__, __may_alias__));
832		_Float16 __u = ((struct __mm_load_sh_struct *)__dp)->__u;
	828	_Float16 __u = ((const struct __mm_load_sh_struct *)__dp)->__u;
833	829	return (__m128h){__u, 0, 0, 0, 0, 0, 0, 0};
834	830	}
835	831
...	...	@@ -838,13 +834,13 @@ _mm_mask_load_sh(__m128h __W, __mmask8 __U, const void *__A) {
838	834	__m128h src = (__v8hf)__builtin_shufflevector(
839	835	(__v8hf)__W, (__v8hf)_mm_setzero_ph(), 0, 8, 8, 8, 8, 8, 8, 8);
840	836
841		return (__m128h)__builtin_ia32_loadsh128_mask((__v8hf *)__A, src, __U & 1);
	837	return (__m128h)__builtin_ia32_loadsh128_mask((const __v8hf *)__A, src, __U & 1);
842	838	}
843	839
844	840	static __inline__ __m128h __DEFAULT_FN_ATTRS128
845	841	_mm_maskz_load_sh(__mmask8 __U, const void *__A) {
846	842	return (__m128h)__builtin_ia32_loadsh128_mask(
847		(__v8hf *)__A, (__v8hf)_mm_setzero_ph(), __U & 1);
	843	(const __v8hf *)__A, (__v8hf)_mm_setzero_ph(), __U & 1);
848	844	}
849	845
850	846	static __inline__ __m512h __DEFAULT_FN_ATTRS512
...	...	@@ -3347,3 +3343,4 @@ _mm512_permutexvar_ph(__m512i __A, __m512h __B) {
3347	3343	#undef __DEFAULT_FN_ATTRS512
3348	3344
3349	3345	#endif
	3346	#endif

lib/include/avx512ifmavlintrin.h+13-27

...	...	@@ -18,14 +18,21 @@
18	18	#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl"), __min_vector_width__(128)))
19	19	#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl"), __min_vector_width__(256)))
20	20
	21	#define _mm_madd52hi_epu64(X, Y, Z) \
	22	((__m128i)__builtin_ia32_vpmadd52huq128((__v2di)(X), (__v2di)(Y), \
	23	(__v2di)(Z)))
21	24
	25	#define _mm256_madd52hi_epu64(X, Y, Z) \
	26	((__m256i)__builtin_ia32_vpmadd52huq256((__v4di)(X), (__v4di)(Y), \
	27	(__v4di)(Z)))
22	28
23		static __inline__ __m128i __DEFAULT_FN_ATTRS128
24		_mm_madd52hi_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
25		{
26		return (__m128i)__builtin_ia32_vpmadd52huq128((__v2di) __X, (__v2di) __Y,
27		(__v2di) __Z);
28		}
	29	#define _mm_madd52lo_epu64(X, Y, Z) \
	30	((__m128i)__builtin_ia32_vpmadd52luq128((__v2di)(X), (__v2di)(Y), \
	31	(__v2di)(Z)))
	32
	33	#define _mm256_madd52lo_epu64(X, Y, Z) \
	34	((__m256i)__builtin_ia32_vpmadd52luq256((__v4di)(X), (__v4di)(Y), \
	35	(__v4di)(Z)))
29	36
30	37	static __inline__ __m128i __DEFAULT_FN_ATTRS128
31	38	_mm_mask_madd52hi_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
...	...	@@ -43,13 +50,6 @@ _mm_maskz_madd52hi_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
43	50	(__v2di)_mm_setzero_si128());
44	51	}
45	52
46		static __inline__ __m256i __DEFAULT_FN_ATTRS256
47		_mm256_madd52hi_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
48		{
49		return (__m256i)__builtin_ia32_vpmadd52huq256((__v4di)__X, (__v4di)__Y,
50		(__v4di)__Z);
51		}
52
53	53	static __inline__ __m256i __DEFAULT_FN_ATTRS256
54	54	_mm256_mask_madd52hi_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
55	55	{
...	...	@@ -66,13 +66,6 @@ _mm256_maskz_madd52hi_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z
66	66	(__v4di)_mm256_setzero_si256());
67	67	}
68	68
69		static __inline__ __m128i __DEFAULT_FN_ATTRS128
70		_mm_madd52lo_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
71		{
72		return (__m128i)__builtin_ia32_vpmadd52luq128((__v2di)__X, (__v2di)__Y,
73		(__v2di)__Z);
74		}
75
76	69	static __inline__ __m128i __DEFAULT_FN_ATTRS128
77	70	_mm_mask_madd52lo_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
78	71	{
...	...	@@ -89,13 +82,6 @@ _mm_maskz_madd52lo_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
89	82	(__v2di)_mm_setzero_si128());
90	83	}
91	84
92		static __inline__ __m256i __DEFAULT_FN_ATTRS256
93		_mm256_madd52lo_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
94		{
95		return (__m256i)__builtin_ia32_vpmadd52luq256((__v4di)__X, (__v4di)__Y,
96		(__v4di)__Z);
97		}
98
99	85	static __inline__ __m256i __DEFAULT_FN_ATTRS256
100	86	_mm256_mask_madd52lo_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
101	87	{

lib/include/avx512vlbf16intrin.h+31-38

...	...	@@ -10,11 +10,11 @@
10	10	#error "Never use <avx512vlbf16intrin.h> directly; include <immintrin.h> instead."
11	11	#endif
12	12
	13	#ifdef __SSE2__
	14
13	15	#ifndef __AVX512VLBF16INTRIN_H
14	16	#define __AVX512VLBF16INTRIN_H
15	17
16		typedef short __m128bh __attribute__((__vector_size__(16), __aligned__(16)));
17
18	18	#define __DEFAULT_FN_ATTRS128 \
19	19	__attribute__((__always_inline__, __nodebug__, \
20	20	__target__("avx512vl, avx512bf16"), __min_vector_width__(128)))
...	...	@@ -59,9 +59,9 @@ _mm_cvtne2ps_pbh(__m128 __A, __m128 __B) {
59	59	/// conversion of __B, and higher 64 bits come from conversion of __A.
60	60	static __inline__ __m128bh __DEFAULT_FN_ATTRS128
61	61	_mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
62		return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
63		(__v8hi)_mm_cvtne2ps_pbh(__A, __B),
64		(__v8hi)__W);
	62	return (__m128bh)__builtin_ia32_selectpbf_128((__mmask8)__U,
	63	(__v8bf)_mm_cvtne2ps_pbh(__A, __B),
	64	(__v8bf)__W);
65	65	}
66	66
67	67	/// Convert Two Packed Single Data to One Packed BF16 Data.
...	...	@@ -81,9 +81,9 @@ _mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
81	81	/// conversion of __B, and higher 64 bits come from conversion of __A.
82	82	static __inline__ __m128bh __DEFAULT_FN_ATTRS128
83	83	_mm_maskz_cvtne2ps_pbh(__mmask8 __U, __m128 __A, __m128 __B) {
84		return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
85		(__v8hi)_mm_cvtne2ps_pbh(__A, __B),
86		(__v8hi)_mm_setzero_si128());
	84	return (__m128bh)__builtin_ia32_selectpbf_128((__mmask8)__U,
	85	(__v8bf)_mm_cvtne2ps_pbh(__A, __B),
	86	(__v8bf)_mm_setzero_si128());
87	87	}
88	88
89	89	/// Convert Two Packed Single Data to One Packed BF16 Data.
...	...	@@ -123,9 +123,9 @@ _mm256_cvtne2ps_pbh(__m256 __A, __m256 __B) {
123	123	/// conversion of __B, and higher 128 bits come from conversion of __A.
124	124	static __inline__ __m256bh __DEFAULT_FN_ATTRS256
125	125	_mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
126		return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
127		(__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
128		(__v16hi)__W);
	126	return (__m256bh)__builtin_ia32_selectpbf_256((__mmask16)__U,
	127	(__v16bf)_mm256_cvtne2ps_pbh(__A, __B),
	128	(__v16bf)__W);
129	129	}
130	130
131	131	/// Convert Two Packed Single Data to One Packed BF16 Data.
...	...	@@ -145,9 +145,9 @@ _mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
145	145	/// conversion of __B, and higher 128 bits come from conversion of __A.
146	146	static __inline__ __m256bh __DEFAULT_FN_ATTRS256
147	147	_mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
148		return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
149		(__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
150		(__v16hi)_mm256_setzero_si256());
	148	return (__m256bh)__builtin_ia32_selectpbf_256((__mmask16)__U,
	149	(__v16bf)_mm256_cvtne2ps_pbh(__A, __B),
	150	(__v16bf)_mm256_setzero_si256());
151	151	}
152	152
153	153	/// Convert Packed Single Data to Packed BF16 Data.
...	...	@@ -160,12 +160,8 @@ _mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
160	160	/// A 128-bit vector of [4 x float].
161	161	/// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
162	162	/// conversion of __A, and higher 64 bits are 0.
163		static __inline__ __m128bh __DEFAULT_FN_ATTRS128
164		_mm_cvtneps_pbh(__m128 __A) {
165		return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
166		(__v8hi)_mm_undefined_si128(),
167		(__mmask8)-1);
168		}
	163	#define _mm_cvtneps_pbh(A) \
	164	((__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)(A)))
169	165
170	166	/// Convert Packed Single Data to Packed BF16 Data.
171	167	///
...	...	@@ -185,7 +181,7 @@ _mm_cvtneps_pbh(__m128 __A) {
185	181	static __inline__ __m128bh __DEFAULT_FN_ATTRS128
186	182	_mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
187	183	return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
188		(__v8hi)__W,
	184	(__v8bf)__W,
189	185	(__mmask8)__U);
190	186	}
191	187
...	...	@@ -205,7 +201,7 @@ _mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
205	201	static __inline__ __m128bh __DEFAULT_FN_ATTRS128
206	202	_mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
207	203	return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
208		(__v8hi)_mm_setzero_si128(),
	204	(__v8bf)_mm_setzero_si128(),
209	205	(__mmask8)__U);
210	206	}
211	207
...	...	@@ -218,12 +214,8 @@ _mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
218	214	/// \param __A
219	215	/// A 256-bit vector of [8 x float].
220	216	/// \returns A 128-bit vector of [8 x bfloat] comes from conversion of __A.
221		static __inline__ __m128bh __DEFAULT_FN_ATTRS256
222		_mm256_cvtneps_pbh(__m256 __A) {
223		return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
224		(__v8hi)_mm_undefined_si128(),
225		(__mmask8)-1);
226		}
	217	#define _mm256_cvtneps_pbh(A) \
	218	((__m128bh)__builtin_ia32_vcvtneps2bf16256((__v8sf)(A)))
227	219
228	220	/// Convert Packed Single Data to Packed BF16 Data.
229	221	///
...	...	@@ -242,7 +234,7 @@ _mm256_cvtneps_pbh(__m256 __A) {
242	234	static __inline__ __m128bh __DEFAULT_FN_ATTRS256
243	235	_mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
244	236	return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
245		(__v8hi)__W,
	237	(__v8bf)__W,
246	238	(__mmask8)__U);
247	239	}
248	240
...	...	@@ -261,7 +253,7 @@ _mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
261	253	static __inline__ __m128bh __DEFAULT_FN_ATTRS256
262	254	_mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
263	255	return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
264		(__v8hi)_mm_setzero_si128(),
	256	(__v8bf)_mm_setzero_si128(),
265	257	(__mmask8)__U);
266	258	}
267	259
...	...	@@ -282,8 +274,8 @@ _mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
282	274	static __inline__ __m128 __DEFAULT_FN_ATTRS128
283	275	_mm_dpbf16_ps(__m128 __D, __m128bh __A, __m128bh __B) {
284	276	return (__m128)__builtin_ia32_dpbf16ps_128((__v4sf)__D,
285		(__v4si)__A,
286		(__v4si)__B);
	277	(__v8bf)__A,
	278	(__v8bf)__B);
287	279	}
288	280
289	281	/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
...	...	@@ -351,8 +343,8 @@ _mm_maskz_dpbf16_ps(__mmask8 __U, __m128 __D, __m128bh __A, __m128bh __B) {
351	343	static __inline__ __m256 __DEFAULT_FN_ATTRS256
352	344	_mm256_dpbf16_ps(__m256 __D, __m256bh __A, __m256bh __B) {
353	345	return (__m256)__builtin_ia32_dpbf16ps_256((__v8sf)__D,
354		(__v8si)__A,
355		(__v8si)__B);
	346	(__v16bf)__A,
	347	(__v16bf)__B);
356	348	}
357	349
358	350	/// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
...	...	@@ -413,11 +405,11 @@ _mm256_maskz_dpbf16_ps(__mmask8 __U, __m256 __D, __m256bh __A, __m256bh __B) {
413	405	/// A float data.
414	406	/// \returns A bf16 data whose sign field and exponent field keep unchanged,
415	407	/// and fraction field is truncated to 7 bits.
416		static __inline__ __bfloat16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
	408	static __inline__ __bf16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
417	409	__v4sf __V = {__A, 0, 0, 0};
418		__v8hi __R = __builtin_ia32_cvtneps2bf16_128_mask(
419		(__v4sf)__V, (__v8hi)_mm_undefined_si128(), (__mmask8)-1);
420		return (__bfloat16)__R[0];
	410	__v8bf __R = __builtin_ia32_cvtneps2bf16_128_mask(
	411	(__v4sf)__V, (__v8bf)_mm_undefined_si128(), (__mmask8)-1);
	412	return (__bf16)__R[0];
421	413	}
422	414
423	415	/// Convert Packed BF16 Data to Packed float Data.
...	...	@@ -520,3 +512,4 @@ _mm256_mask_cvtpbh_ps(__m256 __S, __mmask8 __U, __m128bh __A) {
520	512	#undef __DEFAULT_FN_ATTRS256
521	513
522	514	#endif
	515	#endif

lib/include/avx512vlbwintrin.h+352

...	...	@@ -2803,6 +2803,358 @@ _mm256_mask_permutexvar_epi16 (__m256i __W, __mmask16 __M, __m256i __A,
2803	2803	(__v16hi)_mm256_dbsad_epu8((A), (B), (imm)), \
2804	2804	(__v16hi)_mm256_setzero_si256()))
2805	2805
	2806	static __inline__ short __DEFAULT_FN_ATTRS128
	2807	_mm_reduce_add_epi16(__m128i __W) {
	2808	return __builtin_reduce_add((__v8hi)__W);
	2809	}
	2810
	2811	static __inline__ short __DEFAULT_FN_ATTRS128
	2812	_mm_reduce_mul_epi16(__m128i __W) {
	2813	return __builtin_reduce_mul((__v8hi)__W);
	2814	}
	2815
	2816	static __inline__ short __DEFAULT_FN_ATTRS128
	2817	_mm_reduce_and_epi16(__m128i __W) {
	2818	return __builtin_reduce_and((__v8hi)__W);
	2819	}
	2820
	2821	static __inline__ short __DEFAULT_FN_ATTRS128
	2822	_mm_reduce_or_epi16(__m128i __W) {
	2823	return __builtin_reduce_or((__v8hi)__W);
	2824	}
	2825
	2826	static __inline__ short __DEFAULT_FN_ATTRS128
	2827	_mm_mask_reduce_add_epi16( __mmask8 __M, __m128i __W) {
	2828	__W = _mm_maskz_mov_epi16(__M, __W);
	2829	return __builtin_reduce_add((__v8hi)__W);
	2830	}
	2831
	2832	static __inline__ short __DEFAULT_FN_ATTRS128
	2833	_mm_mask_reduce_mul_epi16( __mmask8 __M, __m128i __W) {
	2834	__W = _mm_mask_mov_epi16(_mm_set1_epi16(1), __M, __W);
	2835	return __builtin_reduce_mul((__v8hi)__W);
	2836	}
	2837
	2838	static __inline__ short __DEFAULT_FN_ATTRS128
	2839	_mm_mask_reduce_and_epi16( __mmask8 __M, __m128i __W) {
	2840	__W = _mm_mask_mov_epi16(_mm_set1_epi16(-1), __M, __W);
	2841	return __builtin_reduce_and((__v8hi)__W);
	2842	}
	2843
	2844	static __inline__ short __DEFAULT_FN_ATTRS128
	2845	_mm_mask_reduce_or_epi16(__mmask8 __M, __m128i __W) {
	2846	__W = _mm_maskz_mov_epi16(__M, __W);
	2847	return __builtin_reduce_or((__v8hi)__W);
	2848	}
	2849
	2850	static __inline__ short __DEFAULT_FN_ATTRS128
	2851	_mm_reduce_max_epi16(__m128i __V) {
	2852	return __builtin_reduce_max((__v8hi)__V);
	2853	}
	2854
	2855	static __inline__ unsigned short __DEFAULT_FN_ATTRS128
	2856	_mm_reduce_max_epu16(__m128i __V) {
	2857	return __builtin_reduce_max((__v8hu)__V);
	2858	}
	2859
	2860	static __inline__ short __DEFAULT_FN_ATTRS128
	2861	_mm_reduce_min_epi16(__m128i __V) {
	2862	return __builtin_reduce_min((__v8hi)__V);
	2863	}
	2864
	2865	static __inline__ unsigned short __DEFAULT_FN_ATTRS128
	2866	_mm_reduce_min_epu16(__m128i __V) {
	2867	return __builtin_reduce_min((__v8hu)__V);
	2868	}
	2869
	2870	static __inline__ short __DEFAULT_FN_ATTRS128
	2871	_mm_mask_reduce_max_epi16(__mmask16 __M, __m128i __V) {
	2872	__V = _mm_mask_mov_epi16(_mm_set1_epi16(-32767-1), __M, __V);
	2873	return __builtin_reduce_max((__v8hi)__V);
	2874	}
	2875
	2876	static __inline__ unsigned short __DEFAULT_FN_ATTRS128
	2877	_mm_mask_reduce_max_epu16(__mmask16 __M, __m128i __V) {
	2878	__V = _mm_maskz_mov_epi16(__M, __V);
	2879	return __builtin_reduce_max((__v8hu)__V);
	2880	}
	2881
	2882	static __inline__ short __DEFAULT_FN_ATTRS128
	2883	_mm_mask_reduce_min_epi16(__mmask16 __M, __m128i __V) {
	2884	__V = _mm_mask_mov_epi16(_mm_set1_epi16(32767), __M, __V);
	2885	return __builtin_reduce_min((__v8hi)__V);
	2886	}
	2887
	2888	static __inline__ unsigned short __DEFAULT_FN_ATTRS128
	2889	_mm_mask_reduce_min_epu16(__mmask16 __M, __m128i __V) {
	2890	__V = _mm_mask_mov_epi16(_mm_set1_epi16(-1), __M, __V);
	2891	return __builtin_reduce_min((__v8hu)__V);
	2892	}
	2893
	2894	static __inline__ short __DEFAULT_FN_ATTRS256
	2895	_mm256_reduce_add_epi16(__m256i __W) {
	2896	return __builtin_reduce_add((__v16hi)__W);
	2897	}
	2898
	2899	static __inline__ short __DEFAULT_FN_ATTRS256
	2900	_mm256_reduce_mul_epi16(__m256i __W) {
	2901	return __builtin_reduce_mul((__v16hi)__W);
	2902	}
	2903
	2904	static __inline__ short __DEFAULT_FN_ATTRS256
	2905	_mm256_reduce_and_epi16(__m256i __W) {
	2906	return __builtin_reduce_and((__v16hi)__W);
	2907	}
	2908
	2909	static __inline__ short __DEFAULT_FN_ATTRS256
	2910	_mm256_reduce_or_epi16(__m256i __W) {
	2911	return __builtin_reduce_or((__v16hi)__W);
	2912	}
	2913
	2914	static __inline__ short __DEFAULT_FN_ATTRS256
	2915	_mm256_mask_reduce_add_epi16( __mmask16 __M, __m256i __W) {
	2916	__W = _mm256_maskz_mov_epi16(__M, __W);
	2917	return __builtin_reduce_add((__v16hi)__W);
	2918	}
	2919
	2920	static __inline__ short __DEFAULT_FN_ATTRS256
	2921	_mm256_mask_reduce_mul_epi16( __mmask16 __M, __m256i __W) {
	2922	__W = _mm256_mask_mov_epi16(_mm256_set1_epi16(1), __M, __W);
	2923	return __builtin_reduce_mul((__v16hi)__W);
	2924	}
	2925
	2926	static __inline__ short __DEFAULT_FN_ATTRS256
	2927	_mm256_mask_reduce_and_epi16( __mmask16 __M, __m256i __W) {
	2928	__W = _mm256_mask_mov_epi16(_mm256_set1_epi16(-1), __M, __W);
	2929	return __builtin_reduce_and((__v16hi)__W);
	2930	}
	2931
	2932	static __inline__ short __DEFAULT_FN_ATTRS256
	2933	_mm256_mask_reduce_or_epi16(__mmask16 __M, __m256i __W) {
	2934	__W = _mm256_maskz_mov_epi16(__M, __W);
	2935	return __builtin_reduce_or((__v16hi)__W);
	2936	}
	2937
	2938	static __inline__ short __DEFAULT_FN_ATTRS256
	2939	_mm256_reduce_max_epi16(__m256i __V) {
	2940	return __builtin_reduce_max((__v16hi)__V);
	2941	}
	2942
	2943	static __inline__ unsigned short __DEFAULT_FN_ATTRS256
	2944	_mm256_reduce_max_epu16(__m256i __V) {
	2945	return __builtin_reduce_max((__v16hu)__V);
	2946	}
	2947
	2948	static __inline__ short __DEFAULT_FN_ATTRS256
	2949	_mm256_reduce_min_epi16(__m256i __V) {
	2950	return __builtin_reduce_min((__v16hi)__V);
	2951	}
	2952
	2953	static __inline__ unsigned short __DEFAULT_FN_ATTRS256
	2954	_mm256_reduce_min_epu16(__m256i __V) {
	2955	return __builtin_reduce_min((__v16hu)__V);
	2956	}
	2957
	2958	static __inline__ short __DEFAULT_FN_ATTRS256
	2959	_mm256_mask_reduce_max_epi16(__mmask16 __M, __m256i __V) {
	2960	__V = _mm256_mask_mov_epi16(_mm256_set1_epi16(-32767-1), __M, __V);
	2961	return __builtin_reduce_max((__v16hi)__V);
	2962	}
	2963
	2964	static __inline__ unsigned short __DEFAULT_FN_ATTRS256
	2965	_mm256_mask_reduce_max_epu16(__mmask16 __M, __m256i __V) {
	2966	__V = _mm256_maskz_mov_epi16(__M, __V);
	2967	return __builtin_reduce_max((__v16hu)__V);
	2968	}
	2969
	2970	static __inline__ short __DEFAULT_FN_ATTRS256
	2971	_mm256_mask_reduce_min_epi16(__mmask16 __M, __m256i __V) {
	2972	__V = _mm256_mask_mov_epi16(_mm256_set1_epi16(32767), __M, __V);
	2973	return __builtin_reduce_min((__v16hi)__V);
	2974	}
	2975
	2976	static __inline__ unsigned short __DEFAULT_FN_ATTRS256
	2977	_mm256_mask_reduce_min_epu16(__mmask16 __M, __m256i __V) {
	2978	__V = _mm256_mask_mov_epi16(_mm256_set1_epi16(-1), __M, __V);
	2979	return __builtin_reduce_min((__v16hu)__V);
	2980	}
	2981
	2982	static __inline__ signed char __DEFAULT_FN_ATTRS128
	2983	_mm_reduce_add_epi8(__m128i __W) {
	2984	return __builtin_reduce_add((__v16qs)__W);
	2985	}
	2986
	2987	static __inline__ signed char __DEFAULT_FN_ATTRS128
	2988	_mm_reduce_mul_epi8(__m128i __W) {
	2989	return __builtin_reduce_mul((__v16qs)__W);
	2990	}
	2991
	2992	static __inline__ signed char __DEFAULT_FN_ATTRS128
	2993	_mm_reduce_and_epi8(__m128i __W) {
	2994	return __builtin_reduce_and((__v16qs)__W);
	2995	}
	2996
	2997	static __inline__ signed char __DEFAULT_FN_ATTRS128
	2998	_mm_reduce_or_epi8(__m128i __W) {
	2999	return __builtin_reduce_or((__v16qs)__W);
	3000	}
	3001
	3002	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3003	_mm_mask_reduce_add_epi8(__mmask16 __M, __m128i __W) {
	3004	__W = _mm_maskz_mov_epi8(__M, __W);
	3005	return __builtin_reduce_add((__v16qs)__W);
	3006	}
	3007
	3008	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3009	_mm_mask_reduce_mul_epi8(__mmask16 __M, __m128i __W) {
	3010	__W = _mm_mask_mov_epi8(_mm_set1_epi8(1), __M, __W);
	3011	return __builtin_reduce_mul((__v16qs)__W);
	3012	}
	3013
	3014	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3015	_mm_mask_reduce_and_epi8(__mmask16 __M, __m128i __W) {
	3016	__W = _mm_mask_mov_epi8(_mm_set1_epi8(-1), __M, __W);
	3017	return __builtin_reduce_and((__v16qs)__W);
	3018	}
	3019
	3020	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3021	_mm_mask_reduce_or_epi8(__mmask16 __M, __m128i __W) {
	3022	__W = _mm_maskz_mov_epi8(__M, __W);
	3023	return __builtin_reduce_or((__v16qs)__W);
	3024	}
	3025
	3026	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3027	_mm_reduce_max_epi8(__m128i __V) {
	3028	return __builtin_reduce_max((__v16qs)__V);
	3029	}
	3030
	3031	static __inline__ unsigned char __DEFAULT_FN_ATTRS128
	3032	_mm_reduce_max_epu8(__m128i __V) {
	3033	return __builtin_reduce_max((__v16qu)__V);
	3034	}
	3035
	3036	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3037	_mm_reduce_min_epi8(__m128i __V) {
	3038	return __builtin_reduce_min((__v16qs)__V);
	3039	}
	3040
	3041	static __inline__ unsigned char __DEFAULT_FN_ATTRS128
	3042	_mm_reduce_min_epu8(__m128i __V) {
	3043	return __builtin_reduce_min((__v16qu)__V);
	3044	}
	3045
	3046	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3047	_mm_mask_reduce_max_epi8(__mmask16 __M, __m128i __V) {
	3048	__V = _mm_mask_mov_epi8(_mm_set1_epi8(-127-1), __M, __V);
	3049	return __builtin_reduce_max((__v16qs)__V);
	3050	}
	3051
	3052	static __inline__ unsigned char __DEFAULT_FN_ATTRS128
	3053	_mm_mask_reduce_max_epu8(__mmask16 __M, __m128i __V) {
	3054	__V = _mm_maskz_mov_epi8(__M, __V);
	3055	return __builtin_reduce_max((__v16qu)__V);
	3056	}
	3057
	3058	static __inline__ signed char __DEFAULT_FN_ATTRS128
	3059	_mm_mask_reduce_min_epi8(__mmask16 __M, __m128i __V) {
	3060	__V = _mm_mask_mov_epi8(_mm_set1_epi8(127), __M, __V);
	3061	return __builtin_reduce_min((__v16qs)__V);
	3062	}
	3063
	3064	static __inline__ unsigned char __DEFAULT_FN_ATTRS128
	3065	_mm_mask_reduce_min_epu8(__mmask16 __M, __m128i __V) {
	3066	__V = _mm_mask_mov_epi8(_mm_set1_epi8(-1), __M, __V);
	3067	return __builtin_reduce_min((__v16qu)__V);
	3068	}
	3069
	3070	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3071	_mm256_reduce_add_epi8(__m256i __W) {
	3072	return __builtin_reduce_add((__v32qs)__W);
	3073	}
	3074
	3075	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3076	_mm256_reduce_mul_epi8(__m256i __W) {
	3077	return __builtin_reduce_mul((__v32qs)__W);
	3078	}
	3079
	3080	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3081	_mm256_reduce_and_epi8(__m256i __W) {
	3082	return __builtin_reduce_and((__v32qs)__W);
	3083	}
	3084
	3085	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3086	_mm256_reduce_or_epi8(__m256i __W) {
	3087	return __builtin_reduce_or((__v32qs)__W);
	3088	}
	3089
	3090	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3091	_mm256_mask_reduce_add_epi8(__mmask32 __M, __m256i __W) {
	3092	__W = _mm256_maskz_mov_epi8(__M, __W);
	3093	return __builtin_reduce_add((__v32qs)__W);
	3094	}
	3095
	3096	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3097	_mm256_mask_reduce_mul_epi8(__mmask32 __M, __m256i __W) {
	3098	__W = _mm256_mask_mov_epi8(_mm256_set1_epi8(1), __M, __W);
	3099	return __builtin_reduce_mul((__v32qs)__W);
	3100	}
	3101
	3102	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3103	_mm256_mask_reduce_and_epi8(__mmask32 __M, __m256i __W) {
	3104	__W = _mm256_mask_mov_epi8(_mm256_set1_epi8(-1), __M, __W);
	3105	return __builtin_reduce_and((__v32qs)__W);
	3106	}
	3107
	3108	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3109	_mm256_mask_reduce_or_epi8(__mmask32 __M, __m256i __W) {
	3110	__W = _mm256_maskz_mov_epi8(__M, __W);
	3111	return __builtin_reduce_or((__v32qs)__W);
	3112	}
	3113
	3114	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3115	_mm256_reduce_max_epi8(__m256i __V) {
	3116	return __builtin_reduce_max((__v32qs)__V);
	3117	}
	3118
	3119	static __inline__ unsigned char __DEFAULT_FN_ATTRS256
	3120	_mm256_reduce_max_epu8(__m256i __V) {
	3121	return __builtin_reduce_max((__v32qu)__V);
	3122	}
	3123
	3124	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3125	_mm256_reduce_min_epi8(__m256i __V) {
	3126	return __builtin_reduce_min((__v32qs)__V);
	3127	}
	3128
	3129	static __inline__ unsigned char __DEFAULT_FN_ATTRS256
	3130	_mm256_reduce_min_epu8(__m256i __V) {
	3131	return __builtin_reduce_min((__v32qu)__V);
	3132	}
	3133
	3134	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3135	_mm256_mask_reduce_max_epi8(__mmask32 __M, __m256i __V) {
	3136	__V = _mm256_mask_mov_epi8(_mm256_set1_epi8(-127-1), __M, __V);
	3137	return __builtin_reduce_max((__v32qs)__V);
	3138	}
	3139
	3140	static __inline__ unsigned char __DEFAULT_FN_ATTRS256
	3141	_mm256_mask_reduce_max_epu8(__mmask32 __M, __m256i __V) {
	3142	__V = _mm256_maskz_mov_epi8(__M, __V);
	3143	return __builtin_reduce_max((__v32qu)__V);
	3144	}
	3145
	3146	static __inline__ signed char __DEFAULT_FN_ATTRS256
	3147	_mm256_mask_reduce_min_epi8(__mmask32 __M, __m256i __V) {
	3148	__V = _mm256_mask_mov_epi8(_mm256_set1_epi8(127), __M, __V);
	3149	return __builtin_reduce_min((__v32qs)__V);
	3150	}
	3151
	3152	static __inline__ unsigned char __DEFAULT_FN_ATTRS256
	3153	_mm256_mask_reduce_min_epu8(__mmask32 __M, __m256i __V) {
	3154	__V = _mm256_mask_mov_epi8(_mm256_set1_epi8(-1), __M, __V);
	3155	return __builtin_reduce_min((__v32qu)__V);
	3156	}
	3157
2806	3158	#undef __DEFAULT_FN_ATTRS128
2807	3159	#undef __DEFAULT_FN_ATTRS256
2808	3160

lib/include/avx512vlfp16intrin.h+3

...	...	@@ -11,6 +11,8 @@
11	11	"Never use <avx512vlfp16intrin.h> directly; include <immintrin.h> instead."
12	12	#endif
13	13
	14	#ifdef __SSE2__
	15
14	16	#ifndef __AVX512VLFP16INTRIN_H
15	17	#define __AVX512VLFP16INTRIN_H
16	18
...	...	@@ -2066,3 +2068,4 @@ _mm_reduce_min_ph(__m128h __V) {
2066	2068	#undef __DEFAULT_FN_ATTRS256
2067	2069
2068	2070	#endif
	2071	#endif

lib/include/avxifmaintrin.h created+177

...	...	@@ -0,0 +1,177 @@
	1	/*===----------------- avxifmaintrin.h - IFMA intrinsics -------------------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9
	10	#ifndef __IMMINTRIN_H
	11	#error "Never use <avxifmaintrin.h> directly; include <immintrin.h> instead."
	12	#endif
	13
	14	#ifndef __AVXIFMAINTRIN_H
	15	#define __AVXIFMAINTRIN_H
	16
	17	/* Define the default attributes for the functions in this file. */
	18	#define __DEFAULT_FN_ATTRS128 \
	19	__attribute__((__always_inline__, __nodebug__, __target__("avxifma"), \
	20	__min_vector_width__(128)))
	21	#define __DEFAULT_FN_ATTRS256 \
	22	__attribute__((__always_inline__, __nodebug__, __target__("avxifma"), \
	23	__min_vector_width__(256)))
	24
	25	// must vex-encoding
	26
	27	/// Multiply packed unsigned 52-bit integers in each 64-bit element of \a __Y
	28	/// and \a __Z to form a 104-bit intermediate result. Add the high 52-bit
	29	/// unsigned integer from the intermediate result with the corresponding
	30	/// unsigned 64-bit integer in \a __X, and store the results in \a dst.
	31	///
	32	/// \headerfile <immintrin.h>
	33	///
	34	/// \code
	35	/// __m128i
	36	/// _mm_madd52hi_avx_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
	37	/// \endcode
	38	///
	39	/// This intrinsic corresponds to the \c VPMADD52HUQ instruction.
	40	///
	41	/// \return
	42	/// return __m128i dst.
	43	/// \param __X
	44	/// A 128-bit vector of [2 x i64]
	45	/// \param __Y
	46	/// A 128-bit vector of [2 x i64]
	47	/// \param __Z
	48	/// A 128-bit vector of [2 x i64]
	49	///
	50	/// \code{.operation}
	51	/// FOR j := 0 to 1
	52	/// i := j*64
	53	/// tmp[127:0] := ZeroExtend64(__Y[i+51:i]) * ZeroExtend64(__Z[i+51:i])
	54	/// dst[i+63:i] := __X[i+63:i] + ZeroExtend64(tmp[103:52])
	55	/// ENDFOR
	56	/// dst[MAX:128] := 0
	57	/// \endcode
	58	static __inline__ __m128i __DEFAULT_FN_ATTRS128
	59	_mm_madd52hi_avx_epu64(__m128i __X, __m128i __Y, __m128i __Z) {
	60	return (__m128i)__builtin_ia32_vpmadd52huq128((__v2di)__X, (__v2di)__Y,
	61	(__v2di)__Z);
	62	}
	63
	64	/// Multiply packed unsigned 52-bit integers in each 64-bit element of \a __Y
	65	/// and \a __Z to form a 104-bit intermediate result. Add the high 52-bit
	66	/// unsigned integer from the intermediate result with the corresponding
	67	/// unsigned 64-bit integer in \a __X, and store the results in \a dst.
	68	///
	69	/// \headerfile <immintrin.h>
	70	///
	71	/// \code
	72	/// __m256i
	73	/// _mm256_madd52hi_avx_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
	74	/// \endcode
	75	///
	76	/// This intrinsic corresponds to the \c VPMADD52HUQ instruction.
	77	///
	78	/// \return
	79	/// return __m256i dst.
	80	/// \param __X
	81	/// A 256-bit vector of [4 x i64]
	82	/// \param __Y
	83	/// A 256-bit vector of [4 x i64]
	84	/// \param __Z
	85	/// A 256-bit vector of [4 x i64]
	86	///
	87	/// \code{.operation}
	88	/// FOR j := 0 to 3
	89	/// i := j*64
	90	/// tmp[127:0] := ZeroExtend64(__Y[i+51:i]) * ZeroExtend64(__Z[i+51:i])
	91	/// dst[i+63:i] := __X[i+63:i] + ZeroExtend64(tmp[103:52])
	92	/// ENDFOR
	93	/// dst[MAX:256] := 0
	94	/// \endcode
	95	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	96	_mm256_madd52hi_avx_epu64(__m256i __X, __m256i __Y, __m256i __Z) {
	97	return (__m256i)__builtin_ia32_vpmadd52huq256((__v4di)__X, (__v4di)__Y,
	98	(__v4di)__Z);
	99	}
	100
	101	/// Multiply packed unsigned 52-bit integers in each 64-bit element of \a __Y
	102	/// and \a __Z to form a 104-bit intermediate result. Add the low 52-bit
	103	/// unsigned integer from the intermediate result with the corresponding
	104	/// unsigned 64-bit integer in \a __X, and store the results in \a dst.
	105	///
	106	/// \headerfile <immintrin.h>
	107	///
	108	/// \code
	109	/// __m128i
	110	/// _mm_madd52lo_avx_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
	111	/// \endcode
	112	///
	113	/// This intrinsic corresponds to the \c VPMADD52LUQ instruction.
	114	///
	115	/// \return
	116	/// return __m128i dst.
	117	/// \param __X
	118	/// A 128-bit vector of [2 x i64]
	119	/// \param __Y
	120	/// A 128-bit vector of [2 x i64]
	121	/// \param __Z
	122	/// A 128-bit vector of [2 x i64]
	123	///
	124	/// \code{.operation}
	125	/// FOR j := 0 to 1
	126	/// i := j*64
	127	/// tmp[127:0] := ZeroExtend64(__Y[i+51:i]) * ZeroExtend64(__Z[i+51:i])
	128	/// dst[i+63:i] := __X[i+63:i] + ZeroExtend64(tmp[51:0])
	129	/// ENDFOR
	130	/// dst[MAX:128] := 0
	131	/// \endcode
	132	static __inline__ __m128i __DEFAULT_FN_ATTRS128
	133	_mm_madd52lo_avx_epu64(__m128i __X, __m128i __Y, __m128i __Z) {
	134	return (__m128i)__builtin_ia32_vpmadd52luq128((__v2di)__X, (__v2di)__Y,
	135	(__v2di)__Z);
	136	}
	137
	138	/// Multiply packed unsigned 52-bit integers in each 64-bit element of \a __Y
	139	/// and \a __Z to form a 104-bit intermediate result. Add the low 52-bit
	140	/// unsigned integer from the intermediate result with the corresponding
	141	/// unsigned 64-bit integer in \a __X, and store the results in \a dst.
	142	///
	143	/// \headerfile <immintrin.h>
	144	///
	145	/// \code
	146	/// __m256i
	147	/// _mm256_madd52lo_avx_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
	148	/// \endcode
	149	///
	150	/// This intrinsic corresponds to the \c VPMADD52LUQ instruction.
	151	///
	152	/// \return
	153	/// return __m256i dst.
	154	/// \param __X
	155	/// A 256-bit vector of [4 x i64]
	156	/// \param __Y
	157	/// A 256-bit vector of [4 x i64]
	158	/// \param __Z
	159	/// A 256-bit vector of [4 x i64]
	160	///
	161	/// \code{.operation}
	162	/// FOR j := 0 to 3
	163	/// i := j*64
	164	/// tmp[127:0] := ZeroExtend64(__Y[i+51:i]) * ZeroExtend64(__Z[i+51:i])
	165	/// dst[i+63:i] := __X[i+63:i] + ZeroExtend64(tmp[51:0])
	166	/// ENDFOR
	167	/// dst[MAX:256] := 0
	168	/// \endcode
	169	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	170	_mm256_madd52lo_avx_epu64(__m256i __X, __m256i __Y, __m256i __Z) {
	171	return (__m256i)__builtin_ia32_vpmadd52luq256((__v4di)__X, (__v4di)__Y,
	172	(__v4di)__Z);
	173	}
	174	#undef __DEFAULT_FN_ATTRS128
	175	#undef __DEFAULT_FN_ATTRS256
	176
	177	#endif // __AVXIFMAINTRIN_H

lib/include/avxintrin.h+12-2

...	...	@@ -39,6 +39,16 @@ typedef float __m256_u __attribute__ ((__vector_size__ (32), __aligned__(1)));
39	39	typedef double __m256d_u __attribute__((__vector_size__(32), __aligned__(1)));
40	40	typedef long long __m256i_u __attribute__((__vector_size__(32), __aligned__(1)));
41	41
	42	#ifdef __SSE2__
	43	/* Both _Float16 and __bf16 require SSE2 being enabled. */
	44	typedef _Float16 __v16hf __attribute__((__vector_size__(32), __aligned__(32)));
	45	typedef _Float16 __m256h __attribute__((__vector_size__(32), __aligned__(32)));
	46	typedef _Float16 __m256h_u __attribute__((__vector_size__(32), __aligned__(1)));
	47
	48	typedef __bf16 __v16bf __attribute__((__vector_size__(32), __aligned__(32)));
	49	typedef __bf16 __m256bh __attribute__((__vector_size__(32), __aligned__(32)));
	50	#endif
	51
42	52	/* Define the default attributes for the functions in this file. */
43	53	#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx"), __min_vector_width__(256)))
44	54	#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx"), __min_vector_width__(128)))
...	...	@@ -4288,7 +4298,7 @@ _mm256_set1_epi64x(long long __q)
4288	4298	static __inline __m256d __DEFAULT_FN_ATTRS
4289	4299	_mm256_setzero_pd(void)
4290	4300	{
4291		return __extension__ (__m256d){ 0, 0, 0, 0 };
	4301	return __extension__ (__m256d){ 0.0, 0.0, 0.0, 0.0 };
4292	4302	}
4293	4303
4294	4304	/// Constructs a 256-bit floating-point vector of [8 x float] with all
...	...	@@ -4302,7 +4312,7 @@ _mm256_setzero_pd(void)
4302	4312	static __inline __m256 __DEFAULT_FN_ATTRS
4303	4313	_mm256_setzero_ps(void)
4304	4314	{
4305		return __extension__ (__m256){ 0, 0, 0, 0, 0, 0, 0, 0 };
	4315	return __extension__ (__m256){ 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f };
4306	4316	}
4307	4317
4308	4318	/// Constructs a 256-bit integer vector initialized to zero.

lib/include/avxneconvertintrin.h created+484

...	...	@@ -0,0 +1,484 @@
	1	/*===-------------- avxneconvertintrin.h - AVXNECONVERT --------------------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9
	10	#ifndef __IMMINTRIN_H
	11	#error \
	12	"Never use <avxneconvertintrin.h> directly; include <immintrin.h> instead."
	13	#endif // __IMMINTRIN_H
	14
	15	#ifdef __SSE2__
	16
	17	#ifndef __AVXNECONVERTINTRIN_H
	18	#define __AVXNECONVERTINTRIN_H
	19
	20	/* Define the default attributes for the functions in this file. */
	21	#define __DEFAULT_FN_ATTRS128 \
	22	__attribute__((__always_inline__, __nodebug__, __target__("avxneconvert"), \
	23	__min_vector_width__(128)))
	24	#define __DEFAULT_FN_ATTRS256 \
	25	__attribute__((__always_inline__, __nodebug__, __target__("avxneconvert"), \
	26	__min_vector_width__(256)))
	27
	28	/// Convert scalar BF16 (16-bit) floating-point element
	29	/// stored at memory locations starting at location \a __A to a
	30	/// single-precision (32-bit) floating-point, broadcast it to packed
	31	/// single-precision (32-bit) floating-point elements, and store the results in
	32	/// \a dst.
	33	///
	34	/// \headerfile <x86intrin.h>
	35	///
	36	/// \code
	37	/// _mm_bcstnebf16_ps(const void *__A);
	38	/// \endcode
	39	///
	40	/// This intrinsic corresponds to the \c VBCSTNEBF162PS instruction.
	41	///
	42	/// \param __A
	43	/// A pointer to a 16-bit memory location. The address of the memory
	44	/// location does not have to be aligned.
	45	/// \returns
	46	/// A 128-bit vector of [4 x float].
	47	///
	48	/// \code{.operation}
	49	/// b := Convert_BF16_To_FP32(MEM[__A+15:__A])
	50	/// FOR j := 0 to 3
	51	/// m := j*32
	52	/// dst[m+31:m] := b
	53	/// ENDFOR
	54	/// dst[MAX:128] := 0
	55	/// \endcode
	56	static __inline__ __m128 __DEFAULT_FN_ATTRS128
	57	_mm_bcstnebf16_ps(const void *__A) {
	58	return (__m128)__builtin_ia32_vbcstnebf162ps128((const __bf16 *)__A);
	59	}
	60
	61	/// Convert scalar BF16 (16-bit) floating-point element
	62	/// stored at memory locations starting at location \a __A to a
	63	/// single-precision (32-bit) floating-point, broadcast it to packed
	64	/// single-precision (32-bit) floating-point elements, and store the results in
	65	/// \a dst.
	66	///
	67	/// \headerfile <x86intrin.h>
	68	///
	69	/// \code
	70	/// _mm256_bcstnebf16_ps(const void *__A);
	71	/// \endcode
	72	///
	73	/// This intrinsic corresponds to the \c VBCSTNEBF162PS instruction.
	74	///
	75	/// \param __A
	76	/// A pointer to a 16-bit memory location. The address of the memory
	77	/// location does not have to be aligned.
	78	/// \returns
	79	/// A 256-bit vector of [8 x float].
	80	///
	81	/// \code{.operation}
	82	/// b := Convert_BF16_To_FP32(MEM[__A+15:__A])
	83	/// FOR j := 0 to 7
	84	/// m := j*32
	85	/// dst[m+31:m] := b
	86	/// ENDFOR
	87	/// dst[MAX:256] := 0
	88	/// \endcode
	89	static __inline__ __m256 __DEFAULT_FN_ATTRS256
	90	_mm256_bcstnebf16_ps(const void *__A) {
	91	return (__m256)__builtin_ia32_vbcstnebf162ps256((const __bf16 *)__A);
	92	}
	93
	94	/// Convert scalar half-precision (16-bit) floating-point element
	95	/// stored at memory locations starting at location \a __A to a
	96	/// single-precision (32-bit) floating-point, broadcast it to packed
	97	/// single-precision (32-bit) floating-point elements, and store the results in
	98	/// \a dst.
	99	///
	100	/// \headerfile <x86intrin.h>
	101	///
	102	/// \code
	103	/// _mm_bcstnesh_ps(const void *__A);
	104	/// \endcode
	105	///
	106	/// This intrinsic corresponds to the \c VBCSTNESH2PS instruction.
	107	///
	108	/// \param __A
	109	/// A pointer to a 16-bit memory location. The address of the memory
	110	/// location does not have to be aligned.
	111	/// \returns
	112	/// A 128-bit vector of [4 x float].
	113	///
	114	/// \code{.operation}
	115	/// b := Convert_FP16_To_FP32(MEM[__A+15:__A])
	116	/// FOR j := 0 to 3
	117	/// m := j*32
	118	/// dst[m+31:m] := b
	119	/// ENDFOR
	120	/// dst[MAX:128] := 0
	121	/// \endcode
	122	static __inline__ __m128 __DEFAULT_FN_ATTRS128
	123	_mm_bcstnesh_ps(const void *__A) {
	124	return (__m128)__builtin_ia32_vbcstnesh2ps128((const _Float16 *)__A);
	125	}
	126
	127	/// Convert scalar half-precision (16-bit) floating-point element
	128	/// stored at memory locations starting at location \a __A to a
	129	/// single-precision (32-bit) floating-point, broadcast it to packed
	130	/// single-precision (32-bit) floating-point elements, and store the results in
	131	/// \a dst.
	132	///
	133	/// \headerfile <x86intrin.h>
	134	///
	135	/// \code
	136	/// _mm256_bcstnesh_ps(const void *__A);
	137	/// \endcode
	138	///
	139	/// This intrinsic corresponds to the \c VBCSTNESH2PS instruction.
	140	///
	141	/// \param __A
	142	/// A pointer to a 16-bit memory location. The address of the memory
	143	/// location does not have to be aligned.
	144	/// \returns
	145	/// A 256-bit vector of [8 x float].
	146	///
	147	/// \code{.operation}
	148	/// b := Convert_FP16_To_FP32(MEM[__A+15:__A])
	149	/// FOR j := 0 to 7
	150	/// m := j*32
	151	/// dst[m+31:m] := b
	152	/// ENDFOR
	153	/// dst[MAX:256] := 0
	154	/// \endcode
	155	static __inline__ __m256 __DEFAULT_FN_ATTRS256
	156	_mm256_bcstnesh_ps(const void *__A) {
	157	return (__m256)__builtin_ia32_vbcstnesh2ps256((const _Float16 *)__A);
	158	}
	159
	160	/// Convert packed BF16 (16-bit) floating-point even-indexed elements
	161	/// stored at memory locations starting at location \a __A to packed
	162	/// single-precision (32-bit) floating-point elements, and store the results in
	163	/// \a dst.
	164	///
	165	/// \headerfile <x86intrin.h>
	166	///
	167	/// \code
	168	/// _mm_cvtneebf16_ps(const __m128bh *__A);
	169	/// \endcode
	170	///
	171	/// This intrinsic corresponds to the \c VCVTNEEBF162PS instruction.
	172	///
	173	/// \param __A
	174	/// A pointer to a 128-bit memory location containing 8 consecutive
	175	/// BF16 (16-bit) floating-point values.
	176	/// \returns
	177	/// A 128-bit vector of [4 x float].
	178	///
	179	/// \code{.operation}
	180	/// FOR j := 0 to 3
	181	/// k := j*2
	182	/// i := k*16
	183	/// m := j*32
	184	/// dst[m+31:m] := Convert_BF16_To_FP32(MEM[__A+i+15:__A+i])
	185	/// ENDFOR
	186	/// dst[MAX:128] := 0
	187	/// \endcode
	188	static __inline__ __m128 __DEFAULT_FN_ATTRS128
	189	_mm_cvtneebf16_ps(const __m128bh *__A) {
	190	return (__m128)__builtin_ia32_vcvtneebf162ps128((const __v8bf *)__A);
	191	}
	192
	193	/// Convert packed BF16 (16-bit) floating-point even-indexed elements
	194	/// stored at memory locations starting at location \a __A to packed
	195	/// single-precision (32-bit) floating-point elements, and store the results in
	196	/// \a dst.
	197	///
	198	/// \headerfile <x86intrin.h>
	199	///
	200	/// \code
	201	/// _mm256_cvtneebf16_ps(const __m256bh *__A);
	202	/// \endcode
	203	///
	204	/// This intrinsic corresponds to the \c VCVTNEEBF162PS instruction.
	205	///
	206	/// \param __A
	207	/// A pointer to a 256-bit memory location containing 16 consecutive
	208	/// BF16 (16-bit) floating-point values.
	209	/// \returns
	210	/// A 256-bit vector of [8 x float].
	211	///
	212	/// \code{.operation}
	213	/// FOR j := 0 to 7
	214	/// k := j*2
	215	/// i := k*16
	216	/// m := j*32
	217	/// dst[m+31:m] := Convert_BF16_To_FP32(MEM[__A+i+15:__A+i])
	218	/// ENDFOR
	219	/// dst[MAX:256] := 0
	220	/// \endcode
	221	static __inline__ __m256 __DEFAULT_FN_ATTRS256
	222	_mm256_cvtneebf16_ps(const __m256bh *__A) {
	223	return (__m256)__builtin_ia32_vcvtneebf162ps256((const __v16bf *)__A);
	224	}
	225
	226	/// Convert packed half-precision (16-bit) floating-point even-indexed elements
	227	/// stored at memory locations starting at location \a __A to packed
	228	/// single-precision (32-bit) floating-point elements, and store the results in
	229	/// \a dst.
	230	///
	231	/// \headerfile <x86intrin.h>
	232	///
	233	/// \code
	234	/// _mm_cvtneeph_ps(const __m128h *__A);
	235	/// \endcode
	236	///
	237	/// This intrinsic corresponds to the \c VCVTNEEPH2PS instruction.
	238	///
	239	/// \param __A
	240	/// A pointer to a 128-bit memory location containing 8 consecutive
	241	/// half-precision (16-bit) floating-point values.
	242	/// \returns
	243	/// A 128-bit vector of [4 x float].
	244	///
	245	/// \code{.operation}
	246	/// FOR j := 0 to 3
	247	/// k := j*2
	248	/// i := k*16
	249	/// m := j*32
	250	/// dst[m+31:m] := Convert_FP16_To_FP32(MEM[__A+i+15:__A+i])
	251	/// ENDFOR
	252	/// dst[MAX:128] := 0
	253	/// \endcode
	254	static __inline__ __m128 __DEFAULT_FN_ATTRS128
	255	_mm_cvtneeph_ps(const __m128h *__A) {
	256	return (__m128)__builtin_ia32_vcvtneeph2ps128((const __v8hf *)__A);
	257	}
	258
	259	/// Convert packed half-precision (16-bit) floating-point even-indexed elements
	260	/// stored at memory locations starting at location \a __A to packed
	261	/// single-precision (32-bit) floating-point elements, and store the results in
	262	/// \a dst.
	263	///
	264	/// \headerfile <x86intrin.h>
	265	///
	266	/// \code
	267	/// _mm256_cvtneeph_ps(const __m256h *__A);
	268	/// \endcode
	269	///
	270	/// This intrinsic corresponds to the \c VCVTNEEPH2PS instruction.
	271	///
	272	/// \param __A
	273	/// A pointer to a 256-bit memory location containing 16 consecutive
	274	/// half-precision (16-bit) floating-point values.
	275	/// \returns
	276	/// A 256-bit vector of [8 x float].
	277	///
	278	/// \code{.operation}
	279	/// FOR j := 0 to 7
	280	/// k := j*2
	281	/// i := k*16
	282	/// m := j*32
	283	/// dst[m+31:m] := Convert_FP16_To_FP32(MEM[__A+i+15:__A+i])
	284	/// ENDFOR
	285	/// dst[MAX:256] := 0
	286	/// \endcode
	287	static __inline__ __m256 __DEFAULT_FN_ATTRS256
	288	_mm256_cvtneeph_ps(const __m256h *__A) {
	289	return (__m256)__builtin_ia32_vcvtneeph2ps256((const __v16hf *)__A);
	290	}
	291
	292	/// Convert packed BF16 (16-bit) floating-point odd-indexed elements
	293	/// stored at memory locations starting at location \a __A to packed
	294	/// single-precision (32-bit) floating-point elements, and store the results in
	295	/// \a dst.
	296	///
	297	/// \headerfile <x86intrin.h>
	298	///
	299	/// \code
	300	/// _mm_cvtneobf16_ps(const __m128bh *__A);
	301	/// \endcode
	302	///
	303	/// This intrinsic corresponds to the \c VCVTNEOBF162PS instruction.
	304	///
	305	/// \param __A
	306	/// A pointer to a 128-bit memory location containing 8 consecutive
	307	/// BF16 (16-bit) floating-point values.
	308	/// \returns
	309	/// A 128-bit vector of [4 x float].
	310	///
	311	/// \code{.operation}
	312	/// FOR j := 0 to 3
	313	/// k := j*2+1
	314	/// i := k*16
	315	/// m := j*32
	316	/// dst[m+31:m] := Convert_BF16_To_FP32(MEM[__A+i+15:__A+i])
	317	/// ENDFOR
	318	/// dst[MAX:128] := 0
	319	/// \endcode
	320	static __inline__ __m128 __DEFAULT_FN_ATTRS128
	321	_mm_cvtneobf16_ps(const __m128bh *__A) {
	322	return (__m128)__builtin_ia32_vcvtneobf162ps128((const __v8bf *)__A);
	323	}
	324
	325	/// Convert packed BF16 (16-bit) floating-point odd-indexed elements
	326	/// stored at memory locations starting at location \a __A to packed
	327	/// single-precision (32-bit) floating-point elements, and store the results in
	328	/// \a dst.
	329	///
	330	/// \headerfile <x86intrin.h>
	331	///
	332	/// \code
	333	/// _mm256_cvtneobf16_ps(const __m256bh *__A);
	334	/// \endcode
	335	///
	336	/// This intrinsic corresponds to the \c VCVTNEOBF162PS instruction.
	337	///
	338	/// \param __A
	339	/// A pointer to a 256-bit memory location containing 16 consecutive
	340	/// BF16 (16-bit) floating-point values.
	341	/// \returns
	342	/// A 256-bit vector of [8 x float].
	343	///
	344	/// \code{.operation}
	345	/// FOR j := 0 to 7
	346	/// k := j*2+1
	347	/// i := k*16
	348	/// m := j*32
	349	/// dst[m+31:m] := Convert_BF16_To_FP32(MEM[__A+i+15:__A+i])
	350	/// ENDFOR
	351	/// dst[MAX:256] := 0
	352	/// \endcode
	353	static __inline__ __m256 __DEFAULT_FN_ATTRS256
	354	_mm256_cvtneobf16_ps(const __m256bh *__A) {
	355	return (__m256)__builtin_ia32_vcvtneobf162ps256((const __v16bf *)__A);
	356	}
	357
	358	/// Convert packed half-precision (16-bit) floating-point odd-indexed elements
	359	/// stored at memory locations starting at location \a __A to packed
	360	/// single-precision (32-bit) floating-point elements, and store the results in
	361	/// \a dst.
	362	///
	363	/// \headerfile <x86intrin.h>
	364	///
	365	/// \code
	366	/// _mm_cvtneoph_ps(const __m128h *__A);
	367	/// \endcode
	368	///
	369	/// This intrinsic corresponds to the \c VCVTNEOPH2PS instruction.
	370	///
	371	/// \param __A
	372	/// A pointer to a 128-bit memory location containing 8 consecutive
	373	/// half-precision (16-bit) floating-point values.
	374	/// \returns
	375	/// A 128-bit vector of [4 x float].
	376	///
	377	/// \code{.operation}
	378	/// FOR j := 0 to 3
	379	/// k := j*2+1
	380	/// i := k*16
	381	/// m := j*32
	382	/// dst[m+31:m] := Convert_FP16_To_FP32(MEM[__A+i+15:__A+i])
	383	/// ENDFOR
	384	/// dst[MAX:128] := 0
	385	/// \endcode
	386	static __inline__ __m128 __DEFAULT_FN_ATTRS128
	387	_mm_cvtneoph_ps(const __m128h *__A) {
	388	return (__m128)__builtin_ia32_vcvtneoph2ps128((const __v8hf *)__A);
	389	}
	390
	391	/// Convert packed half-precision (16-bit) floating-point odd-indexed elements
	392	/// stored at memory locations starting at location \a __A to packed
	393	/// single-precision (32-bit) floating-point elements, and store the results in
	394	/// \a dst.
	395	///
	396	/// \headerfile <x86intrin.h>
	397	///
	398	/// \code
	399	/// _mm256_cvtneoph_ps(const __m256h *__A);
	400	/// \endcode
	401	///
	402	/// This intrinsic corresponds to the \c VCVTNEOPH2PS instruction.
	403	///
	404	/// \param __A
	405	/// A pointer to a 256-bit memory location containing 16 consecutive
	406	/// half-precision (16-bit) floating-point values.
	407	/// \returns
	408	/// A 256-bit vector of [8 x float].
	409	///
	410	/// \code{.operation}
	411	/// FOR j := 0 to 7
	412	/// k := j*2+1
	413	/// i := k*16
	414	/// m := j*32
	415	/// dst[m+31:m] := Convert_FP16_To_FP32(MEM[__A+i+15:__A+i])
	416	/// ENDFOR
	417	/// dst[MAX:256] := 0
	418	/// \endcode
	419	static __inline__ __m256 __DEFAULT_FN_ATTRS256
	420	_mm256_cvtneoph_ps(const __m256h *__A) {
	421	return (__m256)__builtin_ia32_vcvtneoph2ps256((const __v16hf *)__A);
	422	}
	423
	424	/// Convert packed single-precision (32-bit) floating-point elements in \a __A
	425	/// to packed BF16 (16-bit) floating-point elements, and store the results in \a
	426	/// dst.
	427	///
	428	/// \headerfile <x86intrin.h>
	429	///
	430	/// \code
	431	/// _mm_cvtneps_avx_pbh(__m128 __A);
	432	/// \endcode
	433	///
	434	/// This intrinsic corresponds to the \c VCVTNEPS2BF16 instruction.
	435	///
	436	/// \param __A
	437	/// A 128-bit vector of [4 x float].
	438	/// \returns
	439	/// A 128-bit vector of [8 x bfloat].
	440	///
	441	/// \code{.operation}
	442	/// FOR j := 0 to 3
	443	/// dst.word[j] := Convert_FP32_To_BF16(__A.fp32[j])
	444	/// ENDFOR
	445	/// dst[MAX:128] := 0
	446	/// \endcode
	447	static __inline__ __m128bh __DEFAULT_FN_ATTRS128
	448	_mm_cvtneps_avx_pbh(__m128 __A) {
	449	return (__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)__A);
	450	}
	451
	452	/// Convert packed single-precision (32-bit) floating-point elements in \a __A
	453	/// to packed BF16 (16-bit) floating-point elements, and store the results in \a
	454	/// dst.
	455	///
	456	/// \headerfile <x86intrin.h>
	457	///
	458	/// \code
	459	/// _mm256_cvtneps_avx_pbh(__m256 __A);
	460	/// \endcode
	461	///
	462	/// This intrinsic corresponds to the \c VCVTNEPS2BF16 instruction.
	463	///
	464	/// \param __A
	465	/// A 256-bit vector of [8 x float].
	466	/// \returns
	467	/// A 128-bit vector of [8 x bfloat].
	468	///
	469	/// \code{.operation}
	470	/// FOR j := 0 to 7
	471	/// dst.word[j] := Convert_FP32_To_BF16(a.fp32[j])
	472	/// ENDFOR
	473	/// dst[MAX:128] := 0
	474	/// \endcode
	475	static __inline__ __m128bh __DEFAULT_FN_ATTRS256
	476	_mm256_cvtneps_avx_pbh(__m256 __A) {
	477	return (__m128bh)__builtin_ia32_vcvtneps2bf16256((__v8sf)__A);
	478	}
	479
	480	#undef __DEFAULT_FN_ATTRS128
	481	#undef __DEFAULT_FN_ATTRS256
	482
	483	#endif // __AVXNECONVERTINTRIN_H
	484	#endif // __SSE2__

lib/include/avxvnniint8intrin.h created+471

...	...	@@ -0,0 +1,471 @@
	1	/*===-------- avxvnniint8intrin.h - AVXVNNIINT8 intrinsics -----------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9	#ifndef __IMMINTRIN_H
	10	#error \
	11	"Never use <avxvnniint8intrin.h> directly; include <immintrin.h> instead."
	12	#endif
	13
	14	#ifndef __AVXVNNIINT8INTRIN_H
	15	#define __AVXVNNIINT8INTRIN_H
	16
	17	/* Define the default attributes for the functions in this file. */
	18	#define __DEFAULT_FN_ATTRS256 \
	19	__attribute__((__always_inline__, __nodebug__, __target__("avxvnniint8"), \
	20	__min_vector_width__(256)))
	21	#define __DEFAULT_FN_ATTRS128 \
	22	__attribute__((__always_inline__, __nodebug__, __target__("avxvnniint8"), \
	23	__min_vector_width__(128)))
	24
	25	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	26	/// corresponding signed 8-bit integers in \a __B, producing 4 intermediate
	27	/// signed 16-bit results. Sum these 4 results with the corresponding
	28	/// 32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
	29	///
	30	/// \headerfile <x86intrin.h>
	31	///
	32	/// \code
	33	/// _mm_dpbssd_epi32(__m128i __W, __m128i __A, __m128i __B);
	34	/// \endcode
	35	///
	36	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	37	///
	38	/// \param __A
	39	/// A 128-bit vector of [16 x char].
	40	/// \param __B
	41	/// A 128-bit vector of [16 x char].
	42	/// \returns
	43	/// A 128-bit vector of [4 x int].
	44	///
	45	/// \code{.operation}
	46	/// FOR j := 0 to 3
	47	/// tmp1.word := SignExtend16(__A.byte[4j]) SignExtend16(__B.byte[4*j])
	48	/// tmp2.word := SignExtend16(__A.byte[4j+1]) SignExtend16(__B.byte[4*j+1])
	49	/// tmp3.word := SignExtend16(__A.byte[4j+2]) SignExtend16(__B.byte[4*j+2])
	50	/// tmp4.word := SignExtend16(__A.byte[4j+3]) SignExtend16(__B.byte[4*j+3])
	51	/// dst.dword[j] := __W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4
	52	/// ENDFOR
	53	/// dst[MAX:128] := 0
	54	/// \endcode
	55	static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpbssd_epi32(__m128i __W,
	56	__m128i __A,
	57	__m128i __B) {
	58	return (__m128i)__builtin_ia32_vpdpbssd128((__v4si)__W, (__v4si)__A,
	59	(__v4si)__B);
	60	}
	61
	62	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	63	/// corresponding signed 8-bit integers in \a __B, producing 4 intermediate
	64	/// signed 16-bit results. Sum these 4 results with the corresponding
	65	/// 32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
	66	///
	67	/// \headerfile <x86intrin.h>
	68	///
	69	/// \code
	70	/// _mm256_dpbssd_epi32(__m256i __W, __m256i __A, __m256i __B);
	71	/// \endcode
	72	///
	73	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	74	///
	75	/// \param __A
	76	/// A 256-bit vector of [32 x char].
	77	/// \param __B
	78	/// A 256-bit vector of [32 x char].
	79	/// \returns
	80	/// A 256-bit vector of [8 x int].
	81	///
	82	/// \code{.operation}
	83	/// FOR j := 0 to 7
	84	/// tmp1.word := SignExtend16(__A.byte[4j]) SignExtend16(__B.byte[4*j])
	85	/// tmp2.word := SignExtend16(__A.byte[4j+1]) SignExtend16(__B.byte[4*j+1])
	86	/// tmp3.word := SignExtend16(__A.byte[4j+2]) SignExtend16(__B.byte[4*j+2])
	87	/// tmp4.word := SignExtend16(__A.byte[4j+3]) SignExtend16(__B.byte[4*j+3])
	88	/// dst.dword[j] := __W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4
	89	/// ENDFOR
	90	/// dst[MAX:256] := 0
	91	/// \endcode
	92	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	93	_mm256_dpbssd_epi32(__m256i __W, __m256i __A, __m256i __B) {
	94	return (__m256i)__builtin_ia32_vpdpbssd256((__v8si)__W, (__v8si)__A,
	95	(__v8si)__B);
	96	}
	97
	98	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	99	/// corresponding signed 8-bit integers in \a __B, producing 4 intermediate
	100	/// signed 16-bit results. Sum these 4 results with the corresponding
	101	/// 32-bit integer in \a __W with signed saturation, and store the packed
	102	/// 32-bit results in \a dst.
	103	///
	104	/// \headerfile <x86intrin.h>
	105	///
	106	/// \code
	107	/// _mm_dpbssds_epi32( __m128i __W, __m128i __A, __m128i __B);
	108	/// \endcode
	109	///
	110	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	111	///
	112	/// \param __A
	113	/// A 128-bit vector of [16 x char].
	114	/// \param __B
	115	/// A 128-bit vector of [16 x char].
	116	/// \returns
	117	/// A 128-bit vector of [4 x int].
	118	///
	119	/// \code{.operation}
	120	/// FOR j := 0 to 3
	121	/// tmp1.word := SignExtend16(__A.byte[4j]) SignExtend16(__B.byte[4*j])
	122	/// tmp2.word := SignExtend16(__A.byte[4j+1]) SignExtend16(__B.byte[4*j+1])
	123	/// tmp3.word := SignExtend16(__A.byte[4j+2]) SignExtend16(__B.byte[4*j+2])
	124	/// tmp4.word := SignExtend16(__A.byte[4j+3]) SignExtend16(__B.byte[4*j+3])
	125	/// dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4)
	126	/// ENDFOR
	127	/// dst[MAX:128] := 0
	128	/// \endcode
	129	static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpbssds_epi32(__m128i __W,
	130	__m128i __A,
	131	__m128i __B) {
	132	return (__m128i)__builtin_ia32_vpdpbssds128((__v4si)__W, (__v4si)__A,
	133	(__v4si)__B);
	134	}
	135
	136	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	137	/// corresponding signed 8-bit integers in \a __B, producing 4 intermediate
	138	/// signed 16-bit results. Sum these 4 results with the corresponding
	139	/// 32-bit integer in \a __W with signed saturation, and store the packed
	140	/// 32-bit results in \a dst.
	141	///
	142	/// \headerfile <x86intrin.h>
	143	///
	144	/// \code
	145	/// _mm256_dpbssds_epi32(__m256i __W, __m256i __A, __m256i __B);
	146	/// \endcode
	147	///
	148	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	149	///
	150	/// \param __A
	151	/// A 256-bit vector of [32 x char].
	152	/// \param __B
	153	/// A 256-bit vector of [32 x char].
	154	/// \returns
	155	/// A 256-bit vector of [8 x int].
	156	///
	157	/// \code{.operation}
	158	/// FOR j := 0 to 7
	159	/// tmp1.word := SignExtend16(__A.byte[4j]) SignExtend16(__B.byte[4*j])
	160	/// tmp2.word := SignExtend16(__A.byte[4j+1]) SignExtend16(__B.byte[4*j+1])
	161	/// tmp3.word := SignExtend16(__A.byte[4j+2]) SignExtend16(__B.byte[4*j+2])
	162	/// tmp4.word := SignExtend16(__A.byte[4j+3]) SignExtend16(__B.byte[4*j+3])
	163	/// dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4)
	164	/// ENDFOR
	165	/// dst[MAX:256] := 0
	166	/// \endcode
	167	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	168	_mm256_dpbssds_epi32(__m256i __W, __m256i __A, __m256i __B) {
	169	return (__m256i)__builtin_ia32_vpdpbssds256((__v8si)__W, (__v8si)__A,
	170	(__v8si)__B);
	171	}
	172
	173	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	174	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	175	/// signed 16-bit results. Sum these 4 results with the corresponding
	176	/// 32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
	177	///
	178	/// \headerfile <x86intrin.h>
	179	///
	180	/// \code
	181	/// _mm_dpbsud_epi32(__m128i __W, __m128i __A, __m128i __B);
	182	/// \endcode
	183	///
	184	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	185	///
	186	/// \param __A
	187	/// A 128-bit vector of [16 x char].
	188	/// \param __B
	189	/// A 128-bit vector of [16 x unsigned char].
	190	/// \returns
	191	/// A 128-bit vector of [4 x int].
	192	///
	193	/// \code{.operation}
	194	/// FOR j := 0 to 3
	195	/// tmp1.word := Signed(SignExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j]))
	196	/// tmp2.word := Signed(SignExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1]))
	197	/// tmp3.word := Signed(SignExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2]))
	198	/// tmp4.word := Signed(SignExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3]))
	199	/// dst.dword[j] := __W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4
	200	/// ENDFOR
	201	/// dst[MAX:128] := 0
	202	/// \endcode
	203	static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpbsud_epi32(__m128i __W,
	204	__m128i __A,
	205	__m128i __B) {
	206	return (__m128i)__builtin_ia32_vpdpbsud128((__v4si)__W, (__v4si)__A,
	207	(__v4si)__B);
	208	}
	209
	210	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	211	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	212	/// signed 16-bit results. Sum these 4 results with the corresponding
	213	/// 32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
	214	///
	215	/// \headerfile <x86intrin.h>
	216	///
	217	/// \code
	218	/// _mm256_dpbsud_epi32(__m256i __W, __m256i __A, __m256i __B);
	219	/// \endcode
	220	///
	221	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	222	///
	223	/// \param __A
	224	/// A 256-bit vector of [32 x char].
	225	/// \param __B
	226	/// A 256-bit vector of [32 x unsigned char].
	227	/// \returns
	228	/// A 256-bit vector of [8 x int].
	229	///
	230	/// \code{.operation}
	231	/// FOR j := 0 to 7
	232	/// tmp1.word := Signed(SignExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j]))
	233	/// tmp2.word := Signed(SignExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1]))
	234	/// tmp3.word := Signed(SignExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2]))
	235	/// tmp4.word := Signed(SignExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3]))
	236	/// dst.dword[j] := __W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4
	237	/// ENDFOR
	238	/// dst[MAX:256] := 0
	239	/// \endcode
	240	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	241	_mm256_dpbsud_epi32(__m256i __W, __m256i __A, __m256i __B) {
	242	return (__m256i)__builtin_ia32_vpdpbsud256((__v8si)__W, (__v8si)__A,
	243	(__v8si)__B);
	244	}
	245
	246	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	247	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	248	/// signed 16-bit results. Sum these 4 results with the corresponding
	249	/// 32-bit integer in \a __W with signed saturation, and store the packed
	250	/// 32-bit results in \a dst.
	251	///
	252	/// \headerfile <x86intrin.h>
	253	///
	254	/// \code
	255	/// _mm_dpbsuds_epi32( __m128i __W, __m128i __A, __m128i __B);
	256	/// \endcode
	257	///
	258	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	259	///
	260	/// \param __A
	261	/// A 128-bit vector of [16 x char].
	262	/// \param __B
	263	/// A 128-bit vector of [16 x unsigned char].
	264	/// \returns
	265	/// A 128-bit vector of [4 x int].
	266	///
	267	/// \code{.operation}
	268	/// FOR j := 0 to 3
	269	/// tmp1.word := Signed(SignExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j]))
	270	/// tmp2.word := Signed(SignExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1]))
	271	/// tmp3.word := Signed(SignExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2]))
	272	/// tmp4.word := Signed(SignExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3]))
	273	/// dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4)
	274	/// ENDFOR
	275	/// dst[MAX:128] := 0
	276	/// \endcode
	277	static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpbsuds_epi32(__m128i __W,
	278	__m128i __A,
	279	__m128i __B) {
	280	return (__m128i)__builtin_ia32_vpdpbsuds128((__v4si)__W, (__v4si)__A,
	281	(__v4si)__B);
	282	}
	283
	284	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	285	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	286	/// signed 16-bit results. Sum these 4 results with the corresponding
	287	/// 32-bit integer in \a __W with signed saturation, and store the packed
	288	/// 32-bit results in \a dst.
	289	///
	290	/// \headerfile <x86intrin.h>
	291	///
	292	/// \code
	293	/// _mm256_dpbsuds_epi32(__m256i __W, __m256i __A, __m256i __B);
	294	/// \endcode
	295	///
	296	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	297	///
	298	/// \param __A
	299	/// A 256-bit vector of [32 x char].
	300	/// \param __B
	301	/// A 256-bit vector of [32 x unsigned char].
	302	/// \returns
	303	/// A 256-bit vector of [8 x int].
	304	///
	305	/// \code{.operation}
	306	/// FOR j := 0 to 7
	307	/// tmp1.word := Signed(SignExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j]))
	308	/// tmp2.word := Signed(SignExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1]))
	309	/// tmp3.word := Signed(SignExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2]))
	310	/// tmp4.word := Signed(SignExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3]))
	311	/// dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4)
	312	/// ENDFOR
	313	/// dst[MAX:256] := 0
	314	/// \endcode
	315	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	316	_mm256_dpbsuds_epi32(__m256i __W, __m256i __A, __m256i __B) {
	317	return (__m256i)__builtin_ia32_vpdpbsuds256((__v8si)__W, (__v8si)__A,
	318	(__v8si)__B);
	319	}
	320
	321	/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a __A with
	322	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	323	/// signed 16-bit results. Sum these 4 results with the corresponding
	324	/// 32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
	325	///
	326	/// \headerfile <x86intrin.h>
	327	///
	328	/// \code
	329	/// _mm_dpbuud_epi32(__m128i __W, __m128i __A, __m128i __B);
	330	/// \endcode
	331	///
	332	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	333	///
	334	/// \param __A
	335	/// A 128-bit vector of [16 x unsigned char].
	336	/// \param __B
	337	/// A 128-bit vector of [16 x unsigned char].
	338	/// \returns
	339	/// A 128-bit vector of [4 x int].
	340	///
	341	/// \code{.operation}
	342	/// FOR j := 0 to 3
	343	/// tmp1.word := ZeroExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j])
	344	/// tmp2.word := ZeroExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1])
	345	/// tmp3.word := ZeroExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2])
	346	/// tmp4.word := ZeroExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3])
	347	/// dst.dword[j] := __W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4
	348	/// ENDFOR
	349	/// dst[MAX:128] := 0
	350	/// \endcode
	351	static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpbuud_epi32(__m128i __W,
	352	__m128i __A,
	353	__m128i __B) {
	354	return (__m128i)__builtin_ia32_vpdpbuud128((__v4si)__W, (__v4si)__A,
	355	(__v4si)__B);
	356	}
	357
	358	/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a __A with
	359	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	360	/// signed 16-bit results. Sum these 4 results with the corresponding
	361	/// 32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
	362	///
	363	/// \headerfile <x86intrin.h>
	364	///
	365	/// \code
	366	/// _mm256_dpbuud_epi32(__m256i __W, __m256i __A, __m256i __B);
	367	/// \endcode
	368	///
	369	/// This intrinsic corresponds to the \c VPDPBSSD instruction.
	370	///
	371	/// \param __A
	372	/// A 256-bit vector of [32 x unsigned char].
	373	/// \param __B
	374	/// A 256-bit vector of [32 x unsigned char].
	375	/// \returns
	376	/// A 256-bit vector of [8 x int].
	377	///
	378	/// \code{.operation}
	379	/// FOR j := 0 to 7
	380	/// tmp1.word := ZeroExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j])
	381	/// tmp2.word := ZeroExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1])
	382	/// tmp3.word := ZeroExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2])
	383	/// tmp4.word := ZeroExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3])
	384	/// dst.dword[j] := __W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4
	385	/// ENDFOR
	386	/// dst[MAX:256] := 0
	387	/// \endcode
	388	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	389	_mm256_dpbuud_epi32(__m256i __W, __m256i __A, __m256i __B) {
	390	return (__m256i)__builtin_ia32_vpdpbuud256((__v8si)__W, (__v8si)__A,
	391	(__v8si)__B);
	392	}
	393
	394	/// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a __A with
	395	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	396	/// signed 16-bit results. Sum these 4 results with the corresponding
	397	/// 32-bit integer in \a __W with signed saturation, and store the packed
	398	/// 32-bit results in \a dst.
	399	///
	400	/// \headerfile <x86intrin.h>
	401	///
	402	/// \code
	403	/// _mm_dpbuuds_epi32( __m128i __W, __m128i __A, __m128i __B);
	404	/// \endcode
	405	///
	406	/// This intrinsic corresponds to the \c VPDPBUUDS instruction.
	407	///
	408	/// \param __A
	409	/// A 128-bit vector of [16 x unsigned char].
	410	/// \param __B
	411	/// A 128-bit vector of [16 x unsigned char].
	412	/// \returns
	413	/// A 128-bit vector of [4 x int].
	414	///
	415	/// \code{.operation}
	416	/// FOR j := 0 to 3
	417	/// tmp1.word := ZeroExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j])
	418	/// tmp2.word := ZeroExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1])
	419	/// tmp3.word := ZeroExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2])
	420	/// tmp4.word := ZeroExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3])
	421	/// dst.dword[j] := UNSIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4)
	422	/// ENDFOR
	423	/// dst[MAX:128] := 0
	424	/// \endcode
	425	static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpbuuds_epi32(__m128i __W,
	426	__m128i __A,
	427	__m128i __B) {
	428	return (__m128i)__builtin_ia32_vpdpbuuds128((__v4si)__W, (__v4si)__A,
	429	(__v4si)__B);
	430	}
	431
	432	/// Multiply groups of 4 adjacent pairs of signed 8-bit integers in \a __A with
	433	/// corresponding unsigned 8-bit integers in \a __B, producing 4 intermediate
	434	/// signed 16-bit results. Sum these 4 results with the corresponding
	435	/// 32-bit integer in \a __W with signed saturation, and store the packed
	436	/// 32-bit results in \a dst.
	437	///
	438	/// \headerfile <x86intrin.h>
	439	///
	440	/// \code
	441	/// _mm256_dpbuuds_epi32(__m256i __W, __m256i __A, __m256i __B);
	442	/// \endcode
	443	///
	444	/// This intrinsic corresponds to the \c VPDPBUUDS instruction.
	445	///
	446	/// \param __A
	447	/// A 256-bit vector of [32 x unsigned char].
	448	/// \param __B
	449	/// A 256-bit vector of [32 x unsigned char].
	450	/// \returns
	451	/// A 256-bit vector of [8 x int].
	452	///
	453	/// \code{.operation}
	454	/// FOR j := 0 to 7
	455	/// tmp1.word := ZeroExtend16(__A.byte[4j]) ZeroExtend16(__B.byte[4*j])
	456	/// tmp2.word := ZeroExtend16(__A.byte[4j+1]) ZeroExtend16(__B.byte[4*j+1])
	457	/// tmp3.word := ZeroExtend16(__A.byte[4j+2]) ZeroExtend16(__B.byte[4*j+2])
	458	/// tmp4.word := ZeroExtend16(__A.byte[4j+3]) ZeroExtend16(__B.byte[4*j+3])
	459	/// dst.dword[j] := UNSIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2 + tmp3 + tmp4)
	460	/// ENDFOR
	461	/// dst[MAX:256] := 0
	462	/// \endcode
	463	static __inline__ __m256i __DEFAULT_FN_ATTRS256
	464	_mm256_dpbuuds_epi32(__m256i __W, __m256i __A, __m256i __B) {
	465	return (__m256i)__builtin_ia32_vpdpbuuds256((__v8si)__W, (__v8si)__A,
	466	(__v8si)__B);
	467	}
	468	#undef __DEFAULT_FN_ATTRS128
	469	#undef __DEFAULT_FN_ATTRS256
	470
	471	#endif // __AVXVNNIINT8INTRIN_H

lib/include/cmpccxaddintrin.h created+70

...	...	@@ -0,0 +1,70 @@
	1	/*===--------------- cmpccxaddintrin.h - CMPCCXADD intrinsics--------------===
	2	*
	3	*
	4	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	5	* See https://llvm.org/LICENSE.txt for license information.
	6	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	7	*
	8	*===-----------------------------------------------------------------------===
	9	*/
	10	#ifndef __X86GPRINTRIN_H
	11	#error \
	12	"Never use <cmpccxaddintrin.h> directly; include <x86gprintrin.h> instead."
	13	#endif // __X86GPRINTRIN_H
	14
	15	#ifndef __CMPCCXADDINTRIN_H
	16	#define __CMPCCXADDINTRIN_H
	17	#ifdef __x86_64__
	18
	19	typedef enum {
	20	_CMPCCX_O, /* Overflow. */
	21	_CMPCCX_NO, /* No overflow. */
	22	_CMPCCX_B, /* Below. */
	23	_CMPCCX_NB, /* Not below. */
	24	_CMPCCX_Z, /* Zero. */
	25	_CMPCCX_NZ, /* Not zero. */
	26	_CMPCCX_BE, /* Below or equal. */
	27	_CMPCCX_NBE, /* Neither below nor equal. */
	28	_CMPCCX_S, /* Sign. */
	29	_CMPCCX_NS, /* No sign. */
	30	_CMPCCX_P, /* Parity. */
	31	_CMPCCX_NP, /* No parity. */
	32	_CMPCCX_L, /* Less. */
	33	_CMPCCX_NL, /* Not less. */
	34	_CMPCCX_LE, /* Less or equal. */
	35	_CMPCCX_NLE, /* Neither less nor equal. */
	36	} _CMPCCX_ENUM;
	37
	38	/// Compares the value from the memory __A with the value of __B. If the
	39	/// specified condition __D is met, then add the third operand __C to the
	40	/// __A and write it into __A, else the value of __A is unchanged. The return
	41	/// value is the original value of __A.
	42	///
	43	/// \headerfile <immintrin.h>
	44	///
	45	/// This intrinsic corresponds to the \c CMPCCXADD instructions.
	46	///
	47	/// \param __A
	48	/// __A pointer specifying the memory address.
	49	///
	50	/// \param __B
	51	/// A integer operand.
	52	///
	53	/// \param __C
	54	/// A integer operand.
	55	///
	56	/// \param __D
	57	/// The specified condition.
	58	///
	59	/// \returns a integer which is the original value of first operand.
	60
	61	#define _cmpccxadd_epi32(__A, __B, __C, __D) \
	62	((int)(__builtin_ia32_cmpccxadd32((void *)(__A), (int)(__B), (int)(__C), \
	63	(int)(__D))))
	64
	65	#define _cmpccxadd_epi64(__A, __B, __C, __D) \
	66	((long long)(__builtin_ia32_cmpccxadd64((void *)(__A), (long long)(__B), \
	67	(long long)(__C), (int)(__D))))
	68
	69	#endif // __x86_64__
	70	#endif // __CMPCCXADDINTRIN_H

lib/include/cpuid.h+11-1

...	...	@@ -200,9 +200,18 @@
200	200	#define bit_AMXINT8 0x02000000
201	201
202	202	/* Features in %eax for leaf 7 sub-leaf 1 */
	203	#define bit_RAOINT 0x00000008
203	204	#define bit_AVXVNNI 0x00000010
204	205	#define bit_AVX512BF16 0x00000020
	206	#define bit_CMPCCXADD 0x00000080
	207	#define bit_AMXFP16 0x00200000
205	208	#define bit_HRESET 0x00400000
	209	#define bit_AVXIFMA 0x00800000
	210
	211	/* Features in %edx for leaf 7 sub-leaf 1 */
	212	#define bit_AVXVNNIINT8 0x00000010
	213	#define bit_AVXNECONVERT 0x00000020
	214	#define bit_PREFETCHI 0x00004000
206	215
207	216	/* Features in %eax for leaf 13 sub-leaf 1 */
208	217	#define bit_XSAVEOPT 0x00000001
...	...	@@ -261,7 +270,8 @@
261	270	: "0"(__leaf), "2"(__count))
262	271	#endif
263	272
264		static __inline int __get_cpuid_max (unsigned int __leaf, unsigned int *__sig)
	273	static __inline unsigned int __get_cpuid_max (unsigned int __leaf,
	274	unsigned int *__sig)
265	275	{
266	276	unsigned int __eax, __ebx, __ecx, __edx;
267	277	#if __i386__

lib/include/cuda_wrappers/cmath created+90

...	...	@@ -0,0 +1,90 @@
	1	/*===---- cmath - CUDA wrapper for <cmath> ---------------------------------===
	2	*
	3	* Permission is hereby granted, free of charge, to any person obtaining a copy
	4	* of this software and associated documentation files (the "Software"), to deal
	5	* in the Software without restriction, including without limitation the rights
	6	* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
	7	* copies of the Software, and to permit persons to whom the Software is
	8	* furnished to do so, subject to the following conditions:
	9	*
	10	* The above copyright notice and this permission notice shall be included in
	11	* all copies or substantial portions of the Software.
	12	*
	13	* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
	14	* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
	15	* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
	16	* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
	17	* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
	18	* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
	19	* THE SOFTWARE.
	20	*
	21	*===-----------------------------------------------------------------------===
	22	*/
	23
	24	#ifndef __CLANG_CUDA_WRAPPERS_CMATH
	25	#define __CLANG_CUDA_WRAPPERS_CMATH
	26
	27	#include_next <cmath>
	28
	29	#if defined(_LIBCPP_STD_VER)
	30
	31	// libc++ will need long double variants of these functions, but CUDA does not
	32	// provide them. We'll provide their declarations, which should allow the
	33	// headers to parse, but would not allow accidental use of them on a GPU.
	34
	35	__attribute__((device)) long double logb(long double);
	36	__attribute__((device)) long double scalbn(long double, int);
	37
	38	namespace std {
	39
	40	// For __constexpr_fmin/fmax we only need device-side overloads before c++14
	41	// where they are not constexpr.
	42	#if _LIBCPP_STD_VER < 14
	43
	44	__attribute__((device))
	45	inline _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 float __constexpr_fmax(float __x, float __y) _NOEXCEPT {
	46	return __builtin_fmaxf(__x, __y);
	47	}
	48
	49	__attribute__((device))
	50	inline _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 double __constexpr_fmax(double __x, double __y) _NOEXCEPT {
	51	return __builtin_fmax(__x, __y);
	52	}
	53
	54	__attribute__((device))
	55	inline _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 long double
	56	__constexpr_fmax(long double __x, long double __y) _NOEXCEPT {
	57	return __builtin_fmaxl(__x, __y);
	58	}
	59
	60	template <class _Tp, class _Up, __enable_if_t<is_arithmetic<_Tp>::value && is_arithmetic<_Up>::value, int> = 0>
	61	__attribute__((device))
	62	_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 typename __promote<_Tp, _Up>::type
	63	__constexpr_fmax(_Tp __x, _Up __y) _NOEXCEPT {
	64	using __result_type = typename __promote<_Tp, _Up>::type;
	65	return std::__constexpr_fmax(static_cast<__result_type>(__x), static_cast<__result_type>(__y));
	66	}
	67	#endif // _LIBCPP_STD_VER < 14
	68
	69	// For logb/scalbn templates we must always provide device overloads because
	70	// libc++ implementation uses __builtin_XXX which gets translated into a libcall
	71	// which we can't handle on GPU. We need to forward those to CUDA-provided
	72	// implementations.
	73
	74	template <class _Tp>
	75	__attribute__((device))
	76	_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Tp __constexpr_logb(_Tp __x) {
	77	return ::logb(__x);
	78	}
	79
	80	template <class _Tp>
	81	__attribute__((device))
	82	_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX20 _Tp __constexpr_scalbn(_Tp __x, int __exp) {
	83	return ::scalbn(__x, __exp);
	84	}
	85
	86	} // namespace std//
	87
	88	#endif // _LIBCPP_STD_VER
	89
	90	#endif // include guard

lib/include/emmintrin.h+11-1

...	...	@@ -38,6 +38,16 @@ typedef unsigned char __v16qu __attribute__((__vector_size__(16)));
38	38	* appear in the interface though. */
39	39	typedef signed char __v16qs __attribute__((__vector_size__(16)));
40	40
	41	#ifdef __SSE2__
	42	/* Both _Float16 and __bf16 require SSE2 being enabled. */
	43	typedef _Float16 __v8hf __attribute__((__vector_size__(16), __aligned__(16)));
	44	typedef _Float16 __m128h __attribute__((__vector_size__(16), __aligned__(16)));
	45	typedef _Float16 __m128h_u __attribute__((__vector_size__(16), __aligned__(1)));
	46
	47	typedef __bf16 __v8bf __attribute__((__vector_size__(16), __aligned__(16)));
	48	typedef __bf16 __m128bh __attribute__((__vector_size__(16), __aligned__(16)));
	49	#endif
	50
41	51	/* Define the default attributes for the functions in this file. */
42	52	#define __DEFAULT_FN_ATTRS \
43	53	__attribute__((__always_inline__, __nodebug__, __target__("sse2"), \
...	...	@@ -1809,7 +1819,7 @@ static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_setr_pd(double __w,
1809	1819	/// \returns An initialized 128-bit floating-point vector of [2 x double] with
1810	1820	/// all elements set to zero.
1811	1821	static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_setzero_pd(void) {
1812		return __extension__(__m128d){0, 0};
	1822	return __extension__(__m128d){0.0, 0.0};
1813	1823	}
1814	1824
1815	1825	/// Constructs a 128-bit floating-point vector of [2 x double]. The lower

lib/include/float.h+17-10

...	...	@@ -38,9 +38,10 @@
38	38	# undef FLT_MANT_DIG
39	39	# undef DBL_MANT_DIG
40	40	# undef LDBL_MANT_DIG
41		# if __STDC_VERSION__ >= 199901L \|\| !defined(__STRICT_ANSI__) \|\| \
42		__cplusplus >= 201103L \|\| \
43		(__STDC_HOSTED__ && defined(_AIX) && defined(_ALL_SOURCE))
	41	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) \|\| \
	42	!defined(__STRICT_ANSI__) \|\| \
	43	(defined(__cplusplus) && __cplusplus >= 201103L) \|\| \
	44	(__STDC_HOSTED__ && defined(_AIX) && defined(_ALL_SOURCE))
44	45	# undef DECIMAL_DIG
45	46	# endif
46	47	# undef FLT_DIG
...	...	@@ -67,9 +68,10 @@
67	68	# undef FLT_MIN
68	69	# undef DBL_MIN
69	70	# undef LDBL_MIN
70		# if __STDC_VERSION__ >= 201112L \|\| !defined(__STRICT_ANSI__) \|\| \
71		__cplusplus >= 201703L \|\| \
72		(__STDC_HOSTED__ && defined(_AIX) && defined(_ALL_SOURCE))
	71	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L) \|\| \
	72	!defined(__STRICT_ANSI__) \|\| \
	73	(defined(__cplusplus) && __cplusplus >= 201703L) \|\| \
	74	(__STDC_HOSTED__ && defined(_AIX) && defined(_ALL_SOURCE))
73	75	# undef FLT_TRUE_MIN
74	76	# undef DBL_TRUE_MIN
75	77	# undef LDBL_TRUE_MIN
...	...	@@ -84,7 +86,10 @@
84	86
85	87	/* Characteristics of floating point types, C99 5.2.4.2.2 */
86	88
	89	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) \|\| \
	90	(defined(__cplusplus) && __cplusplus >= 201103L)
87	91	#define FLT_EVAL_METHOD __FLT_EVAL_METHOD__
	92	#endif
88	93	#define FLT_ROUNDS (__builtin_flt_rounds())
89	94	#define FLT_RADIX __FLT_RADIX__
90	95
...	...	@@ -92,8 +97,9 @@
92	97	#define DBL_MANT_DIG __DBL_MANT_DIG__
93	98	#define LDBL_MANT_DIG __LDBL_MANT_DIG__
94	99
95		#if __STDC_VERSION__ >= 199901L \|\| !defined(__STRICT_ANSI__) \|\| \
96		__cplusplus >= 201103L \|\| \
	100	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) \|\| \
	101	!defined(__STRICT_ANSI__) \|\| \
	102	(defined(__cplusplus) && __cplusplus >= 201103L) \|\| \
97	103	(__STDC_HOSTED__ && defined(_AIX) && defined(_ALL_SOURCE))
98	104	# define DECIMAL_DIG __DECIMAL_DIG__
99	105	#endif
...	...	@@ -130,8 +136,9 @@
130	136	#define DBL_MIN __DBL_MIN__
131	137	#define LDBL_MIN __LDBL_MIN__
132	138
133		#if __STDC_VERSION__ >= 201112L \|\| !defined(__STRICT_ANSI__) \|\| \
134		__cplusplus >= 201703L \|\| \
	139	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L) \|\| \
	140	!defined(__STRICT_ANSI__) \|\| \
	141	(defined(__cplusplus) && __cplusplus >= 201703L) \|\| \
135	142	(__STDC_HOSTED__ && defined(_AIX) && defined(_ALL_SOURCE))
136	143	# define FLT_TRUE_MIN __FLT_DENORM_MIN__
137	144	# define DBL_TRUE_MIN __DBL_DENORM_MIN__

lib/include/gfniintrin.h+7-5

...	...	@@ -20,10 +20,12 @@
20	20	/* Default attributes for YMM unmasked form. */
21	21	#define __DEFAULT_FN_ATTRS_Y __attribute__((__always_inline__, __nodebug__, __target__("avx,gfni"), __min_vector_width__(256)))
22	22
23		/* Default attributes for ZMM forms. */
24		#define __DEFAULT_FN_ATTRS_Z __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,gfni"), __min_vector_width__(512)))
	23	/* Default attributes for ZMM unmasked forms. */
	24	#define __DEFAULT_FN_ATTRS_Z __attribute__((__always_inline__, __nodebug__, __target__("avx512f,gfni"), __min_vector_width__(512)))
	25	/* Default attributes for ZMM masked forms. */
	26	#define __DEFAULT_FN_ATTRS_Z_MASK __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,gfni"), __min_vector_width__(512)))
25	27
26		/* Default attributes for VLX forms. */
	28	/* Default attributes for VLX masked forms. */
27	29	#define __DEFAULT_FN_ATTRS_VL128 __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,avx512vl,gfni"), __min_vector_width__(128)))
28	30	#define __DEFAULT_FN_ATTRS_VL256 __attribute__((__always_inline__, __nodebug__, __target__("avx512bw,avx512vl,gfni"), __min_vector_width__(256)))
29	31
...	...	@@ -99,7 +101,7 @@ _mm512_gf2p8mul_epi8(__m512i __A, __m512i __B)
99	101	(__v64qi) __B);
100	102	}
101	103
102		static __inline__ __m512i __DEFAULT_FN_ATTRS_Z
	104	static __inline__ __m512i __DEFAULT_FN_ATTRS_Z_MASK
103	105	_mm512_mask_gf2p8mul_epi8(__m512i __S, __mmask64 __U, __m512i __A, __m512i __B)
104	106	{
105	107	return (__m512i) __builtin_ia32_selectb_512(__U,
...	...	@@ -107,7 +109,7 @@ _mm512_mask_gf2p8mul_epi8(__m512i __S, __mmask64 __U, __m512i __A, __m512i __B)
107	109	(__v64qi) __S);
108	110	}
109	111
110		static __inline__ __m512i __DEFAULT_FN_ATTRS_Z
	112	static __inline__ __m512i __DEFAULT_FN_ATTRS_Z_MASK
111	113	_mm512_maskz_gf2p8mul_epi8(__mmask64 __U, __m512i __A, __m512i __B)
112	114	{
113	115	return _mm512_mask_gf2p8mul_epi8((__m512i)_mm512_setzero_si512(),

lib/include/hlsl.h deleted-15

...	...	@@ -1,15 +0,0 @@
1		//===----- hlsl.h - HLSL definitions --------------------------------------===//
2		//
3		// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4		// See https://llvm.org/LICENSE.txt for license information.
5		// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6		//
7		//===----------------------------------------------------------------------===//
8
9		#ifndef _HLSL_H_
10		#define _HLSL_H_
11
12		#include "hlsl/hlsl_basic_types.h"
13		#include "hlsl/hlsl_intrinsics.h"
14
15		#endif //_HLSL_H_

lib/include/hlsl_basic_types.h deleted-64

...	...	@@ -1,64 +0,0 @@
1		//===----- hlsl_basic_types.h - HLSL definitions for basic types ----------===//
2		//
3		// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4		// See https://llvm.org/LICENSE.txt for license information.
5		// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6		//
7		//===----------------------------------------------------------------------===//
8
9		#ifndef _HLSL_HLSL_BASIC_TYPES_H_
10		#define _HLSL_HLSL_BASIC_TYPES_H_
11
12		// built-in scalar data types:
13
14		#ifdef __HLSL_ENABLE_16_BIT
15		// 16-bit integer.
16		typedef unsigned short uint16_t;
17		typedef short int16_t;
18		#endif
19
20		// unsigned 32-bit integer.
21		typedef unsigned int uint;
22
23		// 64-bit integer.
24		typedef unsigned long uint64_t;
25		typedef long int64_t;
26
27		// built-in vector data types:
28
29		#ifdef __HLSL_ENABLE_16_BIT
30		typedef vector<int16_t, 2> int16_t2;
31		typedef vector<int16_t, 3> int16_t3;
32		typedef vector<int16_t, 4> int16_t4;
33		typedef vector<uint16_t, 2> uint16_t2;
34		typedef vector<uint16_t, 3> uint16_t3;
35		typedef vector<uint16_t, 4> uint16_t4;
36		#endif
37
38		typedef vector<int, 2> int2;
39		typedef vector<int, 3> int3;
40		typedef vector<int, 4> int4;
41		typedef vector<uint, 2> uint2;
42		typedef vector<uint, 3> uint3;
43		typedef vector<uint, 4> uint4;
44		typedef vector<int64_t, 2> int64_t2;
45		typedef vector<int64_t, 3> int64_t3;
46		typedef vector<int64_t, 4> int64_t4;
47		typedef vector<uint64_t, 2> uint64_t2;
48		typedef vector<uint64_t, 3> uint64_t3;
49		typedef vector<uint64_t, 4> uint64_t4;
50
51		#ifdef __HLSL_ENABLE_16_BIT
52		typedef vector<half, 2> half2;
53		typedef vector<half, 3> half3;
54		typedef vector<half, 4> half4;
55		#endif
56
57		typedef vector<float, 2> float2;
58		typedef vector<float, 3> float3;
59		typedef vector<float, 4> float4;
60		typedef vector<double, 2> double2;
61		typedef vector<double, 3> double3;
62		typedef vector<double, 4> double4;
63
64		#endif //_HLSL_HLSL_BASIC_TYPES_H_

lib/include/hlsl_intrinsics.h deleted-15

...	...	@@ -1,15 +0,0 @@
1		//===----- hlsl_intrinsics.h - HLSL definitions for intrinsics ----------===//
2		//
3		// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4		// See https://llvm.org/LICENSE.txt for license information.
5		// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6		//
7		//===----------------------------------------------------------------------===//
8
9		#ifndef _HLSL_HLSL_INTRINSICS_H_
10		#define _HLSL_HLSL_INTRINSICS_H_
11
12		__attribute__((clang_builtin_alias(__builtin_hlsl_wave_active_count_bits))) uint
13		WaveActiveCountBits(bool bBit);
14
15		#endif //_HLSL_HLSL_INTRINSICS_H_

lib/include/immintrin.h+40-8

...	...	@@ -189,6 +189,11 @@
189	189	#include <avx512ifmavlintrin.h>
190	190	#endif
191	191
	192	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	193	defined(__AVXIFMA__)
	194	#include <avxifmaintrin.h>
	195	#endif
	196
192	197	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
193	198	defined(__AVX512VBMI__)
194	199	#include <avx512vbmiintrin.h>
...	...	@@ -214,17 +219,13 @@
214	219	#include <avx512pfintrin.h>
215	220	#endif
216	221
217		/*
218		* FIXME: _Float16 type is legal only when HW support float16 operation.
219		* We use __AVX512FP16__ to identify if float16 is supported or not, so
220		* when float16 is not supported, the related header is not included.
221		*
222		*/
223		#if defined(__AVX512FP16__)
	222	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	223	defined(__AVX512FP16__)
224	224	#include <avx512fp16intrin.h>
225	225	#endif
226	226
227		#if defined(__AVX512FP16__) && defined(__AVX512VL__)
	227	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	228	(defined(__AVX512VL__) && defined(__AVX512FP16__))
228	229	#include <avx512vlfp16intrin.h>
229	230	#endif
230	231
...	...	@@ -258,6 +259,16 @@
258	259	#include <gfniintrin.h>
259	260	#endif
260	261
	262	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	263	defined(__AVXVNNIINT8__)
	264	#include <avxvnniint8intrin.h>
	265	#endif
	266
	267	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	268	defined(__AVXNECONVERT__)
	269	#include <avxneconvertintrin.h>
	270	#endif
	271
261	272	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
262	273	defined(__RDPID__)
263	274	/// Returns the value of the IA32_TSC_AUX MSR (0xc0000103).
...	...	@@ -291,6 +302,23 @@ _rdrand64_step(unsigned long long *__p)
291	302	{
292	303	return (int)__builtin_ia32_rdrand64_step(__p);
293	304	}
	305	#else
	306	// We need to emulate the functionality of 64-bit rdrand with 2 32-bit
	307	// rdrand instructions.
	308	static __inline__ int __attribute__((__always_inline__, __nodebug__, __target__("rdrnd")))
	309	_rdrand64_step(unsigned long long *__p)
	310	{
	311	unsigned int __lo, __hi;
	312	unsigned int __res_lo = __builtin_ia32_rdrand32_step(&__lo);
	313	unsigned int __res_hi = __builtin_ia32_rdrand32_step(&__hi);
	314	if (__res_lo && __res_hi) {
	315	*__p = ((unsigned long long)__hi << 32) \| (unsigned long long)__lo;
	316	return 1;
	317	} else {
	318	*__p = 0;
	319	return 0;
	320	}
	321	}
294	322	#endif
295	323	#endif /* __RDRND__ */
296	324
...	...	@@ -495,6 +523,10 @@ _storebe_i64(void * __P, long long __D) {
495	523	defined(__INVPCID__)
496	524	#include <invpcidintrin.h>
497	525	#endif
	526	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	527	defined(__AMXFP16__)
	528	#include <amxfp16intrin.h>
	529	#endif
498	530
499	531	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
500	532	defined(__KL__) \|\| defined(__WIDEKL__)

lib/include/larchintrin.h created+234

...	...	@@ -0,0 +1,234 @@
	1	/*===------------ larchintrin.h - LoongArch intrinsics ---------------------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9
	10	#ifndef _LOONGARCH_BASE_INTRIN_H
	11	#define _LOONGARCH_BASE_INTRIN_H
	12
	13	#ifdef __cplusplus
	14	extern "C" {
	15	#endif
	16
	17	typedef struct rdtime {
	18	unsigned int value;
	19	unsigned int timeid;
	20	} __rdtime_t;
	21
	22	#if __loongarch_grlen == 64
	23	typedef struct drdtime {
	24	unsigned long dvalue;
	25	unsigned long dtimeid;
	26	} __drdtime_t;
	27
	28	extern __inline __drdtime_t
	29	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	30	__rdtime_d(void) {
	31	__drdtime_t __drdtime;
	32	__asm__ volatile(
	33	"rdtime.d %[val], %[tid]\n\t"
	34	: [val] "=&r"(__drdtime.dvalue), [tid] "=&r"(__drdtime.dtimeid));
	35	return __drdtime;
	36	}
	37	#endif
	38
	39	extern __inline __rdtime_t
	40	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	41	__rdtimeh_w(void) {
	42	__rdtime_t __rdtime;
	43	__asm__ volatile("rdtimeh.w %[val], %[tid]\n\t"
	44	: [val] "=&r"(__rdtime.value), [tid] "=&r"(__rdtime.timeid));
	45	return __rdtime;
	46	}
	47
	48	extern __inline __rdtime_t
	49	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	50	__rdtimel_w(void) {
	51	__rdtime_t __rdtime;
	52	__asm__ volatile("rdtimel.w %[val], %[tid]\n\t"
	53	: [val] "=&r"(__rdtime.value), [tid] "=&r"(__rdtime.timeid));
	54	return __rdtime;
	55	}
	56
	57	#if __loongarch_grlen == 64
	58	extern __inline int
	59	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	60	__crc_w_b_w(char _1, int _2) {
	61	return (int)__builtin_loongarch_crc_w_b_w((char)_1, (int)_2);
	62	}
	63
	64	extern __inline int
	65	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	66	__crc_w_h_w(short _1, int _2) {
	67	return (int)__builtin_loongarch_crc_w_h_w((short)_1, (int)_2);
	68	}
	69
	70	extern __inline int
	71	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	72	__crc_w_w_w(int _1, int _2) {
	73	return (int)__builtin_loongarch_crc_w_w_w((int)_1, (int)_2);
	74	}
	75
	76	extern __inline int
	77	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	78	__crc_w_d_w(long int _1, int _2) {
	79	return (int)__builtin_loongarch_crc_w_d_w((long int)_1, (int)_2);
	80	}
	81
	82	extern __inline int
	83	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	84	__crcc_w_b_w(char _1, int _2) {
	85	return (int)__builtin_loongarch_crcc_w_b_w((char)_1, (int)_2);
	86	}
	87
	88	extern __inline int
	89	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	90	__crcc_w_h_w(short _1, int _2) {
	91	return (int)__builtin_loongarch_crcc_w_h_w((short)_1, (int)_2);
	92	}
	93
	94	extern __inline int
	95	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	96	__crcc_w_w_w(int _1, int _2) {
	97	return (int)__builtin_loongarch_crcc_w_w_w((int)_1, (int)_2);
	98	}
	99
	100	extern __inline int
	101	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	102	__crcc_w_d_w(long int _1, int _2) {
	103	return (int)__builtin_loongarch_crcc_w_d_w((long int)_1, (int)_2);
	104	}
	105	#endif
	106
	107	#define __break(/ui15/ _1) __builtin_loongarch_break((_1))
	108
	109	#if __loongarch_grlen == 32
	110	#define __cacop_w(/uimm5/ _1, /unsigned int/ _2, /simm12/ _3) \
	111	((void)__builtin_loongarch_cacop_w((_1), (unsigned int)(_2), (_3)))
	112	#endif
	113
	114	#if __loongarch_grlen == 64
	115	#define __cacop_d(/uimm5/ _1, /unsigned long int/ _2, /simm12/ _3) \
	116	((void)__builtin_loongarch_cacop_d((_1), (unsigned long int)(_2), (_3)))
	117	#endif
	118
	119	#define __dbar(/ui15/ _1) __builtin_loongarch_dbar((_1))
	120
	121	#define __ibar(/ui15/ _1) __builtin_loongarch_ibar((_1))
	122
	123	#define __movfcsr2gr(/ui5/ _1) __builtin_loongarch_movfcsr2gr((_1));
	124
	125	#define __movgr2fcsr(/ui5/ _1, _2) \
	126	__builtin_loongarch_movgr2fcsr((_1), (unsigned int)_2);
	127
	128	#define __syscall(/ui15/ _1) __builtin_loongarch_syscall((_1))
	129
	130	#define __csrrd_w(/ui14/ _1) ((unsigned int)__builtin_loongarch_csrrd_w((_1)))
	131
	132	#define __csrwr_w(/unsigned int/ _1, /ui14/ _2) \
	133	((unsigned int)__builtin_loongarch_csrwr_w((unsigned int)(_1), (_2)))
	134
	135	#define __csrxchg_w(/unsigned int/ _1, /unsigned int/ _2, /ui14/ _3) \
	136	((unsigned int)__builtin_loongarch_csrxchg_w((unsigned int)(_1), \
	137	(unsigned int)(_2), (_3)))
	138
	139	#if __loongarch_grlen == 64
	140	#define __csrrd_d(/ui14/ _1) \
	141	((unsigned long int)__builtin_loongarch_csrrd_d((_1)))
	142
	143	#define __csrwr_d(/unsigned long int/ _1, /ui14/ _2) \
	144	((unsigned long int)__builtin_loongarch_csrwr_d((unsigned long int)(_1), \
	145	(_2)))
	146
	147	#define __csrxchg_d(/unsigned long int/ _1, /unsigned long int/ _2, \
	148	/ui14/ _3) \
	149	((unsigned long int)__builtin_loongarch_csrxchg_d( \
	150	(unsigned long int)(_1), (unsigned long int)(_2), (_3)))
	151	#endif
	152
	153	extern __inline unsigned char
	154	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	155	__iocsrrd_b(unsigned int _1) {
	156	return (unsigned char)__builtin_loongarch_iocsrrd_b((unsigned int)_1);
	157	}
	158
	159	extern __inline unsigned char
	160	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	161	__iocsrrd_h(unsigned int _1) {
	162	return (unsigned short)__builtin_loongarch_iocsrrd_h((unsigned int)_1);
	163	}
	164
	165	extern __inline unsigned int
	166	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	167	__iocsrrd_w(unsigned int _1) {
	168	return (unsigned int)__builtin_loongarch_iocsrrd_w((unsigned int)_1);
	169	}
	170
	171	#if __loongarch_grlen == 64
	172	extern __inline unsigned long int
	173	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	174	__iocsrrd_d(unsigned int _1) {
	175	return (unsigned long int)__builtin_loongarch_iocsrrd_d((unsigned int)_1);
	176	}
	177	#endif
	178
	179	extern __inline void
	180	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	181	__iocsrwr_b(unsigned char _1, unsigned int _2) {
	182	__builtin_loongarch_iocsrwr_b((unsigned char)_1, (unsigned int)_2);
	183	}
	184
	185	extern __inline void
	186	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	187	__iocsrwr_h(unsigned short _1, unsigned int _2) {
	188	__builtin_loongarch_iocsrwr_h((unsigned short)_1, (unsigned int)_2);
	189	}
	190
	191	extern __inline void
	192	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	193	__iocsrwr_w(unsigned int _1, unsigned int _2) {
	194	__builtin_loongarch_iocsrwr_w((unsigned int)_1, (unsigned int)_2);
	195	}
	196
	197	extern __inline unsigned int
	198	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	199	__cpucfg(unsigned int _1) {
	200	return (unsigned int)__builtin_loongarch_cpucfg((unsigned int)_1);
	201	}
	202
	203	#if __loongarch_grlen == 64
	204	extern __inline void
	205	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	206	__iocsrwr_d(unsigned long int _1, unsigned int _2) {
	207	__builtin_loongarch_iocsrwr_d((unsigned long int)_1, (unsigned int)_2);
	208	}
	209
	210	extern __inline void
	211	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	212	__asrtgt_d(long int _1, long int _2) {
	213	__builtin_loongarch_asrtgt_d((long int)_1, (long int)_2);
	214	}
	215
	216	extern __inline void
	217	__attribute__((__gnu_inline__, __always_inline__, __artificial__))
	218	__asrtle_d(long int _1, long int _2) {
	219	__builtin_loongarch_asrtle_d((long int)_1, (long int)_2);
	220	}
	221	#endif
	222
	223	#if __loongarch_grlen == 64
	224	#define __lddir_d(/long int/ _1, /ui5/ _2) \
	225	((long int)__builtin_loongarch_lddir_d((long int)(_1), (_2)))
	226
	227	#define __ldpte_d(/long int/ _1, /ui5/ _2) \
	228	((void)__builtin_loongarch_ldpte_d((long int)(_1), (_2)))
	229	#endif
	230
	231	#ifdef __cplusplus
	232	}
	233	#endif
	234	#endif /* _LOONGARCH_BASE_INTRIN_H */

lib/include/limits.h+3-2

...	...	@@ -65,7 +65,7 @@
65	65	/* C2x 5.2.4.2.1 */
66	66	/* FIXME: This is using the placeholder dates Clang produces for these macros
67	67	in C2x mode; switch to the correct values once they've been published. */
68		#if __STDC_VERSION__ >= 202000L
	68	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
69	69	#define BOOL_WIDTH __BOOL_WIDTH__
70	70	#define CHAR_WIDTH CHAR_BIT
71	71	#define SCHAR_WIDTH CHAR_BIT
...	...	@@ -93,7 +93,8 @@
93	93	/* C99 5.2.4.2.1: Added long long.
94	94	C++11 18.3.3.2: same contents as the Standard C Library header <limits.h>.
95	95	*/
96		#if __STDC_VERSION__ >= 199901L \|\| __cplusplus >= 201103L
	96	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) \|\| \
	97	(defined(__cplusplus) && __cplusplus >= 201103L)
97	98
98	99	#undef LLONG_MIN
99	100	#undef LLONG_MAX

lib/include/opencl-c-base.h+19

...	...	@@ -74,6 +74,25 @@
74	74	#define __opencl_c_atomic_scope_all_devices 1
75	75	#define __opencl_c_read_write_images 1
76	76	#endif // defined(__SPIR__)
	77
	78	// Undefine any feature macros that have been explicitly disabled using
	79	// an __undef_<feature> macro.
	80	#ifdef __undef___opencl_c_work_group_collective_functions
	81	#undef __opencl_c_work_group_collective_functions
	82	#endif
	83	#ifdef __undef___opencl_c_atomic_order_seq_cst
	84	#undef __opencl_c_atomic_order_seq_cst
	85	#endif
	86	#ifdef __undef___opencl_c_atomic_scope_device
	87	#undef __opencl_c_atomic_scope_device
	88	#endif
	89	#ifdef __undef___opencl_c_atomic_scope_all_devices
	90	#undef __opencl_c_atomic_scope_all_devices
	91	#endif
	92	#ifdef __undef___opencl_c_read_write_images
	93	#undef __opencl_c_read_write_images
	94	#endif
	95
77	96	#endif // (__OPENCL_CPP_VERSION__ == 202100 \|\| __OPENCL_C_VERSION__ == 300)
78	97
79	98	#if !defined(__opencl_c_generic_address_space)

lib/include/opencl-c.h+175-145

...	...	@@ -12396,11 +12396,11 @@ void __ovld vstorea_half16_rtn(double16, size_t, __private half *);
12396	12396	* image objects and then want to read the updated data.
12397	12397	*/
12398	12398
12399		void __ovld __conv barrier(cl_mem_fence_flags flags);
	12399	void __ovld __conv barrier(cl_mem_fence_flags);
12400	12400
12401	12401	#if defined(__OPENCL_CPP_VERSION__) \|\| (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
12402		void __ovld __conv work_group_barrier(cl_mem_fence_flags flags, memory_scope);
12403		void __ovld __conv work_group_barrier(cl_mem_fence_flags flags);
	12402	void __ovld __conv work_group_barrier(cl_mem_fence_flags, memory_scope);
	12403	void __ovld __conv work_group_barrier(cl_mem_fence_flags);
12404	12404	#endif //defined(__OPENCL_CPP_VERSION__) \|\| (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
12405	12405
12406	12406	// OpenCL v1.1 s6.11.9, v1.2 s6.12.9 - Explicit Memory Fence Functions
...	...	@@ -12418,7 +12418,7 @@ void __ovld __conv work_group_barrier(cl_mem_fence_flags flags);
12418	12418	* CLK_LOCAL_MEM_FENCE
12419	12419	* CLK_GLOBAL_MEM_FENCE.
12420	12420	*/
12421		void __ovld mem_fence(cl_mem_fence_flags flags);
	12421	void __ovld mem_fence(cl_mem_fence_flags);
12422	12422
12423	12423	/**
12424	12424	* Read memory barrier that orders only
...	...	@@ -12430,7 +12430,7 @@ void __ovld mem_fence(cl_mem_fence_flags flags);
12430	12430	* CLK_LOCAL_MEM_FENCE
12431	12431	* CLK_GLOBAL_MEM_FENCE.
12432	12432	*/
12433		void __ovld read_mem_fence(cl_mem_fence_flags flags);
	12433	void __ovld read_mem_fence(cl_mem_fence_flags);
12434	12434
12435	12435	/**
12436	12436	* Write memory barrier that orders only
...	...	@@ -12442,7 +12442,7 @@ void __ovld read_mem_fence(cl_mem_fence_flags flags);
12442	12442	* CLK_LOCAL_MEM_FENCE
12443	12443	* CLK_GLOBAL_MEM_FENCE.
12444	12444	*/
12445		void __ovld write_mem_fence(cl_mem_fence_flags flags);
	12445	void __ovld write_mem_fence(cl_mem_fence_flags);
12446	12446
12447	12447	// OpenCL v2.0 s6.13.9 - Address Space Qualifier Functions
12448	12448
...	...	@@ -12891,29 +12891,29 @@ void __ovld prefetch(const __global half16 *, size_t);
12891	12891	* (old + val) and store result at location
12892	12892	* pointed by p. The function returns old.
12893	12893	*/
12894		int __ovld atomic_add(volatile __global int *p, int val);
12895		uint __ovld atomic_add(volatile __global uint *p, uint val);
12896		int __ovld atomic_add(volatile __local int *p, int val);
12897		uint __ovld atomic_add(volatile __local uint *p, uint val);
	12894	int __ovld atomic_add(volatile __global int *, int);
	12895	uint __ovld atomic_add(volatile __global uint *, uint);
	12896	int __ovld atomic_add(volatile __local int *, int);
	12897	uint __ovld atomic_add(volatile __local uint *, uint);
12898	12898	#ifdef __OPENCL_CPP_VERSION__
12899		int __ovld atomic_add(volatile int *p, int val);
12900		uint __ovld atomic_add(volatile uint *p, uint val);
	12899	int __ovld atomic_add(volatile int *, int);
	12900	uint __ovld atomic_add(volatile uint *, uint);
12901	12901	#endif
12902	12902
12903	12903	#if defined(cl_khr_global_int32_base_atomics)
12904		int __ovld atom_add(volatile __global int *p, int val);
12905		uint __ovld atom_add(volatile __global uint *p, uint val);
	12904	int __ovld atom_add(volatile __global int *, int);
	12905	uint __ovld atom_add(volatile __global uint *, uint);
12906	12906	#endif
12907	12907	#if defined(cl_khr_local_int32_base_atomics)
12908		int __ovld atom_add(volatile __local int *p, int val);
12909		uint __ovld atom_add(volatile __local uint *p, uint val);
	12908	int __ovld atom_add(volatile __local int *, int);
	12909	uint __ovld atom_add(volatile __local uint *, uint);
12910	12910	#endif
12911	12911
12912	12912	#if defined(cl_khr_int64_base_atomics)
12913		long __ovld atom_add(volatile __global long *p, long val);
12914		ulong __ovld atom_add(volatile __global ulong *p, ulong val);
12915		long __ovld atom_add(volatile __local long *p, long val);
12916		ulong __ovld atom_add(volatile __local ulong *p, ulong val);
	12913	long __ovld atom_add(volatile __global long *, long);
	12914	ulong __ovld atom_add(volatile __global ulong *, ulong);
	12915	long __ovld atom_add(volatile __local long *, long);
	12916	ulong __ovld atom_add(volatile __local ulong *, ulong);
12917	12917	#endif
12918	12918
12919	12919	/**
...	...	@@ -12921,29 +12921,29 @@ ulong __ovld atom_add(volatile __local ulong *p, ulong val);
12921	12921	* Compute (old - val) and store result at location pointed by p. The function
12922	12922	* returns old.
12923	12923	*/
12924		int __ovld atomic_sub(volatile __global int *p, int val);
12925		uint __ovld atomic_sub(volatile __global uint *p, uint val);
12926		int __ovld atomic_sub(volatile __local int *p, int val);
12927		uint __ovld atomic_sub(volatile __local uint *p, uint val);
	12924	int __ovld atomic_sub(volatile __global int *, int);
	12925	uint __ovld atomic_sub(volatile __global uint *, uint);
	12926	int __ovld atomic_sub(volatile __local int *, int);
	12927	uint __ovld atomic_sub(volatile __local uint *, uint);
12928	12928	#ifdef __OPENCL_CPP_VERSION__
12929		int __ovld atomic_sub(volatile int *p, int val);
12930		uint __ovld atomic_sub(volatile uint *p, uint val);
	12929	int __ovld atomic_sub(volatile int *, int);
	12930	uint __ovld atomic_sub(volatile uint *, uint);
12931	12931	#endif
12932	12932
12933	12933	#if defined(cl_khr_global_int32_base_atomics)
12934		int __ovld atom_sub(volatile __global int *p, int val);
12935		uint __ovld atom_sub(volatile __global uint *p, uint val);
	12934	int __ovld atom_sub(volatile __global int *, int);
	12935	uint __ovld atom_sub(volatile __global uint *, uint);
12936	12936	#endif
12937	12937	#if defined(cl_khr_local_int32_base_atomics)
12938		int __ovld atom_sub(volatile __local int *p, int val);
12939		uint __ovld atom_sub(volatile __local uint *p, uint val);
	12938	int __ovld atom_sub(volatile __local int *, int);
	12939	uint __ovld atom_sub(volatile __local uint *, uint);
12940	12940	#endif
12941	12941
12942	12942	#if defined(cl_khr_int64_base_atomics)
12943		long __ovld atom_sub(volatile __global long *p, long val);
12944		ulong __ovld atom_sub(volatile __global ulong *p, ulong val);
12945		long __ovld atom_sub(volatile __local long *p, long val);
12946		ulong __ovld atom_sub(volatile __local ulong *p, ulong val);
	12943	long __ovld atom_sub(volatile __global long *, long);
	12944	ulong __ovld atom_sub(volatile __global ulong *, ulong);
	12945	long __ovld atom_sub(volatile __local long *, long);
	12946	ulong __ovld atom_sub(volatile __local ulong *, ulong);
12947	12947	#endif
12948	12948
12949	12949	/**
...	...	@@ -12951,32 +12951,32 @@ ulong __ovld atom_sub(volatile __local ulong *p, ulong val);
12951	12951	* with new value given by val. Returns old
12952	12952	* value.
12953	12953	*/
12954		int __ovld atomic_xchg(volatile __global int *p, int val);
12955		uint __ovld atomic_xchg(volatile __global uint *p, uint val);
12956		int __ovld atomic_xchg(volatile __local int *p, int val);
12957		uint __ovld atomic_xchg(volatile __local uint *p, uint val);
12958		float __ovld atomic_xchg(volatile __global float *p, float val);
12959		float __ovld atomic_xchg(volatile __local float *p, float val);
	12954	int __ovld atomic_xchg(volatile __global int *, int);
	12955	uint __ovld atomic_xchg(volatile __global uint *, uint);
	12956	int __ovld atomic_xchg(volatile __local int *, int);
	12957	uint __ovld atomic_xchg(volatile __local uint *, uint);
	12958	float __ovld atomic_xchg(volatile __global float *, float);
	12959	float __ovld atomic_xchg(volatile __local float *, float);
12960	12960	#ifdef __OPENCL_CPP_VERSION__
12961		int __ovld atomic_xchg(volatile int *p, int val);
12962		uint __ovld atomic_xchg(volatile uint *p, uint val);
12963		float __ovld atomic_xchg(volatile float *p, float val);
	12961	int __ovld atomic_xchg(volatile int *, int);
	12962	uint __ovld atomic_xchg(volatile uint *, uint);
	12963	float __ovld atomic_xchg(volatile float *, float);
12964	12964	#endif
12965	12965
12966	12966	#if defined(cl_khr_global_int32_base_atomics)
12967		int __ovld atom_xchg(volatile __global int *p, int val);
12968		uint __ovld atom_xchg(volatile __global uint *p, uint val);
	12967	int __ovld atom_xchg(volatile __global int *, int);
	12968	uint __ovld atom_xchg(volatile __global uint *, uint);
12969	12969	#endif
12970	12970	#if defined(cl_khr_local_int32_base_atomics)
12971		int __ovld atom_xchg(volatile __local int *p, int val);
12972		uint __ovld atom_xchg(volatile __local uint *p, uint val);
	12971	int __ovld atom_xchg(volatile __local int *, int);
	12972	uint __ovld atom_xchg(volatile __local uint *, uint);
12973	12973	#endif
12974	12974
12975	12975	#if defined(cl_khr_int64_base_atomics)
12976		long __ovld atom_xchg(volatile __global long *p, long val);
12977		long __ovld atom_xchg(volatile __local long *p, long val);
12978		ulong __ovld atom_xchg(volatile __global ulong *p, ulong val);
12979		ulong __ovld atom_xchg(volatile __local ulong *p, ulong val);
	12976	long __ovld atom_xchg(volatile __global long *, long);
	12977	long __ovld atom_xchg(volatile __local long *, long);
	12978	ulong __ovld atom_xchg(volatile __global ulong *, ulong);
	12979	ulong __ovld atom_xchg(volatile __local ulong *, ulong);
12980	12980	#endif
12981	12981
12982	12982	/**
...	...	@@ -13048,29 +13048,29 @@ ulong __ovld atom_dec(volatile __local ulong *);
13048	13048	* location pointed by p. The function
13049	13049	* returns old.
13050	13050	*/
13051		int __ovld atomic_cmpxchg(volatile __global int *p, int cmp, int val);
13052		uint __ovld atomic_cmpxchg(volatile __global uint *p, uint cmp, uint val);
13053		int __ovld atomic_cmpxchg(volatile __local int *p, int cmp, int val);
13054		uint __ovld atomic_cmpxchg(volatile __local uint *p, uint cmp, uint val);
	13051	int __ovld atomic_cmpxchg(volatile __global int *, int, int);
	13052	uint __ovld atomic_cmpxchg(volatile __global uint *, uint, uint);
	13053	int __ovld atomic_cmpxchg(volatile __local int *, int, int);
	13054	uint __ovld atomic_cmpxchg(volatile __local uint *, uint, uint);
13055	13055	#ifdef __OPENCL_CPP_VERSION__
13056		int __ovld atomic_cmpxchg(volatile int *p, int cmp, int val);
13057		uint __ovld atomic_cmpxchg(volatile uint *p, uint cmp, uint val);
	13056	int __ovld atomic_cmpxchg(volatile int *, int, int);
	13057	uint __ovld atomic_cmpxchg(volatile uint *, uint, uint);
13058	13058	#endif
13059	13059
13060	13060	#if defined(cl_khr_global_int32_base_atomics)
13061		int __ovld atom_cmpxchg(volatile __global int *p, int cmp, int val);
13062		uint __ovld atom_cmpxchg(volatile __global uint *p, uint cmp, uint val);
	13061	int __ovld atom_cmpxchg(volatile __global int *, int, int);
	13062	uint __ovld atom_cmpxchg(volatile __global uint *, uint, uint);
13063	13063	#endif
13064	13064	#if defined(cl_khr_local_int32_base_atomics)
13065		int __ovld atom_cmpxchg(volatile __local int *p, int cmp, int val);
13066		uint __ovld atom_cmpxchg(volatile __local uint *p, uint cmp, uint val);
	13065	int __ovld atom_cmpxchg(volatile __local int *, int, int);
	13066	uint __ovld atom_cmpxchg(volatile __local uint *, uint, uint);
13067	13067	#endif
13068	13068
13069	13069	#if defined(cl_khr_int64_base_atomics)
13070		long __ovld atom_cmpxchg(volatile __global long *p, long cmp, long val);
13071		ulong __ovld atom_cmpxchg(volatile __global ulong *p, ulong cmp, ulong val);
13072		long __ovld atom_cmpxchg(volatile __local long *p, long cmp, long val);
13073		ulong __ovld atom_cmpxchg(volatile __local ulong *p, ulong cmp, ulong val);
	13070	long __ovld atom_cmpxchg(volatile __global long *, long, long);
	13071	ulong __ovld atom_cmpxchg(volatile __global ulong *, ulong, ulong);
	13072	long __ovld atom_cmpxchg(volatile __local long *, long, long);
	13073	ulong __ovld atom_cmpxchg(volatile __local ulong *, ulong, ulong);
13074	13074	#endif
13075	13075
13076	13076	/**
...	...	@@ -13080,29 +13080,29 @@ ulong __ovld atom_cmpxchg(volatile __local ulong *p, ulong cmp, ulong val);
13080	13080	* location pointed by p. The function
13081	13081	* returns old.
13082	13082	*/
13083		int __ovld atomic_min(volatile __global int *p, int val);
13084		uint __ovld atomic_min(volatile __global uint *p, uint val);
13085		int __ovld atomic_min(volatile __local int *p, int val);
13086		uint __ovld atomic_min(volatile __local uint *p, uint val);
	13083	int __ovld atomic_min(volatile __global int *, int);
	13084	uint __ovld atomic_min(volatile __global uint *, uint);
	13085	int __ovld atomic_min(volatile __local int *, int);
	13086	uint __ovld atomic_min(volatile __local uint *, uint);
13087	13087	#ifdef __OPENCL_CPP_VERSION__
13088		int __ovld atomic_min(volatile int *p, int val);
13089		uint __ovld atomic_min(volatile uint *p, uint val);
	13088	int __ovld atomic_min(volatile int *, int);
	13089	uint __ovld atomic_min(volatile uint *, uint);
13090	13090	#endif
13091	13091
13092	13092	#if defined(cl_khr_global_int32_extended_atomics)
13093		int __ovld atom_min(volatile __global int *p, int val);
13094		uint __ovld atom_min(volatile __global uint *p, uint val);
	13093	int __ovld atom_min(volatile __global int *, int);
	13094	uint __ovld atom_min(volatile __global uint *, uint);
13095	13095	#endif
13096	13096	#if defined(cl_khr_local_int32_extended_atomics)
13097		int __ovld atom_min(volatile __local int *p, int val);
13098		uint __ovld atom_min(volatile __local uint *p, uint val);
	13097	int __ovld atom_min(volatile __local int *, int);
	13098	uint __ovld atom_min(volatile __local uint *, uint);
13099	13099	#endif
13100	13100
13101	13101	#if defined(cl_khr_int64_extended_atomics)
13102		long __ovld atom_min(volatile __global long *p, long val);
13103		ulong __ovld atom_min(volatile __global ulong *p, ulong val);
13104		long __ovld atom_min(volatile __local long *p, long val);
13105		ulong __ovld atom_min(volatile __local ulong *p, ulong val);
	13102	long __ovld atom_min(volatile __global long *, long);
	13103	ulong __ovld atom_min(volatile __global ulong *, ulong);
	13104	long __ovld atom_min(volatile __local long *, long);
	13105	ulong __ovld atom_min(volatile __local ulong *, ulong);
13106	13106	#endif
13107	13107
13108	13108	/**
...	...	@@ -13112,29 +13112,29 @@ ulong __ovld atom_min(volatile __local ulong *p, ulong val);
13112	13112	* location pointed by p. The function
13113	13113	* returns old.
13114	13114	*/
13115		int __ovld atomic_max(volatile __global int *p, int val);
13116		uint __ovld atomic_max(volatile __global uint *p, uint val);
13117		int __ovld atomic_max(volatile __local int *p, int val);
13118		uint __ovld atomic_max(volatile __local uint *p, uint val);
	13115	int __ovld atomic_max(volatile __global int *, int);
	13116	uint __ovld atomic_max(volatile __global uint *, uint);
	13117	int __ovld atomic_max(volatile __local int *, int);
	13118	uint __ovld atomic_max(volatile __local uint *, uint);
13119	13119	#ifdef __OPENCL_CPP_VERSION__
13120		int __ovld atomic_max(volatile int *p, int val);
13121		uint __ovld atomic_max(volatile uint *p, uint val);
	13120	int __ovld atomic_max(volatile int *, int);
	13121	uint __ovld atomic_max(volatile uint *, uint);
13122	13122	#endif
13123	13123
13124	13124	#if defined(cl_khr_global_int32_extended_atomics)
13125		int __ovld atom_max(volatile __global int *p, int val);
13126		uint __ovld atom_max(volatile __global uint *p, uint val);
	13125	int __ovld atom_max(volatile __global int *, int);
	13126	uint __ovld atom_max(volatile __global uint *, uint);
13127	13127	#endif
13128	13128	#if defined(cl_khr_local_int32_extended_atomics)
13129		int __ovld atom_max(volatile __local int *p, int val);
13130		uint __ovld atom_max(volatile __local uint *p, uint val);
	13129	int __ovld atom_max(volatile __local int *, int);
	13130	uint __ovld atom_max(volatile __local uint *, uint);
13131	13131	#endif
13132	13132
13133	13133	#if defined(cl_khr_int64_extended_atomics)
13134		long __ovld atom_max(volatile __global long *p, long val);
13135		ulong __ovld atom_max(volatile __global ulong *p, ulong val);
13136		long __ovld atom_max(volatile __local long *p, long val);
13137		ulong __ovld atom_max(volatile __local ulong *p, ulong val);
	13134	long __ovld atom_max(volatile __global long *, long);
	13135	ulong __ovld atom_max(volatile __global ulong *, ulong);
	13136	long __ovld atom_max(volatile __local long *, long);
	13137	ulong __ovld atom_max(volatile __local ulong *, ulong);
13138	13138	#endif
13139	13139
13140	13140	/**
...	...	@@ -13143,29 +13143,29 @@ ulong __ovld atom_max(volatile __local ulong *p, ulong val);
13143	13143	* (old & val) and store result at location
13144	13144	* pointed by p. The function returns old.
13145	13145	*/
13146		int __ovld atomic_and(volatile __global int *p, int val);
13147		uint __ovld atomic_and(volatile __global uint *p, uint val);
13148		int __ovld atomic_and(volatile __local int *p, int val);
13149		uint __ovld atomic_and(volatile __local uint *p, uint val);
	13146	int __ovld atomic_and(volatile __global int *, int);
	13147	uint __ovld atomic_and(volatile __global uint *, uint);
	13148	int __ovld atomic_and(volatile __local int *, int);
	13149	uint __ovld atomic_and(volatile __local uint *, uint);
13150	13150	#ifdef __OPENCL_CPP_VERSION__
13151		int __ovld atomic_and(volatile int *p, int val);
13152		uint __ovld atomic_and(volatile uint *p, uint val);
	13151	int __ovld atomic_and(volatile int *, int);
	13152	uint __ovld atomic_and(volatile uint *, uint);
13153	13153	#endif
13154	13154
13155	13155	#if defined(cl_khr_global_int32_extended_atomics)
13156		int __ovld atom_and(volatile __global int *p, int val);
13157		uint __ovld atom_and(volatile __global uint *p, uint val);
	13156	int __ovld atom_and(volatile __global int *, int);
	13157	uint __ovld atom_and(volatile __global uint *, uint);
13158	13158	#endif
13159	13159	#if defined(cl_khr_local_int32_extended_atomics)
13160		int __ovld atom_and(volatile __local int *p, int val);
13161		uint __ovld atom_and(volatile __local uint *p, uint val);
	13160	int __ovld atom_and(volatile __local int *, int);
	13161	uint __ovld atom_and(volatile __local uint *, uint);
13162	13162	#endif
13163	13163
13164	13164	#if defined(cl_khr_int64_extended_atomics)
13165		long __ovld atom_and(volatile __global long *p, long val);
13166		ulong __ovld atom_and(volatile __global ulong *p, ulong val);
13167		long __ovld atom_and(volatile __local long *p, long val);
13168		ulong __ovld atom_and(volatile __local ulong *p, ulong val);
	13165	long __ovld atom_and(volatile __global long *, long);
	13166	ulong __ovld atom_and(volatile __global ulong *, ulong);
	13167	long __ovld atom_and(volatile __local long *, long);
	13168	ulong __ovld atom_and(volatile __local ulong *, ulong);
13169	13169	#endif
13170	13170
13171	13171	/**
...	...	@@ -13174,29 +13174,29 @@ ulong __ovld atom_and(volatile __local ulong *p, ulong val);
13174	13174	* (old \| val) and store result at location
13175	13175	* pointed by p. The function returns old.
13176	13176	*/
13177		int __ovld atomic_or(volatile __global int *p, int val);
13178		uint __ovld atomic_or(volatile __global uint *p, uint val);
13179		int __ovld atomic_or(volatile __local int *p, int val);
13180		uint __ovld atomic_or(volatile __local uint *p, uint val);
	13177	int __ovld atomic_or(volatile __global int *, int);
	13178	uint __ovld atomic_or(volatile __global uint *, uint);
	13179	int __ovld atomic_or(volatile __local int *, int);
	13180	uint __ovld atomic_or(volatile __local uint *, uint);
13181	13181	#ifdef __OPENCL_CPP_VERSION__
13182		int __ovld atomic_or(volatile int *p, int val);
13183		uint __ovld atomic_or(volatile uint *p, uint val);
	13182	int __ovld atomic_or(volatile int *, int);
	13183	uint __ovld atomic_or(volatile uint *, uint);
13184	13184	#endif
13185	13185
13186	13186	#if defined(cl_khr_global_int32_extended_atomics)
13187		int __ovld atom_or(volatile __global int *p, int val);
13188		uint __ovld atom_or(volatile __global uint *p, uint val);
	13187	int __ovld atom_or(volatile __global int *, int);
	13188	uint __ovld atom_or(volatile __global uint *, uint);
13189	13189	#endif
13190	13190	#if defined(cl_khr_local_int32_extended_atomics)
13191		int __ovld atom_or(volatile __local int *p, int val);
13192		uint __ovld atom_or(volatile __local uint *p, uint val);
	13191	int __ovld atom_or(volatile __local int *, int);
	13192	uint __ovld atom_or(volatile __local uint *, uint);
13193	13193	#endif
13194	13194
13195	13195	#if defined(cl_khr_int64_extended_atomics)
13196		long __ovld atom_or(volatile __global long *p, long val);
13197		ulong __ovld atom_or(volatile __global ulong *p, ulong val);
13198		long __ovld atom_or(volatile __local long *p, long val);
13199		ulong __ovld atom_or(volatile __local ulong *p, ulong val);
	13196	long __ovld atom_or(volatile __global long *, long);
	13197	ulong __ovld atom_or(volatile __global ulong *, ulong);
	13198	long __ovld atom_or(volatile __local long *, long);
	13199	ulong __ovld atom_or(volatile __local ulong *, ulong);
13200	13200	#endif
13201	13201
13202	13202	/**
...	...	@@ -13205,29 +13205,29 @@ ulong __ovld atom_or(volatile __local ulong *p, ulong val);
13205	13205	* (old ^ val) and store result at location
13206	13206	* pointed by p. The function returns old.
13207	13207	*/
13208		int __ovld atomic_xor(volatile __global int *p, int val);
13209		uint __ovld atomic_xor(volatile __global uint *p, uint val);
13210		int __ovld atomic_xor(volatile __local int *p, int val);
13211		uint __ovld atomic_xor(volatile __local uint *p, uint val);
	13208	int __ovld atomic_xor(volatile __global int *, int);
	13209	uint __ovld atomic_xor(volatile __global uint *, uint);
	13210	int __ovld atomic_xor(volatile __local int *, int);
	13211	uint __ovld atomic_xor(volatile __local uint *, uint);
13212	13212	#ifdef __OPENCL_CPP_VERSION__
13213		int __ovld atomic_xor(volatile int *p, int val);
13214		uint __ovld atomic_xor(volatile uint *p, uint val);
	13213	int __ovld atomic_xor(volatile int *, int);
	13214	uint __ovld atomic_xor(volatile uint *, uint);
13215	13215	#endif
13216	13216
13217	13217	#if defined(cl_khr_global_int32_extended_atomics)
13218		int __ovld atom_xor(volatile __global int *p, int val);
13219		uint __ovld atom_xor(volatile __global uint *p, uint val);
	13218	int __ovld atom_xor(volatile __global int *, int);
	13219	uint __ovld atom_xor(volatile __global uint *, uint);
13220	13220	#endif
13221	13221	#if defined(cl_khr_local_int32_extended_atomics)
13222		int __ovld atom_xor(volatile __local int *p, int val);
13223		uint __ovld atom_xor(volatile __local uint *p, uint val);
	13222	int __ovld atom_xor(volatile __local int *, int);
	13223	uint __ovld atom_xor(volatile __local uint *, uint);
13224	13224	#endif
13225	13225
13226	13226	#if defined(cl_khr_int64_extended_atomics)
13227		long __ovld atom_xor(volatile __global long *p, long val);
13228		ulong __ovld atom_xor(volatile __global ulong *p, ulong val);
13229		long __ovld atom_xor(volatile __local long *p, long val);
13230		ulong __ovld atom_xor(volatile __local ulong *p, ulong val);
	13227	long __ovld atom_xor(volatile __global long *, long);
	13228	ulong __ovld atom_xor(volatile __global ulong *, ulong);
	13229	long __ovld atom_xor(volatile __local long *, long);
	13230	ulong __ovld atom_xor(volatile __local ulong *, ulong);
13231	13231	#endif
13232	13232
13233	13233	#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
...	...	@@ -15257,13 +15257,17 @@ float4 __ovld __purefn read_imagef(read_only image2d_t, sampler_t, float2, float
15257	15257	int4 __ovld __purefn read_imagei(read_only image2d_t, sampler_t, float2, float);
15258	15258	uint4 __ovld __purefn read_imageui(read_only image2d_t, sampler_t, float2, float);
15259	15259
	15260	#ifdef cl_khr_depth_images
15260	15261	float __ovld __purefn read_imagef(read_only image2d_depth_t, sampler_t, float2, float);
	15262	#endif // cl_khr_depth_images
15261	15263
15262	15264	float4 __ovld __purefn read_imagef(read_only image2d_array_t, sampler_t, float4, float);
15263	15265	int4 __ovld __purefn read_imagei(read_only image2d_array_t, sampler_t, float4, float);
15264	15266	uint4 __ovld __purefn read_imageui(read_only image2d_array_t, sampler_t, float4, float);
15265	15267
	15268	#ifdef cl_khr_depth_images
15266	15269	float __ovld __purefn read_imagef(read_only image2d_array_depth_t, sampler_t, float4, float);
	15270	#endif // cl_khr_depth_images
15267	15271
15268	15272	float4 __ovld __purefn read_imagef(read_only image3d_t, sampler_t, float4, float);
15269	15273	int4 __ovld __purefn read_imagei(read_only image3d_t, sampler_t, float4, float);
...	...	@@ -15281,13 +15285,17 @@ float4 __ovld __purefn read_imagef(read_only image2d_t, sampler_t, float2, float
15281	15285	int4 __ovld __purefn read_imagei(read_only image2d_t, sampler_t, float2, float2, float2);
15282	15286	uint4 __ovld __purefn read_imageui(read_only image2d_t, sampler_t, float2, float2, float2);
15283	15287
	15288	#ifdef cl_khr_depth_images
15284	15289	float __ovld __purefn read_imagef(read_only image2d_depth_t, sampler_t, float2, float2, float2);
	15290	#endif // cl_khr_depth_images
15285	15291
15286	15292	float4 __ovld __purefn read_imagef(read_only image2d_array_t, sampler_t, float4, float2, float2);
15287	15293	int4 __ovld __purefn read_imagei(read_only image2d_array_t, sampler_t, float4, float2, float2);
15288	15294	uint4 __ovld __purefn read_imageui(read_only image2d_array_t, sampler_t, float4, float2, float2);
15289	15295
	15296	#ifdef cl_khr_depth_images
15290	15297	float __ovld __purefn read_imagef(read_only image2d_array_depth_t, sampler_t, float4, float2, float2);
	15298	#endif // cl_khr_depth_images
15291	15299
15292	15300	float4 __ovld __purefn read_imagef(read_only image3d_t, sampler_t, float4, float4, float4);
15293	15301	int4 __ovld __purefn read_imagei(read_only image3d_t, sampler_t, float4, float4, float4);
...	...	@@ -15380,9 +15388,11 @@ float4 __ovld __purefn read_imagef(read_write image2d_array_t, int4);
15380	15388	int4 __ovld __purefn read_imagei(read_write image2d_array_t, int4);
15381	15389	uint4 __ovld __purefn read_imageui(read_write image2d_array_t, int4);
15382	15390
	15391	#ifdef cl_khr_3d_image_writes
15383	15392	float4 __ovld __purefn read_imagef(read_write image3d_t, int4);
15384	15393	int4 __ovld __purefn read_imagei(read_write image3d_t, int4);
15385	15394	uint4 __ovld __purefn read_imageui(read_write image3d_t, int4);
	15395	#endif // cl_khr_3d_image_writes
15386	15396
15387	15397	#ifdef cl_khr_depth_images
15388	15398	float __ovld __purefn read_imagef(read_write image2d_depth_t, int2);
...	...	@@ -15423,9 +15433,11 @@ uint4 __ovld __purefn read_imageui(read_write image2d_array_t, sampler_t, float4
15423	15433
15424	15434	float __ovld __purefn read_imagef(read_write image2d_array_depth_t, sampler_t, float4, float);
15425	15435
	15436	#ifdef cl_khr_3d_image_writes
15426	15437	float4 __ovld __purefn read_imagef(read_write image3d_t, sampler_t, float4, float);
15427	15438	int4 __ovld __purefn read_imagei(read_write image3d_t, sampler_t, float4, float);
15428	15439	uint4 __ovld __purefn read_imageui(read_write image3d_t, sampler_t, float4, float);
	15440	#endif // cl_khr_3d_image_writes
15429	15441
15430	15442	float4 __ovld __purefn read_imagef(read_write image1d_t, sampler_t, float, float, float);
15431	15443	int4 __ovld __purefn read_imagei(read_write image1d_t, sampler_t, float, float, float);
...	...	@@ -15447,9 +15459,11 @@ uint4 __ovld __purefn read_imageui(read_write image2d_array_t, sampler_t, float4
15447	15459
15448	15460	float __ovld __purefn read_imagef(read_write image2d_array_depth_t, sampler_t, float4, float2, float2);
15449	15461
	15462	#ifdef cl_khr_3d_image_writes
15450	15463	float4 __ovld __purefn read_imagef(read_write image3d_t, sampler_t, float4, float4, float4);
15451	15464	int4 __ovld __purefn read_imagei(read_write image3d_t, sampler_t, float4, float4, float4);
15452	15465	uint4 __ovld __purefn read_imageui(read_write image3d_t, sampler_t, float4, float4, float4);
	15466	#endif // cl_khr_3d_image_writes
15453	15467
15454	15468	#endif //cl_khr_mipmap_image
15455	15469
...	...	@@ -15457,7 +15471,9 @@ uint4 __ovld __purefn read_imageui(read_write image3d_t, sampler_t, float4, floa
15457	15471	#ifdef cl_khr_fp16
15458	15472	half4 __ovld __purefn read_imageh(read_write image1d_t, int);
15459	15473	half4 __ovld __purefn read_imageh(read_write image2d_t, int2);
	15474	#ifdef cl_khr_3d_image_writes
15460	15475	half4 __ovld __purefn read_imageh(read_write image3d_t, int4);
	15476	#endif // cl_khr_3d_image_writes
15461	15477	half4 __ovld __purefn read_imageh(read_write image1d_array_t, int2);
15462	15478	half4 __ovld __purefn read_imageh(read_write image2d_array_t, int4);
15463	15479	half4 __ovld __purefn read_imageh(read_write image1d_buffer_t, int);
...	...	@@ -15727,7 +15743,9 @@ int __ovld __cnfn get_image_width(write_only image2d_array_msaa_depth_t);
15727	15743	int __ovld __cnfn get_image_width(read_write image1d_t);
15728	15744	int __ovld __cnfn get_image_width(read_write image1d_buffer_t);
15729	15745	int __ovld __cnfn get_image_width(read_write image2d_t);
	15746	#ifdef cl_khr_3d_image_writes
15730	15747	int __ovld __cnfn get_image_width(read_write image3d_t);
	15748	#endif // cl_khr_3d_image_writes
15731	15749	int __ovld __cnfn get_image_width(read_write image1d_array_t);
15732	15750	int __ovld __cnfn get_image_width(read_write image2d_array_t);
15733	15751	#ifdef cl_khr_depth_images
...	...	@@ -15777,7 +15795,9 @@ int __ovld __cnfn get_image_height(write_only image2d_array_msaa_depth_t);
15777	15795
15778	15796	#if defined(__opencl_c_read_write_images)
15779	15797	int __ovld __cnfn get_image_height(read_write image2d_t);
	15798	#ifdef cl_khr_3d_image_writes
15780	15799	int __ovld __cnfn get_image_height(read_write image3d_t);
	15800	#endif // cl_khr_3d_image_writes
15781	15801	int __ovld __cnfn get_image_height(read_write image2d_array_t);
15782	15802	#ifdef cl_khr_depth_images
15783	15803	int __ovld __cnfn get_image_height(read_write image2d_depth_t);
...	...	@@ -15798,11 +15818,11 @@ int __ovld __cnfn get_image_depth(read_only image3d_t);
15798	15818
15799	15819	#ifdef cl_khr_3d_image_writes
15800	15820	int __ovld __cnfn get_image_depth(write_only image3d_t);
15801		#endif
15802	15821
15803	15822	#if defined(__opencl_c_read_write_images)
15804	15823	int __ovld __cnfn get_image_depth(read_write image3d_t);
15805	15824	#endif //defined(__opencl_c_read_write_images)
	15825	#endif // cl_khr_3d_image_writes
15806	15826
15807	15827	// OpenCL Extension v2.0 s9.18 - Mipmaps
15808	15828	#if defined(__OPENCL_CPP_VERSION__) \|\| (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
...	...	@@ -15824,24 +15844,32 @@ int __ovld get_image_num_mip_levels(write_only image3d_t);
15824	15844	#if defined(__opencl_c_read_write_images)
15825	15845	int __ovld get_image_num_mip_levels(read_write image1d_t);
15826	15846	int __ovld get_image_num_mip_levels(read_write image2d_t);
	15847	#ifdef cl_khr_3d_image_writes
15827	15848	int __ovld get_image_num_mip_levels(read_write image3d_t);
	15849	#endif // cl_khr_3d_image_writes
15828	15850	#endif //defined(__opencl_c_read_write_images)
15829	15851
15830	15852	int __ovld get_image_num_mip_levels(read_only image1d_array_t);
15831	15853	int __ovld get_image_num_mip_levels(read_only image2d_array_t);
	15854	#ifdef cl_khr_depth_images
15832	15855	int __ovld get_image_num_mip_levels(read_only image2d_array_depth_t);
15833	15856	int __ovld get_image_num_mip_levels(read_only image2d_depth_t);
	15857	#endif // cl_khr_depth_images
15834	15858
15835	15859	int __ovld get_image_num_mip_levels(write_only image1d_array_t);
15836	15860	int __ovld get_image_num_mip_levels(write_only image2d_array_t);
	15861	#ifdef cl_khr_depth_images
15837	15862	int __ovld get_image_num_mip_levels(write_only image2d_array_depth_t);
15838	15863	int __ovld get_image_num_mip_levels(write_only image2d_depth_t);
	15864	#endif // cl_khr_depth_images
15839	15865
15840	15866	#if defined(__opencl_c_read_write_images)
15841	15867	int __ovld get_image_num_mip_levels(read_write image1d_array_t);
15842	15868	int __ovld get_image_num_mip_levels(read_write image2d_array_t);
	15869	#ifdef cl_khr_depth_images
15843	15870	int __ovld get_image_num_mip_levels(read_write image2d_array_depth_t);
15844	15871	int __ovld get_image_num_mip_levels(read_write image2d_depth_t);
	15872	#endif // cl_khr_depth_images
15845	15873	#endif //defined(__opencl_c_read_write_images)
15846	15874
15847	15875	#endif //cl_khr_mipmap_image
...	...	@@ -15906,7 +15934,9 @@ int __ovld __cnfn get_image_channel_data_type(write_only image2d_array_msaa_dept
15906	15934	int __ovld __cnfn get_image_channel_data_type(read_write image1d_t);
15907	15935	int __ovld __cnfn get_image_channel_data_type(read_write image1d_buffer_t);
15908	15936	int __ovld __cnfn get_image_channel_data_type(read_write image2d_t);
	15937	#ifdef cl_khr_3d_image_writes
15909	15938	int __ovld __cnfn get_image_channel_data_type(read_write image3d_t);
	15939	#endif // cl_khr_3d_image_writes
15910	15940	int __ovld __cnfn get_image_channel_data_type(read_write image1d_array_t);
15911	15941	int __ovld __cnfn get_image_channel_data_type(read_write image2d_array_t);
15912	15942	#ifdef cl_khr_depth_images
...	...	@@ -15978,7 +16008,9 @@ int __ovld __cnfn get_image_channel_order(write_only image2d_array_msaa_depth_t)
15978	16008	int __ovld __cnfn get_image_channel_order(read_write image1d_t);
15979	16009	int __ovld __cnfn get_image_channel_order(read_write image1d_buffer_t);
15980	16010	int __ovld __cnfn get_image_channel_order(read_write image2d_t);
	16011	#ifdef cl_khr_3d_image_writes
15981	16012	int __ovld __cnfn get_image_channel_order(read_write image3d_t);
	16013	#endif // cl_khr_3d_image_writes
15982	16014	int __ovld __cnfn get_image_channel_order(read_write image1d_array_t);
15983	16015	int __ovld __cnfn get_image_channel_order(read_write image2d_array_t);
15984	16016	#ifdef cl_khr_depth_images
...	...	@@ -16048,10 +16080,10 @@ int2 __ovld __cnfn get_image_dim(read_write image2d_array_msaa_depth_t);
16048	16080	int4 __ovld __cnfn get_image_dim(read_only image3d_t);
16049	16081	#ifdef cl_khr_3d_image_writes
16050	16082	int4 __ovld __cnfn get_image_dim(write_only image3d_t);
16051		#endif
16052	16083	#if defined(__opencl_c_read_write_images)
16053	16084	int4 __ovld __cnfn get_image_dim(read_write image3d_t);
16054	16085	#endif //defined(__opencl_c_read_write_images)
	16086	#endif // cl_khr_3d_image_writes
16055	16087
16056	16088	/**
16057	16089	* Return the image array size.
...	...	@@ -16266,9 +16298,9 @@ uint __ovld get_enqueued_num_sub_groups(void);
16266	16298	uint __ovld get_sub_group_id(void);
16267	16299	uint __ovld get_sub_group_local_id(void);
16268	16300
16269		void __ovld __conv sub_group_barrier(cl_mem_fence_flags flags);
	16301	void __ovld __conv sub_group_barrier(cl_mem_fence_flags);
16270	16302	#if defined(__OPENCL_CPP_VERSION__) \|\| (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
16271		void __ovld __conv sub_group_barrier(cl_mem_fence_flags flags, memory_scope);
	16303	void __ovld __conv sub_group_barrier(cl_mem_fence_flags, memory_scope);
16272	16304	#endif //defined(__OPENCL_CPP_VERSION__) \|\| (__OPENCL_C_VERSION__ >= CL_VERSION_2_0)
16273	16305
16274	16306	int __ovld __conv sub_group_all(int predicate);
...	...	@@ -17847,15 +17879,13 @@ intel_sub_group_avc_sic_configure_skc(
17847	17879	uint skip_block_partition_type, uint skip_motion_vector_mask,
17848	17880	ulong motion_vectors, uchar bidirectional_weight, uchar skip_sad_adjustment,
17849	17881	intel_sub_group_avc_sic_payload_t payload);
17850		intel_sub_group_avc_sic_payload_t __ovld
17851		intel_sub_group_avc_sic_configure_ipe(
17852		uchar luma_intra_partition_mask, uchar intra_neighbour_availabilty,
	17882	intel_sub_group_avc_sic_payload_t __ovld intel_sub_group_avc_sic_configure_ipe(
	17883	uchar luma_intra_partition_mask, uchar intra_neighbour_availability,
17853	17884	uchar left_edge_luma_pixels, uchar upper_left_corner_luma_pixel,
17854	17885	uchar upper_edge_luma_pixels, uchar upper_right_edge_luma_pixels,
17855	17886	uchar intra_sad_adjustment, intel_sub_group_avc_sic_payload_t payload);
17856		intel_sub_group_avc_sic_payload_t __ovld
17857		intel_sub_group_avc_sic_configure_ipe(
17858		uchar luma_intra_partition_mask, uchar intra_neighbour_availabilty,
	17887	intel_sub_group_avc_sic_payload_t __ovld intel_sub_group_avc_sic_configure_ipe(
	17888	uchar luma_intra_partition_mask, uchar intra_neighbour_availability,
17859	17889	uchar left_edge_luma_pixels, uchar upper_left_corner_luma_pixel,
17860	17890	uchar upper_edge_luma_pixels, uchar upper_right_edge_luma_pixels,
17861	17891	ushort left_edge_chroma_pixels, ushort upper_left_corner_chroma_pixel,

lib/include/ppc_wrappers/emmintrin.h+2-2

...	...	@@ -36,7 +36,7 @@
36	36	#ifndef EMMINTRIN_H_
37	37	#define EMMINTRIN_H_
38	38
39		#if defined(__ppc64__) && \
	39	#if defined(__powerpc64__) && \
40	40	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
41	41
42	42	#include <altivec.h>
...	...	@@ -2262,7 +2262,7 @@ extern __inline __m128d
2262	2262
2263	2263	#else
2264	2264	#include_next <emmintrin.h>
2265		#endif /* defined(__ppc64__) &&
	2265	#endif /* defined(__powerpc64__) && \
2266	2266	* (defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX)) */
2267	2267
2268	2268	#endif /* EMMINTRIN_H_ */

lib/include/ppc_wrappers/mm_malloc.h+1-1

...	...	@@ -10,7 +10,7 @@
10	10	#ifndef _MM_MALLOC_H_INCLUDED
11	11	#define _MM_MALLOC_H_INCLUDED
12	12
13		#if defined(__ppc64__) && \
	13	#if defined(__powerpc64__) && \
14	14	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
15	15
16	16	#include <stdlib.h>

lib/include/ppc_wrappers/mmintrin.h+2-2

...	...	@@ -35,7 +35,7 @@
35	35	#ifndef _MMINTRIN_H_INCLUDED
36	36	#define _MMINTRIN_H_INCLUDED
37	37
38		#if defined(__ppc64__) && \
	38	#if defined(__powerpc64__) && \
39	39	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
40	40
41	41	#include <altivec.h>
...	...	@@ -1447,7 +1447,7 @@ extern __inline __m64
1447	1447
1448	1448	#else
1449	1449	#include_next <mmintrin.h>
1450		#endif /* defined(__ppc64__) &&
	1450	#endif /* defined(__powerpc64__) && \
1451	1451	* (defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX)) */
1452	1452
1453	1453	#endif /* _MMINTRIN_H_INCLUDED */

lib/include/ppc_wrappers/pmmintrin.h+2-2

...	...	@@ -39,7 +39,7 @@
39	39	#ifndef PMMINTRIN_H_
40	40	#define PMMINTRIN_H_
41	41
42		#if defined(__ppc64__) && \
	42	#if defined(__powerpc64__) && \
43	43	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
44	44
45	45	/* We need definitions from the SSE2 and SSE header files*/
...	...	@@ -139,7 +139,7 @@ extern __inline __m128i
139	139
140	140	#else
141	141	#include_next <pmmintrin.h>
142		#endif /* defined(__ppc64__) &&
	142	#endif /* defined(__powerpc64__) && \
143	143	* (defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX)) */
144	144
145	145	#endif /* PMMINTRIN_H_ */

lib/include/ppc_wrappers/smmintrin.h+2-2

...	...	@@ -29,7 +29,7 @@
29	29	#ifndef SMMINTRIN_H_
30	30	#define SMMINTRIN_H_
31	31
32		#if defined(__ppc64__) && \
	32	#if defined(__powerpc64__) && \
33	33	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
34	34
35	35	#include <altivec.h>
...	...	@@ -657,7 +657,7 @@ extern __inline __m128i
657	657
658	658	#else
659	659	#include_next <smmintrin.h>
660		#endif /* defined(__ppc64__) &&
	660	#endif /* defined(__powerpc64__) && \
661	661	* (defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX)) */
662	662
663	663	#endif /* SMMINTRIN_H_ */

lib/include/ppc_wrappers/tmmintrin.h+2-2

...	...	@@ -25,7 +25,7 @@
25	25	#ifndef TMMINTRIN_H_
26	26	#define TMMINTRIN_H_
27	27
28		#if defined(__ppc64__) && \
	28	#if defined(__powerpc64__) && \
29	29	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
30	30
31	31	#include <altivec.h>
...	...	@@ -447,7 +447,7 @@ extern __inline __m64
447	447
448	448	#else
449	449	#include_next <tmmintrin.h>
450		#endif /* defined(__ppc64__) &&
	450	#endif /* defined(__powerpc64__) && \
451	451	* (defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX)) */
452	452
453	453	#endif /* TMMINTRIN_H_ */

lib/include/ppc_wrappers/xmmintrin.h+2-2

...	...	@@ -35,7 +35,7 @@
35	35	#ifndef XMMINTRIN_H_
36	36	#define XMMINTRIN_H_
37	37
38		#if defined(__ppc64__) && \
	38	#if defined(__powerpc64__) && \
39	39	(defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX))
40	40
41	41	/* Define four value permute mask */
...	...	@@ -1821,7 +1821,7 @@ extern __inline void
1821	1821
1822	1822	#else
1823	1823	#include_next <xmmintrin.h>
1824		#endif /* defined(__ppc64__) &&
	1824	#endif /* defined(__powerpc64__) && \
1825	1825	* (defined(__linux__) \|\| defined(__FreeBSD__) \|\| defined(_AIX)) */
1826	1826
1827	1827	#endif /* XMMINTRIN_H_ */

lib/include/prfchiintrin.h created+61

...	...	@@ -0,0 +1,61 @@
	1	/*===---- prfchiintrin.h - PREFETCHI intrinsic -----------------------------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9
	10	#ifndef __PRFCHIINTRIN_H
	11	#define __PRFCHIINTRIN_H
	12
	13	#ifdef __x86_64__
	14
	15	/* Define the default attributes for the functions in this file. */
	16	#define __DEFAULT_FN_ATTRS \
	17	__attribute__((__always_inline__, __nodebug__, __target__("prefetchi")))
	18
	19	/// Loads an instruction sequence containing the specified memory address into
	20	/// all level cache.
	21	///
	22	/// Note that the effect of this intrinsic is dependent on the processor
	23	/// implementation.
	24	///
	25	/// \headerfile <x86intrin.h>
	26	///
	27	/// This intrinsic corresponds to the \c PREFETCHIT0 instruction.
	28	///
	29	/// \param __P
	30	/// A pointer specifying the memory address to be prefetched.
	31	static __inline__ void __DEFAULT_FN_ATTRS
	32	_m_prefetchit0(volatile const void *__P) {
	33	#pragma clang diagnostic push
	34	#pragma clang diagnostic ignored "-Wcast-qual"
	35	__builtin_ia32_prefetchi((const void )__P, 3 / _MM_HINT_T0 */);
	36	#pragma clang diagnostic pop
	37	}
	38
	39	/// Loads an instruction sequence containing the specified memory address into
	40	/// all but the first-level cache.
	41	///
	42	/// Note that the effect of this intrinsic is dependent on the processor
	43	/// implementation.
	44	///
	45	/// \headerfile <x86intrin.h>
	46	///
	47	/// This intrinsic corresponds to the \c PREFETCHIT1 instruction.
	48	///
	49	/// \param __P
	50	/// A pointer specifying the memory address to be prefetched.
	51	static __inline__ void __DEFAULT_FN_ATTRS
	52	_m_prefetchit1(volatile const void *__P) {
	53	#pragma clang diagnostic push
	54	#pragma clang diagnostic ignored "-Wcast-qual"
	55	__builtin_ia32_prefetchi((const void )__P, 2 / _MM_HINT_T1 */);
	56	#pragma clang diagnostic pop
	57	}
	58	#endif /* __x86_64__ */
	59	#undef __DEFAULT_FN_ATTRS
	60
	61	#endif /* __PRFCHWINTRIN_H */

lib/include/raointintrin.h created+203

...	...	@@ -0,0 +1,203 @@
	1	/*===----------------------- raointintrin.h - RAOINT ------------------------===
	2	*
	3	* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
	4	* See https://llvm.org/LICENSE.txt for license information.
	5	* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
	6	*
	7	*===-----------------------------------------------------------------------===
	8	*/
	9
	10	#ifndef __X86GPRINTRIN_H
	11	#error "Never use <raointintrin.h> directly; include <x86gprintrin.h> instead."
	12	#endif // __X86GPRINTRIN_H
	13
	14	#ifndef __RAOINTINTRIN_H
	15	#define __RAOINTINTRIN_H
	16
	17	#define __DEFAULT_FN_ATTRS \
	18	__attribute__((__always_inline__, __nodebug__, __target__("raoint")))
	19
	20	/// Atomically add a 32-bit value at memory operand \a __A and a 32-bit \a __B,
	21	/// and store the result to the same memory location.
	22	///
	23	/// This intrinsic should be used for contention or weak ordering. It may
	24	/// result in bad performance for hot data used by single thread only.
	25	///
	26	/// \headerfile <x86intrin.h>
	27	///
	28	/// This intrinsic corresponds to the \c AADD instruction.
	29	///
	30	/// \param __A
	31	/// A pointer to a 32-bit memory location.
	32	/// \param __B
	33	/// A 32-bit integer value.
	34	///
	35	/// \code{.operation}
	36	/// MEM[__A+31:__A] := MEM[__A+31:__A] + __B[31:0]
	37	/// \endcode
	38	static __inline__ void __DEFAULT_FN_ATTRS _aadd_i32(int *__A, int __B) {
	39	__builtin_ia32_aadd32((int *)__A, __B);
	40	}
	41
	42	/// Atomically and a 32-bit value at memory operand \a __A and a 32-bit \a __B,
	43	/// and store the result to the same memory location.
	44	///
	45	/// This intrinsic should be used for contention or weak ordering. It may
	46	/// result in bad performance for hot data used by single thread only.
	47	///
	48	/// \headerfile <x86intrin.h>
	49	///
	50	/// This intrinsic corresponds to the \c AAND instruction.
	51	///
	52	/// \param __A
	53	/// A pointer to a 32-bit memory location.
	54	/// \param __B
	55	/// A 32-bit integer value.
	56	///
	57	/// \code{.operation}
	58	/// MEM[__A+31:__A] := MEM[__A+31:__A] AND __B[31:0]
	59	/// \endcode
	60	static __inline__ void __DEFAULT_FN_ATTRS _aand_i32(int *__A, int __B) {
	61	__builtin_ia32_aand32((int *)__A, __B);
	62	}
	63
	64	/// Atomically or a 32-bit value at memory operand \a __A and a 32-bit \a __B,
	65	/// and store the result to the same memory location.
	66	///
	67	/// This intrinsic should be used for contention or weak ordering. It may
	68	/// result in bad performance for hot data used by single thread only.
	69	///
	70	/// \headerfile <x86intrin.h>
	71	///
	72	/// This intrinsic corresponds to the \c AOR instruction.
	73	///
	74	/// \param __A
	75	/// A pointer to a 32-bit memory location.
	76	/// \param __B
	77	/// A 32-bit integer value.
	78	///
	79	/// \code{.operation}
	80	/// MEM[__A+31:__A] := MEM[__A+31:__A] OR __B[31:0]
	81	/// \endcode
	82	static __inline__ void __DEFAULT_FN_ATTRS _aor_i32(int *__A, int __B) {
	83	__builtin_ia32_aor32((int *)__A, __B);
	84	}
	85
	86	/// Atomically xor a 32-bit value at memory operand \a __A and a 32-bit \a __B,
	87	/// and store the result to the same memory location.
	88	///
	89	/// This intrinsic should be used for contention or weak ordering. It may
	90	/// result in bad performance for hot data used by single thread only.
	91	///
	92	/// \headerfile <x86intrin.h>
	93	///
	94	/// This intrinsic corresponds to the \c AXOR instruction.
	95	///
	96	/// \param __A
	97	/// A pointer to a 32-bit memory location.
	98	/// \param __B
	99	/// A 32-bit integer value.
	100	///
	101	/// \code{.operation}
	102	/// MEM[__A+31:__A] := MEM[__A+31:__A] XOR __B[31:0]
	103	/// \endcode
	104	static __inline__ void __DEFAULT_FN_ATTRS _axor_i32(int *__A, int __B) {
	105	__builtin_ia32_axor32((int *)__A, __B);
	106	}
	107
	108	#ifdef __x86_64__
	109	/// Atomically add a 64-bit value at memory operand \a __A and a 64-bit \a __B,
	110	/// and store the result to the same memory location.
	111	///
	112	/// This intrinsic should be used for contention or weak ordering. It may
	113	/// result in bad performance for hot data used by single thread only.
	114	///
	115	/// \headerfile <x86intrin.h>
	116	///
	117	/// This intrinsic corresponds to the \c AADD instruction.
	118	///
	119	/// \param __A
	120	/// A pointer to a 64-bit memory location.
	121	/// \param __B
	122	/// A 64-bit integer value.
	123	///
	124	/// \code{.operation}
	125	/// MEM[__A+63:__A] := MEM[__A+63:__A] + __B[63:0]
	126	/// \endcode
	127	static __inline__ void __DEFAULT_FN_ATTRS _aadd_i64(long long *__A,
	128	long long __B) {
	129	__builtin_ia32_aadd64((long long *)__A, __B);
	130	}
	131
	132	/// Atomically and a 64-bit value at memory operand \a __A and a 64-bit \a __B,
	133	/// and store the result to the same memory location.
	134	///
	135	/// This intrinsic should be used for contention or weak ordering. It may
	136	/// result in bad performance for hot data used by single thread only.
	137	///
	138	/// \headerfile <x86intrin.h>
	139	///
	140	/// This intrinsic corresponds to the \c AAND instruction.
	141	///
	142	/// \param __A
	143	/// A pointer to a 64-bit memory location.
	144	/// \param __B
	145	/// A 64-bit integer value.
	146	///
	147	/// \code{.operation}
	148	/// MEM[__A+63:__A] := MEM[__A+63:__A] AND __B[63:0]
	149	/// \endcode
	150	static __inline__ void __DEFAULT_FN_ATTRS _aand_i64(long long *__A,
	151	long long __B) {
	152	__builtin_ia32_aand64((long long *)__A, __B);
	153	}
	154
	155	/// Atomically or a 64-bit value at memory operand \a __A and a 64-bit \a __B,
	156	/// and store the result to the same memory location.
	157	///
	158	/// This intrinsic should be used for contention or weak ordering. It may
	159	/// result in bad performance for hot data used by single thread only.
	160	///
	161	/// \headerfile <x86intrin.h>
	162	///
	163	/// This intrinsic corresponds to the \c AOR instruction.
	164	///
	165	/// \param __A
	166	/// A pointer to a 64-bit memory location.
	167	/// \param __B
	168	/// A 64-bit integer value.
	169	///
	170	/// \code{.operation}
	171	/// MEM[__A+63:__A] := MEM[__A+63:__A] OR __B[63:0]
	172	/// \endcode
	173	static __inline__ void __DEFAULT_FN_ATTRS _aor_i64(long long *__A,
	174	long long __B) {
	175	__builtin_ia32_aor64((long long *)__A, __B);
	176	}
	177
	178	/// Atomically xor a 64-bit value at memory operand \a __A and a 64-bit \a __B,
	179	/// and store the result to the same memory location.
	180	///
	181	/// This intrinsic should be used for contention or weak ordering. It may
	182	/// result in bad performance for hot data used by single thread only.
	183	///
	184	/// \headerfile <x86intrin.h>
	185	///
	186	/// This intrinsic corresponds to the \c AXOR instruction.
	187	///
	188	/// \param __A
	189	/// A pointer to a 64-bit memory location.
	190	/// \param __B
	191	/// A 64-bit integer value.
	192	///
	193	/// \code{.operation}
	194	/// MEM[__A+63:__A] := MEM[__A+63:__A] XOR __B[63:0]
	195	/// \endcode
	196	static __inline__ void __DEFAULT_FN_ATTRS _axor_i64(long long *__A,
	197	long long __B) {
	198	__builtin_ia32_axor64((long long *)__A, __B);
	199	}
	200	#endif // __x86_64__
	201
	202	#undef __DEFAULT_FN_ATTRS
	203	#endif // __RAOINTINTRIN_H

lib/include/riscv_vector.h+15-6

...	...	@@ -25,6 +25,8 @@ extern "C" {
25	25	#pragma clang riscv intrinsic vector
26	26
27	27
	28	#define vlenb() __builtin_rvv_vlenb()
	29
28	30	enum RVV_CSR {
29	31	RVV_VSTART = 0,
30	32	RVV_VXSAT,
...	...	@@ -70,7 +72,6 @@ void vwrite_csr(enum RVV_CSR __csr, unsigned long __value) {
70	72	}
71	73	}
72	74
73		#define vsetvl_e8mf8(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 5)
74	75	#define vsetvl_e8mf4(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 6)
75	76	#define vsetvl_e8mf2(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 7)
76	77	#define vsetvl_e8m1(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 0)
...	...	@@ -78,25 +79,28 @@ void vwrite_csr(enum RVV_CSR __csr, unsigned long __value) {
78	79	#define vsetvl_e8m4(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 2)
79	80	#define vsetvl_e8m8(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 3)
80	81
81		#define vsetvl_e16mf4(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 6)
82	82	#define vsetvl_e16mf2(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 7)
83	83	#define vsetvl_e16m1(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 0)
84	84	#define vsetvl_e16m2(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 1)
85	85	#define vsetvl_e16m4(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 2)
86	86	#define vsetvl_e16m8(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 3)
87	87
88		#define vsetvl_e32mf2(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 7)
89	88	#define vsetvl_e32m1(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 0)
90	89	#define vsetvl_e32m2(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 1)
91	90	#define vsetvl_e32m4(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 2)
92	91	#define vsetvl_e32m8(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 3)
93	92
	93	#if __riscv_v_elen >= 64
	94	#define vsetvl_e8mf8(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 5)
	95	#define vsetvl_e16mf4(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 6)
	96	#define vsetvl_e32mf2(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 7)
	97
94	98	#define vsetvl_e64m1(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 0)
95	99	#define vsetvl_e64m2(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 1)
96	100	#define vsetvl_e64m4(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 2)
97	101	#define vsetvl_e64m8(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 3)
	102	#endif
98	103
99		#define vsetvlmax_e8mf8() __builtin_rvv_vsetvlimax(0, 5)
100	104	#define vsetvlmax_e8mf4() __builtin_rvv_vsetvlimax(0, 6)
101	105	#define vsetvlmax_e8mf2() __builtin_rvv_vsetvlimax(0, 7)
102	106	#define vsetvlmax_e8m1() __builtin_rvv_vsetvlimax(0, 0)
...	...	@@ -104,23 +108,28 @@ void vwrite_csr(enum RVV_CSR __csr, unsigned long __value) {
104	108	#define vsetvlmax_e8m4() __builtin_rvv_vsetvlimax(0, 2)
105	109	#define vsetvlmax_e8m8() __builtin_rvv_vsetvlimax(0, 3)
106	110
107		#define vsetvlmax_e16mf4() __builtin_rvv_vsetvlimax(1, 6)
108	111	#define vsetvlmax_e16mf2() __builtin_rvv_vsetvlimax(1, 7)
109	112	#define vsetvlmax_e16m1() __builtin_rvv_vsetvlimax(1, 0)
110	113	#define vsetvlmax_e16m2() __builtin_rvv_vsetvlimax(1, 1)
111	114	#define vsetvlmax_e16m4() __builtin_rvv_vsetvlimax(1, 2)
112	115	#define vsetvlmax_e16m8() __builtin_rvv_vsetvlimax(1, 3)
113	116
114		#define vsetvlmax_e32mf2() __builtin_rvv_vsetvlimax(2, 7)
115	117	#define vsetvlmax_e32m1() __builtin_rvv_vsetvlimax(2, 0)
116	118	#define vsetvlmax_e32m2() __builtin_rvv_vsetvlimax(2, 1)
117	119	#define vsetvlmax_e32m4() __builtin_rvv_vsetvlimax(2, 2)
118	120	#define vsetvlmax_e32m8() __builtin_rvv_vsetvlimax(2, 3)
119	121
	122	#if __riscv_v_elen >= 64
	123	#define vsetvlmax_e8mf8() __builtin_rvv_vsetvlimax(0, 5)
	124	#define vsetvlmax_e16mf4() __builtin_rvv_vsetvlimax(1, 6)
	125	#define vsetvlmax_e32mf2() __builtin_rvv_vsetvlimax(2, 7)
	126
120	127	#define vsetvlmax_e64m1() __builtin_rvv_vsetvlimax(3, 0)
121	128	#define vsetvlmax_e64m2() __builtin_rvv_vsetvlimax(3, 1)
122	129	#define vsetvlmax_e64m4() __builtin_rvv_vsetvlimax(3, 2)
123	130	#define vsetvlmax_e64m8() __builtin_rvv_vsetvlimax(3, 3)
	131	#endif
	132
124	133	typedef __rvv_bool64_t vbool64_t;
125	134	typedef __rvv_bool32_t vbool32_t;
126	135	typedef __rvv_bool16_t vbool16_t;

lib/include/smmintrin.h+1-1

...	...	@@ -818,7 +818,7 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_max_epu32(__m128i __V1,
818	818	/// parameter, is copied to the result.
819	819	/// \param N
820	820	/// Specifies which bits from operand \a Y will be copied, which bits in the
821		/// result they will be be copied to, and which bits in the result will be
	821	/// result they will be copied to, and which bits in the result will be
822	822	/// cleared. The following assignments are made: \n
823	823	/// Bits [7:6] specify the bits to copy from operand \a Y: \n
824	824	/// 00: Selects bits [31:0] from operand \a Y. \n

lib/include/stdarg.h+23-7

...	...	@@ -8,13 +8,30 @@
8	8	*/
9	9
10	10	#ifndef __STDARG_H
11		#define __STDARG_H
12	11
	12	#ifndef __GNUC_VA_LIST
	13	#define __GNUC_VA_LIST
	14	typedef __builtin_va_list __gnuc_va_list;
	15	#endif
	16
	17	#ifdef __need___va_list
	18	#undef __need___va_list
	19	#else
	20	#define __STDARG_H
13	21	#ifndef _VA_LIST
14	22	typedef __builtin_va_list va_list;
15	23	#define _VA_LIST
16	24	#endif
	25
	26	/* FIXME: This is using the placeholder dates Clang produces for these macros
	27	in C2x mode; switch to the correct values once they've been published. */
	28	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
	29	/* C2x does not require the second parameter for va_start. */
	30	#define va_start(ap, ...) __builtin_va_start(ap, 0)
	31	#else
	32	/* Versions before C2x do require the second parameter. */
17	33	#define va_start(ap, param) __builtin_va_start(ap, param)
	34	#endif
18	35	#define va_end(ap) __builtin_va_end(ap)
19	36	#define va_arg(ap, type) __builtin_va_arg(ap, type)
20	37
...	...	@@ -23,13 +40,12 @@ typedef __builtin_va_list va_list;
23	40	*/
24	41	#define __va_copy(d,s) __builtin_va_copy(d,s)
25	42
26		#if __STDC_VERSION__ >= 199901L \|\| __cplusplus >= 201103L \|\| !defined(__STRICT_ANSI__)
	43	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) \|\| \
	44	(defined(__cplusplus) && __cplusplus >= 201103L) \|\| \
	45	!defined(__STRICT_ANSI__)
27	46	#define va_copy(dest, src) __builtin_va_copy(dest, src)
28	47	#endif
29	48
30		#ifndef __GNUC_VA_LIST
31		#define __GNUC_VA_LIST 1
32		typedef __builtin_va_list __gnuc_va_list;
33		#endif
34
35	49	#endif /* __STDARG_H */
	50
	51	#endif /* not __STDARG_H */

lib/include/stdatomic.h+6-3

...	...	@@ -15,10 +15,12 @@
15	15	*
16	16	* Exclude the MSVC path as well as the MSVC header as of the 14.31.30818
17	17	* explicitly disallows `stdatomic.h` in the C mode via an `#error`. Fallback
18		* to the clang resource header until that is fully supported.
	18	* to the clang resource header until that is fully supported. The
	19	* `stdatomic.h` header requires C++ 23 or newer.
19	20	*/
20	21	#if __STDC_HOSTED__ && \
21		__has_include_next(<stdatomic.h>) && !(defined(_MSC_VER) && !defined(__cplusplus))
	22	__has_include_next(<stdatomic.h>) && \
	23	(!defined(_MSC_VER) \|\| (defined(__cplusplus) && __cplusplus >= 202002L))
22	24	# include_next <stdatomic.h>
23	25	#else
24	26
...	...	@@ -45,7 +47,8 @@ extern "C" {
45	47	/* 7.17.2 Initialization */
46	48
47	49	#define ATOMIC_VAR_INIT(value) (value)
48		#if (__STDC_VERSION__ >= 201710L \|\| __cplusplus >= 202002L) && \
	50	#if ((defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201710L) \|\| \
	51	(defined(__cplusplus) && __cplusplus >= 202002L)) && \
49	52	!defined(_CLANG_DISABLE_CRT_DEPRECATION_WARNINGS)
50	53	/* ATOMIC_VAR_INIT was deprecated in C17 and C++20. */
51	54	#pragma clang deprecated(ATOMIC_VAR_INIT)

lib/include/stdbool.h+2-2

...	...	@@ -12,7 +12,7 @@
12	12
13	13	#define __bool_true_false_are_defined 1
14	14
15		#if __STDC_VERSION__ > 201710L
	15	#if defined(__STDC_VERSION__) && __STDC_VERSION__ > 201710L
16	16	/* FIXME: We should be issuing a deprecation warning here, but cannot yet due
17	17	* to system headers which include this header file unconditionally.
18	18	*/
...	...	@@ -23,7 +23,7 @@
23	23	#elif defined(__GNUC__) && !defined(__STRICT_ANSI__)
24	24	/* Define _Bool as a GNU extension. */
25	25	#define _Bool bool
26		#if __cplusplus < 201103L
	26	#if defined(__cplusplus) && __cplusplus < 201103L
27	27	/* For C++98, define bool, false, true as a GNU extension. */
28	28	#define bool bool
29	29	#define false false

lib/include/stddef.h+8-1

...	...	@@ -97,8 +97,15 @@ using ::std::nullptr_t;
97	97	#undef __need_NULL
98	98	#endif /* defined(__need_NULL) */
99	99
	100	/* FIXME: This is using the placeholder dates Clang produces for these macros
	101	in C2x mode; switch to the correct values once they've been published. */
	102	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
	103	typedef typeof(nullptr) nullptr_t;
	104	#endif /* defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L */
	105
100	106	#if defined(__need_STDDEF_H_misc)
101		#if __STDC_VERSION__ >= 201112L \|\| __cplusplus >= 201103L
	107	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L) \|\| \
	108	(defined(__cplusplus) && __cplusplus >= 201103L)
102	109	#include "__stddef_max_align_t.h"
103	110	#endif
104	111	#define offsetof(t, d) __builtin_offsetof(t, d)

lib/include/stdint.h+158-40

...	...	@@ -96,13 +96,21 @@
96	96	typedef __INT64_TYPE__ int64_t;
97	97	# endif /* __int8_t_defined */
98	98	typedef __UINT64_TYPE__ uint64_t;
	99	# undef __int_least64_t
99	100	# define __int_least64_t int64_t
	101	# undef __uint_least64_t
100	102	# define __uint_least64_t uint64_t
	103	# undef __int_least32_t
101	104	# define __int_least32_t int64_t
	105	# undef __uint_least32_t
102	106	# define __uint_least32_t uint64_t
	107	# undef __int_least16_t
103	108	# define __int_least16_t int64_t
	109	# undef __uint_least16_t
104	110	# define __uint_least16_t uint64_t
	111	# undef __int_least8_t
105	112	# define __int_least8_t int64_t
	113	# undef __uint_least8_t
106	114	# define __uint_least8_t uint64_t
107	115	#endif /* __INT64_TYPE__ */
108	116
...	...	@@ -120,11 +128,17 @@ typedef int56_t int_least56_t;
120	128	typedef uint56_t uint_least56_t;
121	129	typedef int56_t int_fast56_t;
122	130	typedef uint56_t uint_fast56_t;
	131	# undef __int_least32_t
123	132	# define __int_least32_t int56_t
	133	# undef __uint_least32_t
124	134	# define __uint_least32_t uint56_t
	135	# undef __int_least16_t
125	136	# define __int_least16_t int56_t
	137	# undef __uint_least16_t
126	138	# define __uint_least16_t uint56_t
	139	# undef __int_least8_t
127	140	# define __int_least8_t int56_t
	141	# undef __uint_least8_t
128	142	# define __uint_least8_t uint56_t
129	143	#endif /* __INT56_TYPE__ */
130	144
...	...	@@ -136,11 +150,17 @@ typedef int48_t int_least48_t;
136	150	typedef uint48_t uint_least48_t;
137	151	typedef int48_t int_fast48_t;
138	152	typedef uint48_t uint_fast48_t;
	153	# undef __int_least32_t
139	154	# define __int_least32_t int48_t
	155	# undef __uint_least32_t
140	156	# define __uint_least32_t uint48_t
	157	# undef __int_least16_t
141	158	# define __int_least16_t int48_t
	159	# undef __uint_least16_t
142	160	# define __uint_least16_t uint48_t
	161	# undef __int_least8_t
143	162	# define __int_least8_t int48_t
	163	# undef __uint_least8_t
144	164	# define __uint_least8_t uint48_t
145	165	#endif /* __INT48_TYPE__ */
146	166
...	...	@@ -152,11 +172,17 @@ typedef int40_t int_least40_t;
152	172	typedef uint40_t uint_least40_t;
153	173	typedef int40_t int_fast40_t;
154	174	typedef uint40_t uint_fast40_t;
	175	# undef __int_least32_t
155	176	# define __int_least32_t int40_t
	177	# undef __uint_least32_t
156	178	# define __uint_least32_t uint40_t
	179	# undef __int_least16_t
157	180	# define __int_least16_t int40_t
	181	# undef __uint_least16_t
158	182	# define __uint_least16_t uint40_t
	183	# undef __int_least8_t
159	184	# define __int_least8_t int40_t
	185	# undef __uint_least8_t
160	186	# define __uint_least8_t uint40_t
161	187	#endif /* __INT40_TYPE__ */
162	188
...	...	@@ -172,11 +198,17 @@ typedef __INT32_TYPE__ int32_t;
172	198	typedef __UINT32_TYPE__ uint32_t;
173	199	# endif /* __uint32_t_defined */
174	200
	201	# undef __int_least32_t
175	202	# define __int_least32_t int32_t
	203	# undef __uint_least32_t
176	204	# define __uint_least32_t uint32_t
	205	# undef __int_least16_t
177	206	# define __int_least16_t int32_t
	207	# undef __uint_least16_t
178	208	# define __uint_least16_t uint32_t
	209	# undef __int_least8_t
179	210	# define __int_least8_t int32_t
	211	# undef __uint_least8_t
180	212	# define __uint_least8_t uint32_t
181	213	#endif /* __INT32_TYPE__ */
182	214
...	...	@@ -194,9 +226,13 @@ typedef int24_t int_least24_t;
194	226	typedef uint24_t uint_least24_t;
195	227	typedef int24_t int_fast24_t;
196	228	typedef uint24_t uint_fast24_t;
	229	# undef __int_least16_t
197	230	# define __int_least16_t int24_t
	231	# undef __uint_least16_t
198	232	# define __uint_least16_t uint24_t
	233	# undef __int_least8_t
199	234	# define __int_least8_t int24_t
	235	# undef __uint_least8_t
200	236	# define __uint_least8_t uint24_t
201	237	#endif /* __INT24_TYPE__ */
202	238
...	...	@@ -205,9 +241,13 @@ typedef uint24_t uint_fast24_t;
205	241	typedef __INT16_TYPE__ int16_t;
206	242	#endif /* __int8_t_defined */
207	243	typedef __UINT16_TYPE__ uint16_t;
	244	# undef __int_least16_t
208	245	# define __int_least16_t int16_t
	246	# undef __uint_least16_t
209	247	# define __uint_least16_t uint16_t
	248	# undef __int_least8_t
210	249	# define __int_least8_t int16_t
	250	# undef __uint_least8_t
211	251	# define __uint_least8_t uint16_t
212	252	#endif /* __INT16_TYPE__ */
213	253
...	...	@@ -224,7 +264,9 @@ typedef __uint_least16_t uint_fast16_t;
224	264	typedef __INT8_TYPE__ int8_t;
225	265	#endif /* __int8_t_defined */
226	266	typedef __UINT8_TYPE__ uint8_t;
	267	# undef __int_least8_t
227	268	# define __int_least8_t int8_t
	269	# undef __uint_least8_t
228	270	# define __uint_least8_t uint8_t
229	271	#endif /* __INT8_TYPE__ */
230	272
...	...	@@ -285,16 +327,15 @@ typedef __UINTMAX_TYPE__ uintmax_t;
285	327
286	328
287	329	#ifdef __INT64_TYPE__
	330	# undef __int64_c_suffix
	331	# undef __int32_c_suffix
	332	# undef __int16_c_suffix
	333	# undef __int8_c_suffix
288	334	# ifdef __INT64_C_SUFFIX__
289	335	# define __int64_c_suffix __INT64_C_SUFFIX__
290	336	# define __int32_c_suffix __INT64_C_SUFFIX__
291	337	# define __int16_c_suffix __INT64_C_SUFFIX__
292	338	# define __int8_c_suffix __INT64_C_SUFFIX__
293		# else
294		# undef __int64_c_suffix
295		# undef __int32_c_suffix
296		# undef __int16_c_suffix
297		# undef __int8_c_suffix
298	339	# endif /* __INT64_C_SUFFIX__ */
299	340	#endif /* __INT64_TYPE__ */
300	341
...	...	@@ -310,6 +351,9 @@ typedef __UINTMAX_TYPE__ uintmax_t;
310	351
311	352
312	353	#ifdef __INT56_TYPE__
	354	# undef __int32_c_suffix
	355	# undef __int16_c_suffix
	356	# undef __int8_c_suffix
313	357	# ifdef __INT56_C_SUFFIX__
314	358	# define INT56_C(v) __int_c(v, __INT56_C_SUFFIX__)
315	359	# define UINT56_C(v) __uint_c(v, __INT56_C_SUFFIX__)
...	...	@@ -319,14 +363,14 @@ typedef __UINTMAX_TYPE__ uintmax_t;
319	363	# else
320	364	# define INT56_C(v) v
321	365	# define UINT56_C(v) v ## U
322		# undef __int32_c_suffix
323		# undef __int16_c_suffix
324		# undef __int8_c_suffix
325	366	# endif /* __INT56_C_SUFFIX__ */
326	367	#endif /* __INT56_TYPE__ */
327	368
328	369
329	370	#ifdef __INT48_TYPE__
	371	# undef __int32_c_suffix
	372	# undef __int16_c_suffix
	373	# undef __int8_c_suffix
330	374	# ifdef __INT48_C_SUFFIX__
331	375	# define INT48_C(v) __int_c(v, __INT48_C_SUFFIX__)
332	376	# define UINT48_C(v) __uint_c(v, __INT48_C_SUFFIX__)
...	...	@@ -336,14 +380,14 @@ typedef __UINTMAX_TYPE__ uintmax_t;
336	380	# else
337	381	# define INT48_C(v) v
338	382	# define UINT48_C(v) v ## U
339		# undef __int32_c_suffix
340		# undef __int16_c_suffix
341		# undef __int8_c_suffix
342	383	# endif /* __INT48_C_SUFFIX__ */
343	384	#endif /* __INT48_TYPE__ */
344	385
345	386
346	387	#ifdef __INT40_TYPE__
	388	# undef __int32_c_suffix
	389	# undef __int16_c_suffix
	390	# undef __int8_c_suffix
347	391	# ifdef __INT40_C_SUFFIX__
348	392	# define INT40_C(v) __int_c(v, __INT40_C_SUFFIX__)
349	393	# define UINT40_C(v) __uint_c(v, __INT40_C_SUFFIX__)
...	...	@@ -353,22 +397,18 @@ typedef __UINTMAX_TYPE__ uintmax_t;
353	397	# else
354	398	# define INT40_C(v) v
355	399	# define UINT40_C(v) v ## U
356		# undef __int32_c_suffix
357		# undef __int16_c_suffix
358		# undef __int8_c_suffix
359	400	# endif /* __INT40_C_SUFFIX__ */
360	401	#endif /* __INT40_TYPE__ */
361	402
362	403
363	404	#ifdef __INT32_TYPE__
	405	# undef __int32_c_suffix
	406	# undef __int16_c_suffix
	407	# undef __int8_c_suffix
364	408	# ifdef __INT32_C_SUFFIX__
365	409	# define __int32_c_suffix __INT32_C_SUFFIX__
366	410	# define __int16_c_suffix __INT32_C_SUFFIX__
367	411	# define __int8_c_suffix __INT32_C_SUFFIX__
368		#else
369		# undef __int32_c_suffix
370		# undef __int16_c_suffix
371		# undef __int8_c_suffix
372	412	# endif /* __INT32_C_SUFFIX__ */
373	413	#endif /* __INT32_TYPE__ */
374	414
...	...	@@ -384,6 +424,8 @@ typedef __UINTMAX_TYPE__ uintmax_t;
384	424
385	425
386	426	#ifdef __INT24_TYPE__
	427	# undef __int16_c_suffix
	428	# undef __int8_c_suffix
387	429	# ifdef __INT24_C_SUFFIX__
388	430	# define INT24_C(v) __int_c(v, __INT24_C_SUFFIX__)
389	431	# define UINT24_C(v) __uint_c(v, __INT24_C_SUFFIX__)
...	...	@@ -392,19 +434,16 @@ typedef __UINTMAX_TYPE__ uintmax_t;
392	434	# else
393	435	# define INT24_C(v) v
394	436	# define UINT24_C(v) v ## U
395		# undef __int16_c_suffix
396		# undef __int8_c_suffix
397	437	# endif /* __INT24_C_SUFFIX__ */
398	438	#endif /* __INT24_TYPE__ */
399	439
400	440
401	441	#ifdef __INT16_TYPE__
	442	# undef __int16_c_suffix
	443	# undef __int8_c_suffix
402	444	# ifdef __INT16_C_SUFFIX__
403	445	# define __int16_c_suffix __INT16_C_SUFFIX__
404	446	# define __int8_c_suffix __INT16_C_SUFFIX__
405		#else
406		# undef __int16_c_suffix
407		# undef __int8_c_suffix
408	447	# endif /* __INT16_C_SUFFIX__ */
409	448	#endif /* __INT16_TYPE__ */
410	449
...	...	@@ -420,10 +459,9 @@ typedef __UINTMAX_TYPE__ uintmax_t;
420	459
421	460
422	461	#ifdef __INT8_TYPE__
	462	# undef __int8_c_suffix
423	463	# ifdef __INT8_C_SUFFIX__
424	464	# define __int8_c_suffix __INT8_C_SUFFIX__
425		#else
426		# undef __int8_c_suffix
427	465	# endif /* __INT8_C_SUFFIX__ */
428	466	#endif /* __INT8_TYPE__ */
429	467
...	...	@@ -463,27 +501,39 @@ typedef __UINTMAX_TYPE__ uintmax_t;
463	501	# define UINT64_MAX UINT64_C(18446744073709551615)
464	502	/* FIXME: This is using the placeholder dates Clang produces for these macros
465	503	in C2x mode; switch to the correct values once they've been published. */
466		#if __STDC_VERSION__ >= 202000L
	504	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
467	505	# define UINT64_WIDTH 64
468	506	# define INT64_WIDTH UINT64_WIDTH
469	507
470	508	# define __UINT_LEAST64_WIDTH UINT64_WIDTH
	509	# undef __UINT_LEAST32_WIDTH
471	510	# define __UINT_LEAST32_WIDTH UINT64_WIDTH
	511	# undef __UINT_LEAST16_WIDTH
472	512	# define __UINT_LEAST16_WIDTH UINT64_WIDTH
	513	# undef __UINT_LEAST8_MAX
473	514	# define __UINT_LEAST8_MAX UINT64_MAX
474	515	#endif /* __STDC_VERSION__ */
475	516
476	517	# define __INT_LEAST64_MIN INT64_MIN
477	518	# define __INT_LEAST64_MAX INT64_MAX
478	519	# define __UINT_LEAST64_MAX UINT64_MAX
	520	# undef __INT_LEAST32_MIN
479	521	# define __INT_LEAST32_MIN INT64_MIN
	522	# undef __INT_LEAST32_MAX
480	523	# define __INT_LEAST32_MAX INT64_MAX
	524	# undef __UINT_LEAST32_MAX
481	525	# define __UINT_LEAST32_MAX UINT64_MAX
	526	# undef __INT_LEAST16_MIN
482	527	# define __INT_LEAST16_MIN INT64_MIN
	528	# undef __INT_LEAST16_MAX
483	529	# define __INT_LEAST16_MAX INT64_MAX
	530	# undef __UINT_LEAST16_MAX
484	531	# define __UINT_LEAST16_MAX UINT64_MAX
	532	# undef __INT_LEAST8_MIN
485	533	# define __INT_LEAST8_MIN INT64_MIN
	534	# undef __INT_LEAST8_MAX
486	535	# define __INT_LEAST8_MAX INT64_MAX
	536	# undef __UINT_LEAST8_MAX
487	537	# define __UINT_LEAST8_MAX UINT64_MAX
488	538	#endif /* __INT64_TYPE__ */
489	539
...	...	@@ -497,7 +547,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
497	547
498	548	/* FIXME: This is using the placeholder dates Clang produces for these macros
499	549	in C2x mode; switch to the correct values once they've been published. */
500		#if __STDC_VERSION__ >= 202000L
	550	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
501	551	# define UINT_LEAST64_WIDTH __UINT_LEAST64_WIDTH
502	552	# define INT_LEAST64_WIDTH UINT_LEAST64_WIDTH
503	553	# define UINT_FAST64_WIDTH __UINT_LEAST64_WIDTH
...	...	@@ -517,27 +567,39 @@ typedef __UINTMAX_TYPE__ uintmax_t;
517	567	# define INT_FAST56_MAX INT56_MAX
518	568	# define UINT_FAST56_MAX UINT56_MAX
519	569
	570	# undef __INT_LEAST32_MIN
520	571	# define __INT_LEAST32_MIN INT56_MIN
	572	# undef __INT_LEAST32_MAX
521	573	# define __INT_LEAST32_MAX INT56_MAX
	574	# undef __UINT_LEAST32_MAX
522	575	# define __UINT_LEAST32_MAX UINT56_MAX
	576	# undef __INT_LEAST16_MIN
523	577	# define __INT_LEAST16_MIN INT56_MIN
	578	# undef __INT_LEAST16_MAX
524	579	# define __INT_LEAST16_MAX INT56_MAX
	580	# undef __UINT_LEAST16_MAX
525	581	# define __UINT_LEAST16_MAX UINT56_MAX
	582	# undef __INT_LEAST8_MIN
526	583	# define __INT_LEAST8_MIN INT56_MIN
	584	# undef __INT_LEAST8_MAX
527	585	# define __INT_LEAST8_MAX INT56_MAX
	586	# undef __UINT_LEAST8_MAX
528	587	# define __UINT_LEAST8_MAX UINT56_MAX
529	588
530	589	/* FIXME: This is using the placeholder dates Clang produces for these macros
531	590	in C2x mode; switch to the correct values once they've been published. */
532		#if __STDC_VERSION__ >= 202000L
	591	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
533	592	# define UINT56_WIDTH 56
534	593	# define INT56_WIDTH UINT56_WIDTH
535	594	# define UINT_LEAST56_WIDTH UINT56_WIDTH
536	595	# define INT_LEAST56_WIDTH UINT_LEAST56_WIDTH
537	596	# define UINT_FAST56_WIDTH UINT56_WIDTH
538	597	# define INT_FAST56_WIDTH UINT_FAST56_WIDTH
	598	# undef __UINT_LEAST32_WIDTH
539	599	# define __UINT_LEAST32_WIDTH UINT56_WIDTH
	600	# undef __UINT_LEAST16_WIDTH
540	601	# define __UINT_LEAST16_WIDTH UINT56_WIDTH
	602	# undef __UINT_LEAST8_WIDTH
541	603	# define __UINT_LEAST8_WIDTH UINT56_WIDTH
542	604	#endif /* __STDC_VERSION__ */
543	605	#endif /* __INT56_TYPE__ */
...	...	@@ -554,27 +616,39 @@ typedef __UINTMAX_TYPE__ uintmax_t;
554	616	# define INT_FAST48_MAX INT48_MAX
555	617	# define UINT_FAST48_MAX UINT48_MAX
556	618
	619	# undef __INT_LEAST32_MIN
557	620	# define __INT_LEAST32_MIN INT48_MIN
	621	# undef __INT_LEAST32_MAX
558	622	# define __INT_LEAST32_MAX INT48_MAX
	623	# undef __UINT_LEAST32_MAX
559	624	# define __UINT_LEAST32_MAX UINT48_MAX
	625	# undef __INT_LEAST16_MIN
560	626	# define __INT_LEAST16_MIN INT48_MIN
	627	# undef __INT_LEAST16_MAX
561	628	# define __INT_LEAST16_MAX INT48_MAX
	629	# undef __UINT_LEAST16_MAX
562	630	# define __UINT_LEAST16_MAX UINT48_MAX
	631	# undef __INT_LEAST8_MIN
563	632	# define __INT_LEAST8_MIN INT48_MIN
	633	# undef __INT_LEAST8_MAX
564	634	# define __INT_LEAST8_MAX INT48_MAX
	635	# undef __UINT_LEAST8_MAX
565	636	# define __UINT_LEAST8_MAX UINT48_MAX
566	637
567	638	/* FIXME: This is using the placeholder dates Clang produces for these macros
568	639	in C2x mode; switch to the correct values once they've been published. */
569		#if __STDC_VERSION__ >= 202000L
	640	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
570	641	#define UINT48_WIDTH 48
571	642	#define INT48_WIDTH UINT48_WIDTH
572	643	#define UINT_LEAST48_WIDTH UINT48_WIDTH
573	644	#define INT_LEAST48_WIDTH UINT_LEAST48_WIDTH
574	645	#define UINT_FAST48_WIDTH UINT48_WIDTH
575	646	#define INT_FAST48_WIDTH UINT_FAST48_WIDTH
	647	#undef __UINT_LEAST32_WIDTH
576	648	#define __UINT_LEAST32_WIDTH UINT48_WIDTH
	649	# undef __UINT_LEAST16_WIDTH
577	650	#define __UINT_LEAST16_WIDTH UINT48_WIDTH
	651	# undef __UINT_LEAST8_WIDTH
578	652	#define __UINT_LEAST8_WIDTH UINT48_WIDTH
579	653	#endif /* __STDC_VERSION__ */
580	654	#endif /* __INT48_TYPE__ */
...	...	@@ -591,27 +665,39 @@ typedef __UINTMAX_TYPE__ uintmax_t;
591	665	# define INT_FAST40_MAX INT40_MAX
592	666	# define UINT_FAST40_MAX UINT40_MAX
593	667
	668	# undef __INT_LEAST32_MIN
594	669	# define __INT_LEAST32_MIN INT40_MIN
	670	# undef __INT_LEAST32_MAX
595	671	# define __INT_LEAST32_MAX INT40_MAX
	672	# undef __UINT_LEAST32_MAX
596	673	# define __UINT_LEAST32_MAX UINT40_MAX
	674	# undef __INT_LEAST16_MIN
597	675	# define __INT_LEAST16_MIN INT40_MIN
	676	# undef __INT_LEAST16_MAX
598	677	# define __INT_LEAST16_MAX INT40_MAX
	678	# undef __UINT_LEAST16_MAX
599	679	# define __UINT_LEAST16_MAX UINT40_MAX
	680	# undef __INT_LEAST8_MIN
600	681	# define __INT_LEAST8_MIN INT40_MIN
	682	# undef __INT_LEAST8_MAX
601	683	# define __INT_LEAST8_MAX INT40_MAX
	684	# undef __UINT_LEAST8_MAX
602	685	# define __UINT_LEAST8_MAX UINT40_MAX
603	686
604	687	/* FIXME: This is using the placeholder dates Clang produces for these macros
605	688	in C2x mode; switch to the correct values once they've been published. */
606		#if __STDC_VERSION__ >= 202000L
	689	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
607	690	# define UINT40_WIDTH 40
608	691	# define INT40_WIDTH UINT40_WIDTH
609	692	# define UINT_LEAST40_WIDTH UINT40_WIDTH
610	693	# define INT_LEAST40_WIDTH UINT_LEAST40_WIDTH
611	694	# define UINT_FAST40_WIDTH UINT40_WIDTH
612	695	# define INT_FAST40_WIDTH UINT_FAST40_WIDTH
	696	# undef __UINT_LEAST32_WIDTH
613	697	# define __UINT_LEAST32_WIDTH UINT40_WIDTH
	698	# undef __UINT_LEAST16_WIDTH
614	699	# define __UINT_LEAST16_WIDTH UINT40_WIDTH
	700	# undef __UINT_LEAST8_WIDTH
615	701	# define __UINT_LEAST8_WIDTH UINT40_WIDTH
616	702	#endif /* __STDC_VERSION__ */
617	703	#endif /* __INT40_TYPE__ */
...	...	@@ -622,23 +708,35 @@ typedef __UINTMAX_TYPE__ uintmax_t;
622	708	# define INT32_MIN (-INT32_C(2147483647)-1)
623	709	# define UINT32_MAX UINT32_C(4294967295)
624	710
	711	# undef __INT_LEAST32_MIN
625	712	# define __INT_LEAST32_MIN INT32_MIN
	713	# undef __INT_LEAST32_MAX
626	714	# define __INT_LEAST32_MAX INT32_MAX
	715	# undef __UINT_LEAST32_MAX
627	716	# define __UINT_LEAST32_MAX UINT32_MAX
	717	# undef __INT_LEAST16_MIN
628	718	# define __INT_LEAST16_MIN INT32_MIN
	719	# undef __INT_LEAST16_MAX
629	720	# define __INT_LEAST16_MAX INT32_MAX
	721	# undef __UINT_LEAST16_MAX
630	722	# define __UINT_LEAST16_MAX UINT32_MAX
	723	# undef __INT_LEAST8_MIN
631	724	# define __INT_LEAST8_MIN INT32_MIN
	725	# undef __INT_LEAST8_MAX
632	726	# define __INT_LEAST8_MAX INT32_MAX
	727	# undef __UINT_LEAST8_MAX
633	728	# define __UINT_LEAST8_MAX UINT32_MAX
634	729
635	730	/* FIXME: This is using the placeholder dates Clang produces for these macros
636	731	in C2x mode; switch to the correct values once they've been published. */
637		#if __STDC_VERSION__ >= 202000L
	732	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
638	733	# define UINT32_WIDTH 32
639	734	# define INT32_WIDTH UINT32_WIDTH
	735	# undef __UINT_LEAST32_WIDTH
640	736	# define __UINT_LEAST32_WIDTH UINT32_WIDTH
	737	# undef __UINT_LEAST16_WIDTH
641	738	# define __UINT_LEAST16_WIDTH UINT32_WIDTH
	739	# undef __UINT_LEAST8_WIDTH
642	740	# define __UINT_LEAST8_WIDTH UINT32_WIDTH
643	741	#endif /* __STDC_VERSION__ */
644	742	#endif /* __INT32_TYPE__ */
...	...	@@ -653,7 +751,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
653	751
654	752	/* FIXME: This is using the placeholder dates Clang produces for these macros
655	753	in C2x mode; switch to the correct values once they've been published. */
656		#if __STDC_VERSION__ >= 202000L
	754	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
657	755	# define UINT_LEAST32_WIDTH __UINT_LEAST32_WIDTH
658	756	# define INT_LEAST32_WIDTH UINT_LEAST32_WIDTH
659	757	# define UINT_FAST32_WIDTH __UINT_LEAST32_WIDTH
...	...	@@ -673,23 +771,31 @@ typedef __UINTMAX_TYPE__ uintmax_t;
673	771	# define INT_FAST24_MAX INT24_MAX
674	772	# define UINT_FAST24_MAX UINT24_MAX
675	773
	774	# undef __INT_LEAST16_MIN
676	775	# define __INT_LEAST16_MIN INT24_MIN
	776	# undef __INT_LEAST16_MAX
677	777	# define __INT_LEAST16_MAX INT24_MAX
	778	# undef __UINT_LEAST16_MAX
678	779	# define __UINT_LEAST16_MAX UINT24_MAX
	780	# undef __INT_LEAST8_MIN
679	781	# define __INT_LEAST8_MIN INT24_MIN
	782	# undef __INT_LEAST8_MAX
680	783	# define __INT_LEAST8_MAX INT24_MAX
	784	# undef __UINT_LEAST8_MAX
681	785	# define __UINT_LEAST8_MAX UINT24_MAX
682	786
683	787	/* FIXME: This is using the placeholder dates Clang produces for these macros
684	788	in C2x mode; switch to the correct values once they've been published. */
685		#if __STDC_VERSION__ >= 202000L
	789	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
686	790	# define UINT24_WIDTH 24
687	791	# define INT24_WIDTH UINT24_WIDTH
688	792	# define UINT_LEAST24_WIDTH UINT24_WIDTH
689	793	# define INT_LEAST24_WIDTH UINT_LEAST24_WIDTH
690	794	# define UINT_FAST24_WIDTH UINT24_WIDTH
691	795	# define INT_FAST24_WIDTH UINT_FAST24_WIDTH
	796	# undef __UINT_LEAST16_WIDTH
692	797	# define __UINT_LEAST16_WIDTH UINT24_WIDTH
	798	# undef __UINT_LEAST8_WIDTH
693	799	# define __UINT_LEAST8_WIDTH UINT24_WIDTH
694	800	#endif /* __STDC_VERSION__ */
695	801	#endif /* __INT24_TYPE__ */
...	...	@@ -700,19 +806,27 @@ typedef __UINTMAX_TYPE__ uintmax_t;
700	806	#define INT16_MIN (-INT16_C(32767)-1)
701	807	#define UINT16_MAX UINT16_C(65535)
702	808
	809	# undef __INT_LEAST16_MIN
703	810	# define __INT_LEAST16_MIN INT16_MIN
	811	# undef __INT_LEAST16_MAX
704	812	# define __INT_LEAST16_MAX INT16_MAX
	813	# undef __UINT_LEAST16_MAX
705	814	# define __UINT_LEAST16_MAX UINT16_MAX
	815	# undef __INT_LEAST8_MIN
706	816	# define __INT_LEAST8_MIN INT16_MIN
	817	# undef __INT_LEAST8_MAX
707	818	# define __INT_LEAST8_MAX INT16_MAX
	819	# undef __UINT_LEAST8_MAX
708	820	# define __UINT_LEAST8_MAX UINT16_MAX
709	821
710	822	/* FIXME: This is using the placeholder dates Clang produces for these macros
711	823	in C2x mode; switch to the correct values once they've been published. */
712		#if __STDC_VERSION__ >= 202000L
	824	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
713	825	# define UINT16_WIDTH 16
714	826	# define INT16_WIDTH UINT16_WIDTH
	827	# undef __UINT_LEAST16_WIDTH
715	828	# define __UINT_LEAST16_WIDTH UINT16_WIDTH
	829	# undef __UINT_LEAST8_WIDTH
716	830	# define __UINT_LEAST8_WIDTH UINT16_WIDTH
717	831	#endif /* __STDC_VERSION__ */
718	832	#endif /* __INT16_TYPE__ */
...	...	@@ -727,7 +841,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
727	841
728	842	/* FIXME: This is using the placeholder dates Clang produces for these macros
729	843	in C2x mode; switch to the correct values once they've been published. */
730		#if __STDC_VERSION__ >= 202000L
	844	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
731	845	# define UINT_LEAST16_WIDTH __UINT_LEAST16_WIDTH
732	846	# define INT_LEAST16_WIDTH UINT_LEAST16_WIDTH
733	847	# define UINT_FAST16_WIDTH __UINT_LEAST16_WIDTH
...	...	@@ -741,15 +855,19 @@ typedef __UINTMAX_TYPE__ uintmax_t;
741	855	# define INT8_MIN (-INT8_C(127)-1)
742	856	# define UINT8_MAX UINT8_C(255)
743	857
	858	# undef __INT_LEAST8_MIN
744	859	# define __INT_LEAST8_MIN INT8_MIN
	860	# undef __INT_LEAST8_MAX
745	861	# define __INT_LEAST8_MAX INT8_MAX
	862	# undef __UINT_LEAST8_MAX
746	863	# define __UINT_LEAST8_MAX UINT8_MAX
747	864
748	865	/* FIXME: This is using the placeholder dates Clang produces for these macros
749	866	in C2x mode; switch to the correct values once they've been published. */
750		#if __STDC_VERSION__ >= 202000L
	867	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
751	868	# define UINT8_WIDTH 8
752	869	# define INT8_WIDTH UINT8_WIDTH
	870	# undef __UINT_LEAST8_WIDTH
753	871	# define __UINT_LEAST8_WIDTH UINT8_WIDTH
754	872	#endif /* __STDC_VERSION__ */
755	873	#endif /* __INT8_TYPE__ */
...	...	@@ -764,7 +882,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
764	882
765	883	/* FIXME: This is using the placeholder dates Clang produces for these macros
766	884	in C2x mode; switch to the correct values once they've been published. */
767		#if __STDC_VERSION__ >= 202000L
	885	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
768	886	# define UINT_LEAST8_WIDTH __UINT_LEAST8_WIDTH
769	887	# define INT_LEAST8_WIDTH UINT_LEAST8_WIDTH
770	888	# define UINT_FAST8_WIDTH __UINT_LEAST8_WIDTH
...	...	@@ -792,7 +910,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
792	910	/* C2x 7.20.2.4 Width of integer types capable of holding object pointers. */
793	911	/* FIXME: This is using the placeholder dates Clang produces for these macros
794	912	in C2x mode; switch to the correct values once they've been published. */
795		#if __STDC_VERSION__ >= 202000L
	913	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
796	914	/* NB: The C standard requires that these be the same value, but the compiler
797	915	exposes separate internal width macros. */
798	916	#define INTPTR_WIDTH __INTPTR_WIDTH__
...	...	@@ -813,7 +931,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
813	931	/* C2x 7.20.2.5 Width of greatest-width integer types. */
814	932	/* FIXME: This is using the placeholder dates Clang produces for these macros
815	933	in C2x mode; switch to the correct values once they've been published. */
816		#if __STDC_VERSION__ >= 202000L
	934	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
817	935	/* NB: The C standard requires that these be the same value, but the compiler
818	936	exposes separate internal width macros. */
819	937	#define INTMAX_WIDTH __INTMAX_WIDTH__
...	...	@@ -849,7 +967,7 @@ typedef __UINTMAX_TYPE__ uintmax_t;
849	967	/* C2x 7.20.3.x Width of other integer types. */
850	968	/* FIXME: This is using the placeholder dates Clang produces for these macros
851	969	in C2x mode; switch to the correct values once they've been published. */
852		#if __STDC_VERSION__ >= 202000L
	970	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202000L
853	971	#define PTRDIFF_WIDTH __PTRDIFF_WIDTH__
854	972	#define SIG_ATOMIC_WIDTH __SIG_ATOMIC_WIDTH__
855	973	#define SIZE_WIDTH __SIZE_WIDTH__

lib/include/stdnoreturn.h+1-1

...	...	@@ -13,7 +13,7 @@
13	13	#define noreturn _Noreturn
14	14	#define __noreturn_is_defined 1
15	15
16		#if __STDC_VERSION__ > 201710L && \
	16	#if (defined(__STDC_VERSION__) && __STDC_VERSION__ > 201710L) && \
17	17	!defined(_CLANG_DISABLE_CRT_DEPRECATION_WARNINGS)
18	18	/* The noreturn macro is deprecated in C2x. We do not mark it as such because
19	19	including the header file in C2x is also deprecated and we do not want to

lib/include/unwind.h+2-1

...	...	@@ -65,7 +65,8 @@ struct _Unwind_Context;
65	65	#if defined(__arm__) && !(defined(__USING_SJLJ_EXCEPTIONS__) \|\| \
66	66	defined(__ARM_DWARF_EH__) \|\| defined(__SEH__))
67	67	struct _Unwind_Control_Block;
68		typedef struct _Unwind_Control_Block _Unwind_Exception; /* Alias */
	68	typedef struct _Unwind_Control_Block _Unwind_Control_Block;
	69	#define _Unwind_Exception _Unwind_Control_Block /* Alias */
69	70	#else
70	71	struct _Unwind_Exception;
71	72	typedef struct _Unwind_Exception _Unwind_Exception;

lib/include/velintrin.h+1-1

...	...	@@ -13,7 +13,7 @@
13	13	typedef double __vr __attribute__((__vector_size__(2048)));
14	14
15	15	// Vector mask registers
16		#if __STDC_VERSION__ >= 199901L
	16	#if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L
17	17	// For C99
18	18	typedef _Bool __vm __attribute__((ext_vector_type(256)));
19	19	typedef _Bool __vm256 __attribute__((ext_vector_type(256)));

lib/include/x86gprintrin.h+19-7

...	...	@@ -25,23 +25,35 @@
25	25	#include <crc32intrin.h>
26	26	#endif
27	27
	28	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	29	defined(__PRFCHI__)
	30	#include <prfchiintrin.h>
	31	#endif
	32
	33	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	34	defined(__RAOINT__)
	35	#include <raointintrin.h>
	36	#endif
	37
	38	#if !(defined(_MSC_VER) \|\| defined(__SCE__)) \|\| __has_feature(modules) \|\| \
	39	defined(__CMPCCXADD__)
	40	#include <cmpccxaddintrin.h>
	41	#endif
	42
28	43	#if defined(__i386__)
29		#define __FULLBX "ebx"
	44	#define __SAVE_GPRBX "mov {%%ebx, %%eax \|eax, ebx};"
	45	#define __RESTORE_GPRBX "mov {%%eax, %%ebx \|ebx, eax};"
30	46	#define __TMPGPR "eax"
31	47	#else
32	48	// When in 64-bit target, the 32-bit operands generate a 32-bit result,
33	49	// zero-extended to a 64-bit result in the destination general-purpose,
34	50	// It means "mov x %ebx" will clobber the higher 32 bits of rbx, so we
35	51	// should preserve the 64-bit register rbx.
36		#define __FULLBX "rbx"
	52	#define __SAVE_GPRBX "mov {%%rbx, %%rax \|rax, rbx};"
	53	#define __RESTORE_GPRBX "mov {%%rax, %%rbx \|rbx, rax};"
37	54	#define __TMPGPR "rax"
38	55	#endif
39	56
40		#define __MOVEGPR(__r1, __r2) "mov {%%"__r1 ", %%"__r2 "\|"__r2 ", "__r1"};"
41
42		#define __SAVE_GPRBX __MOVEGPR(__FULLBX, __TMPGPR)
43		#define __RESTORE_GPRBX __MOVEGPR(__TMPGPR, __FULLBX)
44
45	57	#define __SSC_MARK(__Tag) \
46	58	__asm__ __volatile__( __SAVE_GPRBX \
47	59	"mov {%0, %%ebx\|ebx, %0}; " \

lib/include/xmmintrin.h+1-2

...	...	@@ -1906,7 +1906,7 @@ _mm_setr_ps(float __z, float __y, float __x, float __w)
1906	1906	static __inline__ __m128 __DEFAULT_FN_ATTRS
1907	1907	_mm_setzero_ps(void)
1908	1908	{
1909		return __extension__ (__m128){ 0, 0, 0, 0 };
	1909	return __extension__ (__m128){ 0.0f, 0.0f, 0.0f, 0.0f };
1910	1910	}
1911	1911
1912	1912	/// Stores the upper 64 bits of a 128-bit vector of [4 x float] to a
...	...	@@ -3005,7 +3005,6 @@ do { \
3005	3005	#define _m_pavgw _mm_avg_pu16
3006	3006	#define _m_psadbw _mm_sad_pu8
3007	3007	#define _m_ _mm_
3008		#define _m_ _mm_
3009	3008
3010	3009	#undef __DEFAULT_FN_ATTRS
3011	3010	#undef __DEFAULT_FN_ATTRS_MMX