authorgravatar for andrew@ziglang.orgAndrew Kelley <andrew@ziglang.org> 2022-07-03 15:58:39-07:00
committergravatar for andrew@ziglang.orgAndrew Kelley <andrew@ziglang.org> 2022-07-03 15:58:39-07:00
logd588f88b1a46fc3719db50ae81e2354461c1d0da
tree702481c2957193d740a09405a14859ebf9984445
parentcbdd21cd9ac8ccab923c1f27f472af747489f661

update C headers to LLVM 14.0.6

6 files changed, 4098 insertions(+), 4090 deletions(-)

lib/include/__clang_cuda_intrinsics.h+1-1
......@@ -509,7 +509,7 @@ __device__ inline void *__nv_cvta_constant_to_generic_impl(size_t __ptr) {
509509__device__ inline void *__nv_cvta_local_to_generic_impl(size_t __ptr) {
510510 return (void *)(void __attribute__((address_space(5))) *)__ptr;
511511}
512__device__ inline uint32_t __nvvm_get_smem_pointer(void *__ptr) {
512__device__ inline cuuint32_t __nvvm_get_smem_pointer(void *__ptr) {
513513 return __nv_cvta_generic_to_shared_impl(__ptr);
514514}
515515} // extern "C"
lib/include/arm_fp16.h+89-89
......@@ -31,561 +31,561 @@ typedef __fp16 float16_t;
3131
3232#if defined(__ARM_FEATURE_FP16_SCALAR_ARITHMETIC) && defined(__aarch64__)
3333#define vabdh_f16(__p0, __p1) __extension__ ({ \
34 float16_t __ret; \
3435 float16_t __s0 = __p0; \
3536 float16_t __s1 = __p1; \
36 float16_t __ret; \
3737 __ret = (float16_t) __builtin_neon_vabdh_f16(__s0, __s1); \
3838 __ret; \
3939})
4040#define vabsh_f16(__p0) __extension__ ({ \
41 float16_t __s0 = __p0; \
4241 float16_t __ret; \
42 float16_t __s0 = __p0; \
4343 __ret = (float16_t) __builtin_neon_vabsh_f16(__s0); \
4444 __ret; \
4545})
4646#define vaddh_f16(__p0, __p1) __extension__ ({ \
47 float16_t __ret; \
4748 float16_t __s0 = __p0; \
4849 float16_t __s1 = __p1; \
49 float16_t __ret; \
5050 __ret = (float16_t) __builtin_neon_vaddh_f16(__s0, __s1); \
5151 __ret; \
5252})
5353#define vcageh_f16(__p0, __p1) __extension__ ({ \
54 uint16_t __ret; \
5455 float16_t __s0 = __p0; \
5556 float16_t __s1 = __p1; \
56 uint16_t __ret; \
5757 __ret = (uint16_t) __builtin_neon_vcageh_f16(__s0, __s1); \
5858 __ret; \
5959})
6060#define vcagth_f16(__p0, __p1) __extension__ ({ \
61 uint16_t __ret; \
6162 float16_t __s0 = __p0; \
6263 float16_t __s1 = __p1; \
63 uint16_t __ret; \
6464 __ret = (uint16_t) __builtin_neon_vcagth_f16(__s0, __s1); \
6565 __ret; \
6666})
6767#define vcaleh_f16(__p0, __p1) __extension__ ({ \
68 uint16_t __ret; \
6869 float16_t __s0 = __p0; \
6970 float16_t __s1 = __p1; \
70 uint16_t __ret; \
7171 __ret = (uint16_t) __builtin_neon_vcaleh_f16(__s0, __s1); \
7272 __ret; \
7373})
7474#define vcalth_f16(__p0, __p1) __extension__ ({ \
75 uint16_t __ret; \
7576 float16_t __s0 = __p0; \
7677 float16_t __s1 = __p1; \
77 uint16_t __ret; \
7878 __ret = (uint16_t) __builtin_neon_vcalth_f16(__s0, __s1); \
7979 __ret; \
8080})
8181#define vceqh_f16(__p0, __p1) __extension__ ({ \
82 uint16_t __ret; \
8283 float16_t __s0 = __p0; \
8384 float16_t __s1 = __p1; \
84 uint16_t __ret; \
8585 __ret = (uint16_t) __builtin_neon_vceqh_f16(__s0, __s1); \
8686 __ret; \
8787})
8888#define vceqzh_f16(__p0) __extension__ ({ \
89 float16_t __s0 = __p0; \
9089 uint16_t __ret; \
90 float16_t __s0 = __p0; \
9191 __ret = (uint16_t) __builtin_neon_vceqzh_f16(__s0); \
9292 __ret; \
9393})
9494#define vcgeh_f16(__p0, __p1) __extension__ ({ \
95 uint16_t __ret; \
9596 float16_t __s0 = __p0; \
9697 float16_t __s1 = __p1; \
97 uint16_t __ret; \
9898 __ret = (uint16_t) __builtin_neon_vcgeh_f16(__s0, __s1); \
9999 __ret; \
100100})
101101#define vcgezh_f16(__p0) __extension__ ({ \
102 float16_t __s0 = __p0; \
103102 uint16_t __ret; \
103 float16_t __s0 = __p0; \
104104 __ret = (uint16_t) __builtin_neon_vcgezh_f16(__s0); \
105105 __ret; \
106106})
107107#define vcgth_f16(__p0, __p1) __extension__ ({ \
108 uint16_t __ret; \
108109 float16_t __s0 = __p0; \
109110 float16_t __s1 = __p1; \
110 uint16_t __ret; \
111111 __ret = (uint16_t) __builtin_neon_vcgth_f16(__s0, __s1); \
112112 __ret; \
113113})
114114#define vcgtzh_f16(__p0) __extension__ ({ \
115 float16_t __s0 = __p0; \
116115 uint16_t __ret; \
116 float16_t __s0 = __p0; \
117117 __ret = (uint16_t) __builtin_neon_vcgtzh_f16(__s0); \
118118 __ret; \
119119})
120120#define vcleh_f16(__p0, __p1) __extension__ ({ \
121 uint16_t __ret; \
121122 float16_t __s0 = __p0; \
122123 float16_t __s1 = __p1; \
123 uint16_t __ret; \
124124 __ret = (uint16_t) __builtin_neon_vcleh_f16(__s0, __s1); \
125125 __ret; \
126126})
127127#define vclezh_f16(__p0) __extension__ ({ \
128 float16_t __s0 = __p0; \
129128 uint16_t __ret; \
129 float16_t __s0 = __p0; \
130130 __ret = (uint16_t) __builtin_neon_vclezh_f16(__s0); \
131131 __ret; \
132132})
133133#define vclth_f16(__p0, __p1) __extension__ ({ \
134 uint16_t __ret; \
134135 float16_t __s0 = __p0; \
135136 float16_t __s1 = __p1; \
136 uint16_t __ret; \
137137 __ret = (uint16_t) __builtin_neon_vclth_f16(__s0, __s1); \
138138 __ret; \
139139})
140140#define vcltzh_f16(__p0) __extension__ ({ \
141 float16_t __s0 = __p0; \
142141 uint16_t __ret; \
142 float16_t __s0 = __p0; \
143143 __ret = (uint16_t) __builtin_neon_vcltzh_f16(__s0); \
144144 __ret; \
145145})
146146#define vcvth_n_s16_f16(__p0, __p1) __extension__ ({ \
147 float16_t __s0 = __p0; \
148147 int16_t __ret; \
148 float16_t __s0 = __p0; \
149149 __ret = (int16_t) __builtin_neon_vcvth_n_s16_f16(__s0, __p1); \
150150 __ret; \
151151})
152152#define vcvth_n_s32_f16(__p0, __p1) __extension__ ({ \
153 float16_t __s0 = __p0; \
154153 int32_t __ret; \
154 float16_t __s0 = __p0; \
155155 __ret = (int32_t) __builtin_neon_vcvth_n_s32_f16(__s0, __p1); \
156156 __ret; \
157157})
158158#define vcvth_n_s64_f16(__p0, __p1) __extension__ ({ \
159 float16_t __s0 = __p0; \
160159 int64_t __ret; \
160 float16_t __s0 = __p0; \
161161 __ret = (int64_t) __builtin_neon_vcvth_n_s64_f16(__s0, __p1); \
162162 __ret; \
163163})
164164#define vcvth_n_u16_f16(__p0, __p1) __extension__ ({ \
165 float16_t __s0 = __p0; \
166165 uint16_t __ret; \
166 float16_t __s0 = __p0; \
167167 __ret = (uint16_t) __builtin_neon_vcvth_n_u16_f16(__s0, __p1); \
168168 __ret; \
169169})
170170#define vcvth_n_u32_f16(__p0, __p1) __extension__ ({ \
171 float16_t __s0 = __p0; \
172171 uint32_t __ret; \
172 float16_t __s0 = __p0; \
173173 __ret = (uint32_t) __builtin_neon_vcvth_n_u32_f16(__s0, __p1); \
174174 __ret; \
175175})
176176#define vcvth_n_u64_f16(__p0, __p1) __extension__ ({ \
177 float16_t __s0 = __p0; \
178177 uint64_t __ret; \
178 float16_t __s0 = __p0; \
179179 __ret = (uint64_t) __builtin_neon_vcvth_n_u64_f16(__s0, __p1); \
180180 __ret; \
181181})
182182#define vcvth_s16_f16(__p0) __extension__ ({ \
183 float16_t __s0 = __p0; \
184183 int16_t __ret; \
184 float16_t __s0 = __p0; \
185185 __ret = (int16_t) __builtin_neon_vcvth_s16_f16(__s0); \
186186 __ret; \
187187})
188188#define vcvth_s32_f16(__p0) __extension__ ({ \
189 float16_t __s0 = __p0; \
190189 int32_t __ret; \
190 float16_t __s0 = __p0; \
191191 __ret = (int32_t) __builtin_neon_vcvth_s32_f16(__s0); \
192192 __ret; \
193193})
194194#define vcvth_s64_f16(__p0) __extension__ ({ \
195 float16_t __s0 = __p0; \
196195 int64_t __ret; \
196 float16_t __s0 = __p0; \
197197 __ret = (int64_t) __builtin_neon_vcvth_s64_f16(__s0); \
198198 __ret; \
199199})
200200#define vcvth_u16_f16(__p0) __extension__ ({ \
201 float16_t __s0 = __p0; \
202201 uint16_t __ret; \
202 float16_t __s0 = __p0; \
203203 __ret = (uint16_t) __builtin_neon_vcvth_u16_f16(__s0); \
204204 __ret; \
205205})
206206#define vcvth_u32_f16(__p0) __extension__ ({ \
207 float16_t __s0 = __p0; \
208207 uint32_t __ret; \
208 float16_t __s0 = __p0; \
209209 __ret = (uint32_t) __builtin_neon_vcvth_u32_f16(__s0); \
210210 __ret; \
211211})
212212#define vcvth_u64_f16(__p0) __extension__ ({ \
213 float16_t __s0 = __p0; \
214213 uint64_t __ret; \
214 float16_t __s0 = __p0; \
215215 __ret = (uint64_t) __builtin_neon_vcvth_u64_f16(__s0); \
216216 __ret; \
217217})
218218#define vcvtah_s16_f16(__p0) __extension__ ({ \
219 float16_t __s0 = __p0; \
220219 int16_t __ret; \
220 float16_t __s0 = __p0; \
221221 __ret = (int16_t) __builtin_neon_vcvtah_s16_f16(__s0); \
222222 __ret; \
223223})
224224#define vcvtah_s32_f16(__p0) __extension__ ({ \
225 float16_t __s0 = __p0; \
226225 int32_t __ret; \
226 float16_t __s0 = __p0; \
227227 __ret = (int32_t) __builtin_neon_vcvtah_s32_f16(__s0); \
228228 __ret; \
229229})
230230#define vcvtah_s64_f16(__p0) __extension__ ({ \
231 float16_t __s0 = __p0; \
232231 int64_t __ret; \
232 float16_t __s0 = __p0; \
233233 __ret = (int64_t) __builtin_neon_vcvtah_s64_f16(__s0); \
234234 __ret; \
235235})
236236#define vcvtah_u16_f16(__p0) __extension__ ({ \
237 float16_t __s0 = __p0; \
238237 uint16_t __ret; \
238 float16_t __s0 = __p0; \
239239 __ret = (uint16_t) __builtin_neon_vcvtah_u16_f16(__s0); \
240240 __ret; \
241241})
242242#define vcvtah_u32_f16(__p0) __extension__ ({ \
243 float16_t __s0 = __p0; \
244243 uint32_t __ret; \
244 float16_t __s0 = __p0; \
245245 __ret = (uint32_t) __builtin_neon_vcvtah_u32_f16(__s0); \
246246 __ret; \
247247})
248248#define vcvtah_u64_f16(__p0) __extension__ ({ \
249 float16_t __s0 = __p0; \
250249 uint64_t __ret; \
250 float16_t __s0 = __p0; \
251251 __ret = (uint64_t) __builtin_neon_vcvtah_u64_f16(__s0); \
252252 __ret; \
253253})
254254#define vcvth_f16_u16(__p0) __extension__ ({ \
255 uint16_t __s0 = __p0; \
256255 float16_t __ret; \
256 uint16_t __s0 = __p0; \
257257 __ret = (float16_t) __builtin_neon_vcvth_f16_u16(__s0); \
258258 __ret; \
259259})
260260#define vcvth_f16_s16(__p0) __extension__ ({ \
261 int16_t __s0 = __p0; \
262261 float16_t __ret; \
262 int16_t __s0 = __p0; \
263263 __ret = (float16_t) __builtin_neon_vcvth_f16_s16(__s0); \
264264 __ret; \
265265})
266266#define vcvth_f16_u32(__p0) __extension__ ({ \
267 uint32_t __s0 = __p0; \
268267 float16_t __ret; \
268 uint32_t __s0 = __p0; \
269269 __ret = (float16_t) __builtin_neon_vcvth_f16_u32(__s0); \
270270 __ret; \
271271})
272272#define vcvth_f16_s32(__p0) __extension__ ({ \
273 int32_t __s0 = __p0; \
274273 float16_t __ret; \
274 int32_t __s0 = __p0; \
275275 __ret = (float16_t) __builtin_neon_vcvth_f16_s32(__s0); \
276276 __ret; \
277277})
278278#define vcvth_f16_u64(__p0) __extension__ ({ \
279 uint64_t __s0 = __p0; \
280279 float16_t __ret; \
280 uint64_t __s0 = __p0; \
281281 __ret = (float16_t) __builtin_neon_vcvth_f16_u64(__s0); \
282282 __ret; \
283283})
284284#define vcvth_f16_s64(__p0) __extension__ ({ \
285 int64_t __s0 = __p0; \
286285 float16_t __ret; \
286 int64_t __s0 = __p0; \
287287 __ret = (float16_t) __builtin_neon_vcvth_f16_s64(__s0); \
288288 __ret; \
289289})
290290#define vcvth_n_f16_u32(__p0, __p1) __extension__ ({ \
291 uint32_t __s0 = __p0; \
292291 float16_t __ret; \
292 uint32_t __s0 = __p0; \
293293 __ret = (float16_t) __builtin_neon_vcvth_n_f16_u32(__s0, __p1); \
294294 __ret; \
295295})
296296#define vcvth_n_f16_s32(__p0, __p1) __extension__ ({ \
297 int32_t __s0 = __p0; \
298297 float16_t __ret; \
298 int32_t __s0 = __p0; \
299299 __ret = (float16_t) __builtin_neon_vcvth_n_f16_s32(__s0, __p1); \
300300 __ret; \
301301})
302302#define vcvth_n_f16_u64(__p0, __p1) __extension__ ({ \
303 uint64_t __s0 = __p0; \
304303 float16_t __ret; \
304 uint64_t __s0 = __p0; \
305305 __ret = (float16_t) __builtin_neon_vcvth_n_f16_u64(__s0, __p1); \
306306 __ret; \
307307})
308308#define vcvth_n_f16_s64(__p0, __p1) __extension__ ({ \
309 int64_t __s0 = __p0; \
310309 float16_t __ret; \
310 int64_t __s0 = __p0; \
311311 __ret = (float16_t) __builtin_neon_vcvth_n_f16_s64(__s0, __p1); \
312312 __ret; \
313313})
314314#define vcvth_n_f16_u16(__p0, __p1) __extension__ ({ \
315 uint16_t __s0 = __p0; \
316315 float16_t __ret; \
316 uint16_t __s0 = __p0; \
317317 __ret = (float16_t) __builtin_neon_vcvth_n_f16_u16(__s0, __p1); \
318318 __ret; \
319319})
320320#define vcvth_n_f16_s16(__p0, __p1) __extension__ ({ \
321 int16_t __s0 = __p0; \
322321 float16_t __ret; \
322 int16_t __s0 = __p0; \
323323 __ret = (float16_t) __builtin_neon_vcvth_n_f16_s16(__s0, __p1); \
324324 __ret; \
325325})
326326#define vcvtmh_s16_f16(__p0) __extension__ ({ \
327 float16_t __s0 = __p0; \
328327 int16_t __ret; \
328 float16_t __s0 = __p0; \
329329 __ret = (int16_t) __builtin_neon_vcvtmh_s16_f16(__s0); \
330330 __ret; \
331331})
332332#define vcvtmh_s32_f16(__p0) __extension__ ({ \
333 float16_t __s0 = __p0; \
334333 int32_t __ret; \
334 float16_t __s0 = __p0; \
335335 __ret = (int32_t) __builtin_neon_vcvtmh_s32_f16(__s0); \
336336 __ret; \
337337})
338338#define vcvtmh_s64_f16(__p0) __extension__ ({ \
339 float16_t __s0 = __p0; \
340339 int64_t __ret; \
340 float16_t __s0 = __p0; \
341341 __ret = (int64_t) __builtin_neon_vcvtmh_s64_f16(__s0); \
342342 __ret; \
343343})
344344#define vcvtmh_u16_f16(__p0) __extension__ ({ \
345 float16_t __s0 = __p0; \
346345 uint16_t __ret; \
346 float16_t __s0 = __p0; \
347347 __ret = (uint16_t) __builtin_neon_vcvtmh_u16_f16(__s0); \
348348 __ret; \
349349})
350350#define vcvtmh_u32_f16(__p0) __extension__ ({ \
351 float16_t __s0 = __p0; \
352351 uint32_t __ret; \
352 float16_t __s0 = __p0; \
353353 __ret = (uint32_t) __builtin_neon_vcvtmh_u32_f16(__s0); \
354354 __ret; \
355355})
356356#define vcvtmh_u64_f16(__p0) __extension__ ({ \
357 float16_t __s0 = __p0; \
358357 uint64_t __ret; \
358 float16_t __s0 = __p0; \
359359 __ret = (uint64_t) __builtin_neon_vcvtmh_u64_f16(__s0); \
360360 __ret; \
361361})
362362#define vcvtnh_s16_f16(__p0) __extension__ ({ \
363 float16_t __s0 = __p0; \
364363 int16_t __ret; \
364 float16_t __s0 = __p0; \
365365 __ret = (int16_t) __builtin_neon_vcvtnh_s16_f16(__s0); \
366366 __ret; \
367367})
368368#define vcvtnh_s32_f16(__p0) __extension__ ({ \
369 float16_t __s0 = __p0; \
370369 int32_t __ret; \
370 float16_t __s0 = __p0; \
371371 __ret = (int32_t) __builtin_neon_vcvtnh_s32_f16(__s0); \
372372 __ret; \
373373})
374374#define vcvtnh_s64_f16(__p0) __extension__ ({ \
375 float16_t __s0 = __p0; \
376375 int64_t __ret; \
376 float16_t __s0 = __p0; \
377377 __ret = (int64_t) __builtin_neon_vcvtnh_s64_f16(__s0); \
378378 __ret; \
379379})
380380#define vcvtnh_u16_f16(__p0) __extension__ ({ \
381 float16_t __s0 = __p0; \
382381 uint16_t __ret; \
382 float16_t __s0 = __p0; \
383383 __ret = (uint16_t) __builtin_neon_vcvtnh_u16_f16(__s0); \
384384 __ret; \
385385})
386386#define vcvtnh_u32_f16(__p0) __extension__ ({ \
387 float16_t __s0 = __p0; \
388387 uint32_t __ret; \
388 float16_t __s0 = __p0; \
389389 __ret = (uint32_t) __builtin_neon_vcvtnh_u32_f16(__s0); \
390390 __ret; \
391391})
392392#define vcvtnh_u64_f16(__p0) __extension__ ({ \
393 float16_t __s0 = __p0; \
394393 uint64_t __ret; \
394 float16_t __s0 = __p0; \
395395 __ret = (uint64_t) __builtin_neon_vcvtnh_u64_f16(__s0); \
396396 __ret; \
397397})
398398#define vcvtph_s16_f16(__p0) __extension__ ({ \
399 float16_t __s0 = __p0; \
400399 int16_t __ret; \
400 float16_t __s0 = __p0; \
401401 __ret = (int16_t) __builtin_neon_vcvtph_s16_f16(__s0); \
402402 __ret; \
403403})
404404#define vcvtph_s32_f16(__p0) __extension__ ({ \
405 float16_t __s0 = __p0; \
406405 int32_t __ret; \
406 float16_t __s0 = __p0; \
407407 __ret = (int32_t) __builtin_neon_vcvtph_s32_f16(__s0); \
408408 __ret; \
409409})
410410#define vcvtph_s64_f16(__p0) __extension__ ({ \
411 float16_t __s0 = __p0; \
412411 int64_t __ret; \
412 float16_t __s0 = __p0; \
413413 __ret = (int64_t) __builtin_neon_vcvtph_s64_f16(__s0); \
414414 __ret; \
415415})
416416#define vcvtph_u16_f16(__p0) __extension__ ({ \
417 float16_t __s0 = __p0; \
418417 uint16_t __ret; \
418 float16_t __s0 = __p0; \
419419 __ret = (uint16_t) __builtin_neon_vcvtph_u16_f16(__s0); \
420420 __ret; \
421421})
422422#define vcvtph_u32_f16(__p0) __extension__ ({ \
423 float16_t __s0 = __p0; \
424423 uint32_t __ret; \
424 float16_t __s0 = __p0; \
425425 __ret = (uint32_t) __builtin_neon_vcvtph_u32_f16(__s0); \
426426 __ret; \
427427})
428428#define vcvtph_u64_f16(__p0) __extension__ ({ \
429 float16_t __s0 = __p0; \
430429 uint64_t __ret; \
430 float16_t __s0 = __p0; \
431431 __ret = (uint64_t) __builtin_neon_vcvtph_u64_f16(__s0); \
432432 __ret; \
433433})
434434#define vdivh_f16(__p0, __p1) __extension__ ({ \
435 float16_t __ret; \
435436 float16_t __s0 = __p0; \
436437 float16_t __s1 = __p1; \
437 float16_t __ret; \
438438 __ret = (float16_t) __builtin_neon_vdivh_f16(__s0, __s1); \
439439 __ret; \
440440})
441441#define vfmah_f16(__p0, __p1, __p2) __extension__ ({ \
442 float16_t __ret; \
442443 float16_t __s0 = __p0; \
443444 float16_t __s1 = __p1; \
444445 float16_t __s2 = __p2; \
445 float16_t __ret; \
446446 __ret = (float16_t) __builtin_neon_vfmah_f16(__s0, __s1, __s2); \
447447 __ret; \
448448})
449449#define vfmsh_f16(__p0, __p1, __p2) __extension__ ({ \
450 float16_t __ret; \
450451 float16_t __s0 = __p0; \
451452 float16_t __s1 = __p1; \
452453 float16_t __s2 = __p2; \
453 float16_t __ret; \
454454 __ret = (float16_t) __builtin_neon_vfmsh_f16(__s0, __s1, __s2); \
455455 __ret; \
456456})
457457#define vmaxh_f16(__p0, __p1) __extension__ ({ \
458 float16_t __ret; \
458459 float16_t __s0 = __p0; \
459460 float16_t __s1 = __p1; \
460 float16_t __ret; \
461461 __ret = (float16_t) __builtin_neon_vmaxh_f16(__s0, __s1); \
462462 __ret; \
463463})
464464#define vmaxnmh_f16(__p0, __p1) __extension__ ({ \
465 float16_t __ret; \
465466 float16_t __s0 = __p0; \
466467 float16_t __s1 = __p1; \
467 float16_t __ret; \
468468 __ret = (float16_t) __builtin_neon_vmaxnmh_f16(__s0, __s1); \
469469 __ret; \
470470})
471471#define vminh_f16(__p0, __p1) __extension__ ({ \
472 float16_t __ret; \
472473 float16_t __s0 = __p0; \
473474 float16_t __s1 = __p1; \
474 float16_t __ret; \
475475 __ret = (float16_t) __builtin_neon_vminh_f16(__s0, __s1); \
476476 __ret; \
477477})
478478#define vminnmh_f16(__p0, __p1) __extension__ ({ \
479 float16_t __ret; \
479480 float16_t __s0 = __p0; \
480481 float16_t __s1 = __p1; \
481 float16_t __ret; \
482482 __ret = (float16_t) __builtin_neon_vminnmh_f16(__s0, __s1); \
483483 __ret; \
484484})
485485#define vmulh_f16(__p0, __p1) __extension__ ({ \
486 float16_t __ret; \
486487 float16_t __s0 = __p0; \
487488 float16_t __s1 = __p1; \
488 float16_t __ret; \
489489 __ret = (float16_t) __builtin_neon_vmulh_f16(__s0, __s1); \
490490 __ret; \
491491})
492492#define vmulxh_f16(__p0, __p1) __extension__ ({ \
493 float16_t __ret; \
493494 float16_t __s0 = __p0; \
494495 float16_t __s1 = __p1; \
495 float16_t __ret; \
496496 __ret = (float16_t) __builtin_neon_vmulxh_f16(__s0, __s1); \
497497 __ret; \
498498})
499499#define vnegh_f16(__p0) __extension__ ({ \
500 float16_t __s0 = __p0; \
501500 float16_t __ret; \
501 float16_t __s0 = __p0; \
502502 __ret = (float16_t) __builtin_neon_vnegh_f16(__s0); \
503503 __ret; \
504504})
505505#define vrecpeh_f16(__p0) __extension__ ({ \
506 float16_t __s0 = __p0; \
507506 float16_t __ret; \
507 float16_t __s0 = __p0; \
508508 __ret = (float16_t) __builtin_neon_vrecpeh_f16(__s0); \
509509 __ret; \
510510})
511511#define vrecpsh_f16(__p0, __p1) __extension__ ({ \
512 float16_t __ret; \
512513 float16_t __s0 = __p0; \
513514 float16_t __s1 = __p1; \
514 float16_t __ret; \
515515 __ret = (float16_t) __builtin_neon_vrecpsh_f16(__s0, __s1); \
516516 __ret; \
517517})
518518#define vrecpxh_f16(__p0) __extension__ ({ \
519 float16_t __s0 = __p0; \
520519 float16_t __ret; \
520 float16_t __s0 = __p0; \
521521 __ret = (float16_t) __builtin_neon_vrecpxh_f16(__s0); \
522522 __ret; \
523523})
524524#define vrndh_f16(__p0) __extension__ ({ \
525 float16_t __s0 = __p0; \
526525 float16_t __ret; \
526 float16_t __s0 = __p0; \
527527 __ret = (float16_t) __builtin_neon_vrndh_f16(__s0); \
528528 __ret; \
529529})
530530#define vrndah_f16(__p0) __extension__ ({ \
531 float16_t __s0 = __p0; \
532531 float16_t __ret; \
532 float16_t __s0 = __p0; \
533533 __ret = (float16_t) __builtin_neon_vrndah_f16(__s0); \
534534 __ret; \
535535})
536536#define vrndih_f16(__p0) __extension__ ({ \
537 float16_t __s0 = __p0; \
538537 float16_t __ret; \
538 float16_t __s0 = __p0; \
539539 __ret = (float16_t) __builtin_neon_vrndih_f16(__s0); \
540540 __ret; \
541541})
542542#define vrndmh_f16(__p0) __extension__ ({ \
543 float16_t __s0 = __p0; \
544543 float16_t __ret; \
544 float16_t __s0 = __p0; \
545545 __ret = (float16_t) __builtin_neon_vrndmh_f16(__s0); \
546546 __ret; \
547547})
548548#define vrndnh_f16(__p0) __extension__ ({ \
549 float16_t __s0 = __p0; \
550549 float16_t __ret; \
550 float16_t __s0 = __p0; \
551551 __ret = (float16_t) __builtin_neon_vrndnh_f16(__s0); \
552552 __ret; \
553553})
554554#define vrndph_f16(__p0) __extension__ ({ \
555 float16_t __s0 = __p0; \
556555 float16_t __ret; \
556 float16_t __s0 = __p0; \
557557 __ret = (float16_t) __builtin_neon_vrndph_f16(__s0); \
558558 __ret; \
559559})
560560#define vrndxh_f16(__p0) __extension__ ({ \
561 float16_t __s0 = __p0; \
562561 float16_t __ret; \
562 float16_t __s0 = __p0; \
563563 __ret = (float16_t) __builtin_neon_vrndxh_f16(__s0); \
564564 __ret; \
565565})
566566#define vrsqrteh_f16(__p0) __extension__ ({ \
567 float16_t __s0 = __p0; \
568567 float16_t __ret; \
568 float16_t __s0 = __p0; \
569569 __ret = (float16_t) __builtin_neon_vrsqrteh_f16(__s0); \
570570 __ret; \
571571})
572572#define vrsqrtsh_f16(__p0, __p1) __extension__ ({ \
573 float16_t __ret; \
573574 float16_t __s0 = __p0; \
574575 float16_t __s1 = __p1; \
575 float16_t __ret; \
576576 __ret = (float16_t) __builtin_neon_vrsqrtsh_f16(__s0, __s1); \
577577 __ret; \
578578})
579579#define vsqrth_f16(__p0) __extension__ ({ \
580 float16_t __s0 = __p0; \
581580 float16_t __ret; \
581 float16_t __s0 = __p0; \
582582 __ret = (float16_t) __builtin_neon_vsqrth_f16(__s0); \
583583 __ret; \
584584})
585585#define vsubh_f16(__p0, __p1) __extension__ ({ \
586 float16_t __ret; \
586587 float16_t __s0 = __p0; \
587588 float16_t __s1 = __p1; \
588 float16_t __ret; \
589589 __ret = (float16_t) __builtin_neon_vsubh_f16(__s0, __s1); \
590590 __ret; \
591591})
lib/include/arm_neon.h+3999-3999
......@@ -462,53 +462,53 @@ typedef struct bfloat16x8x4_t {
462462
463463#ifdef __LITTLE_ENDIAN__
464464#define splat_lane_p8(__p0, __p1) __extension__ ({ \
465 poly8x8_t __s0 = __p0; \
466465 poly8x8_t __ret; \
466 poly8x8_t __s0 = __p0; \
467467 __ret = (poly8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 4); \
468468 __ret; \
469469})
470470#else
471471#define splat_lane_p8(__p0, __p1) __extension__ ({ \
472 poly8x8_t __ret; \
472473 poly8x8_t __s0 = __p0; \
473474 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
474 poly8x8_t __ret; \
475475 __ret = (poly8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 4); \
476476 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
477477 __ret; \
478478})
479479#define __noswap_splat_lane_p8(__p0, __p1) __extension__ ({ \
480 poly8x8_t __s0 = __p0; \
481480 poly8x8_t __ret; \
481 poly8x8_t __s0 = __p0; \
482482 __ret = (poly8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 4); \
483483 __ret; \
484484})
485485#endif
486486
487487#define splat_lane_p64(__p0, __p1) __extension__ ({ \
488 poly64x1_t __s0 = __p0; \
489488 poly64x1_t __ret; \
489 poly64x1_t __s0 = __p0; \
490490 __ret = (poly64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 6); \
491491 __ret; \
492492})
493493#ifdef __LITTLE_ENDIAN__
494494#define splat_lane_p16(__p0, __p1) __extension__ ({ \
495 poly16x4_t __s0 = __p0; \
496495 poly16x4_t __ret; \
496 poly16x4_t __s0 = __p0; \
497497 __ret = (poly16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 5); \
498498 __ret; \
499499})
500500#else
501501#define splat_lane_p16(__p0, __p1) __extension__ ({ \
502 poly16x4_t __ret; \
502503 poly16x4_t __s0 = __p0; \
503504 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
504 poly16x4_t __ret; \
505505 __ret = (poly16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 5); \
506506 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
507507 __ret; \
508508})
509509#define __noswap_splat_lane_p16(__p0, __p1) __extension__ ({ \
510 poly16x4_t __s0 = __p0; \
511510 poly16x4_t __ret; \
511 poly16x4_t __s0 = __p0; \
512512 __ret = (poly16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 5); \
513513 __ret; \
514514})
......@@ -516,23 +516,23 @@ typedef struct bfloat16x8x4_t {
516516
517517#ifdef __LITTLE_ENDIAN__
518518#define splatq_lane_p8(__p0, __p1) __extension__ ({ \
519 poly8x8_t __s0 = __p0; \
520519 poly8x16_t __ret; \
520 poly8x8_t __s0 = __p0; \
521521 __ret = (poly8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 4); \
522522 __ret; \
523523})
524524#else
525525#define splatq_lane_p8(__p0, __p1) __extension__ ({ \
526 poly8x16_t __ret; \
526527 poly8x8_t __s0 = __p0; \
527528 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
528 poly8x16_t __ret; \
529529 __ret = (poly8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 4); \
530530 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
531531 __ret; \
532532})
533533#define __noswap_splatq_lane_p8(__p0, __p1) __extension__ ({ \
534 poly8x8_t __s0 = __p0; \
535534 poly8x16_t __ret; \
535 poly8x8_t __s0 = __p0; \
536536 __ret = (poly8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 4); \
537537 __ret; \
538538})
......@@ -540,22 +540,22 @@ typedef struct bfloat16x8x4_t {
540540
541541#ifdef __LITTLE_ENDIAN__
542542#define splatq_lane_p64(__p0, __p1) __extension__ ({ \
543 poly64x1_t __s0 = __p0; \
544543 poly64x2_t __ret; \
544 poly64x1_t __s0 = __p0; \
545545 __ret = (poly64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 6); \
546546 __ret; \
547547})
548548#else
549549#define splatq_lane_p64(__p0, __p1) __extension__ ({ \
550 poly64x1_t __s0 = __p0; \
551550 poly64x2_t __ret; \
551 poly64x1_t __s0 = __p0; \
552552 __ret = (poly64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 6); \
553553 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
554554 __ret; \
555555})
556556#define __noswap_splatq_lane_p64(__p0, __p1) __extension__ ({ \
557 poly64x1_t __s0 = __p0; \
558557 poly64x2_t __ret; \
558 poly64x1_t __s0 = __p0; \
559559 __ret = (poly64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 6); \
560560 __ret; \
561561})
......@@ -563,23 +563,23 @@ typedef struct bfloat16x8x4_t {
563563
564564#ifdef __LITTLE_ENDIAN__
565565#define splatq_lane_p16(__p0, __p1) __extension__ ({ \
566 poly16x4_t __s0 = __p0; \
567566 poly16x8_t __ret; \
567 poly16x4_t __s0 = __p0; \
568568 __ret = (poly16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 5); \
569569 __ret; \
570570})
571571#else
572572#define splatq_lane_p16(__p0, __p1) __extension__ ({ \
573 poly16x8_t __ret; \
573574 poly16x4_t __s0 = __p0; \
574575 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
575 poly16x8_t __ret; \
576576 __ret = (poly16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 5); \
577577 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
578578 __ret; \
579579})
580580#define __noswap_splatq_lane_p16(__p0, __p1) __extension__ ({ \
581 poly16x4_t __s0 = __p0; \
582581 poly16x8_t __ret; \
582 poly16x4_t __s0 = __p0; \
583583 __ret = (poly16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 5); \
584584 __ret; \
585585})
......@@ -587,23 +587,23 @@ typedef struct bfloat16x8x4_t {
587587
588588#ifdef __LITTLE_ENDIAN__
589589#define splatq_lane_u8(__p0, __p1) __extension__ ({ \
590 uint8x8_t __s0 = __p0; \
591590 uint8x16_t __ret; \
591 uint8x8_t __s0 = __p0; \
592592 __ret = (uint8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 16); \
593593 __ret; \
594594})
595595#else
596596#define splatq_lane_u8(__p0, __p1) __extension__ ({ \
597 uint8x16_t __ret; \
597598 uint8x8_t __s0 = __p0; \
598599 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
599 uint8x16_t __ret; \
600600 __ret = (uint8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 16); \
601601 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
602602 __ret; \
603603})
604604#define __noswap_splatq_lane_u8(__p0, __p1) __extension__ ({ \
605 uint8x8_t __s0 = __p0; \
606605 uint8x16_t __ret; \
606 uint8x8_t __s0 = __p0; \
607607 __ret = (uint8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 16); \
608608 __ret; \
609609})
......@@ -611,23 +611,23 @@ typedef struct bfloat16x8x4_t {
611611
612612#ifdef __LITTLE_ENDIAN__
613613#define splatq_lane_u32(__p0, __p1) __extension__ ({ \
614 uint32x2_t __s0 = __p0; \
615614 uint32x4_t __ret; \
615 uint32x2_t __s0 = __p0; \
616616 __ret = (uint32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 18); \
617617 __ret; \
618618})
619619#else
620620#define splatq_lane_u32(__p0, __p1) __extension__ ({ \
621 uint32x4_t __ret; \
621622 uint32x2_t __s0 = __p0; \
622623 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
623 uint32x4_t __ret; \
624624 __ret = (uint32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 18); \
625625 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
626626 __ret; \
627627})
628628#define __noswap_splatq_lane_u32(__p0, __p1) __extension__ ({ \
629 uint32x2_t __s0 = __p0; \
630629 uint32x4_t __ret; \
630 uint32x2_t __s0 = __p0; \
631631 __ret = (uint32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 18); \
632632 __ret; \
633633})
......@@ -635,22 +635,22 @@ typedef struct bfloat16x8x4_t {
635635
636636#ifdef __LITTLE_ENDIAN__
637637#define splatq_lane_u64(__p0, __p1) __extension__ ({ \
638 uint64x1_t __s0 = __p0; \
639638 uint64x2_t __ret; \
639 uint64x1_t __s0 = __p0; \
640640 __ret = (uint64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 19); \
641641 __ret; \
642642})
643643#else
644644#define splatq_lane_u64(__p0, __p1) __extension__ ({ \
645 uint64x1_t __s0 = __p0; \
646645 uint64x2_t __ret; \
646 uint64x1_t __s0 = __p0; \
647647 __ret = (uint64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 19); \
648648 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
649649 __ret; \
650650})
651651#define __noswap_splatq_lane_u64(__p0, __p1) __extension__ ({ \
652 uint64x1_t __s0 = __p0; \
653652 uint64x2_t __ret; \
653 uint64x1_t __s0 = __p0; \
654654 __ret = (uint64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 19); \
655655 __ret; \
656656})
......@@ -658,23 +658,23 @@ typedef struct bfloat16x8x4_t {
658658
659659#ifdef __LITTLE_ENDIAN__
660660#define splatq_lane_u16(__p0, __p1) __extension__ ({ \
661 uint16x4_t __s0 = __p0; \
662661 uint16x8_t __ret; \
662 uint16x4_t __s0 = __p0; \
663663 __ret = (uint16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 17); \
664664 __ret; \
665665})
666666#else
667667#define splatq_lane_u16(__p0, __p1) __extension__ ({ \
668 uint16x8_t __ret; \
668669 uint16x4_t __s0 = __p0; \
669670 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
670 uint16x8_t __ret; \
671671 __ret = (uint16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 17); \
672672 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
673673 __ret; \
674674})
675675#define __noswap_splatq_lane_u16(__p0, __p1) __extension__ ({ \
676 uint16x4_t __s0 = __p0; \
677676 uint16x8_t __ret; \
677 uint16x4_t __s0 = __p0; \
678678 __ret = (uint16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 17); \
679679 __ret; \
680680})
......@@ -682,23 +682,23 @@ typedef struct bfloat16x8x4_t {
682682
683683#ifdef __LITTLE_ENDIAN__
684684#define splatq_lane_s8(__p0, __p1) __extension__ ({ \
685 int8x8_t __s0 = __p0; \
686685 int8x16_t __ret; \
686 int8x8_t __s0 = __p0; \
687687 __ret = (int8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 0); \
688688 __ret; \
689689})
690690#else
691691#define splatq_lane_s8(__p0, __p1) __extension__ ({ \
692 int8x16_t __ret; \
692693 int8x8_t __s0 = __p0; \
693694 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
694 int8x16_t __ret; \
695695 __ret = (int8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 0); \
696696 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
697697 __ret; \
698698})
699699#define __noswap_splatq_lane_s8(__p0, __p1) __extension__ ({ \
700 int8x8_t __s0 = __p0; \
701700 int8x16_t __ret; \
701 int8x8_t __s0 = __p0; \
702702 __ret = (int8x16_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 0); \
703703 __ret; \
704704})
......@@ -706,22 +706,22 @@ typedef struct bfloat16x8x4_t {
706706
707707#ifdef __LITTLE_ENDIAN__
708708#define splatq_lane_f64(__p0, __p1) __extension__ ({ \
709 float64x1_t __s0 = __p0; \
710709 float64x2_t __ret; \
710 float64x1_t __s0 = __p0; \
711711 __ret = (float64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 10); \
712712 __ret; \
713713})
714714#else
715715#define splatq_lane_f64(__p0, __p1) __extension__ ({ \
716 float64x1_t __s0 = __p0; \
717716 float64x2_t __ret; \
717 float64x1_t __s0 = __p0; \
718718 __ret = (float64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 10); \
719719 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
720720 __ret; \
721721})
722722#define __noswap_splatq_lane_f64(__p0, __p1) __extension__ ({ \
723 float64x1_t __s0 = __p0; \
724723 float64x2_t __ret; \
724 float64x1_t __s0 = __p0; \
725725 __ret = (float64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 10); \
726726 __ret; \
727727})
......@@ -729,23 +729,23 @@ typedef struct bfloat16x8x4_t {
729729
730730#ifdef __LITTLE_ENDIAN__
731731#define splatq_lane_f32(__p0, __p1) __extension__ ({ \
732 float32x2_t __s0 = __p0; \
733732 float32x4_t __ret; \
733 float32x2_t __s0 = __p0; \
734734 __ret = (float32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 9); \
735735 __ret; \
736736})
737737#else
738738#define splatq_lane_f32(__p0, __p1) __extension__ ({ \
739 float32x4_t __ret; \
739740 float32x2_t __s0 = __p0; \
740741 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
741 float32x4_t __ret; \
742742 __ret = (float32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 9); \
743743 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
744744 __ret; \
745745})
746746#define __noswap_splatq_lane_f32(__p0, __p1) __extension__ ({ \
747 float32x2_t __s0 = __p0; \
748747 float32x4_t __ret; \
748 float32x2_t __s0 = __p0; \
749749 __ret = (float32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 9); \
750750 __ret; \
751751})
......@@ -753,23 +753,23 @@ typedef struct bfloat16x8x4_t {
753753
754754#ifdef __LITTLE_ENDIAN__
755755#define splatq_lane_f16(__p0, __p1) __extension__ ({ \
756 float16x4_t __s0 = __p0; \
757756 float16x8_t __ret; \
757 float16x4_t __s0 = __p0; \
758758 __ret = (float16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 8); \
759759 __ret; \
760760})
761761#else
762762#define splatq_lane_f16(__p0, __p1) __extension__ ({ \
763 float16x8_t __ret; \
763764 float16x4_t __s0 = __p0; \
764765 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
765 float16x8_t __ret; \
766766 __ret = (float16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 8); \
767767 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
768768 __ret; \
769769})
770770#define __noswap_splatq_lane_f16(__p0, __p1) __extension__ ({ \
771 float16x4_t __s0 = __p0; \
772771 float16x8_t __ret; \
772 float16x4_t __s0 = __p0; \
773773 __ret = (float16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 8); \
774774 __ret; \
775775})
......@@ -777,23 +777,23 @@ typedef struct bfloat16x8x4_t {
777777
778778#ifdef __LITTLE_ENDIAN__
779779#define splatq_lane_s32(__p0, __p1) __extension__ ({ \
780 int32x2_t __s0 = __p0; \
781780 int32x4_t __ret; \
781 int32x2_t __s0 = __p0; \
782782 __ret = (int32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 2); \
783783 __ret; \
784784})
785785#else
786786#define splatq_lane_s32(__p0, __p1) __extension__ ({ \
787 int32x4_t __ret; \
787788 int32x2_t __s0 = __p0; \
788789 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
789 int32x4_t __ret; \
790790 __ret = (int32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 2); \
791791 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
792792 __ret; \
793793})
794794#define __noswap_splatq_lane_s32(__p0, __p1) __extension__ ({ \
795 int32x2_t __s0 = __p0; \
796795 int32x4_t __ret; \
796 int32x2_t __s0 = __p0; \
797797 __ret = (int32x4_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 2); \
798798 __ret; \
799799})
......@@ -801,22 +801,22 @@ typedef struct bfloat16x8x4_t {
801801
802802#ifdef __LITTLE_ENDIAN__
803803#define splatq_lane_s64(__p0, __p1) __extension__ ({ \
804 int64x1_t __s0 = __p0; \
805804 int64x2_t __ret; \
805 int64x1_t __s0 = __p0; \
806806 __ret = (int64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 3); \
807807 __ret; \
808808})
809809#else
810810#define splatq_lane_s64(__p0, __p1) __extension__ ({ \
811 int64x1_t __s0 = __p0; \
812811 int64x2_t __ret; \
812 int64x1_t __s0 = __p0; \
813813 __ret = (int64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 3); \
814814 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
815815 __ret; \
816816})
817817#define __noswap_splatq_lane_s64(__p0, __p1) __extension__ ({ \
818 int64x1_t __s0 = __p0; \
819818 int64x2_t __ret; \
819 int64x1_t __s0 = __p0; \
820820 __ret = (int64x2_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 3); \
821821 __ret; \
822822})
......@@ -824,23 +824,23 @@ typedef struct bfloat16x8x4_t {
824824
825825#ifdef __LITTLE_ENDIAN__
826826#define splatq_lane_s16(__p0, __p1) __extension__ ({ \
827 int16x4_t __s0 = __p0; \
828827 int16x8_t __ret; \
828 int16x4_t __s0 = __p0; \
829829 __ret = (int16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 1); \
830830 __ret; \
831831})
832832#else
833833#define splatq_lane_s16(__p0, __p1) __extension__ ({ \
834 int16x8_t __ret; \
834835 int16x4_t __s0 = __p0; \
835836 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
836 int16x8_t __ret; \
837837 __ret = (int16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 1); \
838838 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
839839 __ret; \
840840})
841841#define __noswap_splatq_lane_s16(__p0, __p1) __extension__ ({ \
842 int16x4_t __s0 = __p0; \
843842 int16x8_t __ret; \
843 int16x4_t __s0 = __p0; \
844844 __ret = (int16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 1); \
845845 __ret; \
846846})
......@@ -848,23 +848,23 @@ typedef struct bfloat16x8x4_t {
848848
849849#ifdef __LITTLE_ENDIAN__
850850#define splat_lane_u8(__p0, __p1) __extension__ ({ \
851 uint8x8_t __s0 = __p0; \
852851 uint8x8_t __ret; \
852 uint8x8_t __s0 = __p0; \
853853 __ret = (uint8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 16); \
854854 __ret; \
855855})
856856#else
857857#define splat_lane_u8(__p0, __p1) __extension__ ({ \
858 uint8x8_t __ret; \
858859 uint8x8_t __s0 = __p0; \
859860 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
860 uint8x8_t __ret; \
861861 __ret = (uint8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 16); \
862862 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
863863 __ret; \
864864})
865865#define __noswap_splat_lane_u8(__p0, __p1) __extension__ ({ \
866 uint8x8_t __s0 = __p0; \
867866 uint8x8_t __ret; \
867 uint8x8_t __s0 = __p0; \
868868 __ret = (uint8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 16); \
869869 __ret; \
870870})
......@@ -872,53 +872,53 @@ typedef struct bfloat16x8x4_t {
872872
873873#ifdef __LITTLE_ENDIAN__
874874#define splat_lane_u32(__p0, __p1) __extension__ ({ \
875 uint32x2_t __s0 = __p0; \
876875 uint32x2_t __ret; \
876 uint32x2_t __s0 = __p0; \
877877 __ret = (uint32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 18); \
878878 __ret; \
879879})
880880#else
881881#define splat_lane_u32(__p0, __p1) __extension__ ({ \
882 uint32x2_t __ret; \
882883 uint32x2_t __s0 = __p0; \
883884 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
884 uint32x2_t __ret; \
885885 __ret = (uint32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 18); \
886886 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
887887 __ret; \
888888})
889889#define __noswap_splat_lane_u32(__p0, __p1) __extension__ ({ \
890 uint32x2_t __s0 = __p0; \
891890 uint32x2_t __ret; \
891 uint32x2_t __s0 = __p0; \
892892 __ret = (uint32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 18); \
893893 __ret; \
894894})
895895#endif
896896
897897#define splat_lane_u64(__p0, __p1) __extension__ ({ \
898 uint64x1_t __s0 = __p0; \
899898 uint64x1_t __ret; \
899 uint64x1_t __s0 = __p0; \
900900 __ret = (uint64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 19); \
901901 __ret; \
902902})
903903#ifdef __LITTLE_ENDIAN__
904904#define splat_lane_u16(__p0, __p1) __extension__ ({ \
905 uint16x4_t __s0 = __p0; \
906905 uint16x4_t __ret; \
906 uint16x4_t __s0 = __p0; \
907907 __ret = (uint16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 17); \
908908 __ret; \
909909})
910910#else
911911#define splat_lane_u16(__p0, __p1) __extension__ ({ \
912 uint16x4_t __ret; \
912913 uint16x4_t __s0 = __p0; \
913914 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
914 uint16x4_t __ret; \
915915 __ret = (uint16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 17); \
916916 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
917917 __ret; \
918918})
919919#define __noswap_splat_lane_u16(__p0, __p1) __extension__ ({ \
920 uint16x4_t __s0 = __p0; \
921920 uint16x4_t __ret; \
921 uint16x4_t __s0 = __p0; \
922922 __ret = (uint16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 17); \
923923 __ret; \
924924})
......@@ -926,53 +926,53 @@ typedef struct bfloat16x8x4_t {
926926
927927#ifdef __LITTLE_ENDIAN__
928928#define splat_lane_s8(__p0, __p1) __extension__ ({ \
929 int8x8_t __s0 = __p0; \
930929 int8x8_t __ret; \
930 int8x8_t __s0 = __p0; \
931931 __ret = (int8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 0); \
932932 __ret; \
933933})
934934#else
935935#define splat_lane_s8(__p0, __p1) __extension__ ({ \
936 int8x8_t __ret; \
936937 int8x8_t __s0 = __p0; \
937938 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
938 int8x8_t __ret; \
939939 __ret = (int8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 0); \
940940 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
941941 __ret; \
942942})
943943#define __noswap_splat_lane_s8(__p0, __p1) __extension__ ({ \
944 int8x8_t __s0 = __p0; \
945944 int8x8_t __ret; \
945 int8x8_t __s0 = __p0; \
946946 __ret = (int8x8_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 0); \
947947 __ret; \
948948})
949949#endif
950950
951951#define splat_lane_f64(__p0, __p1) __extension__ ({ \
952 float64x1_t __s0 = __p0; \
953952 float64x1_t __ret; \
953 float64x1_t __s0 = __p0; \
954954 __ret = (float64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 10); \
955955 __ret; \
956956})
957957#ifdef __LITTLE_ENDIAN__
958958#define splat_lane_f32(__p0, __p1) __extension__ ({ \
959 float32x2_t __s0 = __p0; \
960959 float32x2_t __ret; \
960 float32x2_t __s0 = __p0; \
961961 __ret = (float32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 9); \
962962 __ret; \
963963})
964964#else
965965#define splat_lane_f32(__p0, __p1) __extension__ ({ \
966 float32x2_t __ret; \
966967 float32x2_t __s0 = __p0; \
967968 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
968 float32x2_t __ret; \
969969 __ret = (float32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 9); \
970970 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
971971 __ret; \
972972})
973973#define __noswap_splat_lane_f32(__p0, __p1) __extension__ ({ \
974 float32x2_t __s0 = __p0; \
975974 float32x2_t __ret; \
975 float32x2_t __s0 = __p0; \
976976 __ret = (float32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 9); \
977977 __ret; \
978978})
......@@ -980,23 +980,23 @@ typedef struct bfloat16x8x4_t {
980980
981981#ifdef __LITTLE_ENDIAN__
982982#define splat_lane_f16(__p0, __p1) __extension__ ({ \
983 float16x4_t __s0 = __p0; \
984983 float16x4_t __ret; \
984 float16x4_t __s0 = __p0; \
985985 __ret = (float16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 8); \
986986 __ret; \
987987})
988988#else
989989#define splat_lane_f16(__p0, __p1) __extension__ ({ \
990 float16x4_t __ret; \
990991 float16x4_t __s0 = __p0; \
991992 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
992 float16x4_t __ret; \
993993 __ret = (float16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 8); \
994994 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
995995 __ret; \
996996})
997997#define __noswap_splat_lane_f16(__p0, __p1) __extension__ ({ \
998 float16x4_t __s0 = __p0; \
999998 float16x4_t __ret; \
999 float16x4_t __s0 = __p0; \
10001000 __ret = (float16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 8); \
10011001 __ret; \
10021002})
......@@ -1004,53 +1004,53 @@ typedef struct bfloat16x8x4_t {
10041004
10051005#ifdef __LITTLE_ENDIAN__
10061006#define splat_lane_s32(__p0, __p1) __extension__ ({ \
1007 int32x2_t __s0 = __p0; \
10081007 int32x2_t __ret; \
1008 int32x2_t __s0 = __p0; \
10091009 __ret = (int32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 2); \
10101010 __ret; \
10111011})
10121012#else
10131013#define splat_lane_s32(__p0, __p1) __extension__ ({ \
1014 int32x2_t __ret; \
10141015 int32x2_t __s0 = __p0; \
10151016 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1016 int32x2_t __ret; \
10171017 __ret = (int32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 2); \
10181018 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
10191019 __ret; \
10201020})
10211021#define __noswap_splat_lane_s32(__p0, __p1) __extension__ ({ \
1022 int32x2_t __s0 = __p0; \
10231022 int32x2_t __ret; \
1023 int32x2_t __s0 = __p0; \
10241024 __ret = (int32x2_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 2); \
10251025 __ret; \
10261026})
10271027#endif
10281028
10291029#define splat_lane_s64(__p0, __p1) __extension__ ({ \
1030 int64x1_t __s0 = __p0; \
10311030 int64x1_t __ret; \
1031 int64x1_t __s0 = __p0; \
10321032 __ret = (int64x1_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 3); \
10331033 __ret; \
10341034})
10351035#ifdef __LITTLE_ENDIAN__
10361036#define splat_lane_s16(__p0, __p1) __extension__ ({ \
1037 int16x4_t __s0 = __p0; \
10381037 int16x4_t __ret; \
1038 int16x4_t __s0 = __p0; \
10391039 __ret = (int16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 1); \
10401040 __ret; \
10411041})
10421042#else
10431043#define splat_lane_s16(__p0, __p1) __extension__ ({ \
1044 int16x4_t __ret; \
10441045 int16x4_t __s0 = __p0; \
10451046 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1046 int16x4_t __ret; \
10471047 __ret = (int16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 1); \
10481048 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
10491049 __ret; \
10501050})
10511051#define __noswap_splat_lane_s16(__p0, __p1) __extension__ ({ \
1052 int16x4_t __s0 = __p0; \
10531052 int16x4_t __ret; \
1053 int16x4_t __s0 = __p0; \
10541054 __ret = (int16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 1); \
10551055 __ret; \
10561056})
......@@ -1058,23 +1058,23 @@ typedef struct bfloat16x8x4_t {
10581058
10591059#ifdef __LITTLE_ENDIAN__
10601060#define splat_laneq_p8(__p0, __p1) __extension__ ({ \
1061 poly8x16_t __s0 = __p0; \
10621061 poly8x8_t __ret; \
1062 poly8x16_t __s0 = __p0; \
10631063 __ret = (poly8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 36); \
10641064 __ret; \
10651065})
10661066#else
10671067#define splat_laneq_p8(__p0, __p1) __extension__ ({ \
1068 poly8x8_t __ret; \
10681069 poly8x16_t __s0 = __p0; \
10691070 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
1070 poly8x8_t __ret; \
10711071 __ret = (poly8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 36); \
10721072 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
10731073 __ret; \
10741074})
10751075#define __noswap_splat_laneq_p8(__p0, __p1) __extension__ ({ \
1076 poly8x16_t __s0 = __p0; \
10771076 poly8x8_t __ret; \
1077 poly8x16_t __s0 = __p0; \
10781078 __ret = (poly8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 36); \
10791079 __ret; \
10801080})
......@@ -1082,22 +1082,22 @@ typedef struct bfloat16x8x4_t {
10821082
10831083#ifdef __LITTLE_ENDIAN__
10841084#define splat_laneq_p64(__p0, __p1) __extension__ ({ \
1085 poly64x2_t __s0 = __p0; \
10861085 poly64x1_t __ret; \
1086 poly64x2_t __s0 = __p0; \
10871087 __ret = (poly64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 38); \
10881088 __ret; \
10891089})
10901090#else
10911091#define splat_laneq_p64(__p0, __p1) __extension__ ({ \
1092 poly64x1_t __ret; \
10921093 poly64x2_t __s0 = __p0; \
10931094 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1094 poly64x1_t __ret; \
10951095 __ret = (poly64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 38); \
10961096 __ret; \
10971097})
10981098#define __noswap_splat_laneq_p64(__p0, __p1) __extension__ ({ \
1099 poly64x2_t __s0 = __p0; \
11001099 poly64x1_t __ret; \
1100 poly64x2_t __s0 = __p0; \
11011101 __ret = (poly64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 38); \
11021102 __ret; \
11031103})
......@@ -1105,23 +1105,23 @@ typedef struct bfloat16x8x4_t {
11051105
11061106#ifdef __LITTLE_ENDIAN__
11071107#define splat_laneq_p16(__p0, __p1) __extension__ ({ \
1108 poly16x8_t __s0 = __p0; \
11091108 poly16x4_t __ret; \
1109 poly16x8_t __s0 = __p0; \
11101110 __ret = (poly16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 37); \
11111111 __ret; \
11121112})
11131113#else
11141114#define splat_laneq_p16(__p0, __p1) __extension__ ({ \
1115 poly16x4_t __ret; \
11151116 poly16x8_t __s0 = __p0; \
11161117 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1117 poly16x4_t __ret; \
11181118 __ret = (poly16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 37); \
11191119 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
11201120 __ret; \
11211121})
11221122#define __noswap_splat_laneq_p16(__p0, __p1) __extension__ ({ \
1123 poly16x8_t __s0 = __p0; \
11241123 poly16x4_t __ret; \
1124 poly16x8_t __s0 = __p0; \
11251125 __ret = (poly16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 37); \
11261126 __ret; \
11271127})
......@@ -1129,23 +1129,23 @@ typedef struct bfloat16x8x4_t {
11291129
11301130#ifdef __LITTLE_ENDIAN__
11311131#define splatq_laneq_p8(__p0, __p1) __extension__ ({ \
1132 poly8x16_t __s0 = __p0; \
11331132 poly8x16_t __ret; \
1133 poly8x16_t __s0 = __p0; \
11341134 __ret = (poly8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 36); \
11351135 __ret; \
11361136})
11371137#else
11381138#define splatq_laneq_p8(__p0, __p1) __extension__ ({ \
1139 poly8x16_t __ret; \
11391140 poly8x16_t __s0 = __p0; \
11401141 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
1141 poly8x16_t __ret; \
11421142 __ret = (poly8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 36); \
11431143 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
11441144 __ret; \
11451145})
11461146#define __noswap_splatq_laneq_p8(__p0, __p1) __extension__ ({ \
1147 poly8x16_t __s0 = __p0; \
11481147 poly8x16_t __ret; \
1148 poly8x16_t __s0 = __p0; \
11491149 __ret = (poly8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 36); \
11501150 __ret; \
11511151})
......@@ -1153,23 +1153,23 @@ typedef struct bfloat16x8x4_t {
11531153
11541154#ifdef __LITTLE_ENDIAN__
11551155#define splatq_laneq_p64(__p0, __p1) __extension__ ({ \
1156 poly64x2_t __s0 = __p0; \
11571156 poly64x2_t __ret; \
1157 poly64x2_t __s0 = __p0; \
11581158 __ret = (poly64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 38); \
11591159 __ret; \
11601160})
11611161#else
11621162#define splatq_laneq_p64(__p0, __p1) __extension__ ({ \
1163 poly64x2_t __ret; \
11631164 poly64x2_t __s0 = __p0; \
11641165 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1165 poly64x2_t __ret; \
11661166 __ret = (poly64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 38); \
11671167 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
11681168 __ret; \
11691169})
11701170#define __noswap_splatq_laneq_p64(__p0, __p1) __extension__ ({ \
1171 poly64x2_t __s0 = __p0; \
11721171 poly64x2_t __ret; \
1172 poly64x2_t __s0 = __p0; \
11731173 __ret = (poly64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 38); \
11741174 __ret; \
11751175})
......@@ -1177,23 +1177,23 @@ typedef struct bfloat16x8x4_t {
11771177
11781178#ifdef __LITTLE_ENDIAN__
11791179#define splatq_laneq_p16(__p0, __p1) __extension__ ({ \
1180 poly16x8_t __s0 = __p0; \
11811180 poly16x8_t __ret; \
1181 poly16x8_t __s0 = __p0; \
11821182 __ret = (poly16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 37); \
11831183 __ret; \
11841184})
11851185#else
11861186#define splatq_laneq_p16(__p0, __p1) __extension__ ({ \
1187 poly16x8_t __ret; \
11871188 poly16x8_t __s0 = __p0; \
11881189 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1189 poly16x8_t __ret; \
11901190 __ret = (poly16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 37); \
11911191 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
11921192 __ret; \
11931193})
11941194#define __noswap_splatq_laneq_p16(__p0, __p1) __extension__ ({ \
1195 poly16x8_t __s0 = __p0; \
11961195 poly16x8_t __ret; \
1196 poly16x8_t __s0 = __p0; \
11971197 __ret = (poly16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 37); \
11981198 __ret; \
11991199})
......@@ -1201,23 +1201,23 @@ typedef struct bfloat16x8x4_t {
12011201
12021202#ifdef __LITTLE_ENDIAN__
12031203#define splatq_laneq_u8(__p0, __p1) __extension__ ({ \
1204 uint8x16_t __s0 = __p0; \
12051204 uint8x16_t __ret; \
1205 uint8x16_t __s0 = __p0; \
12061206 __ret = (uint8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 48); \
12071207 __ret; \
12081208})
12091209#else
12101210#define splatq_laneq_u8(__p0, __p1) __extension__ ({ \
1211 uint8x16_t __ret; \
12111212 uint8x16_t __s0 = __p0; \
12121213 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
1213 uint8x16_t __ret; \
12141214 __ret = (uint8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 48); \
12151215 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
12161216 __ret; \
12171217})
12181218#define __noswap_splatq_laneq_u8(__p0, __p1) __extension__ ({ \
1219 uint8x16_t __s0 = __p0; \
12201219 uint8x16_t __ret; \
1220 uint8x16_t __s0 = __p0; \
12211221 __ret = (uint8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 48); \
12221222 __ret; \
12231223})
......@@ -1225,23 +1225,23 @@ typedef struct bfloat16x8x4_t {
12251225
12261226#ifdef __LITTLE_ENDIAN__
12271227#define splatq_laneq_u32(__p0, __p1) __extension__ ({ \
1228 uint32x4_t __s0 = __p0; \
12291228 uint32x4_t __ret; \
1229 uint32x4_t __s0 = __p0; \
12301230 __ret = (uint32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 50); \
12311231 __ret; \
12321232})
12331233#else
12341234#define splatq_laneq_u32(__p0, __p1) __extension__ ({ \
1235 uint32x4_t __ret; \
12351236 uint32x4_t __s0 = __p0; \
12361237 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1237 uint32x4_t __ret; \
12381238 __ret = (uint32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 50); \
12391239 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
12401240 __ret; \
12411241})
12421242#define __noswap_splatq_laneq_u32(__p0, __p1) __extension__ ({ \
1243 uint32x4_t __s0 = __p0; \
12441243 uint32x4_t __ret; \
1244 uint32x4_t __s0 = __p0; \
12451245 __ret = (uint32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 50); \
12461246 __ret; \
12471247})
......@@ -1249,23 +1249,23 @@ typedef struct bfloat16x8x4_t {
12491249
12501250#ifdef __LITTLE_ENDIAN__
12511251#define splatq_laneq_u64(__p0, __p1) __extension__ ({ \
1252 uint64x2_t __s0 = __p0; \
12531252 uint64x2_t __ret; \
1253 uint64x2_t __s0 = __p0; \
12541254 __ret = (uint64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 51); \
12551255 __ret; \
12561256})
12571257#else
12581258#define splatq_laneq_u64(__p0, __p1) __extension__ ({ \
1259 uint64x2_t __ret; \
12591260 uint64x2_t __s0 = __p0; \
12601261 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1261 uint64x2_t __ret; \
12621262 __ret = (uint64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 51); \
12631263 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
12641264 __ret; \
12651265})
12661266#define __noswap_splatq_laneq_u64(__p0, __p1) __extension__ ({ \
1267 uint64x2_t __s0 = __p0; \
12681267 uint64x2_t __ret; \
1268 uint64x2_t __s0 = __p0; \
12691269 __ret = (uint64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 51); \
12701270 __ret; \
12711271})
......@@ -1273,23 +1273,23 @@ typedef struct bfloat16x8x4_t {
12731273
12741274#ifdef __LITTLE_ENDIAN__
12751275#define splatq_laneq_u16(__p0, __p1) __extension__ ({ \
1276 uint16x8_t __s0 = __p0; \
12771276 uint16x8_t __ret; \
1277 uint16x8_t __s0 = __p0; \
12781278 __ret = (uint16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 49); \
12791279 __ret; \
12801280})
12811281#else
12821282#define splatq_laneq_u16(__p0, __p1) __extension__ ({ \
1283 uint16x8_t __ret; \
12831284 uint16x8_t __s0 = __p0; \
12841285 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1285 uint16x8_t __ret; \
12861286 __ret = (uint16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 49); \
12871287 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
12881288 __ret; \
12891289})
12901290#define __noswap_splatq_laneq_u16(__p0, __p1) __extension__ ({ \
1291 uint16x8_t __s0 = __p0; \
12921291 uint16x8_t __ret; \
1292 uint16x8_t __s0 = __p0; \
12931293 __ret = (uint16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 49); \
12941294 __ret; \
12951295})
......@@ -1297,23 +1297,23 @@ typedef struct bfloat16x8x4_t {
12971297
12981298#ifdef __LITTLE_ENDIAN__
12991299#define splatq_laneq_s8(__p0, __p1) __extension__ ({ \
1300 int8x16_t __s0 = __p0; \
13011300 int8x16_t __ret; \
1301 int8x16_t __s0 = __p0; \
13021302 __ret = (int8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 32); \
13031303 __ret; \
13041304})
13051305#else
13061306#define splatq_laneq_s8(__p0, __p1) __extension__ ({ \
1307 int8x16_t __ret; \
13071308 int8x16_t __s0 = __p0; \
13081309 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
1309 int8x16_t __ret; \
13101310 __ret = (int8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 32); \
13111311 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
13121312 __ret; \
13131313})
13141314#define __noswap_splatq_laneq_s8(__p0, __p1) __extension__ ({ \
1315 int8x16_t __s0 = __p0; \
13161315 int8x16_t __ret; \
1316 int8x16_t __s0 = __p0; \
13171317 __ret = (int8x16_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 32); \
13181318 __ret; \
13191319})
......@@ -1321,23 +1321,23 @@ typedef struct bfloat16x8x4_t {
13211321
13221322#ifdef __LITTLE_ENDIAN__
13231323#define splatq_laneq_f64(__p0, __p1) __extension__ ({ \
1324 float64x2_t __s0 = __p0; \
13251324 float64x2_t __ret; \
1325 float64x2_t __s0 = __p0; \
13261326 __ret = (float64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 42); \
13271327 __ret; \
13281328})
13291329#else
13301330#define splatq_laneq_f64(__p0, __p1) __extension__ ({ \
1331 float64x2_t __ret; \
13311332 float64x2_t __s0 = __p0; \
13321333 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1333 float64x2_t __ret; \
13341334 __ret = (float64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 42); \
13351335 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
13361336 __ret; \
13371337})
13381338#define __noswap_splatq_laneq_f64(__p0, __p1) __extension__ ({ \
1339 float64x2_t __s0 = __p0; \
13401339 float64x2_t __ret; \
1340 float64x2_t __s0 = __p0; \
13411341 __ret = (float64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 42); \
13421342 __ret; \
13431343})
......@@ -1345,23 +1345,23 @@ typedef struct bfloat16x8x4_t {
13451345
13461346#ifdef __LITTLE_ENDIAN__
13471347#define splatq_laneq_f32(__p0, __p1) __extension__ ({ \
1348 float32x4_t __s0 = __p0; \
13491348 float32x4_t __ret; \
1349 float32x4_t __s0 = __p0; \
13501350 __ret = (float32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 41); \
13511351 __ret; \
13521352})
13531353#else
13541354#define splatq_laneq_f32(__p0, __p1) __extension__ ({ \
1355 float32x4_t __ret; \
13551356 float32x4_t __s0 = __p0; \
13561357 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1357 float32x4_t __ret; \
13581358 __ret = (float32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 41); \
13591359 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
13601360 __ret; \
13611361})
13621362#define __noswap_splatq_laneq_f32(__p0, __p1) __extension__ ({ \
1363 float32x4_t __s0 = __p0; \
13641363 float32x4_t __ret; \
1364 float32x4_t __s0 = __p0; \
13651365 __ret = (float32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 41); \
13661366 __ret; \
13671367})
......@@ -1369,23 +1369,23 @@ typedef struct bfloat16x8x4_t {
13691369
13701370#ifdef __LITTLE_ENDIAN__
13711371#define splatq_laneq_f16(__p0, __p1) __extension__ ({ \
1372 float16x8_t __s0 = __p0; \
13731372 float16x8_t __ret; \
1373 float16x8_t __s0 = __p0; \
13741374 __ret = (float16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 40); \
13751375 __ret; \
13761376})
13771377#else
13781378#define splatq_laneq_f16(__p0, __p1) __extension__ ({ \
1379 float16x8_t __ret; \
13791380 float16x8_t __s0 = __p0; \
13801381 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1381 float16x8_t __ret; \
13821382 __ret = (float16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 40); \
13831383 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
13841384 __ret; \
13851385})
13861386#define __noswap_splatq_laneq_f16(__p0, __p1) __extension__ ({ \
1387 float16x8_t __s0 = __p0; \
13881387 float16x8_t __ret; \
1388 float16x8_t __s0 = __p0; \
13891389 __ret = (float16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 40); \
13901390 __ret; \
13911391})
......@@ -1393,23 +1393,23 @@ typedef struct bfloat16x8x4_t {
13931393
13941394#ifdef __LITTLE_ENDIAN__
13951395#define splatq_laneq_s32(__p0, __p1) __extension__ ({ \
1396 int32x4_t __s0 = __p0; \
13971396 int32x4_t __ret; \
1397 int32x4_t __s0 = __p0; \
13981398 __ret = (int32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 34); \
13991399 __ret; \
14001400})
14011401#else
14021402#define splatq_laneq_s32(__p0, __p1) __extension__ ({ \
1403 int32x4_t __ret; \
14031404 int32x4_t __s0 = __p0; \
14041405 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1405 int32x4_t __ret; \
14061406 __ret = (int32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 34); \
14071407 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
14081408 __ret; \
14091409})
14101410#define __noswap_splatq_laneq_s32(__p0, __p1) __extension__ ({ \
1411 int32x4_t __s0 = __p0; \
14121411 int32x4_t __ret; \
1412 int32x4_t __s0 = __p0; \
14131413 __ret = (int32x4_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 34); \
14141414 __ret; \
14151415})
......@@ -1417,23 +1417,23 @@ typedef struct bfloat16x8x4_t {
14171417
14181418#ifdef __LITTLE_ENDIAN__
14191419#define splatq_laneq_s64(__p0, __p1) __extension__ ({ \
1420 int64x2_t __s0 = __p0; \
14211420 int64x2_t __ret; \
1421 int64x2_t __s0 = __p0; \
14221422 __ret = (int64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 35); \
14231423 __ret; \
14241424})
14251425#else
14261426#define splatq_laneq_s64(__p0, __p1) __extension__ ({ \
1427 int64x2_t __ret; \
14271428 int64x2_t __s0 = __p0; \
14281429 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1429 int64x2_t __ret; \
14301430 __ret = (int64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 35); \
14311431 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
14321432 __ret; \
14331433})
14341434#define __noswap_splatq_laneq_s64(__p0, __p1) __extension__ ({ \
1435 int64x2_t __s0 = __p0; \
14361435 int64x2_t __ret; \
1436 int64x2_t __s0 = __p0; \
14371437 __ret = (int64x2_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 35); \
14381438 __ret; \
14391439})
......@@ -1441,23 +1441,23 @@ typedef struct bfloat16x8x4_t {
14411441
14421442#ifdef __LITTLE_ENDIAN__
14431443#define splatq_laneq_s16(__p0, __p1) __extension__ ({ \
1444 int16x8_t __s0 = __p0; \
14451444 int16x8_t __ret; \
1445 int16x8_t __s0 = __p0; \
14461446 __ret = (int16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 33); \
14471447 __ret; \
14481448})
14491449#else
14501450#define splatq_laneq_s16(__p0, __p1) __extension__ ({ \
1451 int16x8_t __ret; \
14511452 int16x8_t __s0 = __p0; \
14521453 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1453 int16x8_t __ret; \
14541454 __ret = (int16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 33); \
14551455 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
14561456 __ret; \
14571457})
14581458#define __noswap_splatq_laneq_s16(__p0, __p1) __extension__ ({ \
1459 int16x8_t __s0 = __p0; \
14601459 int16x8_t __ret; \
1460 int16x8_t __s0 = __p0; \
14611461 __ret = (int16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 33); \
14621462 __ret; \
14631463})
......@@ -1465,23 +1465,23 @@ typedef struct bfloat16x8x4_t {
14651465
14661466#ifdef __LITTLE_ENDIAN__
14671467#define splat_laneq_u8(__p0, __p1) __extension__ ({ \
1468 uint8x16_t __s0 = __p0; \
14691468 uint8x8_t __ret; \
1469 uint8x16_t __s0 = __p0; \
14701470 __ret = (uint8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 48); \
14711471 __ret; \
14721472})
14731473#else
14741474#define splat_laneq_u8(__p0, __p1) __extension__ ({ \
1475 uint8x8_t __ret; \
14751476 uint8x16_t __s0 = __p0; \
14761477 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
1477 uint8x8_t __ret; \
14781478 __ret = (uint8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 48); \
14791479 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
14801480 __ret; \
14811481})
14821482#define __noswap_splat_laneq_u8(__p0, __p1) __extension__ ({ \
1483 uint8x16_t __s0 = __p0; \
14841483 uint8x8_t __ret; \
1484 uint8x16_t __s0 = __p0; \
14851485 __ret = (uint8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 48); \
14861486 __ret; \
14871487})
......@@ -1489,23 +1489,23 @@ typedef struct bfloat16x8x4_t {
14891489
14901490#ifdef __LITTLE_ENDIAN__
14911491#define splat_laneq_u32(__p0, __p1) __extension__ ({ \
1492 uint32x4_t __s0 = __p0; \
14931492 uint32x2_t __ret; \
1493 uint32x4_t __s0 = __p0; \
14941494 __ret = (uint32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 50); \
14951495 __ret; \
14961496})
14971497#else
14981498#define splat_laneq_u32(__p0, __p1) __extension__ ({ \
1499 uint32x2_t __ret; \
14991500 uint32x4_t __s0 = __p0; \
15001501 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1501 uint32x2_t __ret; \
15021502 __ret = (uint32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 50); \
15031503 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
15041504 __ret; \
15051505})
15061506#define __noswap_splat_laneq_u32(__p0, __p1) __extension__ ({ \
1507 uint32x4_t __s0 = __p0; \
15081507 uint32x2_t __ret; \
1508 uint32x4_t __s0 = __p0; \
15091509 __ret = (uint32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 50); \
15101510 __ret; \
15111511})
......@@ -1513,22 +1513,22 @@ typedef struct bfloat16x8x4_t {
15131513
15141514#ifdef __LITTLE_ENDIAN__
15151515#define splat_laneq_u64(__p0, __p1) __extension__ ({ \
1516 uint64x2_t __s0 = __p0; \
15171516 uint64x1_t __ret; \
1517 uint64x2_t __s0 = __p0; \
15181518 __ret = (uint64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 51); \
15191519 __ret; \
15201520})
15211521#else
15221522#define splat_laneq_u64(__p0, __p1) __extension__ ({ \
1523 uint64x1_t __ret; \
15231524 uint64x2_t __s0 = __p0; \
15241525 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1525 uint64x1_t __ret; \
15261526 __ret = (uint64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 51); \
15271527 __ret; \
15281528})
15291529#define __noswap_splat_laneq_u64(__p0, __p1) __extension__ ({ \
1530 uint64x2_t __s0 = __p0; \
15311530 uint64x1_t __ret; \
1531 uint64x2_t __s0 = __p0; \
15321532 __ret = (uint64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 51); \
15331533 __ret; \
15341534})
......@@ -1536,23 +1536,23 @@ typedef struct bfloat16x8x4_t {
15361536
15371537#ifdef __LITTLE_ENDIAN__
15381538#define splat_laneq_u16(__p0, __p1) __extension__ ({ \
1539 uint16x8_t __s0 = __p0; \
15401539 uint16x4_t __ret; \
1540 uint16x8_t __s0 = __p0; \
15411541 __ret = (uint16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 49); \
15421542 __ret; \
15431543})
15441544#else
15451545#define splat_laneq_u16(__p0, __p1) __extension__ ({ \
1546 uint16x4_t __ret; \
15461547 uint16x8_t __s0 = __p0; \
15471548 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1548 uint16x4_t __ret; \
15491549 __ret = (uint16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 49); \
15501550 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
15511551 __ret; \
15521552})
15531553#define __noswap_splat_laneq_u16(__p0, __p1) __extension__ ({ \
1554 uint16x8_t __s0 = __p0; \
15551554 uint16x4_t __ret; \
1555 uint16x8_t __s0 = __p0; \
15561556 __ret = (uint16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 49); \
15571557 __ret; \
15581558})
......@@ -1560,23 +1560,23 @@ typedef struct bfloat16x8x4_t {
15601560
15611561#ifdef __LITTLE_ENDIAN__
15621562#define splat_laneq_s8(__p0, __p1) __extension__ ({ \
1563 int8x16_t __s0 = __p0; \
15641563 int8x8_t __ret; \
1564 int8x16_t __s0 = __p0; \
15651565 __ret = (int8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 32); \
15661566 __ret; \
15671567})
15681568#else
15691569#define splat_laneq_s8(__p0, __p1) __extension__ ({ \
1570 int8x8_t __ret; \
15701571 int8x16_t __s0 = __p0; \
15711572 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
1572 int8x8_t __ret; \
15731573 __ret = (int8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 32); \
15741574 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
15751575 __ret; \
15761576})
15771577#define __noswap_splat_laneq_s8(__p0, __p1) __extension__ ({ \
1578 int8x16_t __s0 = __p0; \
15791578 int8x8_t __ret; \
1579 int8x16_t __s0 = __p0; \
15801580 __ret = (int8x8_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 32); \
15811581 __ret; \
15821582})
......@@ -1584,22 +1584,22 @@ typedef struct bfloat16x8x4_t {
15841584
15851585#ifdef __LITTLE_ENDIAN__
15861586#define splat_laneq_f64(__p0, __p1) __extension__ ({ \
1587 float64x2_t __s0 = __p0; \
15881587 float64x1_t __ret; \
1588 float64x2_t __s0 = __p0; \
15891589 __ret = (float64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 42); \
15901590 __ret; \
15911591})
15921592#else
15931593#define splat_laneq_f64(__p0, __p1) __extension__ ({ \
1594 float64x1_t __ret; \
15941595 float64x2_t __s0 = __p0; \
15951596 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1596 float64x1_t __ret; \
15971597 __ret = (float64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 42); \
15981598 __ret; \
15991599})
16001600#define __noswap_splat_laneq_f64(__p0, __p1) __extension__ ({ \
1601 float64x2_t __s0 = __p0; \
16021601 float64x1_t __ret; \
1602 float64x2_t __s0 = __p0; \
16031603 __ret = (float64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 42); \
16041604 __ret; \
16051605})
......@@ -1607,23 +1607,23 @@ typedef struct bfloat16x8x4_t {
16071607
16081608#ifdef __LITTLE_ENDIAN__
16091609#define splat_laneq_f32(__p0, __p1) __extension__ ({ \
1610 float32x4_t __s0 = __p0; \
16111610 float32x2_t __ret; \
1611 float32x4_t __s0 = __p0; \
16121612 __ret = (float32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 41); \
16131613 __ret; \
16141614})
16151615#else
16161616#define splat_laneq_f32(__p0, __p1) __extension__ ({ \
1617 float32x2_t __ret; \
16171618 float32x4_t __s0 = __p0; \
16181619 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1619 float32x2_t __ret; \
16201620 __ret = (float32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 41); \
16211621 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
16221622 __ret; \
16231623})
16241624#define __noswap_splat_laneq_f32(__p0, __p1) __extension__ ({ \
1625 float32x4_t __s0 = __p0; \
16261625 float32x2_t __ret; \
1626 float32x4_t __s0 = __p0; \
16271627 __ret = (float32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 41); \
16281628 __ret; \
16291629})
......@@ -1631,23 +1631,23 @@ typedef struct bfloat16x8x4_t {
16311631
16321632#ifdef __LITTLE_ENDIAN__
16331633#define splat_laneq_f16(__p0, __p1) __extension__ ({ \
1634 float16x8_t __s0 = __p0; \
16351634 float16x4_t __ret; \
1635 float16x8_t __s0 = __p0; \
16361636 __ret = (float16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 40); \
16371637 __ret; \
16381638})
16391639#else
16401640#define splat_laneq_f16(__p0, __p1) __extension__ ({ \
1641 float16x4_t __ret; \
16411642 float16x8_t __s0 = __p0; \
16421643 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1643 float16x4_t __ret; \
16441644 __ret = (float16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 40); \
16451645 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
16461646 __ret; \
16471647})
16481648#define __noswap_splat_laneq_f16(__p0, __p1) __extension__ ({ \
1649 float16x8_t __s0 = __p0; \
16501649 float16x4_t __ret; \
1650 float16x8_t __s0 = __p0; \
16511651 __ret = (float16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 40); \
16521652 __ret; \
16531653})
......@@ -1655,23 +1655,23 @@ typedef struct bfloat16x8x4_t {
16551655
16561656#ifdef __LITTLE_ENDIAN__
16571657#define splat_laneq_s32(__p0, __p1) __extension__ ({ \
1658 int32x4_t __s0 = __p0; \
16591658 int32x2_t __ret; \
1659 int32x4_t __s0 = __p0; \
16601660 __ret = (int32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 34); \
16611661 __ret; \
16621662})
16631663#else
16641664#define splat_laneq_s32(__p0, __p1) __extension__ ({ \
1665 int32x2_t __ret; \
16651666 int32x4_t __s0 = __p0; \
16661667 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
1667 int32x2_t __ret; \
16681668 __ret = (int32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 34); \
16691669 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
16701670 __ret; \
16711671})
16721672#define __noswap_splat_laneq_s32(__p0, __p1) __extension__ ({ \
1673 int32x4_t __s0 = __p0; \
16741673 int32x2_t __ret; \
1674 int32x4_t __s0 = __p0; \
16751675 __ret = (int32x2_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 34); \
16761676 __ret; \
16771677})
......@@ -1679,22 +1679,22 @@ typedef struct bfloat16x8x4_t {
16791679
16801680#ifdef __LITTLE_ENDIAN__
16811681#define splat_laneq_s64(__p0, __p1) __extension__ ({ \
1682 int64x2_t __s0 = __p0; \
16831682 int64x1_t __ret; \
1683 int64x2_t __s0 = __p0; \
16841684 __ret = (int64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 35); \
16851685 __ret; \
16861686})
16871687#else
16881688#define splat_laneq_s64(__p0, __p1) __extension__ ({ \
1689 int64x1_t __ret; \
16891690 int64x2_t __s0 = __p0; \
16901691 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
1691 int64x1_t __ret; \
16921692 __ret = (int64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 35); \
16931693 __ret; \
16941694})
16951695#define __noswap_splat_laneq_s64(__p0, __p1) __extension__ ({ \
1696 int64x2_t __s0 = __p0; \
16971696 int64x1_t __ret; \
1697 int64x2_t __s0 = __p0; \
16981698 __ret = (int64x1_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 35); \
16991699 __ret; \
17001700})
......@@ -1702,23 +1702,23 @@ typedef struct bfloat16x8x4_t {
17021702
17031703#ifdef __LITTLE_ENDIAN__
17041704#define splat_laneq_s16(__p0, __p1) __extension__ ({ \
1705 int16x8_t __s0 = __p0; \
17061705 int16x4_t __ret; \
1706 int16x8_t __s0 = __p0; \
17071707 __ret = (int16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 33); \
17081708 __ret; \
17091709})
17101710#else
17111711#define splat_laneq_s16(__p0, __p1) __extension__ ({ \
1712 int16x4_t __ret; \
17121713 int16x8_t __s0 = __p0; \
17131714 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
1714 int16x4_t __ret; \
17151715 __ret = (int16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 33); \
17161716 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
17171717 __ret; \
17181718})
17191719#define __noswap_splat_laneq_s16(__p0, __p1) __extension__ ({ \
1720 int16x8_t __s0 = __p0; \
17211720 int16x4_t __ret; \
1721 int16x8_t __s0 = __p0; \
17221722 __ret = (int16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 33); \
17231723 __ret; \
17241724})
......@@ -1732,9 +1732,9 @@ __ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
17321732}
17331733#else
17341734__ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1735 uint8x16_t __ret;
17351736 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
17361737 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1737 uint8x16_t __ret;
17381738 __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
17391739 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
17401740 return __ret;
......@@ -1754,9 +1754,9 @@ __ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
17541754}
17551755#else
17561756__ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1757 uint32x4_t __ret;
17571758 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
17581759 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1759 uint32x4_t __ret;
17601760 __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
17611761 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
17621762 return __ret;
......@@ -1776,9 +1776,9 @@ __ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
17761776}
17771777#else
17781778__ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1779 uint16x8_t __ret;
17791780 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
17801781 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1781 uint16x8_t __ret;
17821782 __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
17831783 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
17841784 return __ret;
......@@ -1798,9 +1798,9 @@ __ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
17981798}
17991799#else
18001800__ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
1801 int8x16_t __ret;
18011802 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18021803 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1803 int8x16_t __ret;
18041804 __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
18051805 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18061806 return __ret;
......@@ -1820,9 +1820,9 @@ __ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
18201820}
18211821#else
18221822__ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
1823 float32x4_t __ret;
18231824 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
18241825 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1825 float32x4_t __ret;
18261826 __ret = (float32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
18271827 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
18281828 return __ret;
......@@ -1837,9 +1837,9 @@ __ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
18371837}
18381838#else
18391839__ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
1840 int32x4_t __ret;
18401841 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
18411842 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1842 int32x4_t __ret;
18431843 __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
18441844 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
18451845 return __ret;
......@@ -1859,9 +1859,9 @@ __ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
18591859}
18601860#else
18611861__ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
1862 int16x8_t __ret;
18621863 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
18631864 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1864 int16x8_t __ret;
18651865 __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
18661866 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
18671867 return __ret;
......@@ -1881,9 +1881,9 @@ __ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
18811881}
18821882#else
18831883__ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
1884 uint8x8_t __ret;
18841885 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
18851886 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1886 uint8x8_t __ret;
18871887 __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
18881888 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
18891889 return __ret;
......@@ -1903,9 +1903,9 @@ __ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
19031903}
19041904#else
19051905__ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
1906 uint32x2_t __ret;
19061907 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
19071908 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1908 uint32x2_t __ret;
19091909 __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
19101910 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
19111911 return __ret;
......@@ -1925,9 +1925,9 @@ __ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
19251925}
19261926#else
19271927__ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
1928 uint16x4_t __ret;
19281929 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
19291930 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1930 uint16x4_t __ret;
19311931 __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
19321932 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
19331933 return __ret;
......@@ -1947,9 +1947,9 @@ __ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
19471947}
19481948#else
19491949__ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
1950 int8x8_t __ret;
19501951 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
19511952 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1952 int8x8_t __ret;
19531953 __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
19541954 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
19551955 return __ret;
......@@ -1969,9 +1969,9 @@ __ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
19691969}
19701970#else
19711971__ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
1972 float32x2_t __ret;
19721973 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
19731974 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1974 float32x2_t __ret;
19751975 __ret = (float32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
19761976 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
19771977 return __ret;
......@@ -1986,9 +1986,9 @@ __ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
19861986}
19871987#else
19881988__ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
1989 int32x2_t __ret;
19891990 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
19901991 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1991 int32x2_t __ret;
19921992 __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
19931993 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
19941994 return __ret;
......@@ -2008,9 +2008,9 @@ __ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
20082008}
20092009#else
20102010__ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
2011 int16x4_t __ret;
20112012 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
20122013 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2013 int16x4_t __ret;
20142014 __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
20152015 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
20162016 return __ret;
......@@ -2030,8 +2030,8 @@ __ai int8x16_t vabsq_s8(int8x16_t __p0) {
20302030}
20312031#else
20322032__ai int8x16_t vabsq_s8(int8x16_t __p0) {
2033 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20342033 int8x16_t __ret;
2034 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20352035 __ret = (int8x16_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 32);
20362036 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20372037 return __ret;
......@@ -2046,8 +2046,8 @@ __ai float32x4_t vabsq_f32(float32x4_t __p0) {
20462046}
20472047#else
20482048__ai float32x4_t vabsq_f32(float32x4_t __p0) {
2049 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
20502049 float32x4_t __ret;
2050 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
20512051 __ret = (float32x4_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 41);
20522052 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
20532053 return __ret;
......@@ -2062,8 +2062,8 @@ __ai int32x4_t vabsq_s32(int32x4_t __p0) {
20622062}
20632063#else
20642064__ai int32x4_t vabsq_s32(int32x4_t __p0) {
2065 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
20662065 int32x4_t __ret;
2066 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
20672067 __ret = (int32x4_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 34);
20682068 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
20692069 return __ret;
......@@ -2078,8 +2078,8 @@ __ai int16x8_t vabsq_s16(int16x8_t __p0) {
20782078}
20792079#else
20802080__ai int16x8_t vabsq_s16(int16x8_t __p0) {
2081 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
20822081 int16x8_t __ret;
2082 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
20832083 __ret = (int16x8_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 33);
20842084 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
20852085 return __ret;
......@@ -2094,8 +2094,8 @@ __ai int8x8_t vabs_s8(int8x8_t __p0) {
20942094}
20952095#else
20962096__ai int8x8_t vabs_s8(int8x8_t __p0) {
2097 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
20982097 int8x8_t __ret;
2098 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
20992099 __ret = (int8x8_t) __builtin_neon_vabs_v((int8x8_t)__rev0, 0);
21002100 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
21012101 return __ret;
......@@ -2110,8 +2110,8 @@ __ai float32x2_t vabs_f32(float32x2_t __p0) {
21102110}
21112111#else
21122112__ai float32x2_t vabs_f32(float32x2_t __p0) {
2113 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
21142113 float32x2_t __ret;
2114 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
21152115 __ret = (float32x2_t) __builtin_neon_vabs_v((int8x8_t)__rev0, 9);
21162116 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
21172117 return __ret;
......@@ -2126,8 +2126,8 @@ __ai int32x2_t vabs_s32(int32x2_t __p0) {
21262126}
21272127#else
21282128__ai int32x2_t vabs_s32(int32x2_t __p0) {
2129 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
21302129 int32x2_t __ret;
2130 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
21312131 __ret = (int32x2_t) __builtin_neon_vabs_v((int8x8_t)__rev0, 2);
21322132 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
21332133 return __ret;
......@@ -2142,8 +2142,8 @@ __ai int16x4_t vabs_s16(int16x4_t __p0) {
21422142}
21432143#else
21442144__ai int16x4_t vabs_s16(int16x4_t __p0) {
2145 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
21462145 int16x4_t __ret;
2146 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
21472147 __ret = (int16x4_t) __builtin_neon_vabs_v((int8x8_t)__rev0, 1);
21482148 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
21492149 return __ret;
......@@ -2158,9 +2158,9 @@ __ai uint8x16_t vaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
21582158}
21592159#else
21602160__ai uint8x16_t vaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2161 uint8x16_t __ret;
21612162 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
21622163 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2163 uint8x16_t __ret;
21642164 __ret = __rev0 + __rev1;
21652165 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
21662166 return __ret;
......@@ -2175,9 +2175,9 @@ __ai uint32x4_t vaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
21752175}
21762176#else
21772177__ai uint32x4_t vaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2178 uint32x4_t __ret;
21782179 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
21792180 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2180 uint32x4_t __ret;
21812181 __ret = __rev0 + __rev1;
21822182 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
21832183 return __ret;
......@@ -2192,9 +2192,9 @@ __ai uint64x2_t vaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
21922192}
21932193#else
21942194__ai uint64x2_t vaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
2195 uint64x2_t __ret;
21952196 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
21962197 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2197 uint64x2_t __ret;
21982198 __ret = __rev0 + __rev1;
21992199 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
22002200 return __ret;
......@@ -2209,9 +2209,9 @@ __ai uint16x8_t vaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
22092209}
22102210#else
22112211__ai uint16x8_t vaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2212 uint16x8_t __ret;
22122213 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
22132214 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2214 uint16x8_t __ret;
22152215 __ret = __rev0 + __rev1;
22162216 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
22172217 return __ret;
......@@ -2226,9 +2226,9 @@ __ai int8x16_t vaddq_s8(int8x16_t __p0, int8x16_t __p1) {
22262226}
22272227#else
22282228__ai int8x16_t vaddq_s8(int8x16_t __p0, int8x16_t __p1) {
2229 int8x16_t __ret;
22292230 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22302231 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2231 int8x16_t __ret;
22322232 __ret = __rev0 + __rev1;
22332233 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22342234 return __ret;
......@@ -2243,9 +2243,9 @@ __ai float32x4_t vaddq_f32(float32x4_t __p0, float32x4_t __p1) {
22432243}
22442244#else
22452245__ai float32x4_t vaddq_f32(float32x4_t __p0, float32x4_t __p1) {
2246 float32x4_t __ret;
22462247 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
22472248 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2248 float32x4_t __ret;
22492249 __ret = __rev0 + __rev1;
22502250 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
22512251 return __ret;
......@@ -2260,9 +2260,9 @@ __ai int32x4_t vaddq_s32(int32x4_t __p0, int32x4_t __p1) {
22602260}
22612261#else
22622262__ai int32x4_t vaddq_s32(int32x4_t __p0, int32x4_t __p1) {
2263 int32x4_t __ret;
22632264 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
22642265 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2265 int32x4_t __ret;
22662266 __ret = __rev0 + __rev1;
22672267 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
22682268 return __ret;
......@@ -2277,9 +2277,9 @@ __ai int64x2_t vaddq_s64(int64x2_t __p0, int64x2_t __p1) {
22772277}
22782278#else
22792279__ai int64x2_t vaddq_s64(int64x2_t __p0, int64x2_t __p1) {
2280 int64x2_t __ret;
22802281 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
22812282 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2282 int64x2_t __ret;
22832283 __ret = __rev0 + __rev1;
22842284 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
22852285 return __ret;
......@@ -2294,9 +2294,9 @@ __ai int16x8_t vaddq_s16(int16x8_t __p0, int16x8_t __p1) {
22942294}
22952295#else
22962296__ai int16x8_t vaddq_s16(int16x8_t __p0, int16x8_t __p1) {
2297 int16x8_t __ret;
22972298 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
22982299 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2299 int16x8_t __ret;
23002300 __ret = __rev0 + __rev1;
23012301 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
23022302 return __ret;
......@@ -2311,9 +2311,9 @@ __ai uint8x8_t vadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
23112311}
23122312#else
23132313__ai uint8x8_t vadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
2314 uint8x8_t __ret;
23142315 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
23152316 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2316 uint8x8_t __ret;
23172317 __ret = __rev0 + __rev1;
23182318 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
23192319 return __ret;
......@@ -2328,9 +2328,9 @@ __ai uint32x2_t vadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
23282328}
23292329#else
23302330__ai uint32x2_t vadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
2331 uint32x2_t __ret;
23312332 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
23322333 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2333 uint32x2_t __ret;
23342334 __ret = __rev0 + __rev1;
23352335 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
23362336 return __ret;
......@@ -2350,9 +2350,9 @@ __ai uint16x4_t vadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
23502350}
23512351#else
23522352__ai uint16x4_t vadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
2353 uint16x4_t __ret;
23532354 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
23542355 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2355 uint16x4_t __ret;
23562356 __ret = __rev0 + __rev1;
23572357 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
23582358 return __ret;
......@@ -2367,9 +2367,9 @@ __ai int8x8_t vadd_s8(int8x8_t __p0, int8x8_t __p1) {
23672367}
23682368#else
23692369__ai int8x8_t vadd_s8(int8x8_t __p0, int8x8_t __p1) {
2370 int8x8_t __ret;
23702371 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
23712372 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2372 int8x8_t __ret;
23732373 __ret = __rev0 + __rev1;
23742374 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
23752375 return __ret;
......@@ -2384,9 +2384,9 @@ __ai float32x2_t vadd_f32(float32x2_t __p0, float32x2_t __p1) {
23842384}
23852385#else
23862386__ai float32x2_t vadd_f32(float32x2_t __p0, float32x2_t __p1) {
2387 float32x2_t __ret;
23872388 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
23882389 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2389 float32x2_t __ret;
23902390 __ret = __rev0 + __rev1;
23912391 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
23922392 return __ret;
......@@ -2401,9 +2401,9 @@ __ai int32x2_t vadd_s32(int32x2_t __p0, int32x2_t __p1) {
24012401}
24022402#else
24032403__ai int32x2_t vadd_s32(int32x2_t __p0, int32x2_t __p1) {
2404 int32x2_t __ret;
24042405 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
24052406 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2406 int32x2_t __ret;
24072407 __ret = __rev0 + __rev1;
24082408 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
24092409 return __ret;
......@@ -2423,9 +2423,9 @@ __ai int16x4_t vadd_s16(int16x4_t __p0, int16x4_t __p1) {
24232423}
24242424#else
24252425__ai int16x4_t vadd_s16(int16x4_t __p0, int16x4_t __p1) {
2426 int16x4_t __ret;
24262427 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
24272428 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2428 int16x4_t __ret;
24292429 __ret = __rev0 + __rev1;
24302430 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
24312431 return __ret;
......@@ -2440,9 +2440,9 @@ __ai poly8x8_t vadd_p8(poly8x8_t __p0, poly8x8_t __p1) {
24402440}
24412441#else
24422442__ai poly8x8_t vadd_p8(poly8x8_t __p0, poly8x8_t __p1) {
2443 poly8x8_t __ret;
24432444 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
24442445 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2445 poly8x8_t __ret;
24462446 __ret = (poly8x8_t) __builtin_neon_vadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 4);
24472447 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
24482448 return __ret;
......@@ -2462,9 +2462,9 @@ __ai poly16x4_t vadd_p16(poly16x4_t __p0, poly16x4_t __p1) {
24622462}
24632463#else
24642464__ai poly16x4_t vadd_p16(poly16x4_t __p0, poly16x4_t __p1) {
2465 poly16x4_t __ret;
24652466 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
24662467 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2467 poly16x4_t __ret;
24682468 __ret = (poly16x4_t) __builtin_neon_vadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 5);
24692469 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
24702470 return __ret;
......@@ -2479,9 +2479,9 @@ __ai poly8x16_t vaddq_p8(poly8x16_t __p0, poly8x16_t __p1) {
24792479}
24802480#else
24812481__ai poly8x16_t vaddq_p8(poly8x16_t __p0, poly8x16_t __p1) {
2482 poly8x16_t __ret;
24822483 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
24832484 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2484 poly8x16_t __ret;
24852485 __ret = (poly8x16_t) __builtin_neon_vaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 36);
24862486 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
24872487 return __ret;
......@@ -2496,9 +2496,9 @@ __ai poly64x2_t vaddq_p64(poly64x2_t __p0, poly64x2_t __p1) {
24962496}
24972497#else
24982498__ai poly64x2_t vaddq_p64(poly64x2_t __p0, poly64x2_t __p1) {
2499 poly64x2_t __ret;
24992500 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
25002501 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2501 poly64x2_t __ret;
25022502 __ret = (poly64x2_t) __builtin_neon_vaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 38);
25032503 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
25042504 return __ret;
......@@ -2513,9 +2513,9 @@ __ai poly16x8_t vaddq_p16(poly16x8_t __p0, poly16x8_t __p1) {
25132513}
25142514#else
25152515__ai poly16x8_t vaddq_p16(poly16x8_t __p0, poly16x8_t __p1) {
2516 poly16x8_t __ret;
25162517 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
25172518 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2518 poly16x8_t __ret;
25192519 __ret = (poly16x8_t) __builtin_neon_vaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 37);
25202520 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
25212521 return __ret;
......@@ -2530,9 +2530,9 @@ __ai uint16x4_t vaddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
25302530}
25312531#else
25322532__ai uint16x4_t vaddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
2533 uint16x4_t __ret;
25332534 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
25342535 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2535 uint16x4_t __ret;
25362536 __ret = (uint16x4_t) __builtin_neon_vaddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 17);
25372537 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
25382538 return __ret;
......@@ -2552,9 +2552,9 @@ __ai uint32x2_t vaddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
25522552}
25532553#else
25542554__ai uint32x2_t vaddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
2555 uint32x2_t __ret;
25552556 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
25562557 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2557 uint32x2_t __ret;
25582558 __ret = (uint32x2_t) __builtin_neon_vaddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 18);
25592559 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
25602560 return __ret;
......@@ -2574,9 +2574,9 @@ __ai uint8x8_t vaddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
25742574}
25752575#else
25762576__ai uint8x8_t vaddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
2577 uint8x8_t __ret;
25772578 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
25782579 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2579 uint8x8_t __ret;
25802580 __ret = (uint8x8_t) __builtin_neon_vaddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 16);
25812581 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
25822582 return __ret;
......@@ -2596,9 +2596,9 @@ __ai int16x4_t vaddhn_s32(int32x4_t __p0, int32x4_t __p1) {
25962596}
25972597#else
25982598__ai int16x4_t vaddhn_s32(int32x4_t __p0, int32x4_t __p1) {
2599 int16x4_t __ret;
25992600 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
26002601 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2601 int16x4_t __ret;
26022602 __ret = (int16x4_t) __builtin_neon_vaddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 1);
26032603 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
26042604 return __ret;
......@@ -2618,9 +2618,9 @@ __ai int32x2_t vaddhn_s64(int64x2_t __p0, int64x2_t __p1) {
26182618}
26192619#else
26202620__ai int32x2_t vaddhn_s64(int64x2_t __p0, int64x2_t __p1) {
2621 int32x2_t __ret;
26212622 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
26222623 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2623 int32x2_t __ret;
26242624 __ret = (int32x2_t) __builtin_neon_vaddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 2);
26252625 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
26262626 return __ret;
......@@ -2640,9 +2640,9 @@ __ai int8x8_t vaddhn_s16(int16x8_t __p0, int16x8_t __p1) {
26402640}
26412641#else
26422642__ai int8x8_t vaddhn_s16(int16x8_t __p0, int16x8_t __p1) {
2643 int8x8_t __ret;
26432644 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
26442645 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2645 int8x8_t __ret;
26462646 __ret = (int8x8_t) __builtin_neon_vaddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 0);
26472647 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
26482648 return __ret;
......@@ -2662,9 +2662,9 @@ __ai uint8x16_t vandq_u8(uint8x16_t __p0, uint8x16_t __p1) {
26622662}
26632663#else
26642664__ai uint8x16_t vandq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2665 uint8x16_t __ret;
26652666 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
26662667 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2667 uint8x16_t __ret;
26682668 __ret = __rev0 & __rev1;
26692669 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
26702670 return __ret;
......@@ -2679,9 +2679,9 @@ __ai uint32x4_t vandq_u32(uint32x4_t __p0, uint32x4_t __p1) {
26792679}
26802680#else
26812681__ai uint32x4_t vandq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2682 uint32x4_t __ret;
26822683 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
26832684 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2684 uint32x4_t __ret;
26852685 __ret = __rev0 & __rev1;
26862686 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
26872687 return __ret;
......@@ -2696,9 +2696,9 @@ __ai uint64x2_t vandq_u64(uint64x2_t __p0, uint64x2_t __p1) {
26962696}
26972697#else
26982698__ai uint64x2_t vandq_u64(uint64x2_t __p0, uint64x2_t __p1) {
2699 uint64x2_t __ret;
26992700 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
27002701 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2701 uint64x2_t __ret;
27022702 __ret = __rev0 & __rev1;
27032703 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
27042704 return __ret;
......@@ -2713,9 +2713,9 @@ __ai uint16x8_t vandq_u16(uint16x8_t __p0, uint16x8_t __p1) {
27132713}
27142714#else
27152715__ai uint16x8_t vandq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2716 uint16x8_t __ret;
27162717 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
27172718 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2718 uint16x8_t __ret;
27192719 __ret = __rev0 & __rev1;
27202720 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
27212721 return __ret;
......@@ -2730,9 +2730,9 @@ __ai int8x16_t vandq_s8(int8x16_t __p0, int8x16_t __p1) {
27302730}
27312731#else
27322732__ai int8x16_t vandq_s8(int8x16_t __p0, int8x16_t __p1) {
2733 int8x16_t __ret;
27332734 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
27342735 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2735 int8x16_t __ret;
27362736 __ret = __rev0 & __rev1;
27372737 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
27382738 return __ret;
......@@ -2747,9 +2747,9 @@ __ai int32x4_t vandq_s32(int32x4_t __p0, int32x4_t __p1) {
27472747}
27482748#else
27492749__ai int32x4_t vandq_s32(int32x4_t __p0, int32x4_t __p1) {
2750 int32x4_t __ret;
27502751 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
27512752 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2752 int32x4_t __ret;
27532753 __ret = __rev0 & __rev1;
27542754 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
27552755 return __ret;
......@@ -2764,9 +2764,9 @@ __ai int64x2_t vandq_s64(int64x2_t __p0, int64x2_t __p1) {
27642764}
27652765#else
27662766__ai int64x2_t vandq_s64(int64x2_t __p0, int64x2_t __p1) {
2767 int64x2_t __ret;
27672768 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
27682769 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2769 int64x2_t __ret;
27702770 __ret = __rev0 & __rev1;
27712771 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
27722772 return __ret;
......@@ -2781,9 +2781,9 @@ __ai int16x8_t vandq_s16(int16x8_t __p0, int16x8_t __p1) {
27812781}
27822782#else
27832783__ai int16x8_t vandq_s16(int16x8_t __p0, int16x8_t __p1) {
2784 int16x8_t __ret;
27842785 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
27852786 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2786 int16x8_t __ret;
27872787 __ret = __rev0 & __rev1;
27882788 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
27892789 return __ret;
......@@ -2798,9 +2798,9 @@ __ai uint8x8_t vand_u8(uint8x8_t __p0, uint8x8_t __p1) {
27982798}
27992799#else
28002800__ai uint8x8_t vand_u8(uint8x8_t __p0, uint8x8_t __p1) {
2801 uint8x8_t __ret;
28012802 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
28022803 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2803 uint8x8_t __ret;
28042804 __ret = __rev0 & __rev1;
28052805 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
28062806 return __ret;
......@@ -2815,9 +2815,9 @@ __ai uint32x2_t vand_u32(uint32x2_t __p0, uint32x2_t __p1) {
28152815}
28162816#else
28172817__ai uint32x2_t vand_u32(uint32x2_t __p0, uint32x2_t __p1) {
2818 uint32x2_t __ret;
28182819 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
28192820 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2820 uint32x2_t __ret;
28212821 __ret = __rev0 & __rev1;
28222822 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
28232823 return __ret;
......@@ -2837,9 +2837,9 @@ __ai uint16x4_t vand_u16(uint16x4_t __p0, uint16x4_t __p1) {
28372837}
28382838#else
28392839__ai uint16x4_t vand_u16(uint16x4_t __p0, uint16x4_t __p1) {
2840 uint16x4_t __ret;
28402841 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
28412842 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2842 uint16x4_t __ret;
28432843 __ret = __rev0 & __rev1;
28442844 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
28452845 return __ret;
......@@ -2854,9 +2854,9 @@ __ai int8x8_t vand_s8(int8x8_t __p0, int8x8_t __p1) {
28542854}
28552855#else
28562856__ai int8x8_t vand_s8(int8x8_t __p0, int8x8_t __p1) {
2857 int8x8_t __ret;
28572858 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
28582859 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2859 int8x8_t __ret;
28602860 __ret = __rev0 & __rev1;
28612861 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
28622862 return __ret;
......@@ -2871,9 +2871,9 @@ __ai int32x2_t vand_s32(int32x2_t __p0, int32x2_t __p1) {
28712871}
28722872#else
28732873__ai int32x2_t vand_s32(int32x2_t __p0, int32x2_t __p1) {
2874 int32x2_t __ret;
28742875 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
28752876 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2876 int32x2_t __ret;
28772877 __ret = __rev0 & __rev1;
28782878 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
28792879 return __ret;
......@@ -2893,9 +2893,9 @@ __ai int16x4_t vand_s16(int16x4_t __p0, int16x4_t __p1) {
28932893}
28942894#else
28952895__ai int16x4_t vand_s16(int16x4_t __p0, int16x4_t __p1) {
2896 int16x4_t __ret;
28962897 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
28972898 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2898 int16x4_t __ret;
28992899 __ret = __rev0 & __rev1;
29002900 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
29012901 return __ret;
......@@ -2910,9 +2910,9 @@ __ai uint8x16_t vbicq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29102910}
29112911#else
29122912__ai uint8x16_t vbicq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2913 uint8x16_t __ret;
29132914 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29142915 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2915 uint8x16_t __ret;
29162916 __ret = __rev0 & ~__rev1;
29172917 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29182918 return __ret;
......@@ -2927,9 +2927,9 @@ __ai uint32x4_t vbicq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29272927}
29282928#else
29292929__ai uint32x4_t vbicq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2930 uint32x4_t __ret;
29302931 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
29312932 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2932 uint32x4_t __ret;
29332933 __ret = __rev0 & ~__rev1;
29342934 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
29352935 return __ret;
......@@ -2944,9 +2944,9 @@ __ai uint64x2_t vbicq_u64(uint64x2_t __p0, uint64x2_t __p1) {
29442944}
29452945#else
29462946__ai uint64x2_t vbicq_u64(uint64x2_t __p0, uint64x2_t __p1) {
2947 uint64x2_t __ret;
29472948 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
29482949 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2949 uint64x2_t __ret;
29502950 __ret = __rev0 & ~__rev1;
29512951 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
29522952 return __ret;
......@@ -2961,9 +2961,9 @@ __ai uint16x8_t vbicq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29612961}
29622962#else
29632963__ai uint16x8_t vbicq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2964 uint16x8_t __ret;
29642965 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
29652966 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2966 uint16x8_t __ret;
29672967 __ret = __rev0 & ~__rev1;
29682968 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
29692969 return __ret;
......@@ -2978,9 +2978,9 @@ __ai int8x16_t vbicq_s8(int8x16_t __p0, int8x16_t __p1) {
29782978}
29792979#else
29802980__ai int8x16_t vbicq_s8(int8x16_t __p0, int8x16_t __p1) {
2981 int8x16_t __ret;
29812982 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29822983 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2983 int8x16_t __ret;
29842984 __ret = __rev0 & ~__rev1;
29852985 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29862986 return __ret;
......@@ -2995,9 +2995,9 @@ __ai int32x4_t vbicq_s32(int32x4_t __p0, int32x4_t __p1) {
29952995}
29962996#else
29972997__ai int32x4_t vbicq_s32(int32x4_t __p0, int32x4_t __p1) {
2998 int32x4_t __ret;
29982999 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
29993000 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3000 int32x4_t __ret;
30013001 __ret = __rev0 & ~__rev1;
30023002 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
30033003 return __ret;
......@@ -3012,9 +3012,9 @@ __ai int64x2_t vbicq_s64(int64x2_t __p0, int64x2_t __p1) {
30123012}
30133013#else
30143014__ai int64x2_t vbicq_s64(int64x2_t __p0, int64x2_t __p1) {
3015 int64x2_t __ret;
30153016 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
30163017 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3017 int64x2_t __ret;
30183018 __ret = __rev0 & ~__rev1;
30193019 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
30203020 return __ret;
......@@ -3029,9 +3029,9 @@ __ai int16x8_t vbicq_s16(int16x8_t __p0, int16x8_t __p1) {
30293029}
30303030#else
30313031__ai int16x8_t vbicq_s16(int16x8_t __p0, int16x8_t __p1) {
3032 int16x8_t __ret;
30323033 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
30333034 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3034 int16x8_t __ret;
30353035 __ret = __rev0 & ~__rev1;
30363036 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
30373037 return __ret;
......@@ -3046,9 +3046,9 @@ __ai uint8x8_t vbic_u8(uint8x8_t __p0, uint8x8_t __p1) {
30463046}
30473047#else
30483048__ai uint8x8_t vbic_u8(uint8x8_t __p0, uint8x8_t __p1) {
3049 uint8x8_t __ret;
30493050 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
30503051 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3051 uint8x8_t __ret;
30523052 __ret = __rev0 & ~__rev1;
30533053 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
30543054 return __ret;
......@@ -3063,9 +3063,9 @@ __ai uint32x2_t vbic_u32(uint32x2_t __p0, uint32x2_t __p1) {
30633063}
30643064#else
30653065__ai uint32x2_t vbic_u32(uint32x2_t __p0, uint32x2_t __p1) {
3066 uint32x2_t __ret;
30663067 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
30673068 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3068 uint32x2_t __ret;
30693069 __ret = __rev0 & ~__rev1;
30703070 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
30713071 return __ret;
......@@ -3085,9 +3085,9 @@ __ai uint16x4_t vbic_u16(uint16x4_t __p0, uint16x4_t __p1) {
30853085}
30863086#else
30873087__ai uint16x4_t vbic_u16(uint16x4_t __p0, uint16x4_t __p1) {
3088 uint16x4_t __ret;
30883089 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
30893090 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3090 uint16x4_t __ret;
30913091 __ret = __rev0 & ~__rev1;
30923092 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
30933093 return __ret;
......@@ -3102,9 +3102,9 @@ __ai int8x8_t vbic_s8(int8x8_t __p0, int8x8_t __p1) {
31023102}
31033103#else
31043104__ai int8x8_t vbic_s8(int8x8_t __p0, int8x8_t __p1) {
3105 int8x8_t __ret;
31053106 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
31063107 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3107 int8x8_t __ret;
31083108 __ret = __rev0 & ~__rev1;
31093109 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
31103110 return __ret;
......@@ -3119,9 +3119,9 @@ __ai int32x2_t vbic_s32(int32x2_t __p0, int32x2_t __p1) {
31193119}
31203120#else
31213121__ai int32x2_t vbic_s32(int32x2_t __p0, int32x2_t __p1) {
3122 int32x2_t __ret;
31223123 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
31233124 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3124 int32x2_t __ret;
31253125 __ret = __rev0 & ~__rev1;
31263126 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
31273127 return __ret;
......@@ -3141,9 +3141,9 @@ __ai int16x4_t vbic_s16(int16x4_t __p0, int16x4_t __p1) {
31413141}
31423142#else
31433143__ai int16x4_t vbic_s16(int16x4_t __p0, int16x4_t __p1) {
3144 int16x4_t __ret;
31443145 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
31453146 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3146 int16x4_t __ret;
31473147 __ret = __rev0 & ~__rev1;
31483148 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
31493149 return __ret;
......@@ -3158,10 +3158,10 @@ __ai poly8x8_t vbsl_p8(uint8x8_t __p0, poly8x8_t __p1, poly8x8_t __p2) {
31583158}
31593159#else
31603160__ai poly8x8_t vbsl_p8(uint8x8_t __p0, poly8x8_t __p1, poly8x8_t __p2) {
3161 poly8x8_t __ret;
31613162 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
31623163 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31633164 poly8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
3164 poly8x8_t __ret;
31653165 __ret = (poly8x8_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 4);
31663166 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
31673167 return __ret;
......@@ -3176,10 +3176,10 @@ __ai poly16x4_t vbsl_p16(uint16x4_t __p0, poly16x4_t __p1, poly16x4_t __p2) {
31763176}
31773177#else
31783178__ai poly16x4_t vbsl_p16(uint16x4_t __p0, poly16x4_t __p1, poly16x4_t __p2) {
3179 poly16x4_t __ret;
31793180 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
31803181 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31813182 poly16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
3182 poly16x4_t __ret;
31833183 __ret = (poly16x4_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 5);
31843184 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
31853185 return __ret;
......@@ -3194,10 +3194,10 @@ __ai poly8x16_t vbslq_p8(uint8x16_t __p0, poly8x16_t __p1, poly8x16_t __p2) {
31943194}
31953195#else
31963196__ai poly8x16_t vbslq_p8(uint8x16_t __p0, poly8x16_t __p1, poly8x16_t __p2) {
3197 poly8x16_t __ret;
31973198 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31983199 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31993200 poly8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3200 poly8x16_t __ret;
32013201 __ret = (poly8x16_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 36);
32023202 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
32033203 return __ret;
......@@ -3212,10 +3212,10 @@ __ai poly16x8_t vbslq_p16(uint16x8_t __p0, poly16x8_t __p1, poly16x8_t __p2) {
32123212}
32133213#else
32143214__ai poly16x8_t vbslq_p16(uint16x8_t __p0, poly16x8_t __p1, poly16x8_t __p2) {
3215 poly16x8_t __ret;
32153216 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
32163217 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
32173218 poly16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
3218 poly16x8_t __ret;
32193219 __ret = (poly16x8_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 37);
32203220 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
32213221 return __ret;
......@@ -3230,10 +3230,10 @@ __ai uint8x16_t vbslq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
32303230}
32313231#else
32323232__ai uint8x16_t vbslq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
3233 uint8x16_t __ret;
32333234 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
32343235 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
32353236 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3236 uint8x16_t __ret;
32373237 __ret = (uint8x16_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
32383238 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
32393239 return __ret;
......@@ -3248,10 +3248,10 @@ __ai uint32x4_t vbslq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
32483248}
32493249#else
32503250__ai uint32x4_t vbslq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
3251 uint32x4_t __ret;
32513252 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
32523253 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
32533254 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
3254 uint32x4_t __ret;
32553255 __ret = (uint32x4_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
32563256 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
32573257 return __ret;
......@@ -3266,10 +3266,10 @@ __ai uint64x2_t vbslq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
32663266}
32673267#else
32683268__ai uint64x2_t vbslq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
3269 uint64x2_t __ret;
32693270 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
32703271 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
32713272 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
3272 uint64x2_t __ret;
32733273 __ret = (uint64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
32743274 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
32753275 return __ret;
......@@ -3284,10 +3284,10 @@ __ai uint16x8_t vbslq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
32843284}
32853285#else
32863286__ai uint16x8_t vbslq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
3287 uint16x8_t __ret;
32873288 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
32883289 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
32893290 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
3290 uint16x8_t __ret;
32913291 __ret = (uint16x8_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
32923292 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
32933293 return __ret;
......@@ -3302,10 +3302,10 @@ __ai int8x16_t vbslq_s8(uint8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
33023302}
33033303#else
33043304__ai int8x16_t vbslq_s8(uint8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
3305 int8x16_t __ret;
33053306 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
33063307 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
33073308 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3308 int8x16_t __ret;
33093309 __ret = (int8x16_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
33103310 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
33113311 return __ret;
......@@ -3320,10 +3320,10 @@ __ai float32x4_t vbslq_f32(uint32x4_t __p0, float32x4_t __p1, float32x4_t __p2)
33203320}
33213321#else
33223322__ai float32x4_t vbslq_f32(uint32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
3323 float32x4_t __ret;
33233324 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
33243325 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
33253326 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
3326 float32x4_t __ret;
33273327 __ret = (float32x4_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
33283328 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33293329 return __ret;
......@@ -3338,10 +3338,10 @@ __ai int32x4_t vbslq_s32(uint32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
33383338}
33393339#else
33403340__ai int32x4_t vbslq_s32(uint32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
3341 int32x4_t __ret;
33413342 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
33423343 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
33433344 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
3344 int32x4_t __ret;
33453345 __ret = (int32x4_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
33463346 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
33473347 return __ret;
......@@ -3356,10 +3356,10 @@ __ai int64x2_t vbslq_s64(uint64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
33563356}
33573357#else
33583358__ai int64x2_t vbslq_s64(uint64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
3359 int64x2_t __ret;
33593360 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
33603361 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
33613362 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
3362 int64x2_t __ret;
33633363 __ret = (int64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
33643364 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
33653365 return __ret;
......@@ -3374,10 +3374,10 @@ __ai int16x8_t vbslq_s16(uint16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
33743374}
33753375#else
33763376__ai int16x8_t vbslq_s16(uint16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
3377 int16x8_t __ret;
33773378 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
33783379 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
33793380 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
3380 int16x8_t __ret;
33813381 __ret = (int16x8_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
33823382 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
33833383 return __ret;
......@@ -3392,10 +3392,10 @@ __ai uint8x8_t vbsl_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
33923392}
33933393#else
33943394__ai uint8x8_t vbsl_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
3395 uint8x8_t __ret;
33953396 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
33963397 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
33973398 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
3398 uint8x8_t __ret;
33993399 __ret = (uint8x8_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 16);
34003400 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
34013401 return __ret;
......@@ -3410,10 +3410,10 @@ __ai uint32x2_t vbsl_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
34103410}
34113411#else
34123412__ai uint32x2_t vbsl_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
3413 uint32x2_t __ret;
34133414 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34143415 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
34153416 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
3416 uint32x2_t __ret;
34173417 __ret = (uint32x2_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 18);
34183418 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
34193419 return __ret;
......@@ -3433,10 +3433,10 @@ __ai uint16x4_t vbsl_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
34333433}
34343434#else
34353435__ai uint16x4_t vbsl_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
3436 uint16x4_t __ret;
34363437 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
34373438 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
34383439 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
3439 uint16x4_t __ret;
34403440 __ret = (uint16x4_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 17);
34413441 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
34423442 return __ret;
......@@ -3451,10 +3451,10 @@ __ai int8x8_t vbsl_s8(uint8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
34513451}
34523452#else
34533453__ai int8x8_t vbsl_s8(uint8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
3454 int8x8_t __ret;
34543455 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
34553456 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
34563457 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
3457 int8x8_t __ret;
34583458 __ret = (int8x8_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 0);
34593459 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
34603460 return __ret;
......@@ -3469,10 +3469,10 @@ __ai float32x2_t vbsl_f32(uint32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
34693469}
34703470#else
34713471__ai float32x2_t vbsl_f32(uint32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
3472 float32x2_t __ret;
34723473 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34733474 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
34743475 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
3475 float32x2_t __ret;
34763476 __ret = (float32x2_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
34773477 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
34783478 return __ret;
......@@ -3487,10 +3487,10 @@ __ai int32x2_t vbsl_s32(uint32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
34873487}
34883488#else
34893489__ai int32x2_t vbsl_s32(uint32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
3490 int32x2_t __ret;
34903491 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
34913492 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
34923493 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
3493 int32x2_t __ret;
34943494 __ret = (int32x2_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
34953495 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
34963496 return __ret;
......@@ -3510,10 +3510,10 @@ __ai int16x4_t vbsl_s16(uint16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
35103510}
35113511#else
35123512__ai int16x4_t vbsl_s16(uint16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
3513 int16x4_t __ret;
35133514 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35143515 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35153516 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
3516 int16x4_t __ret;
35173517 __ret = (int16x4_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
35183518 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35193519 return __ret;
......@@ -3528,9 +3528,9 @@ __ai uint32x4_t vcageq_f32(float32x4_t __p0, float32x4_t __p1) {
35283528}
35293529#else
35303530__ai uint32x4_t vcageq_f32(float32x4_t __p0, float32x4_t __p1) {
3531 uint32x4_t __ret;
35313532 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35323533 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3533 uint32x4_t __ret;
35343534 __ret = (uint32x4_t) __builtin_neon_vcageq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
35353535 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35363536 return __ret;
......@@ -3545,9 +3545,9 @@ __ai uint32x2_t vcage_f32(float32x2_t __p0, float32x2_t __p1) {
35453545}
35463546#else
35473547__ai uint32x2_t vcage_f32(float32x2_t __p0, float32x2_t __p1) {
3548 uint32x2_t __ret;
35483549 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35493550 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3550 uint32x2_t __ret;
35513551 __ret = (uint32x2_t) __builtin_neon_vcage_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
35523552 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35533553 return __ret;
......@@ -3562,9 +3562,9 @@ __ai uint32x4_t vcagtq_f32(float32x4_t __p0, float32x4_t __p1) {
35623562}
35633563#else
35643564__ai uint32x4_t vcagtq_f32(float32x4_t __p0, float32x4_t __p1) {
3565 uint32x4_t __ret;
35653566 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
35663567 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3567 uint32x4_t __ret;
35683568 __ret = (uint32x4_t) __builtin_neon_vcagtq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
35693569 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
35703570 return __ret;
......@@ -3579,9 +3579,9 @@ __ai uint32x2_t vcagt_f32(float32x2_t __p0, float32x2_t __p1) {
35793579}
35803580#else
35813581__ai uint32x2_t vcagt_f32(float32x2_t __p0, float32x2_t __p1) {
3582 uint32x2_t __ret;
35823583 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
35833584 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3584 uint32x2_t __ret;
35853585 __ret = (uint32x2_t) __builtin_neon_vcagt_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
35863586 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
35873587 return __ret;
......@@ -3596,9 +3596,9 @@ __ai uint32x4_t vcaleq_f32(float32x4_t __p0, float32x4_t __p1) {
35963596}
35973597#else
35983598__ai uint32x4_t vcaleq_f32(float32x4_t __p0, float32x4_t __p1) {
3599 uint32x4_t __ret;
35993600 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36003601 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3601 uint32x4_t __ret;
36023602 __ret = (uint32x4_t) __builtin_neon_vcaleq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
36033603 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36043604 return __ret;
......@@ -3613,9 +3613,9 @@ __ai uint32x2_t vcale_f32(float32x2_t __p0, float32x2_t __p1) {
36133613}
36143614#else
36153615__ai uint32x2_t vcale_f32(float32x2_t __p0, float32x2_t __p1) {
3616 uint32x2_t __ret;
36163617 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36173618 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3618 uint32x2_t __ret;
36193619 __ret = (uint32x2_t) __builtin_neon_vcale_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
36203620 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36213621 return __ret;
......@@ -3630,9 +3630,9 @@ __ai uint32x4_t vcaltq_f32(float32x4_t __p0, float32x4_t __p1) {
36303630}
36313631#else
36323632__ai uint32x4_t vcaltq_f32(float32x4_t __p0, float32x4_t __p1) {
3633 uint32x4_t __ret;
36333634 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
36343635 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3635 uint32x4_t __ret;
36363636 __ret = (uint32x4_t) __builtin_neon_vcaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
36373637 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
36383638 return __ret;
......@@ -3647,9 +3647,9 @@ __ai uint32x2_t vcalt_f32(float32x2_t __p0, float32x2_t __p1) {
36473647}
36483648#else
36493649__ai uint32x2_t vcalt_f32(float32x2_t __p0, float32x2_t __p1) {
3650 uint32x2_t __ret;
36503651 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
36513652 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3652 uint32x2_t __ret;
36533653 __ret = (uint32x2_t) __builtin_neon_vcalt_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
36543654 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
36553655 return __ret;
......@@ -3664,9 +3664,9 @@ __ai uint8x8_t vceq_p8(poly8x8_t __p0, poly8x8_t __p1) {
36643664}
36653665#else
36663666__ai uint8x8_t vceq_p8(poly8x8_t __p0, poly8x8_t __p1) {
3667 uint8x8_t __ret;
36673668 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
36683669 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3669 uint8x8_t __ret;
36703670 __ret = (uint8x8_t)(__rev0 == __rev1);
36713671 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
36723672 return __ret;
......@@ -3681,9 +3681,9 @@ __ai uint8x16_t vceqq_p8(poly8x16_t __p0, poly8x16_t __p1) {
36813681}
36823682#else
36833683__ai uint8x16_t vceqq_p8(poly8x16_t __p0, poly8x16_t __p1) {
3684 uint8x16_t __ret;
36843685 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36853686 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3686 uint8x16_t __ret;
36873687 __ret = (uint8x16_t)(__rev0 == __rev1);
36883688 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36893689 return __ret;
......@@ -3698,9 +3698,9 @@ __ai uint8x16_t vceqq_u8(uint8x16_t __p0, uint8x16_t __p1) {
36983698}
36993699#else
37003700__ai uint8x16_t vceqq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3701 uint8x16_t __ret;
37013702 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
37023703 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3703 uint8x16_t __ret;
37043704 __ret = (uint8x16_t)(__rev0 == __rev1);
37053705 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
37063706 return __ret;
......@@ -3715,9 +3715,9 @@ __ai uint32x4_t vceqq_u32(uint32x4_t __p0, uint32x4_t __p1) {
37153715}
37163716#else
37173717__ai uint32x4_t vceqq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3718 uint32x4_t __ret;
37183719 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
37193720 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3720 uint32x4_t __ret;
37213721 __ret = (uint32x4_t)(__rev0 == __rev1);
37223722 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
37233723 return __ret;
......@@ -3732,9 +3732,9 @@ __ai uint16x8_t vceqq_u16(uint16x8_t __p0, uint16x8_t __p1) {
37323732}
37333733#else
37343734__ai uint16x8_t vceqq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3735 uint16x8_t __ret;
37353736 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
37363737 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3737 uint16x8_t __ret;
37383738 __ret = (uint16x8_t)(__rev0 == __rev1);
37393739 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
37403740 return __ret;
......@@ -3749,9 +3749,9 @@ __ai uint8x16_t vceqq_s8(int8x16_t __p0, int8x16_t __p1) {
37493749}
37503750#else
37513751__ai uint8x16_t vceqq_s8(int8x16_t __p0, int8x16_t __p1) {
3752 uint8x16_t __ret;
37523753 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
37533754 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3754 uint8x16_t __ret;
37553755 __ret = (uint8x16_t)(__rev0 == __rev1);
37563756 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
37573757 return __ret;
......@@ -3766,9 +3766,9 @@ __ai uint32x4_t vceqq_f32(float32x4_t __p0, float32x4_t __p1) {
37663766}
37673767#else
37683768__ai uint32x4_t vceqq_f32(float32x4_t __p0, float32x4_t __p1) {
3769 uint32x4_t __ret;
37693770 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
37703771 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3771 uint32x4_t __ret;
37723772 __ret = (uint32x4_t)(__rev0 == __rev1);
37733773 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
37743774 return __ret;
......@@ -3783,9 +3783,9 @@ __ai uint32x4_t vceqq_s32(int32x4_t __p0, int32x4_t __p1) {
37833783}
37843784#else
37853785__ai uint32x4_t vceqq_s32(int32x4_t __p0, int32x4_t __p1) {
3786 uint32x4_t __ret;
37863787 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
37873788 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3788 uint32x4_t __ret;
37893789 __ret = (uint32x4_t)(__rev0 == __rev1);
37903790 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
37913791 return __ret;
......@@ -3800,9 +3800,9 @@ __ai uint16x8_t vceqq_s16(int16x8_t __p0, int16x8_t __p1) {
38003800}
38013801#else
38023802__ai uint16x8_t vceqq_s16(int16x8_t __p0, int16x8_t __p1) {
3803 uint16x8_t __ret;
38033804 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
38043805 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3805 uint16x8_t __ret;
38063806 __ret = (uint16x8_t)(__rev0 == __rev1);
38073807 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
38083808 return __ret;
......@@ -3817,9 +3817,9 @@ __ai uint8x8_t vceq_u8(uint8x8_t __p0, uint8x8_t __p1) {
38173817}
38183818#else
38193819__ai uint8x8_t vceq_u8(uint8x8_t __p0, uint8x8_t __p1) {
3820 uint8x8_t __ret;
38203821 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
38213822 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3822 uint8x8_t __ret;
38233823 __ret = (uint8x8_t)(__rev0 == __rev1);
38243824 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
38253825 return __ret;
......@@ -3834,9 +3834,9 @@ __ai uint32x2_t vceq_u32(uint32x2_t __p0, uint32x2_t __p1) {
38343834}
38353835#else
38363836__ai uint32x2_t vceq_u32(uint32x2_t __p0, uint32x2_t __p1) {
3837 uint32x2_t __ret;
38373838 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38383839 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3839 uint32x2_t __ret;
38403840 __ret = (uint32x2_t)(__rev0 == __rev1);
38413841 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38423842 return __ret;
......@@ -3851,9 +3851,9 @@ __ai uint16x4_t vceq_u16(uint16x4_t __p0, uint16x4_t __p1) {
38513851}
38523852#else
38533853__ai uint16x4_t vceq_u16(uint16x4_t __p0, uint16x4_t __p1) {
3854 uint16x4_t __ret;
38543855 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
38553856 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3856 uint16x4_t __ret;
38573857 __ret = (uint16x4_t)(__rev0 == __rev1);
38583858 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
38593859 return __ret;
......@@ -3868,9 +3868,9 @@ __ai uint8x8_t vceq_s8(int8x8_t __p0, int8x8_t __p1) {
38683868}
38693869#else
38703870__ai uint8x8_t vceq_s8(int8x8_t __p0, int8x8_t __p1) {
3871 uint8x8_t __ret;
38713872 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
38723873 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3873 uint8x8_t __ret;
38743874 __ret = (uint8x8_t)(__rev0 == __rev1);
38753875 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
38763876 return __ret;
......@@ -3885,9 +3885,9 @@ __ai uint32x2_t vceq_f32(float32x2_t __p0, float32x2_t __p1) {
38853885}
38863886#else
38873887__ai uint32x2_t vceq_f32(float32x2_t __p0, float32x2_t __p1) {
3888 uint32x2_t __ret;
38883889 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
38893890 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3890 uint32x2_t __ret;
38913891 __ret = (uint32x2_t)(__rev0 == __rev1);
38923892 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
38933893 return __ret;
......@@ -3902,9 +3902,9 @@ __ai uint32x2_t vceq_s32(int32x2_t __p0, int32x2_t __p1) {
39023902}
39033903#else
39043904__ai uint32x2_t vceq_s32(int32x2_t __p0, int32x2_t __p1) {
3905 uint32x2_t __ret;
39053906 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
39063907 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3907 uint32x2_t __ret;
39083908 __ret = (uint32x2_t)(__rev0 == __rev1);
39093909 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
39103910 return __ret;
......@@ -3919,9 +3919,9 @@ __ai uint16x4_t vceq_s16(int16x4_t __p0, int16x4_t __p1) {
39193919}
39203920#else
39213921__ai uint16x4_t vceq_s16(int16x4_t __p0, int16x4_t __p1) {
3922 uint16x4_t __ret;
39223923 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39233924 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3924 uint16x4_t __ret;
39253925 __ret = (uint16x4_t)(__rev0 == __rev1);
39263926 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39273927 return __ret;
......@@ -3936,9 +3936,9 @@ __ai uint8x16_t vcgeq_u8(uint8x16_t __p0, uint8x16_t __p1) {
39363936}
39373937#else
39383938__ai uint8x16_t vcgeq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3939 uint8x16_t __ret;
39393940 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39403941 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3941 uint8x16_t __ret;
39423942 __ret = (uint8x16_t)(__rev0 >= __rev1);
39433943 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39443944 return __ret;
......@@ -3953,9 +3953,9 @@ __ai uint32x4_t vcgeq_u32(uint32x4_t __p0, uint32x4_t __p1) {
39533953}
39543954#else
39553955__ai uint32x4_t vcgeq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3956 uint32x4_t __ret;
39563957 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
39573958 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3958 uint32x4_t __ret;
39593959 __ret = (uint32x4_t)(__rev0 >= __rev1);
39603960 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
39613961 return __ret;
......@@ -3970,9 +3970,9 @@ __ai uint16x8_t vcgeq_u16(uint16x8_t __p0, uint16x8_t __p1) {
39703970}
39713971#else
39723972__ai uint16x8_t vcgeq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3973 uint16x8_t __ret;
39733974 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
39743975 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3975 uint16x8_t __ret;
39763976 __ret = (uint16x8_t)(__rev0 >= __rev1);
39773977 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
39783978 return __ret;
......@@ -3987,9 +3987,9 @@ __ai uint8x16_t vcgeq_s8(int8x16_t __p0, int8x16_t __p1) {
39873987}
39883988#else
39893989__ai uint8x16_t vcgeq_s8(int8x16_t __p0, int8x16_t __p1) {
3990 uint8x16_t __ret;
39903991 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39913992 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3992 uint8x16_t __ret;
39933993 __ret = (uint8x16_t)(__rev0 >= __rev1);
39943994 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39953995 return __ret;
......@@ -4004,9 +4004,9 @@ __ai uint32x4_t vcgeq_f32(float32x4_t __p0, float32x4_t __p1) {
40044004}
40054005#else
40064006__ai uint32x4_t vcgeq_f32(float32x4_t __p0, float32x4_t __p1) {
4007 uint32x4_t __ret;
40074008 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
40084009 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4009 uint32x4_t __ret;
40104010 __ret = (uint32x4_t)(__rev0 >= __rev1);
40114011 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
40124012 return __ret;
......@@ -4021,9 +4021,9 @@ __ai uint32x4_t vcgeq_s32(int32x4_t __p0, int32x4_t __p1) {
40214021}
40224022#else
40234023__ai uint32x4_t vcgeq_s32(int32x4_t __p0, int32x4_t __p1) {
4024 uint32x4_t __ret;
40244025 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
40254026 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4026 uint32x4_t __ret;
40274027 __ret = (uint32x4_t)(__rev0 >= __rev1);
40284028 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
40294029 return __ret;
......@@ -4038,9 +4038,9 @@ __ai uint16x8_t vcgeq_s16(int16x8_t __p0, int16x8_t __p1) {
40384038}
40394039#else
40404040__ai uint16x8_t vcgeq_s16(int16x8_t __p0, int16x8_t __p1) {
4041 uint16x8_t __ret;
40414042 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
40424043 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4043 uint16x8_t __ret;
40444044 __ret = (uint16x8_t)(__rev0 >= __rev1);
40454045 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
40464046 return __ret;
......@@ -4055,9 +4055,9 @@ __ai uint8x8_t vcge_u8(uint8x8_t __p0, uint8x8_t __p1) {
40554055}
40564056#else
40574057__ai uint8x8_t vcge_u8(uint8x8_t __p0, uint8x8_t __p1) {
4058 uint8x8_t __ret;
40584059 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
40594060 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4060 uint8x8_t __ret;
40614061 __ret = (uint8x8_t)(__rev0 >= __rev1);
40624062 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
40634063 return __ret;
......@@ -4072,9 +4072,9 @@ __ai uint32x2_t vcge_u32(uint32x2_t __p0, uint32x2_t __p1) {
40724072}
40734073#else
40744074__ai uint32x2_t vcge_u32(uint32x2_t __p0, uint32x2_t __p1) {
4075 uint32x2_t __ret;
40754076 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
40764077 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4077 uint32x2_t __ret;
40784078 __ret = (uint32x2_t)(__rev0 >= __rev1);
40794079 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
40804080 return __ret;
......@@ -4089,9 +4089,9 @@ __ai uint16x4_t vcge_u16(uint16x4_t __p0, uint16x4_t __p1) {
40894089}
40904090#else
40914091__ai uint16x4_t vcge_u16(uint16x4_t __p0, uint16x4_t __p1) {
4092 uint16x4_t __ret;
40924093 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
40934094 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4094 uint16x4_t __ret;
40954095 __ret = (uint16x4_t)(__rev0 >= __rev1);
40964096 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
40974097 return __ret;
......@@ -4106,9 +4106,9 @@ __ai uint8x8_t vcge_s8(int8x8_t __p0, int8x8_t __p1) {
41064106}
41074107#else
41084108__ai uint8x8_t vcge_s8(int8x8_t __p0, int8x8_t __p1) {
4109 uint8x8_t __ret;
41094110 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
41104111 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4111 uint8x8_t __ret;
41124112 __ret = (uint8x8_t)(__rev0 >= __rev1);
41134113 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
41144114 return __ret;
......@@ -4123,9 +4123,9 @@ __ai uint32x2_t vcge_f32(float32x2_t __p0, float32x2_t __p1) {
41234123}
41244124#else
41254125__ai uint32x2_t vcge_f32(float32x2_t __p0, float32x2_t __p1) {
4126 uint32x2_t __ret;
41264127 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41274128 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4128 uint32x2_t __ret;
41294129 __ret = (uint32x2_t)(__rev0 >= __rev1);
41304130 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41314131 return __ret;
......@@ -4140,9 +4140,9 @@ __ai uint32x2_t vcge_s32(int32x2_t __p0, int32x2_t __p1) {
41404140}
41414141#else
41424142__ai uint32x2_t vcge_s32(int32x2_t __p0, int32x2_t __p1) {
4143 uint32x2_t __ret;
41434144 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
41444145 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4145 uint32x2_t __ret;
41464146 __ret = (uint32x2_t)(__rev0 >= __rev1);
41474147 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
41484148 return __ret;
......@@ -4157,9 +4157,9 @@ __ai uint16x4_t vcge_s16(int16x4_t __p0, int16x4_t __p1) {
41574157}
41584158#else
41594159__ai uint16x4_t vcge_s16(int16x4_t __p0, int16x4_t __p1) {
4160 uint16x4_t __ret;
41604161 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41614162 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4162 uint16x4_t __ret;
41634163 __ret = (uint16x4_t)(__rev0 >= __rev1);
41644164 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
41654165 return __ret;
......@@ -4174,9 +4174,9 @@ __ai uint8x16_t vcgtq_u8(uint8x16_t __p0, uint8x16_t __p1) {
41744174}
41754175#else
41764176__ai uint8x16_t vcgtq_u8(uint8x16_t __p0, uint8x16_t __p1) {
4177 uint8x16_t __ret;
41774178 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
41784179 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4179 uint8x16_t __ret;
41804180 __ret = (uint8x16_t)(__rev0 > __rev1);
41814181 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
41824182 return __ret;
......@@ -4191,9 +4191,9 @@ __ai uint32x4_t vcgtq_u32(uint32x4_t __p0, uint32x4_t __p1) {
41914191}
41924192#else
41934193__ai uint32x4_t vcgtq_u32(uint32x4_t __p0, uint32x4_t __p1) {
4194 uint32x4_t __ret;
41944195 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
41954196 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4196 uint32x4_t __ret;
41974197 __ret = (uint32x4_t)(__rev0 > __rev1);
41984198 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
41994199 return __ret;
......@@ -4208,9 +4208,9 @@ __ai uint16x8_t vcgtq_u16(uint16x8_t __p0, uint16x8_t __p1) {
42084208}
42094209#else
42104210__ai uint16x8_t vcgtq_u16(uint16x8_t __p0, uint16x8_t __p1) {
4211 uint16x8_t __ret;
42114212 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42124213 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4213 uint16x8_t __ret;
42144214 __ret = (uint16x8_t)(__rev0 > __rev1);
42154215 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
42164216 return __ret;
......@@ -4225,9 +4225,9 @@ __ai uint8x16_t vcgtq_s8(int8x16_t __p0, int8x16_t __p1) {
42254225}
42264226#else
42274227__ai uint8x16_t vcgtq_s8(int8x16_t __p0, int8x16_t __p1) {
4228 uint8x16_t __ret;
42284229 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
42294230 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4230 uint8x16_t __ret;
42314231 __ret = (uint8x16_t)(__rev0 > __rev1);
42324232 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
42334233 return __ret;
......@@ -4242,9 +4242,9 @@ __ai uint32x4_t vcgtq_f32(float32x4_t __p0, float32x4_t __p1) {
42424242}
42434243#else
42444244__ai uint32x4_t vcgtq_f32(float32x4_t __p0, float32x4_t __p1) {
4245 uint32x4_t __ret;
42454246 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
42464247 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4247 uint32x4_t __ret;
42484248 __ret = (uint32x4_t)(__rev0 > __rev1);
42494249 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
42504250 return __ret;
......@@ -4259,9 +4259,9 @@ __ai uint32x4_t vcgtq_s32(int32x4_t __p0, int32x4_t __p1) {
42594259}
42604260#else
42614261__ai uint32x4_t vcgtq_s32(int32x4_t __p0, int32x4_t __p1) {
4262 uint32x4_t __ret;
42624263 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
42634264 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4264 uint32x4_t __ret;
42654265 __ret = (uint32x4_t)(__rev0 > __rev1);
42664266 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
42674267 return __ret;
......@@ -4276,9 +4276,9 @@ __ai uint16x8_t vcgtq_s16(int16x8_t __p0, int16x8_t __p1) {
42764276}
42774277#else
42784278__ai uint16x8_t vcgtq_s16(int16x8_t __p0, int16x8_t __p1) {
4279 uint16x8_t __ret;
42794280 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42804281 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4281 uint16x8_t __ret;
42824282 __ret = (uint16x8_t)(__rev0 > __rev1);
42834283 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
42844284 return __ret;
......@@ -4293,9 +4293,9 @@ __ai uint8x8_t vcgt_u8(uint8x8_t __p0, uint8x8_t __p1) {
42934293}
42944294#else
42954295__ai uint8x8_t vcgt_u8(uint8x8_t __p0, uint8x8_t __p1) {
4296 uint8x8_t __ret;
42964297 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
42974298 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4298 uint8x8_t __ret;
42994299 __ret = (uint8x8_t)(__rev0 > __rev1);
43004300 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43014301 return __ret;
......@@ -4310,9 +4310,9 @@ __ai uint32x2_t vcgt_u32(uint32x2_t __p0, uint32x2_t __p1) {
43104310}
43114311#else
43124312__ai uint32x2_t vcgt_u32(uint32x2_t __p0, uint32x2_t __p1) {
4313 uint32x2_t __ret;
43134314 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43144315 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4315 uint32x2_t __ret;
43164316 __ret = (uint32x2_t)(__rev0 > __rev1);
43174317 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43184318 return __ret;
......@@ -4327,9 +4327,9 @@ __ai uint16x4_t vcgt_u16(uint16x4_t __p0, uint16x4_t __p1) {
43274327}
43284328#else
43294329__ai uint16x4_t vcgt_u16(uint16x4_t __p0, uint16x4_t __p1) {
4330 uint16x4_t __ret;
43304331 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43314332 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4332 uint16x4_t __ret;
43334333 __ret = (uint16x4_t)(__rev0 > __rev1);
43344334 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
43354335 return __ret;
......@@ -4344,9 +4344,9 @@ __ai uint8x8_t vcgt_s8(int8x8_t __p0, int8x8_t __p1) {
43444344}
43454345#else
43464346__ai uint8x8_t vcgt_s8(int8x8_t __p0, int8x8_t __p1) {
4347 uint8x8_t __ret;
43474348 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
43484349 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4349 uint8x8_t __ret;
43504350 __ret = (uint8x8_t)(__rev0 > __rev1);
43514351 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
43524352 return __ret;
......@@ -4361,9 +4361,9 @@ __ai uint32x2_t vcgt_f32(float32x2_t __p0, float32x2_t __p1) {
43614361}
43624362#else
43634363__ai uint32x2_t vcgt_f32(float32x2_t __p0, float32x2_t __p1) {
4364 uint32x2_t __ret;
43644365 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43654366 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4366 uint32x2_t __ret;
43674367 __ret = (uint32x2_t)(__rev0 > __rev1);
43684368 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43694369 return __ret;
......@@ -4378,9 +4378,9 @@ __ai uint32x2_t vcgt_s32(int32x2_t __p0, int32x2_t __p1) {
43784378}
43794379#else
43804380__ai uint32x2_t vcgt_s32(int32x2_t __p0, int32x2_t __p1) {
4381 uint32x2_t __ret;
43814382 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
43824383 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4383 uint32x2_t __ret;
43844384 __ret = (uint32x2_t)(__rev0 > __rev1);
43854385 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
43864386 return __ret;
......@@ -4395,9 +4395,9 @@ __ai uint16x4_t vcgt_s16(int16x4_t __p0, int16x4_t __p1) {
43954395}
43964396#else
43974397__ai uint16x4_t vcgt_s16(int16x4_t __p0, int16x4_t __p1) {
4398 uint16x4_t __ret;
43984399 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
43994400 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4400 uint16x4_t __ret;
44014401 __ret = (uint16x4_t)(__rev0 > __rev1);
44024402 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44034403 return __ret;
......@@ -4412,9 +4412,9 @@ __ai uint8x16_t vcleq_u8(uint8x16_t __p0, uint8x16_t __p1) {
44124412}
44134413#else
44144414__ai uint8x16_t vcleq_u8(uint8x16_t __p0, uint8x16_t __p1) {
4415 uint8x16_t __ret;
44154416 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
44164417 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4417 uint8x16_t __ret;
44184418 __ret = (uint8x16_t)(__rev0 <= __rev1);
44194419 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
44204420 return __ret;
......@@ -4429,9 +4429,9 @@ __ai uint32x4_t vcleq_u32(uint32x4_t __p0, uint32x4_t __p1) {
44294429}
44304430#else
44314431__ai uint32x4_t vcleq_u32(uint32x4_t __p0, uint32x4_t __p1) {
4432 uint32x4_t __ret;
44324433 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44334434 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4434 uint32x4_t __ret;
44354435 __ret = (uint32x4_t)(__rev0 <= __rev1);
44364436 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44374437 return __ret;
......@@ -4446,9 +4446,9 @@ __ai uint16x8_t vcleq_u16(uint16x8_t __p0, uint16x8_t __p1) {
44464446}
44474447#else
44484448__ai uint16x8_t vcleq_u16(uint16x8_t __p0, uint16x8_t __p1) {
4449 uint16x8_t __ret;
44494450 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
44504451 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4451 uint16x8_t __ret;
44524452 __ret = (uint16x8_t)(__rev0 <= __rev1);
44534453 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
44544454 return __ret;
......@@ -4463,9 +4463,9 @@ __ai uint8x16_t vcleq_s8(int8x16_t __p0, int8x16_t __p1) {
44634463}
44644464#else
44654465__ai uint8x16_t vcleq_s8(int8x16_t __p0, int8x16_t __p1) {
4466 uint8x16_t __ret;
44664467 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
44674468 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4468 uint8x16_t __ret;
44694469 __ret = (uint8x16_t)(__rev0 <= __rev1);
44704470 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
44714471 return __ret;
......@@ -4480,9 +4480,9 @@ __ai uint32x4_t vcleq_f32(float32x4_t __p0, float32x4_t __p1) {
44804480}
44814481#else
44824482__ai uint32x4_t vcleq_f32(float32x4_t __p0, float32x4_t __p1) {
4483 uint32x4_t __ret;
44834484 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
44844485 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4485 uint32x4_t __ret;
44864486 __ret = (uint32x4_t)(__rev0 <= __rev1);
44874487 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
44884488 return __ret;
......@@ -4497,9 +4497,9 @@ __ai uint32x4_t vcleq_s32(int32x4_t __p0, int32x4_t __p1) {
44974497}
44984498#else
44994499__ai uint32x4_t vcleq_s32(int32x4_t __p0, int32x4_t __p1) {
4500 uint32x4_t __ret;
45004501 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45014502 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4502 uint32x4_t __ret;
45034503 __ret = (uint32x4_t)(__rev0 <= __rev1);
45044504 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45054505 return __ret;
......@@ -4514,9 +4514,9 @@ __ai uint16x8_t vcleq_s16(int16x8_t __p0, int16x8_t __p1) {
45144514}
45154515#else
45164516__ai uint16x8_t vcleq_s16(int16x8_t __p0, int16x8_t __p1) {
4517 uint16x8_t __ret;
45174518 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45184519 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4519 uint16x8_t __ret;
45204520 __ret = (uint16x8_t)(__rev0 <= __rev1);
45214521 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45224522 return __ret;
......@@ -4531,9 +4531,9 @@ __ai uint8x8_t vcle_u8(uint8x8_t __p0, uint8x8_t __p1) {
45314531}
45324532#else
45334533__ai uint8x8_t vcle_u8(uint8x8_t __p0, uint8x8_t __p1) {
4534 uint8x8_t __ret;
45344535 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45354536 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4536 uint8x8_t __ret;
45374537 __ret = (uint8x8_t)(__rev0 <= __rev1);
45384538 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45394539 return __ret;
......@@ -4548,9 +4548,9 @@ __ai uint32x2_t vcle_u32(uint32x2_t __p0, uint32x2_t __p1) {
45484548}
45494549#else
45504550__ai uint32x2_t vcle_u32(uint32x2_t __p0, uint32x2_t __p1) {
4551 uint32x2_t __ret;
45514552 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
45524553 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4553 uint32x2_t __ret;
45544554 __ret = (uint32x2_t)(__rev0 <= __rev1);
45554555 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
45564556 return __ret;
......@@ -4565,9 +4565,9 @@ __ai uint16x4_t vcle_u16(uint16x4_t __p0, uint16x4_t __p1) {
45654565}
45664566#else
45674567__ai uint16x4_t vcle_u16(uint16x4_t __p0, uint16x4_t __p1) {
4568 uint16x4_t __ret;
45684569 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
45694570 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4570 uint16x4_t __ret;
45714571 __ret = (uint16x4_t)(__rev0 <= __rev1);
45724572 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
45734573 return __ret;
......@@ -4582,9 +4582,9 @@ __ai uint8x8_t vcle_s8(int8x8_t __p0, int8x8_t __p1) {
45824582}
45834583#else
45844584__ai uint8x8_t vcle_s8(int8x8_t __p0, int8x8_t __p1) {
4585 uint8x8_t __ret;
45854586 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
45864587 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4587 uint8x8_t __ret;
45884588 __ret = (uint8x8_t)(__rev0 <= __rev1);
45894589 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
45904590 return __ret;
......@@ -4599,9 +4599,9 @@ __ai uint32x2_t vcle_f32(float32x2_t __p0, float32x2_t __p1) {
45994599}
46004600#else
46014601__ai uint32x2_t vcle_f32(float32x2_t __p0, float32x2_t __p1) {
4602 uint32x2_t __ret;
46024603 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
46034604 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4604 uint32x2_t __ret;
46054605 __ret = (uint32x2_t)(__rev0 <= __rev1);
46064606 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46074607 return __ret;
......@@ -4616,9 +4616,9 @@ __ai uint32x2_t vcle_s32(int32x2_t __p0, int32x2_t __p1) {
46164616}
46174617#else
46184618__ai uint32x2_t vcle_s32(int32x2_t __p0, int32x2_t __p1) {
4619 uint32x2_t __ret;
46194620 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
46204621 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4621 uint32x2_t __ret;
46224622 __ret = (uint32x2_t)(__rev0 <= __rev1);
46234623 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
46244624 return __ret;
......@@ -4633,9 +4633,9 @@ __ai uint16x4_t vcle_s16(int16x4_t __p0, int16x4_t __p1) {
46334633}
46344634#else
46354635__ai uint16x4_t vcle_s16(int16x4_t __p0, int16x4_t __p1) {
4636 uint16x4_t __ret;
46364637 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46374638 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4638 uint16x4_t __ret;
46394639 __ret = (uint16x4_t)(__rev0 <= __rev1);
46404640 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46414641 return __ret;
......@@ -4650,8 +4650,8 @@ __ai int8x16_t vclsq_u8(uint8x16_t __p0) {
46504650}
46514651#else
46524652__ai int8x16_t vclsq_u8(uint8x16_t __p0) {
4653 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46544653 int8x16_t __ret;
4654 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46554655 __ret = (int8x16_t) __builtin_neon_vclsq_v((int8x16_t)__rev0, 32);
46564656 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46574657 return __ret;
......@@ -4666,8 +4666,8 @@ __ai int32x4_t vclsq_u32(uint32x4_t __p0) {
46664666}
46674667#else
46684668__ai int32x4_t vclsq_u32(uint32x4_t __p0) {
4669 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46704669 int32x4_t __ret;
4670 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
46714671 __ret = (int32x4_t) __builtin_neon_vclsq_v((int8x16_t)__rev0, 34);
46724672 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
46734673 return __ret;
......@@ -4682,8 +4682,8 @@ __ai int16x8_t vclsq_u16(uint16x8_t __p0) {
46824682}
46834683#else
46844684__ai int16x8_t vclsq_u16(uint16x8_t __p0) {
4685 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46864685 int16x8_t __ret;
4686 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
46874687 __ret = (int16x8_t) __builtin_neon_vclsq_v((int8x16_t)__rev0, 33);
46884688 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
46894689 return __ret;
......@@ -4698,8 +4698,8 @@ __ai int8x16_t vclsq_s8(int8x16_t __p0) {
46984698}
46994699#else
47004700__ai int8x16_t vclsq_s8(int8x16_t __p0) {
4701 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47024701 int8x16_t __ret;
4702 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47034703 __ret = (int8x16_t) __builtin_neon_vclsq_v((int8x16_t)__rev0, 32);
47044704 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47054705 return __ret;
......@@ -4714,8 +4714,8 @@ __ai int32x4_t vclsq_s32(int32x4_t __p0) {
47144714}
47154715#else
47164716__ai int32x4_t vclsq_s32(int32x4_t __p0) {
4717 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47184717 int32x4_t __ret;
4718 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47194719 __ret = (int32x4_t) __builtin_neon_vclsq_v((int8x16_t)__rev0, 34);
47204720 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47214721 return __ret;
......@@ -4730,8 +4730,8 @@ __ai int16x8_t vclsq_s16(int16x8_t __p0) {
47304730}
47314731#else
47324732__ai int16x8_t vclsq_s16(int16x8_t __p0) {
4733 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47344733 int16x8_t __ret;
4734 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47354735 __ret = (int16x8_t) __builtin_neon_vclsq_v((int8x16_t)__rev0, 33);
47364736 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
47374737 return __ret;
......@@ -4746,8 +4746,8 @@ __ai int8x8_t vcls_u8(uint8x8_t __p0) {
47464746}
47474747#else
47484748__ai int8x8_t vcls_u8(uint8x8_t __p0) {
4749 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47504749 int8x8_t __ret;
4750 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47514751 __ret = (int8x8_t) __builtin_neon_vcls_v((int8x8_t)__rev0, 0);
47524752 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
47534753 return __ret;
......@@ -4762,8 +4762,8 @@ __ai int32x2_t vcls_u32(uint32x2_t __p0) {
47624762}
47634763#else
47644764__ai int32x2_t vcls_u32(uint32x2_t __p0) {
4765 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47664765 int32x2_t __ret;
4766 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
47674767 __ret = (int32x2_t) __builtin_neon_vcls_v((int8x8_t)__rev0, 2);
47684768 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
47694769 return __ret;
......@@ -4778,8 +4778,8 @@ __ai int16x4_t vcls_u16(uint16x4_t __p0) {
47784778}
47794779#else
47804780__ai int16x4_t vcls_u16(uint16x4_t __p0) {
4781 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47824781 int16x4_t __ret;
4782 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
47834783 __ret = (int16x4_t) __builtin_neon_vcls_v((int8x8_t)__rev0, 1);
47844784 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
47854785 return __ret;
......@@ -4794,8 +4794,8 @@ __ai int8x8_t vcls_s8(int8x8_t __p0) {
47944794}
47954795#else
47964796__ai int8x8_t vcls_s8(int8x8_t __p0) {
4797 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47984797 int8x8_t __ret;
4798 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
47994799 __ret = (int8x8_t) __builtin_neon_vcls_v((int8x8_t)__rev0, 0);
48004800 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
48014801 return __ret;
......@@ -4810,8 +4810,8 @@ __ai int32x2_t vcls_s32(int32x2_t __p0) {
48104810}
48114811#else
48124812__ai int32x2_t vcls_s32(int32x2_t __p0) {
4813 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48144813 int32x2_t __ret;
4814 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
48154815 __ret = (int32x2_t) __builtin_neon_vcls_v((int8x8_t)__rev0, 2);
48164816 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
48174817 return __ret;
......@@ -4826,8 +4826,8 @@ __ai int16x4_t vcls_s16(int16x4_t __p0) {
48264826}
48274827#else
48284828__ai int16x4_t vcls_s16(int16x4_t __p0) {
4829 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48304829 int16x4_t __ret;
4830 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48314831 __ret = (int16x4_t) __builtin_neon_vcls_v((int8x8_t)__rev0, 1);
48324832 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
48334833 return __ret;
......@@ -4842,9 +4842,9 @@ __ai uint8x16_t vcltq_u8(uint8x16_t __p0, uint8x16_t __p1) {
48424842}
48434843#else
48444844__ai uint8x16_t vcltq_u8(uint8x16_t __p0, uint8x16_t __p1) {
4845 uint8x16_t __ret;
48454846 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48464847 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4847 uint8x16_t __ret;
48484848 __ret = (uint8x16_t)(__rev0 < __rev1);
48494849 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48504850 return __ret;
......@@ -4859,9 +4859,9 @@ __ai uint32x4_t vcltq_u32(uint32x4_t __p0, uint32x4_t __p1) {
48594859}
48604860#else
48614861__ai uint32x4_t vcltq_u32(uint32x4_t __p0, uint32x4_t __p1) {
4862 uint32x4_t __ret;
48624863 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
48634864 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4864 uint32x4_t __ret;
48654865 __ret = (uint32x4_t)(__rev0 < __rev1);
48664866 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
48674867 return __ret;
......@@ -4876,9 +4876,9 @@ __ai uint16x8_t vcltq_u16(uint16x8_t __p0, uint16x8_t __p1) {
48764876}
48774877#else
48784878__ai uint16x8_t vcltq_u16(uint16x8_t __p0, uint16x8_t __p1) {
4879 uint16x8_t __ret;
48794880 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
48804881 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4881 uint16x8_t __ret;
48824882 __ret = (uint16x8_t)(__rev0 < __rev1);
48834883 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
48844884 return __ret;
......@@ -4893,9 +4893,9 @@ __ai uint8x16_t vcltq_s8(int8x16_t __p0, int8x16_t __p1) {
48934893}
48944894#else
48954895__ai uint8x16_t vcltq_s8(int8x16_t __p0, int8x16_t __p1) {
4896 uint8x16_t __ret;
48964897 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
48974898 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4898 uint8x16_t __ret;
48994899 __ret = (uint8x16_t)(__rev0 < __rev1);
49004900 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
49014901 return __ret;
......@@ -4910,9 +4910,9 @@ __ai uint32x4_t vcltq_f32(float32x4_t __p0, float32x4_t __p1) {
49104910}
49114911#else
49124912__ai uint32x4_t vcltq_f32(float32x4_t __p0, float32x4_t __p1) {
4913 uint32x4_t __ret;
49134914 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49144915 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4915 uint32x4_t __ret;
49164916 __ret = (uint32x4_t)(__rev0 < __rev1);
49174917 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49184918 return __ret;
......@@ -4927,9 +4927,9 @@ __ai uint32x4_t vcltq_s32(int32x4_t __p0, int32x4_t __p1) {
49274927}
49284928#else
49294929__ai uint32x4_t vcltq_s32(int32x4_t __p0, int32x4_t __p1) {
4930 uint32x4_t __ret;
49304931 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49314932 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4932 uint32x4_t __ret;
49334933 __ret = (uint32x4_t)(__rev0 < __rev1);
49344934 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
49354935 return __ret;
......@@ -4944,9 +4944,9 @@ __ai uint16x8_t vcltq_s16(int16x8_t __p0, int16x8_t __p1) {
49444944}
49454945#else
49464946__ai uint16x8_t vcltq_s16(int16x8_t __p0, int16x8_t __p1) {
4947 uint16x8_t __ret;
49474948 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49484949 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4949 uint16x8_t __ret;
49504950 __ret = (uint16x8_t)(__rev0 < __rev1);
49514951 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49524952 return __ret;
......@@ -4961,9 +4961,9 @@ __ai uint8x8_t vclt_u8(uint8x8_t __p0, uint8x8_t __p1) {
49614961}
49624962#else
49634963__ai uint8x8_t vclt_u8(uint8x8_t __p0, uint8x8_t __p1) {
4964 uint8x8_t __ret;
49644965 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
49654966 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4966 uint8x8_t __ret;
49674967 __ret = (uint8x8_t)(__rev0 < __rev1);
49684968 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
49694969 return __ret;
......@@ -4978,9 +4978,9 @@ __ai uint32x2_t vclt_u32(uint32x2_t __p0, uint32x2_t __p1) {
49784978}
49794979#else
49804980__ai uint32x2_t vclt_u32(uint32x2_t __p0, uint32x2_t __p1) {
4981 uint32x2_t __ret;
49814982 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
49824983 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4983 uint32x2_t __ret;
49844984 __ret = (uint32x2_t)(__rev0 < __rev1);
49854985 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
49864986 return __ret;
......@@ -4995,9 +4995,9 @@ __ai uint16x4_t vclt_u16(uint16x4_t __p0, uint16x4_t __p1) {
49954995}
49964996#else
49974997__ai uint16x4_t vclt_u16(uint16x4_t __p0, uint16x4_t __p1) {
4998 uint16x4_t __ret;
49984999 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
49995000 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5000 uint16x4_t __ret;
50015001 __ret = (uint16x4_t)(__rev0 < __rev1);
50025002 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
50035003 return __ret;
......@@ -5012,9 +5012,9 @@ __ai uint8x8_t vclt_s8(int8x8_t __p0, int8x8_t __p1) {
50125012}
50135013#else
50145014__ai uint8x8_t vclt_s8(int8x8_t __p0, int8x8_t __p1) {
5015 uint8x8_t __ret;
50155016 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
50165017 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5017 uint8x8_t __ret;
50185018 __ret = (uint8x8_t)(__rev0 < __rev1);
50195019 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
50205020 return __ret;
......@@ -5029,9 +5029,9 @@ __ai uint32x2_t vclt_f32(float32x2_t __p0, float32x2_t __p1) {
50295029}
50305030#else
50315031__ai uint32x2_t vclt_f32(float32x2_t __p0, float32x2_t __p1) {
5032 uint32x2_t __ret;
50325033 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
50335034 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5034 uint32x2_t __ret;
50355035 __ret = (uint32x2_t)(__rev0 < __rev1);
50365036 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50375037 return __ret;
......@@ -5046,9 +5046,9 @@ __ai uint32x2_t vclt_s32(int32x2_t __p0, int32x2_t __p1) {
50465046}
50475047#else
50485048__ai uint32x2_t vclt_s32(int32x2_t __p0, int32x2_t __p1) {
5049 uint32x2_t __ret;
50495050 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
50505051 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5051 uint32x2_t __ret;
50525052 __ret = (uint32x2_t)(__rev0 < __rev1);
50535053 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
50545054 return __ret;
......@@ -5063,9 +5063,9 @@ __ai uint16x4_t vclt_s16(int16x4_t __p0, int16x4_t __p1) {
50635063}
50645064#else
50655065__ai uint16x4_t vclt_s16(int16x4_t __p0, int16x4_t __p1) {
5066 uint16x4_t __ret;
50665067 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
50675068 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5068 uint16x4_t __ret;
50695069 __ret = (uint16x4_t)(__rev0 < __rev1);
50705070 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
50715071 return __ret;
......@@ -5080,8 +5080,8 @@ __ai uint8x16_t vclzq_u8(uint8x16_t __p0) {
50805080}
50815081#else
50825082__ai uint8x16_t vclzq_u8(uint8x16_t __p0) {
5083 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
50845083 uint8x16_t __ret;
5084 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
50855085 __ret = (uint8x16_t) __builtin_neon_vclzq_v((int8x16_t)__rev0, 48);
50865086 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
50875087 return __ret;
......@@ -5096,8 +5096,8 @@ __ai uint32x4_t vclzq_u32(uint32x4_t __p0) {
50965096}
50975097#else
50985098__ai uint32x4_t vclzq_u32(uint32x4_t __p0) {
5099 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51005099 uint32x4_t __ret;
5100 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51015101 __ret = (uint32x4_t) __builtin_neon_vclzq_v((int8x16_t)__rev0, 50);
51025102 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51035103 return __ret;
......@@ -5112,8 +5112,8 @@ __ai uint16x8_t vclzq_u16(uint16x8_t __p0) {
51125112}
51135113#else
51145114__ai uint16x8_t vclzq_u16(uint16x8_t __p0) {
5115 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51165115 uint16x8_t __ret;
5116 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51175117 __ret = (uint16x8_t) __builtin_neon_vclzq_v((int8x16_t)__rev0, 49);
51185118 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
51195119 return __ret;
......@@ -5128,8 +5128,8 @@ __ai int8x16_t vclzq_s8(int8x16_t __p0) {
51285128}
51295129#else
51305130__ai int8x16_t vclzq_s8(int8x16_t __p0) {
5131 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
51325131 int8x16_t __ret;
5132 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
51335133 __ret = (int8x16_t) __builtin_neon_vclzq_v((int8x16_t)__rev0, 32);
51345134 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
51355135 return __ret;
......@@ -5144,8 +5144,8 @@ __ai int32x4_t vclzq_s32(int32x4_t __p0) {
51445144}
51455145#else
51465146__ai int32x4_t vclzq_s32(int32x4_t __p0) {
5147 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51485147 int32x4_t __ret;
5148 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
51495149 __ret = (int32x4_t) __builtin_neon_vclzq_v((int8x16_t)__rev0, 34);
51505150 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
51515151 return __ret;
......@@ -5160,8 +5160,8 @@ __ai int16x8_t vclzq_s16(int16x8_t __p0) {
51605160}
51615161#else
51625162__ai int16x8_t vclzq_s16(int16x8_t __p0) {
5163 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51645163 int16x8_t __ret;
5164 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51655165 __ret = (int16x8_t) __builtin_neon_vclzq_v((int8x16_t)__rev0, 33);
51665166 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
51675167 return __ret;
......@@ -5176,8 +5176,8 @@ __ai uint8x8_t vclz_u8(uint8x8_t __p0) {
51765176}
51775177#else
51785178__ai uint8x8_t vclz_u8(uint8x8_t __p0) {
5179 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51805179 uint8x8_t __ret;
5180 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
51815181 __ret = (uint8x8_t) __builtin_neon_vclz_v((int8x8_t)__rev0, 16);
51825182 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
51835183 return __ret;
......@@ -5192,8 +5192,8 @@ __ai uint32x2_t vclz_u32(uint32x2_t __p0) {
51925192}
51935193#else
51945194__ai uint32x2_t vclz_u32(uint32x2_t __p0) {
5195 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51965195 uint32x2_t __ret;
5196 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
51975197 __ret = (uint32x2_t) __builtin_neon_vclz_v((int8x8_t)__rev0, 18);
51985198 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
51995199 return __ret;
......@@ -5208,8 +5208,8 @@ __ai uint16x4_t vclz_u16(uint16x4_t __p0) {
52085208}
52095209#else
52105210__ai uint16x4_t vclz_u16(uint16x4_t __p0) {
5211 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
52125211 uint16x4_t __ret;
5212 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
52135213 __ret = (uint16x4_t) __builtin_neon_vclz_v((int8x8_t)__rev0, 17);
52145214 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
52155215 return __ret;
......@@ -5224,8 +5224,8 @@ __ai int8x8_t vclz_s8(int8x8_t __p0) {
52245224}
52255225#else
52265226__ai int8x8_t vclz_s8(int8x8_t __p0) {
5227 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
52285227 int8x8_t __ret;
5228 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
52295229 __ret = (int8x8_t) __builtin_neon_vclz_v((int8x8_t)__rev0, 0);
52305230 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
52315231 return __ret;
......@@ -5240,8 +5240,8 @@ __ai int32x2_t vclz_s32(int32x2_t __p0) {
52405240}
52415241#else
52425242__ai int32x2_t vclz_s32(int32x2_t __p0) {
5243 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
52445243 int32x2_t __ret;
5244 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
52455245 __ret = (int32x2_t) __builtin_neon_vclz_v((int8x8_t)__rev0, 2);
52465246 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
52475247 return __ret;
......@@ -5256,8 +5256,8 @@ __ai int16x4_t vclz_s16(int16x4_t __p0) {
52565256}
52575257#else
52585258__ai int16x4_t vclz_s16(int16x4_t __p0) {
5259 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
52605259 int16x4_t __ret;
5260 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
52615261 __ret = (int16x4_t) __builtin_neon_vclz_v((int8x8_t)__rev0, 1);
52625262 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
52635263 return __ret;
......@@ -5272,8 +5272,8 @@ __ai poly8x8_t vcnt_p8(poly8x8_t __p0) {
52725272}
52735273#else
52745274__ai poly8x8_t vcnt_p8(poly8x8_t __p0) {
5275 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
52765275 poly8x8_t __ret;
5276 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
52775277 __ret = (poly8x8_t) __builtin_neon_vcnt_v((int8x8_t)__rev0, 4);
52785278 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
52795279 return __ret;
......@@ -5288,8 +5288,8 @@ __ai poly8x16_t vcntq_p8(poly8x16_t __p0) {
52885288}
52895289#else
52905290__ai poly8x16_t vcntq_p8(poly8x16_t __p0) {
5291 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
52925291 poly8x16_t __ret;
5292 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
52935293 __ret = (poly8x16_t) __builtin_neon_vcntq_v((int8x16_t)__rev0, 36);
52945294 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
52955295 return __ret;
......@@ -5304,8 +5304,8 @@ __ai uint8x16_t vcntq_u8(uint8x16_t __p0) {
53045304}
53055305#else
53065306__ai uint8x16_t vcntq_u8(uint8x16_t __p0) {
5307 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53085307 uint8x16_t __ret;
5308 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53095309 __ret = (uint8x16_t) __builtin_neon_vcntq_v((int8x16_t)__rev0, 48);
53105310 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53115311 return __ret;
......@@ -5320,8 +5320,8 @@ __ai int8x16_t vcntq_s8(int8x16_t __p0) {
53205320}
53215321#else
53225322__ai int8x16_t vcntq_s8(int8x16_t __p0) {
5323 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53245323 int8x16_t __ret;
5324 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53255325 __ret = (int8x16_t) __builtin_neon_vcntq_v((int8x16_t)__rev0, 32);
53265326 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53275327 return __ret;
......@@ -5336,8 +5336,8 @@ __ai uint8x8_t vcnt_u8(uint8x8_t __p0) {
53365336}
53375337#else
53385338__ai uint8x8_t vcnt_u8(uint8x8_t __p0) {
5339 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
53405339 uint8x8_t __ret;
5340 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
53415341 __ret = (uint8x8_t) __builtin_neon_vcnt_v((int8x8_t)__rev0, 16);
53425342 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
53435343 return __ret;
......@@ -5352,8 +5352,8 @@ __ai int8x8_t vcnt_s8(int8x8_t __p0) {
53525352}
53535353#else
53545354__ai int8x8_t vcnt_s8(int8x8_t __p0) {
5355 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
53565355 int8x8_t __ret;
5356 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
53575357 __ret = (int8x8_t) __builtin_neon_vcnt_v((int8x8_t)__rev0, 0);
53585358 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
53595359 return __ret;
......@@ -5368,9 +5368,9 @@ __ai poly8x16_t vcombine_p8(poly8x8_t __p0, poly8x8_t __p1) {
53685368}
53695369#else
53705370__ai poly8x16_t vcombine_p8(poly8x8_t __p0, poly8x8_t __p1) {
5371 poly8x16_t __ret;
53715372 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
53725373 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5373 poly8x16_t __ret;
53745374 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
53755375 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53765376 return __ret;
......@@ -5385,9 +5385,9 @@ __ai poly16x8_t vcombine_p16(poly16x4_t __p0, poly16x4_t __p1) {
53855385}
53865386#else
53875387__ai poly16x8_t vcombine_p16(poly16x4_t __p0, poly16x4_t __p1) {
5388 poly16x8_t __ret;
53885389 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
53895390 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5390 poly16x8_t __ret;
53915391 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
53925392 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
53935393 return __ret;
......@@ -5402,9 +5402,9 @@ __ai uint8x16_t vcombine_u8(uint8x8_t __p0, uint8x8_t __p1) {
54025402}
54035403#else
54045404__ai uint8x16_t vcombine_u8(uint8x8_t __p0, uint8x8_t __p1) {
5405 uint8x16_t __ret;
54055406 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
54065407 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5407 uint8x16_t __ret;
54085408 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
54095409 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54105410 return __ret;
......@@ -5424,9 +5424,9 @@ __ai uint32x4_t vcombine_u32(uint32x2_t __p0, uint32x2_t __p1) {
54245424}
54255425#else
54265426__ai uint32x4_t vcombine_u32(uint32x2_t __p0, uint32x2_t __p1) {
5427 uint32x4_t __ret;
54275428 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
54285429 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5429 uint32x4_t __ret;
54305430 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3);
54315431 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
54325432 return __ret;
......@@ -5461,9 +5461,9 @@ __ai uint16x8_t vcombine_u16(uint16x4_t __p0, uint16x4_t __p1) {
54615461}
54625462#else
54635463__ai uint16x8_t vcombine_u16(uint16x4_t __p0, uint16x4_t __p1) {
5464 uint16x8_t __ret;
54645465 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
54655466 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5466 uint16x8_t __ret;
54675467 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
54685468 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
54695469 return __ret;
......@@ -5483,9 +5483,9 @@ __ai int8x16_t vcombine_s8(int8x8_t __p0, int8x8_t __p1) {
54835483}
54845484#else
54855485__ai int8x16_t vcombine_s8(int8x8_t __p0, int8x8_t __p1) {
5486 int8x16_t __ret;
54865487 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
54875488 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5488 int8x16_t __ret;
54895489 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
54905490 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54915491 return __ret;
......@@ -5505,9 +5505,9 @@ __ai float32x4_t vcombine_f32(float32x2_t __p0, float32x2_t __p1) {
55055505}
55065506#else
55075507__ai float32x4_t vcombine_f32(float32x2_t __p0, float32x2_t __p1) {
5508 float32x4_t __ret;
55085509 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55095510 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5510 float32x4_t __ret;
55115511 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3);
55125512 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
55135513 return __ret;
......@@ -5527,9 +5527,9 @@ __ai float16x8_t vcombine_f16(float16x4_t __p0, float16x4_t __p1) {
55275527}
55285528#else
55295529__ai float16x8_t vcombine_f16(float16x4_t __p0, float16x4_t __p1) {
5530 float16x8_t __ret;
55305531 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55315532 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5532 float16x8_t __ret;
55335533 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
55345534 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
55355535 return __ret;
......@@ -5549,9 +5549,9 @@ __ai int32x4_t vcombine_s32(int32x2_t __p0, int32x2_t __p1) {
55495549}
55505550#else
55515551__ai int32x4_t vcombine_s32(int32x2_t __p0, int32x2_t __p1) {
5552 int32x4_t __ret;
55525553 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
55535554 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5554 int32x4_t __ret;
55555555 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3);
55565556 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
55575557 return __ret;
......@@ -5586,9 +5586,9 @@ __ai int16x8_t vcombine_s16(int16x4_t __p0, int16x4_t __p1) {
55865586}
55875587#else
55885588__ai int16x8_t vcombine_s16(int16x4_t __p0, int16x4_t __p1) {
5589 int16x8_t __ret;
55895590 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
55905591 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5591 int16x8_t __ret;
55925592 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
55935593 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
55945594 return __ret;
......@@ -5680,8 +5680,8 @@ __ai float32x4_t vcvtq_f32_u32(uint32x4_t __p0) {
56805680}
56815681#else
56825682__ai float32x4_t vcvtq_f32_u32(uint32x4_t __p0) {
5683 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
56845683 float32x4_t __ret;
5684 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
56855685 __ret = (float32x4_t) __builtin_neon_vcvtq_f32_v((int8x16_t)__rev0, 50);
56865686 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
56875687 return __ret;
......@@ -5696,8 +5696,8 @@ __ai float32x4_t vcvtq_f32_s32(int32x4_t __p0) {
56965696}
56975697#else
56985698__ai float32x4_t vcvtq_f32_s32(int32x4_t __p0) {
5699 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57005699 float32x4_t __ret;
5700 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
57015701 __ret = (float32x4_t) __builtin_neon_vcvtq_f32_v((int8x16_t)__rev0, 34);
57025702 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
57035703 return __ret;
......@@ -5712,8 +5712,8 @@ __ai float32x2_t vcvt_f32_u32(uint32x2_t __p0) {
57125712}
57135713#else
57145714__ai float32x2_t vcvt_f32_u32(uint32x2_t __p0) {
5715 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57165715 float32x2_t __ret;
5716 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57175717 __ret = (float32x2_t) __builtin_neon_vcvt_f32_v((int8x8_t)__rev0, 18);
57185718 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
57195719 return __ret;
......@@ -5728,8 +5728,8 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
57285728}
57295729#else
57305730__ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
5731 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57325731 float32x2_t __ret;
5732 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
57335733 __ret = (float32x2_t) __builtin_neon_vcvt_f32_v((int8x8_t)__rev0, 2);
57345734 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
57355735 return __ret;
......@@ -5738,16 +5738,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
57385738
57395739#ifdef __LITTLE_ENDIAN__
57405740#define vcvtq_n_f32_u32(__p0, __p1) __extension__ ({ \
5741 uint32x4_t __s0 = __p0; \
57425741 float32x4_t __ret; \
5742 uint32x4_t __s0 = __p0; \
57435743 __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__s0, __p1, 50); \
57445744 __ret; \
57455745})
57465746#else
57475747#define vcvtq_n_f32_u32(__p0, __p1) __extension__ ({ \
5748 float32x4_t __ret; \
57485749 uint32x4_t __s0 = __p0; \
57495750 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5750 float32x4_t __ret; \
57515751 __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__rev0, __p1, 50); \
57525752 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
57535753 __ret; \
......@@ -5756,16 +5756,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
57565756
57575757#ifdef __LITTLE_ENDIAN__
57585758#define vcvtq_n_f32_s32(__p0, __p1) __extension__ ({ \
5759 int32x4_t __s0 = __p0; \
57605759 float32x4_t __ret; \
5760 int32x4_t __s0 = __p0; \
57615761 __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__s0, __p1, 34); \
57625762 __ret; \
57635763})
57645764#else
57655765#define vcvtq_n_f32_s32(__p0, __p1) __extension__ ({ \
5766 float32x4_t __ret; \
57665767 int32x4_t __s0 = __p0; \
57675768 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5768 float32x4_t __ret; \
57695769 __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__rev0, __p1, 34); \
57705770 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
57715771 __ret; \
......@@ -5774,16 +5774,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
57745774
57755775#ifdef __LITTLE_ENDIAN__
57765776#define vcvt_n_f32_u32(__p0, __p1) __extension__ ({ \
5777 uint32x2_t __s0 = __p0; \
57785777 float32x2_t __ret; \
5778 uint32x2_t __s0 = __p0; \
57795779 __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__s0, __p1, 18); \
57805780 __ret; \
57815781})
57825782#else
57835783#define vcvt_n_f32_u32(__p0, __p1) __extension__ ({ \
5784 float32x2_t __ret; \
57845785 uint32x2_t __s0 = __p0; \
57855786 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5786 float32x2_t __ret; \
57875787 __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__rev0, __p1, 18); \
57885788 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
57895789 __ret; \
......@@ -5792,16 +5792,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
57925792
57935793#ifdef __LITTLE_ENDIAN__
57945794#define vcvt_n_f32_s32(__p0, __p1) __extension__ ({ \
5795 int32x2_t __s0 = __p0; \
57965795 float32x2_t __ret; \
5796 int32x2_t __s0 = __p0; \
57975797 __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__s0, __p1, 2); \
57985798 __ret; \
57995799})
58005800#else
58015801#define vcvt_n_f32_s32(__p0, __p1) __extension__ ({ \
5802 float32x2_t __ret; \
58025803 int32x2_t __s0 = __p0; \
58035804 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5804 float32x2_t __ret; \
58055805 __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__rev0, __p1, 2); \
58065806 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
58075807 __ret; \
......@@ -5810,16 +5810,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
58105810
58115811#ifdef __LITTLE_ENDIAN__
58125812#define vcvtq_n_s32_f32(__p0, __p1) __extension__ ({ \
5813 float32x4_t __s0 = __p0; \
58145813 int32x4_t __ret; \
5814 float32x4_t __s0 = __p0; \
58155815 __ret = (int32x4_t) __builtin_neon_vcvtq_n_s32_v((int8x16_t)__s0, __p1, 34); \
58165816 __ret; \
58175817})
58185818#else
58195819#define vcvtq_n_s32_f32(__p0, __p1) __extension__ ({ \
5820 int32x4_t __ret; \
58205821 float32x4_t __s0 = __p0; \
58215822 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5822 int32x4_t __ret; \
58235823 __ret = (int32x4_t) __builtin_neon_vcvtq_n_s32_v((int8x16_t)__rev0, __p1, 34); \
58245824 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
58255825 __ret; \
......@@ -5828,16 +5828,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
58285828
58295829#ifdef __LITTLE_ENDIAN__
58305830#define vcvt_n_s32_f32(__p0, __p1) __extension__ ({ \
5831 float32x2_t __s0 = __p0; \
58325831 int32x2_t __ret; \
5832 float32x2_t __s0 = __p0; \
58335833 __ret = (int32x2_t) __builtin_neon_vcvt_n_s32_v((int8x8_t)__s0, __p1, 2); \
58345834 __ret; \
58355835})
58365836#else
58375837#define vcvt_n_s32_f32(__p0, __p1) __extension__ ({ \
5838 int32x2_t __ret; \
58385839 float32x2_t __s0 = __p0; \
58395840 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5840 int32x2_t __ret; \
58415841 __ret = (int32x2_t) __builtin_neon_vcvt_n_s32_v((int8x8_t)__rev0, __p1, 2); \
58425842 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
58435843 __ret; \
......@@ -5846,16 +5846,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
58465846
58475847#ifdef __LITTLE_ENDIAN__
58485848#define vcvtq_n_u32_f32(__p0, __p1) __extension__ ({ \
5849 float32x4_t __s0 = __p0; \
58505849 uint32x4_t __ret; \
5850 float32x4_t __s0 = __p0; \
58515851 __ret = (uint32x4_t) __builtin_neon_vcvtq_n_u32_v((int8x16_t)__s0, __p1, 50); \
58525852 __ret; \
58535853})
58545854#else
58555855#define vcvtq_n_u32_f32(__p0, __p1) __extension__ ({ \
5856 uint32x4_t __ret; \
58565857 float32x4_t __s0 = __p0; \
58575858 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5858 uint32x4_t __ret; \
58595859 __ret = (uint32x4_t) __builtin_neon_vcvtq_n_u32_v((int8x16_t)__rev0, __p1, 50); \
58605860 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
58615861 __ret; \
......@@ -5864,16 +5864,16 @@ __ai float32x2_t vcvt_f32_s32(int32x2_t __p0) {
58645864
58655865#ifdef __LITTLE_ENDIAN__
58665866#define vcvt_n_u32_f32(__p0, __p1) __extension__ ({ \
5867 float32x2_t __s0 = __p0; \
58685867 uint32x2_t __ret; \
5868 float32x2_t __s0 = __p0; \
58695869 __ret = (uint32x2_t) __builtin_neon_vcvt_n_u32_v((int8x8_t)__s0, __p1, 18); \
58705870 __ret; \
58715871})
58725872#else
58735873#define vcvt_n_u32_f32(__p0, __p1) __extension__ ({ \
5874 uint32x2_t __ret; \
58745875 float32x2_t __s0 = __p0; \
58755876 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5876 uint32x2_t __ret; \
58775877 __ret = (uint32x2_t) __builtin_neon_vcvt_n_u32_v((int8x8_t)__rev0, __p1, 18); \
58785878 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
58795879 __ret; \
......@@ -5888,8 +5888,8 @@ __ai int32x4_t vcvtq_s32_f32(float32x4_t __p0) {
58885888}
58895889#else
58905890__ai int32x4_t vcvtq_s32_f32(float32x4_t __p0) {
5891 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58925891 int32x4_t __ret;
5892 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
58935893 __ret = (int32x4_t) __builtin_neon_vcvtq_s32_v((int8x16_t)__rev0, 34);
58945894 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
58955895 return __ret;
......@@ -5904,8 +5904,8 @@ __ai int32x2_t vcvt_s32_f32(float32x2_t __p0) {
59045904}
59055905#else
59065906__ai int32x2_t vcvt_s32_f32(float32x2_t __p0) {
5907 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59085907 int32x2_t __ret;
5908 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59095909 __ret = (int32x2_t) __builtin_neon_vcvt_s32_v((int8x8_t)__rev0, 2);
59105910 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
59115911 return __ret;
......@@ -5920,8 +5920,8 @@ __ai uint32x4_t vcvtq_u32_f32(float32x4_t __p0) {
59205920}
59215921#else
59225922__ai uint32x4_t vcvtq_u32_f32(float32x4_t __p0) {
5923 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59245923 uint32x4_t __ret;
5924 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
59255925 __ret = (uint32x4_t) __builtin_neon_vcvtq_u32_v((int8x16_t)__rev0, 50);
59265926 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
59275927 return __ret;
......@@ -5936,8 +5936,8 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
59365936}
59375937#else
59385938__ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
5939 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59405939 uint32x2_t __ret;
5940 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
59415941 __ret = (uint32x2_t) __builtin_neon_vcvt_u32_v((int8x8_t)__rev0, 18);
59425942 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
59435943 return __ret;
......@@ -5946,16 +5946,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
59465946
59475947#ifdef __LITTLE_ENDIAN__
59485948#define vdup_lane_p8(__p0_0, __p1_0) __extension__ ({ \
5949 poly8x8_t __s0_0 = __p0_0; \
59505949 poly8x8_t __ret_0; \
5950 poly8x8_t __s0_0 = __p0_0; \
59515951 __ret_0 = splat_lane_p8(__s0_0, __p1_0); \
59525952 __ret_0; \
59535953})
59545954#else
59555955#define vdup_lane_p8(__p0_1, __p1_1) __extension__ ({ \
5956 poly8x8_t __ret_1; \
59565957 poly8x8_t __s0_1 = __p0_1; \
59575958 poly8x8_t __rev0_1; __rev0_1 = __builtin_shufflevector(__s0_1, __s0_1, 7, 6, 5, 4, 3, 2, 1, 0); \
5958 poly8x8_t __ret_1; \
59595959 __ret_1 = __noswap_splat_lane_p8(__rev0_1, __p1_1); \
59605960 __ret_1 = __builtin_shufflevector(__ret_1, __ret_1, 7, 6, 5, 4, 3, 2, 1, 0); \
59615961 __ret_1; \
......@@ -5964,16 +5964,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
59645964
59655965#ifdef __LITTLE_ENDIAN__
59665966#define vdup_lane_p16(__p0_2, __p1_2) __extension__ ({ \
5967 poly16x4_t __s0_2 = __p0_2; \
59685967 poly16x4_t __ret_2; \
5968 poly16x4_t __s0_2 = __p0_2; \
59695969 __ret_2 = splat_lane_p16(__s0_2, __p1_2); \
59705970 __ret_2; \
59715971})
59725972#else
59735973#define vdup_lane_p16(__p0_3, __p1_3) __extension__ ({ \
5974 poly16x4_t __ret_3; \
59745975 poly16x4_t __s0_3 = __p0_3; \
59755976 poly16x4_t __rev0_3; __rev0_3 = __builtin_shufflevector(__s0_3, __s0_3, 3, 2, 1, 0); \
5976 poly16x4_t __ret_3; \
59775977 __ret_3 = __noswap_splat_lane_p16(__rev0_3, __p1_3); \
59785978 __ret_3 = __builtin_shufflevector(__ret_3, __ret_3, 3, 2, 1, 0); \
59795979 __ret_3; \
......@@ -5982,16 +5982,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
59825982
59835983#ifdef __LITTLE_ENDIAN__
59845984#define vdupq_lane_p8(__p0_4, __p1_4) __extension__ ({ \
5985 poly8x8_t __s0_4 = __p0_4; \
59865985 poly8x16_t __ret_4; \
5986 poly8x8_t __s0_4 = __p0_4; \
59875987 __ret_4 = splatq_lane_p8(__s0_4, __p1_4); \
59885988 __ret_4; \
59895989})
59905990#else
59915991#define vdupq_lane_p8(__p0_5, __p1_5) __extension__ ({ \
5992 poly8x16_t __ret_5; \
59925993 poly8x8_t __s0_5 = __p0_5; \
59935994 poly8x8_t __rev0_5; __rev0_5 = __builtin_shufflevector(__s0_5, __s0_5, 7, 6, 5, 4, 3, 2, 1, 0); \
5994 poly8x16_t __ret_5; \
59955995 __ret_5 = __noswap_splatq_lane_p8(__rev0_5, __p1_5); \
59965996 __ret_5 = __builtin_shufflevector(__ret_5, __ret_5, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
59975997 __ret_5; \
......@@ -6000,16 +6000,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
60006000
60016001#ifdef __LITTLE_ENDIAN__
60026002#define vdupq_lane_p16(__p0_6, __p1_6) __extension__ ({ \
6003 poly16x4_t __s0_6 = __p0_6; \
60046003 poly16x8_t __ret_6; \
6004 poly16x4_t __s0_6 = __p0_6; \
60056005 __ret_6 = splatq_lane_p16(__s0_6, __p1_6); \
60066006 __ret_6; \
60076007})
60086008#else
60096009#define vdupq_lane_p16(__p0_7, __p1_7) __extension__ ({ \
6010 poly16x8_t __ret_7; \
60106011 poly16x4_t __s0_7 = __p0_7; \
60116012 poly16x4_t __rev0_7; __rev0_7 = __builtin_shufflevector(__s0_7, __s0_7, 3, 2, 1, 0); \
6012 poly16x8_t __ret_7; \
60136013 __ret_7 = __noswap_splatq_lane_p16(__rev0_7, __p1_7); \
60146014 __ret_7 = __builtin_shufflevector(__ret_7, __ret_7, 7, 6, 5, 4, 3, 2, 1, 0); \
60156015 __ret_7; \
......@@ -6018,16 +6018,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
60186018
60196019#ifdef __LITTLE_ENDIAN__
60206020#define vdupq_lane_u8(__p0_8, __p1_8) __extension__ ({ \
6021 uint8x8_t __s0_8 = __p0_8; \
60226021 uint8x16_t __ret_8; \
6022 uint8x8_t __s0_8 = __p0_8; \
60236023 __ret_8 = splatq_lane_u8(__s0_8, __p1_8); \
60246024 __ret_8; \
60256025})
60266026#else
60276027#define vdupq_lane_u8(__p0_9, __p1_9) __extension__ ({ \
6028 uint8x16_t __ret_9; \
60286029 uint8x8_t __s0_9 = __p0_9; \
60296030 uint8x8_t __rev0_9; __rev0_9 = __builtin_shufflevector(__s0_9, __s0_9, 7, 6, 5, 4, 3, 2, 1, 0); \
6030 uint8x16_t __ret_9; \
60316031 __ret_9 = __noswap_splatq_lane_u8(__rev0_9, __p1_9); \
60326032 __ret_9 = __builtin_shufflevector(__ret_9, __ret_9, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
60336033 __ret_9; \
......@@ -6036,16 +6036,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
60366036
60376037#ifdef __LITTLE_ENDIAN__
60386038#define vdupq_lane_u32(__p0_10, __p1_10) __extension__ ({ \
6039 uint32x2_t __s0_10 = __p0_10; \
60406039 uint32x4_t __ret_10; \
6040 uint32x2_t __s0_10 = __p0_10; \
60416041 __ret_10 = splatq_lane_u32(__s0_10, __p1_10); \
60426042 __ret_10; \
60436043})
60446044#else
60456045#define vdupq_lane_u32(__p0_11, __p1_11) __extension__ ({ \
6046 uint32x4_t __ret_11; \
60466047 uint32x2_t __s0_11 = __p0_11; \
60476048 uint32x2_t __rev0_11; __rev0_11 = __builtin_shufflevector(__s0_11, __s0_11, 1, 0); \
6048 uint32x4_t __ret_11; \
60496049 __ret_11 = __noswap_splatq_lane_u32(__rev0_11, __p1_11); \
60506050 __ret_11 = __builtin_shufflevector(__ret_11, __ret_11, 3, 2, 1, 0); \
60516051 __ret_11; \
......@@ -6054,15 +6054,15 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
60546054
60556055#ifdef __LITTLE_ENDIAN__
60566056#define vdupq_lane_u64(__p0_12, __p1_12) __extension__ ({ \
6057 uint64x1_t __s0_12 = __p0_12; \
60586057 uint64x2_t __ret_12; \
6058 uint64x1_t __s0_12 = __p0_12; \
60596059 __ret_12 = splatq_lane_u64(__s0_12, __p1_12); \
60606060 __ret_12; \
60616061})
60626062#else
60636063#define vdupq_lane_u64(__p0_13, __p1_13) __extension__ ({ \
6064 uint64x1_t __s0_13 = __p0_13; \
60656064 uint64x2_t __ret_13; \
6065 uint64x1_t __s0_13 = __p0_13; \
60666066 __ret_13 = __noswap_splatq_lane_u64(__s0_13, __p1_13); \
60676067 __ret_13 = __builtin_shufflevector(__ret_13, __ret_13, 1, 0); \
60686068 __ret_13; \
......@@ -6071,16 +6071,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
60716071
60726072#ifdef __LITTLE_ENDIAN__
60736073#define vdupq_lane_u16(__p0_14, __p1_14) __extension__ ({ \
6074 uint16x4_t __s0_14 = __p0_14; \
60756074 uint16x8_t __ret_14; \
6075 uint16x4_t __s0_14 = __p0_14; \
60766076 __ret_14 = splatq_lane_u16(__s0_14, __p1_14); \
60776077 __ret_14; \
60786078})
60796079#else
60806080#define vdupq_lane_u16(__p0_15, __p1_15) __extension__ ({ \
6081 uint16x8_t __ret_15; \
60816082 uint16x4_t __s0_15 = __p0_15; \
60826083 uint16x4_t __rev0_15; __rev0_15 = __builtin_shufflevector(__s0_15, __s0_15, 3, 2, 1, 0); \
6083 uint16x8_t __ret_15; \
60846084 __ret_15 = __noswap_splatq_lane_u16(__rev0_15, __p1_15); \
60856085 __ret_15 = __builtin_shufflevector(__ret_15, __ret_15, 7, 6, 5, 4, 3, 2, 1, 0); \
60866086 __ret_15; \
......@@ -6089,16 +6089,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
60896089
60906090#ifdef __LITTLE_ENDIAN__
60916091#define vdupq_lane_s8(__p0_16, __p1_16) __extension__ ({ \
6092 int8x8_t __s0_16 = __p0_16; \
60936092 int8x16_t __ret_16; \
6093 int8x8_t __s0_16 = __p0_16; \
60946094 __ret_16 = splatq_lane_s8(__s0_16, __p1_16); \
60956095 __ret_16; \
60966096})
60976097#else
60986098#define vdupq_lane_s8(__p0_17, __p1_17) __extension__ ({ \
6099 int8x16_t __ret_17; \
60996100 int8x8_t __s0_17 = __p0_17; \
61006101 int8x8_t __rev0_17; __rev0_17 = __builtin_shufflevector(__s0_17, __s0_17, 7, 6, 5, 4, 3, 2, 1, 0); \
6101 int8x16_t __ret_17; \
61026102 __ret_17 = __noswap_splatq_lane_s8(__rev0_17, __p1_17); \
61036103 __ret_17 = __builtin_shufflevector(__ret_17, __ret_17, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
61046104 __ret_17; \
......@@ -6107,16 +6107,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
61076107
61086108#ifdef __LITTLE_ENDIAN__
61096109#define vdupq_lane_f32(__p0_18, __p1_18) __extension__ ({ \
6110 float32x2_t __s0_18 = __p0_18; \
61116110 float32x4_t __ret_18; \
6111 float32x2_t __s0_18 = __p0_18; \
61126112 __ret_18 = splatq_lane_f32(__s0_18, __p1_18); \
61136113 __ret_18; \
61146114})
61156115#else
61166116#define vdupq_lane_f32(__p0_19, __p1_19) __extension__ ({ \
6117 float32x4_t __ret_19; \
61176118 float32x2_t __s0_19 = __p0_19; \
61186119 float32x2_t __rev0_19; __rev0_19 = __builtin_shufflevector(__s0_19, __s0_19, 1, 0); \
6119 float32x4_t __ret_19; \
61206120 __ret_19 = __noswap_splatq_lane_f32(__rev0_19, __p1_19); \
61216121 __ret_19 = __builtin_shufflevector(__ret_19, __ret_19, 3, 2, 1, 0); \
61226122 __ret_19; \
......@@ -6125,16 +6125,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
61256125
61266126#ifdef __LITTLE_ENDIAN__
61276127#define vdupq_lane_s32(__p0_20, __p1_20) __extension__ ({ \
6128 int32x2_t __s0_20 = __p0_20; \
61296128 int32x4_t __ret_20; \
6129 int32x2_t __s0_20 = __p0_20; \
61306130 __ret_20 = splatq_lane_s32(__s0_20, __p1_20); \
61316131 __ret_20; \
61326132})
61336133#else
61346134#define vdupq_lane_s32(__p0_21, __p1_21) __extension__ ({ \
6135 int32x4_t __ret_21; \
61356136 int32x2_t __s0_21 = __p0_21; \
61366137 int32x2_t __rev0_21; __rev0_21 = __builtin_shufflevector(__s0_21, __s0_21, 1, 0); \
6137 int32x4_t __ret_21; \
61386138 __ret_21 = __noswap_splatq_lane_s32(__rev0_21, __p1_21); \
61396139 __ret_21 = __builtin_shufflevector(__ret_21, __ret_21, 3, 2, 1, 0); \
61406140 __ret_21; \
......@@ -6143,15 +6143,15 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
61436143
61446144#ifdef __LITTLE_ENDIAN__
61456145#define vdupq_lane_s64(__p0_22, __p1_22) __extension__ ({ \
6146 int64x1_t __s0_22 = __p0_22; \
61476146 int64x2_t __ret_22; \
6147 int64x1_t __s0_22 = __p0_22; \
61486148 __ret_22 = splatq_lane_s64(__s0_22, __p1_22); \
61496149 __ret_22; \
61506150})
61516151#else
61526152#define vdupq_lane_s64(__p0_23, __p1_23) __extension__ ({ \
6153 int64x1_t __s0_23 = __p0_23; \
61546153 int64x2_t __ret_23; \
6154 int64x1_t __s0_23 = __p0_23; \
61556155 __ret_23 = __noswap_splatq_lane_s64(__s0_23, __p1_23); \
61566156 __ret_23 = __builtin_shufflevector(__ret_23, __ret_23, 1, 0); \
61576157 __ret_23; \
......@@ -6160,16 +6160,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
61606160
61616161#ifdef __LITTLE_ENDIAN__
61626162#define vdupq_lane_s16(__p0_24, __p1_24) __extension__ ({ \
6163 int16x4_t __s0_24 = __p0_24; \
61646163 int16x8_t __ret_24; \
6164 int16x4_t __s0_24 = __p0_24; \
61656165 __ret_24 = splatq_lane_s16(__s0_24, __p1_24); \
61666166 __ret_24; \
61676167})
61686168#else
61696169#define vdupq_lane_s16(__p0_25, __p1_25) __extension__ ({ \
6170 int16x8_t __ret_25; \
61706171 int16x4_t __s0_25 = __p0_25; \
61716172 int16x4_t __rev0_25; __rev0_25 = __builtin_shufflevector(__s0_25, __s0_25, 3, 2, 1, 0); \
6172 int16x8_t __ret_25; \
61736173 __ret_25 = __noswap_splatq_lane_s16(__rev0_25, __p1_25); \
61746174 __ret_25 = __builtin_shufflevector(__ret_25, __ret_25, 7, 6, 5, 4, 3, 2, 1, 0); \
61756175 __ret_25; \
......@@ -6178,16 +6178,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
61786178
61796179#ifdef __LITTLE_ENDIAN__
61806180#define vdup_lane_u8(__p0_26, __p1_26) __extension__ ({ \
6181 uint8x8_t __s0_26 = __p0_26; \
61826181 uint8x8_t __ret_26; \
6182 uint8x8_t __s0_26 = __p0_26; \
61836183 __ret_26 = splat_lane_u8(__s0_26, __p1_26); \
61846184 __ret_26; \
61856185})
61866186#else
61876187#define vdup_lane_u8(__p0_27, __p1_27) __extension__ ({ \
6188 uint8x8_t __ret_27; \
61886189 uint8x8_t __s0_27 = __p0_27; \
61896190 uint8x8_t __rev0_27; __rev0_27 = __builtin_shufflevector(__s0_27, __s0_27, 7, 6, 5, 4, 3, 2, 1, 0); \
6190 uint8x8_t __ret_27; \
61916191 __ret_27 = __noswap_splat_lane_u8(__rev0_27, __p1_27); \
61926192 __ret_27 = __builtin_shufflevector(__ret_27, __ret_27, 7, 6, 5, 4, 3, 2, 1, 0); \
61936193 __ret_27; \
......@@ -6196,16 +6196,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
61966196
61976197#ifdef __LITTLE_ENDIAN__
61986198#define vdup_lane_u32(__p0_28, __p1_28) __extension__ ({ \
6199 uint32x2_t __s0_28 = __p0_28; \
62006199 uint32x2_t __ret_28; \
6200 uint32x2_t __s0_28 = __p0_28; \
62016201 __ret_28 = splat_lane_u32(__s0_28, __p1_28); \
62026202 __ret_28; \
62036203})
62046204#else
62056205#define vdup_lane_u32(__p0_29, __p1_29) __extension__ ({ \
6206 uint32x2_t __ret_29; \
62066207 uint32x2_t __s0_29 = __p0_29; \
62076208 uint32x2_t __rev0_29; __rev0_29 = __builtin_shufflevector(__s0_29, __s0_29, 1, 0); \
6208 uint32x2_t __ret_29; \
62096209 __ret_29 = __noswap_splat_lane_u32(__rev0_29, __p1_29); \
62106210 __ret_29 = __builtin_shufflevector(__ret_29, __ret_29, 1, 0); \
62116211 __ret_29; \
......@@ -6213,23 +6213,23 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
62136213#endif
62146214
62156215#define vdup_lane_u64(__p0_30, __p1_30) __extension__ ({ \
6216 uint64x1_t __s0_30 = __p0_30; \
62176216 uint64x1_t __ret_30; \
6217 uint64x1_t __s0_30 = __p0_30; \
62186218 __ret_30 = splat_lane_u64(__s0_30, __p1_30); \
62196219 __ret_30; \
62206220})
62216221#ifdef __LITTLE_ENDIAN__
62226222#define vdup_lane_u16(__p0_31, __p1_31) __extension__ ({ \
6223 uint16x4_t __s0_31 = __p0_31; \
62246223 uint16x4_t __ret_31; \
6224 uint16x4_t __s0_31 = __p0_31; \
62256225 __ret_31 = splat_lane_u16(__s0_31, __p1_31); \
62266226 __ret_31; \
62276227})
62286228#else
62296229#define vdup_lane_u16(__p0_32, __p1_32) __extension__ ({ \
6230 uint16x4_t __ret_32; \
62306231 uint16x4_t __s0_32 = __p0_32; \
62316232 uint16x4_t __rev0_32; __rev0_32 = __builtin_shufflevector(__s0_32, __s0_32, 3, 2, 1, 0); \
6232 uint16x4_t __ret_32; \
62336233 __ret_32 = __noswap_splat_lane_u16(__rev0_32, __p1_32); \
62346234 __ret_32 = __builtin_shufflevector(__ret_32, __ret_32, 3, 2, 1, 0); \
62356235 __ret_32; \
......@@ -6238,16 +6238,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
62386238
62396239#ifdef __LITTLE_ENDIAN__
62406240#define vdup_lane_s8(__p0_33, __p1_33) __extension__ ({ \
6241 int8x8_t __s0_33 = __p0_33; \
62426241 int8x8_t __ret_33; \
6242 int8x8_t __s0_33 = __p0_33; \
62436243 __ret_33 = splat_lane_s8(__s0_33, __p1_33); \
62446244 __ret_33; \
62456245})
62466246#else
62476247#define vdup_lane_s8(__p0_34, __p1_34) __extension__ ({ \
6248 int8x8_t __ret_34; \
62486249 int8x8_t __s0_34 = __p0_34; \
62496250 int8x8_t __rev0_34; __rev0_34 = __builtin_shufflevector(__s0_34, __s0_34, 7, 6, 5, 4, 3, 2, 1, 0); \
6250 int8x8_t __ret_34; \
62516251 __ret_34 = __noswap_splat_lane_s8(__rev0_34, __p1_34); \
62526252 __ret_34 = __builtin_shufflevector(__ret_34, __ret_34, 7, 6, 5, 4, 3, 2, 1, 0); \
62536253 __ret_34; \
......@@ -6256,16 +6256,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
62566256
62576257#ifdef __LITTLE_ENDIAN__
62586258#define vdup_lane_f32(__p0_35, __p1_35) __extension__ ({ \
6259 float32x2_t __s0_35 = __p0_35; \
62606259 float32x2_t __ret_35; \
6260 float32x2_t __s0_35 = __p0_35; \
62616261 __ret_35 = splat_lane_f32(__s0_35, __p1_35); \
62626262 __ret_35; \
62636263})
62646264#else
62656265#define vdup_lane_f32(__p0_36, __p1_36) __extension__ ({ \
6266 float32x2_t __ret_36; \
62666267 float32x2_t __s0_36 = __p0_36; \
62676268 float32x2_t __rev0_36; __rev0_36 = __builtin_shufflevector(__s0_36, __s0_36, 1, 0); \
6268 float32x2_t __ret_36; \
62696269 __ret_36 = __noswap_splat_lane_f32(__rev0_36, __p1_36); \
62706270 __ret_36 = __builtin_shufflevector(__ret_36, __ret_36, 1, 0); \
62716271 __ret_36; \
......@@ -6274,16 +6274,16 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
62746274
62756275#ifdef __LITTLE_ENDIAN__
62766276#define vdup_lane_s32(__p0_37, __p1_37) __extension__ ({ \
6277 int32x2_t __s0_37 = __p0_37; \
62786277 int32x2_t __ret_37; \
6278 int32x2_t __s0_37 = __p0_37; \
62796279 __ret_37 = splat_lane_s32(__s0_37, __p1_37); \
62806280 __ret_37; \
62816281})
62826282#else
62836283#define vdup_lane_s32(__p0_38, __p1_38) __extension__ ({ \
6284 int32x2_t __ret_38; \
62846285 int32x2_t __s0_38 = __p0_38; \
62856286 int32x2_t __rev0_38; __rev0_38 = __builtin_shufflevector(__s0_38, __s0_38, 1, 0); \
6286 int32x2_t __ret_38; \
62876287 __ret_38 = __noswap_splat_lane_s32(__rev0_38, __p1_38); \
62886288 __ret_38 = __builtin_shufflevector(__ret_38, __ret_38, 1, 0); \
62896289 __ret_38; \
......@@ -6291,23 +6291,23 @@ __ai uint32x2_t vcvt_u32_f32(float32x2_t __p0) {
62916291#endif
62926292
62936293#define vdup_lane_s64(__p0_39, __p1_39) __extension__ ({ \
6294 int64x1_t __s0_39 = __p0_39; \
62956294 int64x1_t __ret_39; \
6295 int64x1_t __s0_39 = __p0_39; \
62966296 __ret_39 = splat_lane_s64(__s0_39, __p1_39); \
62976297 __ret_39; \
62986298})
62996299#ifdef __LITTLE_ENDIAN__
63006300#define vdup_lane_s16(__p0_40, __p1_40) __extension__ ({ \
6301 int16x4_t __s0_40 = __p0_40; \
63026301 int16x4_t __ret_40; \
6302 int16x4_t __s0_40 = __p0_40; \
63036303 __ret_40 = splat_lane_s16(__s0_40, __p1_40); \
63046304 __ret_40; \
63056305})
63066306#else
63076307#define vdup_lane_s16(__p0_41, __p1_41) __extension__ ({ \
6308 int16x4_t __ret_41; \
63086309 int16x4_t __s0_41 = __p0_41; \
63096310 int16x4_t __rev0_41; __rev0_41 = __builtin_shufflevector(__s0_41, __s0_41, 3, 2, 1, 0); \
6310 int16x4_t __ret_41; \
63116311 __ret_41 = __noswap_splat_lane_s16(__rev0_41, __p1_41); \
63126312 __ret_41 = __builtin_shufflevector(__ret_41, __ret_41, 3, 2, 1, 0); \
63136313 __ret_41; \
......@@ -6466,15 +6466,15 @@ __ai float32x4_t vdupq_n_f32(float32_t __p0) {
64666466
64676467#ifdef __LITTLE_ENDIAN__
64686468#define vdupq_n_f16(__p0) __extension__ ({ \
6469 float16_t __s0 = __p0; \
64706469 float16x8_t __ret; \
6470 float16_t __s0 = __p0; \
64716471 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
64726472 __ret; \
64736473})
64746474#else
64756475#define vdupq_n_f16(__p0) __extension__ ({ \
6476 float16_t __s0 = __p0; \
64776476 float16x8_t __ret; \
6477 float16_t __s0 = __p0; \
64786478 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
64796479 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
64806480 __ret; \
......@@ -6608,15 +6608,15 @@ __ai float32x2_t vdup_n_f32(float32_t __p0) {
66086608
66096609#ifdef __LITTLE_ENDIAN__
66106610#define vdup_n_f16(__p0) __extension__ ({ \
6611 float16_t __s0 = __p0; \
66126611 float16x4_t __ret; \
6612 float16_t __s0 = __p0; \
66136613 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
66146614 __ret; \
66156615})
66166616#else
66176617#define vdup_n_f16(__p0) __extension__ ({ \
6618 float16_t __s0 = __p0; \
66196618 float16x4_t __ret; \
6619 float16_t __s0 = __p0; \
66206620 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
66216621 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
66226622 __ret; \
......@@ -6666,9 +6666,9 @@ __ai uint8x16_t veorq_u8(uint8x16_t __p0, uint8x16_t __p1) {
66666666}
66676667#else
66686668__ai uint8x16_t veorq_u8(uint8x16_t __p0, uint8x16_t __p1) {
6669 uint8x16_t __ret;
66696670 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66706671 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6671 uint8x16_t __ret;
66726672 __ret = __rev0 ^ __rev1;
66736673 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66746674 return __ret;
......@@ -6683,9 +6683,9 @@ __ai uint32x4_t veorq_u32(uint32x4_t __p0, uint32x4_t __p1) {
66836683}
66846684#else
66856685__ai uint32x4_t veorq_u32(uint32x4_t __p0, uint32x4_t __p1) {
6686 uint32x4_t __ret;
66866687 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
66876688 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6688 uint32x4_t __ret;
66896689 __ret = __rev0 ^ __rev1;
66906690 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
66916691 return __ret;
......@@ -6700,9 +6700,9 @@ __ai uint64x2_t veorq_u64(uint64x2_t __p0, uint64x2_t __p1) {
67006700}
67016701#else
67026702__ai uint64x2_t veorq_u64(uint64x2_t __p0, uint64x2_t __p1) {
6703 uint64x2_t __ret;
67036704 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
67046705 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6705 uint64x2_t __ret;
67066706 __ret = __rev0 ^ __rev1;
67076707 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
67086708 return __ret;
......@@ -6717,9 +6717,9 @@ __ai uint16x8_t veorq_u16(uint16x8_t __p0, uint16x8_t __p1) {
67176717}
67186718#else
67196719__ai uint16x8_t veorq_u16(uint16x8_t __p0, uint16x8_t __p1) {
6720 uint16x8_t __ret;
67206721 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
67216722 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6722 uint16x8_t __ret;
67236723 __ret = __rev0 ^ __rev1;
67246724 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
67256725 return __ret;
......@@ -6734,9 +6734,9 @@ __ai int8x16_t veorq_s8(int8x16_t __p0, int8x16_t __p1) {
67346734}
67356735#else
67366736__ai int8x16_t veorq_s8(int8x16_t __p0, int8x16_t __p1) {
6737 int8x16_t __ret;
67376738 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67386739 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6739 int8x16_t __ret;
67406740 __ret = __rev0 ^ __rev1;
67416741 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67426742 return __ret;
......@@ -6751,9 +6751,9 @@ __ai int32x4_t veorq_s32(int32x4_t __p0, int32x4_t __p1) {
67516751}
67526752#else
67536753__ai int32x4_t veorq_s32(int32x4_t __p0, int32x4_t __p1) {
6754 int32x4_t __ret;
67546755 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
67556756 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6756 int32x4_t __ret;
67576757 __ret = __rev0 ^ __rev1;
67586758 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
67596759 return __ret;
......@@ -6768,9 +6768,9 @@ __ai int64x2_t veorq_s64(int64x2_t __p0, int64x2_t __p1) {
67686768}
67696769#else
67706770__ai int64x2_t veorq_s64(int64x2_t __p0, int64x2_t __p1) {
6771 int64x2_t __ret;
67716772 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
67726773 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6773 int64x2_t __ret;
67746774 __ret = __rev0 ^ __rev1;
67756775 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
67766776 return __ret;
......@@ -6785,9 +6785,9 @@ __ai int16x8_t veorq_s16(int16x8_t __p0, int16x8_t __p1) {
67856785}
67866786#else
67876787__ai int16x8_t veorq_s16(int16x8_t __p0, int16x8_t __p1) {
6788 int16x8_t __ret;
67886789 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
67896790 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6790 int16x8_t __ret;
67916791 __ret = __rev0 ^ __rev1;
67926792 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
67936793 return __ret;
......@@ -6802,9 +6802,9 @@ __ai uint8x8_t veor_u8(uint8x8_t __p0, uint8x8_t __p1) {
68026802}
68036803#else
68046804__ai uint8x8_t veor_u8(uint8x8_t __p0, uint8x8_t __p1) {
6805 uint8x8_t __ret;
68056806 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
68066807 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6807 uint8x8_t __ret;
68086808 __ret = __rev0 ^ __rev1;
68096809 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
68106810 return __ret;
......@@ -6819,9 +6819,9 @@ __ai uint32x2_t veor_u32(uint32x2_t __p0, uint32x2_t __p1) {
68196819}
68206820#else
68216821__ai uint32x2_t veor_u32(uint32x2_t __p0, uint32x2_t __p1) {
6822 uint32x2_t __ret;
68226823 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
68236824 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6824 uint32x2_t __ret;
68256825 __ret = __rev0 ^ __rev1;
68266826 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
68276827 return __ret;
......@@ -6841,9 +6841,9 @@ __ai uint16x4_t veor_u16(uint16x4_t __p0, uint16x4_t __p1) {
68416841}
68426842#else
68436843__ai uint16x4_t veor_u16(uint16x4_t __p0, uint16x4_t __p1) {
6844 uint16x4_t __ret;
68446845 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
68456846 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6846 uint16x4_t __ret;
68476847 __ret = __rev0 ^ __rev1;
68486848 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
68496849 return __ret;
......@@ -6858,9 +6858,9 @@ __ai int8x8_t veor_s8(int8x8_t __p0, int8x8_t __p1) {
68586858}
68596859#else
68606860__ai int8x8_t veor_s8(int8x8_t __p0, int8x8_t __p1) {
6861 int8x8_t __ret;
68616862 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
68626863 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6863 int8x8_t __ret;
68646864 __ret = __rev0 ^ __rev1;
68656865 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
68666866 return __ret;
......@@ -6875,9 +6875,9 @@ __ai int32x2_t veor_s32(int32x2_t __p0, int32x2_t __p1) {
68756875}
68766876#else
68776877__ai int32x2_t veor_s32(int32x2_t __p0, int32x2_t __p1) {
6878 int32x2_t __ret;
68786879 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
68796880 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6880 int32x2_t __ret;
68816881 __ret = __rev0 ^ __rev1;
68826882 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
68836883 return __ret;
......@@ -6897,9 +6897,9 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
68976897}
68986898#else
68996899__ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
6900 int16x4_t __ret;
69006901 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
69016902 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6902 int16x4_t __ret;
69036903 __ret = __rev0 ^ __rev1;
69046904 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
69056905 return __ret;
......@@ -6908,19 +6908,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
69086908
69096909#ifdef __LITTLE_ENDIAN__
69106910#define vext_p8(__p0, __p1, __p2) __extension__ ({ \
6911 poly8x8_t __ret; \
69116912 poly8x8_t __s0 = __p0; \
69126913 poly8x8_t __s1 = __p1; \
6913 poly8x8_t __ret; \
69146914 __ret = (poly8x8_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 4); \
69156915 __ret; \
69166916})
69176917#else
69186918#define vext_p8(__p0, __p1, __p2) __extension__ ({ \
6919 poly8x8_t __ret; \
69196920 poly8x8_t __s0 = __p0; \
69206921 poly8x8_t __s1 = __p1; \
69216922 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
69226923 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
6923 poly8x8_t __ret; \
69246924 __ret = (poly8x8_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 4); \
69256925 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
69266926 __ret; \
......@@ -6929,19 +6929,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
69296929
69306930#ifdef __LITTLE_ENDIAN__
69316931#define vext_p16(__p0, __p1, __p2) __extension__ ({ \
6932 poly16x4_t __ret; \
69326933 poly16x4_t __s0 = __p0; \
69336934 poly16x4_t __s1 = __p1; \
6934 poly16x4_t __ret; \
69356935 __ret = (poly16x4_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 5); \
69366936 __ret; \
69376937})
69386938#else
69396939#define vext_p16(__p0, __p1, __p2) __extension__ ({ \
6940 poly16x4_t __ret; \
69406941 poly16x4_t __s0 = __p0; \
69416942 poly16x4_t __s1 = __p1; \
69426943 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
69436944 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
6944 poly16x4_t __ret; \
69456945 __ret = (poly16x4_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 5); \
69466946 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
69476947 __ret; \
......@@ -6950,19 +6950,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
69506950
69516951#ifdef __LITTLE_ENDIAN__
69526952#define vextq_p8(__p0, __p1, __p2) __extension__ ({ \
6953 poly8x16_t __ret; \
69536954 poly8x16_t __s0 = __p0; \
69546955 poly8x16_t __s1 = __p1; \
6955 poly8x16_t __ret; \
69566956 __ret = (poly8x16_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 36); \
69576957 __ret; \
69586958})
69596959#else
69606960#define vextq_p8(__p0, __p1, __p2) __extension__ ({ \
6961 poly8x16_t __ret; \
69616962 poly8x16_t __s0 = __p0; \
69626963 poly8x16_t __s1 = __p1; \
69636964 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
69646965 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6965 poly8x16_t __ret; \
69666966 __ret = (poly8x16_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 36); \
69676967 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
69686968 __ret; \
......@@ -6971,19 +6971,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
69716971
69726972#ifdef __LITTLE_ENDIAN__
69736973#define vextq_p16(__p0, __p1, __p2) __extension__ ({ \
6974 poly16x8_t __ret; \
69746975 poly16x8_t __s0 = __p0; \
69756976 poly16x8_t __s1 = __p1; \
6976 poly16x8_t __ret; \
69776977 __ret = (poly16x8_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 37); \
69786978 __ret; \
69796979})
69806980#else
69816981#define vextq_p16(__p0, __p1, __p2) __extension__ ({ \
6982 poly16x8_t __ret; \
69826983 poly16x8_t __s0 = __p0; \
69836984 poly16x8_t __s1 = __p1; \
69846985 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
69856986 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
6986 poly16x8_t __ret; \
69876987 __ret = (poly16x8_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 37); \
69886988 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
69896989 __ret; \
......@@ -6992,19 +6992,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
69926992
69936993#ifdef __LITTLE_ENDIAN__
69946994#define vextq_u8(__p0, __p1, __p2) __extension__ ({ \
6995 uint8x16_t __ret; \
69956996 uint8x16_t __s0 = __p0; \
69966997 uint8x16_t __s1 = __p1; \
6997 uint8x16_t __ret; \
69986998 __ret = (uint8x16_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 48); \
69996999 __ret; \
70007000})
70017001#else
70027002#define vextq_u8(__p0, __p1, __p2) __extension__ ({ \
7003 uint8x16_t __ret; \
70037004 uint8x16_t __s0 = __p0; \
70047005 uint8x16_t __s1 = __p1; \
70057006 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
70067007 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
7007 uint8x16_t __ret; \
70087008 __ret = (uint8x16_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 48); \
70097009 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
70107010 __ret; \
......@@ -7013,19 +7013,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
70137013
70147014#ifdef __LITTLE_ENDIAN__
70157015#define vextq_u32(__p0, __p1, __p2) __extension__ ({ \
7016 uint32x4_t __ret; \
70167017 uint32x4_t __s0 = __p0; \
70177018 uint32x4_t __s1 = __p1; \
7018 uint32x4_t __ret; \
70197019 __ret = (uint32x4_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 50); \
70207020 __ret; \
70217021})
70227022#else
70237023#define vextq_u32(__p0, __p1, __p2) __extension__ ({ \
7024 uint32x4_t __ret; \
70247025 uint32x4_t __s0 = __p0; \
70257026 uint32x4_t __s1 = __p1; \
70267027 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
70277028 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
7028 uint32x4_t __ret; \
70297029 __ret = (uint32x4_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 50); \
70307030 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
70317031 __ret; \
......@@ -7034,19 +7034,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
70347034
70357035#ifdef __LITTLE_ENDIAN__
70367036#define vextq_u64(__p0, __p1, __p2) __extension__ ({ \
7037 uint64x2_t __ret; \
70377038 uint64x2_t __s0 = __p0; \
70387039 uint64x2_t __s1 = __p1; \
7039 uint64x2_t __ret; \
70407040 __ret = (uint64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
70417041 __ret; \
70427042})
70437043#else
70447044#define vextq_u64(__p0, __p1, __p2) __extension__ ({ \
7045 uint64x2_t __ret; \
70457046 uint64x2_t __s0 = __p0; \
70467047 uint64x2_t __s1 = __p1; \
70477048 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
70487049 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
7049 uint64x2_t __ret; \
70507050 __ret = (uint64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
70517051 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
70527052 __ret; \
......@@ -7055,19 +7055,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
70557055
70567056#ifdef __LITTLE_ENDIAN__
70577057#define vextq_u16(__p0, __p1, __p2) __extension__ ({ \
7058 uint16x8_t __ret; \
70587059 uint16x8_t __s0 = __p0; \
70597060 uint16x8_t __s1 = __p1; \
7060 uint16x8_t __ret; \
70617061 __ret = (uint16x8_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 49); \
70627062 __ret; \
70637063})
70647064#else
70657065#define vextq_u16(__p0, __p1, __p2) __extension__ ({ \
7066 uint16x8_t __ret; \
70667067 uint16x8_t __s0 = __p0; \
70677068 uint16x8_t __s1 = __p1; \
70687069 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
70697070 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
7070 uint16x8_t __ret; \
70717071 __ret = (uint16x8_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 49); \
70727072 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
70737073 __ret; \
......@@ -7076,19 +7076,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
70767076
70777077#ifdef __LITTLE_ENDIAN__
70787078#define vextq_s8(__p0, __p1, __p2) __extension__ ({ \
7079 int8x16_t __ret; \
70797080 int8x16_t __s0 = __p0; \
70807081 int8x16_t __s1 = __p1; \
7081 int8x16_t __ret; \
70827082 __ret = (int8x16_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 32); \
70837083 __ret; \
70847084})
70857085#else
70867086#define vextq_s8(__p0, __p1, __p2) __extension__ ({ \
7087 int8x16_t __ret; \
70877088 int8x16_t __s0 = __p0; \
70887089 int8x16_t __s1 = __p1; \
70897090 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
70907091 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
7091 int8x16_t __ret; \
70927092 __ret = (int8x16_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 32); \
70937093 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
70947094 __ret; \
......@@ -7097,19 +7097,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
70977097
70987098#ifdef __LITTLE_ENDIAN__
70997099#define vextq_f32(__p0, __p1, __p2) __extension__ ({ \
7100 float32x4_t __ret; \
71007101 float32x4_t __s0 = __p0; \
71017102 float32x4_t __s1 = __p1; \
7102 float32x4_t __ret; \
71037103 __ret = (float32x4_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 41); \
71047104 __ret; \
71057105})
71067106#else
71077107#define vextq_f32(__p0, __p1, __p2) __extension__ ({ \
7108 float32x4_t __ret; \
71087109 float32x4_t __s0 = __p0; \
71097110 float32x4_t __s1 = __p1; \
71107111 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
71117112 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
7112 float32x4_t __ret; \
71137113 __ret = (float32x4_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 41); \
71147114 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
71157115 __ret; \
......@@ -7118,19 +7118,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
71187118
71197119#ifdef __LITTLE_ENDIAN__
71207120#define vextq_s32(__p0, __p1, __p2) __extension__ ({ \
7121 int32x4_t __ret; \
71217122 int32x4_t __s0 = __p0; \
71227123 int32x4_t __s1 = __p1; \
7123 int32x4_t __ret; \
71247124 __ret = (int32x4_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
71257125 __ret; \
71267126})
71277127#else
71287128#define vextq_s32(__p0, __p1, __p2) __extension__ ({ \
7129 int32x4_t __ret; \
71297130 int32x4_t __s0 = __p0; \
71307131 int32x4_t __s1 = __p1; \
71317132 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
71327133 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
7133 int32x4_t __ret; \
71347134 __ret = (int32x4_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
71357135 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
71367136 __ret; \
......@@ -7139,19 +7139,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
71397139
71407140#ifdef __LITTLE_ENDIAN__
71417141#define vextq_s64(__p0, __p1, __p2) __extension__ ({ \
7142 int64x2_t __ret; \
71427143 int64x2_t __s0 = __p0; \
71437144 int64x2_t __s1 = __p1; \
7144 int64x2_t __ret; \
71457145 __ret = (int64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 35); \
71467146 __ret; \
71477147})
71487148#else
71497149#define vextq_s64(__p0, __p1, __p2) __extension__ ({ \
7150 int64x2_t __ret; \
71507151 int64x2_t __s0 = __p0; \
71517152 int64x2_t __s1 = __p1; \
71527153 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
71537154 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
7154 int64x2_t __ret; \
71557155 __ret = (int64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 35); \
71567156 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
71577157 __ret; \
......@@ -7160,19 +7160,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
71607160
71617161#ifdef __LITTLE_ENDIAN__
71627162#define vextq_s16(__p0, __p1, __p2) __extension__ ({ \
7163 int16x8_t __ret; \
71637164 int16x8_t __s0 = __p0; \
71647165 int16x8_t __s1 = __p1; \
7165 int16x8_t __ret; \
71667166 __ret = (int16x8_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
71677167 __ret; \
71687168})
71697169#else
71707170#define vextq_s16(__p0, __p1, __p2) __extension__ ({ \
7171 int16x8_t __ret; \
71717172 int16x8_t __s0 = __p0; \
71727173 int16x8_t __s1 = __p1; \
71737174 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
71747175 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
7175 int16x8_t __ret; \
71767176 __ret = (int16x8_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
71777177 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
71787178 __ret; \
......@@ -7181,19 +7181,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
71817181
71827182#ifdef __LITTLE_ENDIAN__
71837183#define vext_u8(__p0, __p1, __p2) __extension__ ({ \
7184 uint8x8_t __ret; \
71847185 uint8x8_t __s0 = __p0; \
71857186 uint8x8_t __s1 = __p1; \
7186 uint8x8_t __ret; \
71877187 __ret = (uint8x8_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 16); \
71887188 __ret; \
71897189})
71907190#else
71917191#define vext_u8(__p0, __p1, __p2) __extension__ ({ \
7192 uint8x8_t __ret; \
71927193 uint8x8_t __s0 = __p0; \
71937194 uint8x8_t __s1 = __p1; \
71947195 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
71957196 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
7196 uint8x8_t __ret; \
71977197 __ret = (uint8x8_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 16); \
71987198 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
71997199 __ret; \
......@@ -7202,19 +7202,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
72027202
72037203#ifdef __LITTLE_ENDIAN__
72047204#define vext_u32(__p0, __p1, __p2) __extension__ ({ \
7205 uint32x2_t __ret; \
72057206 uint32x2_t __s0 = __p0; \
72067207 uint32x2_t __s1 = __p1; \
7207 uint32x2_t __ret; \
72087208 __ret = (uint32x2_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 18); \
72097209 __ret; \
72107210})
72117211#else
72127212#define vext_u32(__p0, __p1, __p2) __extension__ ({ \
7213 uint32x2_t __ret; \
72137214 uint32x2_t __s0 = __p0; \
72147215 uint32x2_t __s1 = __p1; \
72157216 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
72167217 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
7217 uint32x2_t __ret; \
72187218 __ret = (uint32x2_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 18); \
72197219 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
72207220 __ret; \
......@@ -7222,27 +7222,27 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
72227222#endif
72237223
72247224#define vext_u64(__p0, __p1, __p2) __extension__ ({ \
7225 uint64x1_t __ret; \
72257226 uint64x1_t __s0 = __p0; \
72267227 uint64x1_t __s1 = __p1; \
7227 uint64x1_t __ret; \
72287228 __ret = (uint64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
72297229 __ret; \
72307230})
72317231#ifdef __LITTLE_ENDIAN__
72327232#define vext_u16(__p0, __p1, __p2) __extension__ ({ \
7233 uint16x4_t __ret; \
72337234 uint16x4_t __s0 = __p0; \
72347235 uint16x4_t __s1 = __p1; \
7235 uint16x4_t __ret; \
72367236 __ret = (uint16x4_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 17); \
72377237 __ret; \
72387238})
72397239#else
72407240#define vext_u16(__p0, __p1, __p2) __extension__ ({ \
7241 uint16x4_t __ret; \
72417242 uint16x4_t __s0 = __p0; \
72427243 uint16x4_t __s1 = __p1; \
72437244 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
72447245 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
7245 uint16x4_t __ret; \
72467246 __ret = (uint16x4_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 17); \
72477247 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
72487248 __ret; \
......@@ -7251,19 +7251,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
72517251
72527252#ifdef __LITTLE_ENDIAN__
72537253#define vext_s8(__p0, __p1, __p2) __extension__ ({ \
7254 int8x8_t __ret; \
72547255 int8x8_t __s0 = __p0; \
72557256 int8x8_t __s1 = __p1; \
7256 int8x8_t __ret; \
72577257 __ret = (int8x8_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 0); \
72587258 __ret; \
72597259})
72607260#else
72617261#define vext_s8(__p0, __p1, __p2) __extension__ ({ \
7262 int8x8_t __ret; \
72627263 int8x8_t __s0 = __p0; \
72637264 int8x8_t __s1 = __p1; \
72647265 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
72657266 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
7266 int8x8_t __ret; \
72677267 __ret = (int8x8_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 0); \
72687268 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
72697269 __ret; \
......@@ -7272,19 +7272,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
72727272
72737273#ifdef __LITTLE_ENDIAN__
72747274#define vext_f32(__p0, __p1, __p2) __extension__ ({ \
7275 float32x2_t __ret; \
72757276 float32x2_t __s0 = __p0; \
72767277 float32x2_t __s1 = __p1; \
7277 float32x2_t __ret; \
72787278 __ret = (float32x2_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 9); \
72797279 __ret; \
72807280})
72817281#else
72827282#define vext_f32(__p0, __p1, __p2) __extension__ ({ \
7283 float32x2_t __ret; \
72837284 float32x2_t __s0 = __p0; \
72847285 float32x2_t __s1 = __p1; \
72857286 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
72867287 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
7287 float32x2_t __ret; \
72887288 __ret = (float32x2_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 9); \
72897289 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
72907290 __ret; \
......@@ -7293,19 +7293,19 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
72937293
72947294#ifdef __LITTLE_ENDIAN__
72957295#define vext_s32(__p0, __p1, __p2) __extension__ ({ \
7296 int32x2_t __ret; \
72967297 int32x2_t __s0 = __p0; \
72977298 int32x2_t __s1 = __p1; \
7298 int32x2_t __ret; \
72997299 __ret = (int32x2_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
73007300 __ret; \
73017301})
73027302#else
73037303#define vext_s32(__p0, __p1, __p2) __extension__ ({ \
7304 int32x2_t __ret; \
73047305 int32x2_t __s0 = __p0; \
73057306 int32x2_t __s1 = __p1; \
73067307 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
73077308 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
7308 int32x2_t __ret; \
73097309 __ret = (int32x2_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
73107310 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
73117311 __ret; \
......@@ -7313,27 +7313,27 @@ __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
73137313#endif
73147314
73157315#define vext_s64(__p0, __p1, __p2) __extension__ ({ \
7316 int64x1_t __ret; \
73167317 int64x1_t __s0 = __p0; \
73177318 int64x1_t __s1 = __p1; \
7318 int64x1_t __ret; \
73197319 __ret = (int64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
73207320 __ret; \
73217321})
73227322#ifdef __LITTLE_ENDIAN__
73237323#define vext_s16(__p0, __p1, __p2) __extension__ ({ \
7324 int16x4_t __ret; \
73247325 int16x4_t __s0 = __p0; \
73257326 int16x4_t __s1 = __p1; \
7326 int16x4_t __ret; \
73277327 __ret = (int16x4_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
73287328 __ret; \
73297329})
73307330#else
73317331#define vext_s16(__p0, __p1, __p2) __extension__ ({ \
7332 int16x4_t __ret; \
73327333 int16x4_t __s0 = __p0; \
73337334 int16x4_t __s1 = __p1; \
73347335 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
73357336 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
7336 int16x4_t __ret; \
73377337 __ret = (int16x4_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
73387338 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
73397339 __ret; \
......@@ -7348,8 +7348,8 @@ __ai poly8x8_t vget_high_p8(poly8x16_t __p0) {
73487348}
73497349#else
73507350__ai poly8x8_t vget_high_p8(poly8x16_t __p0) {
7351 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
73527351 poly8x8_t __ret;
7352 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
73537353 __ret = __builtin_shufflevector(__rev0, __rev0, 8, 9, 10, 11, 12, 13, 14, 15);
73547354 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
73557355 return __ret;
......@@ -7369,8 +7369,8 @@ __ai poly16x4_t vget_high_p16(poly16x8_t __p0) {
73697369}
73707370#else
73717371__ai poly16x4_t vget_high_p16(poly16x8_t __p0) {
7372 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
73737372 poly16x4_t __ret;
7373 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
73747374 __ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
73757375 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
73767376 return __ret;
......@@ -7385,8 +7385,8 @@ __ai uint8x8_t vget_high_u8(uint8x16_t __p0) {
73857385}
73867386#else
73877387__ai uint8x8_t vget_high_u8(uint8x16_t __p0) {
7388 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
73897388 uint8x8_t __ret;
7389 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
73907390 __ret = __builtin_shufflevector(__rev0, __rev0, 8, 9, 10, 11, 12, 13, 14, 15);
73917391 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
73927392 return __ret;
......@@ -7406,8 +7406,8 @@ __ai uint32x2_t vget_high_u32(uint32x4_t __p0) {
74067406}
74077407#else
74087408__ai uint32x2_t vget_high_u32(uint32x4_t __p0) {
7409 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
74107409 uint32x2_t __ret;
7410 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
74117411 __ret = __builtin_shufflevector(__rev0, __rev0, 2, 3);
74127412 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
74137413 return __ret;
......@@ -7427,8 +7427,8 @@ __ai uint64x1_t vget_high_u64(uint64x2_t __p0) {
74277427}
74287428#else
74297429__ai uint64x1_t vget_high_u64(uint64x2_t __p0) {
7430 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
74317430 uint64x1_t __ret;
7431 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
74327432 __ret = __builtin_shufflevector(__rev0, __rev0, 1);
74337433 return __ret;
74347434}
......@@ -7442,8 +7442,8 @@ __ai uint16x4_t vget_high_u16(uint16x8_t __p0) {
74427442}
74437443#else
74447444__ai uint16x4_t vget_high_u16(uint16x8_t __p0) {
7445 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
74467445 uint16x4_t __ret;
7446 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
74477447 __ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
74487448 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
74497449 return __ret;
......@@ -7463,8 +7463,8 @@ __ai int8x8_t vget_high_s8(int8x16_t __p0) {
74637463}
74647464#else
74657465__ai int8x8_t vget_high_s8(int8x16_t __p0) {
7466 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
74677466 int8x8_t __ret;
7467 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
74687468 __ret = __builtin_shufflevector(__rev0, __rev0, 8, 9, 10, 11, 12, 13, 14, 15);
74697469 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
74707470 return __ret;
......@@ -7484,8 +7484,8 @@ __ai float32x2_t vget_high_f32(float32x4_t __p0) {
74847484}
74857485#else
74867486__ai float32x2_t vget_high_f32(float32x4_t __p0) {
7487 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
74887487 float32x2_t __ret;
7488 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
74897489 __ret = __builtin_shufflevector(__rev0, __rev0, 2, 3);
74907490 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
74917491 return __ret;
......@@ -7505,8 +7505,8 @@ __ai float16x4_t vget_high_f16(float16x8_t __p0) {
75057505}
75067506#else
75077507__ai float16x4_t vget_high_f16(float16x8_t __p0) {
7508 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
75097508 float16x4_t __ret;
7509 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
75107510 __ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
75117511 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
75127512 return __ret;
......@@ -7526,8 +7526,8 @@ __ai int32x2_t vget_high_s32(int32x4_t __p0) {
75267526}
75277527#else
75287528__ai int32x2_t vget_high_s32(int32x4_t __p0) {
7529 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
75307529 int32x2_t __ret;
7530 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
75317531 __ret = __builtin_shufflevector(__rev0, __rev0, 2, 3);
75327532 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
75337533 return __ret;
......@@ -7547,8 +7547,8 @@ __ai int64x1_t vget_high_s64(int64x2_t __p0) {
75477547}
75487548#else
75497549__ai int64x1_t vget_high_s64(int64x2_t __p0) {
7550 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
75517550 int64x1_t __ret;
7551 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
75527552 __ret = __builtin_shufflevector(__rev0, __rev0, 1);
75537553 return __ret;
75547554}
......@@ -7562,8 +7562,8 @@ __ai int16x4_t vget_high_s16(int16x8_t __p0) {
75627562}
75637563#else
75647564__ai int16x4_t vget_high_s16(int16x8_t __p0) {
7565 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
75667565 int16x4_t __ret;
7566 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
75677567 __ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
75687568 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
75697569 return __ret;
......@@ -7577,22 +7577,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
75777577
75787578#ifdef __LITTLE_ENDIAN__
75797579#define vget_lane_p8(__p0, __p1) __extension__ ({ \
7580 poly8x8_t __s0 = __p0; \
75817580 poly8_t __ret; \
7581 poly8x8_t __s0 = __p0; \
75827582 __ret = (poly8_t) __builtin_neon_vget_lane_i8((poly8x8_t)__s0, __p1); \
75837583 __ret; \
75847584})
75857585#else
75867586#define vget_lane_p8(__p0, __p1) __extension__ ({ \
7587 poly8_t __ret; \
75877588 poly8x8_t __s0 = __p0; \
75887589 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
7589 poly8_t __ret; \
75907590 __ret = (poly8_t) __builtin_neon_vget_lane_i8((poly8x8_t)__rev0, __p1); \
75917591 __ret; \
75927592})
75937593#define __noswap_vget_lane_p8(__p0, __p1) __extension__ ({ \
7594 poly8x8_t __s0 = __p0; \
75957594 poly8_t __ret; \
7595 poly8x8_t __s0 = __p0; \
75967596 __ret = (poly8_t) __builtin_neon_vget_lane_i8((poly8x8_t)__s0, __p1); \
75977597 __ret; \
75987598})
......@@ -7600,22 +7600,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
76007600
76017601#ifdef __LITTLE_ENDIAN__
76027602#define vget_lane_p16(__p0, __p1) __extension__ ({ \
7603 poly16x4_t __s0 = __p0; \
76047603 poly16_t __ret; \
7604 poly16x4_t __s0 = __p0; \
76057605 __ret = (poly16_t) __builtin_neon_vget_lane_i16((poly16x4_t)__s0, __p1); \
76067606 __ret; \
76077607})
76087608#else
76097609#define vget_lane_p16(__p0, __p1) __extension__ ({ \
7610 poly16_t __ret; \
76107611 poly16x4_t __s0 = __p0; \
76117612 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
7612 poly16_t __ret; \
76137613 __ret = (poly16_t) __builtin_neon_vget_lane_i16((poly16x4_t)__rev0, __p1); \
76147614 __ret; \
76157615})
76167616#define __noswap_vget_lane_p16(__p0, __p1) __extension__ ({ \
7617 poly16x4_t __s0 = __p0; \
76187617 poly16_t __ret; \
7618 poly16x4_t __s0 = __p0; \
76197619 __ret = (poly16_t) __builtin_neon_vget_lane_i16((poly16x4_t)__s0, __p1); \
76207620 __ret; \
76217621})
......@@ -7623,22 +7623,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
76237623
76247624#ifdef __LITTLE_ENDIAN__
76257625#define vgetq_lane_p8(__p0, __p1) __extension__ ({ \
7626 poly8x16_t __s0 = __p0; \
76277626 poly8_t __ret; \
7627 poly8x16_t __s0 = __p0; \
76287628 __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((poly8x16_t)__s0, __p1); \
76297629 __ret; \
76307630})
76317631#else
76327632#define vgetq_lane_p8(__p0, __p1) __extension__ ({ \
7633 poly8_t __ret; \
76337634 poly8x16_t __s0 = __p0; \
76347635 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
7635 poly8_t __ret; \
76367636 __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((poly8x16_t)__rev0, __p1); \
76377637 __ret; \
76387638})
76397639#define __noswap_vgetq_lane_p8(__p0, __p1) __extension__ ({ \
7640 poly8x16_t __s0 = __p0; \
76417640 poly8_t __ret; \
7641 poly8x16_t __s0 = __p0; \
76427642 __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((poly8x16_t)__s0, __p1); \
76437643 __ret; \
76447644})
......@@ -7646,22 +7646,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
76467646
76477647#ifdef __LITTLE_ENDIAN__
76487648#define vgetq_lane_p16(__p0, __p1) __extension__ ({ \
7649 poly16x8_t __s0 = __p0; \
76507649 poly16_t __ret; \
7650 poly16x8_t __s0 = __p0; \
76517651 __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((poly16x8_t)__s0, __p1); \
76527652 __ret; \
76537653})
76547654#else
76557655#define vgetq_lane_p16(__p0, __p1) __extension__ ({ \
7656 poly16_t __ret; \
76567657 poly16x8_t __s0 = __p0; \
76577658 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
7658 poly16_t __ret; \
76597659 __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((poly16x8_t)__rev0, __p1); \
76607660 __ret; \
76617661})
76627662#define __noswap_vgetq_lane_p16(__p0, __p1) __extension__ ({ \
7663 poly16x8_t __s0 = __p0; \
76647663 poly16_t __ret; \
7664 poly16x8_t __s0 = __p0; \
76657665 __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((poly16x8_t)__s0, __p1); \
76667666 __ret; \
76677667})
......@@ -7669,22 +7669,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
76697669
76707670#ifdef __LITTLE_ENDIAN__
76717671#define vgetq_lane_u8(__p0, __p1) __extension__ ({ \
7672 uint8x16_t __s0 = __p0; \
76737672 uint8_t __ret; \
7673 uint8x16_t __s0 = __p0; \
76747674 __ret = (uint8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
76757675 __ret; \
76767676})
76777677#else
76787678#define vgetq_lane_u8(__p0, __p1) __extension__ ({ \
7679 uint8_t __ret; \
76797680 uint8x16_t __s0 = __p0; \
76807681 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
7681 uint8_t __ret; \
76827682 __ret = (uint8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__rev0, __p1); \
76837683 __ret; \
76847684})
76857685#define __noswap_vgetq_lane_u8(__p0, __p1) __extension__ ({ \
7686 uint8x16_t __s0 = __p0; \
76877686 uint8_t __ret; \
7687 uint8x16_t __s0 = __p0; \
76887688 __ret = (uint8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
76897689 __ret; \
76907690})
......@@ -7692,22 +7692,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
76927692
76937693#ifdef __LITTLE_ENDIAN__
76947694#define vgetq_lane_u32(__p0, __p1) __extension__ ({ \
7695 uint32x4_t __s0 = __p0; \
76967695 uint32_t __ret; \
7696 uint32x4_t __s0 = __p0; \
76977697 __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
76987698 __ret; \
76997699})
77007700#else
77017701#define vgetq_lane_u32(__p0, __p1) __extension__ ({ \
7702 uint32_t __ret; \
77027703 uint32x4_t __s0 = __p0; \
77037704 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
7704 uint32_t __ret; \
77057705 __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__rev0, __p1); \
77067706 __ret; \
77077707})
77087708#define __noswap_vgetq_lane_u32(__p0, __p1) __extension__ ({ \
7709 uint32x4_t __s0 = __p0; \
77107709 uint32_t __ret; \
7710 uint32x4_t __s0 = __p0; \
77117711 __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
77127712 __ret; \
77137713})
......@@ -7715,22 +7715,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
77157715
77167716#ifdef __LITTLE_ENDIAN__
77177717#define vgetq_lane_u64(__p0, __p1) __extension__ ({ \
7718 uint64x2_t __s0 = __p0; \
77197718 uint64_t __ret; \
7719 uint64x2_t __s0 = __p0; \
77207720 __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
77217721 __ret; \
77227722})
77237723#else
77247724#define vgetq_lane_u64(__p0, __p1) __extension__ ({ \
7725 uint64_t __ret; \
77257726 uint64x2_t __s0 = __p0; \
77267727 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
7727 uint64_t __ret; \
77287728 __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__rev0, __p1); \
77297729 __ret; \
77307730})
77317731#define __noswap_vgetq_lane_u64(__p0, __p1) __extension__ ({ \
7732 uint64x2_t __s0 = __p0; \
77337732 uint64_t __ret; \
7733 uint64x2_t __s0 = __p0; \
77347734 __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
77357735 __ret; \
77367736})
......@@ -7738,22 +7738,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
77387738
77397739#ifdef __LITTLE_ENDIAN__
77407740#define vgetq_lane_u16(__p0, __p1) __extension__ ({ \
7741 uint16x8_t __s0 = __p0; \
77427741 uint16_t __ret; \
7742 uint16x8_t __s0 = __p0; \
77437743 __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
77447744 __ret; \
77457745})
77467746#else
77477747#define vgetq_lane_u16(__p0, __p1) __extension__ ({ \
7748 uint16_t __ret; \
77487749 uint16x8_t __s0 = __p0; \
77497750 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
7750 uint16_t __ret; \
77517751 __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__rev0, __p1); \
77527752 __ret; \
77537753})
77547754#define __noswap_vgetq_lane_u16(__p0, __p1) __extension__ ({ \
7755 uint16x8_t __s0 = __p0; \
77567755 uint16_t __ret; \
7756 uint16x8_t __s0 = __p0; \
77577757 __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
77587758 __ret; \
77597759})
......@@ -7761,22 +7761,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
77617761
77627762#ifdef __LITTLE_ENDIAN__
77637763#define vgetq_lane_s8(__p0, __p1) __extension__ ({ \
7764 int8x16_t __s0 = __p0; \
77657764 int8_t __ret; \
7765 int8x16_t __s0 = __p0; \
77667766 __ret = (int8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
77677767 __ret; \
77687768})
77697769#else
77707770#define vgetq_lane_s8(__p0, __p1) __extension__ ({ \
7771 int8_t __ret; \
77717772 int8x16_t __s0 = __p0; \
77727773 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
7773 int8_t __ret; \
77747774 __ret = (int8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__rev0, __p1); \
77757775 __ret; \
77767776})
77777777#define __noswap_vgetq_lane_s8(__p0, __p1) __extension__ ({ \
7778 int8x16_t __s0 = __p0; \
77797778 int8_t __ret; \
7779 int8x16_t __s0 = __p0; \
77807780 __ret = (int8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
77817781 __ret; \
77827782})
......@@ -7784,22 +7784,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
77847784
77857785#ifdef __LITTLE_ENDIAN__
77867786#define vgetq_lane_f32(__p0, __p1) __extension__ ({ \
7787 float32x4_t __s0 = __p0; \
77887787 float32_t __ret; \
7788 float32x4_t __s0 = __p0; \
77897789 __ret = (float32_t) __builtin_neon_vgetq_lane_f32((float32x4_t)__s0, __p1); \
77907790 __ret; \
77917791})
77927792#else
77937793#define vgetq_lane_f32(__p0, __p1) __extension__ ({ \
7794 float32_t __ret; \
77947795 float32x4_t __s0 = __p0; \
77957796 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
7796 float32_t __ret; \
77977797 __ret = (float32_t) __builtin_neon_vgetq_lane_f32((float32x4_t)__rev0, __p1); \
77987798 __ret; \
77997799})
78007800#define __noswap_vgetq_lane_f32(__p0, __p1) __extension__ ({ \
7801 float32x4_t __s0 = __p0; \
78027801 float32_t __ret; \
7802 float32x4_t __s0 = __p0; \
78037803 __ret = (float32_t) __builtin_neon_vgetq_lane_f32((float32x4_t)__s0, __p1); \
78047804 __ret; \
78057805})
......@@ -7807,22 +7807,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
78077807
78087808#ifdef __LITTLE_ENDIAN__
78097809#define vgetq_lane_s32(__p0, __p1) __extension__ ({ \
7810 int32x4_t __s0 = __p0; \
78117810 int32_t __ret; \
7811 int32x4_t __s0 = __p0; \
78127812 __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
78137813 __ret; \
78147814})
78157815#else
78167816#define vgetq_lane_s32(__p0, __p1) __extension__ ({ \
7817 int32_t __ret; \
78177818 int32x4_t __s0 = __p0; \
78187819 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
7819 int32_t __ret; \
78207820 __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__rev0, __p1); \
78217821 __ret; \
78227822})
78237823#define __noswap_vgetq_lane_s32(__p0, __p1) __extension__ ({ \
7824 int32x4_t __s0 = __p0; \
78257824 int32_t __ret; \
7825 int32x4_t __s0 = __p0; \
78267826 __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int32x4_t)__s0, __p1); \
78277827 __ret; \
78287828})
......@@ -7830,22 +7830,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
78307830
78317831#ifdef __LITTLE_ENDIAN__
78327832#define vgetq_lane_s64(__p0, __p1) __extension__ ({ \
7833 int64x2_t __s0 = __p0; \
78347833 int64_t __ret; \
7834 int64x2_t __s0 = __p0; \
78357835 __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
78367836 __ret; \
78377837})
78387838#else
78397839#define vgetq_lane_s64(__p0, __p1) __extension__ ({ \
7840 int64_t __ret; \
78407841 int64x2_t __s0 = __p0; \
78417842 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
7842 int64_t __ret; \
78437843 __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__rev0, __p1); \
78447844 __ret; \
78457845})
78467846#define __noswap_vgetq_lane_s64(__p0, __p1) __extension__ ({ \
7847 int64x2_t __s0 = __p0; \
78487847 int64_t __ret; \
7848 int64x2_t __s0 = __p0; \
78497849 __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int64x2_t)__s0, __p1); \
78507850 __ret; \
78517851})
......@@ -7853,22 +7853,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
78537853
78547854#ifdef __LITTLE_ENDIAN__
78557855#define vgetq_lane_s16(__p0, __p1) __extension__ ({ \
7856 int16x8_t __s0 = __p0; \
78577856 int16_t __ret; \
7857 int16x8_t __s0 = __p0; \
78587858 __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
78597859 __ret; \
78607860})
78617861#else
78627862#define vgetq_lane_s16(__p0, __p1) __extension__ ({ \
7863 int16_t __ret; \
78637864 int16x8_t __s0 = __p0; \
78647865 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
7865 int16_t __ret; \
78667866 __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__rev0, __p1); \
78677867 __ret; \
78687868})
78697869#define __noswap_vgetq_lane_s16(__p0, __p1) __extension__ ({ \
7870 int16x8_t __s0 = __p0; \
78717870 int16_t __ret; \
7871 int16x8_t __s0 = __p0; \
78727872 __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int16x8_t)__s0, __p1); \
78737873 __ret; \
78747874})
......@@ -7876,22 +7876,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
78767876
78777877#ifdef __LITTLE_ENDIAN__
78787878#define vget_lane_u8(__p0, __p1) __extension__ ({ \
7879 uint8x8_t __s0 = __p0; \
78807879 uint8_t __ret; \
7880 uint8x8_t __s0 = __p0; \
78817881 __ret = (uint8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
78827882 __ret; \
78837883})
78847884#else
78857885#define vget_lane_u8(__p0, __p1) __extension__ ({ \
7886 uint8_t __ret; \
78867887 uint8x8_t __s0 = __p0; \
78877888 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
7888 uint8_t __ret; \
78897889 __ret = (uint8_t) __builtin_neon_vget_lane_i8((int8x8_t)__rev0, __p1); \
78907890 __ret; \
78917891})
78927892#define __noswap_vget_lane_u8(__p0, __p1) __extension__ ({ \
7893 uint8x8_t __s0 = __p0; \
78947893 uint8_t __ret; \
7894 uint8x8_t __s0 = __p0; \
78957895 __ret = (uint8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
78967896 __ret; \
78977897})
......@@ -7899,51 +7899,51 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
78997899
79007900#ifdef __LITTLE_ENDIAN__
79017901#define vget_lane_u32(__p0, __p1) __extension__ ({ \
7902 uint32x2_t __s0 = __p0; \
79037902 uint32_t __ret; \
7903 uint32x2_t __s0 = __p0; \
79047904 __ret = (uint32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
79057905 __ret; \
79067906})
79077907#else
79087908#define vget_lane_u32(__p0, __p1) __extension__ ({ \
7909 uint32_t __ret; \
79097910 uint32x2_t __s0 = __p0; \
79107911 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
7911 uint32_t __ret; \
79127912 __ret = (uint32_t) __builtin_neon_vget_lane_i32((int32x2_t)__rev0, __p1); \
79137913 __ret; \
79147914})
79157915#define __noswap_vget_lane_u32(__p0, __p1) __extension__ ({ \
7916 uint32x2_t __s0 = __p0; \
79177916 uint32_t __ret; \
7917 uint32x2_t __s0 = __p0; \
79187918 __ret = (uint32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
79197919 __ret; \
79207920})
79217921#endif
79227922
79237923#define vget_lane_u64(__p0, __p1) __extension__ ({ \
7924 uint64x1_t __s0 = __p0; \
79257924 uint64_t __ret; \
7925 uint64x1_t __s0 = __p0; \
79267926 __ret = (uint64_t) __builtin_neon_vget_lane_i64((int64x1_t)__s0, __p1); \
79277927 __ret; \
79287928})
79297929#ifdef __LITTLE_ENDIAN__
79307930#define vget_lane_u16(__p0, __p1) __extension__ ({ \
7931 uint16x4_t __s0 = __p0; \
79327931 uint16_t __ret; \
7932 uint16x4_t __s0 = __p0; \
79337933 __ret = (uint16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
79347934 __ret; \
79357935})
79367936#else
79377937#define vget_lane_u16(__p0, __p1) __extension__ ({ \
7938 uint16_t __ret; \
79387939 uint16x4_t __s0 = __p0; \
79397940 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
7940 uint16_t __ret; \
79417941 __ret = (uint16_t) __builtin_neon_vget_lane_i16((int16x4_t)__rev0, __p1); \
79427942 __ret; \
79437943})
79447944#define __noswap_vget_lane_u16(__p0, __p1) __extension__ ({ \
7945 uint16x4_t __s0 = __p0; \
79467945 uint16_t __ret; \
7946 uint16x4_t __s0 = __p0; \
79477947 __ret = (uint16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
79487948 __ret; \
79497949})
......@@ -7951,22 +7951,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
79517951
79527952#ifdef __LITTLE_ENDIAN__
79537953#define vget_lane_s8(__p0, __p1) __extension__ ({ \
7954 int8x8_t __s0 = __p0; \
79557954 int8_t __ret; \
7955 int8x8_t __s0 = __p0; \
79567956 __ret = (int8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
79577957 __ret; \
79587958})
79597959#else
79607960#define vget_lane_s8(__p0, __p1) __extension__ ({ \
7961 int8_t __ret; \
79617962 int8x8_t __s0 = __p0; \
79627963 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
7963 int8_t __ret; \
79647964 __ret = (int8_t) __builtin_neon_vget_lane_i8((int8x8_t)__rev0, __p1); \
79657965 __ret; \
79667966})
79677967#define __noswap_vget_lane_s8(__p0, __p1) __extension__ ({ \
7968 int8x8_t __s0 = __p0; \
79697968 int8_t __ret; \
7969 int8x8_t __s0 = __p0; \
79707970 __ret = (int8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
79717971 __ret; \
79727972})
......@@ -7974,22 +7974,22 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
79747974
79757975#ifdef __LITTLE_ENDIAN__
79767976#define vget_lane_f32(__p0, __p1) __extension__ ({ \
7977 float32x2_t __s0 = __p0; \
79787977 float32_t __ret; \
7978 float32x2_t __s0 = __p0; \
79797979 __ret = (float32_t) __builtin_neon_vget_lane_f32((float32x2_t)__s0, __p1); \
79807980 __ret; \
79817981})
79827982#else
79837983#define vget_lane_f32(__p0, __p1) __extension__ ({ \
7984 float32_t __ret; \
79847985 float32x2_t __s0 = __p0; \
79857986 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
7986 float32_t __ret; \
79877987 __ret = (float32_t) __builtin_neon_vget_lane_f32((float32x2_t)__rev0, __p1); \
79887988 __ret; \
79897989})
79907990#define __noswap_vget_lane_f32(__p0, __p1) __extension__ ({ \
7991 float32x2_t __s0 = __p0; \
79927991 float32_t __ret; \
7992 float32x2_t __s0 = __p0; \
79937993 __ret = (float32_t) __builtin_neon_vget_lane_f32((float32x2_t)__s0, __p1); \
79947994 __ret; \
79957995})
......@@ -7997,51 +7997,51 @@ __ai int16x4_t __noswap_vget_high_s16(int16x8_t __p0) {
79977997
79987998#ifdef __LITTLE_ENDIAN__
79997999#define vget_lane_s32(__p0, __p1) __extension__ ({ \
8000 int32x2_t __s0 = __p0; \
80018000 int32_t __ret; \
8001 int32x2_t __s0 = __p0; \
80028002 __ret = (int32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
80038003 __ret; \
80048004})
80058005#else
80068006#define vget_lane_s32(__p0, __p1) __extension__ ({ \
8007 int32_t __ret; \
80078008 int32x2_t __s0 = __p0; \
80088009 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
8009 int32_t __ret; \
80108010 __ret = (int32_t) __builtin_neon_vget_lane_i32((int32x2_t)__rev0, __p1); \
80118011 __ret; \
80128012})
80138013#define __noswap_vget_lane_s32(__p0, __p1) __extension__ ({ \
8014 int32x2_t __s0 = __p0; \
80158014 int32_t __ret; \
8015 int32x2_t __s0 = __p0; \
80168016 __ret = (int32_t) __builtin_neon_vget_lane_i32((int32x2_t)__s0, __p1); \
80178017 __ret; \
80188018})
80198019#endif
80208020
80218021#define vget_lane_s64(__p0, __p1) __extension__ ({ \
8022 int64x1_t __s0 = __p0; \
80238022 int64_t __ret; \
8023 int64x1_t __s0 = __p0; \
80248024 __ret = (int64_t) __builtin_neon_vget_lane_i64((int64x1_t)__s0, __p1); \
80258025 __ret; \
80268026})
80278027#ifdef __LITTLE_ENDIAN__
80288028#define vget_lane_s16(__p0, __p1) __extension__ ({ \
8029 int16x4_t __s0 = __p0; \
80308029 int16_t __ret; \
8030 int16x4_t __s0 = __p0; \
80318031 __ret = (int16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
80328032 __ret; \
80338033})
80348034#else
80358035#define vget_lane_s16(__p0, __p1) __extension__ ({ \
8036 int16_t __ret; \
80368037 int16x4_t __s0 = __p0; \
80378038 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
8038 int16_t __ret; \
80398039 __ret = (int16_t) __builtin_neon_vget_lane_i16((int16x4_t)__rev0, __p1); \
80408040 __ret; \
80418041})
80428042#define __noswap_vget_lane_s16(__p0, __p1) __extension__ ({ \
8043 int16x4_t __s0 = __p0; \
80448043 int16_t __ret; \
8044 int16x4_t __s0 = __p0; \
80458045 __ret = (int16_t) __builtin_neon_vget_lane_i16((int16x4_t)__s0, __p1); \
80468046 __ret; \
80478047})
......@@ -8055,8 +8055,8 @@ __ai poly8x8_t vget_low_p8(poly8x16_t __p0) {
80558055}
80568056#else
80578057__ai poly8x8_t vget_low_p8(poly8x16_t __p0) {
8058 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
80598058 poly8x8_t __ret;
8059 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
80608060 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3, 4, 5, 6, 7);
80618061 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
80628062 return __ret;
......@@ -8071,8 +8071,8 @@ __ai poly16x4_t vget_low_p16(poly16x8_t __p0) {
80718071}
80728072#else
80738073__ai poly16x4_t vget_low_p16(poly16x8_t __p0) {
8074 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
80758074 poly16x4_t __ret;
8075 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
80768076 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
80778077 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
80788078 return __ret;
......@@ -8087,8 +8087,8 @@ __ai uint8x8_t vget_low_u8(uint8x16_t __p0) {
80878087}
80888088#else
80898089__ai uint8x8_t vget_low_u8(uint8x16_t __p0) {
8090 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
80918090 uint8x8_t __ret;
8091 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
80928092 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3, 4, 5, 6, 7);
80938093 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
80948094 return __ret;
......@@ -8103,8 +8103,8 @@ __ai uint32x2_t vget_low_u32(uint32x4_t __p0) {
81038103}
81048104#else
81058105__ai uint32x2_t vget_low_u32(uint32x4_t __p0) {
8106 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
81078106 uint32x2_t __ret;
8107 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
81088108 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1);
81098109 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
81108110 return __ret;
......@@ -8119,8 +8119,8 @@ __ai uint64x1_t vget_low_u64(uint64x2_t __p0) {
81198119}
81208120#else
81218121__ai uint64x1_t vget_low_u64(uint64x2_t __p0) {
8122 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
81238122 uint64x1_t __ret;
8123 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
81248124 __ret = __builtin_shufflevector(__rev0, __rev0, 0);
81258125 return __ret;
81268126}
......@@ -8134,8 +8134,8 @@ __ai uint16x4_t vget_low_u16(uint16x8_t __p0) {
81348134}
81358135#else
81368136__ai uint16x4_t vget_low_u16(uint16x8_t __p0) {
8137 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
81388137 uint16x4_t __ret;
8138 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
81398139 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
81408140 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
81418141 return __ret;
......@@ -8150,8 +8150,8 @@ __ai int8x8_t vget_low_s8(int8x16_t __p0) {
81508150}
81518151#else
81528152__ai int8x8_t vget_low_s8(int8x16_t __p0) {
8153 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
81548153 int8x8_t __ret;
8154 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
81558155 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3, 4, 5, 6, 7);
81568156 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
81578157 return __ret;
......@@ -8166,8 +8166,8 @@ __ai float32x2_t vget_low_f32(float32x4_t __p0) {
81668166}
81678167#else
81688168__ai float32x2_t vget_low_f32(float32x4_t __p0) {
8169 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
81708169 float32x2_t __ret;
8170 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
81718171 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1);
81728172 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
81738173 return __ret;
......@@ -8182,8 +8182,8 @@ __ai float16x4_t vget_low_f16(float16x8_t __p0) {
81828182}
81838183#else
81848184__ai float16x4_t vget_low_f16(float16x8_t __p0) {
8185 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
81868185 float16x4_t __ret;
8186 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
81878187 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
81888188 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
81898189 return __ret;
......@@ -8198,8 +8198,8 @@ __ai int32x2_t vget_low_s32(int32x4_t __p0) {
81988198}
81998199#else
82008200__ai int32x2_t vget_low_s32(int32x4_t __p0) {
8201 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
82028201 int32x2_t __ret;
8202 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
82038203 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1);
82048204 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
82058205 return __ret;
......@@ -8214,8 +8214,8 @@ __ai int64x1_t vget_low_s64(int64x2_t __p0) {
82148214}
82158215#else
82168216__ai int64x1_t vget_low_s64(int64x2_t __p0) {
8217 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
82188217 int64x1_t __ret;
8218 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
82198219 __ret = __builtin_shufflevector(__rev0, __rev0, 0);
82208220 return __ret;
82218221}
......@@ -8229,8 +8229,8 @@ __ai int16x4_t vget_low_s16(int16x8_t __p0) {
82298229}
82308230#else
82318231__ai int16x4_t vget_low_s16(int16x8_t __p0) {
8232 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
82338232 int16x4_t __ret;
8233 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
82348234 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
82358235 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
82368236 return __ret;
......@@ -8245,9 +8245,9 @@ __ai uint8x16_t vhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
82458245}
82468246#else
82478247__ai uint8x16_t vhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
8248 uint8x16_t __ret;
82488249 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
82498250 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
8250 uint8x16_t __ret;
82518251 __ret = (uint8x16_t) __builtin_neon_vhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
82528252 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
82538253 return __ret;
......@@ -8262,9 +8262,9 @@ __ai uint32x4_t vhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
82628262}
82638263#else
82648264__ai uint32x4_t vhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
8265 uint32x4_t __ret;
82658266 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
82668267 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8267 uint32x4_t __ret;
82688268 __ret = (uint32x4_t) __builtin_neon_vhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
82698269 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
82708270 return __ret;
......@@ -8279,9 +8279,9 @@ __ai uint16x8_t vhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
82798279}
82808280#else
82818281__ai uint16x8_t vhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
8282 uint16x8_t __ret;
82828283 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
82838284 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8284 uint16x8_t __ret;
82858285 __ret = (uint16x8_t) __builtin_neon_vhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
82868286 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
82878287 return __ret;
......@@ -8296,9 +8296,9 @@ __ai int8x16_t vhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
82968296}
82978297#else
82988298__ai int8x16_t vhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
8299 int8x16_t __ret;
82998300 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
83008301 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
8301 int8x16_t __ret;
83028302 __ret = (int8x16_t) __builtin_neon_vhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
83038303 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
83048304 return __ret;
......@@ -8313,9 +8313,9 @@ __ai int32x4_t vhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
83138313}
83148314#else
83158315__ai int32x4_t vhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
8316 int32x4_t __ret;
83168317 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
83178318 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8318 int32x4_t __ret;
83198319 __ret = (int32x4_t) __builtin_neon_vhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
83208320 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
83218321 return __ret;
......@@ -8330,9 +8330,9 @@ __ai int16x8_t vhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
83308330}
83318331#else
83328332__ai int16x8_t vhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
8333 int16x8_t __ret;
83338334 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
83348335 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8335 int16x8_t __ret;
83368336 __ret = (int16x8_t) __builtin_neon_vhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
83378337 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
83388338 return __ret;
......@@ -8347,9 +8347,9 @@ __ai uint8x8_t vhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
83478347}
83488348#else
83498349__ai uint8x8_t vhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
8350 uint8x8_t __ret;
83508351 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
83518352 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8352 uint8x8_t __ret;
83538353 __ret = (uint8x8_t) __builtin_neon_vhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
83548354 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
83558355 return __ret;
......@@ -8364,9 +8364,9 @@ __ai uint32x2_t vhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
83648364}
83658365#else
83668366__ai uint32x2_t vhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
8367 uint32x2_t __ret;
83678368 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
83688369 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
8369 uint32x2_t __ret;
83708370 __ret = (uint32x2_t) __builtin_neon_vhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
83718371 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
83728372 return __ret;
......@@ -8381,9 +8381,9 @@ __ai uint16x4_t vhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
83818381}
83828382#else
83838383__ai uint16x4_t vhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
8384 uint16x4_t __ret;
83848385 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
83858386 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8386 uint16x4_t __ret;
83878387 __ret = (uint16x4_t) __builtin_neon_vhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
83888388 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
83898389 return __ret;
......@@ -8398,9 +8398,9 @@ __ai int8x8_t vhadd_s8(int8x8_t __p0, int8x8_t __p1) {
83988398}
83998399#else
84008400__ai int8x8_t vhadd_s8(int8x8_t __p0, int8x8_t __p1) {
8401 int8x8_t __ret;
84018402 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
84028403 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8403 int8x8_t __ret;
84048404 __ret = (int8x8_t) __builtin_neon_vhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
84058405 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
84068406 return __ret;
......@@ -8415,9 +8415,9 @@ __ai int32x2_t vhadd_s32(int32x2_t __p0, int32x2_t __p1) {
84158415}
84168416#else
84178417__ai int32x2_t vhadd_s32(int32x2_t __p0, int32x2_t __p1) {
8418 int32x2_t __ret;
84188419 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
84198420 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
8420 int32x2_t __ret;
84218421 __ret = (int32x2_t) __builtin_neon_vhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
84228422 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
84238423 return __ret;
......@@ -8432,9 +8432,9 @@ __ai int16x4_t vhadd_s16(int16x4_t __p0, int16x4_t __p1) {
84328432}
84338433#else
84348434__ai int16x4_t vhadd_s16(int16x4_t __p0, int16x4_t __p1) {
8435 int16x4_t __ret;
84358436 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
84368437 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8437 int16x4_t __ret;
84388438 __ret = (int16x4_t) __builtin_neon_vhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
84398439 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
84408440 return __ret;
......@@ -8449,9 +8449,9 @@ __ai uint8x16_t vhsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
84498449}
84508450#else
84518451__ai uint8x16_t vhsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
8452 uint8x16_t __ret;
84528453 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
84538454 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
8454 uint8x16_t __ret;
84558455 __ret = (uint8x16_t) __builtin_neon_vhsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
84568456 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
84578457 return __ret;
......@@ -8466,9 +8466,9 @@ __ai uint32x4_t vhsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
84668466}
84678467#else
84688468__ai uint32x4_t vhsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
8469 uint32x4_t __ret;
84698470 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
84708471 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8471 uint32x4_t __ret;
84728472 __ret = (uint32x4_t) __builtin_neon_vhsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
84738473 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
84748474 return __ret;
......@@ -8483,9 +8483,9 @@ __ai uint16x8_t vhsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
84838483}
84848484#else
84858485__ai uint16x8_t vhsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
8486 uint16x8_t __ret;
84868487 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
84878488 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8488 uint16x8_t __ret;
84898489 __ret = (uint16x8_t) __builtin_neon_vhsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
84908490 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
84918491 return __ret;
......@@ -8500,9 +8500,9 @@ __ai int8x16_t vhsubq_s8(int8x16_t __p0, int8x16_t __p1) {
85008500}
85018501#else
85028502__ai int8x16_t vhsubq_s8(int8x16_t __p0, int8x16_t __p1) {
8503 int8x16_t __ret;
85038504 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
85048505 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
8505 int8x16_t __ret;
85068506 __ret = (int8x16_t) __builtin_neon_vhsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
85078507 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
85088508 return __ret;
......@@ -8517,9 +8517,9 @@ __ai int32x4_t vhsubq_s32(int32x4_t __p0, int32x4_t __p1) {
85178517}
85188518#else
85198519__ai int32x4_t vhsubq_s32(int32x4_t __p0, int32x4_t __p1) {
8520 int32x4_t __ret;
85208521 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
85218522 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8522 int32x4_t __ret;
85238523 __ret = (int32x4_t) __builtin_neon_vhsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
85248524 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
85258525 return __ret;
......@@ -8534,9 +8534,9 @@ __ai int16x8_t vhsubq_s16(int16x8_t __p0, int16x8_t __p1) {
85348534}
85358535#else
85368536__ai int16x8_t vhsubq_s16(int16x8_t __p0, int16x8_t __p1) {
8537 int16x8_t __ret;
85378538 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
85388539 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8539 int16x8_t __ret;
85408540 __ret = (int16x8_t) __builtin_neon_vhsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
85418541 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
85428542 return __ret;
......@@ -8551,9 +8551,9 @@ __ai uint8x8_t vhsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
85518551}
85528552#else
85538553__ai uint8x8_t vhsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
8554 uint8x8_t __ret;
85548555 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
85558556 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8556 uint8x8_t __ret;
85578557 __ret = (uint8x8_t) __builtin_neon_vhsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
85588558 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
85598559 return __ret;
......@@ -8568,9 +8568,9 @@ __ai uint32x2_t vhsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
85688568}
85698569#else
85708570__ai uint32x2_t vhsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
8571 uint32x2_t __ret;
85718572 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
85728573 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
8573 uint32x2_t __ret;
85748574 __ret = (uint32x2_t) __builtin_neon_vhsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
85758575 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
85768576 return __ret;
......@@ -8585,9 +8585,9 @@ __ai uint16x4_t vhsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
85858585}
85868586#else
85878587__ai uint16x4_t vhsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
8588 uint16x4_t __ret;
85888589 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
85898590 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8590 uint16x4_t __ret;
85918591 __ret = (uint16x4_t) __builtin_neon_vhsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
85928592 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
85938593 return __ret;
......@@ -8602,9 +8602,9 @@ __ai int8x8_t vhsub_s8(int8x8_t __p0, int8x8_t __p1) {
86028602}
86038603#else
86048604__ai int8x8_t vhsub_s8(int8x8_t __p0, int8x8_t __p1) {
8605 int8x8_t __ret;
86058606 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
86068607 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
8607 int8x8_t __ret;
86088608 __ret = (int8x8_t) __builtin_neon_vhsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
86098609 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
86108610 return __ret;
......@@ -8619,9 +8619,9 @@ __ai int32x2_t vhsub_s32(int32x2_t __p0, int32x2_t __p1) {
86198619}
86208620#else
86218621__ai int32x2_t vhsub_s32(int32x2_t __p0, int32x2_t __p1) {
8622 int32x2_t __ret;
86228623 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
86238624 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
8624 int32x2_t __ret;
86258625 __ret = (int32x2_t) __builtin_neon_vhsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
86268626 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
86278627 return __ret;
......@@ -8636,9 +8636,9 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
86368636}
86378637#else
86388638__ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
8639 int16x4_t __ret;
86398640 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
86408641 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8641 int16x4_t __ret;
86428642 __ret = (int16x4_t) __builtin_neon_vhsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
86438643 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
86448644 return __ret;
......@@ -9267,16 +9267,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
92679267
92689268#ifdef __LITTLE_ENDIAN__
92699269#define vld1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
9270 poly8x8_t __s1 = __p1; \
92719270 poly8x8_t __ret; \
9271 poly8x8_t __s1 = __p1; \
92729272 __ret = (poly8x8_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 4); \
92739273 __ret; \
92749274})
92759275#else
92769276#define vld1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
9277 poly8x8_t __ret; \
92779278 poly8x8_t __s1 = __p1; \
92789279 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
9279 poly8x8_t __ret; \
92809280 __ret = (poly8x8_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 4); \
92819281 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
92829282 __ret; \
......@@ -9285,16 +9285,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
92859285
92869286#ifdef __LITTLE_ENDIAN__
92879287#define vld1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9288 poly16x4_t __s1 = __p1; \
92899288 poly16x4_t __ret; \
9289 poly16x4_t __s1 = __p1; \
92909290 __ret = (poly16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 5); \
92919291 __ret; \
92929292})
92939293#else
92949294#define vld1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9295 poly16x4_t __ret; \
92959296 poly16x4_t __s1 = __p1; \
92969297 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
9297 poly16x4_t __ret; \
92989298 __ret = (poly16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 5); \
92999299 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
93009300 __ret; \
......@@ -9303,16 +9303,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
93039303
93049304#ifdef __LITTLE_ENDIAN__
93059305#define vld1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
9306 poly8x16_t __s1 = __p1; \
93079306 poly8x16_t __ret; \
9307 poly8x16_t __s1 = __p1; \
93089308 __ret = (poly8x16_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 36); \
93099309 __ret; \
93109310})
93119311#else
93129312#define vld1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
9313 poly8x16_t __ret; \
93139314 poly8x16_t __s1 = __p1; \
93149315 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
9315 poly8x16_t __ret; \
93169316 __ret = (poly8x16_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 36); \
93179317 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
93189318 __ret; \
......@@ -9321,16 +9321,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
93219321
93229322#ifdef __LITTLE_ENDIAN__
93239323#define vld1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9324 poly16x8_t __s1 = __p1; \
93259324 poly16x8_t __ret; \
9325 poly16x8_t __s1 = __p1; \
93269326 __ret = (poly16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 37); \
93279327 __ret; \
93289328})
93299329#else
93309330#define vld1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9331 poly16x8_t __ret; \
93319332 poly16x8_t __s1 = __p1; \
93329333 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
9333 poly16x8_t __ret; \
93349334 __ret = (poly16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 37); \
93359335 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
93369336 __ret; \
......@@ -9339,16 +9339,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
93399339
93409340#ifdef __LITTLE_ENDIAN__
93419341#define vld1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
9342 uint8x16_t __s1 = __p1; \
93439342 uint8x16_t __ret; \
9343 uint8x16_t __s1 = __p1; \
93449344 __ret = (uint8x16_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 48); \
93459345 __ret; \
93469346})
93479347#else
93489348#define vld1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
9349 uint8x16_t __ret; \
93499350 uint8x16_t __s1 = __p1; \
93509351 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
9351 uint8x16_t __ret; \
93529352 __ret = (uint8x16_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 48); \
93539353 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
93549354 __ret; \
......@@ -9357,16 +9357,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
93579357
93589358#ifdef __LITTLE_ENDIAN__
93599359#define vld1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9360 uint32x4_t __s1 = __p1; \
93619360 uint32x4_t __ret; \
9361 uint32x4_t __s1 = __p1; \
93629362 __ret = (uint32x4_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 50); \
93639363 __ret; \
93649364})
93659365#else
93669366#define vld1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9367 uint32x4_t __ret; \
93679368 uint32x4_t __s1 = __p1; \
93689369 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
9369 uint32x4_t __ret; \
93709370 __ret = (uint32x4_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 50); \
93719371 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
93729372 __ret; \
......@@ -9375,16 +9375,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
93759375
93769376#ifdef __LITTLE_ENDIAN__
93779377#define vld1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
9378 uint64x2_t __s1 = __p1; \
93799378 uint64x2_t __ret; \
9379 uint64x2_t __s1 = __p1; \
93809380 __ret = (uint64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 51); \
93819381 __ret; \
93829382})
93839383#else
93849384#define vld1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
9385 uint64x2_t __ret; \
93859386 uint64x2_t __s1 = __p1; \
93869387 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
9387 uint64x2_t __ret; \
93889388 __ret = (uint64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 51); \
93899389 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
93909390 __ret; \
......@@ -9393,16 +9393,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
93939393
93949394#ifdef __LITTLE_ENDIAN__
93959395#define vld1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9396 uint16x8_t __s1 = __p1; \
93979396 uint16x8_t __ret; \
9397 uint16x8_t __s1 = __p1; \
93989398 __ret = (uint16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 49); \
93999399 __ret; \
94009400})
94019401#else
94029402#define vld1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9403 uint16x8_t __ret; \
94039404 uint16x8_t __s1 = __p1; \
94049405 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
9405 uint16x8_t __ret; \
94069406 __ret = (uint16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 49); \
94079407 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
94089408 __ret; \
......@@ -9411,16 +9411,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
94119411
94129412#ifdef __LITTLE_ENDIAN__
94139413#define vld1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
9414 int8x16_t __s1 = __p1; \
94159414 int8x16_t __ret; \
9415 int8x16_t __s1 = __p1; \
94169416 __ret = (int8x16_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 32); \
94179417 __ret; \
94189418})
94199419#else
94209420#define vld1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
9421 int8x16_t __ret; \
94219422 int8x16_t __s1 = __p1; \
94229423 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
9423 int8x16_t __ret; \
94249424 __ret = (int8x16_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 32); \
94259425 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
94269426 __ret; \
......@@ -9429,16 +9429,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
94299429
94309430#ifdef __LITTLE_ENDIAN__
94319431#define vld1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9432 float32x4_t __s1 = __p1; \
94339432 float32x4_t __ret; \
9433 float32x4_t __s1 = __p1; \
94349434 __ret = (float32x4_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 41); \
94359435 __ret; \
94369436})
94379437#else
94389438#define vld1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9439 float32x4_t __ret; \
94399440 float32x4_t __s1 = __p1; \
94409441 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
9441 float32x4_t __ret; \
94429442 __ret = (float32x4_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 41); \
94439443 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
94449444 __ret; \
......@@ -9447,16 +9447,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
94479447
94489448#ifdef __LITTLE_ENDIAN__
94499449#define vld1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9450 int32x4_t __s1 = __p1; \
94519450 int32x4_t __ret; \
9451 int32x4_t __s1 = __p1; \
94529452 __ret = (int32x4_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 34); \
94539453 __ret; \
94549454})
94559455#else
94569456#define vld1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9457 int32x4_t __ret; \
94579458 int32x4_t __s1 = __p1; \
94589459 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
9459 int32x4_t __ret; \
94609460 __ret = (int32x4_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 34); \
94619461 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
94629462 __ret; \
......@@ -9465,16 +9465,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
94659465
94669466#ifdef __LITTLE_ENDIAN__
94679467#define vld1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
9468 int64x2_t __s1 = __p1; \
94699468 int64x2_t __ret; \
9469 int64x2_t __s1 = __p1; \
94709470 __ret = (int64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 35); \
94719471 __ret; \
94729472})
94739473#else
94749474#define vld1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
9475 int64x2_t __ret; \
94759476 int64x2_t __s1 = __p1; \
94769477 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
9477 int64x2_t __ret; \
94789478 __ret = (int64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 35); \
94799479 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
94809480 __ret; \
......@@ -9483,16 +9483,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
94839483
94849484#ifdef __LITTLE_ENDIAN__
94859485#define vld1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9486 int16x8_t __s1 = __p1; \
94879486 int16x8_t __ret; \
9487 int16x8_t __s1 = __p1; \
94889488 __ret = (int16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 33); \
94899489 __ret; \
94909490})
94919491#else
94929492#define vld1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9493 int16x8_t __ret; \
94939494 int16x8_t __s1 = __p1; \
94949495 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
9495 int16x8_t __ret; \
94969496 __ret = (int16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 33); \
94979497 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
94989498 __ret; \
......@@ -9501,16 +9501,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
95019501
95029502#ifdef __LITTLE_ENDIAN__
95039503#define vld1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
9504 uint8x8_t __s1 = __p1; \
95059504 uint8x8_t __ret; \
9505 uint8x8_t __s1 = __p1; \
95069506 __ret = (uint8x8_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 16); \
95079507 __ret; \
95089508})
95099509#else
95109510#define vld1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
9511 uint8x8_t __ret; \
95119512 uint8x8_t __s1 = __p1; \
95129513 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
9513 uint8x8_t __ret; \
95149514 __ret = (uint8x8_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 16); \
95159515 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
95169516 __ret; \
......@@ -9519,16 +9519,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
95199519
95209520#ifdef __LITTLE_ENDIAN__
95219521#define vld1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9522 uint32x2_t __s1 = __p1; \
95239522 uint32x2_t __ret; \
9523 uint32x2_t __s1 = __p1; \
95249524 __ret = (uint32x2_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 18); \
95259525 __ret; \
95269526})
95279527#else
95289528#define vld1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9529 uint32x2_t __ret; \
95299530 uint32x2_t __s1 = __p1; \
95309531 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
9531 uint32x2_t __ret; \
95329532 __ret = (uint32x2_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 18); \
95339533 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
95349534 __ret; \
......@@ -9536,23 +9536,23 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
95369536#endif
95379537
95389538#define vld1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
9539 uint64x1_t __s1 = __p1; \
95409539 uint64x1_t __ret; \
9540 uint64x1_t __s1 = __p1; \
95419541 __ret = (uint64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 19); \
95429542 __ret; \
95439543})
95449544#ifdef __LITTLE_ENDIAN__
95459545#define vld1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9546 uint16x4_t __s1 = __p1; \
95479546 uint16x4_t __ret; \
9547 uint16x4_t __s1 = __p1; \
95489548 __ret = (uint16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 17); \
95499549 __ret; \
95509550})
95519551#else
95529552#define vld1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9553 uint16x4_t __ret; \
95539554 uint16x4_t __s1 = __p1; \
95549555 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
9555 uint16x4_t __ret; \
95569556 __ret = (uint16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 17); \
95579557 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
95589558 __ret; \
......@@ -9561,16 +9561,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
95619561
95629562#ifdef __LITTLE_ENDIAN__
95639563#define vld1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
9564 int8x8_t __s1 = __p1; \
95659564 int8x8_t __ret; \
9565 int8x8_t __s1 = __p1; \
95669566 __ret = (int8x8_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 0); \
95679567 __ret; \
95689568})
95699569#else
95709570#define vld1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
9571 int8x8_t __ret; \
95719572 int8x8_t __s1 = __p1; \
95729573 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
9573 int8x8_t __ret; \
95749574 __ret = (int8x8_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 0); \
95759575 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
95769576 __ret; \
......@@ -9579,16 +9579,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
95799579
95809580#ifdef __LITTLE_ENDIAN__
95819581#define vld1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9582 float32x2_t __s1 = __p1; \
95839582 float32x2_t __ret; \
9583 float32x2_t __s1 = __p1; \
95849584 __ret = (float32x2_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 9); \
95859585 __ret; \
95869586})
95879587#else
95889588#define vld1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9589 float32x2_t __ret; \
95899590 float32x2_t __s1 = __p1; \
95909591 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
9591 float32x2_t __ret; \
95929592 __ret = (float32x2_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 9); \
95939593 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
95949594 __ret; \
......@@ -9597,16 +9597,16 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
95979597
95989598#ifdef __LITTLE_ENDIAN__
95999599#define vld1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9600 int32x2_t __s1 = __p1; \
96019600 int32x2_t __ret; \
9601 int32x2_t __s1 = __p1; \
96029602 __ret = (int32x2_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 2); \
96039603 __ret; \
96049604})
96059605#else
96069606#define vld1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9607 int32x2_t __ret; \
96079608 int32x2_t __s1 = __p1; \
96089609 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
9609 int32x2_t __ret; \
96109610 __ret = (int32x2_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 2); \
96119611 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
96129612 __ret; \
......@@ -9614,23 +9614,23 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
96149614#endif
96159615
96169616#define vld1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
9617 int64x1_t __s1 = __p1; \
96189617 int64x1_t __ret; \
9618 int64x1_t __s1 = __p1; \
96199619 __ret = (int64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 3); \
96209620 __ret; \
96219621})
96229622#ifdef __LITTLE_ENDIAN__
96239623#define vld1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9624 int16x4_t __s1 = __p1; \
96259624 int16x4_t __ret; \
9625 int16x4_t __s1 = __p1; \
96269626 __ret = (int16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 1); \
96279627 __ret; \
96289628})
96299629#else
96309630#define vld1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9631 int16x4_t __ret; \
96319632 int16x4_t __s1 = __p1; \
96329633 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
9633 int16x4_t __ret; \
96349634 __ret = (int16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 1); \
96359635 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
96369636 __ret; \
......@@ -11415,18 +11415,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1141511415
1141611416#ifdef __LITTLE_ENDIAN__
1141711417#define vld2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
11418 poly8x8x2_t __s1 = __p1; \
1141911418 poly8x8x2_t __ret; \
11419 poly8x8x2_t __s1 = __p1; \
1142011420 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 4); \
1142111421 __ret; \
1142211422})
1142311423#else
1142411424#define vld2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
11425 poly8x8x2_t __ret; \
1142511426 poly8x8x2_t __s1 = __p1; \
1142611427 poly8x8x2_t __rev1; \
1142711428 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1142811429 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
11429 poly8x8x2_t __ret; \
1143011430 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 4); \
1143111431 \
1143211432 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -11437,18 +11437,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1143711437
1143811438#ifdef __LITTLE_ENDIAN__
1143911439#define vld2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11440 poly16x4x2_t __s1 = __p1; \
1144111440 poly16x4x2_t __ret; \
11441 poly16x4x2_t __s1 = __p1; \
1144211442 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 5); \
1144311443 __ret; \
1144411444})
1144511445#else
1144611446#define vld2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11447 poly16x4x2_t __ret; \
1144711448 poly16x4x2_t __s1 = __p1; \
1144811449 poly16x4x2_t __rev1; \
1144911450 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1145011451 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
11451 poly16x4x2_t __ret; \
1145211452 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 5); \
1145311453 \
1145411454 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -11459,18 +11459,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1145911459
1146011460#ifdef __LITTLE_ENDIAN__
1146111461#define vld2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11462 poly16x8x2_t __s1 = __p1; \
1146311462 poly16x8x2_t __ret; \
11463 poly16x8x2_t __s1 = __p1; \
1146411464 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 37); \
1146511465 __ret; \
1146611466})
1146711467#else
1146811468#define vld2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11469 poly16x8x2_t __ret; \
1146911470 poly16x8x2_t __s1 = __p1; \
1147011471 poly16x8x2_t __rev1; \
1147111472 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1147211473 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
11473 poly16x8x2_t __ret; \
1147411474 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 37); \
1147511475 \
1147611476 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -11481,18 +11481,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1148111481
1148211482#ifdef __LITTLE_ENDIAN__
1148311483#define vld2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11484 uint32x4x2_t __s1 = __p1; \
1148511484 uint32x4x2_t __ret; \
11485 uint32x4x2_t __s1 = __p1; \
1148611486 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 50); \
1148711487 __ret; \
1148811488})
1148911489#else
1149011490#define vld2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11491 uint32x4x2_t __ret; \
1149111492 uint32x4x2_t __s1 = __p1; \
1149211493 uint32x4x2_t __rev1; \
1149311494 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1149411495 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
11495 uint32x4x2_t __ret; \
1149611496 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 50); \
1149711497 \
1149811498 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -11503,18 +11503,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1150311503
1150411504#ifdef __LITTLE_ENDIAN__
1150511505#define vld2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11506 uint16x8x2_t __s1 = __p1; \
1150711506 uint16x8x2_t __ret; \
11507 uint16x8x2_t __s1 = __p1; \
1150811508 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 49); \
1150911509 __ret; \
1151011510})
1151111511#else
1151211512#define vld2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11513 uint16x8x2_t __ret; \
1151311514 uint16x8x2_t __s1 = __p1; \
1151411515 uint16x8x2_t __rev1; \
1151511516 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1151611517 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
11517 uint16x8x2_t __ret; \
1151811518 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 49); \
1151911519 \
1152011520 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -11525,18 +11525,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1152511525
1152611526#ifdef __LITTLE_ENDIAN__
1152711527#define vld2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11528 float32x4x2_t __s1 = __p1; \
1152911528 float32x4x2_t __ret; \
11529 float32x4x2_t __s1 = __p1; \
1153011530 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 41); \
1153111531 __ret; \
1153211532})
1153311533#else
1153411534#define vld2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11535 float32x4x2_t __ret; \
1153511536 float32x4x2_t __s1 = __p1; \
1153611537 float32x4x2_t __rev1; \
1153711538 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1153811539 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
11539 float32x4x2_t __ret; \
1154011540 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 41); \
1154111541 \
1154211542 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -11547,18 +11547,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1154711547
1154811548#ifdef __LITTLE_ENDIAN__
1154911549#define vld2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11550 int32x4x2_t __s1 = __p1; \
1155111550 int32x4x2_t __ret; \
11551 int32x4x2_t __s1 = __p1; \
1155211552 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 34); \
1155311553 __ret; \
1155411554})
1155511555#else
1155611556#define vld2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11557 int32x4x2_t __ret; \
1155711558 int32x4x2_t __s1 = __p1; \
1155811559 int32x4x2_t __rev1; \
1155911560 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1156011561 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
11561 int32x4x2_t __ret; \
1156211562 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 34); \
1156311563 \
1156411564 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -11569,18 +11569,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1156911569
1157011570#ifdef __LITTLE_ENDIAN__
1157111571#define vld2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11572 int16x8x2_t __s1 = __p1; \
1157311572 int16x8x2_t __ret; \
11573 int16x8x2_t __s1 = __p1; \
1157411574 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 33); \
1157511575 __ret; \
1157611576})
1157711577#else
1157811578#define vld2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11579 int16x8x2_t __ret; \
1157911580 int16x8x2_t __s1 = __p1; \
1158011581 int16x8x2_t __rev1; \
1158111582 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1158211583 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
11583 int16x8x2_t __ret; \
1158411584 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 33); \
1158511585 \
1158611586 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -11591,18 +11591,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1159111591
1159211592#ifdef __LITTLE_ENDIAN__
1159311593#define vld2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
11594 uint8x8x2_t __s1 = __p1; \
1159511594 uint8x8x2_t __ret; \
11595 uint8x8x2_t __s1 = __p1; \
1159611596 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 16); \
1159711597 __ret; \
1159811598})
1159911599#else
1160011600#define vld2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
11601 uint8x8x2_t __ret; \
1160111602 uint8x8x2_t __s1 = __p1; \
1160211603 uint8x8x2_t __rev1; \
1160311604 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1160411605 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
11605 uint8x8x2_t __ret; \
1160611606 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 16); \
1160711607 \
1160811608 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -11613,18 +11613,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1161311613
1161411614#ifdef __LITTLE_ENDIAN__
1161511615#define vld2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11616 uint32x2x2_t __s1 = __p1; \
1161711616 uint32x2x2_t __ret; \
11617 uint32x2x2_t __s1 = __p1; \
1161811618 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 18); \
1161911619 __ret; \
1162011620})
1162111621#else
1162211622#define vld2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11623 uint32x2x2_t __ret; \
1162311624 uint32x2x2_t __s1 = __p1; \
1162411625 uint32x2x2_t __rev1; \
1162511626 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1162611627 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
11627 uint32x2x2_t __ret; \
1162811628 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 18); \
1162911629 \
1163011630 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -11635,18 +11635,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1163511635
1163611636#ifdef __LITTLE_ENDIAN__
1163711637#define vld2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11638 uint16x4x2_t __s1 = __p1; \
1163911638 uint16x4x2_t __ret; \
11639 uint16x4x2_t __s1 = __p1; \
1164011640 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 17); \
1164111641 __ret; \
1164211642})
1164311643#else
1164411644#define vld2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11645 uint16x4x2_t __ret; \
1164511646 uint16x4x2_t __s1 = __p1; \
1164611647 uint16x4x2_t __rev1; \
1164711648 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1164811649 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
11649 uint16x4x2_t __ret; \
1165011650 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 17); \
1165111651 \
1165211652 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -11657,18 +11657,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1165711657
1165811658#ifdef __LITTLE_ENDIAN__
1165911659#define vld2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
11660 int8x8x2_t __s1 = __p1; \
1166111660 int8x8x2_t __ret; \
11661 int8x8x2_t __s1 = __p1; \
1166211662 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 0); \
1166311663 __ret; \
1166411664})
1166511665#else
1166611666#define vld2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
11667 int8x8x2_t __ret; \
1166711668 int8x8x2_t __s1 = __p1; \
1166811669 int8x8x2_t __rev1; \
1166911670 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1167011671 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
11671 int8x8x2_t __ret; \
1167211672 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 0); \
1167311673 \
1167411674 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -11679,18 +11679,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1167911679
1168011680#ifdef __LITTLE_ENDIAN__
1168111681#define vld2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11682 float32x2x2_t __s1 = __p1; \
1168311682 float32x2x2_t __ret; \
11683 float32x2x2_t __s1 = __p1; \
1168411684 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 9); \
1168511685 __ret; \
1168611686})
1168711687#else
1168811688#define vld2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11689 float32x2x2_t __ret; \
1168911690 float32x2x2_t __s1 = __p1; \
1169011691 float32x2x2_t __rev1; \
1169111692 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1169211693 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
11693 float32x2x2_t __ret; \
1169411694 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 9); \
1169511695 \
1169611696 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -11701,18 +11701,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1170111701
1170211702#ifdef __LITTLE_ENDIAN__
1170311703#define vld2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11704 int32x2x2_t __s1 = __p1; \
1170511704 int32x2x2_t __ret; \
11705 int32x2x2_t __s1 = __p1; \
1170611706 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 2); \
1170711707 __ret; \
1170811708})
1170911709#else
1171011710#define vld2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11711 int32x2x2_t __ret; \
1171111712 int32x2x2_t __s1 = __p1; \
1171211713 int32x2x2_t __rev1; \
1171311714 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1171411715 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
11715 int32x2x2_t __ret; \
1171611716 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 2); \
1171711717 \
1171811718 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -11723,18 +11723,18 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1172311723
1172411724#ifdef __LITTLE_ENDIAN__
1172511725#define vld2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11726 int16x4x2_t __s1 = __p1; \
1172711726 int16x4x2_t __ret; \
11727 int16x4x2_t __s1 = __p1; \
1172811728 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 1); \
1172911729 __ret; \
1173011730})
1173111731#else
1173211732#define vld2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11733 int16x4x2_t __ret; \
1173311734 int16x4x2_t __s1 = __p1; \
1173411735 int16x4x2_t __rev1; \
1173511736 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1173611737 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
11737 int16x4x2_t __ret; \
1173811738 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 1); \
1173911739 \
1174011740 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -12449,19 +12449,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1244912449
1245012450#ifdef __LITTLE_ENDIAN__
1245112451#define vld3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
12452 poly8x8x3_t __s1 = __p1; \
1245312452 poly8x8x3_t __ret; \
12453 poly8x8x3_t __s1 = __p1; \
1245412454 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 4); \
1245512455 __ret; \
1245612456})
1245712457#else
1245812458#define vld3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
12459 poly8x8x3_t __ret; \
1245912460 poly8x8x3_t __s1 = __p1; \
1246012461 poly8x8x3_t __rev1; \
1246112462 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1246212463 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1246312464 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
12464 poly8x8x3_t __ret; \
1246512465 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 4); \
1246612466 \
1246712467 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -12473,19 +12473,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1247312473
1247412474#ifdef __LITTLE_ENDIAN__
1247512475#define vld3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
12476 poly16x4x3_t __s1 = __p1; \
1247712476 poly16x4x3_t __ret; \
12477 poly16x4x3_t __s1 = __p1; \
1247812478 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 5); \
1247912479 __ret; \
1248012480})
1248112481#else
1248212482#define vld3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
12483 poly16x4x3_t __ret; \
1248312484 poly16x4x3_t __s1 = __p1; \
1248412485 poly16x4x3_t __rev1; \
1248512486 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1248612487 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1248712488 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
12488 poly16x4x3_t __ret; \
1248912489 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 5); \
1249012490 \
1249112491 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -12497,19 +12497,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1249712497
1249812498#ifdef __LITTLE_ENDIAN__
1249912499#define vld3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
12500 poly16x8x3_t __s1 = __p1; \
1250112500 poly16x8x3_t __ret; \
12501 poly16x8x3_t __s1 = __p1; \
1250212502 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 37); \
1250312503 __ret; \
1250412504})
1250512505#else
1250612506#define vld3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
12507 poly16x8x3_t __ret; \
1250712508 poly16x8x3_t __s1 = __p1; \
1250812509 poly16x8x3_t __rev1; \
1250912510 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1251012511 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1251112512 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
12512 poly16x8x3_t __ret; \
1251312513 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 37); \
1251412514 \
1251512515 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -12521,19 +12521,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1252112521
1252212522#ifdef __LITTLE_ENDIAN__
1252312523#define vld3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
12524 uint32x4x3_t __s1 = __p1; \
1252512524 uint32x4x3_t __ret; \
12525 uint32x4x3_t __s1 = __p1; \
1252612526 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 50); \
1252712527 __ret; \
1252812528})
1252912529#else
1253012530#define vld3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
12531 uint32x4x3_t __ret; \
1253112532 uint32x4x3_t __s1 = __p1; \
1253212533 uint32x4x3_t __rev1; \
1253312534 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1253412535 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1253512536 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
12536 uint32x4x3_t __ret; \
1253712537 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 50); \
1253812538 \
1253912539 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -12545,19 +12545,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1254512545
1254612546#ifdef __LITTLE_ENDIAN__
1254712547#define vld3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
12548 uint16x8x3_t __s1 = __p1; \
1254912548 uint16x8x3_t __ret; \
12549 uint16x8x3_t __s1 = __p1; \
1255012550 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 49); \
1255112551 __ret; \
1255212552})
1255312553#else
1255412554#define vld3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
12555 uint16x8x3_t __ret; \
1255512556 uint16x8x3_t __s1 = __p1; \
1255612557 uint16x8x3_t __rev1; \
1255712558 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1255812559 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1255912560 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
12560 uint16x8x3_t __ret; \
1256112561 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 49); \
1256212562 \
1256312563 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -12569,19 +12569,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1256912569
1257012570#ifdef __LITTLE_ENDIAN__
1257112571#define vld3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
12572 float32x4x3_t __s1 = __p1; \
1257312572 float32x4x3_t __ret; \
12573 float32x4x3_t __s1 = __p1; \
1257412574 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 41); \
1257512575 __ret; \
1257612576})
1257712577#else
1257812578#define vld3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
12579 float32x4x3_t __ret; \
1257912580 float32x4x3_t __s1 = __p1; \
1258012581 float32x4x3_t __rev1; \
1258112582 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1258212583 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1258312584 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
12584 float32x4x3_t __ret; \
1258512585 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 41); \
1258612586 \
1258712587 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -12593,19 +12593,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1259312593
1259412594#ifdef __LITTLE_ENDIAN__
1259512595#define vld3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
12596 int32x4x3_t __s1 = __p1; \
1259712596 int32x4x3_t __ret; \
12597 int32x4x3_t __s1 = __p1; \
1259812598 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 34); \
1259912599 __ret; \
1260012600})
1260112601#else
1260212602#define vld3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
12603 int32x4x3_t __ret; \
1260312604 int32x4x3_t __s1 = __p1; \
1260412605 int32x4x3_t __rev1; \
1260512606 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1260612607 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1260712608 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
12608 int32x4x3_t __ret; \
1260912609 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 34); \
1261012610 \
1261112611 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -12617,19 +12617,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1261712617
1261812618#ifdef __LITTLE_ENDIAN__
1261912619#define vld3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
12620 int16x8x3_t __s1 = __p1; \
1262112620 int16x8x3_t __ret; \
12621 int16x8x3_t __s1 = __p1; \
1262212622 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 33); \
1262312623 __ret; \
1262412624})
1262512625#else
1262612626#define vld3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
12627 int16x8x3_t __ret; \
1262712628 int16x8x3_t __s1 = __p1; \
1262812629 int16x8x3_t __rev1; \
1262912630 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1263012631 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1263112632 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
12632 int16x8x3_t __ret; \
1263312633 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 33); \
1263412634 \
1263512635 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -12641,19 +12641,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1264112641
1264212642#ifdef __LITTLE_ENDIAN__
1264312643#define vld3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
12644 uint8x8x3_t __s1 = __p1; \
1264512644 uint8x8x3_t __ret; \
12645 uint8x8x3_t __s1 = __p1; \
1264612646 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 16); \
1264712647 __ret; \
1264812648})
1264912649#else
1265012650#define vld3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
12651 uint8x8x3_t __ret; \
1265112652 uint8x8x3_t __s1 = __p1; \
1265212653 uint8x8x3_t __rev1; \
1265312654 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1265412655 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1265512656 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
12656 uint8x8x3_t __ret; \
1265712657 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 16); \
1265812658 \
1265912659 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -12665,19 +12665,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1266512665
1266612666#ifdef __LITTLE_ENDIAN__
1266712667#define vld3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
12668 uint32x2x3_t __s1 = __p1; \
1266912668 uint32x2x3_t __ret; \
12669 uint32x2x3_t __s1 = __p1; \
1267012670 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 18); \
1267112671 __ret; \
1267212672})
1267312673#else
1267412674#define vld3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
12675 uint32x2x3_t __ret; \
1267512676 uint32x2x3_t __s1 = __p1; \
1267612677 uint32x2x3_t __rev1; \
1267712678 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1267812679 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
1267912680 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
12680 uint32x2x3_t __ret; \
1268112681 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 18); \
1268212682 \
1268312683 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -12689,19 +12689,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1268912689
1269012690#ifdef __LITTLE_ENDIAN__
1269112691#define vld3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
12692 uint16x4x3_t __s1 = __p1; \
1269312692 uint16x4x3_t __ret; \
12693 uint16x4x3_t __s1 = __p1; \
1269412694 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 17); \
1269512695 __ret; \
1269612696})
1269712697#else
1269812698#define vld3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
12699 uint16x4x3_t __ret; \
1269912700 uint16x4x3_t __s1 = __p1; \
1270012701 uint16x4x3_t __rev1; \
1270112702 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1270212703 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1270312704 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
12704 uint16x4x3_t __ret; \
1270512705 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 17); \
1270612706 \
1270712707 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -12713,19 +12713,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1271312713
1271412714#ifdef __LITTLE_ENDIAN__
1271512715#define vld3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
12716 int8x8x3_t __s1 = __p1; \
1271712716 int8x8x3_t __ret; \
12717 int8x8x3_t __s1 = __p1; \
1271812718 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 0); \
1271912719 __ret; \
1272012720})
1272112721#else
1272212722#define vld3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
12723 int8x8x3_t __ret; \
1272312724 int8x8x3_t __s1 = __p1; \
1272412725 int8x8x3_t __rev1; \
1272512726 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1272612727 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1272712728 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
12728 int8x8x3_t __ret; \
1272912729 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 0); \
1273012730 \
1273112731 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -12737,19 +12737,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1273712737
1273812738#ifdef __LITTLE_ENDIAN__
1273912739#define vld3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
12740 float32x2x3_t __s1 = __p1; \
1274112740 float32x2x3_t __ret; \
12741 float32x2x3_t __s1 = __p1; \
1274212742 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 9); \
1274312743 __ret; \
1274412744})
1274512745#else
1274612746#define vld3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
12747 float32x2x3_t __ret; \
1274712748 float32x2x3_t __s1 = __p1; \
1274812749 float32x2x3_t __rev1; \
1274912750 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1275012751 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
1275112752 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
12752 float32x2x3_t __ret; \
1275312753 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 9); \
1275412754 \
1275512755 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -12761,19 +12761,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1276112761
1276212762#ifdef __LITTLE_ENDIAN__
1276312763#define vld3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
12764 int32x2x3_t __s1 = __p1; \
1276512764 int32x2x3_t __ret; \
12765 int32x2x3_t __s1 = __p1; \
1276612766 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 2); \
1276712767 __ret; \
1276812768})
1276912769#else
1277012770#define vld3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
12771 int32x2x3_t __ret; \
1277112772 int32x2x3_t __s1 = __p1; \
1277212773 int32x2x3_t __rev1; \
1277312774 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1277412775 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
1277512776 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
12776 int32x2x3_t __ret; \
1277712777 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 2); \
1277812778 \
1277912779 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -12785,19 +12785,19 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1278512785
1278612786#ifdef __LITTLE_ENDIAN__
1278712787#define vld3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
12788 int16x4x3_t __s1 = __p1; \
1278912788 int16x4x3_t __ret; \
12789 int16x4x3_t __s1 = __p1; \
1279012790 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 1); \
1279112791 __ret; \
1279212792})
1279312793#else
1279412794#define vld3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
12795 int16x4x3_t __ret; \
1279512796 int16x4x3_t __s1 = __p1; \
1279612797 int16x4x3_t __rev1; \
1279712798 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1279812799 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1279912800 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
12800 int16x4x3_t __ret; \
1280112801 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 1); \
1280212802 \
1280312803 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13551,20 +13551,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1355113551
1355213552#ifdef __LITTLE_ENDIAN__
1355313553#define vld4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
13554 poly8x8x4_t __s1 = __p1; \
1355513554 poly8x8x4_t __ret; \
13555 poly8x8x4_t __s1 = __p1; \
1355613556 __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, 4); \
1355713557 __ret; \
1355813558})
1355913559#else
1356013560#define vld4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
13561 poly8x8x4_t __ret; \
1356113562 poly8x8x4_t __s1 = __p1; \
1356213563 poly8x8x4_t __rev1; \
1356313564 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1356413565 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1356513566 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
1356613567 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
13567 poly8x8x4_t __ret; \
1356813568 __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, 4); \
1356913569 \
1357013570 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -13577,20 +13577,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1357713577
1357813578#ifdef __LITTLE_ENDIAN__
1357913579#define vld4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
13580 poly16x4x4_t __s1 = __p1; \
1358113580 poly16x4x4_t __ret; \
13581 poly16x4x4_t __s1 = __p1; \
1358213582 __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, 5); \
1358313583 __ret; \
1358413584})
1358513585#else
1358613586#define vld4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
13587 poly16x4x4_t __ret; \
1358713588 poly16x4x4_t __s1 = __p1; \
1358813589 poly16x4x4_t __rev1; \
1358913590 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1359013591 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1359113592 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
1359213593 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
13593 poly16x4x4_t __ret; \
1359413594 __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, 5); \
1359513595 \
1359613596 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13603,20 +13603,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1360313603
1360413604#ifdef __LITTLE_ENDIAN__
1360513605#define vld4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
13606 poly16x8x4_t __s1 = __p1; \
1360713606 poly16x8x4_t __ret; \
13607 poly16x8x4_t __s1 = __p1; \
1360813608 __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, 37); \
1360913609 __ret; \
1361013610})
1361113611#else
1361213612#define vld4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
13613 poly16x8x4_t __ret; \
1361313614 poly16x8x4_t __s1 = __p1; \
1361413615 poly16x8x4_t __rev1; \
1361513616 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1361613617 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1361713618 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
1361813619 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
13619 poly16x8x4_t __ret; \
1362013620 __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, 37); \
1362113621 \
1362213622 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -13629,20 +13629,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1362913629
1363013630#ifdef __LITTLE_ENDIAN__
1363113631#define vld4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
13632 uint32x4x4_t __s1 = __p1; \
1363313632 uint32x4x4_t __ret; \
13633 uint32x4x4_t __s1 = __p1; \
1363413634 __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, 50); \
1363513635 __ret; \
1363613636})
1363713637#else
1363813638#define vld4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
13639 uint32x4x4_t __ret; \
1363913640 uint32x4x4_t __s1 = __p1; \
1364013641 uint32x4x4_t __rev1; \
1364113642 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1364213643 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1364313644 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
1364413645 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
13645 uint32x4x4_t __ret; \
1364613646 __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, 50); \
1364713647 \
1364813648 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13655,20 +13655,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1365513655
1365613656#ifdef __LITTLE_ENDIAN__
1365713657#define vld4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
13658 uint16x8x4_t __s1 = __p1; \
1365913658 uint16x8x4_t __ret; \
13659 uint16x8x4_t __s1 = __p1; \
1366013660 __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, 49); \
1366113661 __ret; \
1366213662})
1366313663#else
1366413664#define vld4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
13665 uint16x8x4_t __ret; \
1366513666 uint16x8x4_t __s1 = __p1; \
1366613667 uint16x8x4_t __rev1; \
1366713668 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1366813669 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1366913670 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
1367013671 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
13671 uint16x8x4_t __ret; \
1367213672 __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, 49); \
1367313673 \
1367413674 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -13681,20 +13681,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1368113681
1368213682#ifdef __LITTLE_ENDIAN__
1368313683#define vld4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
13684 float32x4x4_t __s1 = __p1; \
1368513684 float32x4x4_t __ret; \
13685 float32x4x4_t __s1 = __p1; \
1368613686 __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, 41); \
1368713687 __ret; \
1368813688})
1368913689#else
1369013690#define vld4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
13691 float32x4x4_t __ret; \
1369113692 float32x4x4_t __s1 = __p1; \
1369213693 float32x4x4_t __rev1; \
1369313694 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1369413695 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1369513696 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
1369613697 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
13697 float32x4x4_t __ret; \
1369813698 __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, 41); \
1369913699 \
1370013700 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13707,20 +13707,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1370713707
1370813708#ifdef __LITTLE_ENDIAN__
1370913709#define vld4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
13710 int32x4x4_t __s1 = __p1; \
1371113710 int32x4x4_t __ret; \
13711 int32x4x4_t __s1 = __p1; \
1371213712 __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, 34); \
1371313713 __ret; \
1371413714})
1371513715#else
1371613716#define vld4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
13717 int32x4x4_t __ret; \
1371713718 int32x4x4_t __s1 = __p1; \
1371813719 int32x4x4_t __rev1; \
1371913720 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1372013721 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1372113722 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
1372213723 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
13723 int32x4x4_t __ret; \
1372413724 __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, 34); \
1372513725 \
1372613726 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13733,20 +13733,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1373313733
1373413734#ifdef __LITTLE_ENDIAN__
1373513735#define vld4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
13736 int16x8x4_t __s1 = __p1; \
1373713736 int16x8x4_t __ret; \
13737 int16x8x4_t __s1 = __p1; \
1373813738 __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, 33); \
1373913739 __ret; \
1374013740})
1374113741#else
1374213742#define vld4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
13743 int16x8x4_t __ret; \
1374313744 int16x8x4_t __s1 = __p1; \
1374413745 int16x8x4_t __rev1; \
1374513746 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1374613747 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1374713748 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
1374813749 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
13749 int16x8x4_t __ret; \
1375013750 __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, 33); \
1375113751 \
1375213752 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -13759,20 +13759,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1375913759
1376013760#ifdef __LITTLE_ENDIAN__
1376113761#define vld4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
13762 uint8x8x4_t __s1 = __p1; \
1376313762 uint8x8x4_t __ret; \
13763 uint8x8x4_t __s1 = __p1; \
1376413764 __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, 16); \
1376513765 __ret; \
1376613766})
1376713767#else
1376813768#define vld4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
13769 uint8x8x4_t __ret; \
1376913770 uint8x8x4_t __s1 = __p1; \
1377013771 uint8x8x4_t __rev1; \
1377113772 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1377213773 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1377313774 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
1377413775 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
13775 uint8x8x4_t __ret; \
1377613776 __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, 16); \
1377713777 \
1377813778 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -13785,20 +13785,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1378513785
1378613786#ifdef __LITTLE_ENDIAN__
1378713787#define vld4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
13788 uint32x2x4_t __s1 = __p1; \
1378913788 uint32x2x4_t __ret; \
13789 uint32x2x4_t __s1 = __p1; \
1379013790 __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, 18); \
1379113791 __ret; \
1379213792})
1379313793#else
1379413794#define vld4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
13795 uint32x2x4_t __ret; \
1379513796 uint32x2x4_t __s1 = __p1; \
1379613797 uint32x2x4_t __rev1; \
1379713798 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1379813799 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
1379913800 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
1380013801 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
13801 uint32x2x4_t __ret; \
1380213802 __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, 18); \
1380313803 \
1380413804 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -13811,20 +13811,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1381113811
1381213812#ifdef __LITTLE_ENDIAN__
1381313813#define vld4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
13814 uint16x4x4_t __s1 = __p1; \
1381513814 uint16x4x4_t __ret; \
13815 uint16x4x4_t __s1 = __p1; \
1381613816 __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, 17); \
1381713817 __ret; \
1381813818})
1381913819#else
1382013820#define vld4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
13821 uint16x4x4_t __ret; \
1382113822 uint16x4x4_t __s1 = __p1; \
1382213823 uint16x4x4_t __rev1; \
1382313824 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1382413825 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1382513826 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
1382613827 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
13827 uint16x4x4_t __ret; \
1382813828 __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, 17); \
1382913829 \
1383013830 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13837,20 +13837,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1383713837
1383813838#ifdef __LITTLE_ENDIAN__
1383913839#define vld4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
13840 int8x8x4_t __s1 = __p1; \
1384113840 int8x8x4_t __ret; \
13841 int8x8x4_t __s1 = __p1; \
1384213842 __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, 0); \
1384313843 __ret; \
1384413844})
1384513845#else
1384613846#define vld4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
13847 int8x8x4_t __ret; \
1384713848 int8x8x4_t __s1 = __p1; \
1384813849 int8x8x4_t __rev1; \
1384913850 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
1385013851 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
1385113852 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
1385213853 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
13853 int8x8x4_t __ret; \
1385413854 __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, 0); \
1385513855 \
1385613856 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -13863,20 +13863,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1386313863
1386413864#ifdef __LITTLE_ENDIAN__
1386513865#define vld4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
13866 float32x2x4_t __s1 = __p1; \
1386713866 float32x2x4_t __ret; \
13867 float32x2x4_t __s1 = __p1; \
1386813868 __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, 9); \
1386913869 __ret; \
1387013870})
1387113871#else
1387213872#define vld4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
13873 float32x2x4_t __ret; \
1387313874 float32x2x4_t __s1 = __p1; \
1387413875 float32x2x4_t __rev1; \
1387513876 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1387613877 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
1387713878 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
1387813879 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
13879 float32x2x4_t __ret; \
1388013880 __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, 9); \
1388113881 \
1388213882 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -13889,20 +13889,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1388913889
1389013890#ifdef __LITTLE_ENDIAN__
1389113891#define vld4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
13892 int32x2x4_t __s1 = __p1; \
1389313892 int32x2x4_t __ret; \
13893 int32x2x4_t __s1 = __p1; \
1389413894 __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, 2); \
1389513895 __ret; \
1389613896})
1389713897#else
1389813898#define vld4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
13899 int32x2x4_t __ret; \
1389913900 int32x2x4_t __s1 = __p1; \
1390013901 int32x2x4_t __rev1; \
1390113902 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
1390213903 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
1390313904 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
1390413905 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
13905 int32x2x4_t __ret; \
1390613906 __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, 2); \
1390713907 \
1390813908 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -13915,20 +13915,20 @@ __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
1391513915
1391613916#ifdef __LITTLE_ENDIAN__
1391713917#define vld4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
13918 int16x4x4_t __s1 = __p1; \
1391913918 int16x4x4_t __ret; \
13919 int16x4x4_t __s1 = __p1; \
1392013920 __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, 1); \
1392113921 __ret; \
1392213922})
1392313923#else
1392413924#define vld4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
13925 int16x4x4_t __ret; \
1392513926 int16x4x4_t __s1 = __p1; \
1392613927 int16x4x4_t __rev1; \
1392713928 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
1392813929 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
1392913930 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
1393013931 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
13931 int16x4x4_t __ret; \
1393213932 __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, 1); \
1393313933 \
1393413934 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -13947,9 +13947,9 @@ __ai uint8x16_t vmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1394713947}
1394813948#else
1394913949__ai uint8x16_t vmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
13950 uint8x16_t __ret;
1395013951 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1395113952 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
13952 uint8x16_t __ret;
1395313953 __ret = (uint8x16_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
1395413954 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1395513955 return __ret;
......@@ -13964,9 +13964,9 @@ __ai uint32x4_t vmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1396413964}
1396513965#else
1396613966__ai uint32x4_t vmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
13967 uint32x4_t __ret;
1396713968 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1396813969 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13969 uint32x4_t __ret;
1397013970 __ret = (uint32x4_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
1397113971 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1397213972 return __ret;
......@@ -13981,9 +13981,9 @@ __ai uint16x8_t vmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1398113981}
1398213982#else
1398313983__ai uint16x8_t vmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
13984 uint16x8_t __ret;
1398413985 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1398513986 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
13986 uint16x8_t __ret;
1398713987 __ret = (uint16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
1398813988 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1398913989 return __ret;
......@@ -13998,9 +13998,9 @@ __ai int8x16_t vmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
1399813998}
1399913999#else
1400014000__ai int8x16_t vmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
14001 int8x16_t __ret;
1400114002 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1400214003 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14003 int8x16_t __ret;
1400414004 __ret = (int8x16_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
1400514005 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1400614006 return __ret;
......@@ -14015,9 +14015,9 @@ __ai float32x4_t vmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
1401514015}
1401614016#else
1401714017__ai float32x4_t vmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
14018 float32x4_t __ret;
1401814019 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1401914020 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14020 float32x4_t __ret;
1402114021 __ret = (float32x4_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
1402214022 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1402314023 return __ret;
......@@ -14032,9 +14032,9 @@ __ai int32x4_t vmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
1403214032}
1403314033#else
1403414034__ai int32x4_t vmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
14035 int32x4_t __ret;
1403514036 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1403614037 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14037 int32x4_t __ret;
1403814038 __ret = (int32x4_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
1403914039 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1404014040 return __ret;
......@@ -14049,9 +14049,9 @@ __ai int16x8_t vmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
1404914049}
1405014050#else
1405114051__ai int16x8_t vmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
14052 int16x8_t __ret;
1405214053 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1405314054 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14054 int16x8_t __ret;
1405514055 __ret = (int16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
1405614056 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1405714057 return __ret;
......@@ -14066,9 +14066,9 @@ __ai uint8x8_t vmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
1406614066}
1406714067#else
1406814068__ai uint8x8_t vmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
14069 uint8x8_t __ret;
1406914070 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1407014071 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14071 uint8x8_t __ret;
1407214072 __ret = (uint8x8_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
1407314073 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1407414074 return __ret;
......@@ -14083,9 +14083,9 @@ __ai uint32x2_t vmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
1408314083}
1408414084#else
1408514085__ai uint32x2_t vmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
14086 uint32x2_t __ret;
1408614087 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1408714088 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14088 uint32x2_t __ret;
1408914089 __ret = (uint32x2_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1409014090 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1409114091 return __ret;
......@@ -14100,9 +14100,9 @@ __ai uint16x4_t vmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
1410014100}
1410114101#else
1410214102__ai uint16x4_t vmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
14103 uint16x4_t __ret;
1410314104 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1410414105 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14105 uint16x4_t __ret;
1410614106 __ret = (uint16x4_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1410714107 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1410814108 return __ret;
......@@ -14117,9 +14117,9 @@ __ai int8x8_t vmax_s8(int8x8_t __p0, int8x8_t __p1) {
1411714117}
1411814118#else
1411914119__ai int8x8_t vmax_s8(int8x8_t __p0, int8x8_t __p1) {
14120 int8x8_t __ret;
1412014121 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1412114122 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14122 int8x8_t __ret;
1412314123 __ret = (int8x8_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
1412414124 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1412514125 return __ret;
......@@ -14134,9 +14134,9 @@ __ai float32x2_t vmax_f32(float32x2_t __p0, float32x2_t __p1) {
1413414134}
1413514135#else
1413614136__ai float32x2_t vmax_f32(float32x2_t __p0, float32x2_t __p1) {
14137 float32x2_t __ret;
1413714138 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1413814139 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14139 float32x2_t __ret;
1414014140 __ret = (float32x2_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
1414114141 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1414214142 return __ret;
......@@ -14151,9 +14151,9 @@ __ai int32x2_t vmax_s32(int32x2_t __p0, int32x2_t __p1) {
1415114151}
1415214152#else
1415314153__ai int32x2_t vmax_s32(int32x2_t __p0, int32x2_t __p1) {
14154 int32x2_t __ret;
1415414155 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1415514156 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14156 int32x2_t __ret;
1415714157 __ret = (int32x2_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1415814158 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1415914159 return __ret;
......@@ -14168,9 +14168,9 @@ __ai int16x4_t vmax_s16(int16x4_t __p0, int16x4_t __p1) {
1416814168}
1416914169#else
1417014170__ai int16x4_t vmax_s16(int16x4_t __p0, int16x4_t __p1) {
14171 int16x4_t __ret;
1417114172 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1417214173 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14173 int16x4_t __ret;
1417414174 __ret = (int16x4_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1417514175 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1417614176 return __ret;
......@@ -14185,9 +14185,9 @@ __ai uint8x16_t vminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1418514185}
1418614186#else
1418714187__ai uint8x16_t vminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
14188 uint8x16_t __ret;
1418814189 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1418914190 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14190 uint8x16_t __ret;
1419114191 __ret = (uint8x16_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
1419214192 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1419314193 return __ret;
......@@ -14202,9 +14202,9 @@ __ai uint32x4_t vminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1420214202}
1420314203#else
1420414204__ai uint32x4_t vminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
14205 uint32x4_t __ret;
1420514206 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1420614207 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14207 uint32x4_t __ret;
1420814208 __ret = (uint32x4_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
1420914209 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1421014210 return __ret;
......@@ -14219,9 +14219,9 @@ __ai uint16x8_t vminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1421914219}
1422014220#else
1422114221__ai uint16x8_t vminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
14222 uint16x8_t __ret;
1422214223 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1422314224 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14224 uint16x8_t __ret;
1422514225 __ret = (uint16x8_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
1422614226 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1422714227 return __ret;
......@@ -14236,9 +14236,9 @@ __ai int8x16_t vminq_s8(int8x16_t __p0, int8x16_t __p1) {
1423614236}
1423714237#else
1423814238__ai int8x16_t vminq_s8(int8x16_t __p0, int8x16_t __p1) {
14239 int8x16_t __ret;
1423914240 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1424014241 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14241 int8x16_t __ret;
1424214242 __ret = (int8x16_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
1424314243 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1424414244 return __ret;
......@@ -14253,9 +14253,9 @@ __ai float32x4_t vminq_f32(float32x4_t __p0, float32x4_t __p1) {
1425314253}
1425414254#else
1425514255__ai float32x4_t vminq_f32(float32x4_t __p0, float32x4_t __p1) {
14256 float32x4_t __ret;
1425614257 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1425714258 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14258 float32x4_t __ret;
1425914259 __ret = (float32x4_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
1426014260 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1426114261 return __ret;
......@@ -14270,9 +14270,9 @@ __ai int32x4_t vminq_s32(int32x4_t __p0, int32x4_t __p1) {
1427014270}
1427114271#else
1427214272__ai int32x4_t vminq_s32(int32x4_t __p0, int32x4_t __p1) {
14273 int32x4_t __ret;
1427314274 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1427414275 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14275 int32x4_t __ret;
1427614276 __ret = (int32x4_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
1427714277 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1427814278 return __ret;
......@@ -14287,9 +14287,9 @@ __ai int16x8_t vminq_s16(int16x8_t __p0, int16x8_t __p1) {
1428714287}
1428814288#else
1428914289__ai int16x8_t vminq_s16(int16x8_t __p0, int16x8_t __p1) {
14290 int16x8_t __ret;
1429014291 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1429114292 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14292 int16x8_t __ret;
1429314293 __ret = (int16x8_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
1429414294 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1429514295 return __ret;
......@@ -14304,9 +14304,9 @@ __ai uint8x8_t vmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
1430414304}
1430514305#else
1430614306__ai uint8x8_t vmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
14307 uint8x8_t __ret;
1430714308 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1430814309 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14309 uint8x8_t __ret;
1431014310 __ret = (uint8x8_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
1431114311 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1431214312 return __ret;
......@@ -14321,9 +14321,9 @@ __ai uint32x2_t vmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
1432114321}
1432214322#else
1432314323__ai uint32x2_t vmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
14324 uint32x2_t __ret;
1432414325 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1432514326 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14326 uint32x2_t __ret;
1432714327 __ret = (uint32x2_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1432814328 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1432914329 return __ret;
......@@ -14338,9 +14338,9 @@ __ai uint16x4_t vmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
1433814338}
1433914339#else
1434014340__ai uint16x4_t vmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
14341 uint16x4_t __ret;
1434114342 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1434214343 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14343 uint16x4_t __ret;
1434414344 __ret = (uint16x4_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1434514345 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1434614346 return __ret;
......@@ -14355,9 +14355,9 @@ __ai int8x8_t vmin_s8(int8x8_t __p0, int8x8_t __p1) {
1435514355}
1435614356#else
1435714357__ai int8x8_t vmin_s8(int8x8_t __p0, int8x8_t __p1) {
14358 int8x8_t __ret;
1435814359 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1435914360 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14360 int8x8_t __ret;
1436114361 __ret = (int8x8_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
1436214362 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1436314363 return __ret;
......@@ -14372,9 +14372,9 @@ __ai float32x2_t vmin_f32(float32x2_t __p0, float32x2_t __p1) {
1437214372}
1437314373#else
1437414374__ai float32x2_t vmin_f32(float32x2_t __p0, float32x2_t __p1) {
14375 float32x2_t __ret;
1437514376 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1437614377 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14377 float32x2_t __ret;
1437814378 __ret = (float32x2_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
1437914379 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1438014380 return __ret;
......@@ -14389,9 +14389,9 @@ __ai int32x2_t vmin_s32(int32x2_t __p0, int32x2_t __p1) {
1438914389}
1439014390#else
1439114391__ai int32x2_t vmin_s32(int32x2_t __p0, int32x2_t __p1) {
14392 int32x2_t __ret;
1439214393 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1439314394 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14394 int32x2_t __ret;
1439514395 __ret = (int32x2_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1439614396 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1439714397 return __ret;
......@@ -14406,9 +14406,9 @@ __ai int16x4_t vmin_s16(int16x4_t __p0, int16x4_t __p1) {
1440614406}
1440714407#else
1440814408__ai int16x4_t vmin_s16(int16x4_t __p0, int16x4_t __p1) {
14409 int16x4_t __ret;
1440914410 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1441014411 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14411 int16x4_t __ret;
1441214412 __ret = (int16x4_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1441314413 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1441414414 return __ret;
......@@ -14423,10 +14423,10 @@ __ai uint8x16_t vmlaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
1442314423}
1442414424#else
1442514425__ai uint8x16_t vmlaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
14426 uint8x16_t __ret;
1442614427 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1442714428 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1442814429 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14429 uint8x16_t __ret;
1443014430 __ret = __rev0 + __rev1 * __rev2;
1443114431 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1443214432 return __ret;
......@@ -14441,10 +14441,10 @@ __ai uint32x4_t vmlaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
1444114441}
1444214442#else
1444314443__ai uint32x4_t vmlaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
14444 uint32x4_t __ret;
1444414445 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1444514446 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1444614447 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
14447 uint32x4_t __ret;
1444814448 __ret = __rev0 + __rev1 * __rev2;
1444914449 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1445014450 return __ret;
......@@ -14459,10 +14459,10 @@ __ai uint16x8_t vmlaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
1445914459}
1446014460#else
1446114461__ai uint16x8_t vmlaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
14462 uint16x8_t __ret;
1446214463 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1446314464 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1446414465 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
14465 uint16x8_t __ret;
1446614466 __ret = __rev0 + __rev1 * __rev2;
1446714467 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1446814468 return __ret;
......@@ -14477,10 +14477,10 @@ __ai int8x16_t vmlaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
1447714477}
1447814478#else
1447914479__ai int8x16_t vmlaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
14480 int8x16_t __ret;
1448014481 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1448114482 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1448214483 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14483 int8x16_t __ret;
1448414484 __ret = __rev0 + __rev1 * __rev2;
1448514485 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1448614486 return __ret;
......@@ -14495,10 +14495,10 @@ __ai float32x4_t vmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2)
1449514495}
1449614496#else
1449714497__ai float32x4_t vmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
14498 float32x4_t __ret;
1449814499 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1449914500 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1450014501 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
14501 float32x4_t __ret;
1450214502 __ret = __rev0 + __rev1 * __rev2;
1450314503 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1450414504 return __ret;
......@@ -14513,10 +14513,10 @@ __ai int32x4_t vmlaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
1451314513}
1451414514#else
1451514515__ai int32x4_t vmlaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
14516 int32x4_t __ret;
1451614517 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1451714518 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1451814519 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
14519 int32x4_t __ret;
1452014520 __ret = __rev0 + __rev1 * __rev2;
1452114521 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1452214522 return __ret;
......@@ -14531,10 +14531,10 @@ __ai int16x8_t vmlaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
1453114531}
1453214532#else
1453314533__ai int16x8_t vmlaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
14534 int16x8_t __ret;
1453414535 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1453514536 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1453614537 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
14537 int16x8_t __ret;
1453814538 __ret = __rev0 + __rev1 * __rev2;
1453914539 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1454014540 return __ret;
......@@ -14549,10 +14549,10 @@ __ai uint8x8_t vmla_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
1454914549}
1455014550#else
1455114551__ai uint8x8_t vmla_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
14552 uint8x8_t __ret;
1455214553 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1455314554 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1455414555 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
14555 uint8x8_t __ret;
1455614556 __ret = __rev0 + __rev1 * __rev2;
1455714557 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1455814558 return __ret;
......@@ -14567,10 +14567,10 @@ __ai uint32x2_t vmla_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
1456714567}
1456814568#else
1456914569__ai uint32x2_t vmla_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
14570 uint32x2_t __ret;
1457014571 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1457114572 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1457214573 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
14573 uint32x2_t __ret;
1457414574 __ret = __rev0 + __rev1 * __rev2;
1457514575 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1457614576 return __ret;
......@@ -14585,10 +14585,10 @@ __ai uint16x4_t vmla_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
1458514585}
1458614586#else
1458714587__ai uint16x4_t vmla_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
14588 uint16x4_t __ret;
1458814589 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1458914590 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1459014591 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
14591 uint16x4_t __ret;
1459214592 __ret = __rev0 + __rev1 * __rev2;
1459314593 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1459414594 return __ret;
......@@ -14603,10 +14603,10 @@ __ai int8x8_t vmla_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
1460314603}
1460414604#else
1460514605__ai int8x8_t vmla_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
14606 int8x8_t __ret;
1460614607 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1460714608 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1460814609 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
14609 int8x8_t __ret;
1461014610 __ret = __rev0 + __rev1 * __rev2;
1461114611 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1461214612 return __ret;
......@@ -14621,10 +14621,10 @@ __ai float32x2_t vmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2)
1462114621}
1462214622#else
1462314623__ai float32x2_t vmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
14624 float32x2_t __ret;
1462414625 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1462514626 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1462614627 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
14627 float32x2_t __ret;
1462814628 __ret = __rev0 + __rev1 * __rev2;
1462914629 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1463014630 return __ret;
......@@ -14639,10 +14639,10 @@ __ai int32x2_t vmla_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
1463914639}
1464014640#else
1464114641__ai int32x2_t vmla_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
14642 int32x2_t __ret;
1464214643 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1464314644 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1464414645 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
14645 int32x2_t __ret;
1464614646 __ret = __rev0 + __rev1 * __rev2;
1464714647 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1464814648 return __ret;
......@@ -14657,10 +14657,10 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1465714657}
1465814658#else
1465914659__ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
14660 int16x4_t __ret;
1466014661 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1466114662 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1466214663 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
14663 int16x4_t __ret;
1466414664 __ret = __rev0 + __rev1 * __rev2;
1466514665 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1466614666 return __ret;
......@@ -14669,22 +14669,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1466914669
1467014670#ifdef __LITTLE_ENDIAN__
1467114671#define vmlaq_lane_u32(__p0_42, __p1_42, __p2_42, __p3_42) __extension__ ({ \
14672 uint32x4_t __ret_42; \
1467214673 uint32x4_t __s0_42 = __p0_42; \
1467314674 uint32x4_t __s1_42 = __p1_42; \
1467414675 uint32x2_t __s2_42 = __p2_42; \
14675 uint32x4_t __ret_42; \
1467614676 __ret_42 = __s0_42 + __s1_42 * splatq_lane_u32(__s2_42, __p3_42); \
1467714677 __ret_42; \
1467814678})
1467914679#else
1468014680#define vmlaq_lane_u32(__p0_43, __p1_43, __p2_43, __p3_43) __extension__ ({ \
14681 uint32x4_t __ret_43; \
1468114682 uint32x4_t __s0_43 = __p0_43; \
1468214683 uint32x4_t __s1_43 = __p1_43; \
1468314684 uint32x2_t __s2_43 = __p2_43; \
1468414685 uint32x4_t __rev0_43; __rev0_43 = __builtin_shufflevector(__s0_43, __s0_43, 3, 2, 1, 0); \
1468514686 uint32x4_t __rev1_43; __rev1_43 = __builtin_shufflevector(__s1_43, __s1_43, 3, 2, 1, 0); \
1468614687 uint32x2_t __rev2_43; __rev2_43 = __builtin_shufflevector(__s2_43, __s2_43, 1, 0); \
14687 uint32x4_t __ret_43; \
1468814688 __ret_43 = __rev0_43 + __rev1_43 * __noswap_splatq_lane_u32(__rev2_43, __p3_43); \
1468914689 __ret_43 = __builtin_shufflevector(__ret_43, __ret_43, 3, 2, 1, 0); \
1469014690 __ret_43; \
......@@ -14693,22 +14693,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1469314693
1469414694#ifdef __LITTLE_ENDIAN__
1469514695#define vmlaq_lane_u16(__p0_44, __p1_44, __p2_44, __p3_44) __extension__ ({ \
14696 uint16x8_t __ret_44; \
1469614697 uint16x8_t __s0_44 = __p0_44; \
1469714698 uint16x8_t __s1_44 = __p1_44; \
1469814699 uint16x4_t __s2_44 = __p2_44; \
14699 uint16x8_t __ret_44; \
1470014700 __ret_44 = __s0_44 + __s1_44 * splatq_lane_u16(__s2_44, __p3_44); \
1470114701 __ret_44; \
1470214702})
1470314703#else
1470414704#define vmlaq_lane_u16(__p0_45, __p1_45, __p2_45, __p3_45) __extension__ ({ \
14705 uint16x8_t __ret_45; \
1470514706 uint16x8_t __s0_45 = __p0_45; \
1470614707 uint16x8_t __s1_45 = __p1_45; \
1470714708 uint16x4_t __s2_45 = __p2_45; \
1470814709 uint16x8_t __rev0_45; __rev0_45 = __builtin_shufflevector(__s0_45, __s0_45, 7, 6, 5, 4, 3, 2, 1, 0); \
1470914710 uint16x8_t __rev1_45; __rev1_45 = __builtin_shufflevector(__s1_45, __s1_45, 7, 6, 5, 4, 3, 2, 1, 0); \
1471014711 uint16x4_t __rev2_45; __rev2_45 = __builtin_shufflevector(__s2_45, __s2_45, 3, 2, 1, 0); \
14711 uint16x8_t __ret_45; \
1471214712 __ret_45 = __rev0_45 + __rev1_45 * __noswap_splatq_lane_u16(__rev2_45, __p3_45); \
1471314713 __ret_45 = __builtin_shufflevector(__ret_45, __ret_45, 7, 6, 5, 4, 3, 2, 1, 0); \
1471414714 __ret_45; \
......@@ -14717,22 +14717,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1471714717
1471814718#ifdef __LITTLE_ENDIAN__
1471914719#define vmlaq_lane_f32(__p0_46, __p1_46, __p2_46, __p3_46) __extension__ ({ \
14720 float32x4_t __ret_46; \
1472014721 float32x4_t __s0_46 = __p0_46; \
1472114722 float32x4_t __s1_46 = __p1_46; \
1472214723 float32x2_t __s2_46 = __p2_46; \
14723 float32x4_t __ret_46; \
1472414724 __ret_46 = __s0_46 + __s1_46 * splatq_lane_f32(__s2_46, __p3_46); \
1472514725 __ret_46; \
1472614726})
1472714727#else
1472814728#define vmlaq_lane_f32(__p0_47, __p1_47, __p2_47, __p3_47) __extension__ ({ \
14729 float32x4_t __ret_47; \
1472914730 float32x4_t __s0_47 = __p0_47; \
1473014731 float32x4_t __s1_47 = __p1_47; \
1473114732 float32x2_t __s2_47 = __p2_47; \
1473214733 float32x4_t __rev0_47; __rev0_47 = __builtin_shufflevector(__s0_47, __s0_47, 3, 2, 1, 0); \
1473314734 float32x4_t __rev1_47; __rev1_47 = __builtin_shufflevector(__s1_47, __s1_47, 3, 2, 1, 0); \
1473414735 float32x2_t __rev2_47; __rev2_47 = __builtin_shufflevector(__s2_47, __s2_47, 1, 0); \
14735 float32x4_t __ret_47; \
1473614736 __ret_47 = __rev0_47 + __rev1_47 * __noswap_splatq_lane_f32(__rev2_47, __p3_47); \
1473714737 __ret_47 = __builtin_shufflevector(__ret_47, __ret_47, 3, 2, 1, 0); \
1473814738 __ret_47; \
......@@ -14741,22 +14741,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1474114741
1474214742#ifdef __LITTLE_ENDIAN__
1474314743#define vmlaq_lane_s32(__p0_48, __p1_48, __p2_48, __p3_48) __extension__ ({ \
14744 int32x4_t __ret_48; \
1474414745 int32x4_t __s0_48 = __p0_48; \
1474514746 int32x4_t __s1_48 = __p1_48; \
1474614747 int32x2_t __s2_48 = __p2_48; \
14747 int32x4_t __ret_48; \
1474814748 __ret_48 = __s0_48 + __s1_48 * splatq_lane_s32(__s2_48, __p3_48); \
1474914749 __ret_48; \
1475014750})
1475114751#else
1475214752#define vmlaq_lane_s32(__p0_49, __p1_49, __p2_49, __p3_49) __extension__ ({ \
14753 int32x4_t __ret_49; \
1475314754 int32x4_t __s0_49 = __p0_49; \
1475414755 int32x4_t __s1_49 = __p1_49; \
1475514756 int32x2_t __s2_49 = __p2_49; \
1475614757 int32x4_t __rev0_49; __rev0_49 = __builtin_shufflevector(__s0_49, __s0_49, 3, 2, 1, 0); \
1475714758 int32x4_t __rev1_49; __rev1_49 = __builtin_shufflevector(__s1_49, __s1_49, 3, 2, 1, 0); \
1475814759 int32x2_t __rev2_49; __rev2_49 = __builtin_shufflevector(__s2_49, __s2_49, 1, 0); \
14759 int32x4_t __ret_49; \
1476014760 __ret_49 = __rev0_49 + __rev1_49 * __noswap_splatq_lane_s32(__rev2_49, __p3_49); \
1476114761 __ret_49 = __builtin_shufflevector(__ret_49, __ret_49, 3, 2, 1, 0); \
1476214762 __ret_49; \
......@@ -14765,22 +14765,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1476514765
1476614766#ifdef __LITTLE_ENDIAN__
1476714767#define vmlaq_lane_s16(__p0_50, __p1_50, __p2_50, __p3_50) __extension__ ({ \
14768 int16x8_t __ret_50; \
1476814769 int16x8_t __s0_50 = __p0_50; \
1476914770 int16x8_t __s1_50 = __p1_50; \
1477014771 int16x4_t __s2_50 = __p2_50; \
14771 int16x8_t __ret_50; \
1477214772 __ret_50 = __s0_50 + __s1_50 * splatq_lane_s16(__s2_50, __p3_50); \
1477314773 __ret_50; \
1477414774})
1477514775#else
1477614776#define vmlaq_lane_s16(__p0_51, __p1_51, __p2_51, __p3_51) __extension__ ({ \
14777 int16x8_t __ret_51; \
1477714778 int16x8_t __s0_51 = __p0_51; \
1477814779 int16x8_t __s1_51 = __p1_51; \
1477914780 int16x4_t __s2_51 = __p2_51; \
1478014781 int16x8_t __rev0_51; __rev0_51 = __builtin_shufflevector(__s0_51, __s0_51, 7, 6, 5, 4, 3, 2, 1, 0); \
1478114782 int16x8_t __rev1_51; __rev1_51 = __builtin_shufflevector(__s1_51, __s1_51, 7, 6, 5, 4, 3, 2, 1, 0); \
1478214783 int16x4_t __rev2_51; __rev2_51 = __builtin_shufflevector(__s2_51, __s2_51, 3, 2, 1, 0); \
14783 int16x8_t __ret_51; \
1478414784 __ret_51 = __rev0_51 + __rev1_51 * __noswap_splatq_lane_s16(__rev2_51, __p3_51); \
1478514785 __ret_51 = __builtin_shufflevector(__ret_51, __ret_51, 7, 6, 5, 4, 3, 2, 1, 0); \
1478614786 __ret_51; \
......@@ -14789,22 +14789,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1478914789
1479014790#ifdef __LITTLE_ENDIAN__
1479114791#define vmla_lane_u32(__p0_52, __p1_52, __p2_52, __p3_52) __extension__ ({ \
14792 uint32x2_t __ret_52; \
1479214793 uint32x2_t __s0_52 = __p0_52; \
1479314794 uint32x2_t __s1_52 = __p1_52; \
1479414795 uint32x2_t __s2_52 = __p2_52; \
14795 uint32x2_t __ret_52; \
1479614796 __ret_52 = __s0_52 + __s1_52 * splat_lane_u32(__s2_52, __p3_52); \
1479714797 __ret_52; \
1479814798})
1479914799#else
1480014800#define vmla_lane_u32(__p0_53, __p1_53, __p2_53, __p3_53) __extension__ ({ \
14801 uint32x2_t __ret_53; \
1480114802 uint32x2_t __s0_53 = __p0_53; \
1480214803 uint32x2_t __s1_53 = __p1_53; \
1480314804 uint32x2_t __s2_53 = __p2_53; \
1480414805 uint32x2_t __rev0_53; __rev0_53 = __builtin_shufflevector(__s0_53, __s0_53, 1, 0); \
1480514806 uint32x2_t __rev1_53; __rev1_53 = __builtin_shufflevector(__s1_53, __s1_53, 1, 0); \
1480614807 uint32x2_t __rev2_53; __rev2_53 = __builtin_shufflevector(__s2_53, __s2_53, 1, 0); \
14807 uint32x2_t __ret_53; \
1480814808 __ret_53 = __rev0_53 + __rev1_53 * __noswap_splat_lane_u32(__rev2_53, __p3_53); \
1480914809 __ret_53 = __builtin_shufflevector(__ret_53, __ret_53, 1, 0); \
1481014810 __ret_53; \
......@@ -14813,22 +14813,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1481314813
1481414814#ifdef __LITTLE_ENDIAN__
1481514815#define vmla_lane_u16(__p0_54, __p1_54, __p2_54, __p3_54) __extension__ ({ \
14816 uint16x4_t __ret_54; \
1481614817 uint16x4_t __s0_54 = __p0_54; \
1481714818 uint16x4_t __s1_54 = __p1_54; \
1481814819 uint16x4_t __s2_54 = __p2_54; \
14819 uint16x4_t __ret_54; \
1482014820 __ret_54 = __s0_54 + __s1_54 * splat_lane_u16(__s2_54, __p3_54); \
1482114821 __ret_54; \
1482214822})
1482314823#else
1482414824#define vmla_lane_u16(__p0_55, __p1_55, __p2_55, __p3_55) __extension__ ({ \
14825 uint16x4_t __ret_55; \
1482514826 uint16x4_t __s0_55 = __p0_55; \
1482614827 uint16x4_t __s1_55 = __p1_55; \
1482714828 uint16x4_t __s2_55 = __p2_55; \
1482814829 uint16x4_t __rev0_55; __rev0_55 = __builtin_shufflevector(__s0_55, __s0_55, 3, 2, 1, 0); \
1482914830 uint16x4_t __rev1_55; __rev1_55 = __builtin_shufflevector(__s1_55, __s1_55, 3, 2, 1, 0); \
1483014831 uint16x4_t __rev2_55; __rev2_55 = __builtin_shufflevector(__s2_55, __s2_55, 3, 2, 1, 0); \
14831 uint16x4_t __ret_55; \
1483214832 __ret_55 = __rev0_55 + __rev1_55 * __noswap_splat_lane_u16(__rev2_55, __p3_55); \
1483314833 __ret_55 = __builtin_shufflevector(__ret_55, __ret_55, 3, 2, 1, 0); \
1483414834 __ret_55; \
......@@ -14837,22 +14837,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1483714837
1483814838#ifdef __LITTLE_ENDIAN__
1483914839#define vmla_lane_f32(__p0_56, __p1_56, __p2_56, __p3_56) __extension__ ({ \
14840 float32x2_t __ret_56; \
1484014841 float32x2_t __s0_56 = __p0_56; \
1484114842 float32x2_t __s1_56 = __p1_56; \
1484214843 float32x2_t __s2_56 = __p2_56; \
14843 float32x2_t __ret_56; \
1484414844 __ret_56 = __s0_56 + __s1_56 * splat_lane_f32(__s2_56, __p3_56); \
1484514845 __ret_56; \
1484614846})
1484714847#else
1484814848#define vmla_lane_f32(__p0_57, __p1_57, __p2_57, __p3_57) __extension__ ({ \
14849 float32x2_t __ret_57; \
1484914850 float32x2_t __s0_57 = __p0_57; \
1485014851 float32x2_t __s1_57 = __p1_57; \
1485114852 float32x2_t __s2_57 = __p2_57; \
1485214853 float32x2_t __rev0_57; __rev0_57 = __builtin_shufflevector(__s0_57, __s0_57, 1, 0); \
1485314854 float32x2_t __rev1_57; __rev1_57 = __builtin_shufflevector(__s1_57, __s1_57, 1, 0); \
1485414855 float32x2_t __rev2_57; __rev2_57 = __builtin_shufflevector(__s2_57, __s2_57, 1, 0); \
14855 float32x2_t __ret_57; \
1485614856 __ret_57 = __rev0_57 + __rev1_57 * __noswap_splat_lane_f32(__rev2_57, __p3_57); \
1485714857 __ret_57 = __builtin_shufflevector(__ret_57, __ret_57, 1, 0); \
1485814858 __ret_57; \
......@@ -14861,22 +14861,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1486114861
1486214862#ifdef __LITTLE_ENDIAN__
1486314863#define vmla_lane_s32(__p0_58, __p1_58, __p2_58, __p3_58) __extension__ ({ \
14864 int32x2_t __ret_58; \
1486414865 int32x2_t __s0_58 = __p0_58; \
1486514866 int32x2_t __s1_58 = __p1_58; \
1486614867 int32x2_t __s2_58 = __p2_58; \
14867 int32x2_t __ret_58; \
1486814868 __ret_58 = __s0_58 + __s1_58 * splat_lane_s32(__s2_58, __p3_58); \
1486914869 __ret_58; \
1487014870})
1487114871#else
1487214872#define vmla_lane_s32(__p0_59, __p1_59, __p2_59, __p3_59) __extension__ ({ \
14873 int32x2_t __ret_59; \
1487314874 int32x2_t __s0_59 = __p0_59; \
1487414875 int32x2_t __s1_59 = __p1_59; \
1487514876 int32x2_t __s2_59 = __p2_59; \
1487614877 int32x2_t __rev0_59; __rev0_59 = __builtin_shufflevector(__s0_59, __s0_59, 1, 0); \
1487714878 int32x2_t __rev1_59; __rev1_59 = __builtin_shufflevector(__s1_59, __s1_59, 1, 0); \
1487814879 int32x2_t __rev2_59; __rev2_59 = __builtin_shufflevector(__s2_59, __s2_59, 1, 0); \
14879 int32x2_t __ret_59; \
1488014880 __ret_59 = __rev0_59 + __rev1_59 * __noswap_splat_lane_s32(__rev2_59, __p3_59); \
1488114881 __ret_59 = __builtin_shufflevector(__ret_59, __ret_59, 1, 0); \
1488214882 __ret_59; \
......@@ -14885,22 +14885,22 @@ __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1488514885
1488614886#ifdef __LITTLE_ENDIAN__
1488714887#define vmla_lane_s16(__p0_60, __p1_60, __p2_60, __p3_60) __extension__ ({ \
14888 int16x4_t __ret_60; \
1488814889 int16x4_t __s0_60 = __p0_60; \
1488914890 int16x4_t __s1_60 = __p1_60; \
1489014891 int16x4_t __s2_60 = __p2_60; \
14891 int16x4_t __ret_60; \
1489214892 __ret_60 = __s0_60 + __s1_60 * splat_lane_s16(__s2_60, __p3_60); \
1489314893 __ret_60; \
1489414894})
1489514895#else
1489614896#define vmla_lane_s16(__p0_61, __p1_61, __p2_61, __p3_61) __extension__ ({ \
14897 int16x4_t __ret_61; \
1489714898 int16x4_t __s0_61 = __p0_61; \
1489814899 int16x4_t __s1_61 = __p1_61; \
1489914900 int16x4_t __s2_61 = __p2_61; \
1490014901 int16x4_t __rev0_61; __rev0_61 = __builtin_shufflevector(__s0_61, __s0_61, 3, 2, 1, 0); \
1490114902 int16x4_t __rev1_61; __rev1_61 = __builtin_shufflevector(__s1_61, __s1_61, 3, 2, 1, 0); \
1490214903 int16x4_t __rev2_61; __rev2_61 = __builtin_shufflevector(__s2_61, __s2_61, 3, 2, 1, 0); \
14903 int16x4_t __ret_61; \
1490414904 __ret_61 = __rev0_61 + __rev1_61 * __noswap_splat_lane_s16(__rev2_61, __p3_61); \
1490514905 __ret_61 = __builtin_shufflevector(__ret_61, __ret_61, 3, 2, 1, 0); \
1490614906 __ret_61; \
......@@ -14915,9 +14915,9 @@ __ai uint32x4_t vmlaq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
1491514915}
1491614916#else
1491714917__ai uint32x4_t vmlaq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
14918 uint32x4_t __ret;
1491814919 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1491914920 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14920 uint32x4_t __ret;
1492114921 __ret = __rev0 + __rev1 * (uint32x4_t) {__p2, __p2, __p2, __p2};
1492214922 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1492314923 return __ret;
......@@ -14932,9 +14932,9 @@ __ai uint16x8_t vmlaq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
1493214932}
1493314933#else
1493414934__ai uint16x8_t vmlaq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
14935 uint16x8_t __ret;
1493514936 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1493614937 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14937 uint16x8_t __ret;
1493814938 __ret = __rev0 + __rev1 * (uint16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
1493914939 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1494014940 return __ret;
......@@ -14949,9 +14949,9 @@ __ai float32x4_t vmlaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2)
1494914949}
1495014950#else
1495114951__ai float32x4_t vmlaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
14952 float32x4_t __ret;
1495214953 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1495314954 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14954 float32x4_t __ret;
1495514955 __ret = __rev0 + __rev1 * (float32x4_t) {__p2, __p2, __p2, __p2};
1495614956 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1495714957 return __ret;
......@@ -14966,9 +14966,9 @@ __ai int32x4_t vmlaq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
1496614966}
1496714967#else
1496814968__ai int32x4_t vmlaq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
14969 int32x4_t __ret;
1496914970 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1497014971 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14971 int32x4_t __ret;
1497214972 __ret = __rev0 + __rev1 * (int32x4_t) {__p2, __p2, __p2, __p2};
1497314973 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1497414974 return __ret;
......@@ -14983,9 +14983,9 @@ __ai int16x8_t vmlaq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
1498314983}
1498414984#else
1498514985__ai int16x8_t vmlaq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
14986 int16x8_t __ret;
1498614987 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1498714988 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14988 int16x8_t __ret;
1498914989 __ret = __rev0 + __rev1 * (int16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
1499014990 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1499114991 return __ret;
......@@ -15000,9 +15000,9 @@ __ai uint32x2_t vmla_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
1500015000}
1500115001#else
1500215002__ai uint32x2_t vmla_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
15003 uint32x2_t __ret;
1500315004 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1500415005 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15005 uint32x2_t __ret;
1500615006 __ret = __rev0 + __rev1 * (uint32x2_t) {__p2, __p2};
1500715007 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1500815008 return __ret;
......@@ -15017,9 +15017,9 @@ __ai uint16x4_t vmla_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
1501715017}
1501815018#else
1501915019__ai uint16x4_t vmla_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
15020 uint16x4_t __ret;
1502015021 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1502115022 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15022 uint16x4_t __ret;
1502315023 __ret = __rev0 + __rev1 * (uint16x4_t) {__p2, __p2, __p2, __p2};
1502415024 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1502515025 return __ret;
......@@ -15034,9 +15034,9 @@ __ai float32x2_t vmla_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2)
1503415034}
1503515035#else
1503615036__ai float32x2_t vmla_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
15037 float32x2_t __ret;
1503715038 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1503815039 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15039 float32x2_t __ret;
1504015040 __ret = __rev0 + __rev1 * (float32x2_t) {__p2, __p2};
1504115041 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1504215042 return __ret;
......@@ -15051,9 +15051,9 @@ __ai int32x2_t vmla_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
1505115051}
1505215052#else
1505315053__ai int32x2_t vmla_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
15054 int32x2_t __ret;
1505415055 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1505515056 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15056 int32x2_t __ret;
1505715057 __ret = __rev0 + __rev1 * (int32x2_t) {__p2, __p2};
1505815058 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1505915059 return __ret;
......@@ -15068,9 +15068,9 @@ __ai int16x4_t vmla_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
1506815068}
1506915069#else
1507015070__ai int16x4_t vmla_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
15071 int16x4_t __ret;
1507115072 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1507215073 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15073 int16x4_t __ret;
1507415074 __ret = __rev0 + __rev1 * (int16x4_t) {__p2, __p2, __p2, __p2};
1507515075 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1507615076 return __ret;
......@@ -15085,10 +15085,10 @@ __ai uint8x16_t vmlsq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
1508515085}
1508615086#else
1508715087__ai uint8x16_t vmlsq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
15088 uint8x16_t __ret;
1508815089 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1508915090 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1509015091 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15091 uint8x16_t __ret;
1509215092 __ret = __rev0 - __rev1 * __rev2;
1509315093 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1509415094 return __ret;
......@@ -15103,10 +15103,10 @@ __ai uint32x4_t vmlsq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
1510315103}
1510415104#else
1510515105__ai uint32x4_t vmlsq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
15106 uint32x4_t __ret;
1510615107 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1510715108 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1510815109 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
15109 uint32x4_t __ret;
1511015110 __ret = __rev0 - __rev1 * __rev2;
1511115111 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1511215112 return __ret;
......@@ -15121,10 +15121,10 @@ __ai uint16x8_t vmlsq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
1512115121}
1512215122#else
1512315123__ai uint16x8_t vmlsq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
15124 uint16x8_t __ret;
1512415125 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1512515126 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1512615127 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
15127 uint16x8_t __ret;
1512815128 __ret = __rev0 - __rev1 * __rev2;
1512915129 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1513015130 return __ret;
......@@ -15139,10 +15139,10 @@ __ai int8x16_t vmlsq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
1513915139}
1514015140#else
1514115141__ai int8x16_t vmlsq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
15142 int8x16_t __ret;
1514215143 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1514315144 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1514415145 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15145 int8x16_t __ret;
1514615146 __ret = __rev0 - __rev1 * __rev2;
1514715147 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1514815148 return __ret;
......@@ -15157,10 +15157,10 @@ __ai float32x4_t vmlsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2)
1515715157}
1515815158#else
1515915159__ai float32x4_t vmlsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
15160 float32x4_t __ret;
1516015161 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1516115162 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1516215163 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
15163 float32x4_t __ret;
1516415164 __ret = __rev0 - __rev1 * __rev2;
1516515165 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1516615166 return __ret;
......@@ -15175,10 +15175,10 @@ __ai int32x4_t vmlsq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
1517515175}
1517615176#else
1517715177__ai int32x4_t vmlsq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
15178 int32x4_t __ret;
1517815179 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1517915180 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1518015181 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
15181 int32x4_t __ret;
1518215182 __ret = __rev0 - __rev1 * __rev2;
1518315183 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1518415184 return __ret;
......@@ -15193,10 +15193,10 @@ __ai int16x8_t vmlsq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
1519315193}
1519415194#else
1519515195__ai int16x8_t vmlsq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
15196 int16x8_t __ret;
1519615197 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1519715198 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1519815199 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
15199 int16x8_t __ret;
1520015200 __ret = __rev0 - __rev1 * __rev2;
1520115201 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1520215202 return __ret;
......@@ -15211,10 +15211,10 @@ __ai uint8x8_t vmls_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
1521115211}
1521215212#else
1521315213__ai uint8x8_t vmls_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
15214 uint8x8_t __ret;
1521415215 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1521515216 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1521615217 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
15217 uint8x8_t __ret;
1521815218 __ret = __rev0 - __rev1 * __rev2;
1521915219 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1522015220 return __ret;
......@@ -15229,10 +15229,10 @@ __ai uint32x2_t vmls_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
1522915229}
1523015230#else
1523115231__ai uint32x2_t vmls_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
15232 uint32x2_t __ret;
1523215233 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1523315234 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1523415235 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
15235 uint32x2_t __ret;
1523615236 __ret = __rev0 - __rev1 * __rev2;
1523715237 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1523815238 return __ret;
......@@ -15247,10 +15247,10 @@ __ai uint16x4_t vmls_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
1524715247}
1524815248#else
1524915249__ai uint16x4_t vmls_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
15250 uint16x4_t __ret;
1525015251 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1525115252 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1525215253 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
15253 uint16x4_t __ret;
1525415254 __ret = __rev0 - __rev1 * __rev2;
1525515255 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1525615256 return __ret;
......@@ -15265,10 +15265,10 @@ __ai int8x8_t vmls_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
1526515265}
1526615266#else
1526715267__ai int8x8_t vmls_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
15268 int8x8_t __ret;
1526815269 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1526915270 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1527015271 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
15271 int8x8_t __ret;
1527215272 __ret = __rev0 - __rev1 * __rev2;
1527315273 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1527415274 return __ret;
......@@ -15283,10 +15283,10 @@ __ai float32x2_t vmls_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2)
1528315283}
1528415284#else
1528515285__ai float32x2_t vmls_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
15286 float32x2_t __ret;
1528615287 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1528715288 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1528815289 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
15289 float32x2_t __ret;
1529015290 __ret = __rev0 - __rev1 * __rev2;
1529115291 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1529215292 return __ret;
......@@ -15301,10 +15301,10 @@ __ai int32x2_t vmls_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
1530115301}
1530215302#else
1530315303__ai int32x2_t vmls_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
15304 int32x2_t __ret;
1530415305 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1530515306 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1530615307 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
15307 int32x2_t __ret;
1530815308 __ret = __rev0 - __rev1 * __rev2;
1530915309 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1531015310 return __ret;
......@@ -15319,10 +15319,10 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1531915319}
1532015320#else
1532115321__ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
15322 int16x4_t __ret;
1532215323 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1532315324 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1532415325 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
15325 int16x4_t __ret;
1532615326 __ret = __rev0 - __rev1 * __rev2;
1532715327 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1532815328 return __ret;
......@@ -15331,22 +15331,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1533115331
1533215332#ifdef __LITTLE_ENDIAN__
1533315333#define vmlsq_lane_u32(__p0_62, __p1_62, __p2_62, __p3_62) __extension__ ({ \
15334 uint32x4_t __ret_62; \
1533415335 uint32x4_t __s0_62 = __p0_62; \
1533515336 uint32x4_t __s1_62 = __p1_62; \
1533615337 uint32x2_t __s2_62 = __p2_62; \
15337 uint32x4_t __ret_62; \
1533815338 __ret_62 = __s0_62 - __s1_62 * splatq_lane_u32(__s2_62, __p3_62); \
1533915339 __ret_62; \
1534015340})
1534115341#else
1534215342#define vmlsq_lane_u32(__p0_63, __p1_63, __p2_63, __p3_63) __extension__ ({ \
15343 uint32x4_t __ret_63; \
1534315344 uint32x4_t __s0_63 = __p0_63; \
1534415345 uint32x4_t __s1_63 = __p1_63; \
1534515346 uint32x2_t __s2_63 = __p2_63; \
1534615347 uint32x4_t __rev0_63; __rev0_63 = __builtin_shufflevector(__s0_63, __s0_63, 3, 2, 1, 0); \
1534715348 uint32x4_t __rev1_63; __rev1_63 = __builtin_shufflevector(__s1_63, __s1_63, 3, 2, 1, 0); \
1534815349 uint32x2_t __rev2_63; __rev2_63 = __builtin_shufflevector(__s2_63, __s2_63, 1, 0); \
15349 uint32x4_t __ret_63; \
1535015350 __ret_63 = __rev0_63 - __rev1_63 * __noswap_splatq_lane_u32(__rev2_63, __p3_63); \
1535115351 __ret_63 = __builtin_shufflevector(__ret_63, __ret_63, 3, 2, 1, 0); \
1535215352 __ret_63; \
......@@ -15355,22 +15355,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1535515355
1535615356#ifdef __LITTLE_ENDIAN__
1535715357#define vmlsq_lane_u16(__p0_64, __p1_64, __p2_64, __p3_64) __extension__ ({ \
15358 uint16x8_t __ret_64; \
1535815359 uint16x8_t __s0_64 = __p0_64; \
1535915360 uint16x8_t __s1_64 = __p1_64; \
1536015361 uint16x4_t __s2_64 = __p2_64; \
15361 uint16x8_t __ret_64; \
1536215362 __ret_64 = __s0_64 - __s1_64 * splatq_lane_u16(__s2_64, __p3_64); \
1536315363 __ret_64; \
1536415364})
1536515365#else
1536615366#define vmlsq_lane_u16(__p0_65, __p1_65, __p2_65, __p3_65) __extension__ ({ \
15367 uint16x8_t __ret_65; \
1536715368 uint16x8_t __s0_65 = __p0_65; \
1536815369 uint16x8_t __s1_65 = __p1_65; \
1536915370 uint16x4_t __s2_65 = __p2_65; \
1537015371 uint16x8_t __rev0_65; __rev0_65 = __builtin_shufflevector(__s0_65, __s0_65, 7, 6, 5, 4, 3, 2, 1, 0); \
1537115372 uint16x8_t __rev1_65; __rev1_65 = __builtin_shufflevector(__s1_65, __s1_65, 7, 6, 5, 4, 3, 2, 1, 0); \
1537215373 uint16x4_t __rev2_65; __rev2_65 = __builtin_shufflevector(__s2_65, __s2_65, 3, 2, 1, 0); \
15373 uint16x8_t __ret_65; \
1537415374 __ret_65 = __rev0_65 - __rev1_65 * __noswap_splatq_lane_u16(__rev2_65, __p3_65); \
1537515375 __ret_65 = __builtin_shufflevector(__ret_65, __ret_65, 7, 6, 5, 4, 3, 2, 1, 0); \
1537615376 __ret_65; \
......@@ -15379,22 +15379,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1537915379
1538015380#ifdef __LITTLE_ENDIAN__
1538115381#define vmlsq_lane_f32(__p0_66, __p1_66, __p2_66, __p3_66) __extension__ ({ \
15382 float32x4_t __ret_66; \
1538215383 float32x4_t __s0_66 = __p0_66; \
1538315384 float32x4_t __s1_66 = __p1_66; \
1538415385 float32x2_t __s2_66 = __p2_66; \
15385 float32x4_t __ret_66; \
1538615386 __ret_66 = __s0_66 - __s1_66 * splatq_lane_f32(__s2_66, __p3_66); \
1538715387 __ret_66; \
1538815388})
1538915389#else
1539015390#define vmlsq_lane_f32(__p0_67, __p1_67, __p2_67, __p3_67) __extension__ ({ \
15391 float32x4_t __ret_67; \
1539115392 float32x4_t __s0_67 = __p0_67; \
1539215393 float32x4_t __s1_67 = __p1_67; \
1539315394 float32x2_t __s2_67 = __p2_67; \
1539415395 float32x4_t __rev0_67; __rev0_67 = __builtin_shufflevector(__s0_67, __s0_67, 3, 2, 1, 0); \
1539515396 float32x4_t __rev1_67; __rev1_67 = __builtin_shufflevector(__s1_67, __s1_67, 3, 2, 1, 0); \
1539615397 float32x2_t __rev2_67; __rev2_67 = __builtin_shufflevector(__s2_67, __s2_67, 1, 0); \
15397 float32x4_t __ret_67; \
1539815398 __ret_67 = __rev0_67 - __rev1_67 * __noswap_splatq_lane_f32(__rev2_67, __p3_67); \
1539915399 __ret_67 = __builtin_shufflevector(__ret_67, __ret_67, 3, 2, 1, 0); \
1540015400 __ret_67; \
......@@ -15403,22 +15403,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1540315403
1540415404#ifdef __LITTLE_ENDIAN__
1540515405#define vmlsq_lane_s32(__p0_68, __p1_68, __p2_68, __p3_68) __extension__ ({ \
15406 int32x4_t __ret_68; \
1540615407 int32x4_t __s0_68 = __p0_68; \
1540715408 int32x4_t __s1_68 = __p1_68; \
1540815409 int32x2_t __s2_68 = __p2_68; \
15409 int32x4_t __ret_68; \
1541015410 __ret_68 = __s0_68 - __s1_68 * splatq_lane_s32(__s2_68, __p3_68); \
1541115411 __ret_68; \
1541215412})
1541315413#else
1541415414#define vmlsq_lane_s32(__p0_69, __p1_69, __p2_69, __p3_69) __extension__ ({ \
15415 int32x4_t __ret_69; \
1541515416 int32x4_t __s0_69 = __p0_69; \
1541615417 int32x4_t __s1_69 = __p1_69; \
1541715418 int32x2_t __s2_69 = __p2_69; \
1541815419 int32x4_t __rev0_69; __rev0_69 = __builtin_shufflevector(__s0_69, __s0_69, 3, 2, 1, 0); \
1541915420 int32x4_t __rev1_69; __rev1_69 = __builtin_shufflevector(__s1_69, __s1_69, 3, 2, 1, 0); \
1542015421 int32x2_t __rev2_69; __rev2_69 = __builtin_shufflevector(__s2_69, __s2_69, 1, 0); \
15421 int32x4_t __ret_69; \
1542215422 __ret_69 = __rev0_69 - __rev1_69 * __noswap_splatq_lane_s32(__rev2_69, __p3_69); \
1542315423 __ret_69 = __builtin_shufflevector(__ret_69, __ret_69, 3, 2, 1, 0); \
1542415424 __ret_69; \
......@@ -15427,22 +15427,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1542715427
1542815428#ifdef __LITTLE_ENDIAN__
1542915429#define vmlsq_lane_s16(__p0_70, __p1_70, __p2_70, __p3_70) __extension__ ({ \
15430 int16x8_t __ret_70; \
1543015431 int16x8_t __s0_70 = __p0_70; \
1543115432 int16x8_t __s1_70 = __p1_70; \
1543215433 int16x4_t __s2_70 = __p2_70; \
15433 int16x8_t __ret_70; \
1543415434 __ret_70 = __s0_70 - __s1_70 * splatq_lane_s16(__s2_70, __p3_70); \
1543515435 __ret_70; \
1543615436})
1543715437#else
1543815438#define vmlsq_lane_s16(__p0_71, __p1_71, __p2_71, __p3_71) __extension__ ({ \
15439 int16x8_t __ret_71; \
1543915440 int16x8_t __s0_71 = __p0_71; \
1544015441 int16x8_t __s1_71 = __p1_71; \
1544115442 int16x4_t __s2_71 = __p2_71; \
1544215443 int16x8_t __rev0_71; __rev0_71 = __builtin_shufflevector(__s0_71, __s0_71, 7, 6, 5, 4, 3, 2, 1, 0); \
1544315444 int16x8_t __rev1_71; __rev1_71 = __builtin_shufflevector(__s1_71, __s1_71, 7, 6, 5, 4, 3, 2, 1, 0); \
1544415445 int16x4_t __rev2_71; __rev2_71 = __builtin_shufflevector(__s2_71, __s2_71, 3, 2, 1, 0); \
15445 int16x8_t __ret_71; \
1544615446 __ret_71 = __rev0_71 - __rev1_71 * __noswap_splatq_lane_s16(__rev2_71, __p3_71); \
1544715447 __ret_71 = __builtin_shufflevector(__ret_71, __ret_71, 7, 6, 5, 4, 3, 2, 1, 0); \
1544815448 __ret_71; \
......@@ -15451,22 +15451,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1545115451
1545215452#ifdef __LITTLE_ENDIAN__
1545315453#define vmls_lane_u32(__p0_72, __p1_72, __p2_72, __p3_72) __extension__ ({ \
15454 uint32x2_t __ret_72; \
1545415455 uint32x2_t __s0_72 = __p0_72; \
1545515456 uint32x2_t __s1_72 = __p1_72; \
1545615457 uint32x2_t __s2_72 = __p2_72; \
15457 uint32x2_t __ret_72; \
1545815458 __ret_72 = __s0_72 - __s1_72 * splat_lane_u32(__s2_72, __p3_72); \
1545915459 __ret_72; \
1546015460})
1546115461#else
1546215462#define vmls_lane_u32(__p0_73, __p1_73, __p2_73, __p3_73) __extension__ ({ \
15463 uint32x2_t __ret_73; \
1546315464 uint32x2_t __s0_73 = __p0_73; \
1546415465 uint32x2_t __s1_73 = __p1_73; \
1546515466 uint32x2_t __s2_73 = __p2_73; \
1546615467 uint32x2_t __rev0_73; __rev0_73 = __builtin_shufflevector(__s0_73, __s0_73, 1, 0); \
1546715468 uint32x2_t __rev1_73; __rev1_73 = __builtin_shufflevector(__s1_73, __s1_73, 1, 0); \
1546815469 uint32x2_t __rev2_73; __rev2_73 = __builtin_shufflevector(__s2_73, __s2_73, 1, 0); \
15469 uint32x2_t __ret_73; \
1547015470 __ret_73 = __rev0_73 - __rev1_73 * __noswap_splat_lane_u32(__rev2_73, __p3_73); \
1547115471 __ret_73 = __builtin_shufflevector(__ret_73, __ret_73, 1, 0); \
1547215472 __ret_73; \
......@@ -15475,22 +15475,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1547515475
1547615476#ifdef __LITTLE_ENDIAN__
1547715477#define vmls_lane_u16(__p0_74, __p1_74, __p2_74, __p3_74) __extension__ ({ \
15478 uint16x4_t __ret_74; \
1547815479 uint16x4_t __s0_74 = __p0_74; \
1547915480 uint16x4_t __s1_74 = __p1_74; \
1548015481 uint16x4_t __s2_74 = __p2_74; \
15481 uint16x4_t __ret_74; \
1548215482 __ret_74 = __s0_74 - __s1_74 * splat_lane_u16(__s2_74, __p3_74); \
1548315483 __ret_74; \
1548415484})
1548515485#else
1548615486#define vmls_lane_u16(__p0_75, __p1_75, __p2_75, __p3_75) __extension__ ({ \
15487 uint16x4_t __ret_75; \
1548715488 uint16x4_t __s0_75 = __p0_75; \
1548815489 uint16x4_t __s1_75 = __p1_75; \
1548915490 uint16x4_t __s2_75 = __p2_75; \
1549015491 uint16x4_t __rev0_75; __rev0_75 = __builtin_shufflevector(__s0_75, __s0_75, 3, 2, 1, 0); \
1549115492 uint16x4_t __rev1_75; __rev1_75 = __builtin_shufflevector(__s1_75, __s1_75, 3, 2, 1, 0); \
1549215493 uint16x4_t __rev2_75; __rev2_75 = __builtin_shufflevector(__s2_75, __s2_75, 3, 2, 1, 0); \
15493 uint16x4_t __ret_75; \
1549415494 __ret_75 = __rev0_75 - __rev1_75 * __noswap_splat_lane_u16(__rev2_75, __p3_75); \
1549515495 __ret_75 = __builtin_shufflevector(__ret_75, __ret_75, 3, 2, 1, 0); \
1549615496 __ret_75; \
......@@ -15499,22 +15499,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1549915499
1550015500#ifdef __LITTLE_ENDIAN__
1550115501#define vmls_lane_f32(__p0_76, __p1_76, __p2_76, __p3_76) __extension__ ({ \
15502 float32x2_t __ret_76; \
1550215503 float32x2_t __s0_76 = __p0_76; \
1550315504 float32x2_t __s1_76 = __p1_76; \
1550415505 float32x2_t __s2_76 = __p2_76; \
15505 float32x2_t __ret_76; \
1550615506 __ret_76 = __s0_76 - __s1_76 * splat_lane_f32(__s2_76, __p3_76); \
1550715507 __ret_76; \
1550815508})
1550915509#else
1551015510#define vmls_lane_f32(__p0_77, __p1_77, __p2_77, __p3_77) __extension__ ({ \
15511 float32x2_t __ret_77; \
1551115512 float32x2_t __s0_77 = __p0_77; \
1551215513 float32x2_t __s1_77 = __p1_77; \
1551315514 float32x2_t __s2_77 = __p2_77; \
1551415515 float32x2_t __rev0_77; __rev0_77 = __builtin_shufflevector(__s0_77, __s0_77, 1, 0); \
1551515516 float32x2_t __rev1_77; __rev1_77 = __builtin_shufflevector(__s1_77, __s1_77, 1, 0); \
1551615517 float32x2_t __rev2_77; __rev2_77 = __builtin_shufflevector(__s2_77, __s2_77, 1, 0); \
15517 float32x2_t __ret_77; \
1551815518 __ret_77 = __rev0_77 - __rev1_77 * __noswap_splat_lane_f32(__rev2_77, __p3_77); \
1551915519 __ret_77 = __builtin_shufflevector(__ret_77, __ret_77, 1, 0); \
1552015520 __ret_77; \
......@@ -15523,22 +15523,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1552315523
1552415524#ifdef __LITTLE_ENDIAN__
1552515525#define vmls_lane_s32(__p0_78, __p1_78, __p2_78, __p3_78) __extension__ ({ \
15526 int32x2_t __ret_78; \
1552615527 int32x2_t __s0_78 = __p0_78; \
1552715528 int32x2_t __s1_78 = __p1_78; \
1552815529 int32x2_t __s2_78 = __p2_78; \
15529 int32x2_t __ret_78; \
1553015530 __ret_78 = __s0_78 - __s1_78 * splat_lane_s32(__s2_78, __p3_78); \
1553115531 __ret_78; \
1553215532})
1553315533#else
1553415534#define vmls_lane_s32(__p0_79, __p1_79, __p2_79, __p3_79) __extension__ ({ \
15535 int32x2_t __ret_79; \
1553515536 int32x2_t __s0_79 = __p0_79; \
1553615537 int32x2_t __s1_79 = __p1_79; \
1553715538 int32x2_t __s2_79 = __p2_79; \
1553815539 int32x2_t __rev0_79; __rev0_79 = __builtin_shufflevector(__s0_79, __s0_79, 1, 0); \
1553915540 int32x2_t __rev1_79; __rev1_79 = __builtin_shufflevector(__s1_79, __s1_79, 1, 0); \
1554015541 int32x2_t __rev2_79; __rev2_79 = __builtin_shufflevector(__s2_79, __s2_79, 1, 0); \
15541 int32x2_t __ret_79; \
1554215542 __ret_79 = __rev0_79 - __rev1_79 * __noswap_splat_lane_s32(__rev2_79, __p3_79); \
1554315543 __ret_79 = __builtin_shufflevector(__ret_79, __ret_79, 1, 0); \
1554415544 __ret_79; \
......@@ -15547,22 +15547,22 @@ __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1554715547
1554815548#ifdef __LITTLE_ENDIAN__
1554915549#define vmls_lane_s16(__p0_80, __p1_80, __p2_80, __p3_80) __extension__ ({ \
15550 int16x4_t __ret_80; \
1555015551 int16x4_t __s0_80 = __p0_80; \
1555115552 int16x4_t __s1_80 = __p1_80; \
1555215553 int16x4_t __s2_80 = __p2_80; \
15553 int16x4_t __ret_80; \
1555415554 __ret_80 = __s0_80 - __s1_80 * splat_lane_s16(__s2_80, __p3_80); \
1555515555 __ret_80; \
1555615556})
1555715557#else
1555815558#define vmls_lane_s16(__p0_81, __p1_81, __p2_81, __p3_81) __extension__ ({ \
15559 int16x4_t __ret_81; \
1555915560 int16x4_t __s0_81 = __p0_81; \
1556015561 int16x4_t __s1_81 = __p1_81; \
1556115562 int16x4_t __s2_81 = __p2_81; \
1556215563 int16x4_t __rev0_81; __rev0_81 = __builtin_shufflevector(__s0_81, __s0_81, 3, 2, 1, 0); \
1556315564 int16x4_t __rev1_81; __rev1_81 = __builtin_shufflevector(__s1_81, __s1_81, 3, 2, 1, 0); \
1556415565 int16x4_t __rev2_81; __rev2_81 = __builtin_shufflevector(__s2_81, __s2_81, 3, 2, 1, 0); \
15565 int16x4_t __ret_81; \
1556615566 __ret_81 = __rev0_81 - __rev1_81 * __noswap_splat_lane_s16(__rev2_81, __p3_81); \
1556715567 __ret_81 = __builtin_shufflevector(__ret_81, __ret_81, 3, 2, 1, 0); \
1556815568 __ret_81; \
......@@ -15577,9 +15577,9 @@ __ai uint32x4_t vmlsq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
1557715577}
1557815578#else
1557915579__ai uint32x4_t vmlsq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
15580 uint32x4_t __ret;
1558015581 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1558115582 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15582 uint32x4_t __ret;
1558315583 __ret = __rev0 - __rev1 * (uint32x4_t) {__p2, __p2, __p2, __p2};
1558415584 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1558515585 return __ret;
......@@ -15594,9 +15594,9 @@ __ai uint16x8_t vmlsq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
1559415594}
1559515595#else
1559615596__ai uint16x8_t vmlsq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
15597 uint16x8_t __ret;
1559715598 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1559815599 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15599 uint16x8_t __ret;
1560015600 __ret = __rev0 - __rev1 * (uint16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
1560115601 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1560215602 return __ret;
......@@ -15611,9 +15611,9 @@ __ai float32x4_t vmlsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2)
1561115611}
1561215612#else
1561315613__ai float32x4_t vmlsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
15614 float32x4_t __ret;
1561415615 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1561515616 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15616 float32x4_t __ret;
1561715617 __ret = __rev0 - __rev1 * (float32x4_t) {__p2, __p2, __p2, __p2};
1561815618 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1561915619 return __ret;
......@@ -15628,9 +15628,9 @@ __ai int32x4_t vmlsq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
1562815628}
1562915629#else
1563015630__ai int32x4_t vmlsq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
15631 int32x4_t __ret;
1563115632 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1563215633 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15633 int32x4_t __ret;
1563415634 __ret = __rev0 - __rev1 * (int32x4_t) {__p2, __p2, __p2, __p2};
1563515635 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1563615636 return __ret;
......@@ -15645,9 +15645,9 @@ __ai int16x8_t vmlsq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
1564515645}
1564615646#else
1564715647__ai int16x8_t vmlsq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
15648 int16x8_t __ret;
1564815649 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1564915650 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15650 int16x8_t __ret;
1565115651 __ret = __rev0 - __rev1 * (int16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
1565215652 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1565315653 return __ret;
......@@ -15662,9 +15662,9 @@ __ai uint32x2_t vmls_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
1566215662}
1566315663#else
1566415664__ai uint32x2_t vmls_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
15665 uint32x2_t __ret;
1566515666 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1566615667 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15667 uint32x2_t __ret;
1566815668 __ret = __rev0 - __rev1 * (uint32x2_t) {__p2, __p2};
1566915669 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1567015670 return __ret;
......@@ -15679,9 +15679,9 @@ __ai uint16x4_t vmls_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
1567915679}
1568015680#else
1568115681__ai uint16x4_t vmls_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
15682 uint16x4_t __ret;
1568215683 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1568315684 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15684 uint16x4_t __ret;
1568515685 __ret = __rev0 - __rev1 * (uint16x4_t) {__p2, __p2, __p2, __p2};
1568615686 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1568715687 return __ret;
......@@ -15696,9 +15696,9 @@ __ai float32x2_t vmls_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2)
1569615696}
1569715697#else
1569815698__ai float32x2_t vmls_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
15699 float32x2_t __ret;
1569915700 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1570015701 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15701 float32x2_t __ret;
1570215702 __ret = __rev0 - __rev1 * (float32x2_t) {__p2, __p2};
1570315703 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1570415704 return __ret;
......@@ -15713,9 +15713,9 @@ __ai int32x2_t vmls_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
1571315713}
1571415714#else
1571515715__ai int32x2_t vmls_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
15716 int32x2_t __ret;
1571615717 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1571715718 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15718 int32x2_t __ret;
1571915719 __ret = __rev0 - __rev1 * (int32x2_t) {__p2, __p2};
1572015720 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1572115721 return __ret;
......@@ -15730,9 +15730,9 @@ __ai int16x4_t vmls_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
1573015730}
1573115731#else
1573215732__ai int16x4_t vmls_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
15733 int16x4_t __ret;
1573315734 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1573415735 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15735 int16x4_t __ret;
1573615736 __ret = __rev0 - __rev1 * (int16x4_t) {__p2, __p2, __p2, __p2};
1573715737 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1573815738 return __ret;
......@@ -15891,15 +15891,15 @@ __ai float32x4_t vmovq_n_f32(float32_t __p0) {
1589115891
1589215892#ifdef __LITTLE_ENDIAN__
1589315893#define vmovq_n_f16(__p0) __extension__ ({ \
15894 float16_t __s0 = __p0; \
1589515894 float16x8_t __ret; \
15895 float16_t __s0 = __p0; \
1589615896 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
1589715897 __ret; \
1589815898})
1589915899#else
1590015900#define vmovq_n_f16(__p0) __extension__ ({ \
15901 float16_t __s0 = __p0; \
1590215901 float16x8_t __ret; \
15902 float16_t __s0 = __p0; \
1590315903 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
1590415904 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
1590515905 __ret; \
......@@ -16033,15 +16033,15 @@ __ai float32x2_t vmov_n_f32(float32_t __p0) {
1603316033
1603416034#ifdef __LITTLE_ENDIAN__
1603516035#define vmov_n_f16(__p0) __extension__ ({ \
16036 float16_t __s0 = __p0; \
1603716036 float16x4_t __ret; \
16037 float16_t __s0 = __p0; \
1603816038 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
1603916039 __ret; \
1604016040})
1604116041#else
1604216042#define vmov_n_f16(__p0) __extension__ ({ \
16043 float16_t __s0 = __p0; \
1604416043 float16x4_t __ret; \
16044 float16_t __s0 = __p0; \
1604516045 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
1604616046 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
1604716047 __ret; \
......@@ -16091,8 +16091,8 @@ __ai uint16x8_t vmovl_u8(uint8x8_t __p0) {
1609116091}
1609216092#else
1609316093__ai uint16x8_t vmovl_u8(uint8x8_t __p0) {
16094 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1609516094 uint16x8_t __ret;
16095 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1609616096 __ret = (uint16x8_t) __builtin_neon_vmovl_v((int8x8_t)__rev0, 49);
1609716097 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1609816098 return __ret;
......@@ -16112,8 +16112,8 @@ __ai uint64x2_t vmovl_u32(uint32x2_t __p0) {
1611216112}
1611316113#else
1611416114__ai uint64x2_t vmovl_u32(uint32x2_t __p0) {
16115 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1611616115 uint64x2_t __ret;
16116 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1611716117 __ret = (uint64x2_t) __builtin_neon_vmovl_v((int8x8_t)__rev0, 51);
1611816118 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1611916119 return __ret;
......@@ -16133,8 +16133,8 @@ __ai uint32x4_t vmovl_u16(uint16x4_t __p0) {
1613316133}
1613416134#else
1613516135__ai uint32x4_t vmovl_u16(uint16x4_t __p0) {
16136 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1613716136 uint32x4_t __ret;
16137 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1613816138 __ret = (uint32x4_t) __builtin_neon_vmovl_v((int8x8_t)__rev0, 50);
1613916139 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1614016140 return __ret;
......@@ -16154,8 +16154,8 @@ __ai int16x8_t vmovl_s8(int8x8_t __p0) {
1615416154}
1615516155#else
1615616156__ai int16x8_t vmovl_s8(int8x8_t __p0) {
16157 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1615816157 int16x8_t __ret;
16158 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1615916159 __ret = (int16x8_t) __builtin_neon_vmovl_v((int8x8_t)__rev0, 33);
1616016160 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1616116161 return __ret;
......@@ -16175,8 +16175,8 @@ __ai int64x2_t vmovl_s32(int32x2_t __p0) {
1617516175}
1617616176#else
1617716177__ai int64x2_t vmovl_s32(int32x2_t __p0) {
16178 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1617916178 int64x2_t __ret;
16179 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1618016180 __ret = (int64x2_t) __builtin_neon_vmovl_v((int8x8_t)__rev0, 35);
1618116181 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1618216182 return __ret;
......@@ -16196,8 +16196,8 @@ __ai int32x4_t vmovl_s16(int16x4_t __p0) {
1619616196}
1619716197#else
1619816198__ai int32x4_t vmovl_s16(int16x4_t __p0) {
16199 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1620016199 int32x4_t __ret;
16200 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1620116201 __ret = (int32x4_t) __builtin_neon_vmovl_v((int8x8_t)__rev0, 34);
1620216202 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1620316203 return __ret;
......@@ -16217,8 +16217,8 @@ __ai uint16x4_t vmovn_u32(uint32x4_t __p0) {
1621716217}
1621816218#else
1621916219__ai uint16x4_t vmovn_u32(uint32x4_t __p0) {
16220 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1622116220 uint16x4_t __ret;
16221 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1622216222 __ret = (uint16x4_t) __builtin_neon_vmovn_v((int8x16_t)__rev0, 17);
1622316223 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1622416224 return __ret;
......@@ -16238,8 +16238,8 @@ __ai uint32x2_t vmovn_u64(uint64x2_t __p0) {
1623816238}
1623916239#else
1624016240__ai uint32x2_t vmovn_u64(uint64x2_t __p0) {
16241 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1624216241 uint32x2_t __ret;
16242 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1624316243 __ret = (uint32x2_t) __builtin_neon_vmovn_v((int8x16_t)__rev0, 18);
1624416244 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1624516245 return __ret;
......@@ -16259,8 +16259,8 @@ __ai uint8x8_t vmovn_u16(uint16x8_t __p0) {
1625916259}
1626016260#else
1626116261__ai uint8x8_t vmovn_u16(uint16x8_t __p0) {
16262 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1626316262 uint8x8_t __ret;
16263 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1626416264 __ret = (uint8x8_t) __builtin_neon_vmovn_v((int8x16_t)__rev0, 16);
1626516265 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1626616266 return __ret;
......@@ -16280,8 +16280,8 @@ __ai int16x4_t vmovn_s32(int32x4_t __p0) {
1628016280}
1628116281#else
1628216282__ai int16x4_t vmovn_s32(int32x4_t __p0) {
16283 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1628416283 int16x4_t __ret;
16284 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1628516285 __ret = (int16x4_t) __builtin_neon_vmovn_v((int8x16_t)__rev0, 1);
1628616286 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1628716287 return __ret;
......@@ -16301,8 +16301,8 @@ __ai int32x2_t vmovn_s64(int64x2_t __p0) {
1630116301}
1630216302#else
1630316303__ai int32x2_t vmovn_s64(int64x2_t __p0) {
16304 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1630516304 int32x2_t __ret;
16305 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1630616306 __ret = (int32x2_t) __builtin_neon_vmovn_v((int8x16_t)__rev0, 2);
1630716307 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1630816308 return __ret;
......@@ -16322,8 +16322,8 @@ __ai int8x8_t vmovn_s16(int16x8_t __p0) {
1632216322}
1632316323#else
1632416324__ai int8x8_t vmovn_s16(int16x8_t __p0) {
16325 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1632616325 int8x8_t __ret;
16326 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1632716327 __ret = (int8x8_t) __builtin_neon_vmovn_v((int8x16_t)__rev0, 0);
1632816328 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1632916329 return __ret;
......@@ -16343,9 +16343,9 @@ __ai uint8x16_t vmulq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1634316343}
1634416344#else
1634516345__ai uint8x16_t vmulq_u8(uint8x16_t __p0, uint8x16_t __p1) {
16346 uint8x16_t __ret;
1634616347 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1634716348 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
16348 uint8x16_t __ret;
1634916349 __ret = __rev0 * __rev1;
1635016350 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1635116351 return __ret;
......@@ -16360,9 +16360,9 @@ __ai uint32x4_t vmulq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1636016360}
1636116361#else
1636216362__ai uint32x4_t vmulq_u32(uint32x4_t __p0, uint32x4_t __p1) {
16363 uint32x4_t __ret;
1636316364 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1636416365 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16365 uint32x4_t __ret;
1636616366 __ret = __rev0 * __rev1;
1636716367 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1636816368 return __ret;
......@@ -16377,9 +16377,9 @@ __ai uint16x8_t vmulq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1637716377}
1637816378#else
1637916379__ai uint16x8_t vmulq_u16(uint16x8_t __p0, uint16x8_t __p1) {
16380 uint16x8_t __ret;
1638016381 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1638116382 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16382 uint16x8_t __ret;
1638316383 __ret = __rev0 * __rev1;
1638416384 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1638516385 return __ret;
......@@ -16394,9 +16394,9 @@ __ai int8x16_t vmulq_s8(int8x16_t __p0, int8x16_t __p1) {
1639416394}
1639516395#else
1639616396__ai int8x16_t vmulq_s8(int8x16_t __p0, int8x16_t __p1) {
16397 int8x16_t __ret;
1639716398 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1639816399 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
16399 int8x16_t __ret;
1640016400 __ret = __rev0 * __rev1;
1640116401 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1640216402 return __ret;
......@@ -16411,9 +16411,9 @@ __ai float32x4_t vmulq_f32(float32x4_t __p0, float32x4_t __p1) {
1641116411}
1641216412#else
1641316413__ai float32x4_t vmulq_f32(float32x4_t __p0, float32x4_t __p1) {
16414 float32x4_t __ret;
1641416415 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1641516416 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16416 float32x4_t __ret;
1641716417 __ret = __rev0 * __rev1;
1641816418 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1641916419 return __ret;
......@@ -16428,9 +16428,9 @@ __ai int32x4_t vmulq_s32(int32x4_t __p0, int32x4_t __p1) {
1642816428}
1642916429#else
1643016430__ai int32x4_t vmulq_s32(int32x4_t __p0, int32x4_t __p1) {
16431 int32x4_t __ret;
1643116432 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1643216433 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16433 int32x4_t __ret;
1643416434 __ret = __rev0 * __rev1;
1643516435 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1643616436 return __ret;
......@@ -16445,9 +16445,9 @@ __ai int16x8_t vmulq_s16(int16x8_t __p0, int16x8_t __p1) {
1644516445}
1644616446#else
1644716447__ai int16x8_t vmulq_s16(int16x8_t __p0, int16x8_t __p1) {
16448 int16x8_t __ret;
1644816449 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1644916450 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16450 int16x8_t __ret;
1645116451 __ret = __rev0 * __rev1;
1645216452 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1645316453 return __ret;
......@@ -16462,9 +16462,9 @@ __ai uint8x8_t vmul_u8(uint8x8_t __p0, uint8x8_t __p1) {
1646216462}
1646316463#else
1646416464__ai uint8x8_t vmul_u8(uint8x8_t __p0, uint8x8_t __p1) {
16465 uint8x8_t __ret;
1646516466 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1646616467 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16467 uint8x8_t __ret;
1646816468 __ret = __rev0 * __rev1;
1646916469 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1647016470 return __ret;
......@@ -16479,9 +16479,9 @@ __ai uint32x2_t vmul_u32(uint32x2_t __p0, uint32x2_t __p1) {
1647916479}
1648016480#else
1648116481__ai uint32x2_t vmul_u32(uint32x2_t __p0, uint32x2_t __p1) {
16482 uint32x2_t __ret;
1648216483 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1648316484 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16484 uint32x2_t __ret;
1648516485 __ret = __rev0 * __rev1;
1648616486 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1648716487 return __ret;
......@@ -16496,9 +16496,9 @@ __ai uint16x4_t vmul_u16(uint16x4_t __p0, uint16x4_t __p1) {
1649616496}
1649716497#else
1649816498__ai uint16x4_t vmul_u16(uint16x4_t __p0, uint16x4_t __p1) {
16499 uint16x4_t __ret;
1649916500 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1650016501 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16501 uint16x4_t __ret;
1650216502 __ret = __rev0 * __rev1;
1650316503 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1650416504 return __ret;
......@@ -16513,9 +16513,9 @@ __ai int8x8_t vmul_s8(int8x8_t __p0, int8x8_t __p1) {
1651316513}
1651416514#else
1651516515__ai int8x8_t vmul_s8(int8x8_t __p0, int8x8_t __p1) {
16516 int8x8_t __ret;
1651616517 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1651716518 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16518 int8x8_t __ret;
1651916519 __ret = __rev0 * __rev1;
1652016520 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1652116521 return __ret;
......@@ -16530,9 +16530,9 @@ __ai float32x2_t vmul_f32(float32x2_t __p0, float32x2_t __p1) {
1653016530}
1653116531#else
1653216532__ai float32x2_t vmul_f32(float32x2_t __p0, float32x2_t __p1) {
16533 float32x2_t __ret;
1653316534 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1653416535 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16535 float32x2_t __ret;
1653616536 __ret = __rev0 * __rev1;
1653716537 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1653816538 return __ret;
......@@ -16547,9 +16547,9 @@ __ai int32x2_t vmul_s32(int32x2_t __p0, int32x2_t __p1) {
1654716547}
1654816548#else
1654916549__ai int32x2_t vmul_s32(int32x2_t __p0, int32x2_t __p1) {
16550 int32x2_t __ret;
1655016551 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1655116552 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16552 int32x2_t __ret;
1655316553 __ret = __rev0 * __rev1;
1655416554 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1655516555 return __ret;
......@@ -16564,9 +16564,9 @@ __ai int16x4_t vmul_s16(int16x4_t __p0, int16x4_t __p1) {
1656416564}
1656516565#else
1656616566__ai int16x4_t vmul_s16(int16x4_t __p0, int16x4_t __p1) {
16567 int16x4_t __ret;
1656716568 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1656816569 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16569 int16x4_t __ret;
1657016570 __ret = __rev0 * __rev1;
1657116571 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1657216572 return __ret;
......@@ -16581,9 +16581,9 @@ __ai poly8x8_t vmul_p8(poly8x8_t __p0, poly8x8_t __p1) {
1658116581}
1658216582#else
1658316583__ai poly8x8_t vmul_p8(poly8x8_t __p0, poly8x8_t __p1) {
16584 poly8x8_t __ret;
1658416585 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1658516586 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16586 poly8x8_t __ret;
1658716587 __ret = (poly8x8_t) __builtin_neon_vmul_v((int8x8_t)__rev0, (int8x8_t)__rev1, 4);
1658816588 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1658916589 return __ret;
......@@ -16598,9 +16598,9 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1659816598}
1659916599#else
1660016600__ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
16601 poly8x16_t __ret;
1660116602 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1660216603 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
16603 poly8x16_t __ret;
1660416604 __ret = (poly8x16_t) __builtin_neon_vmulq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 36);
1660516605 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1660616606 return __ret;
......@@ -16609,19 +16609,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1660916609
1661016610#ifdef __LITTLE_ENDIAN__
1661116611#define vmulq_lane_u32(__p0_82, __p1_82, __p2_82) __extension__ ({ \
16612 uint32x4_t __ret_82; \
1661216613 uint32x4_t __s0_82 = __p0_82; \
1661316614 uint32x2_t __s1_82 = __p1_82; \
16614 uint32x4_t __ret_82; \
1661516615 __ret_82 = __s0_82 * splatq_lane_u32(__s1_82, __p2_82); \
1661616616 __ret_82; \
1661716617})
1661816618#else
1661916619#define vmulq_lane_u32(__p0_83, __p1_83, __p2_83) __extension__ ({ \
16620 uint32x4_t __ret_83; \
1662016621 uint32x4_t __s0_83 = __p0_83; \
1662116622 uint32x2_t __s1_83 = __p1_83; \
1662216623 uint32x4_t __rev0_83; __rev0_83 = __builtin_shufflevector(__s0_83, __s0_83, 3, 2, 1, 0); \
1662316624 uint32x2_t __rev1_83; __rev1_83 = __builtin_shufflevector(__s1_83, __s1_83, 1, 0); \
16624 uint32x4_t __ret_83; \
1662516625 __ret_83 = __rev0_83 * __noswap_splatq_lane_u32(__rev1_83, __p2_83); \
1662616626 __ret_83 = __builtin_shufflevector(__ret_83, __ret_83, 3, 2, 1, 0); \
1662716627 __ret_83; \
......@@ -16630,19 +16630,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1663016630
1663116631#ifdef __LITTLE_ENDIAN__
1663216632#define vmulq_lane_u16(__p0_84, __p1_84, __p2_84) __extension__ ({ \
16633 uint16x8_t __ret_84; \
1663316634 uint16x8_t __s0_84 = __p0_84; \
1663416635 uint16x4_t __s1_84 = __p1_84; \
16635 uint16x8_t __ret_84; \
1663616636 __ret_84 = __s0_84 * splatq_lane_u16(__s1_84, __p2_84); \
1663716637 __ret_84; \
1663816638})
1663916639#else
1664016640#define vmulq_lane_u16(__p0_85, __p1_85, __p2_85) __extension__ ({ \
16641 uint16x8_t __ret_85; \
1664116642 uint16x8_t __s0_85 = __p0_85; \
1664216643 uint16x4_t __s1_85 = __p1_85; \
1664316644 uint16x8_t __rev0_85; __rev0_85 = __builtin_shufflevector(__s0_85, __s0_85, 7, 6, 5, 4, 3, 2, 1, 0); \
1664416645 uint16x4_t __rev1_85; __rev1_85 = __builtin_shufflevector(__s1_85, __s1_85, 3, 2, 1, 0); \
16645 uint16x8_t __ret_85; \
1664616646 __ret_85 = __rev0_85 * __noswap_splatq_lane_u16(__rev1_85, __p2_85); \
1664716647 __ret_85 = __builtin_shufflevector(__ret_85, __ret_85, 7, 6, 5, 4, 3, 2, 1, 0); \
1664816648 __ret_85; \
......@@ -16651,19 +16651,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1665116651
1665216652#ifdef __LITTLE_ENDIAN__
1665316653#define vmulq_lane_f32(__p0_86, __p1_86, __p2_86) __extension__ ({ \
16654 float32x4_t __ret_86; \
1665416655 float32x4_t __s0_86 = __p0_86; \
1665516656 float32x2_t __s1_86 = __p1_86; \
16656 float32x4_t __ret_86; \
1665716657 __ret_86 = __s0_86 * splatq_lane_f32(__s1_86, __p2_86); \
1665816658 __ret_86; \
1665916659})
1666016660#else
1666116661#define vmulq_lane_f32(__p0_87, __p1_87, __p2_87) __extension__ ({ \
16662 float32x4_t __ret_87; \
1666216663 float32x4_t __s0_87 = __p0_87; \
1666316664 float32x2_t __s1_87 = __p1_87; \
1666416665 float32x4_t __rev0_87; __rev0_87 = __builtin_shufflevector(__s0_87, __s0_87, 3, 2, 1, 0); \
1666516666 float32x2_t __rev1_87; __rev1_87 = __builtin_shufflevector(__s1_87, __s1_87, 1, 0); \
16666 float32x4_t __ret_87; \
1666716667 __ret_87 = __rev0_87 * __noswap_splatq_lane_f32(__rev1_87, __p2_87); \
1666816668 __ret_87 = __builtin_shufflevector(__ret_87, __ret_87, 3, 2, 1, 0); \
1666916669 __ret_87; \
......@@ -16672,19 +16672,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1667216672
1667316673#ifdef __LITTLE_ENDIAN__
1667416674#define vmulq_lane_s32(__p0_88, __p1_88, __p2_88) __extension__ ({ \
16675 int32x4_t __ret_88; \
1667516676 int32x4_t __s0_88 = __p0_88; \
1667616677 int32x2_t __s1_88 = __p1_88; \
16677 int32x4_t __ret_88; \
1667816678 __ret_88 = __s0_88 * splatq_lane_s32(__s1_88, __p2_88); \
1667916679 __ret_88; \
1668016680})
1668116681#else
1668216682#define vmulq_lane_s32(__p0_89, __p1_89, __p2_89) __extension__ ({ \
16683 int32x4_t __ret_89; \
1668316684 int32x4_t __s0_89 = __p0_89; \
1668416685 int32x2_t __s1_89 = __p1_89; \
1668516686 int32x4_t __rev0_89; __rev0_89 = __builtin_shufflevector(__s0_89, __s0_89, 3, 2, 1, 0); \
1668616687 int32x2_t __rev1_89; __rev1_89 = __builtin_shufflevector(__s1_89, __s1_89, 1, 0); \
16687 int32x4_t __ret_89; \
1668816688 __ret_89 = __rev0_89 * __noswap_splatq_lane_s32(__rev1_89, __p2_89); \
1668916689 __ret_89 = __builtin_shufflevector(__ret_89, __ret_89, 3, 2, 1, 0); \
1669016690 __ret_89; \
......@@ -16693,19 +16693,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1669316693
1669416694#ifdef __LITTLE_ENDIAN__
1669516695#define vmulq_lane_s16(__p0_90, __p1_90, __p2_90) __extension__ ({ \
16696 int16x8_t __ret_90; \
1669616697 int16x8_t __s0_90 = __p0_90; \
1669716698 int16x4_t __s1_90 = __p1_90; \
16698 int16x8_t __ret_90; \
1669916699 __ret_90 = __s0_90 * splatq_lane_s16(__s1_90, __p2_90); \
1670016700 __ret_90; \
1670116701})
1670216702#else
1670316703#define vmulq_lane_s16(__p0_91, __p1_91, __p2_91) __extension__ ({ \
16704 int16x8_t __ret_91; \
1670416705 int16x8_t __s0_91 = __p0_91; \
1670516706 int16x4_t __s1_91 = __p1_91; \
1670616707 int16x8_t __rev0_91; __rev0_91 = __builtin_shufflevector(__s0_91, __s0_91, 7, 6, 5, 4, 3, 2, 1, 0); \
1670716708 int16x4_t __rev1_91; __rev1_91 = __builtin_shufflevector(__s1_91, __s1_91, 3, 2, 1, 0); \
16708 int16x8_t __ret_91; \
1670916709 __ret_91 = __rev0_91 * __noswap_splatq_lane_s16(__rev1_91, __p2_91); \
1671016710 __ret_91 = __builtin_shufflevector(__ret_91, __ret_91, 7, 6, 5, 4, 3, 2, 1, 0); \
1671116711 __ret_91; \
......@@ -16714,19 +16714,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1671416714
1671516715#ifdef __LITTLE_ENDIAN__
1671616716#define vmul_lane_u32(__p0_92, __p1_92, __p2_92) __extension__ ({ \
16717 uint32x2_t __ret_92; \
1671716718 uint32x2_t __s0_92 = __p0_92; \
1671816719 uint32x2_t __s1_92 = __p1_92; \
16719 uint32x2_t __ret_92; \
1672016720 __ret_92 = __s0_92 * splat_lane_u32(__s1_92, __p2_92); \
1672116721 __ret_92; \
1672216722})
1672316723#else
1672416724#define vmul_lane_u32(__p0_93, __p1_93, __p2_93) __extension__ ({ \
16725 uint32x2_t __ret_93; \
1672516726 uint32x2_t __s0_93 = __p0_93; \
1672616727 uint32x2_t __s1_93 = __p1_93; \
1672716728 uint32x2_t __rev0_93; __rev0_93 = __builtin_shufflevector(__s0_93, __s0_93, 1, 0); \
1672816729 uint32x2_t __rev1_93; __rev1_93 = __builtin_shufflevector(__s1_93, __s1_93, 1, 0); \
16729 uint32x2_t __ret_93; \
1673016730 __ret_93 = __rev0_93 * __noswap_splat_lane_u32(__rev1_93, __p2_93); \
1673116731 __ret_93 = __builtin_shufflevector(__ret_93, __ret_93, 1, 0); \
1673216732 __ret_93; \
......@@ -16735,19 +16735,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1673516735
1673616736#ifdef __LITTLE_ENDIAN__
1673716737#define vmul_lane_u16(__p0_94, __p1_94, __p2_94) __extension__ ({ \
16738 uint16x4_t __ret_94; \
1673816739 uint16x4_t __s0_94 = __p0_94; \
1673916740 uint16x4_t __s1_94 = __p1_94; \
16740 uint16x4_t __ret_94; \
1674116741 __ret_94 = __s0_94 * splat_lane_u16(__s1_94, __p2_94); \
1674216742 __ret_94; \
1674316743})
1674416744#else
1674516745#define vmul_lane_u16(__p0_95, __p1_95, __p2_95) __extension__ ({ \
16746 uint16x4_t __ret_95; \
1674616747 uint16x4_t __s0_95 = __p0_95; \
1674716748 uint16x4_t __s1_95 = __p1_95; \
1674816749 uint16x4_t __rev0_95; __rev0_95 = __builtin_shufflevector(__s0_95, __s0_95, 3, 2, 1, 0); \
1674916750 uint16x4_t __rev1_95; __rev1_95 = __builtin_shufflevector(__s1_95, __s1_95, 3, 2, 1, 0); \
16750 uint16x4_t __ret_95; \
1675116751 __ret_95 = __rev0_95 * __noswap_splat_lane_u16(__rev1_95, __p2_95); \
1675216752 __ret_95 = __builtin_shufflevector(__ret_95, __ret_95, 3, 2, 1, 0); \
1675316753 __ret_95; \
......@@ -16756,19 +16756,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1675616756
1675716757#ifdef __LITTLE_ENDIAN__
1675816758#define vmul_lane_f32(__p0_96, __p1_96, __p2_96) __extension__ ({ \
16759 float32x2_t __ret_96; \
1675916760 float32x2_t __s0_96 = __p0_96; \
1676016761 float32x2_t __s1_96 = __p1_96; \
16761 float32x2_t __ret_96; \
1676216762 __ret_96 = __s0_96 * splat_lane_f32(__s1_96, __p2_96); \
1676316763 __ret_96; \
1676416764})
1676516765#else
1676616766#define vmul_lane_f32(__p0_97, __p1_97, __p2_97) __extension__ ({ \
16767 float32x2_t __ret_97; \
1676716768 float32x2_t __s0_97 = __p0_97; \
1676816769 float32x2_t __s1_97 = __p1_97; \
1676916770 float32x2_t __rev0_97; __rev0_97 = __builtin_shufflevector(__s0_97, __s0_97, 1, 0); \
1677016771 float32x2_t __rev1_97; __rev1_97 = __builtin_shufflevector(__s1_97, __s1_97, 1, 0); \
16771 float32x2_t __ret_97; \
1677216772 __ret_97 = __rev0_97 * __noswap_splat_lane_f32(__rev1_97, __p2_97); \
1677316773 __ret_97 = __builtin_shufflevector(__ret_97, __ret_97, 1, 0); \
1677416774 __ret_97; \
......@@ -16777,19 +16777,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1677716777
1677816778#ifdef __LITTLE_ENDIAN__
1677916779#define vmul_lane_s32(__p0_98, __p1_98, __p2_98) __extension__ ({ \
16780 int32x2_t __ret_98; \
1678016781 int32x2_t __s0_98 = __p0_98; \
1678116782 int32x2_t __s1_98 = __p1_98; \
16782 int32x2_t __ret_98; \
1678316783 __ret_98 = __s0_98 * splat_lane_s32(__s1_98, __p2_98); \
1678416784 __ret_98; \
1678516785})
1678616786#else
1678716787#define vmul_lane_s32(__p0_99, __p1_99, __p2_99) __extension__ ({ \
16788 int32x2_t __ret_99; \
1678816789 int32x2_t __s0_99 = __p0_99; \
1678916790 int32x2_t __s1_99 = __p1_99; \
1679016791 int32x2_t __rev0_99; __rev0_99 = __builtin_shufflevector(__s0_99, __s0_99, 1, 0); \
1679116792 int32x2_t __rev1_99; __rev1_99 = __builtin_shufflevector(__s1_99, __s1_99, 1, 0); \
16792 int32x2_t __ret_99; \
1679316793 __ret_99 = __rev0_99 * __noswap_splat_lane_s32(__rev1_99, __p2_99); \
1679416794 __ret_99 = __builtin_shufflevector(__ret_99, __ret_99, 1, 0); \
1679516795 __ret_99; \
......@@ -16798,19 +16798,19 @@ __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
1679816798
1679916799#ifdef __LITTLE_ENDIAN__
1680016800#define vmul_lane_s16(__p0_100, __p1_100, __p2_100) __extension__ ({ \
16801 int16x4_t __ret_100; \
1680116802 int16x4_t __s0_100 = __p0_100; \
1680216803 int16x4_t __s1_100 = __p1_100; \
16803 int16x4_t __ret_100; \
1680416804 __ret_100 = __s0_100 * splat_lane_s16(__s1_100, __p2_100); \
1680516805 __ret_100; \
1680616806})
1680716807#else
1680816808#define vmul_lane_s16(__p0_101, __p1_101, __p2_101) __extension__ ({ \
16809 int16x4_t __ret_101; \
1680916810 int16x4_t __s0_101 = __p0_101; \
1681016811 int16x4_t __s1_101 = __p1_101; \
1681116812 int16x4_t __rev0_101; __rev0_101 = __builtin_shufflevector(__s0_101, __s0_101, 3, 2, 1, 0); \
1681216813 int16x4_t __rev1_101; __rev1_101 = __builtin_shufflevector(__s1_101, __s1_101, 3, 2, 1, 0); \
16813 int16x4_t __ret_101; \
1681416814 __ret_101 = __rev0_101 * __noswap_splat_lane_s16(__rev1_101, __p2_101); \
1681516815 __ret_101 = __builtin_shufflevector(__ret_101, __ret_101, 3, 2, 1, 0); \
1681616816 __ret_101; \
......@@ -16825,8 +16825,8 @@ __ai uint32x4_t vmulq_n_u32(uint32x4_t __p0, uint32_t __p1) {
1682516825}
1682616826#else
1682716827__ai uint32x4_t vmulq_n_u32(uint32x4_t __p0, uint32_t __p1) {
16828 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1682916828 uint32x4_t __ret;
16829 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1683016830 __ret = __rev0 * (uint32x4_t) {__p1, __p1, __p1, __p1};
1683116831 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1683216832 return __ret;
......@@ -16841,8 +16841,8 @@ __ai uint16x8_t vmulq_n_u16(uint16x8_t __p0, uint16_t __p1) {
1684116841}
1684216842#else
1684316843__ai uint16x8_t vmulq_n_u16(uint16x8_t __p0, uint16_t __p1) {
16844 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1684516844 uint16x8_t __ret;
16845 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1684616846 __ret = __rev0 * (uint16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1};
1684716847 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1684816848 return __ret;
......@@ -16857,8 +16857,8 @@ __ai float32x4_t vmulq_n_f32(float32x4_t __p0, float32_t __p1) {
1685716857}
1685816858#else
1685916859__ai float32x4_t vmulq_n_f32(float32x4_t __p0, float32_t __p1) {
16860 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1686116860 float32x4_t __ret;
16861 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1686216862 __ret = __rev0 * (float32x4_t) {__p1, __p1, __p1, __p1};
1686316863 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1686416864 return __ret;
......@@ -16873,8 +16873,8 @@ __ai int32x4_t vmulq_n_s32(int32x4_t __p0, int32_t __p1) {
1687316873}
1687416874#else
1687516875__ai int32x4_t vmulq_n_s32(int32x4_t __p0, int32_t __p1) {
16876 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1687716876 int32x4_t __ret;
16877 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1687816878 __ret = __rev0 * (int32x4_t) {__p1, __p1, __p1, __p1};
1687916879 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1688016880 return __ret;
......@@ -16889,8 +16889,8 @@ __ai int16x8_t vmulq_n_s16(int16x8_t __p0, int16_t __p1) {
1688916889}
1689016890#else
1689116891__ai int16x8_t vmulq_n_s16(int16x8_t __p0, int16_t __p1) {
16892 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1689316892 int16x8_t __ret;
16893 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1689416894 __ret = __rev0 * (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1};
1689516895 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1689616896 return __ret;
......@@ -16905,8 +16905,8 @@ __ai uint32x2_t vmul_n_u32(uint32x2_t __p0, uint32_t __p1) {
1690516905}
1690616906#else
1690716907__ai uint32x2_t vmul_n_u32(uint32x2_t __p0, uint32_t __p1) {
16908 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1690916908 uint32x2_t __ret;
16909 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1691016910 __ret = __rev0 * (uint32x2_t) {__p1, __p1};
1691116911 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1691216912 return __ret;
......@@ -16921,8 +16921,8 @@ __ai uint16x4_t vmul_n_u16(uint16x4_t __p0, uint16_t __p1) {
1692116921}
1692216922#else
1692316923__ai uint16x4_t vmul_n_u16(uint16x4_t __p0, uint16_t __p1) {
16924 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1692516924 uint16x4_t __ret;
16925 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1692616926 __ret = __rev0 * (uint16x4_t) {__p1, __p1, __p1, __p1};
1692716927 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1692816928 return __ret;
......@@ -16937,8 +16937,8 @@ __ai float32x2_t vmul_n_f32(float32x2_t __p0, float32_t __p1) {
1693716937}
1693816938#else
1693916939__ai float32x2_t vmul_n_f32(float32x2_t __p0, float32_t __p1) {
16940 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1694116940 float32x2_t __ret;
16941 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1694216942 __ret = __rev0 * (float32x2_t) {__p1, __p1};
1694316943 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1694416944 return __ret;
......@@ -16953,8 +16953,8 @@ __ai int32x2_t vmul_n_s32(int32x2_t __p0, int32_t __p1) {
1695316953}
1695416954#else
1695516955__ai int32x2_t vmul_n_s32(int32x2_t __p0, int32_t __p1) {
16956 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1695716956 int32x2_t __ret;
16957 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1695816958 __ret = __rev0 * (int32x2_t) {__p1, __p1};
1695916959 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1696016960 return __ret;
......@@ -16969,8 +16969,8 @@ __ai int16x4_t vmul_n_s16(int16x4_t __p0, int16_t __p1) {
1696916969}
1697016970#else
1697116971__ai int16x4_t vmul_n_s16(int16x4_t __p0, int16_t __p1) {
16972 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1697316972 int16x4_t __ret;
16973 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1697416974 __ret = __rev0 * (int16x4_t) {__p1, __p1, __p1, __p1};
1697516975 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1697616976 return __ret;
......@@ -16985,9 +16985,9 @@ __ai poly16x8_t vmull_p8(poly8x8_t __p0, poly8x8_t __p1) {
1698516985}
1698616986#else
1698716987__ai poly16x8_t vmull_p8(poly8x8_t __p0, poly8x8_t __p1) {
16988 poly16x8_t __ret;
1698816989 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1698916990 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16990 poly16x8_t __ret;
1699116991 __ret = (poly16x8_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 37);
1699216992 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1699316993 return __ret;
......@@ -17007,9 +17007,9 @@ __ai uint16x8_t vmull_u8(uint8x8_t __p0, uint8x8_t __p1) {
1700717007}
1700817008#else
1700917009__ai uint16x8_t vmull_u8(uint8x8_t __p0, uint8x8_t __p1) {
17010 uint16x8_t __ret;
1701017011 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1701117012 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17012 uint16x8_t __ret;
1701317013 __ret = (uint16x8_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 49);
1701417014 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1701517015 return __ret;
......@@ -17029,9 +17029,9 @@ __ai uint64x2_t vmull_u32(uint32x2_t __p0, uint32x2_t __p1) {
1702917029}
1703017030#else
1703117031__ai uint64x2_t vmull_u32(uint32x2_t __p0, uint32x2_t __p1) {
17032 uint64x2_t __ret;
1703217033 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1703317034 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17034 uint64x2_t __ret;
1703517035 __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 51);
1703617036 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1703717037 return __ret;
......@@ -17051,9 +17051,9 @@ __ai uint32x4_t vmull_u16(uint16x4_t __p0, uint16x4_t __p1) {
1705117051}
1705217052#else
1705317053__ai uint32x4_t vmull_u16(uint16x4_t __p0, uint16x4_t __p1) {
17054 uint32x4_t __ret;
1705417055 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1705517056 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17056 uint32x4_t __ret;
1705717057 __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 50);
1705817058 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1705917059 return __ret;
......@@ -17073,9 +17073,9 @@ __ai int16x8_t vmull_s8(int8x8_t __p0, int8x8_t __p1) {
1707317073}
1707417074#else
1707517075__ai int16x8_t vmull_s8(int8x8_t __p0, int8x8_t __p1) {
17076 int16x8_t __ret;
1707617077 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1707717078 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17078 int16x8_t __ret;
1707917079 __ret = (int16x8_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 33);
1708017080 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1708117081 return __ret;
......@@ -17095,9 +17095,9 @@ __ai int64x2_t vmull_s32(int32x2_t __p0, int32x2_t __p1) {
1709517095}
1709617096#else
1709717097__ai int64x2_t vmull_s32(int32x2_t __p0, int32x2_t __p1) {
17098 int64x2_t __ret;
1709817099 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1709917100 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17100 int64x2_t __ret;
1710117101 __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 35);
1710217102 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1710317103 return __ret;
......@@ -17117,9 +17117,9 @@ __ai int32x4_t vmull_s16(int16x4_t __p0, int16x4_t __p1) {
1711717117}
1711817118#else
1711917119__ai int32x4_t vmull_s16(int16x4_t __p0, int16x4_t __p1) {
17120 int32x4_t __ret;
1712017121 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1712117122 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17122 int32x4_t __ret;
1712317123 __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 34);
1712417124 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1712517125 return __ret;
......@@ -17133,19 +17133,19 @@ __ai int32x4_t __noswap_vmull_s16(int16x4_t __p0, int16x4_t __p1) {
1713317133
1713417134#ifdef __LITTLE_ENDIAN__
1713517135#define vmull_lane_u32(__p0_102, __p1_102, __p2_102) __extension__ ({ \
17136 uint64x2_t __ret_102; \
1713617137 uint32x2_t __s0_102 = __p0_102; \
1713717138 uint32x2_t __s1_102 = __p1_102; \
17138 uint64x2_t __ret_102; \
1713917139 __ret_102 = vmull_u32(__s0_102, splat_lane_u32(__s1_102, __p2_102)); \
1714017140 __ret_102; \
1714117141})
1714217142#else
1714317143#define vmull_lane_u32(__p0_103, __p1_103, __p2_103) __extension__ ({ \
17144 uint64x2_t __ret_103; \
1714417145 uint32x2_t __s0_103 = __p0_103; \
1714517146 uint32x2_t __s1_103 = __p1_103; \
1714617147 uint32x2_t __rev0_103; __rev0_103 = __builtin_shufflevector(__s0_103, __s0_103, 1, 0); \
1714717148 uint32x2_t __rev1_103; __rev1_103 = __builtin_shufflevector(__s1_103, __s1_103, 1, 0); \
17148 uint64x2_t __ret_103; \
1714917149 __ret_103 = __noswap_vmull_u32(__rev0_103, __noswap_splat_lane_u32(__rev1_103, __p2_103)); \
1715017150 __ret_103 = __builtin_shufflevector(__ret_103, __ret_103, 1, 0); \
1715117151 __ret_103; \
......@@ -17154,19 +17154,19 @@ __ai int32x4_t __noswap_vmull_s16(int16x4_t __p0, int16x4_t __p1) {
1715417154
1715517155#ifdef __LITTLE_ENDIAN__
1715617156#define vmull_lane_u16(__p0_104, __p1_104, __p2_104) __extension__ ({ \
17157 uint32x4_t __ret_104; \
1715717158 uint16x4_t __s0_104 = __p0_104; \
1715817159 uint16x4_t __s1_104 = __p1_104; \
17159 uint32x4_t __ret_104; \
1716017160 __ret_104 = vmull_u16(__s0_104, splat_lane_u16(__s1_104, __p2_104)); \
1716117161 __ret_104; \
1716217162})
1716317163#else
1716417164#define vmull_lane_u16(__p0_105, __p1_105, __p2_105) __extension__ ({ \
17165 uint32x4_t __ret_105; \
1716517166 uint16x4_t __s0_105 = __p0_105; \
1716617167 uint16x4_t __s1_105 = __p1_105; \
1716717168 uint16x4_t __rev0_105; __rev0_105 = __builtin_shufflevector(__s0_105, __s0_105, 3, 2, 1, 0); \
1716817169 uint16x4_t __rev1_105; __rev1_105 = __builtin_shufflevector(__s1_105, __s1_105, 3, 2, 1, 0); \
17169 uint32x4_t __ret_105; \
1717017170 __ret_105 = __noswap_vmull_u16(__rev0_105, __noswap_splat_lane_u16(__rev1_105, __p2_105)); \
1717117171 __ret_105 = __builtin_shufflevector(__ret_105, __ret_105, 3, 2, 1, 0); \
1717217172 __ret_105; \
......@@ -17175,19 +17175,19 @@ __ai int32x4_t __noswap_vmull_s16(int16x4_t __p0, int16x4_t __p1) {
1717517175
1717617176#ifdef __LITTLE_ENDIAN__
1717717177#define vmull_lane_s32(__p0_106, __p1_106, __p2_106) __extension__ ({ \
17178 int64x2_t __ret_106; \
1717817179 int32x2_t __s0_106 = __p0_106; \
1717917180 int32x2_t __s1_106 = __p1_106; \
17180 int64x2_t __ret_106; \
1718117181 __ret_106 = vmull_s32(__s0_106, splat_lane_s32(__s1_106, __p2_106)); \
1718217182 __ret_106; \
1718317183})
1718417184#else
1718517185#define vmull_lane_s32(__p0_107, __p1_107, __p2_107) __extension__ ({ \
17186 int64x2_t __ret_107; \
1718617187 int32x2_t __s0_107 = __p0_107; \
1718717188 int32x2_t __s1_107 = __p1_107; \
1718817189 int32x2_t __rev0_107; __rev0_107 = __builtin_shufflevector(__s0_107, __s0_107, 1, 0); \
1718917190 int32x2_t __rev1_107; __rev1_107 = __builtin_shufflevector(__s1_107, __s1_107, 1, 0); \
17190 int64x2_t __ret_107; \
1719117191 __ret_107 = __noswap_vmull_s32(__rev0_107, __noswap_splat_lane_s32(__rev1_107, __p2_107)); \
1719217192 __ret_107 = __builtin_shufflevector(__ret_107, __ret_107, 1, 0); \
1719317193 __ret_107; \
......@@ -17196,19 +17196,19 @@ __ai int32x4_t __noswap_vmull_s16(int16x4_t __p0, int16x4_t __p1) {
1719617196
1719717197#ifdef __LITTLE_ENDIAN__
1719817198#define vmull_lane_s16(__p0_108, __p1_108, __p2_108) __extension__ ({ \
17199 int32x4_t __ret_108; \
1719917200 int16x4_t __s0_108 = __p0_108; \
1720017201 int16x4_t __s1_108 = __p1_108; \
17201 int32x4_t __ret_108; \
1720217202 __ret_108 = vmull_s16(__s0_108, splat_lane_s16(__s1_108, __p2_108)); \
1720317203 __ret_108; \
1720417204})
1720517205#else
1720617206#define vmull_lane_s16(__p0_109, __p1_109, __p2_109) __extension__ ({ \
17207 int32x4_t __ret_109; \
1720717208 int16x4_t __s0_109 = __p0_109; \
1720817209 int16x4_t __s1_109 = __p1_109; \
1720917210 int16x4_t __rev0_109; __rev0_109 = __builtin_shufflevector(__s0_109, __s0_109, 3, 2, 1, 0); \
1721017211 int16x4_t __rev1_109; __rev1_109 = __builtin_shufflevector(__s1_109, __s1_109, 3, 2, 1, 0); \
17211 int32x4_t __ret_109; \
1721217212 __ret_109 = __noswap_vmull_s16(__rev0_109, __noswap_splat_lane_s16(__rev1_109, __p2_109)); \
1721317213 __ret_109 = __builtin_shufflevector(__ret_109, __ret_109, 3, 2, 1, 0); \
1721417214 __ret_109; \
......@@ -17223,8 +17223,8 @@ __ai uint64x2_t vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
1722317223}
1722417224#else
1722517225__ai uint64x2_t vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
17226 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1722717226 uint64x2_t __ret;
17227 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1722817228 __ret = __noswap_vmull_u32(__rev0, (uint32x2_t) {__p1, __p1});
1722917229 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1723017230 return __ret;
......@@ -17244,8 +17244,8 @@ __ai uint32x4_t vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
1724417244}
1724517245#else
1724617246__ai uint32x4_t vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
17247 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1724817247 uint32x4_t __ret;
17248 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1724917249 __ret = __noswap_vmull_u16(__rev0, (uint16x4_t) {__p1, __p1, __p1, __p1});
1725017250 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1725117251 return __ret;
......@@ -17265,8 +17265,8 @@ __ai int64x2_t vmull_n_s32(int32x2_t __p0, int32_t __p1) {
1726517265}
1726617266#else
1726717267__ai int64x2_t vmull_n_s32(int32x2_t __p0, int32_t __p1) {
17268 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1726917268 int64x2_t __ret;
17269 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1727017270 __ret = __noswap_vmull_s32(__rev0, (int32x2_t) {__p1, __p1});
1727117271 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1727217272 return __ret;
......@@ -17286,8 +17286,8 @@ __ai int32x4_t vmull_n_s16(int16x4_t __p0, int16_t __p1) {
1728617286}
1728717287#else
1728817288__ai int32x4_t vmull_n_s16(int16x4_t __p0, int16_t __p1) {
17289 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1729017289 int32x4_t __ret;
17290 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1729117291 __ret = __noswap_vmull_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
1729217292 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1729317293 return __ret;
......@@ -17307,8 +17307,8 @@ __ai poly8x8_t vmvn_p8(poly8x8_t __p0) {
1730717307}
1730817308#else
1730917309__ai poly8x8_t vmvn_p8(poly8x8_t __p0) {
17310 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1731117310 poly8x8_t __ret;
17311 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1731217312 __ret = ~__rev0;
1731317313 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1731417314 return __ret;
......@@ -17323,8 +17323,8 @@ __ai poly8x16_t vmvnq_p8(poly8x16_t __p0) {
1732317323}
1732417324#else
1732517325__ai poly8x16_t vmvnq_p8(poly8x16_t __p0) {
17326 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1732717326 poly8x16_t __ret;
17327 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1732817328 __ret = ~__rev0;
1732917329 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1733017330 return __ret;
......@@ -17339,8 +17339,8 @@ __ai uint8x16_t vmvnq_u8(uint8x16_t __p0) {
1733917339}
1734017340#else
1734117341__ai uint8x16_t vmvnq_u8(uint8x16_t __p0) {
17342 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1734317342 uint8x16_t __ret;
17343 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1734417344 __ret = ~__rev0;
1734517345 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1734617346 return __ret;
......@@ -17355,8 +17355,8 @@ __ai uint32x4_t vmvnq_u32(uint32x4_t __p0) {
1735517355}
1735617356#else
1735717357__ai uint32x4_t vmvnq_u32(uint32x4_t __p0) {
17358 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1735917358 uint32x4_t __ret;
17359 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1736017360 __ret = ~__rev0;
1736117361 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1736217362 return __ret;
......@@ -17371,8 +17371,8 @@ __ai uint16x8_t vmvnq_u16(uint16x8_t __p0) {
1737117371}
1737217372#else
1737317373__ai uint16x8_t vmvnq_u16(uint16x8_t __p0) {
17374 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1737517374 uint16x8_t __ret;
17375 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1737617376 __ret = ~__rev0;
1737717377 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1737817378 return __ret;
......@@ -17387,8 +17387,8 @@ __ai int8x16_t vmvnq_s8(int8x16_t __p0) {
1738717387}
1738817388#else
1738917389__ai int8x16_t vmvnq_s8(int8x16_t __p0) {
17390 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1739117390 int8x16_t __ret;
17391 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1739217392 __ret = ~__rev0;
1739317393 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1739417394 return __ret;
......@@ -17403,8 +17403,8 @@ __ai int32x4_t vmvnq_s32(int32x4_t __p0) {
1740317403}
1740417404#else
1740517405__ai int32x4_t vmvnq_s32(int32x4_t __p0) {
17406 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1740717406 int32x4_t __ret;
17407 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1740817408 __ret = ~__rev0;
1740917409 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1741017410 return __ret;
......@@ -17419,8 +17419,8 @@ __ai int16x8_t vmvnq_s16(int16x8_t __p0) {
1741917419}
1742017420#else
1742117421__ai int16x8_t vmvnq_s16(int16x8_t __p0) {
17422 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1742317422 int16x8_t __ret;
17423 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1742417424 __ret = ~__rev0;
1742517425 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1742617426 return __ret;
......@@ -17435,8 +17435,8 @@ __ai uint8x8_t vmvn_u8(uint8x8_t __p0) {
1743517435}
1743617436#else
1743717437__ai uint8x8_t vmvn_u8(uint8x8_t __p0) {
17438 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1743917438 uint8x8_t __ret;
17439 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1744017440 __ret = ~__rev0;
1744117441 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1744217442 return __ret;
......@@ -17451,8 +17451,8 @@ __ai uint32x2_t vmvn_u32(uint32x2_t __p0) {
1745117451}
1745217452#else
1745317453__ai uint32x2_t vmvn_u32(uint32x2_t __p0) {
17454 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1745517454 uint32x2_t __ret;
17455 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1745617456 __ret = ~__rev0;
1745717457 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1745817458 return __ret;
......@@ -17467,8 +17467,8 @@ __ai uint16x4_t vmvn_u16(uint16x4_t __p0) {
1746717467}
1746817468#else
1746917469__ai uint16x4_t vmvn_u16(uint16x4_t __p0) {
17470 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1747117470 uint16x4_t __ret;
17471 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1747217472 __ret = ~__rev0;
1747317473 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1747417474 return __ret;
......@@ -17483,8 +17483,8 @@ __ai int8x8_t vmvn_s8(int8x8_t __p0) {
1748317483}
1748417484#else
1748517485__ai int8x8_t vmvn_s8(int8x8_t __p0) {
17486 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1748717486 int8x8_t __ret;
17487 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1748817488 __ret = ~__rev0;
1748917489 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1749017490 return __ret;
......@@ -17499,8 +17499,8 @@ __ai int32x2_t vmvn_s32(int32x2_t __p0) {
1749917499}
1750017500#else
1750117501__ai int32x2_t vmvn_s32(int32x2_t __p0) {
17502 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1750317502 int32x2_t __ret;
17503 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1750417504 __ret = ~__rev0;
1750517505 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1750617506 return __ret;
......@@ -17515,8 +17515,8 @@ __ai int16x4_t vmvn_s16(int16x4_t __p0) {
1751517515}
1751617516#else
1751717517__ai int16x4_t vmvn_s16(int16x4_t __p0) {
17518 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1751917518 int16x4_t __ret;
17519 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1752017520 __ret = ~__rev0;
1752117521 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1752217522 return __ret;
......@@ -17531,8 +17531,8 @@ __ai int8x16_t vnegq_s8(int8x16_t __p0) {
1753117531}
1753217532#else
1753317533__ai int8x16_t vnegq_s8(int8x16_t __p0) {
17534 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1753517534 int8x16_t __ret;
17535 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1753617536 __ret = -__rev0;
1753717537 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1753817538 return __ret;
......@@ -17547,8 +17547,8 @@ __ai float32x4_t vnegq_f32(float32x4_t __p0) {
1754717547}
1754817548#else
1754917549__ai float32x4_t vnegq_f32(float32x4_t __p0) {
17550 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1755117550 float32x4_t __ret;
17551 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1755217552 __ret = -__rev0;
1755317553 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1755417554 return __ret;
......@@ -17563,8 +17563,8 @@ __ai int32x4_t vnegq_s32(int32x4_t __p0) {
1756317563}
1756417564#else
1756517565__ai int32x4_t vnegq_s32(int32x4_t __p0) {
17566 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1756717566 int32x4_t __ret;
17567 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1756817568 __ret = -__rev0;
1756917569 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1757017570 return __ret;
......@@ -17579,8 +17579,8 @@ __ai int16x8_t vnegq_s16(int16x8_t __p0) {
1757917579}
1758017580#else
1758117581__ai int16x8_t vnegq_s16(int16x8_t __p0) {
17582 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1758317582 int16x8_t __ret;
17583 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1758417584 __ret = -__rev0;
1758517585 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1758617586 return __ret;
......@@ -17595,8 +17595,8 @@ __ai int8x8_t vneg_s8(int8x8_t __p0) {
1759517595}
1759617596#else
1759717597__ai int8x8_t vneg_s8(int8x8_t __p0) {
17598 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1759917598 int8x8_t __ret;
17599 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1760017600 __ret = -__rev0;
1760117601 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1760217602 return __ret;
......@@ -17611,8 +17611,8 @@ __ai float32x2_t vneg_f32(float32x2_t __p0) {
1761117611}
1761217612#else
1761317613__ai float32x2_t vneg_f32(float32x2_t __p0) {
17614 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1761517614 float32x2_t __ret;
17615 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1761617616 __ret = -__rev0;
1761717617 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1761817618 return __ret;
......@@ -17627,8 +17627,8 @@ __ai int32x2_t vneg_s32(int32x2_t __p0) {
1762717627}
1762817628#else
1762917629__ai int32x2_t vneg_s32(int32x2_t __p0) {
17630 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1763117630 int32x2_t __ret;
17631 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1763217632 __ret = -__rev0;
1763317633 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1763417634 return __ret;
......@@ -17643,8 +17643,8 @@ __ai int16x4_t vneg_s16(int16x4_t __p0) {
1764317643}
1764417644#else
1764517645__ai int16x4_t vneg_s16(int16x4_t __p0) {
17646 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1764717646 int16x4_t __ret;
17647 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1764817648 __ret = -__rev0;
1764917649 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1765017650 return __ret;
......@@ -17659,9 +17659,9 @@ __ai uint8x16_t vornq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1765917659}
1766017660#else
1766117661__ai uint8x16_t vornq_u8(uint8x16_t __p0, uint8x16_t __p1) {
17662 uint8x16_t __ret;
1766217663 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1766317664 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
17664 uint8x16_t __ret;
1766517665 __ret = __rev0 | ~__rev1;
1766617666 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1766717667 return __ret;
......@@ -17676,9 +17676,9 @@ __ai uint32x4_t vornq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1767617676}
1767717677#else
1767817678__ai uint32x4_t vornq_u32(uint32x4_t __p0, uint32x4_t __p1) {
17679 uint32x4_t __ret;
1767917680 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1768017681 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17681 uint32x4_t __ret;
1768217682 __ret = __rev0 | ~__rev1;
1768317683 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1768417684 return __ret;
......@@ -17693,9 +17693,9 @@ __ai uint64x2_t vornq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1769317693}
1769417694#else
1769517695__ai uint64x2_t vornq_u64(uint64x2_t __p0, uint64x2_t __p1) {
17696 uint64x2_t __ret;
1769617697 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1769717698 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17698 uint64x2_t __ret;
1769917699 __ret = __rev0 | ~__rev1;
1770017700 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1770117701 return __ret;
......@@ -17710,9 +17710,9 @@ __ai uint16x8_t vornq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1771017710}
1771117711#else
1771217712__ai uint16x8_t vornq_u16(uint16x8_t __p0, uint16x8_t __p1) {
17713 uint16x8_t __ret;
1771317714 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1771417715 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17715 uint16x8_t __ret;
1771617716 __ret = __rev0 | ~__rev1;
1771717717 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1771817718 return __ret;
......@@ -17727,9 +17727,9 @@ __ai int8x16_t vornq_s8(int8x16_t __p0, int8x16_t __p1) {
1772717727}
1772817728#else
1772917729__ai int8x16_t vornq_s8(int8x16_t __p0, int8x16_t __p1) {
17730 int8x16_t __ret;
1773017731 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1773117732 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
17732 int8x16_t __ret;
1773317733 __ret = __rev0 | ~__rev1;
1773417734 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1773517735 return __ret;
......@@ -17744,9 +17744,9 @@ __ai int32x4_t vornq_s32(int32x4_t __p0, int32x4_t __p1) {
1774417744}
1774517745#else
1774617746__ai int32x4_t vornq_s32(int32x4_t __p0, int32x4_t __p1) {
17747 int32x4_t __ret;
1774717748 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1774817749 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17749 int32x4_t __ret;
1775017750 __ret = __rev0 | ~__rev1;
1775117751 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1775217752 return __ret;
......@@ -17761,9 +17761,9 @@ __ai int64x2_t vornq_s64(int64x2_t __p0, int64x2_t __p1) {
1776117761}
1776217762#else
1776317763__ai int64x2_t vornq_s64(int64x2_t __p0, int64x2_t __p1) {
17764 int64x2_t __ret;
1776417765 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1776517766 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17766 int64x2_t __ret;
1776717767 __ret = __rev0 | ~__rev1;
1776817768 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1776917769 return __ret;
......@@ -17778,9 +17778,9 @@ __ai int16x8_t vornq_s16(int16x8_t __p0, int16x8_t __p1) {
1777817778}
1777917779#else
1778017780__ai int16x8_t vornq_s16(int16x8_t __p0, int16x8_t __p1) {
17781 int16x8_t __ret;
1778117782 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1778217783 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17783 int16x8_t __ret;
1778417784 __ret = __rev0 | ~__rev1;
1778517785 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1778617786 return __ret;
......@@ -17795,9 +17795,9 @@ __ai uint8x8_t vorn_u8(uint8x8_t __p0, uint8x8_t __p1) {
1779517795}
1779617796#else
1779717797__ai uint8x8_t vorn_u8(uint8x8_t __p0, uint8x8_t __p1) {
17798 uint8x8_t __ret;
1779817799 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1779917800 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17800 uint8x8_t __ret;
1780117801 __ret = __rev0 | ~__rev1;
1780217802 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1780317803 return __ret;
......@@ -17812,9 +17812,9 @@ __ai uint32x2_t vorn_u32(uint32x2_t __p0, uint32x2_t __p1) {
1781217812}
1781317813#else
1781417814__ai uint32x2_t vorn_u32(uint32x2_t __p0, uint32x2_t __p1) {
17815 uint32x2_t __ret;
1781517816 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1781617817 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17817 uint32x2_t __ret;
1781817818 __ret = __rev0 | ~__rev1;
1781917819 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1782017820 return __ret;
......@@ -17834,9 +17834,9 @@ __ai uint16x4_t vorn_u16(uint16x4_t __p0, uint16x4_t __p1) {
1783417834}
1783517835#else
1783617836__ai uint16x4_t vorn_u16(uint16x4_t __p0, uint16x4_t __p1) {
17837 uint16x4_t __ret;
1783717838 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1783817839 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17839 uint16x4_t __ret;
1784017840 __ret = __rev0 | ~__rev1;
1784117841 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1784217842 return __ret;
......@@ -17851,9 +17851,9 @@ __ai int8x8_t vorn_s8(int8x8_t __p0, int8x8_t __p1) {
1785117851}
1785217852#else
1785317853__ai int8x8_t vorn_s8(int8x8_t __p0, int8x8_t __p1) {
17854 int8x8_t __ret;
1785417855 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1785517856 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17856 int8x8_t __ret;
1785717857 __ret = __rev0 | ~__rev1;
1785817858 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1785917859 return __ret;
......@@ -17868,9 +17868,9 @@ __ai int32x2_t vorn_s32(int32x2_t __p0, int32x2_t __p1) {
1786817868}
1786917869#else
1787017870__ai int32x2_t vorn_s32(int32x2_t __p0, int32x2_t __p1) {
17871 int32x2_t __ret;
1787117872 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1787217873 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17873 int32x2_t __ret;
1787417874 __ret = __rev0 | ~__rev1;
1787517875 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1787617876 return __ret;
......@@ -17890,9 +17890,9 @@ __ai int16x4_t vorn_s16(int16x4_t __p0, int16x4_t __p1) {
1789017890}
1789117891#else
1789217892__ai int16x4_t vorn_s16(int16x4_t __p0, int16x4_t __p1) {
17893 int16x4_t __ret;
1789317894 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1789417895 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17895 int16x4_t __ret;
1789617896 __ret = __rev0 | ~__rev1;
1789717897 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1789817898 return __ret;
......@@ -17907,9 +17907,9 @@ __ai uint8x16_t vorrq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1790717907}
1790817908#else
1790917909__ai uint8x16_t vorrq_u8(uint8x16_t __p0, uint8x16_t __p1) {
17910 uint8x16_t __ret;
1791017911 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1791117912 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
17912 uint8x16_t __ret;
1791317913 __ret = __rev0 | __rev1;
1791417914 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1791517915 return __ret;
......@@ -17924,9 +17924,9 @@ __ai uint32x4_t vorrq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1792417924}
1792517925#else
1792617926__ai uint32x4_t vorrq_u32(uint32x4_t __p0, uint32x4_t __p1) {
17927 uint32x4_t __ret;
1792717928 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1792817929 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17929 uint32x4_t __ret;
1793017930 __ret = __rev0 | __rev1;
1793117931 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1793217932 return __ret;
......@@ -17941,9 +17941,9 @@ __ai uint64x2_t vorrq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1794117941}
1794217942#else
1794317943__ai uint64x2_t vorrq_u64(uint64x2_t __p0, uint64x2_t __p1) {
17944 uint64x2_t __ret;
1794417945 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1794517946 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17946 uint64x2_t __ret;
1794717947 __ret = __rev0 | __rev1;
1794817948 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1794917949 return __ret;
......@@ -17958,9 +17958,9 @@ __ai uint16x8_t vorrq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1795817958}
1795917959#else
1796017960__ai uint16x8_t vorrq_u16(uint16x8_t __p0, uint16x8_t __p1) {
17961 uint16x8_t __ret;
1796117962 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1796217963 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17963 uint16x8_t __ret;
1796417964 __ret = __rev0 | __rev1;
1796517965 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1796617966 return __ret;
......@@ -17975,9 +17975,9 @@ __ai int8x16_t vorrq_s8(int8x16_t __p0, int8x16_t __p1) {
1797517975}
1797617976#else
1797717977__ai int8x16_t vorrq_s8(int8x16_t __p0, int8x16_t __p1) {
17978 int8x16_t __ret;
1797817979 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1797917980 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
17980 int8x16_t __ret;
1798117981 __ret = __rev0 | __rev1;
1798217982 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1798317983 return __ret;
......@@ -17992,9 +17992,9 @@ __ai int32x4_t vorrq_s32(int32x4_t __p0, int32x4_t __p1) {
1799217992}
1799317993#else
1799417994__ai int32x4_t vorrq_s32(int32x4_t __p0, int32x4_t __p1) {
17995 int32x4_t __ret;
1799517996 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1799617997 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17997 int32x4_t __ret;
1799817998 __ret = __rev0 | __rev1;
1799917999 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1800018000 return __ret;
......@@ -18009,9 +18009,9 @@ __ai int64x2_t vorrq_s64(int64x2_t __p0, int64x2_t __p1) {
1800918009}
1801018010#else
1801118011__ai int64x2_t vorrq_s64(int64x2_t __p0, int64x2_t __p1) {
18012 int64x2_t __ret;
1801218013 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1801318014 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18014 int64x2_t __ret;
1801518015 __ret = __rev0 | __rev1;
1801618016 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1801718017 return __ret;
......@@ -18026,9 +18026,9 @@ __ai int16x8_t vorrq_s16(int16x8_t __p0, int16x8_t __p1) {
1802618026}
1802718027#else
1802818028__ai int16x8_t vorrq_s16(int16x8_t __p0, int16x8_t __p1) {
18029 int16x8_t __ret;
1802918030 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1803018031 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18031 int16x8_t __ret;
1803218032 __ret = __rev0 | __rev1;
1803318033 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1803418034 return __ret;
......@@ -18043,9 +18043,9 @@ __ai uint8x8_t vorr_u8(uint8x8_t __p0, uint8x8_t __p1) {
1804318043}
1804418044#else
1804518045__ai uint8x8_t vorr_u8(uint8x8_t __p0, uint8x8_t __p1) {
18046 uint8x8_t __ret;
1804618047 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1804718048 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18048 uint8x8_t __ret;
1804918049 __ret = __rev0 | __rev1;
1805018050 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1805118051 return __ret;
......@@ -18060,9 +18060,9 @@ __ai uint32x2_t vorr_u32(uint32x2_t __p0, uint32x2_t __p1) {
1806018060}
1806118061#else
1806218062__ai uint32x2_t vorr_u32(uint32x2_t __p0, uint32x2_t __p1) {
18063 uint32x2_t __ret;
1806318064 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1806418065 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18065 uint32x2_t __ret;
1806618066 __ret = __rev0 | __rev1;
1806718067 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1806818068 return __ret;
......@@ -18082,9 +18082,9 @@ __ai uint16x4_t vorr_u16(uint16x4_t __p0, uint16x4_t __p1) {
1808218082}
1808318083#else
1808418084__ai uint16x4_t vorr_u16(uint16x4_t __p0, uint16x4_t __p1) {
18085 uint16x4_t __ret;
1808518086 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1808618087 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18087 uint16x4_t __ret;
1808818088 __ret = __rev0 | __rev1;
1808918089 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1809018090 return __ret;
......@@ -18099,9 +18099,9 @@ __ai int8x8_t vorr_s8(int8x8_t __p0, int8x8_t __p1) {
1809918099}
1810018100#else
1810118101__ai int8x8_t vorr_s8(int8x8_t __p0, int8x8_t __p1) {
18102 int8x8_t __ret;
1810218103 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1810318104 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18104 int8x8_t __ret;
1810518105 __ret = __rev0 | __rev1;
1810618106 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1810718107 return __ret;
......@@ -18116,9 +18116,9 @@ __ai int32x2_t vorr_s32(int32x2_t __p0, int32x2_t __p1) {
1811618116}
1811718117#else
1811818118__ai int32x2_t vorr_s32(int32x2_t __p0, int32x2_t __p1) {
18119 int32x2_t __ret;
1811918120 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1812018121 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18121 int32x2_t __ret;
1812218122 __ret = __rev0 | __rev1;
1812318123 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1812418124 return __ret;
......@@ -18138,9 +18138,9 @@ __ai int16x4_t vorr_s16(int16x4_t __p0, int16x4_t __p1) {
1813818138}
1813918139#else
1814018140__ai int16x4_t vorr_s16(int16x4_t __p0, int16x4_t __p1) {
18141 int16x4_t __ret;
1814118142 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1814218143 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18143 int16x4_t __ret;
1814418144 __ret = __rev0 | __rev1;
1814518145 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1814618146 return __ret;
......@@ -18155,9 +18155,9 @@ __ai uint16x8_t vpadalq_u8(uint16x8_t __p0, uint8x16_t __p1) {
1815518155}
1815618156#else
1815718157__ai uint16x8_t vpadalq_u8(uint16x8_t __p0, uint8x16_t __p1) {
18158 uint16x8_t __ret;
1815818159 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1815918160 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18160 uint16x8_t __ret;
1816118161 __ret = (uint16x8_t) __builtin_neon_vpadalq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
1816218162 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1816318163 return __ret;
......@@ -18172,9 +18172,9 @@ __ai uint64x2_t vpadalq_u32(uint64x2_t __p0, uint32x4_t __p1) {
1817218172}
1817318173#else
1817418174__ai uint64x2_t vpadalq_u32(uint64x2_t __p0, uint32x4_t __p1) {
18175 uint64x2_t __ret;
1817518176 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1817618177 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18177 uint64x2_t __ret;
1817818178 __ret = (uint64x2_t) __builtin_neon_vpadalq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
1817918179 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1818018180 return __ret;
......@@ -18189,9 +18189,9 @@ __ai uint32x4_t vpadalq_u16(uint32x4_t __p0, uint16x8_t __p1) {
1818918189}
1819018190#else
1819118191__ai uint32x4_t vpadalq_u16(uint32x4_t __p0, uint16x8_t __p1) {
18192 uint32x4_t __ret;
1819218193 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1819318194 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18194 uint32x4_t __ret;
1819518195 __ret = (uint32x4_t) __builtin_neon_vpadalq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
1819618196 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1819718197 return __ret;
......@@ -18206,9 +18206,9 @@ __ai int16x8_t vpadalq_s8(int16x8_t __p0, int8x16_t __p1) {
1820618206}
1820718207#else
1820818208__ai int16x8_t vpadalq_s8(int16x8_t __p0, int8x16_t __p1) {
18209 int16x8_t __ret;
1820918210 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1821018211 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18211 int16x8_t __ret;
1821218212 __ret = (int16x8_t) __builtin_neon_vpadalq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
1821318213 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1821418214 return __ret;
......@@ -18223,9 +18223,9 @@ __ai int64x2_t vpadalq_s32(int64x2_t __p0, int32x4_t __p1) {
1822318223}
1822418224#else
1822518225__ai int64x2_t vpadalq_s32(int64x2_t __p0, int32x4_t __p1) {
18226 int64x2_t __ret;
1822618227 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1822718228 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18228 int64x2_t __ret;
1822918229 __ret = (int64x2_t) __builtin_neon_vpadalq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
1823018230 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1823118231 return __ret;
......@@ -18240,9 +18240,9 @@ __ai int32x4_t vpadalq_s16(int32x4_t __p0, int16x8_t __p1) {
1824018240}
1824118241#else
1824218242__ai int32x4_t vpadalq_s16(int32x4_t __p0, int16x8_t __p1) {
18243 int32x4_t __ret;
1824318244 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1824418245 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18245 int32x4_t __ret;
1824618246 __ret = (int32x4_t) __builtin_neon_vpadalq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
1824718247 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1824818248 return __ret;
......@@ -18257,9 +18257,9 @@ __ai uint16x4_t vpadal_u8(uint16x4_t __p0, uint8x8_t __p1) {
1825718257}
1825818258#else
1825918259__ai uint16x4_t vpadal_u8(uint16x4_t __p0, uint8x8_t __p1) {
18260 uint16x4_t __ret;
1826018261 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1826118262 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18262 uint16x4_t __ret;
1826318263 __ret = (uint16x4_t) __builtin_neon_vpadal_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1826418264 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1826518265 return __ret;
......@@ -18274,8 +18274,8 @@ __ai uint64x1_t vpadal_u32(uint64x1_t __p0, uint32x2_t __p1) {
1827418274}
1827518275#else
1827618276__ai uint64x1_t vpadal_u32(uint64x1_t __p0, uint32x2_t __p1) {
18277 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1827818277 uint64x1_t __ret;
18278 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1827918279 __ret = (uint64x1_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__rev1, 19);
1828018280 return __ret;
1828118281}
......@@ -18289,9 +18289,9 @@ __ai uint32x2_t vpadal_u16(uint32x2_t __p0, uint16x4_t __p1) {
1828918289}
1829018290#else
1829118291__ai uint32x2_t vpadal_u16(uint32x2_t __p0, uint16x4_t __p1) {
18292 uint32x2_t __ret;
1829218293 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1829318294 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18294 uint32x2_t __ret;
1829518295 __ret = (uint32x2_t) __builtin_neon_vpadal_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1829618296 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1829718297 return __ret;
......@@ -18306,9 +18306,9 @@ __ai int16x4_t vpadal_s8(int16x4_t __p0, int8x8_t __p1) {
1830618306}
1830718307#else
1830818308__ai int16x4_t vpadal_s8(int16x4_t __p0, int8x8_t __p1) {
18309 int16x4_t __ret;
1830918310 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1831018311 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18311 int16x4_t __ret;
1831218312 __ret = (int16x4_t) __builtin_neon_vpadal_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1831318313 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1831418314 return __ret;
......@@ -18323,8 +18323,8 @@ __ai int64x1_t vpadal_s32(int64x1_t __p0, int32x2_t __p1) {
1832318323}
1832418324#else
1832518325__ai int64x1_t vpadal_s32(int64x1_t __p0, int32x2_t __p1) {
18326 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1832718326 int64x1_t __ret;
18327 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1832818328 __ret = (int64x1_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__rev1, 3);
1832918329 return __ret;
1833018330}
......@@ -18338,9 +18338,9 @@ __ai int32x2_t vpadal_s16(int32x2_t __p0, int16x4_t __p1) {
1833818338}
1833918339#else
1834018340__ai int32x2_t vpadal_s16(int32x2_t __p0, int16x4_t __p1) {
18341 int32x2_t __ret;
1834118342 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1834218343 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18343 int32x2_t __ret;
1834418344 __ret = (int32x2_t) __builtin_neon_vpadal_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1834518345 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1834618346 return __ret;
......@@ -18355,9 +18355,9 @@ __ai uint8x8_t vpadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
1835518355}
1835618356#else
1835718357__ai uint8x8_t vpadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
18358 uint8x8_t __ret;
1835818359 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1835918360 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18360 uint8x8_t __ret;
1836118361 __ret = (uint8x8_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
1836218362 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1836318363 return __ret;
......@@ -18372,9 +18372,9 @@ __ai uint32x2_t vpadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
1837218372}
1837318373#else
1837418374__ai uint32x2_t vpadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
18375 uint32x2_t __ret;
1837518376 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1837618377 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18377 uint32x2_t __ret;
1837818378 __ret = (uint32x2_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1837918379 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1838018380 return __ret;
......@@ -18389,9 +18389,9 @@ __ai uint16x4_t vpadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
1838918389}
1839018390#else
1839118391__ai uint16x4_t vpadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
18392 uint16x4_t __ret;
1839218393 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1839318394 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18394 uint16x4_t __ret;
1839518395 __ret = (uint16x4_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1839618396 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1839718397 return __ret;
......@@ -18406,9 +18406,9 @@ __ai int8x8_t vpadd_s8(int8x8_t __p0, int8x8_t __p1) {
1840618406}
1840718407#else
1840818408__ai int8x8_t vpadd_s8(int8x8_t __p0, int8x8_t __p1) {
18409 int8x8_t __ret;
1840918410 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1841018411 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18411 int8x8_t __ret;
1841218412 __ret = (int8x8_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
1841318413 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1841418414 return __ret;
......@@ -18423,9 +18423,9 @@ __ai float32x2_t vpadd_f32(float32x2_t __p0, float32x2_t __p1) {
1842318423}
1842418424#else
1842518425__ai float32x2_t vpadd_f32(float32x2_t __p0, float32x2_t __p1) {
18426 float32x2_t __ret;
1842618427 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1842718428 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18428 float32x2_t __ret;
1842918429 __ret = (float32x2_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
1843018430 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1843118431 return __ret;
......@@ -18440,9 +18440,9 @@ __ai int32x2_t vpadd_s32(int32x2_t __p0, int32x2_t __p1) {
1844018440}
1844118441#else
1844218442__ai int32x2_t vpadd_s32(int32x2_t __p0, int32x2_t __p1) {
18443 int32x2_t __ret;
1844318444 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1844418445 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18445 int32x2_t __ret;
1844618446 __ret = (int32x2_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1844718447 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1844818448 return __ret;
......@@ -18457,9 +18457,9 @@ __ai int16x4_t vpadd_s16(int16x4_t __p0, int16x4_t __p1) {
1845718457}
1845818458#else
1845918459__ai int16x4_t vpadd_s16(int16x4_t __p0, int16x4_t __p1) {
18460 int16x4_t __ret;
1846018461 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1846118462 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18462 int16x4_t __ret;
1846318463 __ret = (int16x4_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1846418464 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1846518465 return __ret;
......@@ -18474,8 +18474,8 @@ __ai uint16x8_t vpaddlq_u8(uint8x16_t __p0) {
1847418474}
1847518475#else
1847618476__ai uint16x8_t vpaddlq_u8(uint8x16_t __p0) {
18477 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1847818477 uint16x8_t __ret;
18478 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1847918479 __ret = (uint16x8_t) __builtin_neon_vpaddlq_v((int8x16_t)__rev0, 49);
1848018480 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1848118481 return __ret;
......@@ -18490,8 +18490,8 @@ __ai uint64x2_t vpaddlq_u32(uint32x4_t __p0) {
1849018490}
1849118491#else
1849218492__ai uint64x2_t vpaddlq_u32(uint32x4_t __p0) {
18493 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1849418493 uint64x2_t __ret;
18494 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1849518495 __ret = (uint64x2_t) __builtin_neon_vpaddlq_v((int8x16_t)__rev0, 51);
1849618496 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1849718497 return __ret;
......@@ -18506,8 +18506,8 @@ __ai uint32x4_t vpaddlq_u16(uint16x8_t __p0) {
1850618506}
1850718507#else
1850818508__ai uint32x4_t vpaddlq_u16(uint16x8_t __p0) {
18509 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1851018509 uint32x4_t __ret;
18510 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1851118511 __ret = (uint32x4_t) __builtin_neon_vpaddlq_v((int8x16_t)__rev0, 50);
1851218512 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1851318513 return __ret;
......@@ -18522,8 +18522,8 @@ __ai int16x8_t vpaddlq_s8(int8x16_t __p0) {
1852218522}
1852318523#else
1852418524__ai int16x8_t vpaddlq_s8(int8x16_t __p0) {
18525 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1852618525 int16x8_t __ret;
18526 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1852718527 __ret = (int16x8_t) __builtin_neon_vpaddlq_v((int8x16_t)__rev0, 33);
1852818528 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1852918529 return __ret;
......@@ -18538,8 +18538,8 @@ __ai int64x2_t vpaddlq_s32(int32x4_t __p0) {
1853818538}
1853918539#else
1854018540__ai int64x2_t vpaddlq_s32(int32x4_t __p0) {
18541 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1854218541 int64x2_t __ret;
18542 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1854318543 __ret = (int64x2_t) __builtin_neon_vpaddlq_v((int8x16_t)__rev0, 35);
1854418544 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1854518545 return __ret;
......@@ -18554,8 +18554,8 @@ __ai int32x4_t vpaddlq_s16(int16x8_t __p0) {
1855418554}
1855518555#else
1855618556__ai int32x4_t vpaddlq_s16(int16x8_t __p0) {
18557 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1855818557 int32x4_t __ret;
18558 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1855918559 __ret = (int32x4_t) __builtin_neon_vpaddlq_v((int8x16_t)__rev0, 34);
1856018560 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1856118561 return __ret;
......@@ -18570,8 +18570,8 @@ __ai uint16x4_t vpaddl_u8(uint8x8_t __p0) {
1857018570}
1857118571#else
1857218572__ai uint16x4_t vpaddl_u8(uint8x8_t __p0) {
18573 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1857418573 uint16x4_t __ret;
18574 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1857518575 __ret = (uint16x4_t) __builtin_neon_vpaddl_v((int8x8_t)__rev0, 17);
1857618576 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1857718577 return __ret;
......@@ -18586,8 +18586,8 @@ __ai uint64x1_t vpaddl_u32(uint32x2_t __p0) {
1858618586}
1858718587#else
1858818588__ai uint64x1_t vpaddl_u32(uint32x2_t __p0) {
18589 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1859018589 uint64x1_t __ret;
18590 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1859118591 __ret = (uint64x1_t) __builtin_neon_vpaddl_v((int8x8_t)__rev0, 19);
1859218592 return __ret;
1859318593}
......@@ -18601,8 +18601,8 @@ __ai uint32x2_t vpaddl_u16(uint16x4_t __p0) {
1860118601}
1860218602#else
1860318603__ai uint32x2_t vpaddl_u16(uint16x4_t __p0) {
18604 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1860518604 uint32x2_t __ret;
18605 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1860618606 __ret = (uint32x2_t) __builtin_neon_vpaddl_v((int8x8_t)__rev0, 18);
1860718607 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1860818608 return __ret;
......@@ -18617,8 +18617,8 @@ __ai int16x4_t vpaddl_s8(int8x8_t __p0) {
1861718617}
1861818618#else
1861918619__ai int16x4_t vpaddl_s8(int8x8_t __p0) {
18620 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1862118620 int16x4_t __ret;
18621 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1862218622 __ret = (int16x4_t) __builtin_neon_vpaddl_v((int8x8_t)__rev0, 1);
1862318623 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1862418624 return __ret;
......@@ -18633,8 +18633,8 @@ __ai int64x1_t vpaddl_s32(int32x2_t __p0) {
1863318633}
1863418634#else
1863518635__ai int64x1_t vpaddl_s32(int32x2_t __p0) {
18636 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1863718636 int64x1_t __ret;
18637 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1863818638 __ret = (int64x1_t) __builtin_neon_vpaddl_v((int8x8_t)__rev0, 3);
1863918639 return __ret;
1864018640}
......@@ -18648,8 +18648,8 @@ __ai int32x2_t vpaddl_s16(int16x4_t __p0) {
1864818648}
1864918649#else
1865018650__ai int32x2_t vpaddl_s16(int16x4_t __p0) {
18651 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1865218651 int32x2_t __ret;
18652 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1865318653 __ret = (int32x2_t) __builtin_neon_vpaddl_v((int8x8_t)__rev0, 2);
1865418654 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1865518655 return __ret;
......@@ -18664,9 +18664,9 @@ __ai uint8x8_t vpmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
1866418664}
1866518665#else
1866618666__ai uint8x8_t vpmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
18667 uint8x8_t __ret;
1866718668 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1866818669 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18669 uint8x8_t __ret;
1867018670 __ret = (uint8x8_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
1867118671 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1867218672 return __ret;
......@@ -18681,9 +18681,9 @@ __ai uint32x2_t vpmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
1868118681}
1868218682#else
1868318683__ai uint32x2_t vpmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
18684 uint32x2_t __ret;
1868418685 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1868518686 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18686 uint32x2_t __ret;
1868718687 __ret = (uint32x2_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1868818688 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1868918689 return __ret;
......@@ -18698,9 +18698,9 @@ __ai uint16x4_t vpmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
1869818698}
1869918699#else
1870018700__ai uint16x4_t vpmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
18701 uint16x4_t __ret;
1870118702 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1870218703 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18703 uint16x4_t __ret;
1870418704 __ret = (uint16x4_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1870518705 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1870618706 return __ret;
......@@ -18715,9 +18715,9 @@ __ai int8x8_t vpmax_s8(int8x8_t __p0, int8x8_t __p1) {
1871518715}
1871618716#else
1871718717__ai int8x8_t vpmax_s8(int8x8_t __p0, int8x8_t __p1) {
18718 int8x8_t __ret;
1871818719 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1871918720 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18720 int8x8_t __ret;
1872118721 __ret = (int8x8_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
1872218722 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1872318723 return __ret;
......@@ -18732,9 +18732,9 @@ __ai float32x2_t vpmax_f32(float32x2_t __p0, float32x2_t __p1) {
1873218732}
1873318733#else
1873418734__ai float32x2_t vpmax_f32(float32x2_t __p0, float32x2_t __p1) {
18735 float32x2_t __ret;
1873518736 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1873618737 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18737 float32x2_t __ret;
1873818738 __ret = (float32x2_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
1873918739 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1874018740 return __ret;
......@@ -18749,9 +18749,9 @@ __ai int32x2_t vpmax_s32(int32x2_t __p0, int32x2_t __p1) {
1874918749}
1875018750#else
1875118751__ai int32x2_t vpmax_s32(int32x2_t __p0, int32x2_t __p1) {
18752 int32x2_t __ret;
1875218753 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1875318754 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18754 int32x2_t __ret;
1875518755 __ret = (int32x2_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1875618756 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1875718757 return __ret;
......@@ -18766,9 +18766,9 @@ __ai int16x4_t vpmax_s16(int16x4_t __p0, int16x4_t __p1) {
1876618766}
1876718767#else
1876818768__ai int16x4_t vpmax_s16(int16x4_t __p0, int16x4_t __p1) {
18769 int16x4_t __ret;
1876918770 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1877018771 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18771 int16x4_t __ret;
1877218772 __ret = (int16x4_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1877318773 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1877418774 return __ret;
......@@ -18783,9 +18783,9 @@ __ai uint8x8_t vpmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
1878318783}
1878418784#else
1878518785__ai uint8x8_t vpmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
18786 uint8x8_t __ret;
1878618787 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1878718788 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18788 uint8x8_t __ret;
1878918789 __ret = (uint8x8_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
1879018790 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1879118791 return __ret;
......@@ -18800,9 +18800,9 @@ __ai uint32x2_t vpmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
1880018800}
1880118801#else
1880218802__ai uint32x2_t vpmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
18803 uint32x2_t __ret;
1880318804 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1880418805 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18805 uint32x2_t __ret;
1880618806 __ret = (uint32x2_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1880718807 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1880818808 return __ret;
......@@ -18817,9 +18817,9 @@ __ai uint16x4_t vpmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
1881718817}
1881818818#else
1881918819__ai uint16x4_t vpmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
18820 uint16x4_t __ret;
1882018821 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1882118822 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18822 uint16x4_t __ret;
1882318823 __ret = (uint16x4_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1882418824 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1882518825 return __ret;
......@@ -18834,9 +18834,9 @@ __ai int8x8_t vpmin_s8(int8x8_t __p0, int8x8_t __p1) {
1883418834}
1883518835#else
1883618836__ai int8x8_t vpmin_s8(int8x8_t __p0, int8x8_t __p1) {
18837 int8x8_t __ret;
1883718838 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1883818839 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18839 int8x8_t __ret;
1884018840 __ret = (int8x8_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
1884118841 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1884218842 return __ret;
......@@ -18851,9 +18851,9 @@ __ai float32x2_t vpmin_f32(float32x2_t __p0, float32x2_t __p1) {
1885118851}
1885218852#else
1885318853__ai float32x2_t vpmin_f32(float32x2_t __p0, float32x2_t __p1) {
18854 float32x2_t __ret;
1885418855 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1885518856 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18856 float32x2_t __ret;
1885718857 __ret = (float32x2_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
1885818858 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1885918859 return __ret;
......@@ -18868,9 +18868,9 @@ __ai int32x2_t vpmin_s32(int32x2_t __p0, int32x2_t __p1) {
1886818868}
1886918869#else
1887018870__ai int32x2_t vpmin_s32(int32x2_t __p0, int32x2_t __p1) {
18871 int32x2_t __ret;
1887118872 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1887218873 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18873 int32x2_t __ret;
1887418874 __ret = (int32x2_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1887518875 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1887618876 return __ret;
......@@ -18885,9 +18885,9 @@ __ai int16x4_t vpmin_s16(int16x4_t __p0, int16x4_t __p1) {
1888518885}
1888618886#else
1888718887__ai int16x4_t vpmin_s16(int16x4_t __p0, int16x4_t __p1) {
18888 int16x4_t __ret;
1888818889 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1888918890 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18890 int16x4_t __ret;
1889118891 __ret = (int16x4_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1889218892 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1889318893 return __ret;
......@@ -18902,8 +18902,8 @@ __ai int8x16_t vqabsq_s8(int8x16_t __p0) {
1890218902}
1890318903#else
1890418904__ai int8x16_t vqabsq_s8(int8x16_t __p0) {
18905 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1890618905 int8x16_t __ret;
18906 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1890718907 __ret = (int8x16_t) __builtin_neon_vqabsq_v((int8x16_t)__rev0, 32);
1890818908 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1890918909 return __ret;
......@@ -18918,8 +18918,8 @@ __ai int32x4_t vqabsq_s32(int32x4_t __p0) {
1891818918}
1891918919#else
1892018920__ai int32x4_t vqabsq_s32(int32x4_t __p0) {
18921 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1892218921 int32x4_t __ret;
18922 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1892318923 __ret = (int32x4_t) __builtin_neon_vqabsq_v((int8x16_t)__rev0, 34);
1892418924 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1892518925 return __ret;
......@@ -18934,8 +18934,8 @@ __ai int16x8_t vqabsq_s16(int16x8_t __p0) {
1893418934}
1893518935#else
1893618936__ai int16x8_t vqabsq_s16(int16x8_t __p0) {
18937 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1893818937 int16x8_t __ret;
18938 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1893918939 __ret = (int16x8_t) __builtin_neon_vqabsq_v((int8x16_t)__rev0, 33);
1894018940 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1894118941 return __ret;
......@@ -18950,8 +18950,8 @@ __ai int8x8_t vqabs_s8(int8x8_t __p0) {
1895018950}
1895118951#else
1895218952__ai int8x8_t vqabs_s8(int8x8_t __p0) {
18953 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1895418953 int8x8_t __ret;
18954 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1895518955 __ret = (int8x8_t) __builtin_neon_vqabs_v((int8x8_t)__rev0, 0);
1895618956 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1895718957 return __ret;
......@@ -18966,8 +18966,8 @@ __ai int32x2_t vqabs_s32(int32x2_t __p0) {
1896618966}
1896718967#else
1896818968__ai int32x2_t vqabs_s32(int32x2_t __p0) {
18969 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1897018969 int32x2_t __ret;
18970 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1897118971 __ret = (int32x2_t) __builtin_neon_vqabs_v((int8x8_t)__rev0, 2);
1897218972 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1897318973 return __ret;
......@@ -18982,8 +18982,8 @@ __ai int16x4_t vqabs_s16(int16x4_t __p0) {
1898218982}
1898318983#else
1898418984__ai int16x4_t vqabs_s16(int16x4_t __p0) {
18985 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1898618985 int16x4_t __ret;
18986 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1898718987 __ret = (int16x4_t) __builtin_neon_vqabs_v((int8x8_t)__rev0, 1);
1898818988 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1898918989 return __ret;
......@@ -18998,9 +18998,9 @@ __ai uint8x16_t vqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1899818998}
1899918999#else
1900019000__ai uint8x16_t vqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
19001 uint8x16_t __ret;
1900119002 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1900219003 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
19003 uint8x16_t __ret;
1900419004 __ret = (uint8x16_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
1900519005 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1900619006 return __ret;
......@@ -19015,9 +19015,9 @@ __ai uint32x4_t vqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1901519015}
1901619016#else
1901719017__ai uint32x4_t vqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
19018 uint32x4_t __ret;
1901819019 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1901919020 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19020 uint32x4_t __ret;
1902119021 __ret = (uint32x4_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
1902219022 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1902319023 return __ret;
......@@ -19032,9 +19032,9 @@ __ai uint64x2_t vqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1903219032}
1903319033#else
1903419034__ai uint64x2_t vqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
19035 uint64x2_t __ret;
1903519036 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1903619037 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19037 uint64x2_t __ret;
1903819038 __ret = (uint64x2_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
1903919039 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1904019040 return __ret;
......@@ -19049,9 +19049,9 @@ __ai uint16x8_t vqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1904919049}
1905019050#else
1905119051__ai uint16x8_t vqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
19052 uint16x8_t __ret;
1905219053 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1905319054 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19054 uint16x8_t __ret;
1905519055 __ret = (uint16x8_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
1905619056 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1905719057 return __ret;
......@@ -19066,9 +19066,9 @@ __ai int8x16_t vqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
1906619066}
1906719067#else
1906819068__ai int8x16_t vqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
19069 int8x16_t __ret;
1906919070 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1907019071 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
19071 int8x16_t __ret;
1907219072 __ret = (int8x16_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
1907319073 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1907419074 return __ret;
......@@ -19083,9 +19083,9 @@ __ai int32x4_t vqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
1908319083}
1908419084#else
1908519085__ai int32x4_t vqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
19086 int32x4_t __ret;
1908619087 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1908719088 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19088 int32x4_t __ret;
1908919089 __ret = (int32x4_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
1909019090 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1909119091 return __ret;
......@@ -19100,9 +19100,9 @@ __ai int64x2_t vqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
1910019100}
1910119101#else
1910219102__ai int64x2_t vqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
19103 int64x2_t __ret;
1910319104 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1910419105 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19105 int64x2_t __ret;
1910619106 __ret = (int64x2_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
1910719107 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1910819108 return __ret;
......@@ -19117,9 +19117,9 @@ __ai int16x8_t vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
1911719117}
1911819118#else
1911919119__ai int16x8_t vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
19120 int16x8_t __ret;
1912019121 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1912119122 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19122 int16x8_t __ret;
1912319123 __ret = (int16x8_t) __builtin_neon_vqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
1912419124 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1912519125 return __ret;
......@@ -19134,9 +19134,9 @@ __ai uint8x8_t vqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
1913419134}
1913519135#else
1913619136__ai uint8x8_t vqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
19137 uint8x8_t __ret;
1913719138 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1913819139 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19139 uint8x8_t __ret;
1914019140 __ret = (uint8x8_t) __builtin_neon_vqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
1914119141 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1914219142 return __ret;
......@@ -19151,9 +19151,9 @@ __ai uint32x2_t vqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
1915119151}
1915219152#else
1915319153__ai uint32x2_t vqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
19154 uint32x2_t __ret;
1915419155 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1915519156 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19156 uint32x2_t __ret;
1915719157 __ret = (uint32x2_t) __builtin_neon_vqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
1915819158 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1915919159 return __ret;
......@@ -19173,9 +19173,9 @@ __ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
1917319173}
1917419174#else
1917519175__ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
19176 uint16x4_t __ret;
1917619177 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1917719178 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19178 uint16x4_t __ret;
1917919179 __ret = (uint16x4_t) __builtin_neon_vqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
1918019180 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1918119181 return __ret;
......@@ -19190,9 +19190,9 @@ __ai int8x8_t vqadd_s8(int8x8_t __p0, int8x8_t __p1) {
1919019190}
1919119191#else
1919219192__ai int8x8_t vqadd_s8(int8x8_t __p0, int8x8_t __p1) {
19193 int8x8_t __ret;
1919319194 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1919419195 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19195 int8x8_t __ret;
1919619196 __ret = (int8x8_t) __builtin_neon_vqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
1919719197 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1919819198 return __ret;
......@@ -19207,9 +19207,9 @@ __ai int32x2_t vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
1920719207}
1920819208#else
1920919209__ai int32x2_t vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
19210 int32x2_t __ret;
1921019211 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1921119212 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19212 int32x2_t __ret;
1921319213 __ret = (int32x2_t) __builtin_neon_vqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1921419214 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1921519215 return __ret;
......@@ -19229,9 +19229,9 @@ __ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
1922919229}
1923019230#else
1923119231__ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
19232 int16x4_t __ret;
1923219233 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1923319234 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19234 int16x4_t __ret;
1923519235 __ret = (int16x4_t) __builtin_neon_vqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1923619236 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1923719237 return __ret;
......@@ -19246,10 +19246,10 @@ __ai int64x2_t vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
1924619246}
1924719247#else
1924819248__ai int64x2_t vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
19249 int64x2_t __ret;
1924919250 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1925019251 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1925119252 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
19252 int64x2_t __ret;
1925319253 __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 35);
1925419254 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1925519255 return __ret;
......@@ -19269,10 +19269,10 @@ __ai int32x4_t vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1926919269}
1927019270#else
1927119271__ai int32x4_t vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
19272 int32x4_t __ret;
1927219273 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1927319274 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1927419275 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
19275 int32x4_t __ret;
1927619276 __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 34);
1927719277 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1927819278 return __ret;
......@@ -19286,22 +19286,22 @@ __ai int32x4_t __noswap_vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __
1928619286
1928719287#ifdef __LITTLE_ENDIAN__
1928819288#define vqdmlal_lane_s32(__p0_110, __p1_110, __p2_110, __p3_110) __extension__ ({ \
19289 int64x2_t __ret_110; \
1928919290 int64x2_t __s0_110 = __p0_110; \
1929019291 int32x2_t __s1_110 = __p1_110; \
1929119292 int32x2_t __s2_110 = __p2_110; \
19292 int64x2_t __ret_110; \
1929319293 __ret_110 = vqdmlal_s32(__s0_110, __s1_110, splat_lane_s32(__s2_110, __p3_110)); \
1929419294 __ret_110; \
1929519295})
1929619296#else
1929719297#define vqdmlal_lane_s32(__p0_111, __p1_111, __p2_111, __p3_111) __extension__ ({ \
19298 int64x2_t __ret_111; \
1929819299 int64x2_t __s0_111 = __p0_111; \
1929919300 int32x2_t __s1_111 = __p1_111; \
1930019301 int32x2_t __s2_111 = __p2_111; \
1930119302 int64x2_t __rev0_111; __rev0_111 = __builtin_shufflevector(__s0_111, __s0_111, 1, 0); \
1930219303 int32x2_t __rev1_111; __rev1_111 = __builtin_shufflevector(__s1_111, __s1_111, 1, 0); \
1930319304 int32x2_t __rev2_111; __rev2_111 = __builtin_shufflevector(__s2_111, __s2_111, 1, 0); \
19304 int64x2_t __ret_111; \
1930519305 __ret_111 = __noswap_vqdmlal_s32(__rev0_111, __rev1_111, __noswap_splat_lane_s32(__rev2_111, __p3_111)); \
1930619306 __ret_111 = __builtin_shufflevector(__ret_111, __ret_111, 1, 0); \
1930719307 __ret_111; \
......@@ -19310,22 +19310,22 @@ __ai int32x4_t __noswap_vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __
1931019310
1931119311#ifdef __LITTLE_ENDIAN__
1931219312#define vqdmlal_lane_s16(__p0_112, __p1_112, __p2_112, __p3_112) __extension__ ({ \
19313 int32x4_t __ret_112; \
1931319314 int32x4_t __s0_112 = __p0_112; \
1931419315 int16x4_t __s1_112 = __p1_112; \
1931519316 int16x4_t __s2_112 = __p2_112; \
19316 int32x4_t __ret_112; \
1931719317 __ret_112 = vqdmlal_s16(__s0_112, __s1_112, splat_lane_s16(__s2_112, __p3_112)); \
1931819318 __ret_112; \
1931919319})
1932019320#else
1932119321#define vqdmlal_lane_s16(__p0_113, __p1_113, __p2_113, __p3_113) __extension__ ({ \
19322 int32x4_t __ret_113; \
1932219323 int32x4_t __s0_113 = __p0_113; \
1932319324 int16x4_t __s1_113 = __p1_113; \
1932419325 int16x4_t __s2_113 = __p2_113; \
1932519326 int32x4_t __rev0_113; __rev0_113 = __builtin_shufflevector(__s0_113, __s0_113, 3, 2, 1, 0); \
1932619327 int16x4_t __rev1_113; __rev1_113 = __builtin_shufflevector(__s1_113, __s1_113, 3, 2, 1, 0); \
1932719328 int16x4_t __rev2_113; __rev2_113 = __builtin_shufflevector(__s2_113, __s2_113, 3, 2, 1, 0); \
19328 int32x4_t __ret_113; \
1932919329 __ret_113 = __noswap_vqdmlal_s16(__rev0_113, __rev1_113, __noswap_splat_lane_s16(__rev2_113, __p3_113)); \
1933019330 __ret_113 = __builtin_shufflevector(__ret_113, __ret_113, 3, 2, 1, 0); \
1933119331 __ret_113; \
......@@ -19340,9 +19340,9 @@ __ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
1934019340}
1934119341#else
1934219342__ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
19343 int64x2_t __ret;
1934319344 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1934419345 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19345 int64x2_t __ret;
1934619346 __ret = __noswap_vqdmlal_s32(__rev0, __rev1, (int32x2_t) {__p2, __p2});
1934719347 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1934819348 return __ret;
......@@ -19362,9 +19362,9 @@ __ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
1936219362}
1936319363#else
1936419364__ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
19365 int32x4_t __ret;
1936519366 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1936619367 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19367 int32x4_t __ret;
1936819368 __ret = __noswap_vqdmlal_s16(__rev0, __rev1, (int16x4_t) {__p2, __p2, __p2, __p2});
1936919369 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1937019370 return __ret;
......@@ -19384,10 +19384,10 @@ __ai int64x2_t vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
1938419384}
1938519385#else
1938619386__ai int64x2_t vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
19387 int64x2_t __ret;
1938719388 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1938819389 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1938919390 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
19390 int64x2_t __ret;
1939119391 __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 35);
1939219392 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1939319393 return __ret;
......@@ -19407,10 +19407,10 @@ __ai int32x4_t vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
1940719407}
1940819408#else
1940919409__ai int32x4_t vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
19410 int32x4_t __ret;
1941019411 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1941119412 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1941219413 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
19413 int32x4_t __ret;
1941419414 __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 34);
1941519415 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1941619416 return __ret;
......@@ -19424,22 +19424,22 @@ __ai int32x4_t __noswap_vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __
1942419424
1942519425#ifdef __LITTLE_ENDIAN__
1942619426#define vqdmlsl_lane_s32(__p0_114, __p1_114, __p2_114, __p3_114) __extension__ ({ \
19427 int64x2_t __ret_114; \
1942719428 int64x2_t __s0_114 = __p0_114; \
1942819429 int32x2_t __s1_114 = __p1_114; \
1942919430 int32x2_t __s2_114 = __p2_114; \
19430 int64x2_t __ret_114; \
1943119431 __ret_114 = vqdmlsl_s32(__s0_114, __s1_114, splat_lane_s32(__s2_114, __p3_114)); \
1943219432 __ret_114; \
1943319433})
1943419434#else
1943519435#define vqdmlsl_lane_s32(__p0_115, __p1_115, __p2_115, __p3_115) __extension__ ({ \
19436 int64x2_t __ret_115; \
1943619437 int64x2_t __s0_115 = __p0_115; \
1943719438 int32x2_t __s1_115 = __p1_115; \
1943819439 int32x2_t __s2_115 = __p2_115; \
1943919440 int64x2_t __rev0_115; __rev0_115 = __builtin_shufflevector(__s0_115, __s0_115, 1, 0); \
1944019441 int32x2_t __rev1_115; __rev1_115 = __builtin_shufflevector(__s1_115, __s1_115, 1, 0); \
1944119442 int32x2_t __rev2_115; __rev2_115 = __builtin_shufflevector(__s2_115, __s2_115, 1, 0); \
19442 int64x2_t __ret_115; \
1944319443 __ret_115 = __noswap_vqdmlsl_s32(__rev0_115, __rev1_115, __noswap_splat_lane_s32(__rev2_115, __p3_115)); \
1944419444 __ret_115 = __builtin_shufflevector(__ret_115, __ret_115, 1, 0); \
1944519445 __ret_115; \
......@@ -19448,22 +19448,22 @@ __ai int32x4_t __noswap_vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __
1944819448
1944919449#ifdef __LITTLE_ENDIAN__
1945019450#define vqdmlsl_lane_s16(__p0_116, __p1_116, __p2_116, __p3_116) __extension__ ({ \
19451 int32x4_t __ret_116; \
1945119452 int32x4_t __s0_116 = __p0_116; \
1945219453 int16x4_t __s1_116 = __p1_116; \
1945319454 int16x4_t __s2_116 = __p2_116; \
19454 int32x4_t __ret_116; \
1945519455 __ret_116 = vqdmlsl_s16(__s0_116, __s1_116, splat_lane_s16(__s2_116, __p3_116)); \
1945619456 __ret_116; \
1945719457})
1945819458#else
1945919459#define vqdmlsl_lane_s16(__p0_117, __p1_117, __p2_117, __p3_117) __extension__ ({ \
19460 int32x4_t __ret_117; \
1946019461 int32x4_t __s0_117 = __p0_117; \
1946119462 int16x4_t __s1_117 = __p1_117; \
1946219463 int16x4_t __s2_117 = __p2_117; \
1946319464 int32x4_t __rev0_117; __rev0_117 = __builtin_shufflevector(__s0_117, __s0_117, 3, 2, 1, 0); \
1946419465 int16x4_t __rev1_117; __rev1_117 = __builtin_shufflevector(__s1_117, __s1_117, 3, 2, 1, 0); \
1946519466 int16x4_t __rev2_117; __rev2_117 = __builtin_shufflevector(__s2_117, __s2_117, 3, 2, 1, 0); \
19466 int32x4_t __ret_117; \
1946719467 __ret_117 = __noswap_vqdmlsl_s16(__rev0_117, __rev1_117, __noswap_splat_lane_s16(__rev2_117, __p3_117)); \
1946819468 __ret_117 = __builtin_shufflevector(__ret_117, __ret_117, 3, 2, 1, 0); \
1946919469 __ret_117; \
......@@ -19478,9 +19478,9 @@ __ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
1947819478}
1947919479#else
1948019480__ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
19481 int64x2_t __ret;
1948119482 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1948219483 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19483 int64x2_t __ret;
1948419484 __ret = __noswap_vqdmlsl_s32(__rev0, __rev1, (int32x2_t) {__p2, __p2});
1948519485 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1948619486 return __ret;
......@@ -19500,9 +19500,9 @@ __ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
1950019500}
1950119501#else
1950219502__ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
19503 int32x4_t __ret;
1950319504 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1950419505 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19505 int32x4_t __ret;
1950619506 __ret = __noswap_vqdmlsl_s16(__rev0, __rev1, (int16x4_t) {__p2, __p2, __p2, __p2});
1950719507 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1950819508 return __ret;
......@@ -19522,9 +19522,9 @@ __ai int32x4_t vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
1952219522}
1952319523#else
1952419524__ai int32x4_t vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
19525 int32x4_t __ret;
1952519526 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1952619527 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19527 int32x4_t __ret;
1952819528 __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
1952919529 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1953019530 return __ret;
......@@ -19544,9 +19544,9 @@ __ai int16x8_t vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
1954419544}
1954519545#else
1954619546__ai int16x8_t vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
19547 int16x8_t __ret;
1954719548 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1954819549 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19549 int16x8_t __ret;
1955019550 __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
1955119551 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1955219552 return __ret;
......@@ -19566,9 +19566,9 @@ __ai int32x2_t vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
1956619566}
1956719567#else
1956819568__ai int32x2_t vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
19569 int32x2_t __ret;
1956919570 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1957019571 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19571 int32x2_t __ret;
1957219572 __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
1957319573 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1957419574 return __ret;
......@@ -19588,9 +19588,9 @@ __ai int16x4_t vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
1958819588}
1958919589#else
1959019590__ai int16x4_t vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
19591 int16x4_t __ret;
1959119592 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1959219593 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19593 int16x4_t __ret;
1959419594 __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
1959519595 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1959619596 return __ret;
......@@ -19610,8 +19610,8 @@ __ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
1961019610}
1961119611#else
1961219612__ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
19613 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1961419613 int32x4_t __ret;
19614 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1961519615 __ret = __noswap_vqdmulhq_s32(__rev0, (int32x4_t) {__p1, __p1, __p1, __p1});
1961619616 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1961719617 return __ret;
......@@ -19626,8 +19626,8 @@ __ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
1962619626}
1962719627#else
1962819628__ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
19629 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1963019629 int16x8_t __ret;
19630 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1963119631 __ret = __noswap_vqdmulhq_s16(__rev0, (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1});
1963219632 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1963319633 return __ret;
......@@ -19642,8 +19642,8 @@ __ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
1964219642}
1964319643#else
1964419644__ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
19645 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1964619645 int32x2_t __ret;
19646 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1964719647 __ret = __noswap_vqdmulh_s32(__rev0, (int32x2_t) {__p1, __p1});
1964819648 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1964919649 return __ret;
......@@ -19658,8 +19658,8 @@ __ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
1965819658}
1965919659#else
1966019660__ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
19661 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1966219661 int16x4_t __ret;
19662 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1966319663 __ret = __noswap_vqdmulh_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
1966419664 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1966519665 return __ret;
......@@ -19674,9 +19674,9 @@ __ai int64x2_t vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
1967419674}
1967519675#else
1967619676__ai int64x2_t vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
19677 int64x2_t __ret;
1967719678 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1967819679 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19679 int64x2_t __ret;
1968019680 __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 35);
1968119681 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1968219682 return __ret;
......@@ -19696,9 +19696,9 @@ __ai int32x4_t vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
1969619696}
1969719697#else
1969819698__ai int32x4_t vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
19699 int32x4_t __ret;
1969919700 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1970019701 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19701 int32x4_t __ret;
1970219702 __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)__rev1, 34);
1970319703 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1970419704 return __ret;
......@@ -19712,19 +19712,19 @@ __ai int32x4_t __noswap_vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
1971219712
1971319713#ifdef __LITTLE_ENDIAN__
1971419714#define vqdmull_lane_s32(__p0_118, __p1_118, __p2_118) __extension__ ({ \
19715 int64x2_t __ret_118; \
1971519716 int32x2_t __s0_118 = __p0_118; \
1971619717 int32x2_t __s1_118 = __p1_118; \
19717 int64x2_t __ret_118; \
1971819718 __ret_118 = vqdmull_s32(__s0_118, splat_lane_s32(__s1_118, __p2_118)); \
1971919719 __ret_118; \
1972019720})
1972119721#else
1972219722#define vqdmull_lane_s32(__p0_119, __p1_119, __p2_119) __extension__ ({ \
19723 int64x2_t __ret_119; \
1972319724 int32x2_t __s0_119 = __p0_119; \
1972419725 int32x2_t __s1_119 = __p1_119; \
1972519726 int32x2_t __rev0_119; __rev0_119 = __builtin_shufflevector(__s0_119, __s0_119, 1, 0); \
1972619727 int32x2_t __rev1_119; __rev1_119 = __builtin_shufflevector(__s1_119, __s1_119, 1, 0); \
19727 int64x2_t __ret_119; \
1972819728 __ret_119 = __noswap_vqdmull_s32(__rev0_119, __noswap_splat_lane_s32(__rev1_119, __p2_119)); \
1972919729 __ret_119 = __builtin_shufflevector(__ret_119, __ret_119, 1, 0); \
1973019730 __ret_119; \
......@@ -19733,19 +19733,19 @@ __ai int32x4_t __noswap_vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
1973319733
1973419734#ifdef __LITTLE_ENDIAN__
1973519735#define vqdmull_lane_s16(__p0_120, __p1_120, __p2_120) __extension__ ({ \
19736 int32x4_t __ret_120; \
1973619737 int16x4_t __s0_120 = __p0_120; \
1973719738 int16x4_t __s1_120 = __p1_120; \
19738 int32x4_t __ret_120; \
1973919739 __ret_120 = vqdmull_s16(__s0_120, splat_lane_s16(__s1_120, __p2_120)); \
1974019740 __ret_120; \
1974119741})
1974219742#else
1974319743#define vqdmull_lane_s16(__p0_121, __p1_121, __p2_121) __extension__ ({ \
19744 int32x4_t __ret_121; \
1974419745 int16x4_t __s0_121 = __p0_121; \
1974519746 int16x4_t __s1_121 = __p1_121; \
1974619747 int16x4_t __rev0_121; __rev0_121 = __builtin_shufflevector(__s0_121, __s0_121, 3, 2, 1, 0); \
1974719748 int16x4_t __rev1_121; __rev1_121 = __builtin_shufflevector(__s1_121, __s1_121, 3, 2, 1, 0); \
19748 int32x4_t __ret_121; \
1974919749 __ret_121 = __noswap_vqdmull_s16(__rev0_121, __noswap_splat_lane_s16(__rev1_121, __p2_121)); \
1975019750 __ret_121 = __builtin_shufflevector(__ret_121, __ret_121, 3, 2, 1, 0); \
1975119751 __ret_121; \
......@@ -19760,8 +19760,8 @@ __ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
1976019760}
1976119761#else
1976219762__ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
19763 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1976419763 int64x2_t __ret;
19764 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1976519765 __ret = __noswap_vqdmull_s32(__rev0, (int32x2_t) {__p1, __p1});
1976619766 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1976719767 return __ret;
......@@ -19781,8 +19781,8 @@ __ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
1978119781}
1978219782#else
1978319783__ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
19784 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1978519784 int32x4_t __ret;
19785 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1978619786 __ret = __noswap_vqdmull_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
1978719787 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1978819788 return __ret;
......@@ -19802,8 +19802,8 @@ __ai uint16x4_t vqmovn_u32(uint32x4_t __p0) {
1980219802}
1980319803#else
1980419804__ai uint16x4_t vqmovn_u32(uint32x4_t __p0) {
19805 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1980619805 uint16x4_t __ret;
19806 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1980719807 __ret = (uint16x4_t) __builtin_neon_vqmovn_v((int8x16_t)__rev0, 17);
1980819808 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1980919809 return __ret;
......@@ -19823,8 +19823,8 @@ __ai uint32x2_t vqmovn_u64(uint64x2_t __p0) {
1982319823}
1982419824#else
1982519825__ai uint32x2_t vqmovn_u64(uint64x2_t __p0) {
19826 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1982719826 uint32x2_t __ret;
19827 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1982819828 __ret = (uint32x2_t) __builtin_neon_vqmovn_v((int8x16_t)__rev0, 18);
1982919829 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1983019830 return __ret;
......@@ -19844,8 +19844,8 @@ __ai uint8x8_t vqmovn_u16(uint16x8_t __p0) {
1984419844}
1984519845#else
1984619846__ai uint8x8_t vqmovn_u16(uint16x8_t __p0) {
19847 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1984819847 uint8x8_t __ret;
19848 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1984919849 __ret = (uint8x8_t) __builtin_neon_vqmovn_v((int8x16_t)__rev0, 16);
1985019850 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1985119851 return __ret;
......@@ -19865,8 +19865,8 @@ __ai int16x4_t vqmovn_s32(int32x4_t __p0) {
1986519865}
1986619866#else
1986719867__ai int16x4_t vqmovn_s32(int32x4_t __p0) {
19868 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1986919868 int16x4_t __ret;
19869 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1987019870 __ret = (int16x4_t) __builtin_neon_vqmovn_v((int8x16_t)__rev0, 1);
1987119871 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1987219872 return __ret;
......@@ -19886,8 +19886,8 @@ __ai int32x2_t vqmovn_s64(int64x2_t __p0) {
1988619886}
1988719887#else
1988819888__ai int32x2_t vqmovn_s64(int64x2_t __p0) {
19889 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1989019889 int32x2_t __ret;
19890 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1989119891 __ret = (int32x2_t) __builtin_neon_vqmovn_v((int8x16_t)__rev0, 2);
1989219892 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1989319893 return __ret;
......@@ -19907,8 +19907,8 @@ __ai int8x8_t vqmovn_s16(int16x8_t __p0) {
1990719907}
1990819908#else
1990919909__ai int8x8_t vqmovn_s16(int16x8_t __p0) {
19910 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1991119910 int8x8_t __ret;
19911 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1991219912 __ret = (int8x8_t) __builtin_neon_vqmovn_v((int8x16_t)__rev0, 0);
1991319913 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1991419914 return __ret;
......@@ -19928,8 +19928,8 @@ __ai uint16x4_t vqmovun_s32(int32x4_t __p0) {
1992819928}
1992919929#else
1993019930__ai uint16x4_t vqmovun_s32(int32x4_t __p0) {
19931 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1993219931 uint16x4_t __ret;
19932 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
1993319933 __ret = (uint16x4_t) __builtin_neon_vqmovun_v((int8x16_t)__rev0, 17);
1993419934 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
1993519935 return __ret;
......@@ -19949,8 +19949,8 @@ __ai uint32x2_t vqmovun_s64(int64x2_t __p0) {
1994919949}
1995019950#else
1995119951__ai uint32x2_t vqmovun_s64(int64x2_t __p0) {
19952 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1995319952 uint32x2_t __ret;
19953 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
1995419954 __ret = (uint32x2_t) __builtin_neon_vqmovun_v((int8x16_t)__rev0, 18);
1995519955 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
1995619956 return __ret;
......@@ -19970,8 +19970,8 @@ __ai uint8x8_t vqmovun_s16(int16x8_t __p0) {
1997019970}
1997119971#else
1997219972__ai uint8x8_t vqmovun_s16(int16x8_t __p0) {
19973 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1997419973 uint8x8_t __ret;
19974 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
1997519975 __ret = (uint8x8_t) __builtin_neon_vqmovun_v((int8x16_t)__rev0, 16);
1997619976 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
1997719977 return __ret;
......@@ -19991,8 +19991,8 @@ __ai int8x16_t vqnegq_s8(int8x16_t __p0) {
1999119991}
1999219992#else
1999319993__ai int8x16_t vqnegq_s8(int8x16_t __p0) {
19994 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1999519994 int8x16_t __ret;
19995 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1999619996 __ret = (int8x16_t) __builtin_neon_vqnegq_v((int8x16_t)__rev0, 32);
1999719997 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1999819998 return __ret;
......@@ -20007,8 +20007,8 @@ __ai int32x4_t vqnegq_s32(int32x4_t __p0) {
2000720007}
2000820008#else
2000920009__ai int32x4_t vqnegq_s32(int32x4_t __p0) {
20010 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2001120010 int32x4_t __ret;
20011 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2001220012 __ret = (int32x4_t) __builtin_neon_vqnegq_v((int8x16_t)__rev0, 34);
2001320013 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2001420014 return __ret;
......@@ -20023,8 +20023,8 @@ __ai int16x8_t vqnegq_s16(int16x8_t __p0) {
2002320023}
2002420024#else
2002520025__ai int16x8_t vqnegq_s16(int16x8_t __p0) {
20026 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2002720026 int16x8_t __ret;
20027 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2002820028 __ret = (int16x8_t) __builtin_neon_vqnegq_v((int8x16_t)__rev0, 33);
2002920029 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2003020030 return __ret;
......@@ -20039,8 +20039,8 @@ __ai int8x8_t vqneg_s8(int8x8_t __p0) {
2003920039}
2004020040#else
2004120041__ai int8x8_t vqneg_s8(int8x8_t __p0) {
20042 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2004320042 int8x8_t __ret;
20043 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2004420044 __ret = (int8x8_t) __builtin_neon_vqneg_v((int8x8_t)__rev0, 0);
2004520045 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2004620046 return __ret;
......@@ -20055,8 +20055,8 @@ __ai int32x2_t vqneg_s32(int32x2_t __p0) {
2005520055}
2005620056#else
2005720057__ai int32x2_t vqneg_s32(int32x2_t __p0) {
20058 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2005920058 int32x2_t __ret;
20059 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2006020060 __ret = (int32x2_t) __builtin_neon_vqneg_v((int8x8_t)__rev0, 2);
2006120061 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2006220062 return __ret;
......@@ -20071,8 +20071,8 @@ __ai int16x4_t vqneg_s16(int16x4_t __p0) {
2007120071}
2007220072#else
2007320073__ai int16x4_t vqneg_s16(int16x4_t __p0) {
20074 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2007520074 int16x4_t __ret;
20075 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2007620076 __ret = (int16x4_t) __builtin_neon_vqneg_v((int8x8_t)__rev0, 1);
2007720077 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2007820078 return __ret;
......@@ -20087,9 +20087,9 @@ __ai int32x4_t vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
2008720087}
2008820088#else
2008920089__ai int32x4_t vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
20090 int32x4_t __ret;
2009020091 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2009120092 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20092 int32x4_t __ret;
2009320093 __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2009420094 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2009520095 return __ret;
......@@ -20109,9 +20109,9 @@ __ai int16x8_t vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
2010920109}
2011020110#else
2011120111__ai int16x8_t vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
20112 int16x8_t __ret;
2011220113 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2011320114 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20114 int16x8_t __ret;
2011520115 __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2011620116 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2011720117 return __ret;
......@@ -20131,9 +20131,9 @@ __ai int32x2_t vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
2013120131}
2013220132#else
2013320133__ai int32x2_t vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
20134 int32x2_t __ret;
2013420135 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2013520136 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20136 int32x2_t __ret;
2013720137 __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2013820138 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2013920139 return __ret;
......@@ -20153,9 +20153,9 @@ __ai int16x4_t vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
2015320153}
2015420154#else
2015520155__ai int16x4_t vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
20156 int16x4_t __ret;
2015620157 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2015720158 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20158 int16x4_t __ret;
2015920159 __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2016020160 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2016120161 return __ret;
......@@ -20175,8 +20175,8 @@ __ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
2017520175}
2017620176#else
2017720177__ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
20178 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2017920178 int32x4_t __ret;
20179 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2018020180 __ret = __noswap_vqrdmulhq_s32(__rev0, (int32x4_t) {__p1, __p1, __p1, __p1});
2018120181 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2018220182 return __ret;
......@@ -20191,8 +20191,8 @@ __ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
2019120191}
2019220192#else
2019320193__ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
20194 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2019520194 int16x8_t __ret;
20195 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2019620196 __ret = __noswap_vqrdmulhq_s16(__rev0, (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1});
2019720197 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2019820198 return __ret;
......@@ -20207,8 +20207,8 @@ __ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
2020720207}
2020820208#else
2020920209__ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
20210 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2021120210 int32x2_t __ret;
20211 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2021220212 __ret = __noswap_vqrdmulh_s32(__rev0, (int32x2_t) {__p1, __p1});
2021320213 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2021420214 return __ret;
......@@ -20223,8 +20223,8 @@ __ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
2022320223}
2022420224#else
2022520225__ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
20226 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2022720226 int16x4_t __ret;
20227 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2022820228 __ret = __noswap_vqrdmulh_s16(__rev0, (int16x4_t) {__p1, __p1, __p1, __p1});
2022920229 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2023020230 return __ret;
......@@ -20239,9 +20239,9 @@ __ai uint8x16_t vqrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
2023920239}
2024020240#else
2024120241__ai uint8x16_t vqrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
20242 uint8x16_t __ret;
2024220243 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2024320244 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20244 uint8x16_t __ret;
2024520245 __ret = (uint8x16_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
2024620246 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2024720247 return __ret;
......@@ -20256,9 +20256,9 @@ __ai uint32x4_t vqrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
2025620256}
2025720257#else
2025820258__ai uint32x4_t vqrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
20259 uint32x4_t __ret;
2025920260 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2026020261 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20261 uint32x4_t __ret;
2026220262 __ret = (uint32x4_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
2026320263 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2026420264 return __ret;
......@@ -20273,9 +20273,9 @@ __ai uint64x2_t vqrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
2027320273}
2027420274#else
2027520275__ai uint64x2_t vqrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
20276 uint64x2_t __ret;
2027620277 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2027720278 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20278 uint64x2_t __ret;
2027920279 __ret = (uint64x2_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
2028020280 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2028120281 return __ret;
......@@ -20290,9 +20290,9 @@ __ai uint16x8_t vqrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
2029020290}
2029120291#else
2029220292__ai uint16x8_t vqrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
20293 uint16x8_t __ret;
2029320294 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2029420295 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20295 uint16x8_t __ret;
2029620296 __ret = (uint16x8_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
2029720297 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2029820298 return __ret;
......@@ -20307,9 +20307,9 @@ __ai int8x16_t vqrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
2030720307}
2030820308#else
2030920309__ai int8x16_t vqrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
20310 int8x16_t __ret;
2031020311 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2031120312 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20312 int8x16_t __ret;
2031320313 __ret = (int8x16_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
2031420314 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2031520315 return __ret;
......@@ -20324,9 +20324,9 @@ __ai int32x4_t vqrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
2032420324}
2032520325#else
2032620326__ai int32x4_t vqrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
20327 int32x4_t __ret;
2032720328 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2032820329 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20329 int32x4_t __ret;
2033020330 __ret = (int32x4_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2033120331 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2033220332 return __ret;
......@@ -20341,9 +20341,9 @@ __ai int64x2_t vqrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
2034120341}
2034220342#else
2034320343__ai int64x2_t vqrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
20344 int64x2_t __ret;
2034420345 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2034520346 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20346 int64x2_t __ret;
2034720347 __ret = (int64x2_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
2034820348 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2034920349 return __ret;
......@@ -20358,9 +20358,9 @@ __ai int16x8_t vqrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
2035820358}
2035920359#else
2036020360__ai int16x8_t vqrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
20361 int16x8_t __ret;
2036120362 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2036220363 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20363 int16x8_t __ret;
2036420364 __ret = (int16x8_t) __builtin_neon_vqrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2036520365 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2036620366 return __ret;
......@@ -20375,9 +20375,9 @@ __ai uint8x8_t vqrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
2037520375}
2037620376#else
2037720377__ai uint8x8_t vqrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
20378 uint8x8_t __ret;
2037820379 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2037920380 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20380 uint8x8_t __ret;
2038120381 __ret = (uint8x8_t) __builtin_neon_vqrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
2038220382 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2038320383 return __ret;
......@@ -20392,9 +20392,9 @@ __ai uint32x2_t vqrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
2039220392}
2039320393#else
2039420394__ai uint32x2_t vqrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
20395 uint32x2_t __ret;
2039520396 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2039620397 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20397 uint32x2_t __ret;
2039820398 __ret = (uint32x2_t) __builtin_neon_vqrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
2039920399 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2040020400 return __ret;
......@@ -20414,9 +20414,9 @@ __ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
2041420414}
2041520415#else
2041620416__ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
20417 uint16x4_t __ret;
2041720418 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2041820419 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20419 uint16x4_t __ret;
2042020420 __ret = (uint16x4_t) __builtin_neon_vqrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
2042120421 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2042220422 return __ret;
......@@ -20431,9 +20431,9 @@ __ai int8x8_t vqrshl_s8(int8x8_t __p0, int8x8_t __p1) {
2043120431}
2043220432#else
2043320433__ai int8x8_t vqrshl_s8(int8x8_t __p0, int8x8_t __p1) {
20434 int8x8_t __ret;
2043420435 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2043520436 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20436 int8x8_t __ret;
2043720437 __ret = (int8x8_t) __builtin_neon_vqrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
2043820438 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2043920439 return __ret;
......@@ -20448,9 +20448,9 @@ __ai int32x2_t vqrshl_s32(int32x2_t __p0, int32x2_t __p1) {
2044820448}
2044920449#else
2045020450__ai int32x2_t vqrshl_s32(int32x2_t __p0, int32x2_t __p1) {
20451 int32x2_t __ret;
2045120452 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2045220453 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20453 int32x2_t __ret;
2045420454 __ret = (int32x2_t) __builtin_neon_vqrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2045520455 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2045620456 return __ret;
......@@ -20470,9 +20470,9 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2047020470}
2047120471#else
2047220472__ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
20473 int16x4_t __ret;
2047320474 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2047420475 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20475 int16x4_t __ret;
2047620476 __ret = (int16x4_t) __builtin_neon_vqrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2047720477 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2047820478 return __ret;
......@@ -20481,23 +20481,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2048120481
2048220482#ifdef __LITTLE_ENDIAN__
2048320483#define vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
20484 uint32x4_t __s0 = __p0; \
2048520484 uint16x4_t __ret; \
20485 uint32x4_t __s0 = __p0; \
2048620486 __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 17); \
2048720487 __ret; \
2048820488})
2048920489#else
2049020490#define vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
20491 uint16x4_t __ret; \
2049120492 uint32x4_t __s0 = __p0; \
2049220493 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
20493 uint16x4_t __ret; \
2049420494 __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 17); \
2049520495 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2049620496 __ret; \
2049720497})
2049820498#define __noswap_vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
20499 uint32x4_t __s0 = __p0; \
2050020499 uint16x4_t __ret; \
20500 uint32x4_t __s0 = __p0; \
2050120501 __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 17); \
2050220502 __ret; \
2050320503})
......@@ -20505,23 +20505,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2050520505
2050620506#ifdef __LITTLE_ENDIAN__
2050720507#define vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
20508 uint64x2_t __s0 = __p0; \
2050920508 uint32x2_t __ret; \
20509 uint64x2_t __s0 = __p0; \
2051020510 __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 18); \
2051120511 __ret; \
2051220512})
2051320513#else
2051420514#define vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
20515 uint32x2_t __ret; \
2051520516 uint64x2_t __s0 = __p0; \
2051620517 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
20517 uint32x2_t __ret; \
2051820518 __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 18); \
2051920519 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2052020520 __ret; \
2052120521})
2052220522#define __noswap_vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
20523 uint64x2_t __s0 = __p0; \
2052420523 uint32x2_t __ret; \
20524 uint64x2_t __s0 = __p0; \
2052520525 __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 18); \
2052620526 __ret; \
2052720527})
......@@ -20529,23 +20529,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2052920529
2053020530#ifdef __LITTLE_ENDIAN__
2053120531#define vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
20532 uint16x8_t __s0 = __p0; \
2053320532 uint8x8_t __ret; \
20533 uint16x8_t __s0 = __p0; \
2053420534 __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 16); \
2053520535 __ret; \
2053620536})
2053720537#else
2053820538#define vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
20539 uint8x8_t __ret; \
2053920540 uint16x8_t __s0 = __p0; \
2054020541 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
20541 uint8x8_t __ret; \
2054220542 __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 16); \
2054320543 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2054420544 __ret; \
2054520545})
2054620546#define __noswap_vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
20547 uint16x8_t __s0 = __p0; \
2054820547 uint8x8_t __ret; \
20548 uint16x8_t __s0 = __p0; \
2054920549 __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 16); \
2055020550 __ret; \
2055120551})
......@@ -20553,23 +20553,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2055320553
2055420554#ifdef __LITTLE_ENDIAN__
2055520555#define vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
20556 int32x4_t __s0 = __p0; \
2055720556 int16x4_t __ret; \
20557 int32x4_t __s0 = __p0; \
2055820558 __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 1); \
2055920559 __ret; \
2056020560})
2056120561#else
2056220562#define vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
20563 int16x4_t __ret; \
2056320564 int32x4_t __s0 = __p0; \
2056420565 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
20565 int16x4_t __ret; \
2056620566 __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 1); \
2056720567 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2056820568 __ret; \
2056920569})
2057020570#define __noswap_vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
20571 int32x4_t __s0 = __p0; \
2057220571 int16x4_t __ret; \
20572 int32x4_t __s0 = __p0; \
2057320573 __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 1); \
2057420574 __ret; \
2057520575})
......@@ -20577,23 +20577,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2057720577
2057820578#ifdef __LITTLE_ENDIAN__
2057920579#define vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
20580 int64x2_t __s0 = __p0; \
2058120580 int32x2_t __ret; \
20581 int64x2_t __s0 = __p0; \
2058220582 __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 2); \
2058320583 __ret; \
2058420584})
2058520585#else
2058620586#define vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
20587 int32x2_t __ret; \
2058720588 int64x2_t __s0 = __p0; \
2058820589 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
20589 int32x2_t __ret; \
2059020590 __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 2); \
2059120591 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2059220592 __ret; \
2059320593})
2059420594#define __noswap_vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
20595 int64x2_t __s0 = __p0; \
2059620595 int32x2_t __ret; \
20596 int64x2_t __s0 = __p0; \
2059720597 __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 2); \
2059820598 __ret; \
2059920599})
......@@ -20601,23 +20601,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2060120601
2060220602#ifdef __LITTLE_ENDIAN__
2060320603#define vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
20604 int16x8_t __s0 = __p0; \
2060520604 int8x8_t __ret; \
20605 int16x8_t __s0 = __p0; \
2060620606 __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 0); \
2060720607 __ret; \
2060820608})
2060920609#else
2061020610#define vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
20611 int8x8_t __ret; \
2061120612 int16x8_t __s0 = __p0; \
2061220613 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
20613 int8x8_t __ret; \
2061420614 __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 0); \
2061520615 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2061620616 __ret; \
2061720617})
2061820618#define __noswap_vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
20619 int16x8_t __s0 = __p0; \
2062020619 int8x8_t __ret; \
20620 int16x8_t __s0 = __p0; \
2062120621 __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 0); \
2062220622 __ret; \
2062320623})
......@@ -20625,23 +20625,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2062520625
2062620626#ifdef __LITTLE_ENDIAN__
2062720627#define vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
20628 int32x4_t __s0 = __p0; \
2062920628 uint16x4_t __ret; \
20629 int32x4_t __s0 = __p0; \
2063020630 __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 17); \
2063120631 __ret; \
2063220632})
2063320633#else
2063420634#define vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
20635 uint16x4_t __ret; \
2063520636 int32x4_t __s0 = __p0; \
2063620637 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
20637 uint16x4_t __ret; \
2063820638 __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 17); \
2063920639 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2064020640 __ret; \
2064120641})
2064220642#define __noswap_vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
20643 int32x4_t __s0 = __p0; \
2064420643 uint16x4_t __ret; \
20644 int32x4_t __s0 = __p0; \
2064520645 __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 17); \
2064620646 __ret; \
2064720647})
......@@ -20649,23 +20649,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2064920649
2065020650#ifdef __LITTLE_ENDIAN__
2065120651#define vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
20652 int64x2_t __s0 = __p0; \
2065320652 uint32x2_t __ret; \
20653 int64x2_t __s0 = __p0; \
2065420654 __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 18); \
2065520655 __ret; \
2065620656})
2065720657#else
2065820658#define vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
20659 uint32x2_t __ret; \
2065920660 int64x2_t __s0 = __p0; \
2066020661 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
20661 uint32x2_t __ret; \
2066220662 __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 18); \
2066320663 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2066420664 __ret; \
2066520665})
2066620666#define __noswap_vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
20667 int64x2_t __s0 = __p0; \
2066820667 uint32x2_t __ret; \
20668 int64x2_t __s0 = __p0; \
2066920669 __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 18); \
2067020670 __ret; \
2067120671})
......@@ -20673,23 +20673,23 @@ __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2067320673
2067420674#ifdef __LITTLE_ENDIAN__
2067520675#define vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
20676 int16x8_t __s0 = __p0; \
2067720676 uint8x8_t __ret; \
20677 int16x8_t __s0 = __p0; \
2067820678 __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 16); \
2067920679 __ret; \
2068020680})
2068120681#else
2068220682#define vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
20683 uint8x8_t __ret; \
2068320684 int16x8_t __s0 = __p0; \
2068420685 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
20685 uint8x8_t __ret; \
2068620686 __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 16); \
2068720687 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2068820688 __ret; \
2068920689})
2069020690#define __noswap_vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
20691 int16x8_t __s0 = __p0; \
2069220691 uint8x8_t __ret; \
20692 int16x8_t __s0 = __p0; \
2069320693 __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 16); \
2069420694 __ret; \
2069520695})
......@@ -20703,9 +20703,9 @@ __ai uint8x16_t vqshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
2070320703}
2070420704#else
2070520705__ai uint8x16_t vqshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
20706 uint8x16_t __ret;
2070620707 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2070720708 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20708 uint8x16_t __ret;
2070920709 __ret = (uint8x16_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
2071020710 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2071120711 return __ret;
......@@ -20720,9 +20720,9 @@ __ai uint32x4_t vqshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
2072020720}
2072120721#else
2072220722__ai uint32x4_t vqshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
20723 uint32x4_t __ret;
2072320724 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2072420725 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20725 uint32x4_t __ret;
2072620726 __ret = (uint32x4_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
2072720727 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2072820728 return __ret;
......@@ -20737,9 +20737,9 @@ __ai uint64x2_t vqshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
2073720737}
2073820738#else
2073920739__ai uint64x2_t vqshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
20740 uint64x2_t __ret;
2074020741 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2074120742 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20742 uint64x2_t __ret;
2074320743 __ret = (uint64x2_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
2074420744 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2074520745 return __ret;
......@@ -20754,9 +20754,9 @@ __ai uint16x8_t vqshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
2075420754}
2075520755#else
2075620756__ai uint16x8_t vqshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
20757 uint16x8_t __ret;
2075720758 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2075820759 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20759 uint16x8_t __ret;
2076020760 __ret = (uint16x8_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
2076120761 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2076220762 return __ret;
......@@ -20771,9 +20771,9 @@ __ai int8x16_t vqshlq_s8(int8x16_t __p0, int8x16_t __p1) {
2077120771}
2077220772#else
2077320773__ai int8x16_t vqshlq_s8(int8x16_t __p0, int8x16_t __p1) {
20774 int8x16_t __ret;
2077420775 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2077520776 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20776 int8x16_t __ret;
2077720777 __ret = (int8x16_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
2077820778 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2077920779 return __ret;
......@@ -20788,9 +20788,9 @@ __ai int32x4_t vqshlq_s32(int32x4_t __p0, int32x4_t __p1) {
2078820788}
2078920789#else
2079020790__ai int32x4_t vqshlq_s32(int32x4_t __p0, int32x4_t __p1) {
20791 int32x4_t __ret;
2079120792 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2079220793 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20793 int32x4_t __ret;
2079420794 __ret = (int32x4_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2079520795 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2079620796 return __ret;
......@@ -20805,9 +20805,9 @@ __ai int64x2_t vqshlq_s64(int64x2_t __p0, int64x2_t __p1) {
2080520805}
2080620806#else
2080720807__ai int64x2_t vqshlq_s64(int64x2_t __p0, int64x2_t __p1) {
20808 int64x2_t __ret;
2080820809 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2080920810 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20810 int64x2_t __ret;
2081120811 __ret = (int64x2_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
2081220812 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2081320813 return __ret;
......@@ -20822,9 +20822,9 @@ __ai int16x8_t vqshlq_s16(int16x8_t __p0, int16x8_t __p1) {
2082220822}
2082320823#else
2082420824__ai int16x8_t vqshlq_s16(int16x8_t __p0, int16x8_t __p1) {
20825 int16x8_t __ret;
2082520826 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2082620827 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20827 int16x8_t __ret;
2082820828 __ret = (int16x8_t) __builtin_neon_vqshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2082920829 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2083020830 return __ret;
......@@ -20839,9 +20839,9 @@ __ai uint8x8_t vqshl_u8(uint8x8_t __p0, int8x8_t __p1) {
2083920839}
2084020840#else
2084120841__ai uint8x8_t vqshl_u8(uint8x8_t __p0, int8x8_t __p1) {
20842 uint8x8_t __ret;
2084220843 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2084320844 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20844 uint8x8_t __ret;
2084520845 __ret = (uint8x8_t) __builtin_neon_vqshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
2084620846 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2084720847 return __ret;
......@@ -20856,9 +20856,9 @@ __ai uint32x2_t vqshl_u32(uint32x2_t __p0, int32x2_t __p1) {
2085620856}
2085720857#else
2085820858__ai uint32x2_t vqshl_u32(uint32x2_t __p0, int32x2_t __p1) {
20859 uint32x2_t __ret;
2085920860 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2086020861 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20861 uint32x2_t __ret;
2086220862 __ret = (uint32x2_t) __builtin_neon_vqshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
2086320863 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2086420864 return __ret;
......@@ -20878,9 +20878,9 @@ __ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
2087820878}
2087920879#else
2088020880__ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
20881 uint16x4_t __ret;
2088120882 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2088220883 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20883 uint16x4_t __ret;
2088420884 __ret = (uint16x4_t) __builtin_neon_vqshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
2088520885 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2088620886 return __ret;
......@@ -20895,9 +20895,9 @@ __ai int8x8_t vqshl_s8(int8x8_t __p0, int8x8_t __p1) {
2089520895}
2089620896#else
2089720897__ai int8x8_t vqshl_s8(int8x8_t __p0, int8x8_t __p1) {
20898 int8x8_t __ret;
2089820899 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2089920900 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20900 int8x8_t __ret;
2090120901 __ret = (int8x8_t) __builtin_neon_vqshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
2090220902 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2090320903 return __ret;
......@@ -20912,9 +20912,9 @@ __ai int32x2_t vqshl_s32(int32x2_t __p0, int32x2_t __p1) {
2091220912}
2091320913#else
2091420914__ai int32x2_t vqshl_s32(int32x2_t __p0, int32x2_t __p1) {
20915 int32x2_t __ret;
2091520916 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2091620917 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20917 int32x2_t __ret;
2091820918 __ret = (int32x2_t) __builtin_neon_vqshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2091920919 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2092020920 return __ret;
......@@ -20934,9 +20934,9 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2093420934}
2093520935#else
2093620936__ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
20937 int16x4_t __ret;
2093720938 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2093820939 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20939 int16x4_t __ret;
2094020940 __ret = (int16x4_t) __builtin_neon_vqshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2094120941 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2094220942 return __ret;
......@@ -20945,16 +20945,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2094520945
2094620946#ifdef __LITTLE_ENDIAN__
2094720947#define vqshlq_n_u8(__p0, __p1) __extension__ ({ \
20948 uint8x16_t __s0 = __p0; \
2094920948 uint8x16_t __ret; \
20949 uint8x16_t __s0 = __p0; \
2095020950 __ret = (uint8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 48); \
2095120951 __ret; \
2095220952})
2095320953#else
2095420954#define vqshlq_n_u8(__p0, __p1) __extension__ ({ \
20955 uint8x16_t __ret; \
2095520956 uint8x16_t __s0 = __p0; \
2095620957 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
20957 uint8x16_t __ret; \
2095820958 __ret = (uint8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 48); \
2095920959 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2096020960 __ret; \
......@@ -20963,16 +20963,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2096320963
2096420964#ifdef __LITTLE_ENDIAN__
2096520965#define vqshlq_n_u32(__p0, __p1) __extension__ ({ \
20966 uint32x4_t __s0 = __p0; \
2096720966 uint32x4_t __ret; \
20967 uint32x4_t __s0 = __p0; \
2096820968 __ret = (uint32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 50); \
2096920969 __ret; \
2097020970})
2097120971#else
2097220972#define vqshlq_n_u32(__p0, __p1) __extension__ ({ \
20973 uint32x4_t __ret; \
2097320974 uint32x4_t __s0 = __p0; \
2097420975 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
20975 uint32x4_t __ret; \
2097620976 __ret = (uint32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 50); \
2097720977 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2097820978 __ret; \
......@@ -20981,16 +20981,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2098120981
2098220982#ifdef __LITTLE_ENDIAN__
2098320983#define vqshlq_n_u64(__p0, __p1) __extension__ ({ \
20984 uint64x2_t __s0 = __p0; \
2098520984 uint64x2_t __ret; \
20985 uint64x2_t __s0 = __p0; \
2098620986 __ret = (uint64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 51); \
2098720987 __ret; \
2098820988})
2098920989#else
2099020990#define vqshlq_n_u64(__p0, __p1) __extension__ ({ \
20991 uint64x2_t __ret; \
2099120992 uint64x2_t __s0 = __p0; \
2099220993 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
20993 uint64x2_t __ret; \
2099420994 __ret = (uint64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 51); \
2099520995 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2099620996 __ret; \
......@@ -20999,16 +20999,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2099920999
2100021000#ifdef __LITTLE_ENDIAN__
2100121001#define vqshlq_n_u16(__p0, __p1) __extension__ ({ \
21002 uint16x8_t __s0 = __p0; \
2100321002 uint16x8_t __ret; \
21003 uint16x8_t __s0 = __p0; \
2100421004 __ret = (uint16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 49); \
2100521005 __ret; \
2100621006})
2100721007#else
2100821008#define vqshlq_n_u16(__p0, __p1) __extension__ ({ \
21009 uint16x8_t __ret; \
2100921010 uint16x8_t __s0 = __p0; \
2101021011 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21011 uint16x8_t __ret; \
2101221012 __ret = (uint16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 49); \
2101321013 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2101421014 __ret; \
......@@ -21017,16 +21017,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2101721017
2101821018#ifdef __LITTLE_ENDIAN__
2101921019#define vqshlq_n_s8(__p0, __p1) __extension__ ({ \
21020 int8x16_t __s0 = __p0; \
2102121020 int8x16_t __ret; \
21021 int8x16_t __s0 = __p0; \
2102221022 __ret = (int8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 32); \
2102321023 __ret; \
2102421024})
2102521025#else
2102621026#define vqshlq_n_s8(__p0, __p1) __extension__ ({ \
21027 int8x16_t __ret; \
2102721028 int8x16_t __s0 = __p0; \
2102821029 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
21029 int8x16_t __ret; \
2103021030 __ret = (int8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 32); \
2103121031 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2103221032 __ret; \
......@@ -21035,16 +21035,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2103521035
2103621036#ifdef __LITTLE_ENDIAN__
2103721037#define vqshlq_n_s32(__p0, __p1) __extension__ ({ \
21038 int32x4_t __s0 = __p0; \
2103921038 int32x4_t __ret; \
21039 int32x4_t __s0 = __p0; \
2104021040 __ret = (int32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 34); \
2104121041 __ret; \
2104221042})
2104321043#else
2104421044#define vqshlq_n_s32(__p0, __p1) __extension__ ({ \
21045 int32x4_t __ret; \
2104521046 int32x4_t __s0 = __p0; \
2104621047 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21047 int32x4_t __ret; \
2104821048 __ret = (int32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 34); \
2104921049 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2105021050 __ret; \
......@@ -21053,16 +21053,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2105321053
2105421054#ifdef __LITTLE_ENDIAN__
2105521055#define vqshlq_n_s64(__p0, __p1) __extension__ ({ \
21056 int64x2_t __s0 = __p0; \
2105721056 int64x2_t __ret; \
21057 int64x2_t __s0 = __p0; \
2105821058 __ret = (int64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 35); \
2105921059 __ret; \
2106021060})
2106121061#else
2106221062#define vqshlq_n_s64(__p0, __p1) __extension__ ({ \
21063 int64x2_t __ret; \
2106321064 int64x2_t __s0 = __p0; \
2106421065 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21065 int64x2_t __ret; \
2106621066 __ret = (int64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 35); \
2106721067 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2106821068 __ret; \
......@@ -21071,16 +21071,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2107121071
2107221072#ifdef __LITTLE_ENDIAN__
2107321073#define vqshlq_n_s16(__p0, __p1) __extension__ ({ \
21074 int16x8_t __s0 = __p0; \
2107521074 int16x8_t __ret; \
21075 int16x8_t __s0 = __p0; \
2107621076 __ret = (int16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 33); \
2107721077 __ret; \
2107821078})
2107921079#else
2108021080#define vqshlq_n_s16(__p0, __p1) __extension__ ({ \
21081 int16x8_t __ret; \
2108121082 int16x8_t __s0 = __p0; \
2108221083 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21083 int16x8_t __ret; \
2108421084 __ret = (int16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 33); \
2108521085 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2108621086 __ret; \
......@@ -21089,16 +21089,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2108921089
2109021090#ifdef __LITTLE_ENDIAN__
2109121091#define vqshl_n_u8(__p0, __p1) __extension__ ({ \
21092 uint8x8_t __s0 = __p0; \
2109321092 uint8x8_t __ret; \
21093 uint8x8_t __s0 = __p0; \
2109421094 __ret = (uint8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 16); \
2109521095 __ret; \
2109621096})
2109721097#else
2109821098#define vqshl_n_u8(__p0, __p1) __extension__ ({ \
21099 uint8x8_t __ret; \
2109921100 uint8x8_t __s0 = __p0; \
2110021101 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21101 uint8x8_t __ret; \
2110221102 __ret = (uint8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 16); \
2110321103 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2110421104 __ret; \
......@@ -21107,16 +21107,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2110721107
2110821108#ifdef __LITTLE_ENDIAN__
2110921109#define vqshl_n_u32(__p0, __p1) __extension__ ({ \
21110 uint32x2_t __s0 = __p0; \
2111121110 uint32x2_t __ret; \
21111 uint32x2_t __s0 = __p0; \
2111221112 __ret = (uint32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 18); \
2111321113 __ret; \
2111421114})
2111521115#else
2111621116#define vqshl_n_u32(__p0, __p1) __extension__ ({ \
21117 uint32x2_t __ret; \
2111721118 uint32x2_t __s0 = __p0; \
2111821119 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21119 uint32x2_t __ret; \
2112021120 __ret = (uint32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 18); \
2112121121 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2112221122 __ret; \
......@@ -21124,23 +21124,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2112421124#endif
2112521125
2112621126#define vqshl_n_u64(__p0, __p1) __extension__ ({ \
21127 uint64x1_t __s0 = __p0; \
2112821127 uint64x1_t __ret; \
21128 uint64x1_t __s0 = __p0; \
2112921129 __ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
2113021130 __ret; \
2113121131})
2113221132#ifdef __LITTLE_ENDIAN__
2113321133#define vqshl_n_u16(__p0, __p1) __extension__ ({ \
21134 uint16x4_t __s0 = __p0; \
2113521134 uint16x4_t __ret; \
21135 uint16x4_t __s0 = __p0; \
2113621136 __ret = (uint16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 17); \
2113721137 __ret; \
2113821138})
2113921139#else
2114021140#define vqshl_n_u16(__p0, __p1) __extension__ ({ \
21141 uint16x4_t __ret; \
2114121142 uint16x4_t __s0 = __p0; \
2114221143 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21143 uint16x4_t __ret; \
2114421144 __ret = (uint16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 17); \
2114521145 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2114621146 __ret; \
......@@ -21149,16 +21149,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2114921149
2115021150#ifdef __LITTLE_ENDIAN__
2115121151#define vqshl_n_s8(__p0, __p1) __extension__ ({ \
21152 int8x8_t __s0 = __p0; \
2115321152 int8x8_t __ret; \
21153 int8x8_t __s0 = __p0; \
2115421154 __ret = (int8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 0); \
2115521155 __ret; \
2115621156})
2115721157#else
2115821158#define vqshl_n_s8(__p0, __p1) __extension__ ({ \
21159 int8x8_t __ret; \
2115921160 int8x8_t __s0 = __p0; \
2116021161 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21161 int8x8_t __ret; \
2116221162 __ret = (int8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 0); \
2116321163 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2116421164 __ret; \
......@@ -21167,16 +21167,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2116721167
2116821168#ifdef __LITTLE_ENDIAN__
2116921169#define vqshl_n_s32(__p0, __p1) __extension__ ({ \
21170 int32x2_t __s0 = __p0; \
2117121170 int32x2_t __ret; \
21171 int32x2_t __s0 = __p0; \
2117221172 __ret = (int32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 2); \
2117321173 __ret; \
2117421174})
2117521175#else
2117621176#define vqshl_n_s32(__p0, __p1) __extension__ ({ \
21177 int32x2_t __ret; \
2117721178 int32x2_t __s0 = __p0; \
2117821179 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21179 int32x2_t __ret; \
2118021180 __ret = (int32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 2); \
2118121181 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2118221182 __ret; \
......@@ -21184,23 +21184,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2118421184#endif
2118521185
2118621186#define vqshl_n_s64(__p0, __p1) __extension__ ({ \
21187 int64x1_t __s0 = __p0; \
2118821187 int64x1_t __ret; \
21188 int64x1_t __s0 = __p0; \
2118921189 __ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
2119021190 __ret; \
2119121191})
2119221192#ifdef __LITTLE_ENDIAN__
2119321193#define vqshl_n_s16(__p0, __p1) __extension__ ({ \
21194 int16x4_t __s0 = __p0; \
2119521194 int16x4_t __ret; \
21195 int16x4_t __s0 = __p0; \
2119621196 __ret = (int16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 1); \
2119721197 __ret; \
2119821198})
2119921199#else
2120021200#define vqshl_n_s16(__p0, __p1) __extension__ ({ \
21201 int16x4_t __ret; \
2120121202 int16x4_t __s0 = __p0; \
2120221203 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21203 int16x4_t __ret; \
2120421204 __ret = (int16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 1); \
2120521205 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2120621206 __ret; \
......@@ -21209,16 +21209,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2120921209
2121021210#ifdef __LITTLE_ENDIAN__
2121121211#define vqshluq_n_s8(__p0, __p1) __extension__ ({ \
21212 int8x16_t __s0 = __p0; \
2121321212 uint8x16_t __ret; \
21213 int8x16_t __s0 = __p0; \
2121421214 __ret = (uint8x16_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 48); \
2121521215 __ret; \
2121621216})
2121721217#else
2121821218#define vqshluq_n_s8(__p0, __p1) __extension__ ({ \
21219 uint8x16_t __ret; \
2121921220 int8x16_t __s0 = __p0; \
2122021221 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
21221 uint8x16_t __ret; \
2122221222 __ret = (uint8x16_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 48); \
2122321223 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2122421224 __ret; \
......@@ -21227,16 +21227,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2122721227
2122821228#ifdef __LITTLE_ENDIAN__
2122921229#define vqshluq_n_s32(__p0, __p1) __extension__ ({ \
21230 int32x4_t __s0 = __p0; \
2123121230 uint32x4_t __ret; \
21231 int32x4_t __s0 = __p0; \
2123221232 __ret = (uint32x4_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 50); \
2123321233 __ret; \
2123421234})
2123521235#else
2123621236#define vqshluq_n_s32(__p0, __p1) __extension__ ({ \
21237 uint32x4_t __ret; \
2123721238 int32x4_t __s0 = __p0; \
2123821239 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21239 uint32x4_t __ret; \
2124021240 __ret = (uint32x4_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 50); \
2124121241 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2124221242 __ret; \
......@@ -21245,16 +21245,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2124521245
2124621246#ifdef __LITTLE_ENDIAN__
2124721247#define vqshluq_n_s64(__p0, __p1) __extension__ ({ \
21248 int64x2_t __s0 = __p0; \
2124921248 uint64x2_t __ret; \
21249 int64x2_t __s0 = __p0; \
2125021250 __ret = (uint64x2_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 51); \
2125121251 __ret; \
2125221252})
2125321253#else
2125421254#define vqshluq_n_s64(__p0, __p1) __extension__ ({ \
21255 uint64x2_t __ret; \
2125521256 int64x2_t __s0 = __p0; \
2125621257 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21257 uint64x2_t __ret; \
2125821258 __ret = (uint64x2_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 51); \
2125921259 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2126021260 __ret; \
......@@ -21263,16 +21263,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2126321263
2126421264#ifdef __LITTLE_ENDIAN__
2126521265#define vqshluq_n_s16(__p0, __p1) __extension__ ({ \
21266 int16x8_t __s0 = __p0; \
2126721266 uint16x8_t __ret; \
21267 int16x8_t __s0 = __p0; \
2126821268 __ret = (uint16x8_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 49); \
2126921269 __ret; \
2127021270})
2127121271#else
2127221272#define vqshluq_n_s16(__p0, __p1) __extension__ ({ \
21273 uint16x8_t __ret; \
2127321274 int16x8_t __s0 = __p0; \
2127421275 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21275 uint16x8_t __ret; \
2127621276 __ret = (uint16x8_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 49); \
2127721277 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2127821278 __ret; \
......@@ -21281,16 +21281,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2128121281
2128221282#ifdef __LITTLE_ENDIAN__
2128321283#define vqshlu_n_s8(__p0, __p1) __extension__ ({ \
21284 int8x8_t __s0 = __p0; \
2128521284 uint8x8_t __ret; \
21285 int8x8_t __s0 = __p0; \
2128621286 __ret = (uint8x8_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 16); \
2128721287 __ret; \
2128821288})
2128921289#else
2129021290#define vqshlu_n_s8(__p0, __p1) __extension__ ({ \
21291 uint8x8_t __ret; \
2129121292 int8x8_t __s0 = __p0; \
2129221293 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21293 uint8x8_t __ret; \
2129421294 __ret = (uint8x8_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 16); \
2129521295 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2129621296 __ret; \
......@@ -21299,16 +21299,16 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2129921299
2130021300#ifdef __LITTLE_ENDIAN__
2130121301#define vqshlu_n_s32(__p0, __p1) __extension__ ({ \
21302 int32x2_t __s0 = __p0; \
2130321302 uint32x2_t __ret; \
21303 int32x2_t __s0 = __p0; \
2130421304 __ret = (uint32x2_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 18); \
2130521305 __ret; \
2130621306})
2130721307#else
2130821308#define vqshlu_n_s32(__p0, __p1) __extension__ ({ \
21309 uint32x2_t __ret; \
2130921310 int32x2_t __s0 = __p0; \
2131021311 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21311 uint32x2_t __ret; \
2131221312 __ret = (uint32x2_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 18); \
2131321313 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2131421314 __ret; \
......@@ -21316,23 +21316,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2131621316#endif
2131721317
2131821318#define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
21319 int64x1_t __s0 = __p0; \
2132021319 uint64x1_t __ret; \
21320 int64x1_t __s0 = __p0; \
2132121321 __ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
2132221322 __ret; \
2132321323})
2132421324#ifdef __LITTLE_ENDIAN__
2132521325#define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
21326 int16x4_t __s0 = __p0; \
2132721326 uint16x4_t __ret; \
21327 int16x4_t __s0 = __p0; \
2132821328 __ret = (uint16x4_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 17); \
2132921329 __ret; \
2133021330})
2133121331#else
2133221332#define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
21333 uint16x4_t __ret; \
2133321334 int16x4_t __s0 = __p0; \
2133421335 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21335 uint16x4_t __ret; \
2133621336 __ret = (uint16x4_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 17); \
2133721337 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2133821338 __ret; \
......@@ -21341,23 +21341,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2134121341
2134221342#ifdef __LITTLE_ENDIAN__
2134321343#define vqshrn_n_u32(__p0, __p1) __extension__ ({ \
21344 uint32x4_t __s0 = __p0; \
2134521344 uint16x4_t __ret; \
21345 uint32x4_t __s0 = __p0; \
2134621346 __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 17); \
2134721347 __ret; \
2134821348})
2134921349#else
2135021350#define vqshrn_n_u32(__p0, __p1) __extension__ ({ \
21351 uint16x4_t __ret; \
2135121352 uint32x4_t __s0 = __p0; \
2135221353 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21353 uint16x4_t __ret; \
2135421354 __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 17); \
2135521355 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2135621356 __ret; \
2135721357})
2135821358#define __noswap_vqshrn_n_u32(__p0, __p1) __extension__ ({ \
21359 uint32x4_t __s0 = __p0; \
2136021359 uint16x4_t __ret; \
21360 uint32x4_t __s0 = __p0; \
2136121361 __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 17); \
2136221362 __ret; \
2136321363})
......@@ -21365,23 +21365,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2136521365
2136621366#ifdef __LITTLE_ENDIAN__
2136721367#define vqshrn_n_u64(__p0, __p1) __extension__ ({ \
21368 uint64x2_t __s0 = __p0; \
2136921368 uint32x2_t __ret; \
21369 uint64x2_t __s0 = __p0; \
2137021370 __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 18); \
2137121371 __ret; \
2137221372})
2137321373#else
2137421374#define vqshrn_n_u64(__p0, __p1) __extension__ ({ \
21375 uint32x2_t __ret; \
2137521376 uint64x2_t __s0 = __p0; \
2137621377 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21377 uint32x2_t __ret; \
2137821378 __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 18); \
2137921379 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2138021380 __ret; \
2138121381})
2138221382#define __noswap_vqshrn_n_u64(__p0, __p1) __extension__ ({ \
21383 uint64x2_t __s0 = __p0; \
2138421383 uint32x2_t __ret; \
21384 uint64x2_t __s0 = __p0; \
2138521385 __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 18); \
2138621386 __ret; \
2138721387})
......@@ -21389,23 +21389,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2138921389
2139021390#ifdef __LITTLE_ENDIAN__
2139121391#define vqshrn_n_u16(__p0, __p1) __extension__ ({ \
21392 uint16x8_t __s0 = __p0; \
2139321392 uint8x8_t __ret; \
21393 uint16x8_t __s0 = __p0; \
2139421394 __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 16); \
2139521395 __ret; \
2139621396})
2139721397#else
2139821398#define vqshrn_n_u16(__p0, __p1) __extension__ ({ \
21399 uint8x8_t __ret; \
2139921400 uint16x8_t __s0 = __p0; \
2140021401 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21401 uint8x8_t __ret; \
2140221402 __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 16); \
2140321403 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2140421404 __ret; \
2140521405})
2140621406#define __noswap_vqshrn_n_u16(__p0, __p1) __extension__ ({ \
21407 uint16x8_t __s0 = __p0; \
2140821407 uint8x8_t __ret; \
21408 uint16x8_t __s0 = __p0; \
2140921409 __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 16); \
2141021410 __ret; \
2141121411})
......@@ -21413,23 +21413,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2141321413
2141421414#ifdef __LITTLE_ENDIAN__
2141521415#define vqshrn_n_s32(__p0, __p1) __extension__ ({ \
21416 int32x4_t __s0 = __p0; \
2141721416 int16x4_t __ret; \
21417 int32x4_t __s0 = __p0; \
2141821418 __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 1); \
2141921419 __ret; \
2142021420})
2142121421#else
2142221422#define vqshrn_n_s32(__p0, __p1) __extension__ ({ \
21423 int16x4_t __ret; \
2142321424 int32x4_t __s0 = __p0; \
2142421425 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21425 int16x4_t __ret; \
2142621426 __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 1); \
2142721427 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2142821428 __ret; \
2142921429})
2143021430#define __noswap_vqshrn_n_s32(__p0, __p1) __extension__ ({ \
21431 int32x4_t __s0 = __p0; \
2143221431 int16x4_t __ret; \
21432 int32x4_t __s0 = __p0; \
2143321433 __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 1); \
2143421434 __ret; \
2143521435})
......@@ -21437,23 +21437,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2143721437
2143821438#ifdef __LITTLE_ENDIAN__
2143921439#define vqshrn_n_s64(__p0, __p1) __extension__ ({ \
21440 int64x2_t __s0 = __p0; \
2144121440 int32x2_t __ret; \
21441 int64x2_t __s0 = __p0; \
2144221442 __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 2); \
2144321443 __ret; \
2144421444})
2144521445#else
2144621446#define vqshrn_n_s64(__p0, __p1) __extension__ ({ \
21447 int32x2_t __ret; \
2144721448 int64x2_t __s0 = __p0; \
2144821449 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21449 int32x2_t __ret; \
2145021450 __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 2); \
2145121451 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2145221452 __ret; \
2145321453})
2145421454#define __noswap_vqshrn_n_s64(__p0, __p1) __extension__ ({ \
21455 int64x2_t __s0 = __p0; \
2145621455 int32x2_t __ret; \
21456 int64x2_t __s0 = __p0; \
2145721457 __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 2); \
2145821458 __ret; \
2145921459})
......@@ -21461,23 +21461,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2146121461
2146221462#ifdef __LITTLE_ENDIAN__
2146321463#define vqshrn_n_s16(__p0, __p1) __extension__ ({ \
21464 int16x8_t __s0 = __p0; \
2146521464 int8x8_t __ret; \
21465 int16x8_t __s0 = __p0; \
2146621466 __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 0); \
2146721467 __ret; \
2146821468})
2146921469#else
2147021470#define vqshrn_n_s16(__p0, __p1) __extension__ ({ \
21471 int8x8_t __ret; \
2147121472 int16x8_t __s0 = __p0; \
2147221473 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21473 int8x8_t __ret; \
2147421474 __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 0); \
2147521475 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2147621476 __ret; \
2147721477})
2147821478#define __noswap_vqshrn_n_s16(__p0, __p1) __extension__ ({ \
21479 int16x8_t __s0 = __p0; \
2148021479 int8x8_t __ret; \
21480 int16x8_t __s0 = __p0; \
2148121481 __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 0); \
2148221482 __ret; \
2148321483})
......@@ -21485,23 +21485,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2148521485
2148621486#ifdef __LITTLE_ENDIAN__
2148721487#define vqshrun_n_s32(__p0, __p1) __extension__ ({ \
21488 int32x4_t __s0 = __p0; \
2148921488 uint16x4_t __ret; \
21489 int32x4_t __s0 = __p0; \
2149021490 __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 17); \
2149121491 __ret; \
2149221492})
2149321493#else
2149421494#define vqshrun_n_s32(__p0, __p1) __extension__ ({ \
21495 uint16x4_t __ret; \
2149521496 int32x4_t __s0 = __p0; \
2149621497 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
21497 uint16x4_t __ret; \
2149821498 __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 17); \
2149921499 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2150021500 __ret; \
2150121501})
2150221502#define __noswap_vqshrun_n_s32(__p0, __p1) __extension__ ({ \
21503 int32x4_t __s0 = __p0; \
2150421503 uint16x4_t __ret; \
21504 int32x4_t __s0 = __p0; \
2150521505 __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 17); \
2150621506 __ret; \
2150721507})
......@@ -21509,23 +21509,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2150921509
2151021510#ifdef __LITTLE_ENDIAN__
2151121511#define vqshrun_n_s64(__p0, __p1) __extension__ ({ \
21512 int64x2_t __s0 = __p0; \
2151321512 uint32x2_t __ret; \
21513 int64x2_t __s0 = __p0; \
2151421514 __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 18); \
2151521515 __ret; \
2151621516})
2151721517#else
2151821518#define vqshrun_n_s64(__p0, __p1) __extension__ ({ \
21519 uint32x2_t __ret; \
2151921520 int64x2_t __s0 = __p0; \
2152021521 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
21521 uint32x2_t __ret; \
2152221522 __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 18); \
2152321523 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2152421524 __ret; \
2152521525})
2152621526#define __noswap_vqshrun_n_s64(__p0, __p1) __extension__ ({ \
21527 int64x2_t __s0 = __p0; \
2152821527 uint32x2_t __ret; \
21528 int64x2_t __s0 = __p0; \
2152921529 __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 18); \
2153021530 __ret; \
2153121531})
......@@ -21533,23 +21533,23 @@ __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
2153321533
2153421534#ifdef __LITTLE_ENDIAN__
2153521535#define vqshrun_n_s16(__p0, __p1) __extension__ ({ \
21536 int16x8_t __s0 = __p0; \
2153721536 uint8x8_t __ret; \
21537 int16x8_t __s0 = __p0; \
2153821538 __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 16); \
2153921539 __ret; \
2154021540})
2154121541#else
2154221542#define vqshrun_n_s16(__p0, __p1) __extension__ ({ \
21543 uint8x8_t __ret; \
2154321544 int16x8_t __s0 = __p0; \
2154421545 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
21545 uint8x8_t __ret; \
2154621546 __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 16); \
2154721547 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2154821548 __ret; \
2154921549})
2155021550#define __noswap_vqshrun_n_s16(__p0, __p1) __extension__ ({ \
21551 int16x8_t __s0 = __p0; \
2155221551 uint8x8_t __ret; \
21552 int16x8_t __s0 = __p0; \
2155321553 __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 16); \
2155421554 __ret; \
2155521555})
......@@ -21563,9 +21563,9 @@ __ai uint8x16_t vqsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2156321563}
2156421564#else
2156521565__ai uint8x16_t vqsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
21566 uint8x16_t __ret;
2156621567 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2156721568 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
21568 uint8x16_t __ret;
2156921569 __ret = (uint8x16_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
2157021570 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2157121571 return __ret;
......@@ -21580,9 +21580,9 @@ __ai uint32x4_t vqsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2158021580}
2158121581#else
2158221582__ai uint32x4_t vqsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
21583 uint32x4_t __ret;
2158321584 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2158421585 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21585 uint32x4_t __ret;
2158621586 __ret = (uint32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
2158721587 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2158821588 return __ret;
......@@ -21597,9 +21597,9 @@ __ai uint64x2_t vqsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
2159721597}
2159821598#else
2159921599__ai uint64x2_t vqsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
21600 uint64x2_t __ret;
2160021601 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2160121602 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21602 uint64x2_t __ret;
2160321603 __ret = (uint64x2_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
2160421604 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2160521605 return __ret;
......@@ -21614,9 +21614,9 @@ __ai uint16x8_t vqsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2161421614}
2161521615#else
2161621616__ai uint16x8_t vqsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
21617 uint16x8_t __ret;
2161721618 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2161821619 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21619 uint16x8_t __ret;
2162021620 __ret = (uint16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
2162121621 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2162221622 return __ret;
......@@ -21631,9 +21631,9 @@ __ai int8x16_t vqsubq_s8(int8x16_t __p0, int8x16_t __p1) {
2163121631}
2163221632#else
2163321633__ai int8x16_t vqsubq_s8(int8x16_t __p0, int8x16_t __p1) {
21634 int8x16_t __ret;
2163421635 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2163521636 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
21636 int8x16_t __ret;
2163721637 __ret = (int8x16_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
2163821638 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2163921639 return __ret;
......@@ -21648,9 +21648,9 @@ __ai int32x4_t vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
2164821648}
2164921649#else
2165021650__ai int32x4_t vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
21651 int32x4_t __ret;
2165121652 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2165221653 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21653 int32x4_t __ret;
2165421654 __ret = (int32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2165521655 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2165621656 return __ret;
......@@ -21665,9 +21665,9 @@ __ai int64x2_t vqsubq_s64(int64x2_t __p0, int64x2_t __p1) {
2166521665}
2166621666#else
2166721667__ai int64x2_t vqsubq_s64(int64x2_t __p0, int64x2_t __p1) {
21668 int64x2_t __ret;
2166821669 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2166921670 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21670 int64x2_t __ret;
2167121671 __ret = (int64x2_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
2167221672 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2167321673 return __ret;
......@@ -21682,9 +21682,9 @@ __ai int16x8_t vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
2168221682}
2168321683#else
2168421684__ai int16x8_t vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
21685 int16x8_t __ret;
2168521686 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2168621687 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21687 int16x8_t __ret;
2168821688 __ret = (int16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2168921689 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2169021690 return __ret;
......@@ -21699,9 +21699,9 @@ __ai uint8x8_t vqsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
2169921699}
2170021700#else
2170121701__ai uint8x8_t vqsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
21702 uint8x8_t __ret;
2170221703 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2170321704 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21704 uint8x8_t __ret;
2170521705 __ret = (uint8x8_t) __builtin_neon_vqsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
2170621706 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2170721707 return __ret;
......@@ -21716,9 +21716,9 @@ __ai uint32x2_t vqsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
2171621716}
2171721717#else
2171821718__ai uint32x2_t vqsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
21719 uint32x2_t __ret;
2171921720 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2172021721 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21721 uint32x2_t __ret;
2172221722 __ret = (uint32x2_t) __builtin_neon_vqsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
2172321723 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2172421724 return __ret;
......@@ -21738,9 +21738,9 @@ __ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
2173821738}
2173921739#else
2174021740__ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
21741 uint16x4_t __ret;
2174121742 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2174221743 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21743 uint16x4_t __ret;
2174421744 __ret = (uint16x4_t) __builtin_neon_vqsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
2174521745 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2174621746 return __ret;
......@@ -21755,9 +21755,9 @@ __ai int8x8_t vqsub_s8(int8x8_t __p0, int8x8_t __p1) {
2175521755}
2175621756#else
2175721757__ai int8x8_t vqsub_s8(int8x8_t __p0, int8x8_t __p1) {
21758 int8x8_t __ret;
2175821759 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2175921760 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21760 int8x8_t __ret;
2176121761 __ret = (int8x8_t) __builtin_neon_vqsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
2176221762 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2176321763 return __ret;
......@@ -21772,9 +21772,9 @@ __ai int32x2_t vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
2177221772}
2177321773#else
2177421774__ai int32x2_t vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
21775 int32x2_t __ret;
2177521776 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2177621777 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21777 int32x2_t __ret;
2177821778 __ret = (int32x2_t) __builtin_neon_vqsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2177921779 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2178021780 return __ret;
......@@ -21794,9 +21794,9 @@ __ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
2179421794}
2179521795#else
2179621796__ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
21797 int16x4_t __ret;
2179721798 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2179821799 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21799 int16x4_t __ret;
2180021800 __ret = (int16x4_t) __builtin_neon_vqsub_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2180121801 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2180221802 return __ret;
......@@ -21811,9 +21811,9 @@ __ai uint16x4_t vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
2181121811}
2181221812#else
2181321813__ai uint16x4_t vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
21814 uint16x4_t __ret;
2181421815 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2181521816 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21816 uint16x4_t __ret;
2181721817 __ret = (uint16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 17);
2181821818 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2181921819 return __ret;
......@@ -21833,9 +21833,9 @@ __ai uint32x2_t vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
2183321833}
2183421834#else
2183521835__ai uint32x2_t vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
21836 uint32x2_t __ret;
2183621837 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2183721838 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21838 uint32x2_t __ret;
2183921839 __ret = (uint32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 18);
2184021840 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2184121841 return __ret;
......@@ -21855,9 +21855,9 @@ __ai uint8x8_t vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
2185521855}
2185621856#else
2185721857__ai uint8x8_t vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
21858 uint8x8_t __ret;
2185821859 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2185921860 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21860 uint8x8_t __ret;
2186121861 __ret = (uint8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 16);
2186221862 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2186321863 return __ret;
......@@ -21877,9 +21877,9 @@ __ai int16x4_t vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
2187721877}
2187821878#else
2187921879__ai int16x4_t vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
21880 int16x4_t __ret;
2188021881 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2188121882 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21882 int16x4_t __ret;
2188321883 __ret = (int16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 1);
2188421884 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2188521885 return __ret;
......@@ -21899,9 +21899,9 @@ __ai int32x2_t vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
2189921899}
2190021900#else
2190121901__ai int32x2_t vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
21902 int32x2_t __ret;
2190221903 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2190321904 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21904 int32x2_t __ret;
2190521905 __ret = (int32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 2);
2190621906 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2190721907 return __ret;
......@@ -21921,9 +21921,9 @@ __ai int8x8_t vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
2192121921}
2192221922#else
2192321923__ai int8x8_t vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
21924 int8x8_t __ret;
2192421925 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2192521926 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21926 int8x8_t __ret;
2192721927 __ret = (int8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 0);
2192821928 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2192921929 return __ret;
......@@ -21943,8 +21943,8 @@ __ai uint32x4_t vrecpeq_u32(uint32x4_t __p0) {
2194321943}
2194421944#else
2194521945__ai uint32x4_t vrecpeq_u32(uint32x4_t __p0) {
21946 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2194721946 uint32x4_t __ret;
21947 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2194821948 __ret = (uint32x4_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 50);
2194921949 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2195021950 return __ret;
......@@ -21959,8 +21959,8 @@ __ai float32x4_t vrecpeq_f32(float32x4_t __p0) {
2195921959}
2196021960#else
2196121961__ai float32x4_t vrecpeq_f32(float32x4_t __p0) {
21962 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2196321962 float32x4_t __ret;
21963 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2196421964 __ret = (float32x4_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 41);
2196521965 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2196621966 return __ret;
......@@ -21975,8 +21975,8 @@ __ai uint32x2_t vrecpe_u32(uint32x2_t __p0) {
2197521975}
2197621976#else
2197721977__ai uint32x2_t vrecpe_u32(uint32x2_t __p0) {
21978 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2197921978 uint32x2_t __ret;
21979 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2198021980 __ret = (uint32x2_t) __builtin_neon_vrecpe_v((int8x8_t)__rev0, 18);
2198121981 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2198221982 return __ret;
......@@ -21991,8 +21991,8 @@ __ai float32x2_t vrecpe_f32(float32x2_t __p0) {
2199121991}
2199221992#else
2199321993__ai float32x2_t vrecpe_f32(float32x2_t __p0) {
21994 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2199521994 float32x2_t __ret;
21995 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2199621996 __ret = (float32x2_t) __builtin_neon_vrecpe_v((int8x8_t)__rev0, 9);
2199721997 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2199821998 return __ret;
......@@ -22007,9 +22007,9 @@ __ai float32x4_t vrecpsq_f32(float32x4_t __p0, float32x4_t __p1) {
2200722007}
2200822008#else
2200922009__ai float32x4_t vrecpsq_f32(float32x4_t __p0, float32x4_t __p1) {
22010 float32x4_t __ret;
2201022011 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2201122012 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22012 float32x4_t __ret;
2201322013 __ret = (float32x4_t) __builtin_neon_vrecpsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
2201422014 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2201522015 return __ret;
......@@ -22024,9 +22024,9 @@ __ai float32x2_t vrecps_f32(float32x2_t __p0, float32x2_t __p1) {
2202422024}
2202522025#else
2202622026__ai float32x2_t vrecps_f32(float32x2_t __p0, float32x2_t __p1) {
22027 float32x2_t __ret;
2202722028 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2202822029 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22029 float32x2_t __ret;
2203022030 __ret = (float32x2_t) __builtin_neon_vrecps_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
2203122031 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2203222032 return __ret;
......@@ -22041,8 +22041,8 @@ __ai poly8x8_t vrev16_p8(poly8x8_t __p0) {
2204122041}
2204222042#else
2204322043__ai poly8x8_t vrev16_p8(poly8x8_t __p0) {
22044 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2204522044 poly8x8_t __ret;
22045 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2204622046 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6);
2204722047 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2204822048 return __ret;
......@@ -22057,8 +22057,8 @@ __ai poly8x16_t vrev16q_p8(poly8x16_t __p0) {
2205722057}
2205822058#else
2205922059__ai poly8x16_t vrev16q_p8(poly8x16_t __p0) {
22060 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2206122060 poly8x16_t __ret;
22061 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2206222062 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6, 9, 8, 11, 10, 13, 12, 15, 14);
2206322063 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2206422064 return __ret;
......@@ -22073,8 +22073,8 @@ __ai uint8x16_t vrev16q_u8(uint8x16_t __p0) {
2207322073}
2207422074#else
2207522075__ai uint8x16_t vrev16q_u8(uint8x16_t __p0) {
22076 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2207722076 uint8x16_t __ret;
22077 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2207822078 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6, 9, 8, 11, 10, 13, 12, 15, 14);
2207922079 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2208022080 return __ret;
......@@ -22089,8 +22089,8 @@ __ai int8x16_t vrev16q_s8(int8x16_t __p0) {
2208922089}
2209022090#else
2209122091__ai int8x16_t vrev16q_s8(int8x16_t __p0) {
22092 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2209322092 int8x16_t __ret;
22093 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2209422094 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6, 9, 8, 11, 10, 13, 12, 15, 14);
2209522095 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2209622096 return __ret;
......@@ -22105,8 +22105,8 @@ __ai uint8x8_t vrev16_u8(uint8x8_t __p0) {
2210522105}
2210622106#else
2210722107__ai uint8x8_t vrev16_u8(uint8x8_t __p0) {
22108 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2210922108 uint8x8_t __ret;
22109 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2211022110 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6);
2211122111 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2211222112 return __ret;
......@@ -22121,8 +22121,8 @@ __ai int8x8_t vrev16_s8(int8x8_t __p0) {
2212122121}
2212222122#else
2212322123__ai int8x8_t vrev16_s8(int8x8_t __p0) {
22124 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2212522124 int8x8_t __ret;
22125 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2212622126 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6);
2212722127 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2212822128 return __ret;
......@@ -22137,8 +22137,8 @@ __ai poly8x8_t vrev32_p8(poly8x8_t __p0) {
2213722137}
2213822138#else
2213922139__ai poly8x8_t vrev32_p8(poly8x8_t __p0) {
22140 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2214122140 poly8x8_t __ret;
22141 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2214222142 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
2214322143 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2214422144 return __ret;
......@@ -22153,8 +22153,8 @@ __ai poly16x4_t vrev32_p16(poly16x4_t __p0) {
2215322153}
2215422154#else
2215522155__ai poly16x4_t vrev32_p16(poly16x4_t __p0) {
22156 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2215722156 poly16x4_t __ret;
22157 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2215822158 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2);
2215922159 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2216022160 return __ret;
......@@ -22169,8 +22169,8 @@ __ai poly8x16_t vrev32q_p8(poly8x16_t __p0) {
2216922169}
2217022170#else
2217122171__ai poly8x16_t vrev32q_p8(poly8x16_t __p0) {
22172 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2217322172 poly8x16_t __ret;
22173 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2217422174 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12);
2217522175 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2217622176 return __ret;
......@@ -22185,8 +22185,8 @@ __ai poly16x8_t vrev32q_p16(poly16x8_t __p0) {
2218522185}
2218622186#else
2218722187__ai poly16x8_t vrev32q_p16(poly16x8_t __p0) {
22188 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2218922188 poly16x8_t __ret;
22189 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2219022190 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6);
2219122191 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2219222192 return __ret;
......@@ -22201,8 +22201,8 @@ __ai uint8x16_t vrev32q_u8(uint8x16_t __p0) {
2220122201}
2220222202#else
2220322203__ai uint8x16_t vrev32q_u8(uint8x16_t __p0) {
22204 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2220522204 uint8x16_t __ret;
22205 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2220622206 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12);
2220722207 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2220822208 return __ret;
......@@ -22217,8 +22217,8 @@ __ai uint16x8_t vrev32q_u16(uint16x8_t __p0) {
2221722217}
2221822218#else
2221922219__ai uint16x8_t vrev32q_u16(uint16x8_t __p0) {
22220 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2222122220 uint16x8_t __ret;
22221 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2222222222 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6);
2222322223 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2222422224 return __ret;
......@@ -22233,8 +22233,8 @@ __ai int8x16_t vrev32q_s8(int8x16_t __p0) {
2223322233}
2223422234#else
2223522235__ai int8x16_t vrev32q_s8(int8x16_t __p0) {
22236 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2223722236 int8x16_t __ret;
22237 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2223822238 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12);
2223922239 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2224022240 return __ret;
......@@ -22249,8 +22249,8 @@ __ai int16x8_t vrev32q_s16(int16x8_t __p0) {
2224922249}
2225022250#else
2225122251__ai int16x8_t vrev32q_s16(int16x8_t __p0) {
22252 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2225322252 int16x8_t __ret;
22253 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2225422254 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2, 5, 4, 7, 6);
2225522255 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2225622256 return __ret;
......@@ -22265,8 +22265,8 @@ __ai uint8x8_t vrev32_u8(uint8x8_t __p0) {
2226522265}
2226622266#else
2226722267__ai uint8x8_t vrev32_u8(uint8x8_t __p0) {
22268 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2226922268 uint8x8_t __ret;
22269 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2227022270 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
2227122271 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2227222272 return __ret;
......@@ -22281,8 +22281,8 @@ __ai uint16x4_t vrev32_u16(uint16x4_t __p0) {
2228122281}
2228222282#else
2228322283__ai uint16x4_t vrev32_u16(uint16x4_t __p0) {
22284 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2228522284 uint16x4_t __ret;
22285 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2228622286 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2);
2228722287 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2228822288 return __ret;
......@@ -22297,8 +22297,8 @@ __ai int8x8_t vrev32_s8(int8x8_t __p0) {
2229722297}
2229822298#else
2229922299__ai int8x8_t vrev32_s8(int8x8_t __p0) {
22300 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2230122300 int8x8_t __ret;
22301 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2230222302 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
2230322303 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2230422304 return __ret;
......@@ -22313,8 +22313,8 @@ __ai int16x4_t vrev32_s16(int16x4_t __p0) {
2231322313}
2231422314#else
2231522315__ai int16x4_t vrev32_s16(int16x4_t __p0) {
22316 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2231722316 int16x4_t __ret;
22317 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2231822318 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2);
2231922319 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2232022320 return __ret;
......@@ -22329,8 +22329,8 @@ __ai poly8x8_t vrev64_p8(poly8x8_t __p0) {
2232922329}
2233022330#else
2233122331__ai poly8x8_t vrev64_p8(poly8x8_t __p0) {
22332 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2233322332 poly8x8_t __ret;
22333 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2233422334 __ret = __builtin_shufflevector(__rev0, __rev0, 7, 6, 5, 4, 3, 2, 1, 0);
2233522335 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2233622336 return __ret;
......@@ -22345,8 +22345,8 @@ __ai poly16x4_t vrev64_p16(poly16x4_t __p0) {
2234522345}
2234622346#else
2234722347__ai poly16x4_t vrev64_p16(poly16x4_t __p0) {
22348 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2234922348 poly16x4_t __ret;
22349 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2235022350 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0);
2235122351 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2235222352 return __ret;
......@@ -22361,8 +22361,8 @@ __ai poly8x16_t vrev64q_p8(poly8x16_t __p0) {
2236122361}
2236222362#else
2236322363__ai poly8x16_t vrev64q_p8(poly8x16_t __p0) {
22364 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2236522364 poly8x16_t __ret;
22365 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2236622366 __ret = __builtin_shufflevector(__rev0, __rev0, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14, 13, 12, 11, 10, 9, 8);
2236722367 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2236822368 return __ret;
......@@ -22377,8 +22377,8 @@ __ai poly16x8_t vrev64q_p16(poly16x8_t __p0) {
2237722377}
2237822378#else
2237922379__ai poly16x8_t vrev64q_p16(poly16x8_t __p0) {
22380 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2238122380 poly16x8_t __ret;
22381 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2238222382 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
2238322383 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2238422384 return __ret;
......@@ -22393,8 +22393,8 @@ __ai uint8x16_t vrev64q_u8(uint8x16_t __p0) {
2239322393}
2239422394#else
2239522395__ai uint8x16_t vrev64q_u8(uint8x16_t __p0) {
22396 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2239722396 uint8x16_t __ret;
22397 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2239822398 __ret = __builtin_shufflevector(__rev0, __rev0, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14, 13, 12, 11, 10, 9, 8);
2239922399 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2240022400 return __ret;
......@@ -22409,8 +22409,8 @@ __ai uint32x4_t vrev64q_u32(uint32x4_t __p0) {
2240922409}
2241022410#else
2241122411__ai uint32x4_t vrev64q_u32(uint32x4_t __p0) {
22412 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2241322412 uint32x4_t __ret;
22413 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2241422414 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2);
2241522415 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2241622416 return __ret;
......@@ -22425,8 +22425,8 @@ __ai uint16x8_t vrev64q_u16(uint16x8_t __p0) {
2242522425}
2242622426#else
2242722427__ai uint16x8_t vrev64q_u16(uint16x8_t __p0) {
22428 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2242922428 uint16x8_t __ret;
22429 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2243022430 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
2243122431 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2243222432 return __ret;
......@@ -22441,8 +22441,8 @@ __ai int8x16_t vrev64q_s8(int8x16_t __p0) {
2244122441}
2244222442#else
2244322443__ai int8x16_t vrev64q_s8(int8x16_t __p0) {
22444 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2244522444 int8x16_t __ret;
22445 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2244622446 __ret = __builtin_shufflevector(__rev0, __rev0, 7, 6, 5, 4, 3, 2, 1, 0, 15, 14, 13, 12, 11, 10, 9, 8);
2244722447 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2244822448 return __ret;
......@@ -22457,8 +22457,8 @@ __ai float32x4_t vrev64q_f32(float32x4_t __p0) {
2245722457}
2245822458#else
2245922459__ai float32x4_t vrev64q_f32(float32x4_t __p0) {
22460 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2246122460 float32x4_t __ret;
22461 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2246222462 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2);
2246322463 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2246422464 return __ret;
......@@ -22473,8 +22473,8 @@ __ai int32x4_t vrev64q_s32(int32x4_t __p0) {
2247322473}
2247422474#else
2247522475__ai int32x4_t vrev64q_s32(int32x4_t __p0) {
22476 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2247722476 int32x4_t __ret;
22477 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2247822478 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0, 3, 2);
2247922479 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2248022480 return __ret;
......@@ -22489,8 +22489,8 @@ __ai int16x8_t vrev64q_s16(int16x8_t __p0) {
2248922489}
2249022490#else
2249122491__ai int16x8_t vrev64q_s16(int16x8_t __p0) {
22492 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2249322492 int16x8_t __ret;
22493 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2249422494 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
2249522495 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2249622496 return __ret;
......@@ -22505,8 +22505,8 @@ __ai uint8x8_t vrev64_u8(uint8x8_t __p0) {
2250522505}
2250622506#else
2250722507__ai uint8x8_t vrev64_u8(uint8x8_t __p0) {
22508 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2250922508 uint8x8_t __ret;
22509 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2251022510 __ret = __builtin_shufflevector(__rev0, __rev0, 7, 6, 5, 4, 3, 2, 1, 0);
2251122511 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2251222512 return __ret;
......@@ -22521,8 +22521,8 @@ __ai uint32x2_t vrev64_u32(uint32x2_t __p0) {
2252122521}
2252222522#else
2252322523__ai uint32x2_t vrev64_u32(uint32x2_t __p0) {
22524 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2252522524 uint32x2_t __ret;
22525 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2252622526 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0);
2252722527 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2252822528 return __ret;
......@@ -22537,8 +22537,8 @@ __ai uint16x4_t vrev64_u16(uint16x4_t __p0) {
2253722537}
2253822538#else
2253922539__ai uint16x4_t vrev64_u16(uint16x4_t __p0) {
22540 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2254122540 uint16x4_t __ret;
22541 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2254222542 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0);
2254322543 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2254422544 return __ret;
......@@ -22553,8 +22553,8 @@ __ai int8x8_t vrev64_s8(int8x8_t __p0) {
2255322553}
2255422554#else
2255522555__ai int8x8_t vrev64_s8(int8x8_t __p0) {
22556 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2255722556 int8x8_t __ret;
22557 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2255822558 __ret = __builtin_shufflevector(__rev0, __rev0, 7, 6, 5, 4, 3, 2, 1, 0);
2255922559 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2256022560 return __ret;
......@@ -22569,8 +22569,8 @@ __ai float32x2_t vrev64_f32(float32x2_t __p0) {
2256922569}
2257022570#else
2257122571__ai float32x2_t vrev64_f32(float32x2_t __p0) {
22572 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2257322572 float32x2_t __ret;
22573 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2257422574 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0);
2257522575 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2257622576 return __ret;
......@@ -22585,8 +22585,8 @@ __ai int32x2_t vrev64_s32(int32x2_t __p0) {
2258522585}
2258622586#else
2258722587__ai int32x2_t vrev64_s32(int32x2_t __p0) {
22588 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2258922588 int32x2_t __ret;
22589 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2259022590 __ret = __builtin_shufflevector(__rev0, __rev0, 1, 0);
2259122591 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2259222592 return __ret;
......@@ -22601,8 +22601,8 @@ __ai int16x4_t vrev64_s16(int16x4_t __p0) {
2260122601}
2260222602#else
2260322603__ai int16x4_t vrev64_s16(int16x4_t __p0) {
22604 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2260522604 int16x4_t __ret;
22605 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2260622606 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0);
2260722607 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2260822608 return __ret;
......@@ -22617,9 +22617,9 @@ __ai uint8x16_t vrhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2261722617}
2261822618#else
2261922619__ai uint8x16_t vrhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
22620 uint8x16_t __ret;
2262022621 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2262122622 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22622 uint8x16_t __ret;
2262322623 __ret = (uint8x16_t) __builtin_neon_vrhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
2262422624 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2262522625 return __ret;
......@@ -22634,9 +22634,9 @@ __ai uint32x4_t vrhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2263422634}
2263522635#else
2263622636__ai uint32x4_t vrhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
22637 uint32x4_t __ret;
2263722638 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2263822639 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22639 uint32x4_t __ret;
2264022640 __ret = (uint32x4_t) __builtin_neon_vrhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
2264122641 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2264222642 return __ret;
......@@ -22651,9 +22651,9 @@ __ai uint16x8_t vrhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2265122651}
2265222652#else
2265322653__ai uint16x8_t vrhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
22654 uint16x8_t __ret;
2265422655 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2265522656 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22656 uint16x8_t __ret;
2265722657 __ret = (uint16x8_t) __builtin_neon_vrhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
2265822658 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2265922659 return __ret;
......@@ -22668,9 +22668,9 @@ __ai int8x16_t vrhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
2266822668}
2266922669#else
2267022670__ai int8x16_t vrhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
22671 int8x16_t __ret;
2267122672 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2267222673 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22673 int8x16_t __ret;
2267422674 __ret = (int8x16_t) __builtin_neon_vrhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
2267522675 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2267622676 return __ret;
......@@ -22685,9 +22685,9 @@ __ai int32x4_t vrhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
2268522685}
2268622686#else
2268722687__ai int32x4_t vrhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
22688 int32x4_t __ret;
2268822689 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2268922690 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22690 int32x4_t __ret;
2269122691 __ret = (int32x4_t) __builtin_neon_vrhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2269222692 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2269322693 return __ret;
......@@ -22702,9 +22702,9 @@ __ai int16x8_t vrhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
2270222702}
2270322703#else
2270422704__ai int16x8_t vrhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
22705 int16x8_t __ret;
2270522706 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2270622707 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22707 int16x8_t __ret;
2270822708 __ret = (int16x8_t) __builtin_neon_vrhaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2270922709 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2271022710 return __ret;
......@@ -22719,9 +22719,9 @@ __ai uint8x8_t vrhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
2271922719}
2272022720#else
2272122721__ai uint8x8_t vrhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
22722 uint8x8_t __ret;
2272222723 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2272322724 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22724 uint8x8_t __ret;
2272522725 __ret = (uint8x8_t) __builtin_neon_vrhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
2272622726 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2272722727 return __ret;
......@@ -22736,9 +22736,9 @@ __ai uint32x2_t vrhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
2273622736}
2273722737#else
2273822738__ai uint32x2_t vrhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
22739 uint32x2_t __ret;
2273922740 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2274022741 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22741 uint32x2_t __ret;
2274222742 __ret = (uint32x2_t) __builtin_neon_vrhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
2274322743 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2274422744 return __ret;
......@@ -22753,9 +22753,9 @@ __ai uint16x4_t vrhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
2275322753}
2275422754#else
2275522755__ai uint16x4_t vrhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
22756 uint16x4_t __ret;
2275622757 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2275722758 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22758 uint16x4_t __ret;
2275922759 __ret = (uint16x4_t) __builtin_neon_vrhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
2276022760 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2276122761 return __ret;
......@@ -22770,9 +22770,9 @@ __ai int8x8_t vrhadd_s8(int8x8_t __p0, int8x8_t __p1) {
2277022770}
2277122771#else
2277222772__ai int8x8_t vrhadd_s8(int8x8_t __p0, int8x8_t __p1) {
22773 int8x8_t __ret;
2277322774 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2277422775 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22775 int8x8_t __ret;
2277622776 __ret = (int8x8_t) __builtin_neon_vrhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
2277722777 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2277822778 return __ret;
......@@ -22787,9 +22787,9 @@ __ai int32x2_t vrhadd_s32(int32x2_t __p0, int32x2_t __p1) {
2278722787}
2278822788#else
2278922789__ai int32x2_t vrhadd_s32(int32x2_t __p0, int32x2_t __p1) {
22790 int32x2_t __ret;
2279022791 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2279122792 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22792 int32x2_t __ret;
2279322793 __ret = (int32x2_t) __builtin_neon_vrhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2279422794 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2279522795 return __ret;
......@@ -22804,9 +22804,9 @@ __ai int16x4_t vrhadd_s16(int16x4_t __p0, int16x4_t __p1) {
2280422804}
2280522805#else
2280622806__ai int16x4_t vrhadd_s16(int16x4_t __p0, int16x4_t __p1) {
22807 int16x4_t __ret;
2280722808 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2280822809 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22809 int16x4_t __ret;
2281022810 __ret = (int16x4_t) __builtin_neon_vrhadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2281122811 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2281222812 return __ret;
......@@ -22821,9 +22821,9 @@ __ai uint8x16_t vrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
2282122821}
2282222822#else
2282322823__ai uint8x16_t vrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
22824 uint8x16_t __ret;
2282422825 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2282522826 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22826 uint8x16_t __ret;
2282722827 __ret = (uint8x16_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
2282822828 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2282922829 return __ret;
......@@ -22838,9 +22838,9 @@ __ai uint32x4_t vrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
2283822838}
2283922839#else
2284022840__ai uint32x4_t vrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
22841 uint32x4_t __ret;
2284122842 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2284222843 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22843 uint32x4_t __ret;
2284422844 __ret = (uint32x4_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
2284522845 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2284622846 return __ret;
......@@ -22855,9 +22855,9 @@ __ai uint64x2_t vrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
2285522855}
2285622856#else
2285722857__ai uint64x2_t vrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
22858 uint64x2_t __ret;
2285822859 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2285922860 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22860 uint64x2_t __ret;
2286122861 __ret = (uint64x2_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
2286222862 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2286322863 return __ret;
......@@ -22872,9 +22872,9 @@ __ai uint16x8_t vrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
2287222872}
2287322873#else
2287422874__ai uint16x8_t vrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
22875 uint16x8_t __ret;
2287522876 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2287622877 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22877 uint16x8_t __ret;
2287822878 __ret = (uint16x8_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
2287922879 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2288022880 return __ret;
......@@ -22889,9 +22889,9 @@ __ai int8x16_t vrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
2288922889}
2289022890#else
2289122891__ai int8x16_t vrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
22892 int8x16_t __ret;
2289222893 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2289322894 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22894 int8x16_t __ret;
2289522895 __ret = (int8x16_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
2289622896 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2289722897 return __ret;
......@@ -22906,9 +22906,9 @@ __ai int32x4_t vrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
2290622906}
2290722907#else
2290822908__ai int32x4_t vrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
22909 int32x4_t __ret;
2290922910 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2291022911 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22911 int32x4_t __ret;
2291222912 __ret = (int32x4_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2291322913 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2291422914 return __ret;
......@@ -22923,9 +22923,9 @@ __ai int64x2_t vrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
2292322923}
2292422924#else
2292522925__ai int64x2_t vrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
22926 int64x2_t __ret;
2292622927 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2292722928 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22928 int64x2_t __ret;
2292922929 __ret = (int64x2_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
2293022930 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2293122931 return __ret;
......@@ -22940,9 +22940,9 @@ __ai int16x8_t vrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
2294022940}
2294122941#else
2294222942__ai int16x8_t vrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
22943 int16x8_t __ret;
2294322944 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2294422945 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22945 int16x8_t __ret;
2294622946 __ret = (int16x8_t) __builtin_neon_vrshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2294722947 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2294822948 return __ret;
......@@ -22957,9 +22957,9 @@ __ai uint8x8_t vrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
2295722957}
2295822958#else
2295922959__ai uint8x8_t vrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
22960 uint8x8_t __ret;
2296022961 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2296122962 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22962 uint8x8_t __ret;
2296322963 __ret = (uint8x8_t) __builtin_neon_vrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
2296422964 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2296522965 return __ret;
......@@ -22974,9 +22974,9 @@ __ai uint32x2_t vrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
2297422974}
2297522975#else
2297622976__ai uint32x2_t vrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
22977 uint32x2_t __ret;
2297722978 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2297822979 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22979 uint32x2_t __ret;
2298022980 __ret = (uint32x2_t) __builtin_neon_vrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
2298122981 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2298222982 return __ret;
......@@ -22996,9 +22996,9 @@ __ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
2299622996}
2299722997#else
2299822998__ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
22999 uint16x4_t __ret;
2299923000 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2300023001 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
23001 uint16x4_t __ret;
2300223002 __ret = (uint16x4_t) __builtin_neon_vrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
2300323003 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2300423004 return __ret;
......@@ -23013,9 +23013,9 @@ __ai int8x8_t vrshl_s8(int8x8_t __p0, int8x8_t __p1) {
2301323013}
2301423014#else
2301523015__ai int8x8_t vrshl_s8(int8x8_t __p0, int8x8_t __p1) {
23016 int8x8_t __ret;
2301623017 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2301723018 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
23018 int8x8_t __ret;
2301923019 __ret = (int8x8_t) __builtin_neon_vrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
2302023020 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2302123021 return __ret;
......@@ -23030,9 +23030,9 @@ __ai int32x2_t vrshl_s32(int32x2_t __p0, int32x2_t __p1) {
2303023030}
2303123031#else
2303223032__ai int32x2_t vrshl_s32(int32x2_t __p0, int32x2_t __p1) {
23033 int32x2_t __ret;
2303323034 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2303423035 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
23035 int32x2_t __ret;
2303623036 __ret = (int32x2_t) __builtin_neon_vrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2303723037 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2303823038 return __ret;
......@@ -23052,9 +23052,9 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2305223052}
2305323053#else
2305423054__ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
23055 int16x4_t __ret;
2305523056 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2305623057 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
23057 int16x4_t __ret;
2305823058 __ret = (int16x4_t) __builtin_neon_vrshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2305923059 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2306023060 return __ret;
......@@ -23063,16 +23063,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2306323063
2306423064#ifdef __LITTLE_ENDIAN__
2306523065#define vrshrq_n_u8(__p0, __p1) __extension__ ({ \
23066 uint8x16_t __s0 = __p0; \
2306723066 uint8x16_t __ret; \
23067 uint8x16_t __s0 = __p0; \
2306823068 __ret = (uint8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 48); \
2306923069 __ret; \
2307023070})
2307123071#else
2307223072#define vrshrq_n_u8(__p0, __p1) __extension__ ({ \
23073 uint8x16_t __ret; \
2307323074 uint8x16_t __s0 = __p0; \
2307423075 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
23075 uint8x16_t __ret; \
2307623076 __ret = (uint8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 48); \
2307723077 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2307823078 __ret; \
......@@ -23081,16 +23081,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2308123081
2308223082#ifdef __LITTLE_ENDIAN__
2308323083#define vrshrq_n_u32(__p0, __p1) __extension__ ({ \
23084 uint32x4_t __s0 = __p0; \
2308523084 uint32x4_t __ret; \
23085 uint32x4_t __s0 = __p0; \
2308623086 __ret = (uint32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 50); \
2308723087 __ret; \
2308823088})
2308923089#else
2309023090#define vrshrq_n_u32(__p0, __p1) __extension__ ({ \
23091 uint32x4_t __ret; \
2309123092 uint32x4_t __s0 = __p0; \
2309223093 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
23093 uint32x4_t __ret; \
2309423094 __ret = (uint32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 50); \
2309523095 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2309623096 __ret; \
......@@ -23099,16 +23099,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2309923099
2310023100#ifdef __LITTLE_ENDIAN__
2310123101#define vrshrq_n_u64(__p0, __p1) __extension__ ({ \
23102 uint64x2_t __s0 = __p0; \
2310323102 uint64x2_t __ret; \
23103 uint64x2_t __s0 = __p0; \
2310423104 __ret = (uint64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 51); \
2310523105 __ret; \
2310623106})
2310723107#else
2310823108#define vrshrq_n_u64(__p0, __p1) __extension__ ({ \
23109 uint64x2_t __ret; \
2310923110 uint64x2_t __s0 = __p0; \
2311023111 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
23111 uint64x2_t __ret; \
2311223112 __ret = (uint64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 51); \
2311323113 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2311423114 __ret; \
......@@ -23117,16 +23117,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2311723117
2311823118#ifdef __LITTLE_ENDIAN__
2311923119#define vrshrq_n_u16(__p0, __p1) __extension__ ({ \
23120 uint16x8_t __s0 = __p0; \
2312123120 uint16x8_t __ret; \
23121 uint16x8_t __s0 = __p0; \
2312223122 __ret = (uint16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 49); \
2312323123 __ret; \
2312423124})
2312523125#else
2312623126#define vrshrq_n_u16(__p0, __p1) __extension__ ({ \
23127 uint16x8_t __ret; \
2312723128 uint16x8_t __s0 = __p0; \
2312823129 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
23129 uint16x8_t __ret; \
2313023130 __ret = (uint16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 49); \
2313123131 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2313223132 __ret; \
......@@ -23135,16 +23135,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2313523135
2313623136#ifdef __LITTLE_ENDIAN__
2313723137#define vrshrq_n_s8(__p0, __p1) __extension__ ({ \
23138 int8x16_t __s0 = __p0; \
2313923138 int8x16_t __ret; \
23139 int8x16_t __s0 = __p0; \
2314023140 __ret = (int8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 32); \
2314123141 __ret; \
2314223142})
2314323143#else
2314423144#define vrshrq_n_s8(__p0, __p1) __extension__ ({ \
23145 int8x16_t __ret; \
2314523146 int8x16_t __s0 = __p0; \
2314623147 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
23147 int8x16_t __ret; \
2314823148 __ret = (int8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 32); \
2314923149 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2315023150 __ret; \
......@@ -23153,16 +23153,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2315323153
2315423154#ifdef __LITTLE_ENDIAN__
2315523155#define vrshrq_n_s32(__p0, __p1) __extension__ ({ \
23156 int32x4_t __s0 = __p0; \
2315723156 int32x4_t __ret; \
23157 int32x4_t __s0 = __p0; \
2315823158 __ret = (int32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 34); \
2315923159 __ret; \
2316023160})
2316123161#else
2316223162#define vrshrq_n_s32(__p0, __p1) __extension__ ({ \
23163 int32x4_t __ret; \
2316323164 int32x4_t __s0 = __p0; \
2316423165 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
23165 int32x4_t __ret; \
2316623166 __ret = (int32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 34); \
2316723167 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2316823168 __ret; \
......@@ -23171,16 +23171,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2317123171
2317223172#ifdef __LITTLE_ENDIAN__
2317323173#define vrshrq_n_s64(__p0, __p1) __extension__ ({ \
23174 int64x2_t __s0 = __p0; \
2317523174 int64x2_t __ret; \
23175 int64x2_t __s0 = __p0; \
2317623176 __ret = (int64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 35); \
2317723177 __ret; \
2317823178})
2317923179#else
2318023180#define vrshrq_n_s64(__p0, __p1) __extension__ ({ \
23181 int64x2_t __ret; \
2318123182 int64x2_t __s0 = __p0; \
2318223183 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
23183 int64x2_t __ret; \
2318423184 __ret = (int64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 35); \
2318523185 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2318623186 __ret; \
......@@ -23189,16 +23189,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2318923189
2319023190#ifdef __LITTLE_ENDIAN__
2319123191#define vrshrq_n_s16(__p0, __p1) __extension__ ({ \
23192 int16x8_t __s0 = __p0; \
2319323192 int16x8_t __ret; \
23193 int16x8_t __s0 = __p0; \
2319423194 __ret = (int16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 33); \
2319523195 __ret; \
2319623196})
2319723197#else
2319823198#define vrshrq_n_s16(__p0, __p1) __extension__ ({ \
23199 int16x8_t __ret; \
2319923200 int16x8_t __s0 = __p0; \
2320023201 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
23201 int16x8_t __ret; \
2320223202 __ret = (int16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 33); \
2320323203 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2320423204 __ret; \
......@@ -23207,16 +23207,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2320723207
2320823208#ifdef __LITTLE_ENDIAN__
2320923209#define vrshr_n_u8(__p0, __p1) __extension__ ({ \
23210 uint8x8_t __s0 = __p0; \
2321123210 uint8x8_t __ret; \
23211 uint8x8_t __s0 = __p0; \
2321223212 __ret = (uint8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 16); \
2321323213 __ret; \
2321423214})
2321523215#else
2321623216#define vrshr_n_u8(__p0, __p1) __extension__ ({ \
23217 uint8x8_t __ret; \
2321723218 uint8x8_t __s0 = __p0; \
2321823219 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
23219 uint8x8_t __ret; \
2322023220 __ret = (uint8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 16); \
2322123221 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2322223222 __ret; \
......@@ -23225,16 +23225,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2322523225
2322623226#ifdef __LITTLE_ENDIAN__
2322723227#define vrshr_n_u32(__p0, __p1) __extension__ ({ \
23228 uint32x2_t __s0 = __p0; \
2322923228 uint32x2_t __ret; \
23229 uint32x2_t __s0 = __p0; \
2323023230 __ret = (uint32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 18); \
2323123231 __ret; \
2323223232})
2323323233#else
2323423234#define vrshr_n_u32(__p0, __p1) __extension__ ({ \
23235 uint32x2_t __ret; \
2323523236 uint32x2_t __s0 = __p0; \
2323623237 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
23237 uint32x2_t __ret; \
2323823238 __ret = (uint32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 18); \
2323923239 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2324023240 __ret; \
......@@ -23242,23 +23242,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2324223242#endif
2324323243
2324423244#define vrshr_n_u64(__p0, __p1) __extension__ ({ \
23245 uint64x1_t __s0 = __p0; \
2324623245 uint64x1_t __ret; \
23246 uint64x1_t __s0 = __p0; \
2324723247 __ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
2324823248 __ret; \
2324923249})
2325023250#ifdef __LITTLE_ENDIAN__
2325123251#define vrshr_n_u16(__p0, __p1) __extension__ ({ \
23252 uint16x4_t __s0 = __p0; \
2325323252 uint16x4_t __ret; \
23253 uint16x4_t __s0 = __p0; \
2325423254 __ret = (uint16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 17); \
2325523255 __ret; \
2325623256})
2325723257#else
2325823258#define vrshr_n_u16(__p0, __p1) __extension__ ({ \
23259 uint16x4_t __ret; \
2325923260 uint16x4_t __s0 = __p0; \
2326023261 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
23261 uint16x4_t __ret; \
2326223262 __ret = (uint16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 17); \
2326323263 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2326423264 __ret; \
......@@ -23267,16 +23267,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2326723267
2326823268#ifdef __LITTLE_ENDIAN__
2326923269#define vrshr_n_s8(__p0, __p1) __extension__ ({ \
23270 int8x8_t __s0 = __p0; \
2327123270 int8x8_t __ret; \
23271 int8x8_t __s0 = __p0; \
2327223272 __ret = (int8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 0); \
2327323273 __ret; \
2327423274})
2327523275#else
2327623276#define vrshr_n_s8(__p0, __p1) __extension__ ({ \
23277 int8x8_t __ret; \
2327723278 int8x8_t __s0 = __p0; \
2327823279 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
23279 int8x8_t __ret; \
2328023280 __ret = (int8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 0); \
2328123281 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2328223282 __ret; \
......@@ -23285,16 +23285,16 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2328523285
2328623286#ifdef __LITTLE_ENDIAN__
2328723287#define vrshr_n_s32(__p0, __p1) __extension__ ({ \
23288 int32x2_t __s0 = __p0; \
2328923288 int32x2_t __ret; \
23289 int32x2_t __s0 = __p0; \
2329023290 __ret = (int32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 2); \
2329123291 __ret; \
2329223292})
2329323293#else
2329423294#define vrshr_n_s32(__p0, __p1) __extension__ ({ \
23295 int32x2_t __ret; \
2329523296 int32x2_t __s0 = __p0; \
2329623297 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
23297 int32x2_t __ret; \
2329823298 __ret = (int32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 2); \
2329923299 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2330023300 __ret; \
......@@ -23302,23 +23302,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2330223302#endif
2330323303
2330423304#define vrshr_n_s64(__p0, __p1) __extension__ ({ \
23305 int64x1_t __s0 = __p0; \
2330623305 int64x1_t __ret; \
23306 int64x1_t __s0 = __p0; \
2330723307 __ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
2330823308 __ret; \
2330923309})
2331023310#ifdef __LITTLE_ENDIAN__
2331123311#define vrshr_n_s16(__p0, __p1) __extension__ ({ \
23312 int16x4_t __s0 = __p0; \
2331323312 int16x4_t __ret; \
23313 int16x4_t __s0 = __p0; \
2331423314 __ret = (int16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 1); \
2331523315 __ret; \
2331623316})
2331723317#else
2331823318#define vrshr_n_s16(__p0, __p1) __extension__ ({ \
23319 int16x4_t __ret; \
2331923320 int16x4_t __s0 = __p0; \
2332023321 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
23321 int16x4_t __ret; \
2332223322 __ret = (int16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 1); \
2332323323 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2332423324 __ret; \
......@@ -23327,23 +23327,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2332723327
2332823328#ifdef __LITTLE_ENDIAN__
2332923329#define vrshrn_n_u32(__p0, __p1) __extension__ ({ \
23330 uint32x4_t __s0 = __p0; \
2333123330 uint16x4_t __ret; \
23331 uint32x4_t __s0 = __p0; \
2333223332 __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 17); \
2333323333 __ret; \
2333423334})
2333523335#else
2333623336#define vrshrn_n_u32(__p0, __p1) __extension__ ({ \
23337 uint16x4_t __ret; \
2333723338 uint32x4_t __s0 = __p0; \
2333823339 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
23339 uint16x4_t __ret; \
2334023340 __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 17); \
2334123341 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2334223342 __ret; \
2334323343})
2334423344#define __noswap_vrshrn_n_u32(__p0, __p1) __extension__ ({ \
23345 uint32x4_t __s0 = __p0; \
2334623345 uint16x4_t __ret; \
23346 uint32x4_t __s0 = __p0; \
2334723347 __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 17); \
2334823348 __ret; \
2334923349})
......@@ -23351,23 +23351,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2335123351
2335223352#ifdef __LITTLE_ENDIAN__
2335323353#define vrshrn_n_u64(__p0, __p1) __extension__ ({ \
23354 uint64x2_t __s0 = __p0; \
2335523354 uint32x2_t __ret; \
23355 uint64x2_t __s0 = __p0; \
2335623356 __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 18); \
2335723357 __ret; \
2335823358})
2335923359#else
2336023360#define vrshrn_n_u64(__p0, __p1) __extension__ ({ \
23361 uint32x2_t __ret; \
2336123362 uint64x2_t __s0 = __p0; \
2336223363 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
23363 uint32x2_t __ret; \
2336423364 __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 18); \
2336523365 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2336623366 __ret; \
2336723367})
2336823368#define __noswap_vrshrn_n_u64(__p0, __p1) __extension__ ({ \
23369 uint64x2_t __s0 = __p0; \
2337023369 uint32x2_t __ret; \
23370 uint64x2_t __s0 = __p0; \
2337123371 __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 18); \
2337223372 __ret; \
2337323373})
......@@ -23375,23 +23375,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2337523375
2337623376#ifdef __LITTLE_ENDIAN__
2337723377#define vrshrn_n_u16(__p0, __p1) __extension__ ({ \
23378 uint16x8_t __s0 = __p0; \
2337923378 uint8x8_t __ret; \
23379 uint16x8_t __s0 = __p0; \
2338023380 __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 16); \
2338123381 __ret; \
2338223382})
2338323383#else
2338423384#define vrshrn_n_u16(__p0, __p1) __extension__ ({ \
23385 uint8x8_t __ret; \
2338523386 uint16x8_t __s0 = __p0; \
2338623387 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
23387 uint8x8_t __ret; \
2338823388 __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 16); \
2338923389 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2339023390 __ret; \
2339123391})
2339223392#define __noswap_vrshrn_n_u16(__p0, __p1) __extension__ ({ \
23393 uint16x8_t __s0 = __p0; \
2339423393 uint8x8_t __ret; \
23394 uint16x8_t __s0 = __p0; \
2339523395 __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 16); \
2339623396 __ret; \
2339723397})
......@@ -23399,23 +23399,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2339923399
2340023400#ifdef __LITTLE_ENDIAN__
2340123401#define vrshrn_n_s32(__p0, __p1) __extension__ ({ \
23402 int32x4_t __s0 = __p0; \
2340323402 int16x4_t __ret; \
23403 int32x4_t __s0 = __p0; \
2340423404 __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 1); \
2340523405 __ret; \
2340623406})
2340723407#else
2340823408#define vrshrn_n_s32(__p0, __p1) __extension__ ({ \
23409 int16x4_t __ret; \
2340923410 int32x4_t __s0 = __p0; \
2341023411 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
23411 int16x4_t __ret; \
2341223412 __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 1); \
2341323413 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2341423414 __ret; \
2341523415})
2341623416#define __noswap_vrshrn_n_s32(__p0, __p1) __extension__ ({ \
23417 int32x4_t __s0 = __p0; \
2341823417 int16x4_t __ret; \
23418 int32x4_t __s0 = __p0; \
2341923419 __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 1); \
2342023420 __ret; \
2342123421})
......@@ -23423,23 +23423,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2342323423
2342423424#ifdef __LITTLE_ENDIAN__
2342523425#define vrshrn_n_s64(__p0, __p1) __extension__ ({ \
23426 int64x2_t __s0 = __p0; \
2342723426 int32x2_t __ret; \
23427 int64x2_t __s0 = __p0; \
2342823428 __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 2); \
2342923429 __ret; \
2343023430})
2343123431#else
2343223432#define vrshrn_n_s64(__p0, __p1) __extension__ ({ \
23433 int32x2_t __ret; \
2343323434 int64x2_t __s0 = __p0; \
2343423435 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
23435 int32x2_t __ret; \
2343623436 __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 2); \
2343723437 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2343823438 __ret; \
2343923439})
2344023440#define __noswap_vrshrn_n_s64(__p0, __p1) __extension__ ({ \
23441 int64x2_t __s0 = __p0; \
2344223441 int32x2_t __ret; \
23442 int64x2_t __s0 = __p0; \
2344323443 __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 2); \
2344423444 __ret; \
2344523445})
......@@ -23447,23 +23447,23 @@ __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
2344723447
2344823448#ifdef __LITTLE_ENDIAN__
2344923449#define vrshrn_n_s16(__p0, __p1) __extension__ ({ \
23450 int16x8_t __s0 = __p0; \
2345123450 int8x8_t __ret; \
23451 int16x8_t __s0 = __p0; \
2345223452 __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 0); \
2345323453 __ret; \
2345423454})
2345523455#else
2345623456#define vrshrn_n_s16(__p0, __p1) __extension__ ({ \
23457 int8x8_t __ret; \
2345723458 int16x8_t __s0 = __p0; \
2345823459 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
23459 int8x8_t __ret; \
2346023460 __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 0); \
2346123461 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2346223462 __ret; \
2346323463})
2346423464#define __noswap_vrshrn_n_s16(__p0, __p1) __extension__ ({ \
23465 int16x8_t __s0 = __p0; \
2346623465 int8x8_t __ret; \
23466 int16x8_t __s0 = __p0; \
2346723467 __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 0); \
2346823468 __ret; \
2346923469})
......@@ -23477,8 +23477,8 @@ __ai uint32x4_t vrsqrteq_u32(uint32x4_t __p0) {
2347723477}
2347823478#else
2347923479__ai uint32x4_t vrsqrteq_u32(uint32x4_t __p0) {
23480 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2348123480 uint32x4_t __ret;
23481 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2348223482 __ret = (uint32x4_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 50);
2348323483 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2348423484 return __ret;
......@@ -23493,8 +23493,8 @@ __ai float32x4_t vrsqrteq_f32(float32x4_t __p0) {
2349323493}
2349423494#else
2349523495__ai float32x4_t vrsqrteq_f32(float32x4_t __p0) {
23496 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2349723496 float32x4_t __ret;
23497 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2349823498 __ret = (float32x4_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 41);
2349923499 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2350023500 return __ret;
......@@ -23509,8 +23509,8 @@ __ai uint32x2_t vrsqrte_u32(uint32x2_t __p0) {
2350923509}
2351023510#else
2351123511__ai uint32x2_t vrsqrte_u32(uint32x2_t __p0) {
23512 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2351323512 uint32x2_t __ret;
23513 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2351423514 __ret = (uint32x2_t) __builtin_neon_vrsqrte_v((int8x8_t)__rev0, 18);
2351523515 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2351623516 return __ret;
......@@ -23525,8 +23525,8 @@ __ai float32x2_t vrsqrte_f32(float32x2_t __p0) {
2352523525}
2352623526#else
2352723527__ai float32x2_t vrsqrte_f32(float32x2_t __p0) {
23528 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2352923528 float32x2_t __ret;
23529 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2353023530 __ret = (float32x2_t) __builtin_neon_vrsqrte_v((int8x8_t)__rev0, 9);
2353123531 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2353223532 return __ret;
......@@ -23541,9 +23541,9 @@ __ai float32x4_t vrsqrtsq_f32(float32x4_t __p0, float32x4_t __p1) {
2354123541}
2354223542#else
2354323543__ai float32x4_t vrsqrtsq_f32(float32x4_t __p0, float32x4_t __p1) {
23544 float32x4_t __ret;
2354423545 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2354523546 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
23546 float32x4_t __ret;
2354723547 __ret = (float32x4_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
2354823548 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2354923549 return __ret;
......@@ -23558,9 +23558,9 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2355823558}
2355923559#else
2356023560__ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
23561 float32x2_t __ret;
2356123562 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2356223563 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
23563 float32x2_t __ret;
2356423564 __ret = (float32x2_t) __builtin_neon_vrsqrts_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
2356523565 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2356623566 return __ret;
......@@ -23569,19 +23569,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2356923569
2357023570#ifdef __LITTLE_ENDIAN__
2357123571#define vrsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
23572 uint8x16_t __ret; \
2357223573 uint8x16_t __s0 = __p0; \
2357323574 uint8x16_t __s1 = __p1; \
23574 uint8x16_t __ret; \
2357523575 __ret = (uint8x16_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 48); \
2357623576 __ret; \
2357723577})
2357823578#else
2357923579#define vrsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
23580 uint8x16_t __ret; \
2358023581 uint8x16_t __s0 = __p0; \
2358123582 uint8x16_t __s1 = __p1; \
2358223583 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2358323584 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
23584 uint8x16_t __ret; \
2358523585 __ret = (uint8x16_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 48); \
2358623586 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2358723587 __ret; \
......@@ -23590,19 +23590,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2359023590
2359123591#ifdef __LITTLE_ENDIAN__
2359223592#define vrsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
23593 uint32x4_t __ret; \
2359323594 uint32x4_t __s0 = __p0; \
2359423595 uint32x4_t __s1 = __p1; \
23595 uint32x4_t __ret; \
2359623596 __ret = (uint32x4_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 50); \
2359723597 __ret; \
2359823598})
2359923599#else
2360023600#define vrsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
23601 uint32x4_t __ret; \
2360123602 uint32x4_t __s0 = __p0; \
2360223603 uint32x4_t __s1 = __p1; \
2360323604 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2360423605 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
23605 uint32x4_t __ret; \
2360623606 __ret = (uint32x4_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 50); \
2360723607 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2360823608 __ret; \
......@@ -23611,19 +23611,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2361123611
2361223612#ifdef __LITTLE_ENDIAN__
2361323613#define vrsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
23614 uint64x2_t __ret; \
2361423615 uint64x2_t __s0 = __p0; \
2361523616 uint64x2_t __s1 = __p1; \
23616 uint64x2_t __ret; \
2361723617 __ret = (uint64x2_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
2361823618 __ret; \
2361923619})
2362023620#else
2362123621#define vrsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
23622 uint64x2_t __ret; \
2362223623 uint64x2_t __s0 = __p0; \
2362323624 uint64x2_t __s1 = __p1; \
2362423625 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2362523626 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
23626 uint64x2_t __ret; \
2362723627 __ret = (uint64x2_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
2362823628 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2362923629 __ret; \
......@@ -23632,19 +23632,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2363223632
2363323633#ifdef __LITTLE_ENDIAN__
2363423634#define vrsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
23635 uint16x8_t __ret; \
2363523636 uint16x8_t __s0 = __p0; \
2363623637 uint16x8_t __s1 = __p1; \
23637 uint16x8_t __ret; \
2363823638 __ret = (uint16x8_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 49); \
2363923639 __ret; \
2364023640})
2364123641#else
2364223642#define vrsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
23643 uint16x8_t __ret; \
2364323644 uint16x8_t __s0 = __p0; \
2364423645 uint16x8_t __s1 = __p1; \
2364523646 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2364623647 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
23647 uint16x8_t __ret; \
2364823648 __ret = (uint16x8_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 49); \
2364923649 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2365023650 __ret; \
......@@ -23653,19 +23653,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2365323653
2365423654#ifdef __LITTLE_ENDIAN__
2365523655#define vrsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
23656 int8x16_t __ret; \
2365623657 int8x16_t __s0 = __p0; \
2365723658 int8x16_t __s1 = __p1; \
23658 int8x16_t __ret; \
2365923659 __ret = (int8x16_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 32); \
2366023660 __ret; \
2366123661})
2366223662#else
2366323663#define vrsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
23664 int8x16_t __ret; \
2366423665 int8x16_t __s0 = __p0; \
2366523666 int8x16_t __s1 = __p1; \
2366623667 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2366723668 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
23668 int8x16_t __ret; \
2366923669 __ret = (int8x16_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 32); \
2367023670 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2367123671 __ret; \
......@@ -23674,19 +23674,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2367423674
2367523675#ifdef __LITTLE_ENDIAN__
2367623676#define vrsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
23677 int32x4_t __ret; \
2367723678 int32x4_t __s0 = __p0; \
2367823679 int32x4_t __s1 = __p1; \
23679 int32x4_t __ret; \
2368023680 __ret = (int32x4_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
2368123681 __ret; \
2368223682})
2368323683#else
2368423684#define vrsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
23685 int32x4_t __ret; \
2368523686 int32x4_t __s0 = __p0; \
2368623687 int32x4_t __s1 = __p1; \
2368723688 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2368823689 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
23689 int32x4_t __ret; \
2369023690 __ret = (int32x4_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
2369123691 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2369223692 __ret; \
......@@ -23695,19 +23695,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2369523695
2369623696#ifdef __LITTLE_ENDIAN__
2369723697#define vrsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
23698 int64x2_t __ret; \
2369823699 int64x2_t __s0 = __p0; \
2369923700 int64x2_t __s1 = __p1; \
23700 int64x2_t __ret; \
2370123701 __ret = (int64x2_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 35); \
2370223702 __ret; \
2370323703})
2370423704#else
2370523705#define vrsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
23706 int64x2_t __ret; \
2370623707 int64x2_t __s0 = __p0; \
2370723708 int64x2_t __s1 = __p1; \
2370823709 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2370923710 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
23710 int64x2_t __ret; \
2371123711 __ret = (int64x2_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 35); \
2371223712 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2371323713 __ret; \
......@@ -23716,19 +23716,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2371623716
2371723717#ifdef __LITTLE_ENDIAN__
2371823718#define vrsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
23719 int16x8_t __ret; \
2371923720 int16x8_t __s0 = __p0; \
2372023721 int16x8_t __s1 = __p1; \
23721 int16x8_t __ret; \
2372223722 __ret = (int16x8_t) __builtin_neon_vrsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
2372323723 __ret; \
2372423724})
2372523725#else
2372623726#define vrsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
23727 int16x8_t __ret; \
2372723728 int16x8_t __s0 = __p0; \
2372823729 int16x8_t __s1 = __p1; \
2372923730 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2373023731 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
23731 int16x8_t __ret; \
2373223732 __ret = (int16x8_t) __builtin_neon_vrsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
2373323733 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2373423734 __ret; \
......@@ -23737,19 +23737,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2373723737
2373823738#ifdef __LITTLE_ENDIAN__
2373923739#define vrsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
23740 uint8x8_t __ret; \
2374023741 uint8x8_t __s0 = __p0; \
2374123742 uint8x8_t __s1 = __p1; \
23742 uint8x8_t __ret; \
2374323743 __ret = (uint8x8_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 16); \
2374423744 __ret; \
2374523745})
2374623746#else
2374723747#define vrsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
23748 uint8x8_t __ret; \
2374823749 uint8x8_t __s0 = __p0; \
2374923750 uint8x8_t __s1 = __p1; \
2375023751 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2375123752 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
23752 uint8x8_t __ret; \
2375323753 __ret = (uint8x8_t) __builtin_neon_vrsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 16); \
2375423754 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2375523755 __ret; \
......@@ -23758,19 +23758,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2375823758
2375923759#ifdef __LITTLE_ENDIAN__
2376023760#define vrsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
23761 uint32x2_t __ret; \
2376123762 uint32x2_t __s0 = __p0; \
2376223763 uint32x2_t __s1 = __p1; \
23763 uint32x2_t __ret; \
2376423764 __ret = (uint32x2_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 18); \
2376523765 __ret; \
2376623766})
2376723767#else
2376823768#define vrsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
23769 uint32x2_t __ret; \
2376923770 uint32x2_t __s0 = __p0; \
2377023771 uint32x2_t __s1 = __p1; \
2377123772 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2377223773 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
23773 uint32x2_t __ret; \
2377423774 __ret = (uint32x2_t) __builtin_neon_vrsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 18); \
2377523775 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2377623776 __ret; \
......@@ -23778,27 +23778,27 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2377823778#endif
2377923779
2378023780#define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
23781 uint64x1_t __ret; \
2378123782 uint64x1_t __s0 = __p0; \
2378223783 uint64x1_t __s1 = __p1; \
23783 uint64x1_t __ret; \
2378423784 __ret = (uint64x1_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
2378523785 __ret; \
2378623786})
2378723787#ifdef __LITTLE_ENDIAN__
2378823788#define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
23789 uint16x4_t __ret; \
2378923790 uint16x4_t __s0 = __p0; \
2379023791 uint16x4_t __s1 = __p1; \
23791 uint16x4_t __ret; \
2379223792 __ret = (uint16x4_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 17); \
2379323793 __ret; \
2379423794})
2379523795#else
2379623796#define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
23797 uint16x4_t __ret; \
2379723798 uint16x4_t __s0 = __p0; \
2379823799 uint16x4_t __s1 = __p1; \
2379923800 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2380023801 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
23801 uint16x4_t __ret; \
2380223802 __ret = (uint16x4_t) __builtin_neon_vrsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 17); \
2380323803 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2380423804 __ret; \
......@@ -23807,19 +23807,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2380723807
2380823808#ifdef __LITTLE_ENDIAN__
2380923809#define vrsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
23810 int8x8_t __ret; \
2381023811 int8x8_t __s0 = __p0; \
2381123812 int8x8_t __s1 = __p1; \
23812 int8x8_t __ret; \
2381323813 __ret = (int8x8_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 0); \
2381423814 __ret; \
2381523815})
2381623816#else
2381723817#define vrsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
23818 int8x8_t __ret; \
2381823819 int8x8_t __s0 = __p0; \
2381923820 int8x8_t __s1 = __p1; \
2382023821 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2382123822 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
23822 int8x8_t __ret; \
2382323823 __ret = (int8x8_t) __builtin_neon_vrsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 0); \
2382423824 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2382523825 __ret; \
......@@ -23828,19 +23828,19 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2382823828
2382923829#ifdef __LITTLE_ENDIAN__
2383023830#define vrsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
23831 int32x2_t __ret; \
2383123832 int32x2_t __s0 = __p0; \
2383223833 int32x2_t __s1 = __p1; \
23833 int32x2_t __ret; \
2383423834 __ret = (int32x2_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
2383523835 __ret; \
2383623836})
2383723837#else
2383823838#define vrsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
23839 int32x2_t __ret; \
2383923840 int32x2_t __s0 = __p0; \
2384023841 int32x2_t __s1 = __p1; \
2384123842 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2384223843 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
23843 int32x2_t __ret; \
2384423844 __ret = (int32x2_t) __builtin_neon_vrsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
2384523845 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2384623846 __ret; \
......@@ -23848,27 +23848,27 @@ __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
2384823848#endif
2384923849
2385023850#define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
23851 int64x1_t __ret; \
2385123852 int64x1_t __s0 = __p0; \
2385223853 int64x1_t __s1 = __p1; \
23853 int64x1_t __ret; \
2385423854 __ret = (int64x1_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
2385523855 __ret; \
2385623856})
2385723857#ifdef __LITTLE_ENDIAN__
2385823858#define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
23859 int16x4_t __ret; \
2385923860 int16x4_t __s0 = __p0; \
2386023861 int16x4_t __s1 = __p1; \
23861 int16x4_t __ret; \
2386223862 __ret = (int16x4_t) __builtin_neon_vrsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
2386323863 __ret; \
2386423864})
2386523865#else
2386623866#define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
23867 int16x4_t __ret; \
2386723868 int16x4_t __s0 = __p0; \
2386823869 int16x4_t __s1 = __p1; \
2386923870 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2387023871 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
23871 int16x4_t __ret; \
2387223872 __ret = (int16x4_t) __builtin_neon_vrsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
2387323873 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2387423874 __ret; \
......@@ -23883,9 +23883,9 @@ __ai uint16x4_t vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
2388323883}
2388423884#else
2388523885__ai uint16x4_t vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
23886 uint16x4_t __ret;
2388623887 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2388723888 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
23888 uint16x4_t __ret;
2388923889 __ret = (uint16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 17);
2389023890 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2389123891 return __ret;
......@@ -23905,9 +23905,9 @@ __ai uint32x2_t vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
2390523905}
2390623906#else
2390723907__ai uint32x2_t vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
23908 uint32x2_t __ret;
2390823909 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2390923910 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
23910 uint32x2_t __ret;
2391123911 __ret = (uint32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 18);
2391223912 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2391323913 return __ret;
......@@ -23927,9 +23927,9 @@ __ai uint8x8_t vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
2392723927}
2392823928#else
2392923929__ai uint8x8_t vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
23930 uint8x8_t __ret;
2393023931 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2393123932 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
23932 uint8x8_t __ret;
2393323933 __ret = (uint8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 16);
2393423934 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2393523935 return __ret;
......@@ -23949,9 +23949,9 @@ __ai int16x4_t vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
2394923949}
2395023950#else
2395123951__ai int16x4_t vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
23952 int16x4_t __ret;
2395223953 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2395323954 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
23954 int16x4_t __ret;
2395523955 __ret = (int16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 1);
2395623956 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2395723957 return __ret;
......@@ -23971,9 +23971,9 @@ __ai int32x2_t vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
2397123971}
2397223972#else
2397323973__ai int32x2_t vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
23974 int32x2_t __ret;
2397423975 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2397523976 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
23976 int32x2_t __ret;
2397723977 __ret = (int32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 2);
2397823978 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2397923979 return __ret;
......@@ -23993,9 +23993,9 @@ __ai int8x8_t vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2399323993}
2399423994#else
2399523995__ai int8x8_t vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
23996 int8x8_t __ret;
2399623997 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2399723998 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
23998 int8x8_t __ret;
2399923999 __ret = (int8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 0);
2400024000 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2400124001 return __ret;
......@@ -24009,26 +24009,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2400924009
2401024010#ifdef __LITTLE_ENDIAN__
2401124011#define vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
24012 poly8x8_t __ret; \
2401224013 poly8_t __s0 = __p0; \
2401324014 poly8x8_t __s1 = __p1; \
24014 poly8x8_t __ret; \
2401524015 __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (poly8x8_t)__s1, __p2); \
2401624016 __ret; \
2401724017})
2401824018#else
2401924019#define vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
24020 poly8x8_t __ret; \
2402024021 poly8_t __s0 = __p0; \
2402124022 poly8x8_t __s1 = __p1; \
2402224023 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
24023 poly8x8_t __ret; \
2402424024 __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (poly8x8_t)__rev1, __p2); \
2402524025 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2402624026 __ret; \
2402724027})
2402824028#define __noswap_vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
24029 poly8x8_t __ret; \
2402924030 poly8_t __s0 = __p0; \
2403024031 poly8x8_t __s1 = __p1; \
24031 poly8x8_t __ret; \
2403224032 __ret = (poly8x8_t) __builtin_neon_vset_lane_i8(__s0, (poly8x8_t)__s1, __p2); \
2403324033 __ret; \
2403424034})
......@@ -24036,26 +24036,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2403624036
2403724037#ifdef __LITTLE_ENDIAN__
2403824038#define vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
24039 poly16x4_t __ret; \
2403924040 poly16_t __s0 = __p0; \
2404024041 poly16x4_t __s1 = __p1; \
24041 poly16x4_t __ret; \
2404224042 __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (poly16x4_t)__s1, __p2); \
2404324043 __ret; \
2404424044})
2404524045#else
2404624046#define vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
24047 poly16x4_t __ret; \
2404724048 poly16_t __s0 = __p0; \
2404824049 poly16x4_t __s1 = __p1; \
2404924050 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
24050 poly16x4_t __ret; \
2405124051 __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (poly16x4_t)__rev1, __p2); \
2405224052 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2405324053 __ret; \
2405424054})
2405524055#define __noswap_vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
24056 poly16x4_t __ret; \
2405624057 poly16_t __s0 = __p0; \
2405724058 poly16x4_t __s1 = __p1; \
24058 poly16x4_t __ret; \
2405924059 __ret = (poly16x4_t) __builtin_neon_vset_lane_i16(__s0, (poly16x4_t)__s1, __p2); \
2406024060 __ret; \
2406124061})
......@@ -24063,26 +24063,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2406324063
2406424064#ifdef __LITTLE_ENDIAN__
2406524065#define vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
24066 poly8x16_t __ret; \
2406624067 poly8_t __s0 = __p0; \
2406724068 poly8x16_t __s1 = __p1; \
24068 poly8x16_t __ret; \
2406924069 __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (poly8x16_t)__s1, __p2); \
2407024070 __ret; \
2407124071})
2407224072#else
2407324073#define vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
24074 poly8x16_t __ret; \
2407424075 poly8_t __s0 = __p0; \
2407524076 poly8x16_t __s1 = __p1; \
2407624077 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
24077 poly8x16_t __ret; \
2407824078 __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (poly8x16_t)__rev1, __p2); \
2407924079 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2408024080 __ret; \
2408124081})
2408224082#define __noswap_vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
24083 poly8x16_t __ret; \
2408324084 poly8_t __s0 = __p0; \
2408424085 poly8x16_t __s1 = __p1; \
24085 poly8x16_t __ret; \
2408624086 __ret = (poly8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (poly8x16_t)__s1, __p2); \
2408724087 __ret; \
2408824088})
......@@ -24090,26 +24090,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2409024090
2409124091#ifdef __LITTLE_ENDIAN__
2409224092#define vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
24093 poly16x8_t __ret; \
2409324094 poly16_t __s0 = __p0; \
2409424095 poly16x8_t __s1 = __p1; \
24095 poly16x8_t __ret; \
2409624096 __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (poly16x8_t)__s1, __p2); \
2409724097 __ret; \
2409824098})
2409924099#else
2410024100#define vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
24101 poly16x8_t __ret; \
2410124102 poly16_t __s0 = __p0; \
2410224103 poly16x8_t __s1 = __p1; \
2410324104 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
24104 poly16x8_t __ret; \
2410524105 __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (poly16x8_t)__rev1, __p2); \
2410624106 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2410724107 __ret; \
2410824108})
2410924109#define __noswap_vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
24110 poly16x8_t __ret; \
2411024111 poly16_t __s0 = __p0; \
2411124112 poly16x8_t __s1 = __p1; \
24112 poly16x8_t __ret; \
2411324113 __ret = (poly16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (poly16x8_t)__s1, __p2); \
2411424114 __ret; \
2411524115})
......@@ -24117,26 +24117,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2411724117
2411824118#ifdef __LITTLE_ENDIAN__
2411924119#define vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
24120 uint8x16_t __ret; \
2412024121 uint8_t __s0 = __p0; \
2412124122 uint8x16_t __s1 = __p1; \
24122 uint8x16_t __ret; \
2412324123 __ret = (uint8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__s1, __p2); \
2412424124 __ret; \
2412524125})
2412624126#else
2412724127#define vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
24128 uint8x16_t __ret; \
2412824129 uint8_t __s0 = __p0; \
2412924130 uint8x16_t __s1 = __p1; \
2413024131 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
24131 uint8x16_t __ret; \
2413224132 __ret = (uint8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__rev1, __p2); \
2413324133 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2413424134 __ret; \
2413524135})
2413624136#define __noswap_vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
24137 uint8x16_t __ret; \
2413724138 uint8_t __s0 = __p0; \
2413824139 uint8x16_t __s1 = __p1; \
24139 uint8x16_t __ret; \
2414024140 __ret = (uint8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__s1, __p2); \
2414124141 __ret; \
2414224142})
......@@ -24144,26 +24144,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2414424144
2414524145#ifdef __LITTLE_ENDIAN__
2414624146#define vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
24147 uint32x4_t __ret; \
2414724148 uint32_t __s0 = __p0; \
2414824149 uint32x4_t __s1 = __p1; \
24149 uint32x4_t __ret; \
2415024150 __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
2415124151 __ret; \
2415224152})
2415324153#else
2415424154#define vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
24155 uint32x4_t __ret; \
2415524156 uint32_t __s0 = __p0; \
2415624157 uint32x4_t __s1 = __p1; \
2415724158 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
24158 uint32x4_t __ret; \
2415924159 __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__rev1, __p2); \
2416024160 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2416124161 __ret; \
2416224162})
2416324163#define __noswap_vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
24164 uint32x4_t __ret; \
2416424165 uint32_t __s0 = __p0; \
2416524166 uint32x4_t __s1 = __p1; \
24166 uint32x4_t __ret; \
2416724167 __ret = (uint32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
2416824168 __ret; \
2416924169})
......@@ -24171,26 +24171,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2417124171
2417224172#ifdef __LITTLE_ENDIAN__
2417324173#define vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
24174 uint64x2_t __ret; \
2417424175 uint64_t __s0 = __p0; \
2417524176 uint64x2_t __s1 = __p1; \
24176 uint64x2_t __ret; \
2417724177 __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
2417824178 __ret; \
2417924179})
2418024180#else
2418124181#define vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
24182 uint64x2_t __ret; \
2418224183 uint64_t __s0 = __p0; \
2418324184 uint64x2_t __s1 = __p1; \
2418424185 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
24185 uint64x2_t __ret; \
2418624186 __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__rev1, __p2); \
2418724187 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2418824188 __ret; \
2418924189})
2419024190#define __noswap_vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
24191 uint64x2_t __ret; \
2419124192 uint64_t __s0 = __p0; \
2419224193 uint64x2_t __s1 = __p1; \
24193 uint64x2_t __ret; \
2419424194 __ret = (uint64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
2419524195 __ret; \
2419624196})
......@@ -24198,26 +24198,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2419824198
2419924199#ifdef __LITTLE_ENDIAN__
2420024200#define vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
24201 uint16x8_t __ret; \
2420124202 uint16_t __s0 = __p0; \
2420224203 uint16x8_t __s1 = __p1; \
24203 uint16x8_t __ret; \
2420424204 __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
2420524205 __ret; \
2420624206})
2420724207#else
2420824208#define vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
24209 uint16x8_t __ret; \
2420924210 uint16_t __s0 = __p0; \
2421024211 uint16x8_t __s1 = __p1; \
2421124212 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
24212 uint16x8_t __ret; \
2421324213 __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__rev1, __p2); \
2421424214 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2421524215 __ret; \
2421624216})
2421724217#define __noswap_vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
24218 uint16x8_t __ret; \
2421824219 uint16_t __s0 = __p0; \
2421924220 uint16x8_t __s1 = __p1; \
24220 uint16x8_t __ret; \
2422124221 __ret = (uint16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
2422224222 __ret; \
2422324223})
......@@ -24225,26 +24225,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2422524225
2422624226#ifdef __LITTLE_ENDIAN__
2422724227#define vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
24228 int8x16_t __ret; \
2422824229 int8_t __s0 = __p0; \
2422924230 int8x16_t __s1 = __p1; \
24230 int8x16_t __ret; \
2423124231 __ret = (int8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__s1, __p2); \
2423224232 __ret; \
2423324233})
2423424234#else
2423524235#define vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
24236 int8x16_t __ret; \
2423624237 int8_t __s0 = __p0; \
2423724238 int8x16_t __s1 = __p1; \
2423824239 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
24239 int8x16_t __ret; \
2424024240 __ret = (int8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__rev1, __p2); \
2424124241 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2424224242 __ret; \
2424324243})
2424424244#define __noswap_vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
24245 int8x16_t __ret; \
2424524246 int8_t __s0 = __p0; \
2424624247 int8x16_t __s1 = __p1; \
24247 int8x16_t __ret; \
2424824248 __ret = (int8x16_t) __builtin_neon_vsetq_lane_i8(__s0, (int8x16_t)__s1, __p2); \
2424924249 __ret; \
2425024250})
......@@ -24252,26 +24252,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2425224252
2425324253#ifdef __LITTLE_ENDIAN__
2425424254#define vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
24255 float32x4_t __ret; \
2425524256 float32_t __s0 = __p0; \
2425624257 float32x4_t __s1 = __p1; \
24257 float32x4_t __ret; \
2425824258 __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (float32x4_t)__s1, __p2); \
2425924259 __ret; \
2426024260})
2426124261#else
2426224262#define vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
24263 float32x4_t __ret; \
2426324264 float32_t __s0 = __p0; \
2426424265 float32x4_t __s1 = __p1; \
2426524266 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
24266 float32x4_t __ret; \
2426724267 __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (float32x4_t)__rev1, __p2); \
2426824268 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2426924269 __ret; \
2427024270})
2427124271#define __noswap_vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
24272 float32x4_t __ret; \
2427224273 float32_t __s0 = __p0; \
2427324274 float32x4_t __s1 = __p1; \
24274 float32x4_t __ret; \
2427524275 __ret = (float32x4_t) __builtin_neon_vsetq_lane_f32(__s0, (float32x4_t)__s1, __p2); \
2427624276 __ret; \
2427724277})
......@@ -24279,26 +24279,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2427924279
2428024280#ifdef __LITTLE_ENDIAN__
2428124281#define vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
24282 int32x4_t __ret; \
2428224283 int32_t __s0 = __p0; \
2428324284 int32x4_t __s1 = __p1; \
24284 int32x4_t __ret; \
2428524285 __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
2428624286 __ret; \
2428724287})
2428824288#else
2428924289#define vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
24290 int32x4_t __ret; \
2429024291 int32_t __s0 = __p0; \
2429124292 int32x4_t __s1 = __p1; \
2429224293 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
24293 int32x4_t __ret; \
2429424294 __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__rev1, __p2); \
2429524295 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2429624296 __ret; \
2429724297})
2429824298#define __noswap_vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
24299 int32x4_t __ret; \
2429924300 int32_t __s0 = __p0; \
2430024301 int32x4_t __s1 = __p1; \
24301 int32x4_t __ret; \
2430224302 __ret = (int32x4_t) __builtin_neon_vsetq_lane_i32(__s0, (int32x4_t)__s1, __p2); \
2430324303 __ret; \
2430424304})
......@@ -24306,26 +24306,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2430624306
2430724307#ifdef __LITTLE_ENDIAN__
2430824308#define vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
24309 int64x2_t __ret; \
2430924310 int64_t __s0 = __p0; \
2431024311 int64x2_t __s1 = __p1; \
24311 int64x2_t __ret; \
2431224312 __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
2431324313 __ret; \
2431424314})
2431524315#else
2431624316#define vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
24317 int64x2_t __ret; \
2431724318 int64_t __s0 = __p0; \
2431824319 int64x2_t __s1 = __p1; \
2431924320 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
24320 int64x2_t __ret; \
2432124321 __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__rev1, __p2); \
2432224322 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2432324323 __ret; \
2432424324})
2432524325#define __noswap_vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
24326 int64x2_t __ret; \
2432624327 int64_t __s0 = __p0; \
2432724328 int64x2_t __s1 = __p1; \
24328 int64x2_t __ret; \
2432924329 __ret = (int64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (int64x2_t)__s1, __p2); \
2433024330 __ret; \
2433124331})
......@@ -24333,26 +24333,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2433324333
2433424334#ifdef __LITTLE_ENDIAN__
2433524335#define vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
24336 int16x8_t __ret; \
2433624337 int16_t __s0 = __p0; \
2433724338 int16x8_t __s1 = __p1; \
24338 int16x8_t __ret; \
2433924339 __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
2434024340 __ret; \
2434124341})
2434224342#else
2434324343#define vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
24344 int16x8_t __ret; \
2434424345 int16_t __s0 = __p0; \
2434524346 int16x8_t __s1 = __p1; \
2434624347 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
24347 int16x8_t __ret; \
2434824348 __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__rev1, __p2); \
2434924349 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2435024350 __ret; \
2435124351})
2435224352#define __noswap_vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
24353 int16x8_t __ret; \
2435324354 int16_t __s0 = __p0; \
2435424355 int16x8_t __s1 = __p1; \
24355 int16x8_t __ret; \
2435624356 __ret = (int16x8_t) __builtin_neon_vsetq_lane_i16(__s0, (int16x8_t)__s1, __p2); \
2435724357 __ret; \
2435824358})
......@@ -24360,26 +24360,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2436024360
2436124361#ifdef __LITTLE_ENDIAN__
2436224362#define vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
24363 uint8x8_t __ret; \
2436324364 uint8_t __s0 = __p0; \
2436424365 uint8x8_t __s1 = __p1; \
24365 uint8x8_t __ret; \
2436624366 __ret = (uint8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__s1, __p2); \
2436724367 __ret; \
2436824368})
2436924369#else
2437024370#define vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
24371 uint8x8_t __ret; \
2437124372 uint8_t __s0 = __p0; \
2437224373 uint8x8_t __s1 = __p1; \
2437324374 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
24374 uint8x8_t __ret; \
2437524375 __ret = (uint8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__rev1, __p2); \
2437624376 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2437724377 __ret; \
2437824378})
2437924379#define __noswap_vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
24380 uint8x8_t __ret; \
2438024381 uint8_t __s0 = __p0; \
2438124382 uint8x8_t __s1 = __p1; \
24382 uint8x8_t __ret; \
2438324383 __ret = (uint8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__s1, __p2); \
2438424384 __ret; \
2438524385})
......@@ -24387,60 +24387,60 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2438724387
2438824388#ifdef __LITTLE_ENDIAN__
2438924389#define vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
24390 uint32x2_t __ret; \
2439024391 uint32_t __s0 = __p0; \
2439124392 uint32x2_t __s1 = __p1; \
24392 uint32x2_t __ret; \
2439324393 __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
2439424394 __ret; \
2439524395})
2439624396#else
2439724397#define vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
24398 uint32x2_t __ret; \
2439824399 uint32_t __s0 = __p0; \
2439924400 uint32x2_t __s1 = __p1; \
2440024401 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
24401 uint32x2_t __ret; \
2440224402 __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__rev1, __p2); \
2440324403 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2440424404 __ret; \
2440524405})
2440624406#define __noswap_vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
24407 uint32x2_t __ret; \
2440724408 uint32_t __s0 = __p0; \
2440824409 uint32x2_t __s1 = __p1; \
24409 uint32x2_t __ret; \
2441024410 __ret = (uint32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
2441124411 __ret; \
2441224412})
2441324413#endif
2441424414
2441524415#define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
24416 uint64x1_t __ret; \
2441624417 uint64_t __s0 = __p0; \
2441724418 uint64x1_t __s1 = __p1; \
24418 uint64x1_t __ret; \
2441924419 __ret = (uint64x1_t) __builtin_neon_vset_lane_i64(__s0, (int64x1_t)__s1, __p2); \
2442024420 __ret; \
2442124421})
2442224422#ifdef __LITTLE_ENDIAN__
2442324423#define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
24424 uint16x4_t __ret; \
2442424425 uint16_t __s0 = __p0; \
2442524426 uint16x4_t __s1 = __p1; \
24426 uint16x4_t __ret; \
2442724427 __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
2442824428 __ret; \
2442924429})
2443024430#else
2443124431#define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
24432 uint16x4_t __ret; \
2443224433 uint16_t __s0 = __p0; \
2443324434 uint16x4_t __s1 = __p1; \
2443424435 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
24435 uint16x4_t __ret; \
2443624436 __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__rev1, __p2); \
2443724437 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2443824438 __ret; \
2443924439})
2444024440#define __noswap_vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
24441 uint16x4_t __ret; \
2444124442 uint16_t __s0 = __p0; \
2444224443 uint16x4_t __s1 = __p1; \
24443 uint16x4_t __ret; \
2444424444 __ret = (uint16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
2444524445 __ret; \
2444624446})
......@@ -24448,26 +24448,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2444824448
2444924449#ifdef __LITTLE_ENDIAN__
2445024450#define vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
24451 int8x8_t __ret; \
2445124452 int8_t __s0 = __p0; \
2445224453 int8x8_t __s1 = __p1; \
24453 int8x8_t __ret; \
2445424454 __ret = (int8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__s1, __p2); \
2445524455 __ret; \
2445624456})
2445724457#else
2445824458#define vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
24459 int8x8_t __ret; \
2445924460 int8_t __s0 = __p0; \
2446024461 int8x8_t __s1 = __p1; \
2446124462 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
24462 int8x8_t __ret; \
2446324463 __ret = (int8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__rev1, __p2); \
2446424464 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2446524465 __ret; \
2446624466})
2446724467#define __noswap_vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
24468 int8x8_t __ret; \
2446824469 int8_t __s0 = __p0; \
2446924470 int8x8_t __s1 = __p1; \
24470 int8x8_t __ret; \
2447124471 __ret = (int8x8_t) __builtin_neon_vset_lane_i8(__s0, (int8x8_t)__s1, __p2); \
2447224472 __ret; \
2447324473})
......@@ -24475,26 +24475,26 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2447524475
2447624476#ifdef __LITTLE_ENDIAN__
2447724477#define vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
24478 float32x2_t __ret; \
2447824479 float32_t __s0 = __p0; \
2447924480 float32x2_t __s1 = __p1; \
24480 float32x2_t __ret; \
2448124481 __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (float32x2_t)__s1, __p2); \
2448224482 __ret; \
2448324483})
2448424484#else
2448524485#define vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
24486 float32x2_t __ret; \
2448624487 float32_t __s0 = __p0; \
2448724488 float32x2_t __s1 = __p1; \
2448824489 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
24489 float32x2_t __ret; \
2449024490 __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (float32x2_t)__rev1, __p2); \
2449124491 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2449224492 __ret; \
2449324493})
2449424494#define __noswap_vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
24495 float32x2_t __ret; \
2449524496 float32_t __s0 = __p0; \
2449624497 float32x2_t __s1 = __p1; \
24497 float32x2_t __ret; \
2449824498 __ret = (float32x2_t) __builtin_neon_vset_lane_f32(__s0, (float32x2_t)__s1, __p2); \
2449924499 __ret; \
2450024500})
......@@ -24502,60 +24502,60 @@ __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
2450224502
2450324503#ifdef __LITTLE_ENDIAN__
2450424504#define vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
24505 int32x2_t __ret; \
2450524506 int32_t __s0 = __p0; \
2450624507 int32x2_t __s1 = __p1; \
24507 int32x2_t __ret; \
2450824508 __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
2450924509 __ret; \
2451024510})
2451124511#else
2451224512#define vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
24513 int32x2_t __ret; \
2451324514 int32_t __s0 = __p0; \
2451424515 int32x2_t __s1 = __p1; \
2451524516 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
24516 int32x2_t __ret; \
2451724517 __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__rev1, __p2); \
2451824518 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2451924519 __ret; \
2452024520})
2452124521#define __noswap_vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
24522 int32x2_t __ret; \
2452224523 int32_t __s0 = __p0; \
2452324524 int32x2_t __s1 = __p1; \
24524 int32x2_t __ret; \
2452524525 __ret = (int32x2_t) __builtin_neon_vset_lane_i32(__s0, (int32x2_t)__s1, __p2); \
2452624526 __ret; \
2452724527})
2452824528#endif
2452924529
2453024530#define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
24531 int64x1_t __ret; \
2453124532 int64_t __s0 = __p0; \
2453224533 int64x1_t __s1 = __p1; \
24533 int64x1_t __ret; \
2453424534 __ret = (int64x1_t) __builtin_neon_vset_lane_i64(__s0, (int64x1_t)__s1, __p2); \
2453524535 __ret; \
2453624536})
2453724537#ifdef __LITTLE_ENDIAN__
2453824538#define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
24539 int16x4_t __ret; \
2453924540 int16_t __s0 = __p0; \
2454024541 int16x4_t __s1 = __p1; \
24541 int16x4_t __ret; \
2454224542 __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
2454324543 __ret; \
2454424544})
2454524545#else
2454624546#define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
24547 int16x4_t __ret; \
2454724548 int16_t __s0 = __p0; \
2454824549 int16x4_t __s1 = __p1; \
2454924550 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
24550 int16x4_t __ret; \
2455124551 __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__rev1, __p2); \
2455224552 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2455324553 __ret; \
2455424554})
2455524555#define __noswap_vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
24556 int16x4_t __ret; \
2455624557 int16_t __s0 = __p0; \
2455724558 int16x4_t __s1 = __p1; \
24558 int16x4_t __ret; \
2455924559 __ret = (int16x4_t) __builtin_neon_vset_lane_i16(__s0, (int16x4_t)__s1, __p2); \
2456024560 __ret; \
2456124561})
......@@ -24569,9 +24569,9 @@ __ai uint8x16_t vshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
2456924569}
2457024570#else
2457124571__ai uint8x16_t vshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
24572 uint8x16_t __ret;
2457224573 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2457324574 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
24574 uint8x16_t __ret;
2457524575 __ret = (uint8x16_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
2457624576 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2457724577 return __ret;
......@@ -24586,9 +24586,9 @@ __ai uint32x4_t vshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
2458624586}
2458724587#else
2458824588__ai uint32x4_t vshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
24589 uint32x4_t __ret;
2458924590 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2459024591 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
24591 uint32x4_t __ret;
2459224592 __ret = (uint32x4_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
2459324593 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2459424594 return __ret;
......@@ -24603,9 +24603,9 @@ __ai uint64x2_t vshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
2460324603}
2460424604#else
2460524605__ai uint64x2_t vshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
24606 uint64x2_t __ret;
2460624607 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2460724608 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
24608 uint64x2_t __ret;
2460924609 __ret = (uint64x2_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
2461024610 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2461124611 return __ret;
......@@ -24620,9 +24620,9 @@ __ai uint16x8_t vshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
2462024620}
2462124621#else
2462224622__ai uint16x8_t vshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
24623 uint16x8_t __ret;
2462324624 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2462424625 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
24625 uint16x8_t __ret;
2462624626 __ret = (uint16x8_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
2462724627 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2462824628 return __ret;
......@@ -24637,9 +24637,9 @@ __ai int8x16_t vshlq_s8(int8x16_t __p0, int8x16_t __p1) {
2463724637}
2463824638#else
2463924639__ai int8x16_t vshlq_s8(int8x16_t __p0, int8x16_t __p1) {
24640 int8x16_t __ret;
2464024641 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2464124642 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
24642 int8x16_t __ret;
2464324643 __ret = (int8x16_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
2464424644 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2464524645 return __ret;
......@@ -24654,9 +24654,9 @@ __ai int32x4_t vshlq_s32(int32x4_t __p0, int32x4_t __p1) {
2465424654}
2465524655#else
2465624656__ai int32x4_t vshlq_s32(int32x4_t __p0, int32x4_t __p1) {
24657 int32x4_t __ret;
2465724658 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2465824659 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
24659 int32x4_t __ret;
2466024660 __ret = (int32x4_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
2466124661 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2466224662 return __ret;
......@@ -24671,9 +24671,9 @@ __ai int64x2_t vshlq_s64(int64x2_t __p0, int64x2_t __p1) {
2467124671}
2467224672#else
2467324673__ai int64x2_t vshlq_s64(int64x2_t __p0, int64x2_t __p1) {
24674 int64x2_t __ret;
2467424675 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2467524676 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
24676 int64x2_t __ret;
2467724677 __ret = (int64x2_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
2467824678 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2467924679 return __ret;
......@@ -24688,9 +24688,9 @@ __ai int16x8_t vshlq_s16(int16x8_t __p0, int16x8_t __p1) {
2468824688}
2468924689#else
2469024690__ai int16x8_t vshlq_s16(int16x8_t __p0, int16x8_t __p1) {
24691 int16x8_t __ret;
2469124692 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2469224693 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
24693 int16x8_t __ret;
2469424694 __ret = (int16x8_t) __builtin_neon_vshlq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
2469524695 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2469624696 return __ret;
......@@ -24705,9 +24705,9 @@ __ai uint8x8_t vshl_u8(uint8x8_t __p0, int8x8_t __p1) {
2470524705}
2470624706#else
2470724707__ai uint8x8_t vshl_u8(uint8x8_t __p0, int8x8_t __p1) {
24708 uint8x8_t __ret;
2470824709 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2470924710 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
24710 uint8x8_t __ret;
2471124711 __ret = (uint8x8_t) __builtin_neon_vshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
2471224712 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2471324713 return __ret;
......@@ -24722,9 +24722,9 @@ __ai uint32x2_t vshl_u32(uint32x2_t __p0, int32x2_t __p1) {
2472224722}
2472324723#else
2472424724__ai uint32x2_t vshl_u32(uint32x2_t __p0, int32x2_t __p1) {
24725 uint32x2_t __ret;
2472524726 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2472624727 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
24727 uint32x2_t __ret;
2472824728 __ret = (uint32x2_t) __builtin_neon_vshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
2472924729 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2473024730 return __ret;
......@@ -24744,9 +24744,9 @@ __ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
2474424744}
2474524745#else
2474624746__ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
24747 uint16x4_t __ret;
2474724748 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2474824749 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
24749 uint16x4_t __ret;
2475024750 __ret = (uint16x4_t) __builtin_neon_vshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
2475124751 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2475224752 return __ret;
......@@ -24761,9 +24761,9 @@ __ai int8x8_t vshl_s8(int8x8_t __p0, int8x8_t __p1) {
2476124761}
2476224762#else
2476324763__ai int8x8_t vshl_s8(int8x8_t __p0, int8x8_t __p1) {
24764 int8x8_t __ret;
2476424765 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2476524766 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
24766 int8x8_t __ret;
2476724767 __ret = (int8x8_t) __builtin_neon_vshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
2476824768 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2476924769 return __ret;
......@@ -24778,9 +24778,9 @@ __ai int32x2_t vshl_s32(int32x2_t __p0, int32x2_t __p1) {
2477824778}
2477924779#else
2478024780__ai int32x2_t vshl_s32(int32x2_t __p0, int32x2_t __p1) {
24781 int32x2_t __ret;
2478124782 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2478224783 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
24783 int32x2_t __ret;
2478424784 __ret = (int32x2_t) __builtin_neon_vshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
2478524785 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2478624786 return __ret;
......@@ -24800,9 +24800,9 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2480024800}
2480124801#else
2480224802__ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
24803 int16x4_t __ret;
2480324804 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2480424805 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
24805 int16x4_t __ret;
2480624806 __ret = (int16x4_t) __builtin_neon_vshl_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
2480724807 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2480824808 return __ret;
......@@ -24811,16 +24811,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2481124811
2481224812#ifdef __LITTLE_ENDIAN__
2481324813#define vshlq_n_u8(__p0, __p1) __extension__ ({ \
24814 uint8x16_t __s0 = __p0; \
2481524814 uint8x16_t __ret; \
24815 uint8x16_t __s0 = __p0; \
2481624816 __ret = (uint8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 48); \
2481724817 __ret; \
2481824818})
2481924819#else
2482024820#define vshlq_n_u8(__p0, __p1) __extension__ ({ \
24821 uint8x16_t __ret; \
2482124822 uint8x16_t __s0 = __p0; \
2482224823 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
24823 uint8x16_t __ret; \
2482424824 __ret = (uint8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 48); \
2482524825 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2482624826 __ret; \
......@@ -24829,16 +24829,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2482924829
2483024830#ifdef __LITTLE_ENDIAN__
2483124831#define vshlq_n_u32(__p0, __p1) __extension__ ({ \
24832 uint32x4_t __s0 = __p0; \
2483324832 uint32x4_t __ret; \
24833 uint32x4_t __s0 = __p0; \
2483424834 __ret = (uint32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 50); \
2483524835 __ret; \
2483624836})
2483724837#else
2483824838#define vshlq_n_u32(__p0, __p1) __extension__ ({ \
24839 uint32x4_t __ret; \
2483924840 uint32x4_t __s0 = __p0; \
2484024841 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
24841 uint32x4_t __ret; \
2484224842 __ret = (uint32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 50); \
2484324843 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2484424844 __ret; \
......@@ -24847,16 +24847,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2484724847
2484824848#ifdef __LITTLE_ENDIAN__
2484924849#define vshlq_n_u64(__p0, __p1) __extension__ ({ \
24850 uint64x2_t __s0 = __p0; \
2485124850 uint64x2_t __ret; \
24851 uint64x2_t __s0 = __p0; \
2485224852 __ret = (uint64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 51); \
2485324853 __ret; \
2485424854})
2485524855#else
2485624856#define vshlq_n_u64(__p0, __p1) __extension__ ({ \
24857 uint64x2_t __ret; \
2485724858 uint64x2_t __s0 = __p0; \
2485824859 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
24859 uint64x2_t __ret; \
2486024860 __ret = (uint64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 51); \
2486124861 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2486224862 __ret; \
......@@ -24865,16 +24865,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2486524865
2486624866#ifdef __LITTLE_ENDIAN__
2486724867#define vshlq_n_u16(__p0, __p1) __extension__ ({ \
24868 uint16x8_t __s0 = __p0; \
2486924868 uint16x8_t __ret; \
24869 uint16x8_t __s0 = __p0; \
2487024870 __ret = (uint16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 49); \
2487124871 __ret; \
2487224872})
2487324873#else
2487424874#define vshlq_n_u16(__p0, __p1) __extension__ ({ \
24875 uint16x8_t __ret; \
2487524876 uint16x8_t __s0 = __p0; \
2487624877 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
24877 uint16x8_t __ret; \
2487824878 __ret = (uint16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 49); \
2487924879 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2488024880 __ret; \
......@@ -24883,16 +24883,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2488324883
2488424884#ifdef __LITTLE_ENDIAN__
2488524885#define vshlq_n_s8(__p0, __p1) __extension__ ({ \
24886 int8x16_t __s0 = __p0; \
2488724886 int8x16_t __ret; \
24887 int8x16_t __s0 = __p0; \
2488824888 __ret = (int8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 32); \
2488924889 __ret; \
2489024890})
2489124891#else
2489224892#define vshlq_n_s8(__p0, __p1) __extension__ ({ \
24893 int8x16_t __ret; \
2489324894 int8x16_t __s0 = __p0; \
2489424895 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
24895 int8x16_t __ret; \
2489624896 __ret = (int8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 32); \
2489724897 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2489824898 __ret; \
......@@ -24901,16 +24901,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2490124901
2490224902#ifdef __LITTLE_ENDIAN__
2490324903#define vshlq_n_s32(__p0, __p1) __extension__ ({ \
24904 int32x4_t __s0 = __p0; \
2490524904 int32x4_t __ret; \
24905 int32x4_t __s0 = __p0; \
2490624906 __ret = (int32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 34); \
2490724907 __ret; \
2490824908})
2490924909#else
2491024910#define vshlq_n_s32(__p0, __p1) __extension__ ({ \
24911 int32x4_t __ret; \
2491124912 int32x4_t __s0 = __p0; \
2491224913 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
24913 int32x4_t __ret; \
2491424914 __ret = (int32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 34); \
2491524915 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2491624916 __ret; \
......@@ -24919,16 +24919,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2491924919
2492024920#ifdef __LITTLE_ENDIAN__
2492124921#define vshlq_n_s64(__p0, __p1) __extension__ ({ \
24922 int64x2_t __s0 = __p0; \
2492324922 int64x2_t __ret; \
24923 int64x2_t __s0 = __p0; \
2492424924 __ret = (int64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 35); \
2492524925 __ret; \
2492624926})
2492724927#else
2492824928#define vshlq_n_s64(__p0, __p1) __extension__ ({ \
24929 int64x2_t __ret; \
2492924930 int64x2_t __s0 = __p0; \
2493024931 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
24931 int64x2_t __ret; \
2493224932 __ret = (int64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 35); \
2493324933 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2493424934 __ret; \
......@@ -24937,16 +24937,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2493724937
2493824938#ifdef __LITTLE_ENDIAN__
2493924939#define vshlq_n_s16(__p0, __p1) __extension__ ({ \
24940 int16x8_t __s0 = __p0; \
2494124940 int16x8_t __ret; \
24941 int16x8_t __s0 = __p0; \
2494224942 __ret = (int16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 33); \
2494324943 __ret; \
2494424944})
2494524945#else
2494624946#define vshlq_n_s16(__p0, __p1) __extension__ ({ \
24947 int16x8_t __ret; \
2494724948 int16x8_t __s0 = __p0; \
2494824949 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
24949 int16x8_t __ret; \
2495024950 __ret = (int16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 33); \
2495124951 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2495224952 __ret; \
......@@ -24955,16 +24955,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2495524955
2495624956#ifdef __LITTLE_ENDIAN__
2495724957#define vshl_n_u8(__p0, __p1) __extension__ ({ \
24958 uint8x8_t __s0 = __p0; \
2495924958 uint8x8_t __ret; \
24959 uint8x8_t __s0 = __p0; \
2496024960 __ret = (uint8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 16); \
2496124961 __ret; \
2496224962})
2496324963#else
2496424964#define vshl_n_u8(__p0, __p1) __extension__ ({ \
24965 uint8x8_t __ret; \
2496524966 uint8x8_t __s0 = __p0; \
2496624967 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
24967 uint8x8_t __ret; \
2496824968 __ret = (uint8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 16); \
2496924969 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2497024970 __ret; \
......@@ -24973,16 +24973,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2497324973
2497424974#ifdef __LITTLE_ENDIAN__
2497524975#define vshl_n_u32(__p0, __p1) __extension__ ({ \
24976 uint32x2_t __s0 = __p0; \
2497724976 uint32x2_t __ret; \
24977 uint32x2_t __s0 = __p0; \
2497824978 __ret = (uint32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 18); \
2497924979 __ret; \
2498024980})
2498124981#else
2498224982#define vshl_n_u32(__p0, __p1) __extension__ ({ \
24983 uint32x2_t __ret; \
2498324984 uint32x2_t __s0 = __p0; \
2498424985 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
24985 uint32x2_t __ret; \
2498624986 __ret = (uint32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 18); \
2498724987 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2498824988 __ret; \
......@@ -24990,23 +24990,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2499024990#endif
2499124991
2499224992#define vshl_n_u64(__p0, __p1) __extension__ ({ \
24993 uint64x1_t __s0 = __p0; \
2499424993 uint64x1_t __ret; \
24994 uint64x1_t __s0 = __p0; \
2499524995 __ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
2499624996 __ret; \
2499724997})
2499824998#ifdef __LITTLE_ENDIAN__
2499924999#define vshl_n_u16(__p0, __p1) __extension__ ({ \
25000 uint16x4_t __s0 = __p0; \
2500125000 uint16x4_t __ret; \
25001 uint16x4_t __s0 = __p0; \
2500225002 __ret = (uint16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 17); \
2500325003 __ret; \
2500425004})
2500525005#else
2500625006#define vshl_n_u16(__p0, __p1) __extension__ ({ \
25007 uint16x4_t __ret; \
2500725008 uint16x4_t __s0 = __p0; \
2500825009 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25009 uint16x4_t __ret; \
2501025010 __ret = (uint16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 17); \
2501125011 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2501225012 __ret; \
......@@ -25015,16 +25015,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2501525015
2501625016#ifdef __LITTLE_ENDIAN__
2501725017#define vshl_n_s8(__p0, __p1) __extension__ ({ \
25018 int8x8_t __s0 = __p0; \
2501925018 int8x8_t __ret; \
25019 int8x8_t __s0 = __p0; \
2502025020 __ret = (int8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 0); \
2502125021 __ret; \
2502225022})
2502325023#else
2502425024#define vshl_n_s8(__p0, __p1) __extension__ ({ \
25025 int8x8_t __ret; \
2502525026 int8x8_t __s0 = __p0; \
2502625027 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25027 int8x8_t __ret; \
2502825028 __ret = (int8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 0); \
2502925029 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2503025030 __ret; \
......@@ -25033,16 +25033,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2503325033
2503425034#ifdef __LITTLE_ENDIAN__
2503525035#define vshl_n_s32(__p0, __p1) __extension__ ({ \
25036 int32x2_t __s0 = __p0; \
2503725036 int32x2_t __ret; \
25037 int32x2_t __s0 = __p0; \
2503825038 __ret = (int32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 2); \
2503925039 __ret; \
2504025040})
2504125041#else
2504225042#define vshl_n_s32(__p0, __p1) __extension__ ({ \
25043 int32x2_t __ret; \
2504325044 int32x2_t __s0 = __p0; \
2504425045 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25045 int32x2_t __ret; \
2504625046 __ret = (int32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 2); \
2504725047 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2504825048 __ret; \
......@@ -25050,23 +25050,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2505025050#endif
2505125051
2505225052#define vshl_n_s64(__p0, __p1) __extension__ ({ \
25053 int64x1_t __s0 = __p0; \
2505425053 int64x1_t __ret; \
25054 int64x1_t __s0 = __p0; \
2505525055 __ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
2505625056 __ret; \
2505725057})
2505825058#ifdef __LITTLE_ENDIAN__
2505925059#define vshl_n_s16(__p0, __p1) __extension__ ({ \
25060 int16x4_t __s0 = __p0; \
2506125060 int16x4_t __ret; \
25061 int16x4_t __s0 = __p0; \
2506225062 __ret = (int16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 1); \
2506325063 __ret; \
2506425064})
2506525065#else
2506625066#define vshl_n_s16(__p0, __p1) __extension__ ({ \
25067 int16x4_t __ret; \
2506725068 int16x4_t __s0 = __p0; \
2506825069 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25069 int16x4_t __ret; \
2507025070 __ret = (int16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 1); \
2507125071 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2507225072 __ret; \
......@@ -25075,23 +25075,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2507525075
2507625076#ifdef __LITTLE_ENDIAN__
2507725077#define vshll_n_u8(__p0, __p1) __extension__ ({ \
25078 uint8x8_t __s0 = __p0; \
2507925078 uint16x8_t __ret; \
25079 uint8x8_t __s0 = __p0; \
2508025080 __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 49); \
2508125081 __ret; \
2508225082})
2508325083#else
2508425084#define vshll_n_u8(__p0, __p1) __extension__ ({ \
25085 uint16x8_t __ret; \
2508525086 uint8x8_t __s0 = __p0; \
2508625087 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25087 uint16x8_t __ret; \
2508825088 __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 49); \
2508925089 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2509025090 __ret; \
2509125091})
2509225092#define __noswap_vshll_n_u8(__p0, __p1) __extension__ ({ \
25093 uint8x8_t __s0 = __p0; \
2509425093 uint16x8_t __ret; \
25094 uint8x8_t __s0 = __p0; \
2509525095 __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 49); \
2509625096 __ret; \
2509725097})
......@@ -25099,23 +25099,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2509925099
2510025100#ifdef __LITTLE_ENDIAN__
2510125101#define vshll_n_u32(__p0, __p1) __extension__ ({ \
25102 uint32x2_t __s0 = __p0; \
2510325102 uint64x2_t __ret; \
25103 uint32x2_t __s0 = __p0; \
2510425104 __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 51); \
2510525105 __ret; \
2510625106})
2510725107#else
2510825108#define vshll_n_u32(__p0, __p1) __extension__ ({ \
25109 uint64x2_t __ret; \
2510925110 uint32x2_t __s0 = __p0; \
2511025111 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25111 uint64x2_t __ret; \
2511225112 __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 51); \
2511325113 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2511425114 __ret; \
2511525115})
2511625116#define __noswap_vshll_n_u32(__p0, __p1) __extension__ ({ \
25117 uint32x2_t __s0 = __p0; \
2511825117 uint64x2_t __ret; \
25118 uint32x2_t __s0 = __p0; \
2511925119 __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 51); \
2512025120 __ret; \
2512125121})
......@@ -25123,23 +25123,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2512325123
2512425124#ifdef __LITTLE_ENDIAN__
2512525125#define vshll_n_u16(__p0, __p1) __extension__ ({ \
25126 uint16x4_t __s0 = __p0; \
2512725126 uint32x4_t __ret; \
25127 uint16x4_t __s0 = __p0; \
2512825128 __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 50); \
2512925129 __ret; \
2513025130})
2513125131#else
2513225132#define vshll_n_u16(__p0, __p1) __extension__ ({ \
25133 uint32x4_t __ret; \
2513325134 uint16x4_t __s0 = __p0; \
2513425135 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25135 uint32x4_t __ret; \
2513625136 __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 50); \
2513725137 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2513825138 __ret; \
2513925139})
2514025140#define __noswap_vshll_n_u16(__p0, __p1) __extension__ ({ \
25141 uint16x4_t __s0 = __p0; \
2514225141 uint32x4_t __ret; \
25142 uint16x4_t __s0 = __p0; \
2514325143 __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 50); \
2514425144 __ret; \
2514525145})
......@@ -25147,23 +25147,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2514725147
2514825148#ifdef __LITTLE_ENDIAN__
2514925149#define vshll_n_s8(__p0, __p1) __extension__ ({ \
25150 int8x8_t __s0 = __p0; \
2515125150 int16x8_t __ret; \
25151 int8x8_t __s0 = __p0; \
2515225152 __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 33); \
2515325153 __ret; \
2515425154})
2515525155#else
2515625156#define vshll_n_s8(__p0, __p1) __extension__ ({ \
25157 int16x8_t __ret; \
2515725158 int8x8_t __s0 = __p0; \
2515825159 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25159 int16x8_t __ret; \
2516025160 __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 33); \
2516125161 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2516225162 __ret; \
2516325163})
2516425164#define __noswap_vshll_n_s8(__p0, __p1) __extension__ ({ \
25165 int8x8_t __s0 = __p0; \
2516625165 int16x8_t __ret; \
25166 int8x8_t __s0 = __p0; \
2516725167 __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 33); \
2516825168 __ret; \
2516925169})
......@@ -25171,23 +25171,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2517125171
2517225172#ifdef __LITTLE_ENDIAN__
2517325173#define vshll_n_s32(__p0, __p1) __extension__ ({ \
25174 int32x2_t __s0 = __p0; \
2517525174 int64x2_t __ret; \
25175 int32x2_t __s0 = __p0; \
2517625176 __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 35); \
2517725177 __ret; \
2517825178})
2517925179#else
2518025180#define vshll_n_s32(__p0, __p1) __extension__ ({ \
25181 int64x2_t __ret; \
2518125182 int32x2_t __s0 = __p0; \
2518225183 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25183 int64x2_t __ret; \
2518425184 __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 35); \
2518525185 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2518625186 __ret; \
2518725187})
2518825188#define __noswap_vshll_n_s32(__p0, __p1) __extension__ ({ \
25189 int32x2_t __s0 = __p0; \
2519025189 int64x2_t __ret; \
25190 int32x2_t __s0 = __p0; \
2519125191 __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 35); \
2519225192 __ret; \
2519325193})
......@@ -25195,23 +25195,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2519525195
2519625196#ifdef __LITTLE_ENDIAN__
2519725197#define vshll_n_s16(__p0, __p1) __extension__ ({ \
25198 int16x4_t __s0 = __p0; \
2519925198 int32x4_t __ret; \
25199 int16x4_t __s0 = __p0; \
2520025200 __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 34); \
2520125201 __ret; \
2520225202})
2520325203#else
2520425204#define vshll_n_s16(__p0, __p1) __extension__ ({ \
25205 int32x4_t __ret; \
2520525206 int16x4_t __s0 = __p0; \
2520625207 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25207 int32x4_t __ret; \
2520825208 __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 34); \
2520925209 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2521025210 __ret; \
2521125211})
2521225212#define __noswap_vshll_n_s16(__p0, __p1) __extension__ ({ \
25213 int16x4_t __s0 = __p0; \
2521425213 int32x4_t __ret; \
25214 int16x4_t __s0 = __p0; \
2521525215 __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 34); \
2521625216 __ret; \
2521725217})
......@@ -25219,16 +25219,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2521925219
2522025220#ifdef __LITTLE_ENDIAN__
2522125221#define vshrq_n_u8(__p0, __p1) __extension__ ({ \
25222 uint8x16_t __s0 = __p0; \
2522325222 uint8x16_t __ret; \
25223 uint8x16_t __s0 = __p0; \
2522425224 __ret = (uint8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 48); \
2522525225 __ret; \
2522625226})
2522725227#else
2522825228#define vshrq_n_u8(__p0, __p1) __extension__ ({ \
25229 uint8x16_t __ret; \
2522925230 uint8x16_t __s0 = __p0; \
2523025231 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
25231 uint8x16_t __ret; \
2523225232 __ret = (uint8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 48); \
2523325233 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2523425234 __ret; \
......@@ -25237,16 +25237,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2523725237
2523825238#ifdef __LITTLE_ENDIAN__
2523925239#define vshrq_n_u32(__p0, __p1) __extension__ ({ \
25240 uint32x4_t __s0 = __p0; \
2524125240 uint32x4_t __ret; \
25241 uint32x4_t __s0 = __p0; \
2524225242 __ret = (uint32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 50); \
2524325243 __ret; \
2524425244})
2524525245#else
2524625246#define vshrq_n_u32(__p0, __p1) __extension__ ({ \
25247 uint32x4_t __ret; \
2524725248 uint32x4_t __s0 = __p0; \
2524825249 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25249 uint32x4_t __ret; \
2525025250 __ret = (uint32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 50); \
2525125251 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2525225252 __ret; \
......@@ -25255,16 +25255,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2525525255
2525625256#ifdef __LITTLE_ENDIAN__
2525725257#define vshrq_n_u64(__p0, __p1) __extension__ ({ \
25258 uint64x2_t __s0 = __p0; \
2525925258 uint64x2_t __ret; \
25259 uint64x2_t __s0 = __p0; \
2526025260 __ret = (uint64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 51); \
2526125261 __ret; \
2526225262})
2526325263#else
2526425264#define vshrq_n_u64(__p0, __p1) __extension__ ({ \
25265 uint64x2_t __ret; \
2526525266 uint64x2_t __s0 = __p0; \
2526625267 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25267 uint64x2_t __ret; \
2526825268 __ret = (uint64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 51); \
2526925269 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2527025270 __ret; \
......@@ -25273,16 +25273,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2527325273
2527425274#ifdef __LITTLE_ENDIAN__
2527525275#define vshrq_n_u16(__p0, __p1) __extension__ ({ \
25276 uint16x8_t __s0 = __p0; \
2527725276 uint16x8_t __ret; \
25277 uint16x8_t __s0 = __p0; \
2527825278 __ret = (uint16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 49); \
2527925279 __ret; \
2528025280})
2528125281#else
2528225282#define vshrq_n_u16(__p0, __p1) __extension__ ({ \
25283 uint16x8_t __ret; \
2528325284 uint16x8_t __s0 = __p0; \
2528425285 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25285 uint16x8_t __ret; \
2528625286 __ret = (uint16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 49); \
2528725287 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2528825288 __ret; \
......@@ -25291,16 +25291,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2529125291
2529225292#ifdef __LITTLE_ENDIAN__
2529325293#define vshrq_n_s8(__p0, __p1) __extension__ ({ \
25294 int8x16_t __s0 = __p0; \
2529525294 int8x16_t __ret; \
25295 int8x16_t __s0 = __p0; \
2529625296 __ret = (int8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 32); \
2529725297 __ret; \
2529825298})
2529925299#else
2530025300#define vshrq_n_s8(__p0, __p1) __extension__ ({ \
25301 int8x16_t __ret; \
2530125302 int8x16_t __s0 = __p0; \
2530225303 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
25303 int8x16_t __ret; \
2530425304 __ret = (int8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 32); \
2530525305 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2530625306 __ret; \
......@@ -25309,16 +25309,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2530925309
2531025310#ifdef __LITTLE_ENDIAN__
2531125311#define vshrq_n_s32(__p0, __p1) __extension__ ({ \
25312 int32x4_t __s0 = __p0; \
2531325312 int32x4_t __ret; \
25313 int32x4_t __s0 = __p0; \
2531425314 __ret = (int32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 34); \
2531525315 __ret; \
2531625316})
2531725317#else
2531825318#define vshrq_n_s32(__p0, __p1) __extension__ ({ \
25319 int32x4_t __ret; \
2531925320 int32x4_t __s0 = __p0; \
2532025321 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25321 int32x4_t __ret; \
2532225322 __ret = (int32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 34); \
2532325323 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2532425324 __ret; \
......@@ -25327,16 +25327,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2532725327
2532825328#ifdef __LITTLE_ENDIAN__
2532925329#define vshrq_n_s64(__p0, __p1) __extension__ ({ \
25330 int64x2_t __s0 = __p0; \
2533125330 int64x2_t __ret; \
25331 int64x2_t __s0 = __p0; \
2533225332 __ret = (int64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 35); \
2533325333 __ret; \
2533425334})
2533525335#else
2533625336#define vshrq_n_s64(__p0, __p1) __extension__ ({ \
25337 int64x2_t __ret; \
2533725338 int64x2_t __s0 = __p0; \
2533825339 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25339 int64x2_t __ret; \
2534025340 __ret = (int64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 35); \
2534125341 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2534225342 __ret; \
......@@ -25345,16 +25345,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2534525345
2534625346#ifdef __LITTLE_ENDIAN__
2534725347#define vshrq_n_s16(__p0, __p1) __extension__ ({ \
25348 int16x8_t __s0 = __p0; \
2534925348 int16x8_t __ret; \
25349 int16x8_t __s0 = __p0; \
2535025350 __ret = (int16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 33); \
2535125351 __ret; \
2535225352})
2535325353#else
2535425354#define vshrq_n_s16(__p0, __p1) __extension__ ({ \
25355 int16x8_t __ret; \
2535525356 int16x8_t __s0 = __p0; \
2535625357 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25357 int16x8_t __ret; \
2535825358 __ret = (int16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 33); \
2535925359 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2536025360 __ret; \
......@@ -25363,16 +25363,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2536325363
2536425364#ifdef __LITTLE_ENDIAN__
2536525365#define vshr_n_u8(__p0, __p1) __extension__ ({ \
25366 uint8x8_t __s0 = __p0; \
2536725366 uint8x8_t __ret; \
25367 uint8x8_t __s0 = __p0; \
2536825368 __ret = (uint8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 16); \
2536925369 __ret; \
2537025370})
2537125371#else
2537225372#define vshr_n_u8(__p0, __p1) __extension__ ({ \
25373 uint8x8_t __ret; \
2537325374 uint8x8_t __s0 = __p0; \
2537425375 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25375 uint8x8_t __ret; \
2537625376 __ret = (uint8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 16); \
2537725377 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2537825378 __ret; \
......@@ -25381,16 +25381,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2538125381
2538225382#ifdef __LITTLE_ENDIAN__
2538325383#define vshr_n_u32(__p0, __p1) __extension__ ({ \
25384 uint32x2_t __s0 = __p0; \
2538525384 uint32x2_t __ret; \
25385 uint32x2_t __s0 = __p0; \
2538625386 __ret = (uint32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 18); \
2538725387 __ret; \
2538825388})
2538925389#else
2539025390#define vshr_n_u32(__p0, __p1) __extension__ ({ \
25391 uint32x2_t __ret; \
2539125392 uint32x2_t __s0 = __p0; \
2539225393 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25393 uint32x2_t __ret; \
2539425394 __ret = (uint32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 18); \
2539525395 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2539625396 __ret; \
......@@ -25398,23 +25398,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2539825398#endif
2539925399
2540025400#define vshr_n_u64(__p0, __p1) __extension__ ({ \
25401 uint64x1_t __s0 = __p0; \
2540225401 uint64x1_t __ret; \
25402 uint64x1_t __s0 = __p0; \
2540325403 __ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
2540425404 __ret; \
2540525405})
2540625406#ifdef __LITTLE_ENDIAN__
2540725407#define vshr_n_u16(__p0, __p1) __extension__ ({ \
25408 uint16x4_t __s0 = __p0; \
2540925408 uint16x4_t __ret; \
25409 uint16x4_t __s0 = __p0; \
2541025410 __ret = (uint16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 17); \
2541125411 __ret; \
2541225412})
2541325413#else
2541425414#define vshr_n_u16(__p0, __p1) __extension__ ({ \
25415 uint16x4_t __ret; \
2541525416 uint16x4_t __s0 = __p0; \
2541625417 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25417 uint16x4_t __ret; \
2541825418 __ret = (uint16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 17); \
2541925419 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2542025420 __ret; \
......@@ -25423,16 +25423,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2542325423
2542425424#ifdef __LITTLE_ENDIAN__
2542525425#define vshr_n_s8(__p0, __p1) __extension__ ({ \
25426 int8x8_t __s0 = __p0; \
2542725426 int8x8_t __ret; \
25427 int8x8_t __s0 = __p0; \
2542825428 __ret = (int8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 0); \
2542925429 __ret; \
2543025430})
2543125431#else
2543225432#define vshr_n_s8(__p0, __p1) __extension__ ({ \
25433 int8x8_t __ret; \
2543325434 int8x8_t __s0 = __p0; \
2543425435 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25435 int8x8_t __ret; \
2543625436 __ret = (int8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 0); \
2543725437 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2543825438 __ret; \
......@@ -25441,16 +25441,16 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2544125441
2544225442#ifdef __LITTLE_ENDIAN__
2544325443#define vshr_n_s32(__p0, __p1) __extension__ ({ \
25444 int32x2_t __s0 = __p0; \
2544525444 int32x2_t __ret; \
25445 int32x2_t __s0 = __p0; \
2544625446 __ret = (int32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 2); \
2544725447 __ret; \
2544825448})
2544925449#else
2545025450#define vshr_n_s32(__p0, __p1) __extension__ ({ \
25451 int32x2_t __ret; \
2545125452 int32x2_t __s0 = __p0; \
2545225453 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25453 int32x2_t __ret; \
2545425454 __ret = (int32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 2); \
2545525455 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2545625456 __ret; \
......@@ -25458,23 +25458,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2545825458#endif
2545925459
2546025460#define vshr_n_s64(__p0, __p1) __extension__ ({ \
25461 int64x1_t __s0 = __p0; \
2546225461 int64x1_t __ret; \
25462 int64x1_t __s0 = __p0; \
2546325463 __ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
2546425464 __ret; \
2546525465})
2546625466#ifdef __LITTLE_ENDIAN__
2546725467#define vshr_n_s16(__p0, __p1) __extension__ ({ \
25468 int16x4_t __s0 = __p0; \
2546925468 int16x4_t __ret; \
25469 int16x4_t __s0 = __p0; \
2547025470 __ret = (int16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 1); \
2547125471 __ret; \
2547225472})
2547325473#else
2547425474#define vshr_n_s16(__p0, __p1) __extension__ ({ \
25475 int16x4_t __ret; \
2547525476 int16x4_t __s0 = __p0; \
2547625477 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25477 int16x4_t __ret; \
2547825478 __ret = (int16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 1); \
2547925479 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2548025480 __ret; \
......@@ -25483,23 +25483,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2548325483
2548425484#ifdef __LITTLE_ENDIAN__
2548525485#define vshrn_n_u32(__p0, __p1) __extension__ ({ \
25486 uint32x4_t __s0 = __p0; \
2548725486 uint16x4_t __ret; \
25487 uint32x4_t __s0 = __p0; \
2548825488 __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 17); \
2548925489 __ret; \
2549025490})
2549125491#else
2549225492#define vshrn_n_u32(__p0, __p1) __extension__ ({ \
25493 uint16x4_t __ret; \
2549325494 uint32x4_t __s0 = __p0; \
2549425495 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25495 uint16x4_t __ret; \
2549625496 __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 17); \
2549725497 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2549825498 __ret; \
2549925499})
2550025500#define __noswap_vshrn_n_u32(__p0, __p1) __extension__ ({ \
25501 uint32x4_t __s0 = __p0; \
2550225501 uint16x4_t __ret; \
25502 uint32x4_t __s0 = __p0; \
2550325503 __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 17); \
2550425504 __ret; \
2550525505})
......@@ -25507,23 +25507,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2550725507
2550825508#ifdef __LITTLE_ENDIAN__
2550925509#define vshrn_n_u64(__p0, __p1) __extension__ ({ \
25510 uint64x2_t __s0 = __p0; \
2551125510 uint32x2_t __ret; \
25511 uint64x2_t __s0 = __p0; \
2551225512 __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 18); \
2551325513 __ret; \
2551425514})
2551525515#else
2551625516#define vshrn_n_u64(__p0, __p1) __extension__ ({ \
25517 uint32x2_t __ret; \
2551725518 uint64x2_t __s0 = __p0; \
2551825519 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25519 uint32x2_t __ret; \
2552025520 __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 18); \
2552125521 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2552225522 __ret; \
2552325523})
2552425524#define __noswap_vshrn_n_u64(__p0, __p1) __extension__ ({ \
25525 uint64x2_t __s0 = __p0; \
2552625525 uint32x2_t __ret; \
25526 uint64x2_t __s0 = __p0; \
2552725527 __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 18); \
2552825528 __ret; \
2552925529})
......@@ -25531,23 +25531,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2553125531
2553225532#ifdef __LITTLE_ENDIAN__
2553325533#define vshrn_n_u16(__p0, __p1) __extension__ ({ \
25534 uint16x8_t __s0 = __p0; \
2553525534 uint8x8_t __ret; \
25535 uint16x8_t __s0 = __p0; \
2553625536 __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 16); \
2553725537 __ret; \
2553825538})
2553925539#else
2554025540#define vshrn_n_u16(__p0, __p1) __extension__ ({ \
25541 uint8x8_t __ret; \
2554125542 uint16x8_t __s0 = __p0; \
2554225543 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25543 uint8x8_t __ret; \
2554425544 __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 16); \
2554525545 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2554625546 __ret; \
2554725547})
2554825548#define __noswap_vshrn_n_u16(__p0, __p1) __extension__ ({ \
25549 uint16x8_t __s0 = __p0; \
2555025549 uint8x8_t __ret; \
25550 uint16x8_t __s0 = __p0; \
2555125551 __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 16); \
2555225552 __ret; \
2555325553})
......@@ -25555,23 +25555,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2555525555
2555625556#ifdef __LITTLE_ENDIAN__
2555725557#define vshrn_n_s32(__p0, __p1) __extension__ ({ \
25558 int32x4_t __s0 = __p0; \
2555925558 int16x4_t __ret; \
25559 int32x4_t __s0 = __p0; \
2556025560 __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 1); \
2556125561 __ret; \
2556225562})
2556325563#else
2556425564#define vshrn_n_s32(__p0, __p1) __extension__ ({ \
25565 int16x4_t __ret; \
2556525566 int32x4_t __s0 = __p0; \
2556625567 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
25567 int16x4_t __ret; \
2556825568 __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 1); \
2556925569 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2557025570 __ret; \
2557125571})
2557225572#define __noswap_vshrn_n_s32(__p0, __p1) __extension__ ({ \
25573 int32x4_t __s0 = __p0; \
2557425573 int16x4_t __ret; \
25574 int32x4_t __s0 = __p0; \
2557525575 __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 1); \
2557625576 __ret; \
2557725577})
......@@ -25579,23 +25579,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2557925579
2558025580#ifdef __LITTLE_ENDIAN__
2558125581#define vshrn_n_s64(__p0, __p1) __extension__ ({ \
25582 int64x2_t __s0 = __p0; \
2558325582 int32x2_t __ret; \
25583 int64x2_t __s0 = __p0; \
2558425584 __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 2); \
2558525585 __ret; \
2558625586})
2558725587#else
2558825588#define vshrn_n_s64(__p0, __p1) __extension__ ({ \
25589 int32x2_t __ret; \
2558925590 int64x2_t __s0 = __p0; \
2559025591 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
25591 int32x2_t __ret; \
2559225592 __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 2); \
2559325593 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2559425594 __ret; \
2559525595})
2559625596#define __noswap_vshrn_n_s64(__p0, __p1) __extension__ ({ \
25597 int64x2_t __s0 = __p0; \
2559825597 int32x2_t __ret; \
25598 int64x2_t __s0 = __p0; \
2559925599 __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 2); \
2560025600 __ret; \
2560125601})
......@@ -25603,23 +25603,23 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2560325603
2560425604#ifdef __LITTLE_ENDIAN__
2560525605#define vshrn_n_s16(__p0, __p1) __extension__ ({ \
25606 int16x8_t __s0 = __p0; \
2560725606 int8x8_t __ret; \
25607 int16x8_t __s0 = __p0; \
2560825608 __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 0); \
2560925609 __ret; \
2561025610})
2561125611#else
2561225612#define vshrn_n_s16(__p0, __p1) __extension__ ({ \
25613 int8x8_t __ret; \
2561325614 int16x8_t __s0 = __p0; \
2561425615 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
25615 int8x8_t __ret; \
2561625616 __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 0); \
2561725617 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2561825618 __ret; \
2561925619})
2562025620#define __noswap_vshrn_n_s16(__p0, __p1) __extension__ ({ \
25621 int16x8_t __s0 = __p0; \
2562225621 int8x8_t __ret; \
25622 int16x8_t __s0 = __p0; \
2562325623 __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 0); \
2562425624 __ret; \
2562525625})
......@@ -25627,19 +25627,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2562725627
2562825628#ifdef __LITTLE_ENDIAN__
2562925629#define vsli_n_p8(__p0, __p1, __p2) __extension__ ({ \
25630 poly8x8_t __ret; \
2563025631 poly8x8_t __s0 = __p0; \
2563125632 poly8x8_t __s1 = __p1; \
25632 poly8x8_t __ret; \
2563325633 __ret = (poly8x8_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 4); \
2563425634 __ret; \
2563525635})
2563625636#else
2563725637#define vsli_n_p8(__p0, __p1, __p2) __extension__ ({ \
25638 poly8x8_t __ret; \
2563825639 poly8x8_t __s0 = __p0; \
2563925640 poly8x8_t __s1 = __p1; \
2564025641 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2564125642 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
25642 poly8x8_t __ret; \
2564325643 __ret = (poly8x8_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 4); \
2564425644 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2564525645 __ret; \
......@@ -25648,19 +25648,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2564825648
2564925649#ifdef __LITTLE_ENDIAN__
2565025650#define vsli_n_p16(__p0, __p1, __p2) __extension__ ({ \
25651 poly16x4_t __ret; \
2565125652 poly16x4_t __s0 = __p0; \
2565225653 poly16x4_t __s1 = __p1; \
25653 poly16x4_t __ret; \
2565425654 __ret = (poly16x4_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 5); \
2565525655 __ret; \
2565625656})
2565725657#else
2565825658#define vsli_n_p16(__p0, __p1, __p2) __extension__ ({ \
25659 poly16x4_t __ret; \
2565925660 poly16x4_t __s0 = __p0; \
2566025661 poly16x4_t __s1 = __p1; \
2566125662 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2566225663 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
25663 poly16x4_t __ret; \
2566425664 __ret = (poly16x4_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 5); \
2566525665 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2566625666 __ret; \
......@@ -25669,19 +25669,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2566925669
2567025670#ifdef __LITTLE_ENDIAN__
2567125671#define vsliq_n_p8(__p0, __p1, __p2) __extension__ ({ \
25672 poly8x16_t __ret; \
2567225673 poly8x16_t __s0 = __p0; \
2567325674 poly8x16_t __s1 = __p1; \
25674 poly8x16_t __ret; \
2567525675 __ret = (poly8x16_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 36); \
2567625676 __ret; \
2567725677})
2567825678#else
2567925679#define vsliq_n_p8(__p0, __p1, __p2) __extension__ ({ \
25680 poly8x16_t __ret; \
2568025681 poly8x16_t __s0 = __p0; \
2568125682 poly8x16_t __s1 = __p1; \
2568225683 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2568325684 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
25684 poly8x16_t __ret; \
2568525685 __ret = (poly8x16_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 36); \
2568625686 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2568725687 __ret; \
......@@ -25690,19 +25690,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2569025690
2569125691#ifdef __LITTLE_ENDIAN__
2569225692#define vsliq_n_p16(__p0, __p1, __p2) __extension__ ({ \
25693 poly16x8_t __ret; \
2569325694 poly16x8_t __s0 = __p0; \
2569425695 poly16x8_t __s1 = __p1; \
25695 poly16x8_t __ret; \
2569625696 __ret = (poly16x8_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 37); \
2569725697 __ret; \
2569825698})
2569925699#else
2570025700#define vsliq_n_p16(__p0, __p1, __p2) __extension__ ({ \
25701 poly16x8_t __ret; \
2570125702 poly16x8_t __s0 = __p0; \
2570225703 poly16x8_t __s1 = __p1; \
2570325704 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2570425705 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
25705 poly16x8_t __ret; \
2570625706 __ret = (poly16x8_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 37); \
2570725707 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2570825708 __ret; \
......@@ -25711,19 +25711,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2571125711
2571225712#ifdef __LITTLE_ENDIAN__
2571325713#define vsliq_n_u8(__p0, __p1, __p2) __extension__ ({ \
25714 uint8x16_t __ret; \
2571425715 uint8x16_t __s0 = __p0; \
2571525716 uint8x16_t __s1 = __p1; \
25716 uint8x16_t __ret; \
2571725717 __ret = (uint8x16_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 48); \
2571825718 __ret; \
2571925719})
2572025720#else
2572125721#define vsliq_n_u8(__p0, __p1, __p2) __extension__ ({ \
25722 uint8x16_t __ret; \
2572225723 uint8x16_t __s0 = __p0; \
2572325724 uint8x16_t __s1 = __p1; \
2572425725 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2572525726 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
25726 uint8x16_t __ret; \
2572725727 __ret = (uint8x16_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 48); \
2572825728 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2572925729 __ret; \
......@@ -25732,19 +25732,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2573225732
2573325733#ifdef __LITTLE_ENDIAN__
2573425734#define vsliq_n_u32(__p0, __p1, __p2) __extension__ ({ \
25735 uint32x4_t __ret; \
2573525736 uint32x4_t __s0 = __p0; \
2573625737 uint32x4_t __s1 = __p1; \
25737 uint32x4_t __ret; \
2573825738 __ret = (uint32x4_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 50); \
2573925739 __ret; \
2574025740})
2574125741#else
2574225742#define vsliq_n_u32(__p0, __p1, __p2) __extension__ ({ \
25743 uint32x4_t __ret; \
2574325744 uint32x4_t __s0 = __p0; \
2574425745 uint32x4_t __s1 = __p1; \
2574525746 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2574625747 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
25747 uint32x4_t __ret; \
2574825748 __ret = (uint32x4_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 50); \
2574925749 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2575025750 __ret; \
......@@ -25753,19 +25753,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2575325753
2575425754#ifdef __LITTLE_ENDIAN__
2575525755#define vsliq_n_u64(__p0, __p1, __p2) __extension__ ({ \
25756 uint64x2_t __ret; \
2575625757 uint64x2_t __s0 = __p0; \
2575725758 uint64x2_t __s1 = __p1; \
25758 uint64x2_t __ret; \
2575925759 __ret = (uint64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
2576025760 __ret; \
2576125761})
2576225762#else
2576325763#define vsliq_n_u64(__p0, __p1, __p2) __extension__ ({ \
25764 uint64x2_t __ret; \
2576425765 uint64x2_t __s0 = __p0; \
2576525766 uint64x2_t __s1 = __p1; \
2576625767 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2576725768 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
25768 uint64x2_t __ret; \
2576925769 __ret = (uint64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
2577025770 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2577125771 __ret; \
......@@ -25774,19 +25774,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2577425774
2577525775#ifdef __LITTLE_ENDIAN__
2577625776#define vsliq_n_u16(__p0, __p1, __p2) __extension__ ({ \
25777 uint16x8_t __ret; \
2577725778 uint16x8_t __s0 = __p0; \
2577825779 uint16x8_t __s1 = __p1; \
25779 uint16x8_t __ret; \
2578025780 __ret = (uint16x8_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 49); \
2578125781 __ret; \
2578225782})
2578325783#else
2578425784#define vsliq_n_u16(__p0, __p1, __p2) __extension__ ({ \
25785 uint16x8_t __ret; \
2578525786 uint16x8_t __s0 = __p0; \
2578625787 uint16x8_t __s1 = __p1; \
2578725788 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2578825789 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
25789 uint16x8_t __ret; \
2579025790 __ret = (uint16x8_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 49); \
2579125791 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2579225792 __ret; \
......@@ -25795,19 +25795,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2579525795
2579625796#ifdef __LITTLE_ENDIAN__
2579725797#define vsliq_n_s8(__p0, __p1, __p2) __extension__ ({ \
25798 int8x16_t __ret; \
2579825799 int8x16_t __s0 = __p0; \
2579925800 int8x16_t __s1 = __p1; \
25800 int8x16_t __ret; \
2580125801 __ret = (int8x16_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 32); \
2580225802 __ret; \
2580325803})
2580425804#else
2580525805#define vsliq_n_s8(__p0, __p1, __p2) __extension__ ({ \
25806 int8x16_t __ret; \
2580625807 int8x16_t __s0 = __p0; \
2580725808 int8x16_t __s1 = __p1; \
2580825809 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2580925810 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
25810 int8x16_t __ret; \
2581125811 __ret = (int8x16_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 32); \
2581225812 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2581325813 __ret; \
......@@ -25816,19 +25816,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2581625816
2581725817#ifdef __LITTLE_ENDIAN__
2581825818#define vsliq_n_s32(__p0, __p1, __p2) __extension__ ({ \
25819 int32x4_t __ret; \
2581925820 int32x4_t __s0 = __p0; \
2582025821 int32x4_t __s1 = __p1; \
25821 int32x4_t __ret; \
2582225822 __ret = (int32x4_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
2582325823 __ret; \
2582425824})
2582525825#else
2582625826#define vsliq_n_s32(__p0, __p1, __p2) __extension__ ({ \
25827 int32x4_t __ret; \
2582725828 int32x4_t __s0 = __p0; \
2582825829 int32x4_t __s1 = __p1; \
2582925830 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2583025831 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
25831 int32x4_t __ret; \
2583225832 __ret = (int32x4_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
2583325833 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2583425834 __ret; \
......@@ -25837,19 +25837,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2583725837
2583825838#ifdef __LITTLE_ENDIAN__
2583925839#define vsliq_n_s64(__p0, __p1, __p2) __extension__ ({ \
25840 int64x2_t __ret; \
2584025841 int64x2_t __s0 = __p0; \
2584125842 int64x2_t __s1 = __p1; \
25842 int64x2_t __ret; \
2584325843 __ret = (int64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 35); \
2584425844 __ret; \
2584525845})
2584625846#else
2584725847#define vsliq_n_s64(__p0, __p1, __p2) __extension__ ({ \
25848 int64x2_t __ret; \
2584825849 int64x2_t __s0 = __p0; \
2584925850 int64x2_t __s1 = __p1; \
2585025851 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2585125852 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
25852 int64x2_t __ret; \
2585325853 __ret = (int64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 35); \
2585425854 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2585525855 __ret; \
......@@ -25858,19 +25858,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2585825858
2585925859#ifdef __LITTLE_ENDIAN__
2586025860#define vsliq_n_s16(__p0, __p1, __p2) __extension__ ({ \
25861 int16x8_t __ret; \
2586125862 int16x8_t __s0 = __p0; \
2586225863 int16x8_t __s1 = __p1; \
25863 int16x8_t __ret; \
2586425864 __ret = (int16x8_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
2586525865 __ret; \
2586625866})
2586725867#else
2586825868#define vsliq_n_s16(__p0, __p1, __p2) __extension__ ({ \
25869 int16x8_t __ret; \
2586925870 int16x8_t __s0 = __p0; \
2587025871 int16x8_t __s1 = __p1; \
2587125872 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2587225873 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
25873 int16x8_t __ret; \
2587425874 __ret = (int16x8_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
2587525875 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2587625876 __ret; \
......@@ -25879,19 +25879,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2587925879
2588025880#ifdef __LITTLE_ENDIAN__
2588125881#define vsli_n_u8(__p0, __p1, __p2) __extension__ ({ \
25882 uint8x8_t __ret; \
2588225883 uint8x8_t __s0 = __p0; \
2588325884 uint8x8_t __s1 = __p1; \
25884 uint8x8_t __ret; \
2588525885 __ret = (uint8x8_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 16); \
2588625886 __ret; \
2588725887})
2588825888#else
2588925889#define vsli_n_u8(__p0, __p1, __p2) __extension__ ({ \
25890 uint8x8_t __ret; \
2589025891 uint8x8_t __s0 = __p0; \
2589125892 uint8x8_t __s1 = __p1; \
2589225893 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2589325894 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
25894 uint8x8_t __ret; \
2589525895 __ret = (uint8x8_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 16); \
2589625896 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2589725897 __ret; \
......@@ -25900,19 +25900,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2590025900
2590125901#ifdef __LITTLE_ENDIAN__
2590225902#define vsli_n_u32(__p0, __p1, __p2) __extension__ ({ \
25903 uint32x2_t __ret; \
2590325904 uint32x2_t __s0 = __p0; \
2590425905 uint32x2_t __s1 = __p1; \
25905 uint32x2_t __ret; \
2590625906 __ret = (uint32x2_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 18); \
2590725907 __ret; \
2590825908})
2590925909#else
2591025910#define vsli_n_u32(__p0, __p1, __p2) __extension__ ({ \
25911 uint32x2_t __ret; \
2591125912 uint32x2_t __s0 = __p0; \
2591225913 uint32x2_t __s1 = __p1; \
2591325914 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2591425915 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
25915 uint32x2_t __ret; \
2591625916 __ret = (uint32x2_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 18); \
2591725917 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2591825918 __ret; \
......@@ -25920,27 +25920,27 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2592025920#endif
2592125921
2592225922#define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
25923 uint64x1_t __ret; \
2592325924 uint64x1_t __s0 = __p0; \
2592425925 uint64x1_t __s1 = __p1; \
25925 uint64x1_t __ret; \
2592625926 __ret = (uint64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
2592725927 __ret; \
2592825928})
2592925929#ifdef __LITTLE_ENDIAN__
2593025930#define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
25931 uint16x4_t __ret; \
2593125932 uint16x4_t __s0 = __p0; \
2593225933 uint16x4_t __s1 = __p1; \
25933 uint16x4_t __ret; \
2593425934 __ret = (uint16x4_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 17); \
2593525935 __ret; \
2593625936})
2593725937#else
2593825938#define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
25939 uint16x4_t __ret; \
2593925940 uint16x4_t __s0 = __p0; \
2594025941 uint16x4_t __s1 = __p1; \
2594125942 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2594225943 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
25943 uint16x4_t __ret; \
2594425944 __ret = (uint16x4_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 17); \
2594525945 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2594625946 __ret; \
......@@ -25949,19 +25949,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2594925949
2595025950#ifdef __LITTLE_ENDIAN__
2595125951#define vsli_n_s8(__p0, __p1, __p2) __extension__ ({ \
25952 int8x8_t __ret; \
2595225953 int8x8_t __s0 = __p0; \
2595325954 int8x8_t __s1 = __p1; \
25954 int8x8_t __ret; \
2595525955 __ret = (int8x8_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 0); \
2595625956 __ret; \
2595725957})
2595825958#else
2595925959#define vsli_n_s8(__p0, __p1, __p2) __extension__ ({ \
25960 int8x8_t __ret; \
2596025961 int8x8_t __s0 = __p0; \
2596125962 int8x8_t __s1 = __p1; \
2596225963 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2596325964 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
25964 int8x8_t __ret; \
2596525965 __ret = (int8x8_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 0); \
2596625966 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2596725967 __ret; \
......@@ -25970,19 +25970,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2597025970
2597125971#ifdef __LITTLE_ENDIAN__
2597225972#define vsli_n_s32(__p0, __p1, __p2) __extension__ ({ \
25973 int32x2_t __ret; \
2597325974 int32x2_t __s0 = __p0; \
2597425975 int32x2_t __s1 = __p1; \
25975 int32x2_t __ret; \
2597625976 __ret = (int32x2_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
2597725977 __ret; \
2597825978})
2597925979#else
2598025980#define vsli_n_s32(__p0, __p1, __p2) __extension__ ({ \
25981 int32x2_t __ret; \
2598125982 int32x2_t __s0 = __p0; \
2598225983 int32x2_t __s1 = __p1; \
2598325984 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2598425985 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
25985 int32x2_t __ret; \
2598625986 __ret = (int32x2_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
2598725987 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2598825988 __ret; \
......@@ -25990,27 +25990,27 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2599025990#endif
2599125991
2599225992#define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
25993 int64x1_t __ret; \
2599325994 int64x1_t __s0 = __p0; \
2599425995 int64x1_t __s1 = __p1; \
25995 int64x1_t __ret; \
2599625996 __ret = (int64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
2599725997 __ret; \
2599825998})
2599925999#ifdef __LITTLE_ENDIAN__
2600026000#define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
26001 int16x4_t __ret; \
2600126002 int16x4_t __s0 = __p0; \
2600226003 int16x4_t __s1 = __p1; \
26003 int16x4_t __ret; \
2600426004 __ret = (int16x4_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
2600526005 __ret; \
2600626006})
2600726007#else
2600826008#define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
26009 int16x4_t __ret; \
2600926010 int16x4_t __s0 = __p0; \
2601026011 int16x4_t __s1 = __p1; \
2601126012 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2601226013 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26013 int16x4_t __ret; \
2601426014 __ret = (int16x4_t) __builtin_neon_vsli_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
2601526015 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2601626016 __ret; \
......@@ -26019,19 +26019,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2601926019
2602026020#ifdef __LITTLE_ENDIAN__
2602126021#define vsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
26022 uint8x16_t __ret; \
2602226023 uint8x16_t __s0 = __p0; \
2602326024 uint8x16_t __s1 = __p1; \
26024 uint8x16_t __ret; \
2602526025 __ret = (uint8x16_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 48); \
2602626026 __ret; \
2602726027})
2602826028#else
2602926029#define vsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
26030 uint8x16_t __ret; \
2603026031 uint8x16_t __s0 = __p0; \
2603126032 uint8x16_t __s1 = __p1; \
2603226033 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2603326034 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
26034 uint8x16_t __ret; \
2603526035 __ret = (uint8x16_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 48); \
2603626036 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2603726037 __ret; \
......@@ -26040,19 +26040,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2604026040
2604126041#ifdef __LITTLE_ENDIAN__
2604226042#define vsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
26043 uint32x4_t __ret; \
2604326044 uint32x4_t __s0 = __p0; \
2604426045 uint32x4_t __s1 = __p1; \
26045 uint32x4_t __ret; \
2604626046 __ret = (uint32x4_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 50); \
2604726047 __ret; \
2604826048})
2604926049#else
2605026050#define vsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
26051 uint32x4_t __ret; \
2605126052 uint32x4_t __s0 = __p0; \
2605226053 uint32x4_t __s1 = __p1; \
2605326054 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2605426055 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26055 uint32x4_t __ret; \
2605626056 __ret = (uint32x4_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 50); \
2605726057 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2605826058 __ret; \
......@@ -26061,19 +26061,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2606126061
2606226062#ifdef __LITTLE_ENDIAN__
2606326063#define vsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
26064 uint64x2_t __ret; \
2606426065 uint64x2_t __s0 = __p0; \
2606526066 uint64x2_t __s1 = __p1; \
26066 uint64x2_t __ret; \
2606726067 __ret = (uint64x2_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
2606826068 __ret; \
2606926069})
2607026070#else
2607126071#define vsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
26072 uint64x2_t __ret; \
2607226073 uint64x2_t __s0 = __p0; \
2607326074 uint64x2_t __s1 = __p1; \
2607426075 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2607526076 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26076 uint64x2_t __ret; \
2607726077 __ret = (uint64x2_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
2607826078 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2607926079 __ret; \
......@@ -26082,19 +26082,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2608226082
2608326083#ifdef __LITTLE_ENDIAN__
2608426084#define vsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
26085 uint16x8_t __ret; \
2608526086 uint16x8_t __s0 = __p0; \
2608626087 uint16x8_t __s1 = __p1; \
26087 uint16x8_t __ret; \
2608826088 __ret = (uint16x8_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 49); \
2608926089 __ret; \
2609026090})
2609126091#else
2609226092#define vsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
26093 uint16x8_t __ret; \
2609326094 uint16x8_t __s0 = __p0; \
2609426095 uint16x8_t __s1 = __p1; \
2609526096 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2609626097 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26097 uint16x8_t __ret; \
2609826098 __ret = (uint16x8_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 49); \
2609926099 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2610026100 __ret; \
......@@ -26103,19 +26103,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2610326103
2610426104#ifdef __LITTLE_ENDIAN__
2610526105#define vsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
26106 int8x16_t __ret; \
2610626107 int8x16_t __s0 = __p0; \
2610726108 int8x16_t __s1 = __p1; \
26108 int8x16_t __ret; \
2610926109 __ret = (int8x16_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 32); \
2611026110 __ret; \
2611126111})
2611226112#else
2611326113#define vsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
26114 int8x16_t __ret; \
2611426115 int8x16_t __s0 = __p0; \
2611526116 int8x16_t __s1 = __p1; \
2611626117 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2611726118 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
26118 int8x16_t __ret; \
2611926119 __ret = (int8x16_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 32); \
2612026120 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2612126121 __ret; \
......@@ -26124,19 +26124,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2612426124
2612526125#ifdef __LITTLE_ENDIAN__
2612626126#define vsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
26127 int32x4_t __ret; \
2612726128 int32x4_t __s0 = __p0; \
2612826129 int32x4_t __s1 = __p1; \
26129 int32x4_t __ret; \
2613026130 __ret = (int32x4_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
2613126131 __ret; \
2613226132})
2613326133#else
2613426134#define vsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
26135 int32x4_t __ret; \
2613526136 int32x4_t __s0 = __p0; \
2613626137 int32x4_t __s1 = __p1; \
2613726138 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2613826139 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26139 int32x4_t __ret; \
2614026140 __ret = (int32x4_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
2614126141 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2614226142 __ret; \
......@@ -26145,19 +26145,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2614526145
2614626146#ifdef __LITTLE_ENDIAN__
2614726147#define vsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
26148 int64x2_t __ret; \
2614826149 int64x2_t __s0 = __p0; \
2614926150 int64x2_t __s1 = __p1; \
26150 int64x2_t __ret; \
2615126151 __ret = (int64x2_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 35); \
2615226152 __ret; \
2615326153})
2615426154#else
2615526155#define vsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
26156 int64x2_t __ret; \
2615626157 int64x2_t __s0 = __p0; \
2615726158 int64x2_t __s1 = __p1; \
2615826159 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2615926160 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26160 int64x2_t __ret; \
2616126161 __ret = (int64x2_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 35); \
2616226162 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2616326163 __ret; \
......@@ -26166,19 +26166,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2616626166
2616726167#ifdef __LITTLE_ENDIAN__
2616826168#define vsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
26169 int16x8_t __ret; \
2616926170 int16x8_t __s0 = __p0; \
2617026171 int16x8_t __s1 = __p1; \
26171 int16x8_t __ret; \
2617226172 __ret = (int16x8_t) __builtin_neon_vsraq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
2617326173 __ret; \
2617426174})
2617526175#else
2617626176#define vsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
26177 int16x8_t __ret; \
2617726178 int16x8_t __s0 = __p0; \
2617826179 int16x8_t __s1 = __p1; \
2617926180 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2618026181 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26181 int16x8_t __ret; \
2618226182 __ret = (int16x8_t) __builtin_neon_vsraq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
2618326183 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2618426184 __ret; \
......@@ -26187,19 +26187,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2618726187
2618826188#ifdef __LITTLE_ENDIAN__
2618926189#define vsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
26190 uint8x8_t __ret; \
2619026191 uint8x8_t __s0 = __p0; \
2619126192 uint8x8_t __s1 = __p1; \
26192 uint8x8_t __ret; \
2619326193 __ret = (uint8x8_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 16); \
2619426194 __ret; \
2619526195})
2619626196#else
2619726197#define vsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
26198 uint8x8_t __ret; \
2619826199 uint8x8_t __s0 = __p0; \
2619926200 uint8x8_t __s1 = __p1; \
2620026201 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2620126202 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26202 uint8x8_t __ret; \
2620326203 __ret = (uint8x8_t) __builtin_neon_vsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 16); \
2620426204 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2620526205 __ret; \
......@@ -26208,19 +26208,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2620826208
2620926209#ifdef __LITTLE_ENDIAN__
2621026210#define vsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
26211 uint32x2_t __ret; \
2621126212 uint32x2_t __s0 = __p0; \
2621226213 uint32x2_t __s1 = __p1; \
26213 uint32x2_t __ret; \
2621426214 __ret = (uint32x2_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 18); \
2621526215 __ret; \
2621626216})
2621726217#else
2621826218#define vsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
26219 uint32x2_t __ret; \
2621926220 uint32x2_t __s0 = __p0; \
2622026221 uint32x2_t __s1 = __p1; \
2622126222 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2622226223 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26223 uint32x2_t __ret; \
2622426224 __ret = (uint32x2_t) __builtin_neon_vsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 18); \
2622526225 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2622626226 __ret; \
......@@ -26228,27 +26228,27 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2622826228#endif
2622926229
2623026230#define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
26231 uint64x1_t __ret; \
2623126232 uint64x1_t __s0 = __p0; \
2623226233 uint64x1_t __s1 = __p1; \
26233 uint64x1_t __ret; \
2623426234 __ret = (uint64x1_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
2623526235 __ret; \
2623626236})
2623726237#ifdef __LITTLE_ENDIAN__
2623826238#define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
26239 uint16x4_t __ret; \
2623926240 uint16x4_t __s0 = __p0; \
2624026241 uint16x4_t __s1 = __p1; \
26241 uint16x4_t __ret; \
2624226242 __ret = (uint16x4_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 17); \
2624326243 __ret; \
2624426244})
2624526245#else
2624626246#define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
26247 uint16x4_t __ret; \
2624726248 uint16x4_t __s0 = __p0; \
2624826249 uint16x4_t __s1 = __p1; \
2624926250 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2625026251 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26251 uint16x4_t __ret; \
2625226252 __ret = (uint16x4_t) __builtin_neon_vsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 17); \
2625326253 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2625426254 __ret; \
......@@ -26257,19 +26257,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2625726257
2625826258#ifdef __LITTLE_ENDIAN__
2625926259#define vsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
26260 int8x8_t __ret; \
2626026261 int8x8_t __s0 = __p0; \
2626126262 int8x8_t __s1 = __p1; \
26262 int8x8_t __ret; \
2626326263 __ret = (int8x8_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 0); \
2626426264 __ret; \
2626526265})
2626626266#else
2626726267#define vsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
26268 int8x8_t __ret; \
2626826269 int8x8_t __s0 = __p0; \
2626926270 int8x8_t __s1 = __p1; \
2627026271 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2627126272 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26272 int8x8_t __ret; \
2627326273 __ret = (int8x8_t) __builtin_neon_vsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 0); \
2627426274 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2627526275 __ret; \
......@@ -26278,19 +26278,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2627826278
2627926279#ifdef __LITTLE_ENDIAN__
2628026280#define vsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
26281 int32x2_t __ret; \
2628126282 int32x2_t __s0 = __p0; \
2628226283 int32x2_t __s1 = __p1; \
26283 int32x2_t __ret; \
2628426284 __ret = (int32x2_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
2628526285 __ret; \
2628626286})
2628726287#else
2628826288#define vsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
26289 int32x2_t __ret; \
2628926290 int32x2_t __s0 = __p0; \
2629026291 int32x2_t __s1 = __p1; \
2629126292 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2629226293 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26293 int32x2_t __ret; \
2629426294 __ret = (int32x2_t) __builtin_neon_vsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
2629526295 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2629626296 __ret; \
......@@ -26298,27 +26298,27 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2629826298#endif
2629926299
2630026300#define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
26301 int64x1_t __ret; \
2630126302 int64x1_t __s0 = __p0; \
2630226303 int64x1_t __s1 = __p1; \
26303 int64x1_t __ret; \
2630426304 __ret = (int64x1_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
2630526305 __ret; \
2630626306})
2630726307#ifdef __LITTLE_ENDIAN__
2630826308#define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
26309 int16x4_t __ret; \
2630926310 int16x4_t __s0 = __p0; \
2631026311 int16x4_t __s1 = __p1; \
26311 int16x4_t __ret; \
2631226312 __ret = (int16x4_t) __builtin_neon_vsra_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
2631326313 __ret; \
2631426314})
2631526315#else
2631626316#define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
26317 int16x4_t __ret; \
2631726318 int16x4_t __s0 = __p0; \
2631826319 int16x4_t __s1 = __p1; \
2631926320 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2632026321 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26321 int16x4_t __ret; \
2632226322 __ret = (int16x4_t) __builtin_neon_vsra_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
2632326323 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2632426324 __ret; \
......@@ -26327,19 +26327,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2632726327
2632826328#ifdef __LITTLE_ENDIAN__
2632926329#define vsri_n_p8(__p0, __p1, __p2) __extension__ ({ \
26330 poly8x8_t __ret; \
2633026331 poly8x8_t __s0 = __p0; \
2633126332 poly8x8_t __s1 = __p1; \
26332 poly8x8_t __ret; \
2633326333 __ret = (poly8x8_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 4); \
2633426334 __ret; \
2633526335})
2633626336#else
2633726337#define vsri_n_p8(__p0, __p1, __p2) __extension__ ({ \
26338 poly8x8_t __ret; \
2633826339 poly8x8_t __s0 = __p0; \
2633926340 poly8x8_t __s1 = __p1; \
2634026341 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2634126342 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26342 poly8x8_t __ret; \
2634326343 __ret = (poly8x8_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 4); \
2634426344 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2634526345 __ret; \
......@@ -26348,19 +26348,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2634826348
2634926349#ifdef __LITTLE_ENDIAN__
2635026350#define vsri_n_p16(__p0, __p1, __p2) __extension__ ({ \
26351 poly16x4_t __ret; \
2635126352 poly16x4_t __s0 = __p0; \
2635226353 poly16x4_t __s1 = __p1; \
26353 poly16x4_t __ret; \
2635426354 __ret = (poly16x4_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 5); \
2635526355 __ret; \
2635626356})
2635726357#else
2635826358#define vsri_n_p16(__p0, __p1, __p2) __extension__ ({ \
26359 poly16x4_t __ret; \
2635926360 poly16x4_t __s0 = __p0; \
2636026361 poly16x4_t __s1 = __p1; \
2636126362 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2636226363 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26363 poly16x4_t __ret; \
2636426364 __ret = (poly16x4_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 5); \
2636526365 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2636626366 __ret; \
......@@ -26369,19 +26369,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2636926369
2637026370#ifdef __LITTLE_ENDIAN__
2637126371#define vsriq_n_p8(__p0, __p1, __p2) __extension__ ({ \
26372 poly8x16_t __ret; \
2637226373 poly8x16_t __s0 = __p0; \
2637326374 poly8x16_t __s1 = __p1; \
26374 poly8x16_t __ret; \
2637526375 __ret = (poly8x16_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 36); \
2637626376 __ret; \
2637726377})
2637826378#else
2637926379#define vsriq_n_p8(__p0, __p1, __p2) __extension__ ({ \
26380 poly8x16_t __ret; \
2638026381 poly8x16_t __s0 = __p0; \
2638126382 poly8x16_t __s1 = __p1; \
2638226383 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2638326384 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
26384 poly8x16_t __ret; \
2638526385 __ret = (poly8x16_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 36); \
2638626386 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2638726387 __ret; \
......@@ -26390,19 +26390,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2639026390
2639126391#ifdef __LITTLE_ENDIAN__
2639226392#define vsriq_n_p16(__p0, __p1, __p2) __extension__ ({ \
26393 poly16x8_t __ret; \
2639326394 poly16x8_t __s0 = __p0; \
2639426395 poly16x8_t __s1 = __p1; \
26395 poly16x8_t __ret; \
2639626396 __ret = (poly16x8_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 37); \
2639726397 __ret; \
2639826398})
2639926399#else
2640026400#define vsriq_n_p16(__p0, __p1, __p2) __extension__ ({ \
26401 poly16x8_t __ret; \
2640126402 poly16x8_t __s0 = __p0; \
2640226403 poly16x8_t __s1 = __p1; \
2640326404 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2640426405 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26405 poly16x8_t __ret; \
2640626406 __ret = (poly16x8_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 37); \
2640726407 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2640826408 __ret; \
......@@ -26411,19 +26411,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2641126411
2641226412#ifdef __LITTLE_ENDIAN__
2641326413#define vsriq_n_u8(__p0, __p1, __p2) __extension__ ({ \
26414 uint8x16_t __ret; \
2641426415 uint8x16_t __s0 = __p0; \
2641526416 uint8x16_t __s1 = __p1; \
26416 uint8x16_t __ret; \
2641726417 __ret = (uint8x16_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 48); \
2641826418 __ret; \
2641926419})
2642026420#else
2642126421#define vsriq_n_u8(__p0, __p1, __p2) __extension__ ({ \
26422 uint8x16_t __ret; \
2642226423 uint8x16_t __s0 = __p0; \
2642326424 uint8x16_t __s1 = __p1; \
2642426425 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2642526426 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
26426 uint8x16_t __ret; \
2642726427 __ret = (uint8x16_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 48); \
2642826428 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2642926429 __ret; \
......@@ -26432,19 +26432,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2643226432
2643326433#ifdef __LITTLE_ENDIAN__
2643426434#define vsriq_n_u32(__p0, __p1, __p2) __extension__ ({ \
26435 uint32x4_t __ret; \
2643526436 uint32x4_t __s0 = __p0; \
2643626437 uint32x4_t __s1 = __p1; \
26437 uint32x4_t __ret; \
2643826438 __ret = (uint32x4_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 50); \
2643926439 __ret; \
2644026440})
2644126441#else
2644226442#define vsriq_n_u32(__p0, __p1, __p2) __extension__ ({ \
26443 uint32x4_t __ret; \
2644326444 uint32x4_t __s0 = __p0; \
2644426445 uint32x4_t __s1 = __p1; \
2644526446 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2644626447 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26447 uint32x4_t __ret; \
2644826448 __ret = (uint32x4_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 50); \
2644926449 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2645026450 __ret; \
......@@ -26453,19 +26453,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2645326453
2645426454#ifdef __LITTLE_ENDIAN__
2645526455#define vsriq_n_u64(__p0, __p1, __p2) __extension__ ({ \
26456 uint64x2_t __ret; \
2645626457 uint64x2_t __s0 = __p0; \
2645726458 uint64x2_t __s1 = __p1; \
26458 uint64x2_t __ret; \
2645926459 __ret = (uint64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
2646026460 __ret; \
2646126461})
2646226462#else
2646326463#define vsriq_n_u64(__p0, __p1, __p2) __extension__ ({ \
26464 uint64x2_t __ret; \
2646426465 uint64x2_t __s0 = __p0; \
2646526466 uint64x2_t __s1 = __p1; \
2646626467 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2646726468 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26468 uint64x2_t __ret; \
2646926469 __ret = (uint64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
2647026470 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2647126471 __ret; \
......@@ -26474,19 +26474,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2647426474
2647526475#ifdef __LITTLE_ENDIAN__
2647626476#define vsriq_n_u16(__p0, __p1, __p2) __extension__ ({ \
26477 uint16x8_t __ret; \
2647726478 uint16x8_t __s0 = __p0; \
2647826479 uint16x8_t __s1 = __p1; \
26479 uint16x8_t __ret; \
2648026480 __ret = (uint16x8_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 49); \
2648126481 __ret; \
2648226482})
2648326483#else
2648426484#define vsriq_n_u16(__p0, __p1, __p2) __extension__ ({ \
26485 uint16x8_t __ret; \
2648526486 uint16x8_t __s0 = __p0; \
2648626487 uint16x8_t __s1 = __p1; \
2648726488 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2648826489 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26489 uint16x8_t __ret; \
2649026490 __ret = (uint16x8_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 49); \
2649126491 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2649226492 __ret; \
......@@ -26495,19 +26495,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2649526495
2649626496#ifdef __LITTLE_ENDIAN__
2649726497#define vsriq_n_s8(__p0, __p1, __p2) __extension__ ({ \
26498 int8x16_t __ret; \
2649826499 int8x16_t __s0 = __p0; \
2649926500 int8x16_t __s1 = __p1; \
26500 int8x16_t __ret; \
2650126501 __ret = (int8x16_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 32); \
2650226502 __ret; \
2650326503})
2650426504#else
2650526505#define vsriq_n_s8(__p0, __p1, __p2) __extension__ ({ \
26506 int8x16_t __ret; \
2650626507 int8x16_t __s0 = __p0; \
2650726508 int8x16_t __s1 = __p1; \
2650826509 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2650926510 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
26510 int8x16_t __ret; \
2651126511 __ret = (int8x16_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 32); \
2651226512 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
2651326513 __ret; \
......@@ -26516,19 +26516,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2651626516
2651726517#ifdef __LITTLE_ENDIAN__
2651826518#define vsriq_n_s32(__p0, __p1, __p2) __extension__ ({ \
26519 int32x4_t __ret; \
2651926520 int32x4_t __s0 = __p0; \
2652026521 int32x4_t __s1 = __p1; \
26521 int32x4_t __ret; \
2652226522 __ret = (int32x4_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
2652326523 __ret; \
2652426524})
2652526525#else
2652626526#define vsriq_n_s32(__p0, __p1, __p2) __extension__ ({ \
26527 int32x4_t __ret; \
2652726528 int32x4_t __s0 = __p0; \
2652826529 int32x4_t __s1 = __p1; \
2652926530 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2653026531 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26531 int32x4_t __ret; \
2653226532 __ret = (int32x4_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
2653326533 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2653426534 __ret; \
......@@ -26537,19 +26537,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2653726537
2653826538#ifdef __LITTLE_ENDIAN__
2653926539#define vsriq_n_s64(__p0, __p1, __p2) __extension__ ({ \
26540 int64x2_t __ret; \
2654026541 int64x2_t __s0 = __p0; \
2654126542 int64x2_t __s1 = __p1; \
26542 int64x2_t __ret; \
2654326543 __ret = (int64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 35); \
2654426544 __ret; \
2654526545})
2654626546#else
2654726547#define vsriq_n_s64(__p0, __p1, __p2) __extension__ ({ \
26548 int64x2_t __ret; \
2654826549 int64x2_t __s0 = __p0; \
2654926550 int64x2_t __s1 = __p1; \
2655026551 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2655126552 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26552 int64x2_t __ret; \
2655326553 __ret = (int64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 35); \
2655426554 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2655526555 __ret; \
......@@ -26558,19 +26558,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2655826558
2655926559#ifdef __LITTLE_ENDIAN__
2656026560#define vsriq_n_s16(__p0, __p1, __p2) __extension__ ({ \
26561 int16x8_t __ret; \
2656126562 int16x8_t __s0 = __p0; \
2656226563 int16x8_t __s1 = __p1; \
26563 int16x8_t __ret; \
2656426564 __ret = (int16x8_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
2656526565 __ret; \
2656626566})
2656726567#else
2656826568#define vsriq_n_s16(__p0, __p1, __p2) __extension__ ({ \
26569 int16x8_t __ret; \
2656926570 int16x8_t __s0 = __p0; \
2657026571 int16x8_t __s1 = __p1; \
2657126572 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2657226573 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26573 int16x8_t __ret; \
2657426574 __ret = (int16x8_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
2657526575 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2657626576 __ret; \
......@@ -26579,19 +26579,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2657926579
2658026580#ifdef __LITTLE_ENDIAN__
2658126581#define vsri_n_u8(__p0, __p1, __p2) __extension__ ({ \
26582 uint8x8_t __ret; \
2658226583 uint8x8_t __s0 = __p0; \
2658326584 uint8x8_t __s1 = __p1; \
26584 uint8x8_t __ret; \
2658526585 __ret = (uint8x8_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 16); \
2658626586 __ret; \
2658726587})
2658826588#else
2658926589#define vsri_n_u8(__p0, __p1, __p2) __extension__ ({ \
26590 uint8x8_t __ret; \
2659026591 uint8x8_t __s0 = __p0; \
2659126592 uint8x8_t __s1 = __p1; \
2659226593 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2659326594 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26594 uint8x8_t __ret; \
2659526595 __ret = (uint8x8_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 16); \
2659626596 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2659726597 __ret; \
......@@ -26600,19 +26600,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2660026600
2660126601#ifdef __LITTLE_ENDIAN__
2660226602#define vsri_n_u32(__p0, __p1, __p2) __extension__ ({ \
26603 uint32x2_t __ret; \
2660326604 uint32x2_t __s0 = __p0; \
2660426605 uint32x2_t __s1 = __p1; \
26605 uint32x2_t __ret; \
2660626606 __ret = (uint32x2_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 18); \
2660726607 __ret; \
2660826608})
2660926609#else
2661026610#define vsri_n_u32(__p0, __p1, __p2) __extension__ ({ \
26611 uint32x2_t __ret; \
2661126612 uint32x2_t __s0 = __p0; \
2661226613 uint32x2_t __s1 = __p1; \
2661326614 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2661426615 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26615 uint32x2_t __ret; \
2661626616 __ret = (uint32x2_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 18); \
2661726617 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2661826618 __ret; \
......@@ -26620,27 +26620,27 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2662026620#endif
2662126621
2662226622#define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
26623 uint64x1_t __ret; \
2662326624 uint64x1_t __s0 = __p0; \
2662426625 uint64x1_t __s1 = __p1; \
26625 uint64x1_t __ret; \
2662626626 __ret = (uint64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 19); \
2662726627 __ret; \
2662826628})
2662926629#ifdef __LITTLE_ENDIAN__
2663026630#define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
26631 uint16x4_t __ret; \
2663126632 uint16x4_t __s0 = __p0; \
2663226633 uint16x4_t __s1 = __p1; \
26633 uint16x4_t __ret; \
2663426634 __ret = (uint16x4_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 17); \
2663526635 __ret; \
2663626636})
2663726637#else
2663826638#define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
26639 uint16x4_t __ret; \
2663926640 uint16x4_t __s0 = __p0; \
2664026641 uint16x4_t __s1 = __p1; \
2664126642 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2664226643 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26643 uint16x4_t __ret; \
2664426644 __ret = (uint16x4_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 17); \
2664526645 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2664626646 __ret; \
......@@ -26649,19 +26649,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2664926649
2665026650#ifdef __LITTLE_ENDIAN__
2665126651#define vsri_n_s8(__p0, __p1, __p2) __extension__ ({ \
26652 int8x8_t __ret; \
2665226653 int8x8_t __s0 = __p0; \
2665326654 int8x8_t __s1 = __p1; \
26654 int8x8_t __ret; \
2665526655 __ret = (int8x8_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 0); \
2665626656 __ret; \
2665726657})
2665826658#else
2665926659#define vsri_n_s8(__p0, __p1, __p2) __extension__ ({ \
26660 int8x8_t __ret; \
2666026661 int8x8_t __s0 = __p0; \
2666126662 int8x8_t __s1 = __p1; \
2666226663 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
2666326664 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
26664 int8x8_t __ret; \
2666526665 __ret = (int8x8_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 0); \
2666626666 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
2666726667 __ret; \
......@@ -26670,19 +26670,19 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2667026670
2667126671#ifdef __LITTLE_ENDIAN__
2667226672#define vsri_n_s32(__p0, __p1, __p2) __extension__ ({ \
26673 int32x2_t __ret; \
2667326674 int32x2_t __s0 = __p0; \
2667426675 int32x2_t __s1 = __p1; \
26675 int32x2_t __ret; \
2667626676 __ret = (int32x2_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
2667726677 __ret; \
2667826678})
2667926679#else
2668026680#define vsri_n_s32(__p0, __p1, __p2) __extension__ ({ \
26681 int32x2_t __ret; \
2668126682 int32x2_t __s0 = __p0; \
2668226683 int32x2_t __s1 = __p1; \
2668326684 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
2668426685 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
26685 int32x2_t __ret; \
2668626686 __ret = (int32x2_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
2668726687 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
2668826688 __ret; \
......@@ -26690,27 +26690,27 @@ __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
2669026690#endif
2669126691
2669226692#define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
26693 int64x1_t __ret; \
2669326694 int64x1_t __s0 = __p0; \
2669426695 int64x1_t __s1 = __p1; \
26695 int64x1_t __ret; \
2669626696 __ret = (int64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 3); \
2669726697 __ret; \
2669826698})
2669926699#ifdef __LITTLE_ENDIAN__
2670026700#define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
26701 int16x4_t __ret; \
2670126702 int16x4_t __s0 = __p0; \
2670226703 int16x4_t __s1 = __p1; \
26703 int16x4_t __ret; \
2670426704 __ret = (int16x4_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
2670526705 __ret; \
2670626706})
2670726707#else
2670826708#define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
26709 int16x4_t __ret; \
2670926710 int16x4_t __s0 = __p0; \
2671026711 int16x4_t __s1 = __p1; \
2671126712 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
2671226713 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
26713 int16x4_t __ret; \
2671426714 __ret = (int16x4_t) __builtin_neon_vsri_n_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
2671526715 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
2671626716 __ret; \
......@@ -29853,9 +29853,9 @@ __ai uint8x16_t vsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2985329853}
2985429854#else
2985529855__ai uint8x16_t vsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29856 uint8x16_t __ret;
2985629857 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2985729858 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29858 uint8x16_t __ret;
2985929859 __ret = __rev0 - __rev1;
2986029860 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2986129861 return __ret;
......@@ -29870,9 +29870,9 @@ __ai uint32x4_t vsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2987029870}
2987129871#else
2987229872__ai uint32x4_t vsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29873 uint32x4_t __ret;
2987329874 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2987429875 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29875 uint32x4_t __ret;
2987629876 __ret = __rev0 - __rev1;
2987729877 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2987829878 return __ret;
......@@ -29887,9 +29887,9 @@ __ai uint64x2_t vsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
2988729887}
2988829888#else
2988929889__ai uint64x2_t vsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
29890 uint64x2_t __ret;
2989029891 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2989129892 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29892 uint64x2_t __ret;
2989329893 __ret = __rev0 - __rev1;
2989429894 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2989529895 return __ret;
......@@ -29904,9 +29904,9 @@ __ai uint16x8_t vsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2990429904}
2990529905#else
2990629906__ai uint16x8_t vsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29907 uint16x8_t __ret;
2990729908 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2990829909 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29909 uint16x8_t __ret;
2991029910 __ret = __rev0 - __rev1;
2991129911 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2991229912 return __ret;
......@@ -29921,9 +29921,9 @@ __ai int8x16_t vsubq_s8(int8x16_t __p0, int8x16_t __p1) {
2992129921}
2992229922#else
2992329923__ai int8x16_t vsubq_s8(int8x16_t __p0, int8x16_t __p1) {
29924 int8x16_t __ret;
2992429925 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2992529926 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29926 int8x16_t __ret;
2992729927 __ret = __rev0 - __rev1;
2992829928 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2992929929 return __ret;
......@@ -29938,9 +29938,9 @@ __ai float32x4_t vsubq_f32(float32x4_t __p0, float32x4_t __p1) {
2993829938}
2993929939#else
2994029940__ai float32x4_t vsubq_f32(float32x4_t __p0, float32x4_t __p1) {
29941 float32x4_t __ret;
2994129942 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2994229943 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29943 float32x4_t __ret;
2994429944 __ret = __rev0 - __rev1;
2994529945 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2994629946 return __ret;
......@@ -29955,9 +29955,9 @@ __ai int32x4_t vsubq_s32(int32x4_t __p0, int32x4_t __p1) {
2995529955}
2995629956#else
2995729957__ai int32x4_t vsubq_s32(int32x4_t __p0, int32x4_t __p1) {
29958 int32x4_t __ret;
2995829959 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
2995929960 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29960 int32x4_t __ret;
2996129961 __ret = __rev0 - __rev1;
2996229962 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
2996329963 return __ret;
......@@ -29972,9 +29972,9 @@ __ai int64x2_t vsubq_s64(int64x2_t __p0, int64x2_t __p1) {
2997229972}
2997329973#else
2997429974__ai int64x2_t vsubq_s64(int64x2_t __p0, int64x2_t __p1) {
29975 int64x2_t __ret;
2997529976 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
2997629977 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29977 int64x2_t __ret;
2997829978 __ret = __rev0 - __rev1;
2997929979 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
2998029980 return __ret;
......@@ -29989,9 +29989,9 @@ __ai int16x8_t vsubq_s16(int16x8_t __p0, int16x8_t __p1) {
2998929989}
2999029990#else
2999129991__ai int16x8_t vsubq_s16(int16x8_t __p0, int16x8_t __p1) {
29992 int16x8_t __ret;
2999229993 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
2999329994 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29994 int16x8_t __ret;
2999529995 __ret = __rev0 - __rev1;
2999629996 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
2999729997 return __ret;
......@@ -30006,9 +30006,9 @@ __ai uint8x8_t vsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
3000630006}
3000730007#else
3000830008__ai uint8x8_t vsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
30009 uint8x8_t __ret;
3000930010 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3001030011 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30011 uint8x8_t __ret;
3001230012 __ret = __rev0 - __rev1;
3001330013 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3001430014 return __ret;
......@@ -30023,9 +30023,9 @@ __ai uint32x2_t vsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
3002330023}
3002430024#else
3002530025__ai uint32x2_t vsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
30026 uint32x2_t __ret;
3002630027 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3002730028 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30028 uint32x2_t __ret;
3002930029 __ret = __rev0 - __rev1;
3003030030 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3003130031 return __ret;
......@@ -30045,9 +30045,9 @@ __ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
3004530045}
3004630046#else
3004730047__ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
30048 uint16x4_t __ret;
3004830049 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3004930050 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30050 uint16x4_t __ret;
3005130051 __ret = __rev0 - __rev1;
3005230052 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3005330053 return __ret;
......@@ -30062,9 +30062,9 @@ __ai int8x8_t vsub_s8(int8x8_t __p0, int8x8_t __p1) {
3006230062}
3006330063#else
3006430064__ai int8x8_t vsub_s8(int8x8_t __p0, int8x8_t __p1) {
30065 int8x8_t __ret;
3006530066 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3006630067 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30067 int8x8_t __ret;
3006830068 __ret = __rev0 - __rev1;
3006930069 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3007030070 return __ret;
......@@ -30079,9 +30079,9 @@ __ai float32x2_t vsub_f32(float32x2_t __p0, float32x2_t __p1) {
3007930079}
3008030080#else
3008130081__ai float32x2_t vsub_f32(float32x2_t __p0, float32x2_t __p1) {
30082 float32x2_t __ret;
3008230083 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3008330084 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30084 float32x2_t __ret;
3008530085 __ret = __rev0 - __rev1;
3008630086 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3008730087 return __ret;
......@@ -30096,9 +30096,9 @@ __ai int32x2_t vsub_s32(int32x2_t __p0, int32x2_t __p1) {
3009630096}
3009730097#else
3009830098__ai int32x2_t vsub_s32(int32x2_t __p0, int32x2_t __p1) {
30099 int32x2_t __ret;
3009930100 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3010030101 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30101 int32x2_t __ret;
3010230102 __ret = __rev0 - __rev1;
3010330103 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3010430104 return __ret;
......@@ -30118,9 +30118,9 @@ __ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
3011830118}
3011930119#else
3012030120__ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
30121 int16x4_t __ret;
3012130122 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3012230123 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30123 int16x4_t __ret;
3012430124 __ret = __rev0 - __rev1;
3012530125 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3012630126 return __ret;
......@@ -30135,9 +30135,9 @@ __ai uint16x4_t vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
3013530135}
3013630136#else
3013730137__ai uint16x4_t vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
30138 uint16x4_t __ret;
3013830139 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3013930140 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30140 uint16x4_t __ret;
3014130141 __ret = (uint16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 17);
3014230142 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3014330143 return __ret;
......@@ -30157,9 +30157,9 @@ __ai uint32x2_t vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
3015730157}
3015830158#else
3015930159__ai uint32x2_t vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
30160 uint32x2_t __ret;
3016030161 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3016130162 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30162 uint32x2_t __ret;
3016330163 __ret = (uint32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 18);
3016430164 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3016530165 return __ret;
......@@ -30179,9 +30179,9 @@ __ai uint8x8_t vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
3017930179}
3018030180#else
3018130181__ai uint8x8_t vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
30182 uint8x8_t __ret;
3018230183 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3018330184 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30184 uint8x8_t __ret;
3018530185 __ret = (uint8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 16);
3018630186 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3018730187 return __ret;
......@@ -30201,9 +30201,9 @@ __ai int16x4_t vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
3020130201}
3020230202#else
3020330203__ai int16x4_t vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
30204 int16x4_t __ret;
3020430205 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3020530206 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30206 int16x4_t __ret;
3020730207 __ret = (int16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 1);
3020830208 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3020930209 return __ret;
......@@ -30223,9 +30223,9 @@ __ai int32x2_t vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
3022330223}
3022430224#else
3022530225__ai int32x2_t vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
30226 int32x2_t __ret;
3022630227 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3022730228 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30228 int32x2_t __ret;
3022930229 __ret = (int32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 2);
3023030230 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3023130231 return __ret;
......@@ -30245,9 +30245,9 @@ __ai int8x8_t vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
3024530245}
3024630246#else
3024730247__ai int8x8_t vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
30248 int8x8_t __ret;
3024830249 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3024930250 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30250 int8x8_t __ret;
3025130251 __ret = (int8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__rev0, (int8x16_t)__rev1, 0);
3025230252 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3025330253 return __ret;
......@@ -30267,9 +30267,9 @@ __ai uint16x8_t vsubl_u8(uint8x8_t __p0, uint8x8_t __p1) {
3026730267}
3026830268#else
3026930269__ai uint16x8_t vsubl_u8(uint8x8_t __p0, uint8x8_t __p1) {
30270 uint16x8_t __ret;
3027030271 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3027130272 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30272 uint16x8_t __ret;
3027330273 __ret = __noswap_vmovl_u8(__rev0) - __noswap_vmovl_u8(__rev1);
3027430274 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3027530275 return __ret;
......@@ -30284,9 +30284,9 @@ __ai uint64x2_t vsubl_u32(uint32x2_t __p0, uint32x2_t __p1) {
3028430284}
3028530285#else
3028630286__ai uint64x2_t vsubl_u32(uint32x2_t __p0, uint32x2_t __p1) {
30287 uint64x2_t __ret;
3028730288 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3028830289 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30289 uint64x2_t __ret;
3029030290 __ret = __noswap_vmovl_u32(__rev0) - __noswap_vmovl_u32(__rev1);
3029130291 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3029230292 return __ret;
......@@ -30301,9 +30301,9 @@ __ai uint32x4_t vsubl_u16(uint16x4_t __p0, uint16x4_t __p1) {
3030130301}
3030230302#else
3030330303__ai uint32x4_t vsubl_u16(uint16x4_t __p0, uint16x4_t __p1) {
30304 uint32x4_t __ret;
3030430305 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3030530306 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30306 uint32x4_t __ret;
3030730307 __ret = __noswap_vmovl_u16(__rev0) - __noswap_vmovl_u16(__rev1);
3030830308 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3030930309 return __ret;
......@@ -30318,9 +30318,9 @@ __ai int16x8_t vsubl_s8(int8x8_t __p0, int8x8_t __p1) {
3031830318}
3031930319#else
3032030320__ai int16x8_t vsubl_s8(int8x8_t __p0, int8x8_t __p1) {
30321 int16x8_t __ret;
3032130322 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3032230323 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30323 int16x8_t __ret;
3032430324 __ret = __noswap_vmovl_s8(__rev0) - __noswap_vmovl_s8(__rev1);
3032530325 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3032630326 return __ret;
......@@ -30335,9 +30335,9 @@ __ai int64x2_t vsubl_s32(int32x2_t __p0, int32x2_t __p1) {
3033530335}
3033630336#else
3033730337__ai int64x2_t vsubl_s32(int32x2_t __p0, int32x2_t __p1) {
30338 int64x2_t __ret;
3033830339 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3033930340 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30340 int64x2_t __ret;
3034130341 __ret = __noswap_vmovl_s32(__rev0) - __noswap_vmovl_s32(__rev1);
3034230342 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3034330343 return __ret;
......@@ -30352,9 +30352,9 @@ __ai int32x4_t vsubl_s16(int16x4_t __p0, int16x4_t __p1) {
3035230352}
3035330353#else
3035430354__ai int32x4_t vsubl_s16(int16x4_t __p0, int16x4_t __p1) {
30355 int32x4_t __ret;
3035530356 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3035630357 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30357 int32x4_t __ret;
3035830358 __ret = __noswap_vmovl_s16(__rev0) - __noswap_vmovl_s16(__rev1);
3035930359 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3036030360 return __ret;
......@@ -30369,9 +30369,9 @@ __ai uint16x8_t vsubw_u8(uint16x8_t __p0, uint8x8_t __p1) {
3036930369}
3037030370#else
3037130371__ai uint16x8_t vsubw_u8(uint16x8_t __p0, uint8x8_t __p1) {
30372 uint16x8_t __ret;
3037230373 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3037330374 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30374 uint16x8_t __ret;
3037530375 __ret = __rev0 - __noswap_vmovl_u8(__rev1);
3037630376 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3037730377 return __ret;
......@@ -30386,9 +30386,9 @@ __ai uint64x2_t vsubw_u32(uint64x2_t __p0, uint32x2_t __p1) {
3038630386}
3038730387#else
3038830388__ai uint64x2_t vsubw_u32(uint64x2_t __p0, uint32x2_t __p1) {
30389 uint64x2_t __ret;
3038930390 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3039030391 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30391 uint64x2_t __ret;
3039230392 __ret = __rev0 - __noswap_vmovl_u32(__rev1);
3039330393 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3039430394 return __ret;
......@@ -30403,9 +30403,9 @@ __ai uint32x4_t vsubw_u16(uint32x4_t __p0, uint16x4_t __p1) {
3040330403}
3040430404#else
3040530405__ai uint32x4_t vsubw_u16(uint32x4_t __p0, uint16x4_t __p1) {
30406 uint32x4_t __ret;
3040630407 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3040730408 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30408 uint32x4_t __ret;
3040930409 __ret = __rev0 - __noswap_vmovl_u16(__rev1);
3041030410 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3041130411 return __ret;
......@@ -30420,9 +30420,9 @@ __ai int16x8_t vsubw_s8(int16x8_t __p0, int8x8_t __p1) {
3042030420}
3042130421#else
3042230422__ai int16x8_t vsubw_s8(int16x8_t __p0, int8x8_t __p1) {
30423 int16x8_t __ret;
3042330424 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3042430425 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30425 int16x8_t __ret;
3042630426 __ret = __rev0 - __noswap_vmovl_s8(__rev1);
3042730427 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3042830428 return __ret;
......@@ -30437,9 +30437,9 @@ __ai int64x2_t vsubw_s32(int64x2_t __p0, int32x2_t __p1) {
3043730437}
3043830438#else
3043930439__ai int64x2_t vsubw_s32(int64x2_t __p0, int32x2_t __p1) {
30440 int64x2_t __ret;
3044030441 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3044130442 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
30442 int64x2_t __ret;
3044330443 __ret = __rev0 - __noswap_vmovl_s32(__rev1);
3044430444 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3044530445 return __ret;
......@@ -30454,9 +30454,9 @@ __ai int32x4_t vsubw_s16(int32x4_t __p0, int16x4_t __p1) {
3045430454}
3045530455#else
3045630456__ai int32x4_t vsubw_s16(int32x4_t __p0, int16x4_t __p1) {
30457 int32x4_t __ret;
3045730458 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3045830459 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30459 int32x4_t __ret;
3046030460 __ret = __rev0 - __noswap_vmovl_s16(__rev1);
3046130461 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3046230462 return __ret;
......@@ -30471,9 +30471,9 @@ __ai poly8x8_t vtbl1_p8(poly8x8_t __p0, uint8x8_t __p1) {
3047130471}
3047230472#else
3047330473__ai poly8x8_t vtbl1_p8(poly8x8_t __p0, uint8x8_t __p1) {
30474 poly8x8_t __ret;
3047430475 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3047530476 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30476 poly8x8_t __ret;
3047730477 __ret = (poly8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__rev0, (int8x8_t)__rev1, 4);
3047830478 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3047930479 return __ret;
......@@ -30488,9 +30488,9 @@ __ai uint8x8_t vtbl1_u8(uint8x8_t __p0, uint8x8_t __p1) {
3048830488}
3048930489#else
3049030490__ai uint8x8_t vtbl1_u8(uint8x8_t __p0, uint8x8_t __p1) {
30491 uint8x8_t __ret;
3049130492 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3049230493 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30493 uint8x8_t __ret;
3049430494 __ret = (uint8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3049530495 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3049630496 return __ret;
......@@ -30505,9 +30505,9 @@ __ai int8x8_t vtbl1_s8(int8x8_t __p0, int8x8_t __p1) {
3050530505}
3050630506#else
3050730507__ai int8x8_t vtbl1_s8(int8x8_t __p0, int8x8_t __p1) {
30508 int8x8_t __ret;
3050830509 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3050930510 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30510 int8x8_t __ret;
3051130511 __ret = (int8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
3051230512 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3051330513 return __ret;
......@@ -30522,11 +30522,11 @@ __ai poly8x8_t vtbl2_p8(poly8x8x2_t __p0, uint8x8_t __p1) {
3052230522}
3052330523#else
3052430524__ai poly8x8_t vtbl2_p8(poly8x8x2_t __p0, uint8x8_t __p1) {
30525 poly8x8_t __ret;
3052530526 poly8x8x2_t __rev0;
3052630527 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3052730528 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3052830529 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30529 poly8x8_t __ret;
3053030530 __ret = (poly8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev1, 4);
3053130531 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3053230532 return __ret;
......@@ -30541,11 +30541,11 @@ __ai uint8x8_t vtbl2_u8(uint8x8x2_t __p0, uint8x8_t __p1) {
3054130541}
3054230542#else
3054330543__ai uint8x8_t vtbl2_u8(uint8x8x2_t __p0, uint8x8_t __p1) {
30544 uint8x8_t __ret;
3054430545 uint8x8x2_t __rev0;
3054530546 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3054630547 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3054730548 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30548 uint8x8_t __ret;
3054930549 __ret = (uint8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev1, 16);
3055030550 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3055130551 return __ret;
......@@ -30560,11 +30560,11 @@ __ai int8x8_t vtbl2_s8(int8x8x2_t __p0, int8x8_t __p1) {
3056030560}
3056130561#else
3056230562__ai int8x8_t vtbl2_s8(int8x8x2_t __p0, int8x8_t __p1) {
30563 int8x8_t __ret;
3056330564 int8x8x2_t __rev0;
3056430565 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3056530566 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3056630567 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30567 int8x8_t __ret;
3056830568 __ret = (int8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev1, 0);
3056930569 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3057030570 return __ret;
......@@ -30579,12 +30579,12 @@ __ai poly8x8_t vtbl3_p8(poly8x8x3_t __p0, uint8x8_t __p1) {
3057930579}
3058030580#else
3058130581__ai poly8x8_t vtbl3_p8(poly8x8x3_t __p0, uint8x8_t __p1) {
30582 poly8x8_t __ret;
3058230583 poly8x8x3_t __rev0;
3058330584 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3058430585 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3058530586 __rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3058630587 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30587 poly8x8_t __ret;
3058830588 __ret = (poly8x8_t) __builtin_neon_vtbl3_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev0.val[2], (int8x8_t)__rev1, 4);
3058930589 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3059030590 return __ret;
......@@ -30599,12 +30599,12 @@ __ai uint8x8_t vtbl3_u8(uint8x8x3_t __p0, uint8x8_t __p1) {
3059930599}
3060030600#else
3060130601__ai uint8x8_t vtbl3_u8(uint8x8x3_t __p0, uint8x8_t __p1) {
30602 uint8x8_t __ret;
3060230603 uint8x8x3_t __rev0;
3060330604 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3060430605 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3060530606 __rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3060630607 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30607 uint8x8_t __ret;
3060830608 __ret = (uint8x8_t) __builtin_neon_vtbl3_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev0.val[2], (int8x8_t)__rev1, 16);
3060930609 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3061030610 return __ret;
......@@ -30619,12 +30619,12 @@ __ai int8x8_t vtbl3_s8(int8x8x3_t __p0, int8x8_t __p1) {
3061930619}
3062030620#else
3062130621__ai int8x8_t vtbl3_s8(int8x8x3_t __p0, int8x8_t __p1) {
30622 int8x8_t __ret;
3062230623 int8x8x3_t __rev0;
3062330624 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3062430625 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3062530626 __rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3062630627 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30627 int8x8_t __ret;
3062830628 __ret = (int8x8_t) __builtin_neon_vtbl3_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev0.val[2], (int8x8_t)__rev1, 0);
3062930629 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3063030630 return __ret;
......@@ -30639,13 +30639,13 @@ __ai poly8x8_t vtbl4_p8(poly8x8x4_t __p0, uint8x8_t __p1) {
3063930639}
3064030640#else
3064130641__ai poly8x8_t vtbl4_p8(poly8x8x4_t __p0, uint8x8_t __p1) {
30642 poly8x8_t __ret;
3064230643 poly8x8x4_t __rev0;
3064330644 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3064430645 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3064530646 __rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3064630647 __rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
3064730648 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30648 poly8x8_t __ret;
3064930649 __ret = (poly8x8_t) __builtin_neon_vtbl4_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev0.val[2], (int8x8_t)__rev0.val[3], (int8x8_t)__rev1, 4);
3065030650 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3065130651 return __ret;
......@@ -30660,13 +30660,13 @@ __ai uint8x8_t vtbl4_u8(uint8x8x4_t __p0, uint8x8_t __p1) {
3066030660}
3066130661#else
3066230662__ai uint8x8_t vtbl4_u8(uint8x8x4_t __p0, uint8x8_t __p1) {
30663 uint8x8_t __ret;
3066330664 uint8x8x4_t __rev0;
3066430665 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3066530666 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3066630667 __rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3066730668 __rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
3066830669 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30669 uint8x8_t __ret;
3067030670 __ret = (uint8x8_t) __builtin_neon_vtbl4_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev0.val[2], (int8x8_t)__rev0.val[3], (int8x8_t)__rev1, 16);
3067130671 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3067230672 return __ret;
......@@ -30681,13 +30681,13 @@ __ai int8x8_t vtbl4_s8(int8x8x4_t __p0, int8x8_t __p1) {
3068130681}
3068230682#else
3068330683__ai int8x8_t vtbl4_s8(int8x8x4_t __p0, int8x8_t __p1) {
30684 int8x8_t __ret;
3068430685 int8x8x4_t __rev0;
3068530686 __rev0.val[0] = __builtin_shufflevector(__p0.val[0], __p0.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3068630687 __rev0.val[1] = __builtin_shufflevector(__p0.val[1], __p0.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3068730688 __rev0.val[2] = __builtin_shufflevector(__p0.val[2], __p0.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3068830689 __rev0.val[3] = __builtin_shufflevector(__p0.val[3], __p0.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
3068930690 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30690 int8x8_t __ret;
3069130691 __ret = (int8x8_t) __builtin_neon_vtbl4_v((int8x8_t)__rev0.val[0], (int8x8_t)__rev0.val[1], (int8x8_t)__rev0.val[2], (int8x8_t)__rev0.val[3], (int8x8_t)__rev1, 0);
3069230692 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3069330693 return __ret;
......@@ -30702,10 +30702,10 @@ __ai poly8x8_t vtbx1_p8(poly8x8_t __p0, poly8x8_t __p1, uint8x8_t __p2) {
3070230702}
3070330703#else
3070430704__ai poly8x8_t vtbx1_p8(poly8x8_t __p0, poly8x8_t __p1, uint8x8_t __p2) {
30705 poly8x8_t __ret;
3070530706 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3070630707 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3070730708 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30708 poly8x8_t __ret;
3070930709 __ret = (poly8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 4);
3071030710 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3071130711 return __ret;
......@@ -30720,10 +30720,10 @@ __ai uint8x8_t vtbx1_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
3072030720}
3072130721#else
3072230722__ai uint8x8_t vtbx1_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
30723 uint8x8_t __ret;
3072330724 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3072430725 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3072530726 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30726 uint8x8_t __ret;
3072730727 __ret = (uint8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 16);
3072830728 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3072930729 return __ret;
......@@ -30738,10 +30738,10 @@ __ai int8x8_t vtbx1_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
3073830738}
3073930739#else
3074030740__ai int8x8_t vtbx1_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
30741 int8x8_t __ret;
3074130742 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3074230743 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3074330744 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30744 int8x8_t __ret;
3074530745 __ret = (int8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 0);
3074630746 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3074730747 return __ret;
......@@ -30756,12 +30756,12 @@ __ai poly8x8_t vtbx2_p8(poly8x8_t __p0, poly8x8x2_t __p1, uint8x8_t __p2) {
3075630756}
3075730757#else
3075830758__ai poly8x8_t vtbx2_p8(poly8x8_t __p0, poly8x8x2_t __p1, uint8x8_t __p2) {
30759 poly8x8_t __ret;
3075930760 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3076030761 poly8x8x2_t __rev1;
3076130762 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3076230763 __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3076330764 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30764 poly8x8_t __ret;
3076530765 __ret = (poly8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev2, 4);
3076630766 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3076730767 return __ret;
......@@ -30776,12 +30776,12 @@ __ai uint8x8_t vtbx2_u8(uint8x8_t __p0, uint8x8x2_t __p1, uint8x8_t __p2) {
3077630776}
3077730777#else
3077830778__ai uint8x8_t vtbx2_u8(uint8x8_t __p0, uint8x8x2_t __p1, uint8x8_t __p2) {
30779 uint8x8_t __ret;
3077930780 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3078030781 uint8x8x2_t __rev1;
3078130782 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3078230783 __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3078330784 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30784 uint8x8_t __ret;
3078530785 __ret = (uint8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev2, 16);
3078630786 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3078730787 return __ret;
......@@ -30796,12 +30796,12 @@ __ai int8x8_t vtbx2_s8(int8x8_t __p0, int8x8x2_t __p1, int8x8_t __p2) {
3079630796}
3079730797#else
3079830798__ai int8x8_t vtbx2_s8(int8x8_t __p0, int8x8x2_t __p1, int8x8_t __p2) {
30799 int8x8_t __ret;
3079930800 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3080030801 int8x8x2_t __rev1;
3080130802 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3080230803 __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3080330804 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30804 int8x8_t __ret;
3080530805 __ret = (int8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev2, 0);
3080630806 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3080730807 return __ret;
......@@ -30816,13 +30816,13 @@ __ai poly8x8_t vtbx3_p8(poly8x8_t __p0, poly8x8x3_t __p1, uint8x8_t __p2) {
3081630816}
3081730817#else
3081830818__ai poly8x8_t vtbx3_p8(poly8x8_t __p0, poly8x8x3_t __p1, uint8x8_t __p2) {
30819 poly8x8_t __ret;
3081930820 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3082030821 poly8x8x3_t __rev1;
3082130822 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3082230823 __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3082330824 __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3082430825 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30825 poly8x8_t __ret;
3082630826 __ret = (poly8x8_t) __builtin_neon_vtbx3_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev2, 4);
3082730827 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3082830828 return __ret;
......@@ -30837,13 +30837,13 @@ __ai uint8x8_t vtbx3_u8(uint8x8_t __p0, uint8x8x3_t __p1, uint8x8_t __p2) {
3083730837}
3083830838#else
3083930839__ai uint8x8_t vtbx3_u8(uint8x8_t __p0, uint8x8x3_t __p1, uint8x8_t __p2) {
30840 uint8x8_t __ret;
3084030841 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3084130842 uint8x8x3_t __rev1;
3084230843 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3084330844 __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3084430845 __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3084530846 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30846 uint8x8_t __ret;
3084730847 __ret = (uint8x8_t) __builtin_neon_vtbx3_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev2, 16);
3084830848 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3084930849 return __ret;
......@@ -30858,13 +30858,13 @@ __ai int8x8_t vtbx3_s8(int8x8_t __p0, int8x8x3_t __p1, int8x8_t __p2) {
3085830858}
3085930859#else
3086030860__ai int8x8_t vtbx3_s8(int8x8_t __p0, int8x8x3_t __p1, int8x8_t __p2) {
30861 int8x8_t __ret;
3086130862 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3086230863 int8x8x3_t __rev1;
3086330864 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
3086430865 __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
3086530866 __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3086630867 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30867 int8x8_t __ret;
3086830868 __ret = (int8x8_t) __builtin_neon_vtbx3_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev2, 0);
3086930869 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3087030870 return __ret;
......@@ -30879,6 +30879,7 @@ __ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
3087930879}
3088030880#else
3088130881__ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
30882 poly8x8_t __ret;
3088230883 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3088330884 poly8x8x4_t __rev1;
3088430885 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -30886,7 +30887,6 @@ __ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
3088630887 __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3088730888 __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
3088830889 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30889 poly8x8_t __ret;
3089030890 __ret = (poly8x8_t) __builtin_neon_vtbx4_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], (int8x8_t)__rev2, 4);
3089130891 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3089230892 return __ret;
......@@ -30901,6 +30901,7 @@ __ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
3090130901}
3090230902#else
3090330903__ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
30904 uint8x8_t __ret;
3090430905 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3090530906 uint8x8x4_t __rev1;
3090630907 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -30908,7 +30909,6 @@ __ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
3090830909 __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3090930910 __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
3091030911 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30911 uint8x8_t __ret;
3091230912 __ret = (uint8x8_t) __builtin_neon_vtbx4_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], (int8x8_t)__rev2, 16);
3091330913 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3091430914 return __ret;
......@@ -30923,6 +30923,7 @@ __ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
3092330923}
3092430924#else
3092530925__ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
30926 int8x8_t __ret;
3092630927 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3092730928 int8x8x4_t __rev1;
3092830929 __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -30930,7 +30931,6 @@ __ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
3093030931 __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
3093130932 __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
3093230933 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
30933 int8x8_t __ret;
3093430934 __ret = (int8x8_t) __builtin_neon_vtbx4_v((int8x8_t)__rev0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], (int8x8_t)__rev1.val[3], (int8x8_t)__rev2, 0);
3093530935 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3093630936 return __ret;
......@@ -30945,9 +30945,9 @@ __ai poly8x8x2_t vtrn_p8(poly8x8_t __p0, poly8x8_t __p1) {
3094530945}
3094630946#else
3094730947__ai poly8x8x2_t vtrn_p8(poly8x8_t __p0, poly8x8_t __p1) {
30948 poly8x8x2_t __ret;
3094830949 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3094930950 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
30950 poly8x8x2_t __ret;
3095130951 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 4);
3095230952
3095330953 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -30964,9 +30964,9 @@ __ai poly16x4x2_t vtrn_p16(poly16x4_t __p0, poly16x4_t __p1) {
3096430964}
3096530965#else
3096630966__ai poly16x4x2_t vtrn_p16(poly16x4_t __p0, poly16x4_t __p1) {
30967 poly16x4x2_t __ret;
3096730968 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3096830969 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
30969 poly16x4x2_t __ret;
3097030970 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 5);
3097130971
3097230972 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -30983,9 +30983,9 @@ __ai poly8x16x2_t vtrnq_p8(poly8x16_t __p0, poly8x16_t __p1) {
3098330983}
3098430984#else
3098530985__ai poly8x16x2_t vtrnq_p8(poly8x16_t __p0, poly8x16_t __p1) {
30986 poly8x16x2_t __ret;
3098630987 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3098730988 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
30988 poly8x16x2_t __ret;
3098930989 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 36);
3099030990
3099130991 __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);
......@@ -31002,9 +31002,9 @@ __ai poly16x8x2_t vtrnq_p16(poly16x8_t __p0, poly16x8_t __p1) {
3100231002}
3100331003#else
3100431004__ai poly16x8x2_t vtrnq_p16(poly16x8_t __p0, poly16x8_t __p1) {
31005 poly16x8x2_t __ret;
3100531006 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3100631007 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31007 poly16x8x2_t __ret;
3100831008 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 37);
3100931009
3101031010 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31021,9 +31021,9 @@ __ai uint8x16x2_t vtrnq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3102131021}
3102231022#else
3102331023__ai uint8x16x2_t vtrnq_u8(uint8x16_t __p0, uint8x16_t __p1) {
31024 uint8x16x2_t __ret;
3102431025 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3102531026 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31026 uint8x16x2_t __ret;
3102731027 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3102831028
3102931029 __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);
......@@ -31040,9 +31040,9 @@ __ai uint32x4x2_t vtrnq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3104031040}
3104131041#else
3104231042__ai uint32x4x2_t vtrnq_u32(uint32x4_t __p0, uint32x4_t __p1) {
31043 uint32x4x2_t __ret;
3104331044 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3104431045 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31045 uint32x4x2_t __ret;
3104631046 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3104731047
3104831048 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31059,9 +31059,9 @@ __ai uint16x8x2_t vtrnq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3105931059}
3106031060#else
3106131061__ai uint16x8x2_t vtrnq_u16(uint16x8_t __p0, uint16x8_t __p1) {
31062 uint16x8x2_t __ret;
3106231063 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3106331064 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31064 uint16x8x2_t __ret;
3106531065 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 49);
3106631066
3106731067 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31078,9 +31078,9 @@ __ai int8x16x2_t vtrnq_s8(int8x16_t __p0, int8x16_t __p1) {
3107831078}
3107931079#else
3108031080__ai int8x16x2_t vtrnq_s8(int8x16_t __p0, int8x16_t __p1) {
31081 int8x16x2_t __ret;
3108131082 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3108231083 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31083 int8x16x2_t __ret;
3108431084 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 32);
3108531085
3108631086 __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);
......@@ -31097,9 +31097,9 @@ __ai float32x4x2_t vtrnq_f32(float32x4_t __p0, float32x4_t __p1) {
3109731097}
3109831098#else
3109931099__ai float32x4x2_t vtrnq_f32(float32x4_t __p0, float32x4_t __p1) {
31100 float32x4x2_t __ret;
3110031101 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3110131102 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31102 float32x4x2_t __ret;
3110331103 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 41);
3110431104
3110531105 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31116,9 +31116,9 @@ __ai int32x4x2_t vtrnq_s32(int32x4_t __p0, int32x4_t __p1) {
3111631116}
3111731117#else
3111831118__ai int32x4x2_t vtrnq_s32(int32x4_t __p0, int32x4_t __p1) {
31119 int32x4x2_t __ret;
3111931120 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3112031121 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31121 int32x4x2_t __ret;
3112231122 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 34);
3112331123
3112431124 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31135,9 +31135,9 @@ __ai int16x8x2_t vtrnq_s16(int16x8_t __p0, int16x8_t __p1) {
3113531135}
3113631136#else
3113731137__ai int16x8x2_t vtrnq_s16(int16x8_t __p0, int16x8_t __p1) {
31138 int16x8x2_t __ret;
3113831139 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3113931140 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31140 int16x8x2_t __ret;
3114131141 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 33);
3114231142
3114331143 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31154,9 +31154,9 @@ __ai uint8x8x2_t vtrn_u8(uint8x8_t __p0, uint8x8_t __p1) {
3115431154}
3115531155#else
3115631156__ai uint8x8x2_t vtrn_u8(uint8x8_t __p0, uint8x8_t __p1) {
31157 uint8x8x2_t __ret;
3115731158 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3115831159 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31159 uint8x8x2_t __ret;
3116031160 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3116131161
3116231162 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31173,9 +31173,9 @@ __ai uint32x2x2_t vtrn_u32(uint32x2_t __p0, uint32x2_t __p1) {
3117331173}
3117431174#else
3117531175__ai uint32x2x2_t vtrn_u32(uint32x2_t __p0, uint32x2_t __p1) {
31176 uint32x2x2_t __ret;
3117631177 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3117731178 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31178 uint32x2x2_t __ret;
3117931179 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 18);
3118031180
3118131181 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -31192,9 +31192,9 @@ __ai uint16x4x2_t vtrn_u16(uint16x4_t __p0, uint16x4_t __p1) {
3119231192}
3119331193#else
3119431194__ai uint16x4x2_t vtrn_u16(uint16x4_t __p0, uint16x4_t __p1) {
31195 uint16x4x2_t __ret;
3119531196 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3119631197 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31197 uint16x4x2_t __ret;
3119831198 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 17);
3119931199
3120031200 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31211,9 +31211,9 @@ __ai int8x8x2_t vtrn_s8(int8x8_t __p0, int8x8_t __p1) {
3121131211}
3121231212#else
3121331213__ai int8x8x2_t vtrn_s8(int8x8_t __p0, int8x8_t __p1) {
31214 int8x8x2_t __ret;
3121431215 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3121531216 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31216 int8x8x2_t __ret;
3121731217 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 0);
3121831218
3121931219 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31230,9 +31230,9 @@ __ai float32x2x2_t vtrn_f32(float32x2_t __p0, float32x2_t __p1) {
3123031230}
3123131231#else
3123231232__ai float32x2x2_t vtrn_f32(float32x2_t __p0, float32x2_t __p1) {
31233 float32x2x2_t __ret;
3123331234 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3123431235 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31235 float32x2x2_t __ret;
3123631236 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 9);
3123731237
3123831238 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -31249,9 +31249,9 @@ __ai int32x2x2_t vtrn_s32(int32x2_t __p0, int32x2_t __p1) {
3124931249}
3125031250#else
3125131251__ai int32x2x2_t vtrn_s32(int32x2_t __p0, int32x2_t __p1) {
31252 int32x2x2_t __ret;
3125231253 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3125331254 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31254 int32x2x2_t __ret;
3125531255 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 2);
3125631256
3125731257 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -31268,9 +31268,9 @@ __ai int16x4x2_t vtrn_s16(int16x4_t __p0, int16x4_t __p1) {
3126831268}
3126931269#else
3127031270__ai int16x4x2_t vtrn_s16(int16x4_t __p0, int16x4_t __p1) {
31271 int16x4x2_t __ret;
3127131272 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3127231273 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31273 int16x4x2_t __ret;
3127431274 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 1);
3127531275
3127631276 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31287,9 +31287,9 @@ __ai uint8x8_t vtst_p8(poly8x8_t __p0, poly8x8_t __p1) {
3128731287}
3128831288#else
3128931289__ai uint8x8_t vtst_p8(poly8x8_t __p0, poly8x8_t __p1) {
31290 uint8x8_t __ret;
3129031291 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3129131292 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31292 uint8x8_t __ret;
3129331293 __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3129431294 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3129531295 return __ret;
......@@ -31304,9 +31304,9 @@ __ai uint16x4_t vtst_p16(poly16x4_t __p0, poly16x4_t __p1) {
3130431304}
3130531305#else
3130631306__ai uint16x4_t vtst_p16(poly16x4_t __p0, poly16x4_t __p1) {
31307 uint16x4_t __ret;
3130731308 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3130831309 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31309 uint16x4_t __ret;
3131031310 __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
3131131311 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3131231312 return __ret;
......@@ -31321,9 +31321,9 @@ __ai uint8x16_t vtstq_p8(poly8x16_t __p0, poly8x16_t __p1) {
3132131321}
3132231322#else
3132331323__ai uint8x16_t vtstq_p8(poly8x16_t __p0, poly8x16_t __p1) {
31324 uint8x16_t __ret;
3132431325 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3132531326 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31326 uint8x16_t __ret;
3132731327 __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3132831328 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3132931329 return __ret;
......@@ -31338,9 +31338,9 @@ __ai uint16x8_t vtstq_p16(poly16x8_t __p0, poly16x8_t __p1) {
3133831338}
3133931339#else
3134031340__ai uint16x8_t vtstq_p16(poly16x8_t __p0, poly16x8_t __p1) {
31341 uint16x8_t __ret;
3134131342 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3134231343 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31343 uint16x8_t __ret;
3134431344 __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
3134531345 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3134631346 return __ret;
......@@ -31355,9 +31355,9 @@ __ai uint8x16_t vtstq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3135531355}
3135631356#else
3135731357__ai uint8x16_t vtstq_u8(uint8x16_t __p0, uint8x16_t __p1) {
31358 uint8x16_t __ret;
3135831359 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3135931360 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31360 uint8x16_t __ret;
3136131361 __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3136231362 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3136331363 return __ret;
......@@ -31372,9 +31372,9 @@ __ai uint32x4_t vtstq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3137231372}
3137331373#else
3137431374__ai uint32x4_t vtstq_u32(uint32x4_t __p0, uint32x4_t __p1) {
31375 uint32x4_t __ret;
3137531376 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3137631377 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31377 uint32x4_t __ret;
3137831378 __ret = (uint32x4_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3137931379 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3138031380 return __ret;
......@@ -31389,9 +31389,9 @@ __ai uint16x8_t vtstq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3138931389}
3139031390#else
3139131391__ai uint16x8_t vtstq_u16(uint16x8_t __p0, uint16x8_t __p1) {
31392 uint16x8_t __ret;
3139231393 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3139331394 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31394 uint16x8_t __ret;
3139531395 __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
3139631396 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3139731397 return __ret;
......@@ -31406,9 +31406,9 @@ __ai uint8x16_t vtstq_s8(int8x16_t __p0, int8x16_t __p1) {
3140631406}
3140731407#else
3140831408__ai uint8x16_t vtstq_s8(int8x16_t __p0, int8x16_t __p1) {
31409 uint8x16_t __ret;
3140931410 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3141031411 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31411 uint8x16_t __ret;
3141231412 __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3141331413 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3141431414 return __ret;
......@@ -31423,9 +31423,9 @@ __ai uint32x4_t vtstq_s32(int32x4_t __p0, int32x4_t __p1) {
3142331423}
3142431424#else
3142531425__ai uint32x4_t vtstq_s32(int32x4_t __p0, int32x4_t __p1) {
31426 uint32x4_t __ret;
3142631427 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3142731428 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31428 uint32x4_t __ret;
3142931429 __ret = (uint32x4_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3143031430 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3143131431 return __ret;
......@@ -31440,9 +31440,9 @@ __ai uint16x8_t vtstq_s16(int16x8_t __p0, int16x8_t __p1) {
3144031440}
3144131441#else
3144231442__ai uint16x8_t vtstq_s16(int16x8_t __p0, int16x8_t __p1) {
31443 uint16x8_t __ret;
3144331444 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3144431445 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31445 uint16x8_t __ret;
3144631446 __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
3144731447 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3144831448 return __ret;
......@@ -31457,9 +31457,9 @@ __ai uint8x8_t vtst_u8(uint8x8_t __p0, uint8x8_t __p1) {
3145731457}
3145831458#else
3145931459__ai uint8x8_t vtst_u8(uint8x8_t __p0, uint8x8_t __p1) {
31460 uint8x8_t __ret;
3146031461 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3146131462 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31462 uint8x8_t __ret;
3146331463 __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3146431464 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3146531465 return __ret;
......@@ -31474,9 +31474,9 @@ __ai uint32x2_t vtst_u32(uint32x2_t __p0, uint32x2_t __p1) {
3147431474}
3147531475#else
3147631476__ai uint32x2_t vtst_u32(uint32x2_t __p0, uint32x2_t __p1) {
31477 uint32x2_t __ret;
3147731478 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3147831479 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31479 uint32x2_t __ret;
3148031480 __ret = (uint32x2_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
3148131481 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3148231482 return __ret;
......@@ -31491,9 +31491,9 @@ __ai uint16x4_t vtst_u16(uint16x4_t __p0, uint16x4_t __p1) {
3149131491}
3149231492#else
3149331493__ai uint16x4_t vtst_u16(uint16x4_t __p0, uint16x4_t __p1) {
31494 uint16x4_t __ret;
3149431495 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3149531496 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31496 uint16x4_t __ret;
3149731497 __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
3149831498 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3149931499 return __ret;
......@@ -31508,9 +31508,9 @@ __ai uint8x8_t vtst_s8(int8x8_t __p0, int8x8_t __p1) {
3150831508}
3150931509#else
3151031510__ai uint8x8_t vtst_s8(int8x8_t __p0, int8x8_t __p1) {
31511 uint8x8_t __ret;
3151131512 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3151231513 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31513 uint8x8_t __ret;
3151431514 __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3151531515 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3151631516 return __ret;
......@@ -31525,9 +31525,9 @@ __ai uint32x2_t vtst_s32(int32x2_t __p0, int32x2_t __p1) {
3152531525}
3152631526#else
3152731527__ai uint32x2_t vtst_s32(int32x2_t __p0, int32x2_t __p1) {
31528 uint32x2_t __ret;
3152831529 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3152931530 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31530 uint32x2_t __ret;
3153131531 __ret = (uint32x2_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
3153231532 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3153331533 return __ret;
......@@ -31542,9 +31542,9 @@ __ai uint16x4_t vtst_s16(int16x4_t __p0, int16x4_t __p1) {
3154231542}
3154331543#else
3154431544__ai uint16x4_t vtst_s16(int16x4_t __p0, int16x4_t __p1) {
31545 uint16x4_t __ret;
3154531546 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3154631547 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31547 uint16x4_t __ret;
3154831548 __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
3154931549 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3155031550 return __ret;
......@@ -31559,9 +31559,9 @@ __ai poly8x8x2_t vuzp_p8(poly8x8_t __p0, poly8x8_t __p1) {
3155931559}
3156031560#else
3156131561__ai poly8x8x2_t vuzp_p8(poly8x8_t __p0, poly8x8_t __p1) {
31562 poly8x8x2_t __ret;
3156231563 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3156331564 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31564 poly8x8x2_t __ret;
3156531565 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 4);
3156631566
3156731567 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31578,9 +31578,9 @@ __ai poly16x4x2_t vuzp_p16(poly16x4_t __p0, poly16x4_t __p1) {
3157831578}
3157931579#else
3158031580__ai poly16x4x2_t vuzp_p16(poly16x4_t __p0, poly16x4_t __p1) {
31581 poly16x4x2_t __ret;
3158131582 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3158231583 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31583 poly16x4x2_t __ret;
3158431584 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 5);
3158531585
3158631586 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31597,9 +31597,9 @@ __ai poly8x16x2_t vuzpq_p8(poly8x16_t __p0, poly8x16_t __p1) {
3159731597}
3159831598#else
3159931599__ai poly8x16x2_t vuzpq_p8(poly8x16_t __p0, poly8x16_t __p1) {
31600 poly8x16x2_t __ret;
3160031601 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3160131602 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31602 poly8x16x2_t __ret;
3160331603 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 36);
3160431604
3160531605 __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);
......@@ -31616,9 +31616,9 @@ __ai poly16x8x2_t vuzpq_p16(poly16x8_t __p0, poly16x8_t __p1) {
3161631616}
3161731617#else
3161831618__ai poly16x8x2_t vuzpq_p16(poly16x8_t __p0, poly16x8_t __p1) {
31619 poly16x8x2_t __ret;
3161931620 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3162031621 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31621 poly16x8x2_t __ret;
3162231622 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 37);
3162331623
3162431624 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31635,9 +31635,9 @@ __ai uint8x16x2_t vuzpq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3163531635}
3163631636#else
3163731637__ai uint8x16x2_t vuzpq_u8(uint8x16_t __p0, uint8x16_t __p1) {
31638 uint8x16x2_t __ret;
3163831639 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3163931640 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31640 uint8x16x2_t __ret;
3164131641 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3164231642
3164331643 __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);
......@@ -31654,9 +31654,9 @@ __ai uint32x4x2_t vuzpq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3165431654}
3165531655#else
3165631656__ai uint32x4x2_t vuzpq_u32(uint32x4_t __p0, uint32x4_t __p1) {
31657 uint32x4x2_t __ret;
3165731658 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3165831659 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31659 uint32x4x2_t __ret;
3166031660 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3166131661
3166231662 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31673,9 +31673,9 @@ __ai uint16x8x2_t vuzpq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3167331673}
3167431674#else
3167531675__ai uint16x8x2_t vuzpq_u16(uint16x8_t __p0, uint16x8_t __p1) {
31676 uint16x8x2_t __ret;
3167631677 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3167731678 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31678 uint16x8x2_t __ret;
3167931679 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 49);
3168031680
3168131681 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31692,9 +31692,9 @@ __ai int8x16x2_t vuzpq_s8(int8x16_t __p0, int8x16_t __p1) {
3169231692}
3169331693#else
3169431694__ai int8x16x2_t vuzpq_s8(int8x16_t __p0, int8x16_t __p1) {
31695 int8x16x2_t __ret;
3169531696 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3169631697 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31697 int8x16x2_t __ret;
3169831698 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 32);
3169931699
3170031700 __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);
......@@ -31711,9 +31711,9 @@ __ai float32x4x2_t vuzpq_f32(float32x4_t __p0, float32x4_t __p1) {
3171131711}
3171231712#else
3171331713__ai float32x4x2_t vuzpq_f32(float32x4_t __p0, float32x4_t __p1) {
31714 float32x4x2_t __ret;
3171431715 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3171531716 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31716 float32x4x2_t __ret;
3171731717 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 41);
3171831718
3171931719 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31730,9 +31730,9 @@ __ai int32x4x2_t vuzpq_s32(int32x4_t __p0, int32x4_t __p1) {
3173031730}
3173131731#else
3173231732__ai int32x4x2_t vuzpq_s32(int32x4_t __p0, int32x4_t __p1) {
31733 int32x4x2_t __ret;
3173331734 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3173431735 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31735 int32x4x2_t __ret;
3173631736 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 34);
3173731737
3173831738 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31749,9 +31749,9 @@ __ai int16x8x2_t vuzpq_s16(int16x8_t __p0, int16x8_t __p1) {
3174931749}
3175031750#else
3175131751__ai int16x8x2_t vuzpq_s16(int16x8_t __p0, int16x8_t __p1) {
31752 int16x8x2_t __ret;
3175231753 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3175331754 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31754 int16x8x2_t __ret;
3175531755 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 33);
3175631756
3175731757 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31768,9 +31768,9 @@ __ai uint8x8x2_t vuzp_u8(uint8x8_t __p0, uint8x8_t __p1) {
3176831768}
3176931769#else
3177031770__ai uint8x8x2_t vuzp_u8(uint8x8_t __p0, uint8x8_t __p1) {
31771 uint8x8x2_t __ret;
3177131772 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3177231773 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31773 uint8x8x2_t __ret;
3177431774 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3177531775
3177631776 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31787,9 +31787,9 @@ __ai uint32x2x2_t vuzp_u32(uint32x2_t __p0, uint32x2_t __p1) {
3178731787}
3178831788#else
3178931789__ai uint32x2x2_t vuzp_u32(uint32x2_t __p0, uint32x2_t __p1) {
31790 uint32x2x2_t __ret;
3179031791 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3179131792 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31792 uint32x2x2_t __ret;
3179331793 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 18);
3179431794
3179531795 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -31806,9 +31806,9 @@ __ai uint16x4x2_t vuzp_u16(uint16x4_t __p0, uint16x4_t __p1) {
3180631806}
3180731807#else
3180831808__ai uint16x4x2_t vuzp_u16(uint16x4_t __p0, uint16x4_t __p1) {
31809 uint16x4x2_t __ret;
3180931810 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3181031811 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31811 uint16x4x2_t __ret;
3181231812 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 17);
3181331813
3181431814 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31825,9 +31825,9 @@ __ai int8x8x2_t vuzp_s8(int8x8_t __p0, int8x8_t __p1) {
3182531825}
3182631826#else
3182731827__ai int8x8x2_t vuzp_s8(int8x8_t __p0, int8x8_t __p1) {
31828 int8x8x2_t __ret;
3182831829 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3182931830 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31830 int8x8x2_t __ret;
3183131831 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 0);
3183231832
3183331833 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31844,9 +31844,9 @@ __ai float32x2x2_t vuzp_f32(float32x2_t __p0, float32x2_t __p1) {
3184431844}
3184531845#else
3184631846__ai float32x2x2_t vuzp_f32(float32x2_t __p0, float32x2_t __p1) {
31847 float32x2x2_t __ret;
3184731848 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3184831849 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31849 float32x2x2_t __ret;
3185031850 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 9);
3185131851
3185231852 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -31863,9 +31863,9 @@ __ai int32x2x2_t vuzp_s32(int32x2_t __p0, int32x2_t __p1) {
3186331863}
3186431864#else
3186531865__ai int32x2x2_t vuzp_s32(int32x2_t __p0, int32x2_t __p1) {
31866 int32x2x2_t __ret;
3186631867 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3186731868 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
31868 int32x2x2_t __ret;
3186931869 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 2);
3187031870
3187131871 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -31882,9 +31882,9 @@ __ai int16x4x2_t vuzp_s16(int16x4_t __p0, int16x4_t __p1) {
3188231882}
3188331883#else
3188431884__ai int16x4x2_t vuzp_s16(int16x4_t __p0, int16x4_t __p1) {
31885 int16x4x2_t __ret;
3188531886 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3188631887 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31887 int16x4x2_t __ret;
3188831888 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 1);
3188931889
3189031890 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31901,9 +31901,9 @@ __ai poly8x8x2_t vzip_p8(poly8x8_t __p0, poly8x8_t __p1) {
3190131901}
3190231902#else
3190331903__ai poly8x8x2_t vzip_p8(poly8x8_t __p0, poly8x8_t __p1) {
31904 poly8x8x2_t __ret;
3190431905 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3190531906 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31906 poly8x8x2_t __ret;
3190731907 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 4);
3190831908
3190931909 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31920,9 +31920,9 @@ __ai poly16x4x2_t vzip_p16(poly16x4_t __p0, poly16x4_t __p1) {
3192031920}
3192131921#else
3192231922__ai poly16x4x2_t vzip_p16(poly16x4_t __p0, poly16x4_t __p1) {
31923 poly16x4x2_t __ret;
3192331924 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3192431925 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
31925 poly16x4x2_t __ret;
3192631926 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 5);
3192731927
3192831928 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -31939,9 +31939,9 @@ __ai poly8x16x2_t vzipq_p8(poly8x16_t __p0, poly8x16_t __p1) {
3193931939}
3194031940#else
3194131941__ai poly8x16x2_t vzipq_p8(poly8x16_t __p0, poly8x16_t __p1) {
31942 poly8x16x2_t __ret;
3194231943 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3194331944 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31944 poly8x16x2_t __ret;
3194531945 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 36);
3194631946
3194731947 __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);
......@@ -31958,9 +31958,9 @@ __ai poly16x8x2_t vzipq_p16(poly16x8_t __p0, poly16x8_t __p1) {
3195831958}
3195931959#else
3196031960__ai poly16x8x2_t vzipq_p16(poly16x8_t __p0, poly16x8_t __p1) {
31961 poly16x8x2_t __ret;
3196131962 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3196231963 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
31963 poly16x8x2_t __ret;
3196431964 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 37);
3196531965
3196631966 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -31977,9 +31977,9 @@ __ai uint8x16x2_t vzipq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3197731977}
3197831978#else
3197931979__ai uint8x16x2_t vzipq_u8(uint8x16_t __p0, uint8x16_t __p1) {
31980 uint8x16x2_t __ret;
3198031981 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3198131982 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
31982 uint8x16x2_t __ret;
3198331983 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3198431984
3198531985 __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);
......@@ -31996,9 +31996,9 @@ __ai uint32x4x2_t vzipq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3199631996}
3199731997#else
3199831998__ai uint32x4x2_t vzipq_u32(uint32x4_t __p0, uint32x4_t __p1) {
31999 uint32x4x2_t __ret;
3199932000 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3200032001 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
32001 uint32x4x2_t __ret;
3200232002 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3200332003
3200432004 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -32015,9 +32015,9 @@ __ai uint16x8x2_t vzipq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3201532015}
3201632016#else
3201732017__ai uint16x8x2_t vzipq_u16(uint16x8_t __p0, uint16x8_t __p1) {
32018 uint16x8x2_t __ret;
3201832019 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3201932020 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
32020 uint16x8x2_t __ret;
3202132021 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 49);
3202232022
3202332023 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -32034,9 +32034,9 @@ __ai int8x16x2_t vzipq_s8(int8x16_t __p0, int8x16_t __p1) {
3203432034}
3203532035#else
3203632036__ai int8x16x2_t vzipq_s8(int8x16_t __p0, int8x16_t __p1) {
32037 int8x16x2_t __ret;
3203732038 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3203832039 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
32039 int8x16x2_t __ret;
3204032040 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 32);
3204132041
3204232042 __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);
......@@ -32053,9 +32053,9 @@ __ai float32x4x2_t vzipq_f32(float32x4_t __p0, float32x4_t __p1) {
3205332053}
3205432054#else
3205532055__ai float32x4x2_t vzipq_f32(float32x4_t __p0, float32x4_t __p1) {
32056 float32x4x2_t __ret;
3205632057 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3205732058 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
32058 float32x4x2_t __ret;
3205932059 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 41);
3206032060
3206132061 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -32072,9 +32072,9 @@ __ai int32x4x2_t vzipq_s32(int32x4_t __p0, int32x4_t __p1) {
3207232072}
3207332073#else
3207432074__ai int32x4x2_t vzipq_s32(int32x4_t __p0, int32x4_t __p1) {
32075 int32x4x2_t __ret;
3207532076 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3207632077 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
32077 int32x4x2_t __ret;
3207832078 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 34);
3207932079
3208032080 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -32091,9 +32091,9 @@ __ai int16x8x2_t vzipq_s16(int16x8_t __p0, int16x8_t __p1) {
3209132091}
3209232092#else
3209332093__ai int16x8x2_t vzipq_s16(int16x8_t __p0, int16x8_t __p1) {
32094 int16x8x2_t __ret;
3209432095 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3209532096 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
32096 int16x8x2_t __ret;
3209732097 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 33);
3209832098
3209932099 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -32110,9 +32110,9 @@ __ai uint8x8x2_t vzip_u8(uint8x8_t __p0, uint8x8_t __p1) {
3211032110}
3211132111#else
3211232112__ai uint8x8x2_t vzip_u8(uint8x8_t __p0, uint8x8_t __p1) {
32113 uint8x8x2_t __ret;
3211332114 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3211432115 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
32115 uint8x8x2_t __ret;
3211632116 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 16);
3211732117
3211832118 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -32129,9 +32129,9 @@ __ai uint32x2x2_t vzip_u32(uint32x2_t __p0, uint32x2_t __p1) {
3212932129}
3213032130#else
3213132131__ai uint32x2x2_t vzip_u32(uint32x2_t __p0, uint32x2_t __p1) {
32132 uint32x2x2_t __ret;
3213232133 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3213332134 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
32134 uint32x2x2_t __ret;
3213532135 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 18);
3213632136
3213732137 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -32148,9 +32148,9 @@ __ai uint16x4x2_t vzip_u16(uint16x4_t __p0, uint16x4_t __p1) {
3214832148}
3214932149#else
3215032150__ai uint16x4x2_t vzip_u16(uint16x4_t __p0, uint16x4_t __p1) {
32151 uint16x4x2_t __ret;
3215132152 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3215232153 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
32153 uint16x4x2_t __ret;
3215432154 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 17);
3215532155
3215632156 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -32167,9 +32167,9 @@ __ai int8x8x2_t vzip_s8(int8x8_t __p0, int8x8_t __p1) {
3216732167}
3216832168#else
3216932169__ai int8x8x2_t vzip_s8(int8x8_t __p0, int8x8_t __p1) {
32170 int8x8x2_t __ret;
3217032171 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3217132172 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
32172 int8x8x2_t __ret;
3217332173 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 0);
3217432174
3217532175 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -32186,9 +32186,9 @@ __ai float32x2x2_t vzip_f32(float32x2_t __p0, float32x2_t __p1) {
3218632186}
3218732187#else
3218832188__ai float32x2x2_t vzip_f32(float32x2_t __p0, float32x2_t __p1) {
32189 float32x2x2_t __ret;
3218932190 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3219032191 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
32191 float32x2x2_t __ret;
3219232192 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 9);
3219332193
3219432194 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -32205,9 +32205,9 @@ __ai int32x2x2_t vzip_s32(int32x2_t __p0, int32x2_t __p1) {
3220532205}
3220632206#else
3220732207__ai int32x2x2_t vzip_s32(int32x2_t __p0, int32x2_t __p1) {
32208 int32x2x2_t __ret;
3220832209 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3220932210 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
32210 int32x2x2_t __ret;
3221132211 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 2);
3221232212
3221332213 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0);
......@@ -32224,9 +32224,9 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3222432224}
3222532225#else
3222632226__ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
32227 int16x4x2_t __ret;
3222732228 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3222832229 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
32229 int16x4x2_t __ret;
3223032230 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 1);
3223132231
3223232232 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -32238,16 +32238,16 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3223832238#if !defined(__aarch64__)
3223932239#ifdef __LITTLE_ENDIAN__
3224032240#define vdupq_lane_f16(__p0_122, __p1_122) __extension__ ({ \
32241 float16x4_t __s0_122 = __p0_122; \
3224232241 float16x8_t __ret_122; \
32242 float16x4_t __s0_122 = __p0_122; \
3224332243 __ret_122 = splatq_lane_f16(__s0_122, __p1_122); \
3224432244 __ret_122; \
3224532245})
3224632246#else
3224732247#define vdupq_lane_f16(__p0_123, __p1_123) __extension__ ({ \
32248 float16x8_t __ret_123; \
3224832249 float16x4_t __s0_123 = __p0_123; \
3224932250 float16x4_t __rev0_123; __rev0_123 = __builtin_shufflevector(__s0_123, __s0_123, 3, 2, 1, 0); \
32250 float16x8_t __ret_123; \
3225132251 __ret_123 = __noswap_splatq_lane_f16(__rev0_123, __p1_123); \
3225232252 __ret_123 = __builtin_shufflevector(__ret_123, __ret_123, 7, 6, 5, 4, 3, 2, 1, 0); \
3225332253 __ret_123; \
......@@ -32256,16 +32256,16 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3225632256
3225732257#ifdef __LITTLE_ENDIAN__
3225832258#define vdup_lane_f16(__p0_124, __p1_124) __extension__ ({ \
32259 float16x4_t __s0_124 = __p0_124; \
3226032259 float16x4_t __ret_124; \
32260 float16x4_t __s0_124 = __p0_124; \
3226132261 __ret_124 = splat_lane_f16(__s0_124, __p1_124); \
3226232262 __ret_124; \
3226332263})
3226432264#else
3226532265#define vdup_lane_f16(__p0_125, __p1_125) __extension__ ({ \
32266 float16x4_t __ret_125; \
3226632267 float16x4_t __s0_125 = __p0_125; \
3226732268 float16x4_t __rev0_125; __rev0_125 = __builtin_shufflevector(__s0_125, __s0_125, 3, 2, 1, 0); \
32268 float16x4_t __ret_125; \
3226932269 __ret_125 = __noswap_splat_lane_f16(__rev0_125, __p1_125); \
3227032270 __ret_125 = __builtin_shufflevector(__ret_125, __ret_125, 3, 2, 1, 0); \
3227132271 __ret_125; \
......@@ -32274,15 +32274,15 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3227432274
3227532275#ifdef __LITTLE_ENDIAN__
3227632276#define vdupq_n_f16(__p0) __extension__ ({ \
32277 float16_t __s0 = __p0; \
3227832277 float16x8_t __ret; \
32278 float16_t __s0 = __p0; \
3227932279 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
3228032280 __ret; \
3228132281})
3228232282#else
3228332283#define vdupq_n_f16(__p0) __extension__ ({ \
32284 float16_t __s0 = __p0; \
3228532284 float16x8_t __ret; \
32285 float16_t __s0 = __p0; \
3228632286 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
3228732287 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
3228832288 __ret; \
......@@ -32291,15 +32291,15 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3229132291
3229232292#ifdef __LITTLE_ENDIAN__
3229332293#define vdup_n_f16(__p0) __extension__ ({ \
32294 float16_t __s0 = __p0; \
3229532294 float16x4_t __ret; \
32295 float16_t __s0 = __p0; \
3229632296 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
3229732297 __ret; \
3229832298})
3229932299#else
3230032300#define vdup_n_f16(__p0) __extension__ ({ \
32301 float16_t __s0 = __p0; \
3230232301 float16x4_t __ret; \
32302 float16_t __s0 = __p0; \
3230332303 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
3230432304 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3230532305 __ret; \
......@@ -32308,15 +32308,15 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3230832308
3230932309#ifdef __LITTLE_ENDIAN__
3231032310#define vmovq_n_f16(__p0) __extension__ ({ \
32311 float16_t __s0 = __p0; \
3231232311 float16x8_t __ret; \
32312 float16_t __s0 = __p0; \
3231332313 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
3231432314 __ret; \
3231532315})
3231632316#else
3231732317#define vmovq_n_f16(__p0) __extension__ ({ \
32318 float16_t __s0 = __p0; \
3231932318 float16x8_t __ret; \
32319 float16_t __s0 = __p0; \
3232032320 __ret = (float16x8_t) {__s0, __s0, __s0, __s0, __s0, __s0, __s0, __s0}; \
3232132321 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
3232232322 __ret; \
......@@ -32325,15 +32325,15 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3232532325
3232632326#ifdef __LITTLE_ENDIAN__
3232732327#define vmov_n_f16(__p0) __extension__ ({ \
32328 float16_t __s0 = __p0; \
3232932328 float16x4_t __ret; \
32329 float16_t __s0 = __p0; \
3233032330 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
3233132331 __ret; \
3233232332})
3233332333#else
3233432334#define vmov_n_f16(__p0) __extension__ ({ \
32335 float16_t __s0 = __p0; \
3233632335 float16x4_t __ret; \
32336 float16_t __s0 = __p0; \
3233732337 __ret = (float16x4_t) {__s0, __s0, __s0, __s0}; \
3233832338 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3233932339 __ret; \
......@@ -32342,19 +32342,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3234232342
3234332343#ifdef __LITTLE_ENDIAN__
3234432344#define vqdmulhq_lane_s32(__p0_126, __p1_126, __p2_126) __extension__ ({ \
32345 int32x4_t __ret_126; \
3234532346 int32x4_t __s0_126 = __p0_126; \
3234632347 int32x2_t __s1_126 = __p1_126; \
32347 int32x4_t __ret_126; \
3234832348 __ret_126 = vqdmulhq_s32(__s0_126, splatq_lane_s32(__s1_126, __p2_126)); \
3234932349 __ret_126; \
3235032350})
3235132351#else
3235232352#define vqdmulhq_lane_s32(__p0_127, __p1_127, __p2_127) __extension__ ({ \
32353 int32x4_t __ret_127; \
3235332354 int32x4_t __s0_127 = __p0_127; \
3235432355 int32x2_t __s1_127 = __p1_127; \
3235532356 int32x4_t __rev0_127; __rev0_127 = __builtin_shufflevector(__s0_127, __s0_127, 3, 2, 1, 0); \
3235632357 int32x2_t __rev1_127; __rev1_127 = __builtin_shufflevector(__s1_127, __s1_127, 1, 0); \
32357 int32x4_t __ret_127; \
3235832358 __ret_127 = __noswap_vqdmulhq_s32(__rev0_127, __noswap_splatq_lane_s32(__rev1_127, __p2_127)); \
3235932359 __ret_127 = __builtin_shufflevector(__ret_127, __ret_127, 3, 2, 1, 0); \
3236032360 __ret_127; \
......@@ -32363,19 +32363,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3236332363
3236432364#ifdef __LITTLE_ENDIAN__
3236532365#define vqdmulhq_lane_s16(__p0_128, __p1_128, __p2_128) __extension__ ({ \
32366 int16x8_t __ret_128; \
3236632367 int16x8_t __s0_128 = __p0_128; \
3236732368 int16x4_t __s1_128 = __p1_128; \
32368 int16x8_t __ret_128; \
3236932369 __ret_128 = vqdmulhq_s16(__s0_128, splatq_lane_s16(__s1_128, __p2_128)); \
3237032370 __ret_128; \
3237132371})
3237232372#else
3237332373#define vqdmulhq_lane_s16(__p0_129, __p1_129, __p2_129) __extension__ ({ \
32374 int16x8_t __ret_129; \
3237432375 int16x8_t __s0_129 = __p0_129; \
3237532376 int16x4_t __s1_129 = __p1_129; \
3237632377 int16x8_t __rev0_129; __rev0_129 = __builtin_shufflevector(__s0_129, __s0_129, 7, 6, 5, 4, 3, 2, 1, 0); \
3237732378 int16x4_t __rev1_129; __rev1_129 = __builtin_shufflevector(__s1_129, __s1_129, 3, 2, 1, 0); \
32378 int16x8_t __ret_129; \
3237932379 __ret_129 = __noswap_vqdmulhq_s16(__rev0_129, __noswap_splatq_lane_s16(__rev1_129, __p2_129)); \
3238032380 __ret_129 = __builtin_shufflevector(__ret_129, __ret_129, 7, 6, 5, 4, 3, 2, 1, 0); \
3238132381 __ret_129; \
......@@ -32384,19 +32384,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3238432384
3238532385#ifdef __LITTLE_ENDIAN__
3238632386#define vqdmulh_lane_s32(__p0_130, __p1_130, __p2_130) __extension__ ({ \
32387 int32x2_t __ret_130; \
3238732388 int32x2_t __s0_130 = __p0_130; \
3238832389 int32x2_t __s1_130 = __p1_130; \
32389 int32x2_t __ret_130; \
3239032390 __ret_130 = vqdmulh_s32(__s0_130, splat_lane_s32(__s1_130, __p2_130)); \
3239132391 __ret_130; \
3239232392})
3239332393#else
3239432394#define vqdmulh_lane_s32(__p0_131, __p1_131, __p2_131) __extension__ ({ \
32395 int32x2_t __ret_131; \
3239532396 int32x2_t __s0_131 = __p0_131; \
3239632397 int32x2_t __s1_131 = __p1_131; \
3239732398 int32x2_t __rev0_131; __rev0_131 = __builtin_shufflevector(__s0_131, __s0_131, 1, 0); \
3239832399 int32x2_t __rev1_131; __rev1_131 = __builtin_shufflevector(__s1_131, __s1_131, 1, 0); \
32399 int32x2_t __ret_131; \
3240032400 __ret_131 = __noswap_vqdmulh_s32(__rev0_131, __noswap_splat_lane_s32(__rev1_131, __p2_131)); \
3240132401 __ret_131 = __builtin_shufflevector(__ret_131, __ret_131, 1, 0); \
3240232402 __ret_131; \
......@@ -32405,19 +32405,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3240532405
3240632406#ifdef __LITTLE_ENDIAN__
3240732407#define vqdmulh_lane_s16(__p0_132, __p1_132, __p2_132) __extension__ ({ \
32408 int16x4_t __ret_132; \
3240832409 int16x4_t __s0_132 = __p0_132; \
3240932410 int16x4_t __s1_132 = __p1_132; \
32410 int16x4_t __ret_132; \
3241132411 __ret_132 = vqdmulh_s16(__s0_132, splat_lane_s16(__s1_132, __p2_132)); \
3241232412 __ret_132; \
3241332413})
3241432414#else
3241532415#define vqdmulh_lane_s16(__p0_133, __p1_133, __p2_133) __extension__ ({ \
32416 int16x4_t __ret_133; \
3241632417 int16x4_t __s0_133 = __p0_133; \
3241732418 int16x4_t __s1_133 = __p1_133; \
3241832419 int16x4_t __rev0_133; __rev0_133 = __builtin_shufflevector(__s0_133, __s0_133, 3, 2, 1, 0); \
3241932420 int16x4_t __rev1_133; __rev1_133 = __builtin_shufflevector(__s1_133, __s1_133, 3, 2, 1, 0); \
32420 int16x4_t __ret_133; \
3242132421 __ret_133 = __noswap_vqdmulh_s16(__rev0_133, __noswap_splat_lane_s16(__rev1_133, __p2_133)); \
3242232422 __ret_133 = __builtin_shufflevector(__ret_133, __ret_133, 3, 2, 1, 0); \
3242332423 __ret_133; \
......@@ -32426,19 +32426,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3242632426
3242732427#ifdef __LITTLE_ENDIAN__
3242832428#define vqrdmulhq_lane_s32(__p0_134, __p1_134, __p2_134) __extension__ ({ \
32429 int32x4_t __ret_134; \
3242932430 int32x4_t __s0_134 = __p0_134; \
3243032431 int32x2_t __s1_134 = __p1_134; \
32431 int32x4_t __ret_134; \
3243232432 __ret_134 = vqrdmulhq_s32(__s0_134, splatq_lane_s32(__s1_134, __p2_134)); \
3243332433 __ret_134; \
3243432434})
3243532435#else
3243632436#define vqrdmulhq_lane_s32(__p0_135, __p1_135, __p2_135) __extension__ ({ \
32437 int32x4_t __ret_135; \
3243732438 int32x4_t __s0_135 = __p0_135; \
3243832439 int32x2_t __s1_135 = __p1_135; \
3243932440 int32x4_t __rev0_135; __rev0_135 = __builtin_shufflevector(__s0_135, __s0_135, 3, 2, 1, 0); \
3244032441 int32x2_t __rev1_135; __rev1_135 = __builtin_shufflevector(__s1_135, __s1_135, 1, 0); \
32441 int32x4_t __ret_135; \
3244232442 __ret_135 = __noswap_vqrdmulhq_s32(__rev0_135, __noswap_splatq_lane_s32(__rev1_135, __p2_135)); \
3244332443 __ret_135 = __builtin_shufflevector(__ret_135, __ret_135, 3, 2, 1, 0); \
3244432444 __ret_135; \
......@@ -32447,19 +32447,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3244732447
3244832448#ifdef __LITTLE_ENDIAN__
3244932449#define vqrdmulhq_lane_s16(__p0_136, __p1_136, __p2_136) __extension__ ({ \
32450 int16x8_t __ret_136; \
3245032451 int16x8_t __s0_136 = __p0_136; \
3245132452 int16x4_t __s1_136 = __p1_136; \
32452 int16x8_t __ret_136; \
3245332453 __ret_136 = vqrdmulhq_s16(__s0_136, splatq_lane_s16(__s1_136, __p2_136)); \
3245432454 __ret_136; \
3245532455})
3245632456#else
3245732457#define vqrdmulhq_lane_s16(__p0_137, __p1_137, __p2_137) __extension__ ({ \
32458 int16x8_t __ret_137; \
3245832459 int16x8_t __s0_137 = __p0_137; \
3245932460 int16x4_t __s1_137 = __p1_137; \
3246032461 int16x8_t __rev0_137; __rev0_137 = __builtin_shufflevector(__s0_137, __s0_137, 7, 6, 5, 4, 3, 2, 1, 0); \
3246132462 int16x4_t __rev1_137; __rev1_137 = __builtin_shufflevector(__s1_137, __s1_137, 3, 2, 1, 0); \
32462 int16x8_t __ret_137; \
3246332463 __ret_137 = __noswap_vqrdmulhq_s16(__rev0_137, __noswap_splatq_lane_s16(__rev1_137, __p2_137)); \
3246432464 __ret_137 = __builtin_shufflevector(__ret_137, __ret_137, 7, 6, 5, 4, 3, 2, 1, 0); \
3246532465 __ret_137; \
......@@ -32468,19 +32468,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3246832468
3246932469#ifdef __LITTLE_ENDIAN__
3247032470#define vqrdmulh_lane_s32(__p0_138, __p1_138, __p2_138) __extension__ ({ \
32471 int32x2_t __ret_138; \
3247132472 int32x2_t __s0_138 = __p0_138; \
3247232473 int32x2_t __s1_138 = __p1_138; \
32473 int32x2_t __ret_138; \
3247432474 __ret_138 = vqrdmulh_s32(__s0_138, splat_lane_s32(__s1_138, __p2_138)); \
3247532475 __ret_138; \
3247632476})
3247732477#else
3247832478#define vqrdmulh_lane_s32(__p0_139, __p1_139, __p2_139) __extension__ ({ \
32479 int32x2_t __ret_139; \
3247932480 int32x2_t __s0_139 = __p0_139; \
3248032481 int32x2_t __s1_139 = __p1_139; \
3248132482 int32x2_t __rev0_139; __rev0_139 = __builtin_shufflevector(__s0_139, __s0_139, 1, 0); \
3248232483 int32x2_t __rev1_139; __rev1_139 = __builtin_shufflevector(__s1_139, __s1_139, 1, 0); \
32483 int32x2_t __ret_139; \
3248432484 __ret_139 = __noswap_vqrdmulh_s32(__rev0_139, __noswap_splat_lane_s32(__rev1_139, __p2_139)); \
3248532485 __ret_139 = __builtin_shufflevector(__ret_139, __ret_139, 1, 0); \
3248632486 __ret_139; \
......@@ -32489,19 +32489,19 @@ __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
3248932489
3249032490#ifdef __LITTLE_ENDIAN__
3249132491#define vqrdmulh_lane_s16(__p0_140, __p1_140, __p2_140) __extension__ ({ \
32492 int16x4_t __ret_140; \
3249232493 int16x4_t __s0_140 = __p0_140; \
3249332494 int16x4_t __s1_140 = __p1_140; \
32494 int16x4_t __ret_140; \
3249532495 __ret_140 = vqrdmulh_s16(__s0_140, splat_lane_s16(__s1_140, __p2_140)); \
3249632496 __ret_140; \
3249732497})
3249832498#else
3249932499#define vqrdmulh_lane_s16(__p0_141, __p1_141, __p2_141) __extension__ ({ \
32500 int16x4_t __ret_141; \
3250032501 int16x4_t __s0_141 = __p0_141; \
3250132502 int16x4_t __s1_141 = __p1_141; \
3250232503 int16x4_t __rev0_141; __rev0_141 = __builtin_shufflevector(__s0_141, __s0_141, 3, 2, 1, 0); \
3250332504 int16x4_t __rev1_141; __rev1_141 = __builtin_shufflevector(__s1_141, __s1_141, 3, 2, 1, 0); \
32504 int16x4_t __ret_141; \
3250532505 __ret_141 = __noswap_vqrdmulh_s16(__rev0_141, __noswap_splat_lane_s16(__rev1_141, __p2_141)); \
3250632506 __ret_141 = __builtin_shufflevector(__ret_141, __ret_141, 3, 2, 1, 0); \
3250732507 __ret_141; \
......@@ -33838,8 +33838,8 @@ __ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
3383833838}
3383933839#else
3384033840__ai float16x4_t vcvt_f16_f32(float32x4_t __p0) {
33841 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3384233841 float16x4_t __ret;
33842 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3384333843 __ret = (float16x4_t) __builtin_neon_vcvt_f16_f32((int8x16_t)__rev0, 41);
3384433844 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3384533845 return __ret;
......@@ -33859,8 +33859,8 @@ __ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
3385933859}
3386033860#else
3386133861__ai float32x4_t vcvt_f32_f16(float16x4_t __p0) {
33862 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3386333862 float32x4_t __ret;
33863 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3386433864 __ret = (float32x4_t) __builtin_neon_vcvt_f32_f16((int8x8_t)__rev0, 8);
3386533865 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3386633866 return __ret;
......@@ -33934,16 +33934,16 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3393433934
3393533935#ifdef __LITTLE_ENDIAN__
3393633936#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33937 float16x8_t __s1 = __p1; \
3393833937 float16x8_t __ret; \
33938 float16x8_t __s1 = __p1; \
3393933939 __ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 40); \
3394033940 __ret; \
3394133941})
3394233942#else
3394333943#define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33944 float16x8_t __ret; \
3394433945 float16x8_t __s1 = __p1; \
3394533946 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
33946 float16x8_t __ret; \
3394733947 __ret = (float16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 40); \
3394833948 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
3394933949 __ret; \
......@@ -33952,16 +33952,16 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3395233952
3395333953#ifdef __LITTLE_ENDIAN__
3395433954#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33955 float16x4_t __s1 = __p1; \
3395633955 float16x4_t __ret; \
33956 float16x4_t __s1 = __p1; \
3395733957 __ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 8); \
3395833958 __ret; \
3395933959})
3396033960#else
3396133961#define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
33962 float16x4_t __ret; \
3396233963 float16x4_t __s1 = __p1; \
3396333964 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
33964 float16x4_t __ret; \
3396533965 __ret = (float16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 8); \
3396633966 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3396733967 __ret; \
......@@ -34146,18 +34146,18 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3414634146
3414734147#ifdef __LITTLE_ENDIAN__
3414834148#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34149 float16x8x2_t __s1 = __p1; \
3415034149 float16x8x2_t __ret; \
34150 float16x8x2_t __s1 = __p1; \
3415134151 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 40); \
3415234152 __ret; \
3415334153})
3415434154#else
3415534155#define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34156 float16x8x2_t __ret; \
3415634157 float16x8x2_t __s1 = __p1; \
3415734158 float16x8x2_t __rev1; \
3415834159 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
3415934160 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
34160 float16x8x2_t __ret; \
3416134161 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 40); \
3416234162 \
3416334163 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -34168,18 +34168,18 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3416834168
3416934169#ifdef __LITTLE_ENDIAN__
3417034170#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34171 float16x4x2_t __s1 = __p1; \
3417234171 float16x4x2_t __ret; \
34172 float16x4x2_t __s1 = __p1; \
3417334173 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 8); \
3417434174 __ret; \
3417534175})
3417634176#else
3417734177#define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34178 float16x4x2_t __ret; \
3417834179 float16x4x2_t __s1 = __p1; \
3417934180 float16x4x2_t __rev1; \
3418034181 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
3418134182 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
34182 float16x4x2_t __ret; \
3418334183 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 8); \
3418434184 \
3418534185 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -34262,19 +34262,19 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3426234262
3426334263#ifdef __LITTLE_ENDIAN__
3426434264#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34265 float16x8x3_t __s1 = __p1; \
3426634265 float16x8x3_t __ret; \
34266 float16x8x3_t __s1 = __p1; \
3426734267 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 40); \
3426834268 __ret; \
3426934269})
3427034270#else
3427134271#define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34272 float16x8x3_t __ret; \
3427234273 float16x8x3_t __s1 = __p1; \
3427334274 float16x8x3_t __rev1; \
3427434275 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
3427534276 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
3427634277 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
34277 float16x8x3_t __ret; \
3427834278 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 40); \
3427934279 \
3428034280 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -34286,19 +34286,19 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3428634286
3428734287#ifdef __LITTLE_ENDIAN__
3428834288#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34289 float16x4x3_t __s1 = __p1; \
3429034289 float16x4x3_t __ret; \
34290 float16x4x3_t __s1 = __p1; \
3429134291 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 8); \
3429234292 __ret; \
3429334293})
3429434294#else
3429534295#define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34296 float16x4x3_t __ret; \
3429634297 float16x4x3_t __s1 = __p1; \
3429734298 float16x4x3_t __rev1; \
3429834299 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
3429934300 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
3430034301 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
34301 float16x4x3_t __ret; \
3430234302 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 8); \
3430334303 \
3430434304 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -34386,20 +34386,20 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3438634386
3438734387#ifdef __LITTLE_ENDIAN__
3438834388#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34389 float16x8x4_t __s1 = __p1; \
3439034389 float16x8x4_t __ret; \
34390 float16x8x4_t __s1 = __p1; \
3439134391 __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); \
3439234392 __ret; \
3439334393})
3439434394#else
3439534395#define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34396 float16x8x4_t __ret; \
3439634397 float16x8x4_t __s1 = __p1; \
3439734398 float16x8x4_t __rev1; \
3439834399 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
3439934400 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
3440034401 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
3440134402 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
34402 float16x8x4_t __ret; \
3440334403 __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); \
3440434404 \
3440534405 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -34412,20 +34412,20 @@ __ai float32x4_t __noswap_vcvt_f32_f16(float16x4_t __p0) {
3441234412
3441334413#ifdef __LITTLE_ENDIAN__
3441434414#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34415 float16x4x4_t __s1 = __p1; \
3441634415 float16x4x4_t __ret; \
34416 float16x4x4_t __s1 = __p1; \
3441734417 __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); \
3441834418 __ret; \
3441934419})
3442034420#else
3442134421#define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
34422 float16x4x4_t __ret; \
3442234423 float16x4x4_t __s1 = __p1; \
3442334424 float16x4x4_t __rev1; \
3442434425 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
3442534426 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
3442634427 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
3442734428 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
34428 float16x4x4_t __ret; \
3442934429 __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); \
3443034430 \
3443134431 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -34786,8 +34786,8 @@ __ai int32x4_t vcvtaq_s32_f32(float32x4_t __p0) {
3478634786}
3478734787#else
3478834788__ai int32x4_t vcvtaq_s32_f32(float32x4_t __p0) {
34789 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3479034789 int32x4_t __ret;
34790 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3479134791 __ret = (int32x4_t) __builtin_neon_vcvtaq_s32_v((int8x16_t)__rev0, 34);
3479234792 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3479334793 return __ret;
......@@ -34802,8 +34802,8 @@ __ai int32x2_t vcvta_s32_f32(float32x2_t __p0) {
3480234802}
3480334803#else
3480434804__ai int32x2_t vcvta_s32_f32(float32x2_t __p0) {
34805 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3480634805 int32x2_t __ret;
34806 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3480734807 __ret = (int32x2_t) __builtin_neon_vcvta_s32_v((int8x8_t)__rev0, 2);
3480834808 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3480934809 return __ret;
......@@ -34818,8 +34818,8 @@ __ai uint32x4_t vcvtaq_u32_f32(float32x4_t __p0) {
3481834818}
3481934819#else
3482034820__ai uint32x4_t vcvtaq_u32_f32(float32x4_t __p0) {
34821 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3482234821 uint32x4_t __ret;
34822 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3482334823 __ret = (uint32x4_t) __builtin_neon_vcvtaq_u32_v((int8x16_t)__rev0, 50);
3482434824 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3482534825 return __ret;
......@@ -34834,8 +34834,8 @@ __ai uint32x2_t vcvta_u32_f32(float32x2_t __p0) {
3483434834}
3483534835#else
3483634836__ai uint32x2_t vcvta_u32_f32(float32x2_t __p0) {
34837 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3483834837 uint32x2_t __ret;
34838 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3483934839 __ret = (uint32x2_t) __builtin_neon_vcvta_u32_v((int8x8_t)__rev0, 18);
3484034840 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3484134841 return __ret;
......@@ -34850,8 +34850,8 @@ __ai int32x4_t vcvtmq_s32_f32(float32x4_t __p0) {
3485034850}
3485134851#else
3485234852__ai int32x4_t vcvtmq_s32_f32(float32x4_t __p0) {
34853 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3485434853 int32x4_t __ret;
34854 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3485534855 __ret = (int32x4_t) __builtin_neon_vcvtmq_s32_v((int8x16_t)__rev0, 34);
3485634856 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3485734857 return __ret;
......@@ -34866,8 +34866,8 @@ __ai int32x2_t vcvtm_s32_f32(float32x2_t __p0) {
3486634866}
3486734867#else
3486834868__ai int32x2_t vcvtm_s32_f32(float32x2_t __p0) {
34869 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3487034869 int32x2_t __ret;
34870 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3487134871 __ret = (int32x2_t) __builtin_neon_vcvtm_s32_v((int8x8_t)__rev0, 2);
3487234872 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3487334873 return __ret;
......@@ -34882,8 +34882,8 @@ __ai uint32x4_t vcvtmq_u32_f32(float32x4_t __p0) {
3488234882}
3488334883#else
3488434884__ai uint32x4_t vcvtmq_u32_f32(float32x4_t __p0) {
34885 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3488634885 uint32x4_t __ret;
34886 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3488734887 __ret = (uint32x4_t) __builtin_neon_vcvtmq_u32_v((int8x16_t)__rev0, 50);
3488834888 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3488934889 return __ret;
......@@ -34898,8 +34898,8 @@ __ai uint32x2_t vcvtm_u32_f32(float32x2_t __p0) {
3489834898}
3489934899#else
3490034900__ai uint32x2_t vcvtm_u32_f32(float32x2_t __p0) {
34901 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3490234901 uint32x2_t __ret;
34902 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3490334903 __ret = (uint32x2_t) __builtin_neon_vcvtm_u32_v((int8x8_t)__rev0, 18);
3490434904 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3490534905 return __ret;
......@@ -34914,8 +34914,8 @@ __ai int32x4_t vcvtnq_s32_f32(float32x4_t __p0) {
3491434914}
3491534915#else
3491634916__ai int32x4_t vcvtnq_s32_f32(float32x4_t __p0) {
34917 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3491834917 int32x4_t __ret;
34918 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3491934919 __ret = (int32x4_t) __builtin_neon_vcvtnq_s32_v((int8x16_t)__rev0, 34);
3492034920 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3492134921 return __ret;
......@@ -34930,8 +34930,8 @@ __ai int32x2_t vcvtn_s32_f32(float32x2_t __p0) {
3493034930}
3493134931#else
3493234932__ai int32x2_t vcvtn_s32_f32(float32x2_t __p0) {
34933 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3493434933 int32x2_t __ret;
34934 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3493534935 __ret = (int32x2_t) __builtin_neon_vcvtn_s32_v((int8x8_t)__rev0, 2);
3493634936 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3493734937 return __ret;
......@@ -34946,8 +34946,8 @@ __ai uint32x4_t vcvtnq_u32_f32(float32x4_t __p0) {
3494634946}
3494734947#else
3494834948__ai uint32x4_t vcvtnq_u32_f32(float32x4_t __p0) {
34949 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3495034949 uint32x4_t __ret;
34950 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3495134951 __ret = (uint32x4_t) __builtin_neon_vcvtnq_u32_v((int8x16_t)__rev0, 50);
3495234952 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3495334953 return __ret;
......@@ -34962,8 +34962,8 @@ __ai uint32x2_t vcvtn_u32_f32(float32x2_t __p0) {
3496234962}
3496334963#else
3496434964__ai uint32x2_t vcvtn_u32_f32(float32x2_t __p0) {
34965 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3496634965 uint32x2_t __ret;
34966 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3496734967 __ret = (uint32x2_t) __builtin_neon_vcvtn_u32_v((int8x8_t)__rev0, 18);
3496834968 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3496934969 return __ret;
......@@ -34978,8 +34978,8 @@ __ai int32x4_t vcvtpq_s32_f32(float32x4_t __p0) {
3497834978}
3497934979#else
3498034980__ai int32x4_t vcvtpq_s32_f32(float32x4_t __p0) {
34981 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3498234981 int32x4_t __ret;
34982 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3498334983 __ret = (int32x4_t) __builtin_neon_vcvtpq_s32_v((int8x16_t)__rev0, 34);
3498434984 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3498534985 return __ret;
......@@ -34994,8 +34994,8 @@ __ai int32x2_t vcvtp_s32_f32(float32x2_t __p0) {
3499434994}
3499534995#else
3499634996__ai int32x2_t vcvtp_s32_f32(float32x2_t __p0) {
34997 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3499834997 int32x2_t __ret;
34998 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3499934999 __ret = (int32x2_t) __builtin_neon_vcvtp_s32_v((int8x8_t)__rev0, 2);
3500035000 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3500135001 return __ret;
......@@ -35010,8 +35010,8 @@ __ai uint32x4_t vcvtpq_u32_f32(float32x4_t __p0) {
3501035010}
3501135011#else
3501235012__ai uint32x4_t vcvtpq_u32_f32(float32x4_t __p0) {
35013 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3501435013 uint32x4_t __ret;
35014 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3501535015 __ret = (uint32x4_t) __builtin_neon_vcvtpq_u32_v((int8x16_t)__rev0, 50);
3501635016 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3501735017 return __ret;
......@@ -35026,8 +35026,8 @@ __ai uint32x2_t vcvtp_u32_f32(float32x2_t __p0) {
3502635026}
3502735027#else
3502835028__ai uint32x2_t vcvtp_u32_f32(float32x2_t __p0) {
35029 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3503035029 uint32x2_t __ret;
35030 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3503135031 __ret = (uint32x2_t) __builtin_neon_vcvtp_u32_v((int8x8_t)__rev0, 18);
3503235032 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3503335033 return __ret;
......@@ -35044,9 +35044,9 @@ __ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3504435044}
3504535045#else
3504635046__ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
35047 uint8x16_t __ret;
3504735048 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3504835049 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35049 uint8x16_t __ret;
3505035050 __ret = (uint8x16_t) __builtin_neon_vaesdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3505135051 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3505235052 return __ret;
......@@ -35061,9 +35061,9 @@ __ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3506135061}
3506235062#else
3506335063__ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
35064 uint8x16_t __ret;
3506435065 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3506535066 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35066 uint8x16_t __ret;
3506735067 __ret = (uint8x16_t) __builtin_neon_vaeseq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
3506835068 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3506935069 return __ret;
......@@ -35078,8 +35078,8 @@ __ai uint8x16_t vaesimcq_u8(uint8x16_t __p0) {
3507835078}
3507935079#else
3508035080__ai uint8x16_t vaesimcq_u8(uint8x16_t __p0) {
35081 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3508235081 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);
3508335083 __ret = (uint8x16_t) __builtin_neon_vaesimcq_v((int8x16_t)__rev0, 48);
3508435084 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3508535085 return __ret;
......@@ -35094,8 +35094,8 @@ __ai uint8x16_t vaesmcq_u8(uint8x16_t __p0) {
3509435094}
3509535095#else
3509635096__ai uint8x16_t vaesmcq_u8(uint8x16_t __p0) {
35097 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3509835097 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);
3509935099 __ret = (uint8x16_t) __builtin_neon_vaesmcq_v((int8x16_t)__rev0, 48);
3510035100 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3510135101 return __ret;
......@@ -35112,8 +35112,8 @@ __ai float32x4_t vrndq_f32(float32x4_t __p0) {
3511235112}
3511335113#else
3511435114__ai float32x4_t vrndq_f32(float32x4_t __p0) {
35115 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3511635115 float32x4_t __ret;
35116 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3511735117 __ret = (float32x4_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 41);
3511835118 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3511935119 return __ret;
......@@ -35128,8 +35128,8 @@ __ai float32x2_t vrnd_f32(float32x2_t __p0) {
3512835128}
3512935129#else
3513035130__ai float32x2_t vrnd_f32(float32x2_t __p0) {
35131 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3513235131 float32x2_t __ret;
35132 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3513335133 __ret = (float32x2_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 9);
3513435134 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3513535135 return __ret;
......@@ -35144,8 +35144,8 @@ __ai float32x4_t vrndaq_f32(float32x4_t __p0) {
3514435144}
3514535145#else
3514635146__ai float32x4_t vrndaq_f32(float32x4_t __p0) {
35147 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3514835147 float32x4_t __ret;
35148 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3514935149 __ret = (float32x4_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 41);
3515035150 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3515135151 return __ret;
......@@ -35160,8 +35160,8 @@ __ai float32x2_t vrnda_f32(float32x2_t __p0) {
3516035160}
3516135161#else
3516235162__ai float32x2_t vrnda_f32(float32x2_t __p0) {
35163 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3516435163 float32x2_t __ret;
35164 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3516535165 __ret = (float32x2_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 9);
3516635166 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3516735167 return __ret;
......@@ -35176,8 +35176,8 @@ __ai float32x4_t vrndiq_f32(float32x4_t __p0) {
3517635176}
3517735177#else
3517835178__ai float32x4_t vrndiq_f32(float32x4_t __p0) {
35179 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3518035179 float32x4_t __ret;
35180 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3518135181 __ret = (float32x4_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 41);
3518235182 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3518335183 return __ret;
......@@ -35192,8 +35192,8 @@ __ai float32x2_t vrndi_f32(float32x2_t __p0) {
3519235192}
3519335193#else
3519435194__ai float32x2_t vrndi_f32(float32x2_t __p0) {
35195 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3519635195 float32x2_t __ret;
35196 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3519735197 __ret = (float32x2_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 9);
3519835198 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3519935199 return __ret;
......@@ -35208,8 +35208,8 @@ __ai float32x4_t vrndmq_f32(float32x4_t __p0) {
3520835208}
3520935209#else
3521035210__ai float32x4_t vrndmq_f32(float32x4_t __p0) {
35211 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3521235211 float32x4_t __ret;
35212 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3521335213 __ret = (float32x4_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 41);
3521435214 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3521535215 return __ret;
......@@ -35224,8 +35224,8 @@ __ai float32x2_t vrndm_f32(float32x2_t __p0) {
3522435224}
3522535225#else
3522635226__ai float32x2_t vrndm_f32(float32x2_t __p0) {
35227 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3522835227 float32x2_t __ret;
35228 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3522935229 __ret = (float32x2_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 9);
3523035230 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3523135231 return __ret;
......@@ -35240,8 +35240,8 @@ __ai float32x4_t vrndnq_f32(float32x4_t __p0) {
3524035240}
3524135241#else
3524235242__ai float32x4_t vrndnq_f32(float32x4_t __p0) {
35243 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3524435243 float32x4_t __ret;
35244 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3524535245 __ret = (float32x4_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 41);
3524635246 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3524735247 return __ret;
......@@ -35256,8 +35256,8 @@ __ai float32x2_t vrndn_f32(float32x2_t __p0) {
3525635256}
3525735257#else
3525835258__ai float32x2_t vrndn_f32(float32x2_t __p0) {
35259 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3526035259 float32x2_t __ret;
35260 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3526135261 __ret = (float32x2_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 9);
3526235262 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3526335263 return __ret;
......@@ -35277,8 +35277,8 @@ __ai float32x4_t vrndpq_f32(float32x4_t __p0) {
3527735277}
3527835278#else
3527935279__ai float32x4_t vrndpq_f32(float32x4_t __p0) {
35280 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3528135280 float32x4_t __ret;
35281 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3528235282 __ret = (float32x4_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 41);
3528335283 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3528435284 return __ret;
......@@ -35293,8 +35293,8 @@ __ai float32x2_t vrndp_f32(float32x2_t __p0) {
3529335293}
3529435294#else
3529535295__ai float32x2_t vrndp_f32(float32x2_t __p0) {
35296 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3529735296 float32x2_t __ret;
35297 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3529835298 __ret = (float32x2_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 9);
3529935299 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3530035300 return __ret;
......@@ -35309,8 +35309,8 @@ __ai float32x4_t vrndxq_f32(float32x4_t __p0) {
3530935309}
3531035310#else
3531135311__ai float32x4_t vrndxq_f32(float32x4_t __p0) {
35312 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3531335312 float32x4_t __ret;
35313 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3531435314 __ret = (float32x4_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 41);
3531535315 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3531635316 return __ret;
......@@ -35325,8 +35325,8 @@ __ai float32x2_t vrndx_f32(float32x2_t __p0) {
3532535325}
3532635326#else
3532735327__ai float32x2_t vrndx_f32(float32x2_t __p0) {
35328 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3532935328 float32x2_t __ret;
35329 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3533035330 __ret = (float32x2_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 9);
3533135331 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3533235332 return __ret;
......@@ -35343,8 +35343,8 @@ __ai float16x8_t vrndq_f16(float16x8_t __p0) {
3534335343}
3534435344#else
3534535345__ai float16x8_t vrndq_f16(float16x8_t __p0) {
35346 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3534735346 float16x8_t __ret;
35347 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3534835348 __ret = (float16x8_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 40);
3534935349 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3535035350 return __ret;
......@@ -35359,8 +35359,8 @@ __ai float16x4_t vrnd_f16(float16x4_t __p0) {
3535935359}
3536035360#else
3536135361__ai float16x4_t vrnd_f16(float16x4_t __p0) {
35362 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3536335362 float16x4_t __ret;
35363 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3536435364 __ret = (float16x4_t) __builtin_neon_vrnd_v((int8x8_t)__rev0, 8);
3536535365 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3536635366 return __ret;
......@@ -35375,8 +35375,8 @@ __ai float16x8_t vrndaq_f16(float16x8_t __p0) {
3537535375}
3537635376#else
3537735377__ai float16x8_t vrndaq_f16(float16x8_t __p0) {
35378 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3537935378 float16x8_t __ret;
35379 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3538035380 __ret = (float16x8_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 40);
3538135381 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3538235382 return __ret;
......@@ -35391,8 +35391,8 @@ __ai float16x4_t vrnda_f16(float16x4_t __p0) {
3539135391}
3539235392#else
3539335393__ai float16x4_t vrnda_f16(float16x4_t __p0) {
35394 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3539535394 float16x4_t __ret;
35395 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3539635396 __ret = (float16x4_t) __builtin_neon_vrnda_v((int8x8_t)__rev0, 8);
3539735397 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3539835398 return __ret;
......@@ -35407,8 +35407,8 @@ __ai float16x8_t vrndmq_f16(float16x8_t __p0) {
3540735407}
3540835408#else
3540935409__ai float16x8_t vrndmq_f16(float16x8_t __p0) {
35410 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3541135410 float16x8_t __ret;
35411 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3541235412 __ret = (float16x8_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 40);
3541335413 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3541435414 return __ret;
......@@ -35423,8 +35423,8 @@ __ai float16x4_t vrndm_f16(float16x4_t __p0) {
3542335423}
3542435424#else
3542535425__ai float16x4_t vrndm_f16(float16x4_t __p0) {
35426 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3542735426 float16x4_t __ret;
35427 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3542835428 __ret = (float16x4_t) __builtin_neon_vrndm_v((int8x8_t)__rev0, 8);
3542935429 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3543035430 return __ret;
......@@ -35439,8 +35439,8 @@ __ai float16x8_t vrndnq_f16(float16x8_t __p0) {
3543935439}
3544035440#else
3544135441__ai float16x8_t vrndnq_f16(float16x8_t __p0) {
35442 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3544335442 float16x8_t __ret;
35443 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3544435444 __ret = (float16x8_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 40);
3544535445 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3544635446 return __ret;
......@@ -35455,8 +35455,8 @@ __ai float16x4_t vrndn_f16(float16x4_t __p0) {
3545535455}
3545635456#else
3545735457__ai float16x4_t vrndn_f16(float16x4_t __p0) {
35458 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3545935458 float16x4_t __ret;
35459 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3546035460 __ret = (float16x4_t) __builtin_neon_vrndn_v((int8x8_t)__rev0, 8);
3546135461 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3546235462 return __ret;
......@@ -35471,8 +35471,8 @@ __ai float16x8_t vrndpq_f16(float16x8_t __p0) {
3547135471}
3547235472#else
3547335473__ai float16x8_t vrndpq_f16(float16x8_t __p0) {
35474 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3547535474 float16x8_t __ret;
35475 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3547635476 __ret = (float16x8_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 40);
3547735477 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3547835478 return __ret;
......@@ -35487,8 +35487,8 @@ __ai float16x4_t vrndp_f16(float16x4_t __p0) {
3548735487}
3548835488#else
3548935489__ai float16x4_t vrndp_f16(float16x4_t __p0) {
35490 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3549135490 float16x4_t __ret;
35491 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3549235492 __ret = (float16x4_t) __builtin_neon_vrndp_v((int8x8_t)__rev0, 8);
3549335493 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3549435494 return __ret;
......@@ -35503,8 +35503,8 @@ __ai float16x8_t vrndxq_f16(float16x8_t __p0) {
3550335503}
3550435504#else
3550535505__ai float16x8_t vrndxq_f16(float16x8_t __p0) {
35506 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3550735506 float16x8_t __ret;
35507 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3550835508 __ret = (float16x8_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 40);
3550935509 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3551035510 return __ret;
......@@ -35519,8 +35519,8 @@ __ai float16x4_t vrndx_f16(float16x4_t __p0) {
3551935519}
3552035520#else
3552135521__ai float16x4_t vrndx_f16(float16x4_t __p0) {
35522 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3552335522 float16x4_t __ret;
35523 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3552435524 __ret = (float16x4_t) __builtin_neon_vrndx_v((int8x8_t)__rev0, 8);
3552535525 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3552635526 return __ret;
......@@ -35537,9 +35537,9 @@ __ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
3553735537}
3553835538#else
3553935539__ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
35540 float32x4_t __ret;
3554035541 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3554135542 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35542 float32x4_t __ret;
3554335543 __ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
3554435544 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3554535545 return __ret;
......@@ -35554,9 +35554,9 @@ __ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
3555435554}
3555535555#else
3555635556__ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
35557 float32x2_t __ret;
3555735558 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3555835559 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
35559 float32x2_t __ret;
3556035560 __ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
3556135561 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3556235562 return __ret;
......@@ -35571,9 +35571,9 @@ __ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
3557135571}
3557235572#else
3557335573__ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
35574 float32x4_t __ret;
3557435575 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3557535576 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35576 float32x4_t __ret;
3557735577 __ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
3557835578 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3557935579 return __ret;
......@@ -35588,9 +35588,9 @@ __ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
3558835588}
3558935589#else
3559035590__ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
35591 float32x2_t __ret;
3559135592 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3559235593 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
35593 float32x2_t __ret;
3559435594 __ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
3559535595 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3559635596 return __ret;
......@@ -35607,9 +35607,9 @@ __ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
3560735607}
3560835608#else
3560935609__ai float16x8_t vmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
35610 float16x8_t __ret;
3561035611 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3561135612 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
35612 float16x8_t __ret;
3561335613 __ret = (float16x8_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
3561435614 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3561535615 return __ret;
......@@ -35624,9 +35624,9 @@ __ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
3562435624}
3562535625#else
3562635626__ai float16x4_t vmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
35627 float16x4_t __ret;
3562735628 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3562835629 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35629 float16x4_t __ret;
3563035630 __ret = (float16x4_t) __builtin_neon_vmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
3563135631 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3563235632 return __ret;
......@@ -35641,9 +35641,9 @@ __ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
3564135641}
3564235642#else
3564335643__ai float16x8_t vminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
35644 float16x8_t __ret;
3564435645 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3564535646 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
35646 float16x8_t __ret;
3564735647 __ret = (float16x8_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
3564835648 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3564935649 return __ret;
......@@ -35658,9 +35658,9 @@ __ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
3565835658}
3565935659#else
3566035660__ai float16x4_t vminnm_f16(float16x4_t __p0, float16x4_t __p1) {
35661 float16x4_t __ret;
3566135662 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3566235663 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35663 float16x4_t __ret;
3566435664 __ret = (float16x4_t) __builtin_neon_vminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
3566535665 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3566635666 return __ret;
......@@ -35677,9 +35677,9 @@ __ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
3567735677}
3567835678#else
3567935679__ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35680 uint32x4_t __ret;
3568035681 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3568135682 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35682 uint32x4_t __ret;
3568335683 __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32(__rev0, __p1, __rev2);
3568435684 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3568535685 return __ret;
......@@ -35699,9 +35699,9 @@ __ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
3569935699}
3570035700#else
3570135701__ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35702 uint32x4_t __ret;
3570235703 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3570335704 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35704 uint32x4_t __ret;
3570535705 __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32(__rev0, __p1, __rev2);
3570635706 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3570735707 return __ret;
......@@ -35716,9 +35716,9 @@ __ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
3571635716}
3571735717#else
3571835718__ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
35719 uint32x4_t __ret;
3571935720 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3572035721 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35721 uint32x4_t __ret;
3572235722 __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32(__rev0, __p1, __rev2);
3572335723 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3572435724 return __ret;
......@@ -35733,10 +35733,10 @@ __ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3573335733}
3573435734#else
3573535735__ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35736 uint32x4_t __ret;
3573635737 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3573735738 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3573835739 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35739 uint32x4_t __ret;
3574035740 __ret = (uint32x4_t) __builtin_neon_vsha1su0q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3574135741 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3574235742 return __ret;
......@@ -35751,9 +35751,9 @@ __ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
3575135751}
3575235752#else
3575335753__ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
35754 uint32x4_t __ret;
3575435755 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3575535756 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35756 uint32x4_t __ret;
3575735757 __ret = (uint32x4_t) __builtin_neon_vsha1su1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3575835758 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3575935759 return __ret;
......@@ -35768,10 +35768,10 @@ __ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3576835768}
3576935769#else
3577035770__ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35771 uint32x4_t __ret;
3577135772 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3577235773 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3577335774 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35774 uint32x4_t __ret;
3577535775 __ret = (uint32x4_t) __builtin_neon_vsha256hq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3577635776 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3577735777 return __ret;
......@@ -35786,10 +35786,10 @@ __ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2
3578635786}
3578735787#else
3578835788__ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35789 uint32x4_t __ret;
3578935790 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3579035791 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3579135792 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35792 uint32x4_t __ret;
3579335793 __ret = (uint32x4_t) __builtin_neon_vsha256h2q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3579435794 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3579535795 return __ret;
......@@ -35804,9 +35804,9 @@ __ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
3580435804}
3580535805#else
3580635806__ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
35807 uint32x4_t __ret;
3580735808 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3580835809 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
35809 uint32x4_t __ret;
3581035810 __ret = (uint32x4_t) __builtin_neon_vsha256su0q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3581135811 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3581235812 return __ret;
......@@ -35821,10 +35821,10 @@ __ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p
3582135821}
3582235822#else
3582335823__ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35824 uint32x4_t __ret;
3582435825 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3582535826 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3582635827 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35827 uint32x4_t __ret;
3582835828 __ret = (uint32x4_t) __builtin_neon_vsha256su1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3582935829 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3583035830 return __ret;
......@@ -35841,10 +35841,10 @@ __ai uint8x16_t vbcaxq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
3584135841}
3584235842#else
3584335843__ai uint8x16_t vbcaxq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
35844 uint8x16_t __ret;
3584435845 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3584535846 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3584635847 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35847 uint8x16_t __ret;
3584835848 __ret = (uint8x16_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
3584935849 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3585035850 return __ret;
......@@ -35859,10 +35859,10 @@ __ai uint32x4_t vbcaxq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
3585935859}
3586035860#else
3586135861__ai uint32x4_t vbcaxq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
35862 uint32x4_t __ret;
3586235863 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3586335864 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3586435865 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35865 uint32x4_t __ret;
3586635866 __ret = (uint32x4_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3586735867 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3586835868 return __ret;
......@@ -35877,10 +35877,10 @@ __ai uint64x2_t vbcaxq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
3587735877}
3587835878#else
3587935879__ai uint64x2_t vbcaxq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
35880 uint64x2_t __ret;
3588035881 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3588135882 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3588235883 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
35883 uint64x2_t __ret;
3588435884 __ret = (uint64x2_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
3588535885 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3588635886 return __ret;
......@@ -35895,10 +35895,10 @@ __ai uint16x8_t vbcaxq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
3589535895}
3589635896#else
3589735897__ai uint16x8_t vbcaxq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
35898 uint16x8_t __ret;
3589835899 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3589935900 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3590035901 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
35901 uint16x8_t __ret;
3590235902 __ret = (uint16x8_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
3590335903 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3590435904 return __ret;
......@@ -35913,10 +35913,10 @@ __ai int8x16_t vbcaxq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
3591335913}
3591435914#else
3591535915__ai int8x16_t vbcaxq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
35916 int8x16_t __ret;
3591635917 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3591735918 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3591835919 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35919 int8x16_t __ret;
3592035920 __ret = (int8x16_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
3592135921 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3592235922 return __ret;
......@@ -35931,10 +35931,10 @@ __ai int32x4_t vbcaxq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
3593135931}
3593235932#else
3593335933__ai int32x4_t vbcaxq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
35934 int32x4_t __ret;
3593435935 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3593535936 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3593635937 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
35937 int32x4_t __ret;
3593835938 __ret = (int32x4_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
3593935939 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3594035940 return __ret;
......@@ -35949,10 +35949,10 @@ __ai int64x2_t vbcaxq_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
3594935949}
3595035950#else
3595135951__ai int64x2_t vbcaxq_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
35952 int64x2_t __ret;
3595235953 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3595335954 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3595435955 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
35955 int64x2_t __ret;
3595635956 __ret = (int64x2_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
3595735957 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3595835958 return __ret;
......@@ -35967,10 +35967,10 @@ __ai int16x8_t vbcaxq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
3596735967}
3596835968#else
3596935969__ai int16x8_t vbcaxq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
35970 int16x8_t __ret;
3597035971 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3597135972 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3597235973 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
35973 int16x8_t __ret;
3597435974 __ret = (int16x8_t) __builtin_neon_vbcaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
3597535975 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3597635976 return __ret;
......@@ -35985,10 +35985,10 @@ __ai uint8x16_t veor3q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
3598535985}
3598635986#else
3598735987__ai uint8x16_t veor3q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
35988 uint8x16_t __ret;
3598835989 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3598935990 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3599035991 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
35991 uint8x16_t __ret;
3599235992 __ret = (uint8x16_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
3599335993 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3599435994 return __ret;
......@@ -36003,10 +36003,10 @@ __ai uint32x4_t veor3q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
3600336003}
3600436004#else
3600536005__ai uint32x4_t veor3q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36006 uint32x4_t __ret;
3600636007 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3600736008 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3600836009 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36009 uint32x4_t __ret;
3601036010 __ret = (uint32x4_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3601136011 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3601236012 return __ret;
......@@ -36021,10 +36021,10 @@ __ai uint64x2_t veor3q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
3602136021}
3602236022#else
3602336023__ai uint64x2_t veor3q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36024 uint64x2_t __ret;
3602436025 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3602536026 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3602636027 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36027 uint64x2_t __ret;
3602836028 __ret = (uint64x2_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
3602936029 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3603036030 return __ret;
......@@ -36039,10 +36039,10 @@ __ai uint16x8_t veor3q_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
3603936039}
3604036040#else
3604136041__ai uint16x8_t veor3q_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
36042 uint16x8_t __ret;
3604236043 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3604336044 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3604436045 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
36045 uint16x8_t __ret;
3604636046 __ret = (uint16x8_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 49);
3604736047 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3604836048 return __ret;
......@@ -36057,10 +36057,10 @@ __ai int8x16_t veor3q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
3605736057}
3605836058#else
3605936059__ai int8x16_t veor3q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
36060 int8x16_t __ret;
3606036061 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3606136062 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3606236063 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
36063 int8x16_t __ret;
3606436064 __ret = (int8x16_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
3606536065 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3606636066 return __ret;
......@@ -36075,10 +36075,10 @@ __ai int32x4_t veor3q_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
3607536075}
3607636076#else
3607736077__ai int32x4_t veor3q_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
36078 int32x4_t __ret;
3607836079 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3607936080 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3608036081 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36081 int32x4_t __ret;
3608236082 __ret = (int32x4_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
3608336083 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3608436084 return __ret;
......@@ -36093,10 +36093,10 @@ __ai int64x2_t veor3q_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
3609336093}
3609436094#else
3609536095__ai int64x2_t veor3q_s64(int64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
36096 int64x2_t __ret;
3609636097 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3609736098 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3609836099 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36099 int64x2_t __ret;
3610036100 __ret = (int64x2_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 35);
3610136101 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3610236102 return __ret;
......@@ -36111,10 +36111,10 @@ __ai int16x8_t veor3q_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
3611136111}
3611236112#else
3611336113__ai int16x8_t veor3q_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
36114 int16x8_t __ret;
3611436115 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3611536116 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3611636117 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
36117 int16x8_t __ret;
3611836118 __ret = (int16x8_t) __builtin_neon_veor3q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
3611936119 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3612036120 return __ret;
......@@ -36129,9 +36129,9 @@ __ai uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
3612936129}
3613036130#else
3613136131__ai uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
36132 uint64x2_t __ret;
3613236133 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3613336134 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36134 uint64x2_t __ret;
3613536135 __ret = (uint64x2_t) __builtin_neon_vrax1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
3613636136 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3613736137 return __ret;
......@@ -36140,19 +36140,19 @@ __ai uint64x2_t vrax1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
3614036140
3614136141#ifdef __LITTLE_ENDIAN__
3614236142#define vxarq_u64(__p0, __p1, __p2) __extension__ ({ \
36143 uint64x2_t __ret; \
3614336144 uint64x2_t __s0 = __p0; \
3614436145 uint64x2_t __s1 = __p1; \
36145 uint64x2_t __ret; \
3614636146 __ret = (uint64x2_t) __builtin_neon_vxarq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 51); \
3614736147 __ret; \
3614836148})
3614936149#else
3615036150#define vxarq_u64(__p0, __p1, __p2) __extension__ ({ \
36151 uint64x2_t __ret; \
3615136152 uint64x2_t __s0 = __p0; \
3615236153 uint64x2_t __s1 = __p1; \
3615336154 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
3615436155 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
36155 uint64x2_t __ret; \
3615636156 __ret = (uint64x2_t) __builtin_neon_vxarq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 51); \
3615736157 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
3615836158 __ret; \
......@@ -36169,10 +36169,10 @@ __ai uint64x2_t vsha512hq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2)
3616936169}
3617036170#else
3617136171__ai uint64x2_t vsha512hq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36172 uint64x2_t __ret;
3617236173 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3617336174 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3617436175 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36175 uint64x2_t __ret;
3617636176 __ret = (uint64x2_t) __builtin_neon_vsha512hq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
3617736177 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3617836178 return __ret;
......@@ -36187,10 +36187,10 @@ __ai uint64x2_t vsha512h2q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2
3618736187}
3618836188#else
3618936189__ai uint64x2_t vsha512h2q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36190 uint64x2_t __ret;
3619036191 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3619136192 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3619236193 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36193 uint64x2_t __ret;
3619436194 __ret = (uint64x2_t) __builtin_neon_vsha512h2q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
3619536195 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3619636196 return __ret;
......@@ -36205,9 +36205,9 @@ __ai uint64x2_t vsha512su0q_u64(uint64x2_t __p0, uint64x2_t __p1) {
3620536205}
3620636206#else
3620736207__ai uint64x2_t vsha512su0q_u64(uint64x2_t __p0, uint64x2_t __p1) {
36208 uint64x2_t __ret;
3620836209 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3620936210 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
36210 uint64x2_t __ret;
3621136211 __ret = (uint64x2_t) __builtin_neon_vsha512su0q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
3621236212 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3621336213 return __ret;
......@@ -36222,10 +36222,10 @@ __ai uint64x2_t vsha512su1q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p
3622236222}
3622336223#else
3622436224__ai uint64x2_t vsha512su1q_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
36225 uint64x2_t __ret;
3622536226 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3622636227 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3622736228 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
36228 uint64x2_t __ret;
3622936229 __ret = (uint64x2_t) __builtin_neon_vsha512su1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 51);
3623036230 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3623136231 return __ret;
......@@ -36242,10 +36242,10 @@ __ai uint32x4_t vsm3partw1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p
3624236242}
3624336243#else
3624436244__ai uint32x4_t vsm3partw1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36245 uint32x4_t __ret;
3624536246 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3624636247 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3624736248 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36248 uint32x4_t __ret;
3624936249 __ret = (uint32x4_t) __builtin_neon_vsm3partw1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3625036250 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3625136251 return __ret;
......@@ -36260,10 +36260,10 @@ __ai uint32x4_t vsm3partw2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p
3626036260}
3626136261#else
3626236262__ai uint32x4_t vsm3partw2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36263 uint32x4_t __ret;
3626336264 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3626436265 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3626536266 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36266 uint32x4_t __ret;
3626736267 __ret = (uint32x4_t) __builtin_neon_vsm3partw2q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3626836268 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3626936269 return __ret;
......@@ -36278,10 +36278,10 @@ __ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3627836278}
3627936279#else
3628036280__ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
36281 uint32x4_t __ret;
3628136282 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3628236283 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3628336284 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
36284 uint32x4_t __ret;
3628536285 __ret = (uint32x4_t) __builtin_neon_vsm3ss1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
3628636286 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3628736287 return __ret;
......@@ -36290,22 +36290,22 @@ __ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3629036290
3629136291#ifdef __LITTLE_ENDIAN__
3629236292#define vsm3tt1aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36293 uint32x4_t __ret; \
3629336294 uint32x4_t __s0 = __p0; \
3629436295 uint32x4_t __s1 = __p1; \
3629536296 uint32x4_t __s2 = __p2; \
36296 uint32x4_t __ret; \
3629736297 __ret = (uint32x4_t) __builtin_neon_vsm3tt1aq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
3629836298 __ret; \
3629936299})
3630036300#else
3630136301#define vsm3tt1aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36302 uint32x4_t __ret; \
3630236303 uint32x4_t __s0 = __p0; \
3630336304 uint32x4_t __s1 = __p1; \
3630436305 uint32x4_t __s2 = __p2; \
3630536306 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
3630636307 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
3630736308 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36308 uint32x4_t __ret; \
3630936309 __ret = (uint32x4_t) __builtin_neon_vsm3tt1aq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
3631036310 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3631136311 __ret; \
......@@ -36314,22 +36314,22 @@ __ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3631436314
3631536315#ifdef __LITTLE_ENDIAN__
3631636316#define vsm3tt1bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36317 uint32x4_t __ret; \
3631736318 uint32x4_t __s0 = __p0; \
3631836319 uint32x4_t __s1 = __p1; \
3631936320 uint32x4_t __s2 = __p2; \
36320 uint32x4_t __ret; \
3632136321 __ret = (uint32x4_t) __builtin_neon_vsm3tt1bq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
3632236322 __ret; \
3632336323})
3632436324#else
3632536325#define vsm3tt1bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36326 uint32x4_t __ret; \
3632636327 uint32x4_t __s0 = __p0; \
3632736328 uint32x4_t __s1 = __p1; \
3632836329 uint32x4_t __s2 = __p2; \
3632936330 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
3633036331 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
3633136332 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36332 uint32x4_t __ret; \
3633336333 __ret = (uint32x4_t) __builtin_neon_vsm3tt1bq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
3633436334 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3633536335 __ret; \
......@@ -36338,22 +36338,22 @@ __ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3633836338
3633936339#ifdef __LITTLE_ENDIAN__
3634036340#define vsm3tt2aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36341 uint32x4_t __ret; \
3634136342 uint32x4_t __s0 = __p0; \
3634236343 uint32x4_t __s1 = __p1; \
3634336344 uint32x4_t __s2 = __p2; \
36344 uint32x4_t __ret; \
3634536345 __ret = (uint32x4_t) __builtin_neon_vsm3tt2aq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
3634636346 __ret; \
3634736347})
3634836348#else
3634936349#define vsm3tt2aq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36350 uint32x4_t __ret; \
3635036351 uint32x4_t __s0 = __p0; \
3635136352 uint32x4_t __s1 = __p1; \
3635236353 uint32x4_t __s2 = __p2; \
3635336354 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
3635436355 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
3635536356 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36356 uint32x4_t __ret; \
3635736357 __ret = (uint32x4_t) __builtin_neon_vsm3tt2aq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
3635836358 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3635936359 __ret; \
......@@ -36362,22 +36362,22 @@ __ai uint32x4_t vsm3ss1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2)
3636236362
3636336363#ifdef __LITTLE_ENDIAN__
3636436364#define vsm3tt2bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36365 uint32x4_t __ret; \
3636536366 uint32x4_t __s0 = __p0; \
3636636367 uint32x4_t __s1 = __p1; \
3636736368 uint32x4_t __s2 = __p2; \
36368 uint32x4_t __ret; \
3636936369 __ret = (uint32x4_t) __builtin_neon_vsm3tt2bq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 50); \
3637036370 __ret; \
3637136371})
3637236372#else
3637336373#define vsm3tt2bq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
36374 uint32x4_t __ret; \
3637436375 uint32x4_t __s0 = __p0; \
3637536376 uint32x4_t __s1 = __p1; \
3637636377 uint32x4_t __s2 = __p2; \
3637736378 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
3637836379 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
3637936380 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
36380 uint32x4_t __ret; \
3638136381 __ret = (uint32x4_t) __builtin_neon_vsm3tt2bq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 50); \
3638236382 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3638336383 __ret; \
......@@ -36394,9 +36394,9 @@ __ai uint32x4_t vsm4eq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3639436394}
3639536395#else
3639636396__ai uint32x4_t vsm4eq_u32(uint32x4_t __p0, uint32x4_t __p1) {
36397 uint32x4_t __ret;
3639736398 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3639836399 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36399 uint32x4_t __ret;
3640036400 __ret = (uint32x4_t) __builtin_neon_vsm4eq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3640136401 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3640236402 return __ret;
......@@ -36411,9 +36411,9 @@ __ai uint32x4_t vsm4ekeyq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3641136411}
3641236412#else
3641336413__ai uint32x4_t vsm4ekeyq_u32(uint32x4_t __p0, uint32x4_t __p1) {
36414 uint32x4_t __ret;
3641436415 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3641536416 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
36416 uint32x4_t __ret;
3641736417 __ret = (uint32x4_t) __builtin_neon_vsm4ekeyq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
3641836418 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3641936419 return __ret;
......@@ -36430,8 +36430,8 @@ __ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
3643036430}
3643136431#else
3643236432__ai int64x2_t vcvtaq_s64_f64(float64x2_t __p0) {
36433 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3643436433 int64x2_t __ret;
36434 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3643536435 __ret = (int64x2_t) __builtin_neon_vcvtaq_s64_v((int8x16_t)__rev0, 35);
3643636436 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3643736437 return __ret;
......@@ -36451,8 +36451,8 @@ __ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
3645136451}
3645236452#else
3645336453__ai uint64x2_t vcvtaq_u64_f64(float64x2_t __p0) {
36454 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3645536454 uint64x2_t __ret;
36455 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3645636456 __ret = (uint64x2_t) __builtin_neon_vcvtaq_u64_v((int8x16_t)__rev0, 51);
3645736457 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3645836458 return __ret;
......@@ -36472,8 +36472,8 @@ __ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
3647236472}
3647336473#else
3647436474__ai int64x2_t vcvtmq_s64_f64(float64x2_t __p0) {
36475 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3647636475 int64x2_t __ret;
36476 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3647736477 __ret = (int64x2_t) __builtin_neon_vcvtmq_s64_v((int8x16_t)__rev0, 35);
3647836478 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3647936479 return __ret;
......@@ -36493,8 +36493,8 @@ __ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
3649336493}
3649436494#else
3649536495__ai uint64x2_t vcvtmq_u64_f64(float64x2_t __p0) {
36496 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3649736496 uint64x2_t __ret;
36497 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3649836498 __ret = (uint64x2_t) __builtin_neon_vcvtmq_u64_v((int8x16_t)__rev0, 51);
3649936499 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3650036500 return __ret;
......@@ -36514,8 +36514,8 @@ __ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
3651436514}
3651536515#else
3651636516__ai int64x2_t vcvtnq_s64_f64(float64x2_t __p0) {
36517 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3651836517 int64x2_t __ret;
36518 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3651936519 __ret = (int64x2_t) __builtin_neon_vcvtnq_s64_v((int8x16_t)__rev0, 35);
3652036520 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3652136521 return __ret;
......@@ -36535,8 +36535,8 @@ __ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
3653536535}
3653636536#else
3653736537__ai uint64x2_t vcvtnq_u64_f64(float64x2_t __p0) {
36538 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3653936538 uint64x2_t __ret;
36539 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3654036540 __ret = (uint64x2_t) __builtin_neon_vcvtnq_u64_v((int8x16_t)__rev0, 51);
3654136541 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3654236542 return __ret;
......@@ -36556,8 +36556,8 @@ __ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
3655636556}
3655736557#else
3655836558__ai int64x2_t vcvtpq_s64_f64(float64x2_t __p0) {
36559 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3656036559 int64x2_t __ret;
36560 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3656136561 __ret = (int64x2_t) __builtin_neon_vcvtpq_s64_v((int8x16_t)__rev0, 35);
3656236562 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3656336563 return __ret;
......@@ -36577,8 +36577,8 @@ __ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
3657736577}
3657836578#else
3657936579__ai uint64x2_t vcvtpq_u64_f64(float64x2_t __p0) {
36580 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3658136580 uint64x2_t __ret;
36581 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3658236582 __ret = (uint64x2_t) __builtin_neon_vcvtpq_u64_v((int8x16_t)__rev0, 51);
3658336583 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3658436584 return __ret;
......@@ -38560,8 +38560,8 @@ __ai float64x2_t vrndq_f64(float64x2_t __p0) {
3856038560}
3856138561#else
3856238562__ai float64x2_t vrndq_f64(float64x2_t __p0) {
38563 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3856438563 float64x2_t __ret;
38564 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3856538565 __ret = (float64x2_t) __builtin_neon_vrndq_v((int8x16_t)__rev0, 42);
3856638566 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3856738567 return __ret;
......@@ -38581,8 +38581,8 @@ __ai float64x2_t vrndaq_f64(float64x2_t __p0) {
3858138581}
3858238582#else
3858338583__ai float64x2_t vrndaq_f64(float64x2_t __p0) {
38584 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3858538584 float64x2_t __ret;
38585 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3858638586 __ret = (float64x2_t) __builtin_neon_vrndaq_v((int8x16_t)__rev0, 42);
3858738587 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3858838588 return __ret;
......@@ -38602,8 +38602,8 @@ __ai float64x2_t vrndiq_f64(float64x2_t __p0) {
3860238602}
3860338603#else
3860438604__ai float64x2_t vrndiq_f64(float64x2_t __p0) {
38605 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3860638605 float64x2_t __ret;
38606 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3860738607 __ret = (float64x2_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 42);
3860838608 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3860938609 return __ret;
......@@ -38623,8 +38623,8 @@ __ai float64x2_t vrndmq_f64(float64x2_t __p0) {
3862338623}
3862438624#else
3862538625__ai float64x2_t vrndmq_f64(float64x2_t __p0) {
38626 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3862738626 float64x2_t __ret;
38627 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3862838628 __ret = (float64x2_t) __builtin_neon_vrndmq_v((int8x16_t)__rev0, 42);
3862938629 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3863038630 return __ret;
......@@ -38644,8 +38644,8 @@ __ai float64x2_t vrndnq_f64(float64x2_t __p0) {
3864438644}
3864538645#else
3864638646__ai float64x2_t vrndnq_f64(float64x2_t __p0) {
38647 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3864838647 float64x2_t __ret;
38648 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3864938649 __ret = (float64x2_t) __builtin_neon_vrndnq_v((int8x16_t)__rev0, 42);
3865038650 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3865138651 return __ret;
......@@ -38665,8 +38665,8 @@ __ai float64x2_t vrndpq_f64(float64x2_t __p0) {
3866538665}
3866638666#else
3866738667__ai float64x2_t vrndpq_f64(float64x2_t __p0) {
38668 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3866938668 float64x2_t __ret;
38669 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3867038670 __ret = (float64x2_t) __builtin_neon_vrndpq_v((int8x16_t)__rev0, 42);
3867138671 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3867238672 return __ret;
......@@ -38686,8 +38686,8 @@ __ai float64x2_t vrndxq_f64(float64x2_t __p0) {
3868638686}
3868738687#else
3868838688__ai float64x2_t vrndxq_f64(float64x2_t __p0) {
38689 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3869038689 float64x2_t __ret;
38690 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3869138691 __ret = (float64x2_t) __builtin_neon_vrndxq_v((int8x16_t)__rev0, 42);
3869238692 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3869338693 return __ret;
......@@ -38709,8 +38709,8 @@ __ai float32x4_t vrnd32xq_f32(float32x4_t __p0) {
3870938709}
3871038710#else
3871138711__ai float32x4_t vrnd32xq_f32(float32x4_t __p0) {
38712 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3871338712 float32x4_t __ret;
38713 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3871438714 __ret = (float32x4_t) __builtin_neon_vrnd32xq_v((int8x16_t)__rev0, 41);
3871538715 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3871638716 return __ret;
......@@ -38725,8 +38725,8 @@ __ai float32x2_t vrnd32x_f32(float32x2_t __p0) {
3872538725}
3872638726#else
3872738727__ai float32x2_t vrnd32x_f32(float32x2_t __p0) {
38728 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3872938728 float32x2_t __ret;
38729 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3873038730 __ret = (float32x2_t) __builtin_neon_vrnd32x_v((int8x8_t)__rev0, 9);
3873138731 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3873238732 return __ret;
......@@ -38741,8 +38741,8 @@ __ai float32x4_t vrnd32zq_f32(float32x4_t __p0) {
3874138741}
3874238742#else
3874338743__ai float32x4_t vrnd32zq_f32(float32x4_t __p0) {
38744 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3874538744 float32x4_t __ret;
38745 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3874638746 __ret = (float32x4_t) __builtin_neon_vrnd32zq_v((int8x16_t)__rev0, 41);
3874738747 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3874838748 return __ret;
......@@ -38757,8 +38757,8 @@ __ai float32x2_t vrnd32z_f32(float32x2_t __p0) {
3875738757}
3875838758#else
3875938759__ai float32x2_t vrnd32z_f32(float32x2_t __p0) {
38760 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3876138760 float32x2_t __ret;
38761 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3876238762 __ret = (float32x2_t) __builtin_neon_vrnd32z_v((int8x8_t)__rev0, 9);
3876338763 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3876438764 return __ret;
......@@ -38773,8 +38773,8 @@ __ai float32x4_t vrnd64xq_f32(float32x4_t __p0) {
3877338773}
3877438774#else
3877538775__ai float32x4_t vrnd64xq_f32(float32x4_t __p0) {
38776 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3877738776 float32x4_t __ret;
38777 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3877838778 __ret = (float32x4_t) __builtin_neon_vrnd64xq_v((int8x16_t)__rev0, 41);
3877938779 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3878038780 return __ret;
......@@ -38789,8 +38789,8 @@ __ai float32x2_t vrnd64x_f32(float32x2_t __p0) {
3878938789}
3879038790#else
3879138791__ai float32x2_t vrnd64x_f32(float32x2_t __p0) {
38792 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3879338792 float32x2_t __ret;
38793 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3879438794 __ret = (float32x2_t) __builtin_neon_vrnd64x_v((int8x8_t)__rev0, 9);
3879538795 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3879638796 return __ret;
......@@ -38805,8 +38805,8 @@ __ai float32x4_t vrnd64zq_f32(float32x4_t __p0) {
3880538805}
3880638806#else
3880738807__ai float32x4_t vrnd64zq_f32(float32x4_t __p0) {
38808 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3880938808 float32x4_t __ret;
38809 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3881038810 __ret = (float32x4_t) __builtin_neon_vrnd64zq_v((int8x16_t)__rev0, 41);
3881138811 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3881238812 return __ret;
......@@ -38821,8 +38821,8 @@ __ai float32x2_t vrnd64z_f32(float32x2_t __p0) {
3882138821}
3882238822#else
3882338823__ai float32x2_t vrnd64z_f32(float32x2_t __p0) {
38824 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3882538824 float32x2_t __ret;
38825 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3882638826 __ret = (float32x2_t) __builtin_neon_vrnd64z_v((int8x8_t)__rev0, 9);
3882738827 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3882838828 return __ret;
......@@ -38839,9 +38839,9 @@ __ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
3883938839}
3884038840#else
3884138841__ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
38842 float64x2_t __ret;
3884238843 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3884338844 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
38844 float64x2_t __ret;
3884538845 __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
3884638846 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3884738847 return __ret;
......@@ -38861,9 +38861,9 @@ __ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
3886138861}
3886238862#else
3886338863__ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
38864 float64x2_t __ret;
3886438865 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3886538866 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
38866 float64x2_t __ret;
3886738867 __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
3886838868 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3886938869 return __ret;
......@@ -39433,23 +39433,23 @@ __ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
3943339433#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC)
3943439434#ifdef __LITTLE_ENDIAN__
3943539435#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
39436 bfloat16x4_t __s0 = __p0; \
3943739436 bfloat16x8_t __ret; \
39437 bfloat16x4_t __s0 = __p0; \
3943839438 __ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 11); \
3943939439 __ret; \
3944039440})
3944139441#else
3944239442#define splatq_lane_bf16(__p0, __p1) __extension__ ({ \
39443 bfloat16x8_t __ret; \
3944339444 bfloat16x4_t __s0 = __p0; \
3944439445 bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
39445 bfloat16x8_t __ret; \
3944639446 __ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__rev0, __p1, 11); \
3944739447 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
3944839448 __ret; \
3944939449})
3945039450#define __noswap_splatq_lane_bf16(__p0, __p1) __extension__ ({ \
39451 bfloat16x4_t __s0 = __p0; \
3945239451 bfloat16x8_t __ret; \
39452 bfloat16x4_t __s0 = __p0; \
3945339453 __ret = (bfloat16x8_t) __builtin_neon_splatq_lane_v((int8x8_t)__s0, __p1, 11); \
3945439454 __ret; \
3945539455})
......@@ -39457,23 +39457,23 @@ __ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
3945739457
3945839458#ifdef __LITTLE_ENDIAN__
3945939459#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
39460 bfloat16x4_t __s0 = __p0; \
3946139460 bfloat16x4_t __ret; \
39461 bfloat16x4_t __s0 = __p0; \
3946239462 __ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 11); \
3946339463 __ret; \
3946439464})
3946539465#else
3946639466#define splat_lane_bf16(__p0, __p1) __extension__ ({ \
39467 bfloat16x4_t __ret; \
3946739468 bfloat16x4_t __s0 = __p0; \
3946839469 bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
39469 bfloat16x4_t __ret; \
3947039470 __ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__rev0, __p1, 11); \
3947139471 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3947239472 __ret; \
3947339473})
3947439474#define __noswap_splat_lane_bf16(__p0, __p1) __extension__ ({ \
39475 bfloat16x4_t __s0 = __p0; \
3947639475 bfloat16x4_t __ret; \
39476 bfloat16x4_t __s0 = __p0; \
3947739477 __ret = (bfloat16x4_t) __builtin_neon_splat_lane_v((int8x8_t)__s0, __p1, 11); \
3947839478 __ret; \
3947939479})
......@@ -39481,23 +39481,23 @@ __ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
3948139481
3948239482#ifdef __LITTLE_ENDIAN__
3948339483#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
39484 bfloat16x8_t __s0 = __p0; \
3948539484 bfloat16x8_t __ret; \
39485 bfloat16x8_t __s0 = __p0; \
3948639486 __ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 43); \
3948739487 __ret; \
3948839488})
3948939489#else
3949039490#define splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
39491 bfloat16x8_t __ret; \
3949139492 bfloat16x8_t __s0 = __p0; \
3949239493 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39493 bfloat16x8_t __ret; \
3949439494 __ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__rev0, __p1, 43); \
3949539495 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
3949639496 __ret; \
3949739497})
3949839498#define __noswap_splatq_laneq_bf16(__p0, __p1) __extension__ ({ \
39499 bfloat16x8_t __s0 = __p0; \
3950039499 bfloat16x8_t __ret; \
39500 bfloat16x8_t __s0 = __p0; \
3950139501 __ret = (bfloat16x8_t) __builtin_neon_splatq_laneq_v((int8x16_t)__s0, __p1, 43); \
3950239502 __ret; \
3950339503})
......@@ -39505,23 +39505,23 @@ __ai bfloat16x4_t vreinterpret_bf16_s16(int16x4_t __p0) {
3950539505
3950639506#ifdef __LITTLE_ENDIAN__
3950739507#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
39508 bfloat16x8_t __s0 = __p0; \
3950939508 bfloat16x4_t __ret; \
39509 bfloat16x8_t __s0 = __p0; \
3951039510 __ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 43); \
3951139511 __ret; \
3951239512})
3951339513#else
3951439514#define splat_laneq_bf16(__p0, __p1) __extension__ ({ \
39515 bfloat16x4_t __ret; \
3951539516 bfloat16x8_t __s0 = __p0; \
3951639517 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39517 bfloat16x4_t __ret; \
3951839518 __ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__rev0, __p1, 43); \
3951939519 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
3952039520 __ret; \
3952139521})
3952239522#define __noswap_splat_laneq_bf16(__p0, __p1) __extension__ ({ \
39523 bfloat16x8_t __s0 = __p0; \
3952439523 bfloat16x4_t __ret; \
39524 bfloat16x8_t __s0 = __p0; \
3952539525 __ret = (bfloat16x4_t) __builtin_neon_splat_laneq_v((int8x16_t)__s0, __p1, 43); \
3952639526 __ret; \
3952739527})
......@@ -39535,10 +39535,10 @@ __ai float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t _
3953539535}
3953639536#else
3953739537__ai float32x4_t vbfdotq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39538 float32x4_t __ret;
3953839539 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3953939540 bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3954039541 bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39541 float32x4_t __ret;
3954239542 __ret = (float32x4_t) __builtin_neon_vbfdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
3954339543 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3954439544 return __ret;
......@@ -39558,10 +39558,10 @@ __ai float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __
3955839558}
3955939559#else
3956039560__ai float32x2_t vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat16x4_t __p2) {
39561 float32x2_t __ret;
3956139562 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
3956239563 bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3956339564 bfloat16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
39564 float32x2_t __ret;
3956539565 __ret = (float32x2_t) __builtin_neon_vbfdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
3956639566 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
3956739567 return __ret;
......@@ -39575,10 +39575,10 @@ __ai float32x2_t __noswap_vbfdot_f32(float32x2_t __p0, bfloat16x4_t __p1, bfloat
3957539575
3957639576#ifdef __LITTLE_ENDIAN__
3957739577#define vbfdotq_lane_f32(__p0_142, __p1_142, __p2_142, __p3_142) __extension__ ({ \
39578 float32x4_t __ret_142; \
3957839579 float32x4_t __s0_142 = __p0_142; \
3957939580 bfloat16x8_t __s1_142 = __p1_142; \
3958039581 bfloat16x4_t __s2_142 = __p2_142; \
39581 float32x4_t __ret_142; \
3958239582bfloat16x4_t __reint_142 = __s2_142; \
3958339583float32x4_t __reint1_142 = splatq_lane_f32(*(float32x2_t *) &__reint_142, __p3_142); \
3958439584 __ret_142 = vbfdotq_f32(__s0_142, __s1_142, *(bfloat16x8_t *) &__reint1_142); \
......@@ -39586,13 +39586,13 @@ float32x4_t __reint1_142 = splatq_lane_f32(*(float32x2_t *) &__reint_142, __p3_1
3958639586})
3958739587#else
3958839588#define vbfdotq_lane_f32(__p0_143, __p1_143, __p2_143, __p3_143) __extension__ ({ \
39589 float32x4_t __ret_143; \
3958939590 float32x4_t __s0_143 = __p0_143; \
3959039591 bfloat16x8_t __s1_143 = __p1_143; \
3959139592 bfloat16x4_t __s2_143 = __p2_143; \
3959239593 float32x4_t __rev0_143; __rev0_143 = __builtin_shufflevector(__s0_143, __s0_143, 3, 2, 1, 0); \
3959339594 bfloat16x8_t __rev1_143; __rev1_143 = __builtin_shufflevector(__s1_143, __s1_143, 7, 6, 5, 4, 3, 2, 1, 0); \
3959439595 bfloat16x4_t __rev2_143; __rev2_143 = __builtin_shufflevector(__s2_143, __s2_143, 3, 2, 1, 0); \
39595 float32x4_t __ret_143; \
3959639596bfloat16x4_t __reint_143 = __rev2_143; \
3959739597float32x4_t __reint1_143 = __noswap_splatq_lane_f32(*(float32x2_t *) &__reint_143, __p3_143); \
3959839598 __ret_143 = __noswap_vbfdotq_f32(__rev0_143, __rev1_143, *(bfloat16x8_t *) &__reint1_143); \
......@@ -39603,10 +39603,10 @@ float32x4_t __reint1_143 = __noswap_splatq_lane_f32(*(float32x2_t *) &__reint_14
3960339603
3960439604#ifdef __LITTLE_ENDIAN__
3960539605#define vbfdot_lane_f32(__p0_144, __p1_144, __p2_144, __p3_144) __extension__ ({ \
39606 float32x2_t __ret_144; \
3960639607 float32x2_t __s0_144 = __p0_144; \
3960739608 bfloat16x4_t __s1_144 = __p1_144; \
3960839609 bfloat16x4_t __s2_144 = __p2_144; \
39609 float32x2_t __ret_144; \
3961039610bfloat16x4_t __reint_144 = __s2_144; \
3961139611float32x2_t __reint1_144 = splat_lane_f32(*(float32x2_t *) &__reint_144, __p3_144); \
3961239612 __ret_144 = vbfdot_f32(__s0_144, __s1_144, *(bfloat16x4_t *) &__reint1_144); \
......@@ -39614,13 +39614,13 @@ float32x2_t __reint1_144 = splat_lane_f32(*(float32x2_t *) &__reint_144, __p3_14
3961439614})
3961539615#else
3961639616#define vbfdot_lane_f32(__p0_145, __p1_145, __p2_145, __p3_145) __extension__ ({ \
39617 float32x2_t __ret_145; \
3961739618 float32x2_t __s0_145 = __p0_145; \
3961839619 bfloat16x4_t __s1_145 = __p1_145; \
3961939620 bfloat16x4_t __s2_145 = __p2_145; \
3962039621 float32x2_t __rev0_145; __rev0_145 = __builtin_shufflevector(__s0_145, __s0_145, 1, 0); \
3962139622 bfloat16x4_t __rev1_145; __rev1_145 = __builtin_shufflevector(__s1_145, __s1_145, 3, 2, 1, 0); \
3962239623 bfloat16x4_t __rev2_145; __rev2_145 = __builtin_shufflevector(__s2_145, __s2_145, 3, 2, 1, 0); \
39623 float32x2_t __ret_145; \
3962439624bfloat16x4_t __reint_145 = __rev2_145; \
3962539625float32x2_t __reint1_145 = __noswap_splat_lane_f32(*(float32x2_t *) &__reint_145, __p3_145); \
3962639626 __ret_145 = __noswap_vbfdot_f32(__rev0_145, __rev1_145, *(bfloat16x4_t *) &__reint1_145); \
......@@ -39631,10 +39631,10 @@ float32x2_t __reint1_145 = __noswap_splat_lane_f32(*(float32x2_t *) &__reint_145
3963139631
3963239632#ifdef __LITTLE_ENDIAN__
3963339633#define vbfdotq_laneq_f32(__p0_146, __p1_146, __p2_146, __p3_146) __extension__ ({ \
39634 float32x4_t __ret_146; \
3963439635 float32x4_t __s0_146 = __p0_146; \
3963539636 bfloat16x8_t __s1_146 = __p1_146; \
3963639637 bfloat16x8_t __s2_146 = __p2_146; \
39637 float32x4_t __ret_146; \
3963839638bfloat16x8_t __reint_146 = __s2_146; \
3963939639float32x4_t __reint1_146 = splatq_laneq_f32(*(float32x4_t *) &__reint_146, __p3_146); \
3964039640 __ret_146 = vbfdotq_f32(__s0_146, __s1_146, *(bfloat16x8_t *) &__reint1_146); \
......@@ -39642,13 +39642,13 @@ float32x4_t __reint1_146 = splatq_laneq_f32(*(float32x4_t *) &__reint_146, __p3_
3964239642})
3964339643#else
3964439644#define vbfdotq_laneq_f32(__p0_147, __p1_147, __p2_147, __p3_147) __extension__ ({ \
39645 float32x4_t __ret_147; \
3964539646 float32x4_t __s0_147 = __p0_147; \
3964639647 bfloat16x8_t __s1_147 = __p1_147; \
3964739648 bfloat16x8_t __s2_147 = __p2_147; \
3964839649 float32x4_t __rev0_147; __rev0_147 = __builtin_shufflevector(__s0_147, __s0_147, 3, 2, 1, 0); \
3964939650 bfloat16x8_t __rev1_147; __rev1_147 = __builtin_shufflevector(__s1_147, __s1_147, 7, 6, 5, 4, 3, 2, 1, 0); \
3965039651 bfloat16x8_t __rev2_147; __rev2_147 = __builtin_shufflevector(__s2_147, __s2_147, 7, 6, 5, 4, 3, 2, 1, 0); \
39651 float32x4_t __ret_147; \
3965239652bfloat16x8_t __reint_147 = __rev2_147; \
3965339653float32x4_t __reint1_147 = __noswap_splatq_laneq_f32(*(float32x4_t *) &__reint_147, __p3_147); \
3965439654 __ret_147 = __noswap_vbfdotq_f32(__rev0_147, __rev1_147, *(bfloat16x8_t *) &__reint1_147); \
......@@ -39659,10 +39659,10 @@ float32x4_t __reint1_147 = __noswap_splatq_laneq_f32(*(float32x4_t *) &__reint_1
3965939659
3966039660#ifdef __LITTLE_ENDIAN__
3966139661#define vbfdot_laneq_f32(__p0_148, __p1_148, __p2_148, __p3_148) __extension__ ({ \
39662 float32x2_t __ret_148; \
3966239663 float32x2_t __s0_148 = __p0_148; \
3966339664 bfloat16x4_t __s1_148 = __p1_148; \
3966439665 bfloat16x8_t __s2_148 = __p2_148; \
39665 float32x2_t __ret_148; \
3966639666bfloat16x8_t __reint_148 = __s2_148; \
3966739667float32x2_t __reint1_148 = splat_laneq_f32(*(float32x4_t *) &__reint_148, __p3_148); \
3966839668 __ret_148 = vbfdot_f32(__s0_148, __s1_148, *(bfloat16x4_t *) &__reint1_148); \
......@@ -39670,13 +39670,13 @@ float32x2_t __reint1_148 = splat_laneq_f32(*(float32x4_t *) &__reint_148, __p3_1
3967039670})
3967139671#else
3967239672#define vbfdot_laneq_f32(__p0_149, __p1_149, __p2_149, __p3_149) __extension__ ({ \
39673 float32x2_t __ret_149; \
3967339674 float32x2_t __s0_149 = __p0_149; \
3967439675 bfloat16x4_t __s1_149 = __p1_149; \
3967539676 bfloat16x8_t __s2_149 = __p2_149; \
3967639677 float32x2_t __rev0_149; __rev0_149 = __builtin_shufflevector(__s0_149, __s0_149, 1, 0); \
3967739678 bfloat16x4_t __rev1_149; __rev1_149 = __builtin_shufflevector(__s1_149, __s1_149, 3, 2, 1, 0); \
3967839679 bfloat16x8_t __rev2_149; __rev2_149 = __builtin_shufflevector(__s2_149, __s2_149, 7, 6, 5, 4, 3, 2, 1, 0); \
39679 float32x2_t __ret_149; \
3968039680bfloat16x8_t __reint_149 = __rev2_149; \
3968139681float32x2_t __reint1_149 = __noswap_splat_laneq_f32(*(float32x4_t *) &__reint_149, __p3_149); \
3968239682 __ret_149 = __noswap_vbfdot_f32(__rev0_149, __rev1_149, *(bfloat16x4_t *) &__reint1_149); \
......@@ -39693,10 +39693,10 @@ __ai float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t
3969339693}
3969439694#else
3969539695__ai float32x4_t vbfmlalbq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39696 float32x4_t __ret;
3969639697 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3969739698 bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3969839699 bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39699 float32x4_t __ret;
3970039700 __ret = (float32x4_t) __builtin_neon_vbfmlalbq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
3970139701 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3970239702 return __ret;
......@@ -39716,10 +39716,10 @@ __ai float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t
3971639716}
3971739717#else
3971839718__ai float32x4_t vbfmlaltq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39719 float32x4_t __ret;
3971939720 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3972039721 bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3972139722 bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39722 float32x4_t __ret;
3972339723 __ret = (float32x4_t) __builtin_neon_vbfmlaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
3972439724 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3972539725 return __ret;
......@@ -39739,10 +39739,10 @@ __ai float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t
3973939739}
3974039740#else
3974139741__ai float32x4_t vbfmmlaq_f32(float32x4_t __p0, bfloat16x8_t __p1, bfloat16x8_t __p2) {
39742 float32x4_t __ret;
3974239743 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3974339744 bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3974439745 bfloat16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
39745 float32x4_t __ret;
3974639746 __ret = (float32x4_t) __builtin_neon_vbfmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
3974739747 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3974839748 return __ret;
......@@ -39757,9 +39757,9 @@ __ai bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
3975739757}
3975839758#else
3975939759__ai bfloat16x8_t vcombine_bf16(bfloat16x4_t __p0, bfloat16x4_t __p1) {
39760 bfloat16x8_t __ret;
3976039761 bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
3976139762 bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39762 bfloat16x8_t __ret;
3976339763 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 1, 2, 3, 4, 5, 6, 7);
3976439764 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
3976539765 return __ret;
......@@ -39787,8 +39787,8 @@ int32x4_t __reint1_150 = vshll_n_s16(*(int16x4_t *) &__reint_150, 16);
3978739787}
3978839788#else
3978939789__ai float32x4_t vcvt_f32_bf16(bfloat16x4_t __p0_151) {
39790 bfloat16x4_t __rev0_151; __rev0_151 = __builtin_shufflevector(__p0_151, __p0_151, 3, 2, 1, 0);
3979139790 float32x4_t __ret_151;
39791 bfloat16x4_t __rev0_151; __rev0_151 = __builtin_shufflevector(__p0_151, __p0_151, 3, 2, 1, 0);
3979239792bfloat16x4_t __reint_151 = __rev0_151;
3979339793int32x4_t __reint1_151 = __noswap_vshll_n_s16(*(int16x4_t *) &__reint_151, 16);
3979439794 __ret_151 = *(float32x4_t *) &__reint1_151;
......@@ -39818,16 +39818,16 @@ __ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
3981839818}
3981939819#ifdef __LITTLE_ENDIAN__
3982039820#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
39821 bfloat16x4_t __s0 = __p0; \
3982239821 bfloat16_t __ret; \
39822 bfloat16x4_t __s0 = __p0; \
3982339823 __ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__s0, __p1); \
3982439824 __ret; \
3982539825})
3982639826#else
3982739827#define vduph_lane_bf16(__p0, __p1) __extension__ ({ \
39828 bfloat16_t __ret; \
3982839829 bfloat16x4_t __s0 = __p0; \
3982939830 bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
39830 bfloat16_t __ret; \
3983139831 __ret = (bfloat16_t) __builtin_neon_vduph_lane_bf16((bfloat16x4_t)__rev0, __p1); \
3983239832 __ret; \
3983339833})
......@@ -39835,16 +39835,16 @@ __ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
3983539835
3983639836#ifdef __LITTLE_ENDIAN__
3983739837#define vdupq_lane_bf16(__p0_153, __p1_153) __extension__ ({ \
39838 bfloat16x4_t __s0_153 = __p0_153; \
3983939838 bfloat16x8_t __ret_153; \
39839 bfloat16x4_t __s0_153 = __p0_153; \
3984039840 __ret_153 = splatq_lane_bf16(__s0_153, __p1_153); \
3984139841 __ret_153; \
3984239842})
3984339843#else
3984439844#define vdupq_lane_bf16(__p0_154, __p1_154) __extension__ ({ \
39845 bfloat16x8_t __ret_154; \
3984539846 bfloat16x4_t __s0_154 = __p0_154; \
3984639847 bfloat16x4_t __rev0_154; __rev0_154 = __builtin_shufflevector(__s0_154, __s0_154, 3, 2, 1, 0); \
39847 bfloat16x8_t __ret_154; \
3984839848 __ret_154 = __noswap_splatq_lane_bf16(__rev0_154, __p1_154); \
3984939849 __ret_154 = __builtin_shufflevector(__ret_154, __ret_154, 7, 6, 5, 4, 3, 2, 1, 0); \
3985039850 __ret_154; \
......@@ -39853,16 +39853,16 @@ __ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
3985339853
3985439854#ifdef __LITTLE_ENDIAN__
3985539855#define vdup_lane_bf16(__p0_155, __p1_155) __extension__ ({ \
39856 bfloat16x4_t __s0_155 = __p0_155; \
3985739856 bfloat16x4_t __ret_155; \
39857 bfloat16x4_t __s0_155 = __p0_155; \
3985839858 __ret_155 = splat_lane_bf16(__s0_155, __p1_155); \
3985939859 __ret_155; \
3986039860})
3986139861#else
3986239862#define vdup_lane_bf16(__p0_156, __p1_156) __extension__ ({ \
39863 bfloat16x4_t __ret_156; \
3986339864 bfloat16x4_t __s0_156 = __p0_156; \
3986439865 bfloat16x4_t __rev0_156; __rev0_156 = __builtin_shufflevector(__s0_156, __s0_156, 3, 2, 1, 0); \
39865 bfloat16x4_t __ret_156; \
3986639866 __ret_156 = __noswap_splat_lane_bf16(__rev0_156, __p1_156); \
3986739867 __ret_156 = __builtin_shufflevector(__ret_156, __ret_156, 3, 2, 1, 0); \
3986839868 __ret_156; \
......@@ -39871,16 +39871,16 @@ __ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
3987139871
3987239872#ifdef __LITTLE_ENDIAN__
3987339873#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
39874 bfloat16x8_t __s0 = __p0; \
3987539874 bfloat16_t __ret; \
39875 bfloat16x8_t __s0 = __p0; \
3987639876 __ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__s0, __p1); \
3987739877 __ret; \
3987839878})
3987939879#else
3988039880#define vduph_laneq_bf16(__p0, __p1) __extension__ ({ \
39881 bfloat16_t __ret; \
3988139882 bfloat16x8_t __s0 = __p0; \
3988239883 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39883 bfloat16_t __ret; \
3988439884 __ret = (bfloat16_t) __builtin_neon_vduph_laneq_bf16((bfloat16x8_t)__rev0, __p1); \
3988539885 __ret; \
3988639886})
......@@ -39888,16 +39888,16 @@ __ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
3988839888
3988939889#ifdef __LITTLE_ENDIAN__
3989039890#define vdupq_laneq_bf16(__p0_157, __p1_157) __extension__ ({ \
39891 bfloat16x8_t __s0_157 = __p0_157; \
3989239891 bfloat16x8_t __ret_157; \
39892 bfloat16x8_t __s0_157 = __p0_157; \
3989339893 __ret_157 = splatq_laneq_bf16(__s0_157, __p1_157); \
3989439894 __ret_157; \
3989539895})
3989639896#else
3989739897#define vdupq_laneq_bf16(__p0_158, __p1_158) __extension__ ({ \
39898 bfloat16x8_t __ret_158; \
3989839899 bfloat16x8_t __s0_158 = __p0_158; \
3989939900 bfloat16x8_t __rev0_158; __rev0_158 = __builtin_shufflevector(__s0_158, __s0_158, 7, 6, 5, 4, 3, 2, 1, 0); \
39900 bfloat16x8_t __ret_158; \
3990139901 __ret_158 = __noswap_splatq_laneq_bf16(__rev0_158, __p1_158); \
3990239902 __ret_158 = __builtin_shufflevector(__ret_158, __ret_158, 7, 6, 5, 4, 3, 2, 1, 0); \
3990339903 __ret_158; \
......@@ -39906,16 +39906,16 @@ __ai bfloat16_t vcvth_bf16_f32(float32_t __p0) {
3990639906
3990739907#ifdef __LITTLE_ENDIAN__
3990839908#define vdup_laneq_bf16(__p0_159, __p1_159) __extension__ ({ \
39909 bfloat16x8_t __s0_159 = __p0_159; \
3991039909 bfloat16x4_t __ret_159; \
39910 bfloat16x8_t __s0_159 = __p0_159; \
3991139911 __ret_159 = splat_laneq_bf16(__s0_159, __p1_159); \
3991239912 __ret_159; \
3991339913})
3991439914#else
3991539915#define vdup_laneq_bf16(__p0_160, __p1_160) __extension__ ({ \
39916 bfloat16x4_t __ret_160; \
3991639917 bfloat16x8_t __s0_160 = __p0_160; \
3991739918 bfloat16x8_t __rev0_160; __rev0_160 = __builtin_shufflevector(__s0_160, __s0_160, 7, 6, 5, 4, 3, 2, 1, 0); \
39918 bfloat16x4_t __ret_160; \
3991939919 __ret_160 = __noswap_splat_laneq_bf16(__rev0_160, __p1_160); \
3992039920 __ret_160 = __builtin_shufflevector(__ret_160, __ret_160, 3, 2, 1, 0); \
3992139921 __ret_160; \
......@@ -39960,8 +39960,8 @@ __ai bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
3996039960}
3996139961#else
3996239962__ai bfloat16x4_t vget_high_bf16(bfloat16x8_t __p0) {
39963 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3996439963 bfloat16x4_t __ret;
39964 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
3996539965 __ret = __builtin_shufflevector(__rev0, __rev0, 4, 5, 6, 7);
3996639966 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
3996739967 return __ret;
......@@ -39975,22 +39975,22 @@ __ai bfloat16x4_t __noswap_vget_high_bf16(bfloat16x8_t __p0) {
3997539975
3997639976#ifdef __LITTLE_ENDIAN__
3997739977#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
39978 bfloat16x8_t __s0 = __p0; \
3997939978 bfloat16_t __ret; \
39979 bfloat16x8_t __s0 = __p0; \
3998039980 __ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
3998139981 __ret; \
3998239982})
3998339983#else
3998439984#define vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
39985 bfloat16_t __ret; \
3998539986 bfloat16x8_t __s0 = __p0; \
3998639987 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
39987 bfloat16_t __ret; \
3998839988 __ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__rev0, __p1); \
3998939989 __ret; \
3999039990})
3999139991#define __noswap_vgetq_lane_bf16(__p0, __p1) __extension__ ({ \
39992 bfloat16x8_t __s0 = __p0; \
3999339992 bfloat16_t __ret; \
39993 bfloat16x8_t __s0 = __p0; \
3999439994 __ret = (bfloat16_t) __builtin_neon_vgetq_lane_bf16((bfloat16x8_t)__s0, __p1); \
3999539995 __ret; \
3999639996})
......@@ -39998,22 +39998,22 @@ __ai bfloat16x4_t __noswap_vget_high_bf16(bfloat16x8_t __p0) {
3999839998
3999939999#ifdef __LITTLE_ENDIAN__
4000040000#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
40001 bfloat16x4_t __s0 = __p0; \
4000240001 bfloat16_t __ret; \
40002 bfloat16x4_t __s0 = __p0; \
4000340003 __ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
4000440004 __ret; \
4000540005})
4000640006#else
4000740007#define vget_lane_bf16(__p0, __p1) __extension__ ({ \
40008 bfloat16_t __ret; \
4000840009 bfloat16x4_t __s0 = __p0; \
4000940010 bfloat16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
40010 bfloat16_t __ret; \
4001140011 __ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__rev0, __p1); \
4001240012 __ret; \
4001340013})
4001440014#define __noswap_vget_lane_bf16(__p0, __p1) __extension__ ({ \
40015 bfloat16x4_t __s0 = __p0; \
4001640015 bfloat16_t __ret; \
40016 bfloat16x4_t __s0 = __p0; \
4001740017 __ret = (bfloat16_t) __builtin_neon_vget_lane_bf16((bfloat16x4_t)__s0, __p1); \
4001840018 __ret; \
4001940019})
......@@ -40027,8 +40027,8 @@ __ai bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
4002740027}
4002840028#else
4002940029__ai bfloat16x4_t vget_low_bf16(bfloat16x8_t __p0) {
40030 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4003140030 bfloat16x4_t __ret;
40031 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4003240032 __ret = __builtin_shufflevector(__rev0, __rev0, 0, 1, 2, 3);
4003340033 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4003440034 return __ret;
......@@ -40102,16 +40102,16 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4010240102
4010340103#ifdef __LITTLE_ENDIAN__
4010440104#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40105 bfloat16x8_t __s1 = __p1; \
4010640105 bfloat16x8_t __ret; \
40106 bfloat16x8_t __s1 = __p1; \
4010740107 __ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 43); \
4010840108 __ret; \
4010940109})
4011040110#else
4011140111#define vld1q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40112 bfloat16x8_t __ret; \
4011240113 bfloat16x8_t __s1 = __p1; \
4011340114 bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
40114 bfloat16x8_t __ret; \
4011540115 __ret = (bfloat16x8_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 43); \
4011640116 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4011740117 __ret; \
......@@ -40120,16 +40120,16 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4012040120
4012140121#ifdef __LITTLE_ENDIAN__
4012240122#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40123 bfloat16x4_t __s1 = __p1; \
4012440123 bfloat16x4_t __ret; \
40124 bfloat16x4_t __s1 = __p1; \
4012540125 __ret = (bfloat16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 11); \
4012640126 __ret; \
4012740127})
4012840128#else
4012940129#define vld1_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40130 bfloat16x4_t __ret; \
4013040131 bfloat16x4_t __s1 = __p1; \
4013140132 bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
40132 bfloat16x4_t __ret; \
4013340133 __ret = (bfloat16x4_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__rev1, __p2, 11); \
4013440134 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4013540135 __ret; \
......@@ -40314,18 +40314,18 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4031440314
4031540315#ifdef __LITTLE_ENDIAN__
4031640316#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40317 bfloat16x8x2_t __s1 = __p1; \
4031840317 bfloat16x8x2_t __ret; \
40318 bfloat16x8x2_t __s1 = __p1; \
4031940319 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 43); \
4032040320 __ret; \
4032140321})
4032240322#else
4032340323#define vld2q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40324 bfloat16x8x2_t __ret; \
4032440325 bfloat16x8x2_t __s1 = __p1; \
4032540326 bfloat16x8x2_t __rev1; \
4032640327 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
4032740328 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
40328 bfloat16x8x2_t __ret; \
4032940329 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 43); \
4033040330 \
4033140331 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -40336,18 +40336,18 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4033640336
4033740337#ifdef __LITTLE_ENDIAN__
4033840338#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40339 bfloat16x4x2_t __s1 = __p1; \
4034040339 bfloat16x4x2_t __ret; \
40340 bfloat16x4x2_t __s1 = __p1; \
4034140341 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 11); \
4034240342 __ret; \
4034340343})
4034440344#else
4034540345#define vld2_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40346 bfloat16x4x2_t __ret; \
4034640347 bfloat16x4x2_t __s1 = __p1; \
4034740348 bfloat16x4x2_t __rev1; \
4034840349 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
4034940350 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
40350 bfloat16x4x2_t __ret; \
4035140351 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], __p2, 11); \
4035240352 \
4035340353 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -40430,19 +40430,19 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4043040430
4043140431#ifdef __LITTLE_ENDIAN__
4043240432#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40433 bfloat16x8x3_t __s1 = __p1; \
4043440433 bfloat16x8x3_t __ret; \
40434 bfloat16x8x3_t __s1 = __p1; \
4043540435 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 43); \
4043640436 __ret; \
4043740437})
4043840438#else
4043940439#define vld3q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40440 bfloat16x8x3_t __ret; \
4044040441 bfloat16x8x3_t __s1 = __p1; \
4044140442 bfloat16x8x3_t __rev1; \
4044240443 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
4044340444 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
4044440445 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
40445 bfloat16x8x3_t __ret; \
4044640446 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 43); \
4044740447 \
4044840448 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -40454,19 +40454,19 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4045440454
4045540455#ifdef __LITTLE_ENDIAN__
4045640456#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40457 bfloat16x4x3_t __s1 = __p1; \
4045840457 bfloat16x4x3_t __ret; \
40458 bfloat16x4x3_t __s1 = __p1; \
4045940459 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 11); \
4046040460 __ret; \
4046140461})
4046240462#else
4046340463#define vld3_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40464 bfloat16x4x3_t __ret; \
4046440465 bfloat16x4x3_t __s1 = __p1; \
4046540466 bfloat16x4x3_t __rev1; \
4046640467 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
4046740468 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
4046840469 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
40469 bfloat16x4x3_t __ret; \
4047040470 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__rev1.val[0], (int8x8_t)__rev1.val[1], (int8x8_t)__rev1.val[2], __p2, 11); \
4047140471 \
4047240472 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -40554,20 +40554,20 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4055440554
4055540555#ifdef __LITTLE_ENDIAN__
4055640556#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40557 bfloat16x8x4_t __s1 = __p1; \
4055840557 bfloat16x8x4_t __ret; \
40558 bfloat16x8x4_t __s1 = __p1; \
4055940559 __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); \
4056040560 __ret; \
4056140561})
4056240562#else
4056340563#define vld4q_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40564 bfloat16x8x4_t __ret; \
4056440565 bfloat16x8x4_t __s1 = __p1; \
4056540566 bfloat16x8x4_t __rev1; \
4056640567 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
4056740568 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 7, 6, 5, 4, 3, 2, 1, 0); \
4056840569 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 7, 6, 5, 4, 3, 2, 1, 0); \
4056940570 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 7, 6, 5, 4, 3, 2, 1, 0); \
40570 bfloat16x8x4_t __ret; \
4057140571 __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); \
4057240572 \
4057340573 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0); \
......@@ -40580,20 +40580,20 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4058040580
4058140581#ifdef __LITTLE_ENDIAN__
4058240582#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40583 bfloat16x4x4_t __s1 = __p1; \
4058440583 bfloat16x4x4_t __ret; \
40584 bfloat16x4x4_t __s1 = __p1; \
4058540585 __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); \
4058640586 __ret; \
4058740587})
4058840588#else
4058940589#define vld4_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40590 bfloat16x4x4_t __ret; \
4059040591 bfloat16x4x4_t __s1 = __p1; \
4059140592 bfloat16x4x4_t __rev1; \
4059240593 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 3, 2, 1, 0); \
4059340594 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 3, 2, 1, 0); \
4059440595 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 3, 2, 1, 0); \
4059540596 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 3, 2, 1, 0); \
40596 bfloat16x4x4_t __ret; \
4059740597 __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); \
4059840598 \
4059940599 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0); \
......@@ -40606,26 +40606,26 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4060640606
4060740607#ifdef __LITTLE_ENDIAN__
4060840608#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40609 bfloat16x8_t __ret; \
4060940610 bfloat16_t __s0 = __p0; \
4061040611 bfloat16x8_t __s1 = __p1; \
40611 bfloat16x8_t __ret; \
4061240612 __ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
4061340613 __ret; \
4061440614})
4061540615#else
4061640616#define vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40617 bfloat16x8_t __ret; \
4061740618 bfloat16_t __s0 = __p0; \
4061840619 bfloat16x8_t __s1 = __p1; \
4061940620 bfloat16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
40620 bfloat16x8_t __ret; \
4062140621 __ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__rev1, __p2); \
4062240622 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4062340623 __ret; \
4062440624})
4062540625#define __noswap_vsetq_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40626 bfloat16x8_t __ret; \
4062640627 bfloat16_t __s0 = __p0; \
4062740628 bfloat16x8_t __s1 = __p1; \
40628 bfloat16x8_t __ret; \
4062940629 __ret = (bfloat16x8_t) __builtin_neon_vsetq_lane_bf16(__s0, (bfloat16x8_t)__s1, __p2); \
4063040630 __ret; \
4063140631})
......@@ -40633,26 +40633,26 @@ __ai bfloat16x4_t __noswap_vget_low_bf16(bfloat16x8_t __p0) {
4063340633
4063440634#ifdef __LITTLE_ENDIAN__
4063540635#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40636 bfloat16x4_t __ret; \
4063640637 bfloat16_t __s0 = __p0; \
4063740638 bfloat16x4_t __s1 = __p1; \
40638 bfloat16x4_t __ret; \
4063940639 __ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
4064040640 __ret; \
4064140641})
4064240642#else
4064340643#define vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40644 bfloat16x4_t __ret; \
4064440645 bfloat16_t __s0 = __p0; \
4064540646 bfloat16x4_t __s1 = __p1; \
4064640647 bfloat16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
40647 bfloat16x4_t __ret; \
4064840648 __ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__rev1, __p2); \
4064940649 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4065040650 __ret; \
4065140651})
4065240652#define __noswap_vset_lane_bf16(__p0, __p1, __p2) __extension__ ({ \
40653 bfloat16x4_t __ret; \
4065340654 bfloat16_t __s0 = __p0; \
4065440655 bfloat16x4_t __s1 = __p1; \
40655 bfloat16x4_t __ret; \
4065640656 __ret = (bfloat16x4_t) __builtin_neon_vset_lane_bf16(__s0, (bfloat16x4_t)__s1, __p2); \
4065740657 __ret; \
4065840658})
......@@ -41008,8 +41008,8 @@ __ai bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
4100841008}
4100941009#else
4101041010__ai bfloat16x4_t __a32_vcvt_bf16_f32(float32x4_t __p0) {
41011 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4101241011 bfloat16x4_t __ret;
41012 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4101341013 __ret = (bfloat16x4_t) __builtin_neon___a32_vcvt_bf16_v((int8x16_t)__rev0, 11);
4101441014 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4101541015 return __ret;
......@@ -41029,8 +41029,8 @@ __ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
4102941029}
4103041030#else
4103141031__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
41032 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4103341032 bfloat16x4_t __ret;
41033 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4103441034 __ret = __noswap___a32_vcvt_bf16_f32(__rev0);
4103541035 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4103641036 return __ret;
......@@ -41045,9 +41045,9 @@ __ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
4104541045}
4104641046#else
4104741047__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
41048 bfloat16x8_t __ret;
4104841049 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4104941050 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41050 bfloat16x8_t __ret;
4105141051 __ret = __noswap_vcombine_bf16(__noswap___a32_vcvt_bf16_f32(__rev1), __noswap_vget_low_bf16(__rev0));
4105241052 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4105341053 return __ret;
......@@ -41062,8 +41062,8 @@ __ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
4106241062}
4106341063#else
4106441064__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
41065 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4106641065 bfloat16x8_t __ret;
41066 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4106741067 __ret = __noswap_vcombine_bf16((bfloat16x4_t)(0ULL), __noswap___a32_vcvt_bf16_f32(__rev0));
4106841068 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4106941069 return __ret;
......@@ -41080,8 +41080,8 @@ __ai bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
4108041080}
4108141081#else
4108241082__ai bfloat16x8_t __a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
41083 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4108441083 bfloat16x8_t __ret;
41084 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4108541085 __ret = (bfloat16x8_t) __builtin_neon___a64_vcvtq_low_bf16_v((int8x16_t)__rev0, 43);
4108641086 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4108741087 return __ret;
......@@ -41095,19 +41095,19 @@ __ai bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
4109541095
4109641096#ifdef __LITTLE_ENDIAN__
4109741097#define vcopyq_lane_bf16(__p0_161, __p1_161, __p2_161, __p3_161) __extension__ ({ \
41098 bfloat16x8_t __ret_161; \
4109841099 bfloat16x8_t __s0_161 = __p0_161; \
4109941100 bfloat16x4_t __s2_161 = __p2_161; \
41100 bfloat16x8_t __ret_161; \
4110141101 __ret_161 = vsetq_lane_bf16(vget_lane_bf16(__s2_161, __p3_161), __s0_161, __p1_161); \
4110241102 __ret_161; \
4110341103})
4110441104#else
4110541105#define vcopyq_lane_bf16(__p0_162, __p1_162, __p2_162, __p3_162) __extension__ ({ \
41106 bfloat16x8_t __ret_162; \
4110641107 bfloat16x8_t __s0_162 = __p0_162; \
4110741108 bfloat16x4_t __s2_162 = __p2_162; \
4110841109 bfloat16x8_t __rev0_162; __rev0_162 = __builtin_shufflevector(__s0_162, __s0_162, 7, 6, 5, 4, 3, 2, 1, 0); \
4110941110 bfloat16x4_t __rev2_162; __rev2_162 = __builtin_shufflevector(__s2_162, __s2_162, 3, 2, 1, 0); \
41110 bfloat16x8_t __ret_162; \
4111141111 __ret_162 = __noswap_vsetq_lane_bf16(__noswap_vget_lane_bf16(__rev2_162, __p3_162), __rev0_162, __p1_162); \
4111241112 __ret_162 = __builtin_shufflevector(__ret_162, __ret_162, 7, 6, 5, 4, 3, 2, 1, 0); \
4111341113 __ret_162; \
......@@ -41116,19 +41116,19 @@ __ai bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
4111641116
4111741117#ifdef __LITTLE_ENDIAN__
4111841118#define vcopy_lane_bf16(__p0_163, __p1_163, __p2_163, __p3_163) __extension__ ({ \
41119 bfloat16x4_t __ret_163; \
4111941120 bfloat16x4_t __s0_163 = __p0_163; \
4112041121 bfloat16x4_t __s2_163 = __p2_163; \
41121 bfloat16x4_t __ret_163; \
4112241122 __ret_163 = vset_lane_bf16(vget_lane_bf16(__s2_163, __p3_163), __s0_163, __p1_163); \
4112341123 __ret_163; \
4112441124})
4112541125#else
4112641126#define vcopy_lane_bf16(__p0_164, __p1_164, __p2_164, __p3_164) __extension__ ({ \
41127 bfloat16x4_t __ret_164; \
4112741128 bfloat16x4_t __s0_164 = __p0_164; \
4112841129 bfloat16x4_t __s2_164 = __p2_164; \
4112941130 bfloat16x4_t __rev0_164; __rev0_164 = __builtin_shufflevector(__s0_164, __s0_164, 3, 2, 1, 0); \
4113041131 bfloat16x4_t __rev2_164; __rev2_164 = __builtin_shufflevector(__s2_164, __s2_164, 3, 2, 1, 0); \
41131 bfloat16x4_t __ret_164; \
4113241132 __ret_164 = __noswap_vset_lane_bf16(__noswap_vget_lane_bf16(__rev2_164, __p3_164), __rev0_164, __p1_164); \
4113341133 __ret_164 = __builtin_shufflevector(__ret_164, __ret_164, 3, 2, 1, 0); \
4113441134 __ret_164; \
......@@ -41137,19 +41137,19 @@ __ai bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
4113741137
4113841138#ifdef __LITTLE_ENDIAN__
4113941139#define vcopyq_laneq_bf16(__p0_165, __p1_165, __p2_165, __p3_165) __extension__ ({ \
41140 bfloat16x8_t __ret_165; \
4114041141 bfloat16x8_t __s0_165 = __p0_165; \
4114141142 bfloat16x8_t __s2_165 = __p2_165; \
41142 bfloat16x8_t __ret_165; \
4114341143 __ret_165 = vsetq_lane_bf16(vgetq_lane_bf16(__s2_165, __p3_165), __s0_165, __p1_165); \
4114441144 __ret_165; \
4114541145})
4114641146#else
4114741147#define vcopyq_laneq_bf16(__p0_166, __p1_166, __p2_166, __p3_166) __extension__ ({ \
41148 bfloat16x8_t __ret_166; \
4114841149 bfloat16x8_t __s0_166 = __p0_166; \
4114941150 bfloat16x8_t __s2_166 = __p2_166; \
4115041151 bfloat16x8_t __rev0_166; __rev0_166 = __builtin_shufflevector(__s0_166, __s0_166, 7, 6, 5, 4, 3, 2, 1, 0); \
4115141152 bfloat16x8_t __rev2_166; __rev2_166 = __builtin_shufflevector(__s2_166, __s2_166, 7, 6, 5, 4, 3, 2, 1, 0); \
41152 bfloat16x8_t __ret_166; \
4115341153 __ret_166 = __noswap_vsetq_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_166, __p3_166), __rev0_166, __p1_166); \
4115441154 __ret_166 = __builtin_shufflevector(__ret_166, __ret_166, 7, 6, 5, 4, 3, 2, 1, 0); \
4115541155 __ret_166; \
......@@ -41158,19 +41158,19 @@ __ai bfloat16x8_t __noswap___a64_vcvtq_low_bf16_f32(float32x4_t __p0) {
4115841158
4115941159#ifdef __LITTLE_ENDIAN__
4116041160#define vcopy_laneq_bf16(__p0_167, __p1_167, __p2_167, __p3_167) __extension__ ({ \
41161 bfloat16x4_t __ret_167; \
4116141162 bfloat16x4_t __s0_167 = __p0_167; \
4116241163 bfloat16x8_t __s2_167 = __p2_167; \
41163 bfloat16x4_t __ret_167; \
4116441164 __ret_167 = vset_lane_bf16(vgetq_lane_bf16(__s2_167, __p3_167), __s0_167, __p1_167); \
4116541165 __ret_167; \
4116641166})
4116741167#else
4116841168#define vcopy_laneq_bf16(__p0_168, __p1_168, __p2_168, __p3_168) __extension__ ({ \
41169 bfloat16x4_t __ret_168; \
4116941170 bfloat16x4_t __s0_168 = __p0_168; \
4117041171 bfloat16x8_t __s2_168 = __p2_168; \
4117141172 bfloat16x4_t __rev0_168; __rev0_168 = __builtin_shufflevector(__s0_168, __s0_168, 3, 2, 1, 0); \
4117241173 bfloat16x8_t __rev2_168; __rev2_168 = __builtin_shufflevector(__s2_168, __s2_168, 7, 6, 5, 4, 3, 2, 1, 0); \
41173 bfloat16x4_t __ret_168; \
4117441174 __ret_168 = __noswap_vset_lane_bf16(__noswap_vgetq_lane_bf16(__rev2_168, __p3_168), __rev0_168, __p1_168); \
4117541175 __ret_168 = __builtin_shufflevector(__ret_168, __ret_168, 3, 2, 1, 0); \
4117641176 __ret_168; \
......@@ -41185,8 +41185,8 @@ __ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
4118541185}
4118641186#else
4118741187__ai bfloat16x4_t vcvt_bf16_f32(float32x4_t __p0) {
41188 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4118941188 bfloat16x4_t __ret;
41189 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4119041190 __ret = __noswap_vget_low_bf16(__noswap___a64_vcvtq_low_bf16_f32(__rev0));
4119141191 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4119241192 return __ret;
......@@ -41201,9 +41201,9 @@ __ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
4120141201}
4120241202#else
4120341203__ai bfloat16x8_t vcvtq_high_bf16_f32(bfloat16x8_t __p0, float32x4_t __p1) {
41204 bfloat16x8_t __ret;
4120441205 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4120541206 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41206 bfloat16x8_t __ret;
4120741207 __ret = (bfloat16x8_t) __builtin_neon_vcvtq_high_bf16_v((int8x16_t)__rev0, (int8x16_t)__rev1, 43);
4120841208 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4120941209 return __ret;
......@@ -41218,8 +41218,8 @@ __ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
4121841218}
4121941219#else
4122041220__ai bfloat16x8_t vcvtq_low_bf16_f32(float32x4_t __p0) {
41221 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4122241221 bfloat16x8_t __ret;
41222 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4122341223 __ret = __noswap___a64_vcvtq_low_bf16_f32(__rev0);
4122441224 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4122541225 return __ret;
......@@ -41236,9 +41236,9 @@ __ai float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
4123641236}
4123741237#else
4123841238__ai float32x2_t vcadd_rot270_f32(float32x2_t __p0, float32x2_t __p1) {
41239 float32x2_t __ret;
4123941240 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4124041241 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41241 float32x2_t __ret;
4124241242 __ret = (float32x2_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
4124341243 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4124441244 return __ret;
......@@ -41253,9 +41253,9 @@ __ai float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
4125341253}
4125441254#else
4125541255__ai float32x2_t vcadd_rot90_f32(float32x2_t __p0, float32x2_t __p1) {
41256 float32x2_t __ret;
4125641257 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4125741258 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41258 float32x2_t __ret;
4125941259 __ret = (float32x2_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
4126041260 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4126141261 return __ret;
......@@ -41270,9 +41270,9 @@ __ai float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
4127041270}
4127141271#else
4127241272__ai float32x4_t vcaddq_rot270_f32(float32x4_t __p0, float32x4_t __p1) {
41273 float32x4_t __ret;
4127341274 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4127441275 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41275 float32x4_t __ret;
4127641276 __ret = (float32x4_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
4127741277 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4127841278 return __ret;
......@@ -41287,9 +41287,9 @@ __ai float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
4128741287}
4128841288#else
4128941289__ai float32x4_t vcaddq_rot90_f32(float32x4_t __p0, float32x4_t __p1) {
41290 float32x4_t __ret;
4129041291 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4129141292 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41292 float32x4_t __ret;
4129341293 __ret = (float32x4_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
4129441294 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4129541295 return __ret;
......@@ -41304,10 +41304,10 @@ __ai float32x4_t vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2
4130441304}
4130541305#else
4130641306__ai float32x4_t vcmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41307 float32x4_t __ret;
4130741308 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4130841309 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4130941310 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41310 float32x4_t __ret;
4131141311 __ret = (float32x4_t) __builtin_neon_vcmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4131241312 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4131341313 return __ret;
......@@ -41327,10 +41327,10 @@ __ai float32x2_t vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2)
4132741327}
4132841328#else
4132941329__ai float32x2_t vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41330 float32x2_t __ret;
4133041331 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4133141332 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4133241333 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41333 float32x2_t __ret;
4133441334 __ret = (float32x2_t) __builtin_neon_vcmla_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4133541335 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4133641336 return __ret;
......@@ -41344,10 +41344,10 @@ __ai float32x2_t __noswap_vcmla_f32(float32x2_t __p0, float32x2_t __p1, float32x
4134441344
4134541345#ifdef __LITTLE_ENDIAN__
4134641346#define vcmla_lane_f32(__p0_169, __p1_169, __p2_169, __p3_169) __extension__ ({ \
41347 float32x2_t __ret_169; \
4134741348 float32x2_t __s0_169 = __p0_169; \
4134841349 float32x2_t __s1_169 = __p1_169; \
4134941350 float32x2_t __s2_169 = __p2_169; \
41350 float32x2_t __ret_169; \
4135141351float32x2_t __reint_169 = __s2_169; \
4135241352uint64x1_t __reint1_169 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_169, __p3_169)}; \
4135341353 __ret_169 = vcmla_f32(__s0_169, __s1_169, *(float32x2_t *) &__reint1_169); \
......@@ -41355,13 +41355,13 @@ uint64x1_t __reint1_169 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4135541355})
4135641356#else
4135741357#define vcmla_lane_f32(__p0_170, __p1_170, __p2_170, __p3_170) __extension__ ({ \
41358 float32x2_t __ret_170; \
4135841359 float32x2_t __s0_170 = __p0_170; \
4135941360 float32x2_t __s1_170 = __p1_170; \
4136041361 float32x2_t __s2_170 = __p2_170; \
4136141362 float32x2_t __rev0_170; __rev0_170 = __builtin_shufflevector(__s0_170, __s0_170, 1, 0); \
4136241363 float32x2_t __rev1_170; __rev1_170 = __builtin_shufflevector(__s1_170, __s1_170, 1, 0); \
4136341364 float32x2_t __rev2_170; __rev2_170 = __builtin_shufflevector(__s2_170, __s2_170, 1, 0); \
41364 float32x2_t __ret_170; \
4136541365float32x2_t __reint_170 = __rev2_170; \
4136641366uint64x1_t __reint1_170 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_170, __p3_170)}; \
4136741367 __ret_170 = __noswap_vcmla_f32(__rev0_170, __rev1_170, *(float32x2_t *) &__reint1_170); \
......@@ -41372,10 +41372,10 @@ uint64x1_t __reint1_170 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4137241372
4137341373#ifdef __LITTLE_ENDIAN__
4137441374#define vcmlaq_lane_f32(__p0_171, __p1_171, __p2_171, __p3_171) __extension__ ({ \
41375 float32x4_t __ret_171; \
4137541376 float32x4_t __s0_171 = __p0_171; \
4137641377 float32x4_t __s1_171 = __p1_171; \
4137741378 float32x2_t __s2_171 = __p2_171; \
41378 float32x4_t __ret_171; \
4137941379float32x2_t __reint_171 = __s2_171; \
4138041380uint64x2_t __reint1_171 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_171, __p3_171), vget_lane_u64(*(uint64x1_t *) &__reint_171, __p3_171)}; \
4138141381 __ret_171 = vcmlaq_f32(__s0_171, __s1_171, *(float32x4_t *) &__reint1_171); \
......@@ -41383,13 +41383,13 @@ uint64x2_t __reint1_171 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4138341383})
4138441384#else
4138541385#define vcmlaq_lane_f32(__p0_172, __p1_172, __p2_172, __p3_172) __extension__ ({ \
41386 float32x4_t __ret_172; \
4138641387 float32x4_t __s0_172 = __p0_172; \
4138741388 float32x4_t __s1_172 = __p1_172; \
4138841389 float32x2_t __s2_172 = __p2_172; \
4138941390 float32x4_t __rev0_172; __rev0_172 = __builtin_shufflevector(__s0_172, __s0_172, 3, 2, 1, 0); \
4139041391 float32x4_t __rev1_172; __rev1_172 = __builtin_shufflevector(__s1_172, __s1_172, 3, 2, 1, 0); \
4139141392 float32x2_t __rev2_172; __rev2_172 = __builtin_shufflevector(__s2_172, __s2_172, 1, 0); \
41392 float32x4_t __ret_172; \
4139341393float32x2_t __reint_172 = __rev2_172; \
4139441394uint64x2_t __reint1_172 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_172, __p3_172), vget_lane_u64(*(uint64x1_t *) &__reint_172, __p3_172)}; \
4139541395 __ret_172 = __noswap_vcmlaq_f32(__rev0_172, __rev1_172, *(float32x4_t *) &__reint1_172); \
......@@ -41400,10 +41400,10 @@ uint64x2_t __reint1_172 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4140041400
4140141401#ifdef __LITTLE_ENDIAN__
4140241402#define vcmla_laneq_f32(__p0_173, __p1_173, __p2_173, __p3_173) __extension__ ({ \
41403 float32x2_t __ret_173; \
4140341404 float32x2_t __s0_173 = __p0_173; \
4140441405 float32x2_t __s1_173 = __p1_173; \
4140541406 float32x4_t __s2_173 = __p2_173; \
41406 float32x2_t __ret_173; \
4140741407float32x4_t __reint_173 = __s2_173; \
4140841408uint64x1_t __reint1_173 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_173, __p3_173)}; \
4140941409 __ret_173 = vcmla_f32(__s0_173, __s1_173, *(float32x2_t *) &__reint1_173); \
......@@ -41411,13 +41411,13 @@ uint64x1_t __reint1_173 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4141141411})
4141241412#else
4141341413#define vcmla_laneq_f32(__p0_174, __p1_174, __p2_174, __p3_174) __extension__ ({ \
41414 float32x2_t __ret_174; \
4141441415 float32x2_t __s0_174 = __p0_174; \
4141541416 float32x2_t __s1_174 = __p1_174; \
4141641417 float32x4_t __s2_174 = __p2_174; \
4141741418 float32x2_t __rev0_174; __rev0_174 = __builtin_shufflevector(__s0_174, __s0_174, 1, 0); \
4141841419 float32x2_t __rev1_174; __rev1_174 = __builtin_shufflevector(__s1_174, __s1_174, 1, 0); \
4141941420 float32x4_t __rev2_174; __rev2_174 = __builtin_shufflevector(__s2_174, __s2_174, 3, 2, 1, 0); \
41420 float32x2_t __ret_174; \
4142141421float32x4_t __reint_174 = __rev2_174; \
4142241422uint64x1_t __reint1_174 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *) &__reint_174, __p3_174)}; \
4142341423 __ret_174 = __noswap_vcmla_f32(__rev0_174, __rev1_174, *(float32x2_t *) &__reint1_174); \
......@@ -41428,10 +41428,10 @@ uint64x1_t __reint1_174 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4142841428
4142941429#ifdef __LITTLE_ENDIAN__
4143041430#define vcmlaq_laneq_f32(__p0_175, __p1_175, __p2_175, __p3_175) __extension__ ({ \
41431 float32x4_t __ret_175; \
4143141432 float32x4_t __s0_175 = __p0_175; \
4143241433 float32x4_t __s1_175 = __p1_175; \
4143341434 float32x4_t __s2_175 = __p2_175; \
41434 float32x4_t __ret_175; \
4143541435float32x4_t __reint_175 = __s2_175; \
4143641436uint64x2_t __reint1_175 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_175, __p3_175), vgetq_lane_u64(*(uint64x2_t *) &__reint_175, __p3_175)}; \
4143741437 __ret_175 = vcmlaq_f32(__s0_175, __s1_175, *(float32x4_t *) &__reint1_175); \
......@@ -41439,13 +41439,13 @@ uint64x2_t __reint1_175 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4143941439})
4144041440#else
4144141441#define vcmlaq_laneq_f32(__p0_176, __p1_176, __p2_176, __p3_176) __extension__ ({ \
41442 float32x4_t __ret_176; \
4144241443 float32x4_t __s0_176 = __p0_176; \
4144341444 float32x4_t __s1_176 = __p1_176; \
4144441445 float32x4_t __s2_176 = __p2_176; \
4144541446 float32x4_t __rev0_176; __rev0_176 = __builtin_shufflevector(__s0_176, __s0_176, 3, 2, 1, 0); \
4144641447 float32x4_t __rev1_176; __rev1_176 = __builtin_shufflevector(__s1_176, __s1_176, 3, 2, 1, 0); \
4144741448 float32x4_t __rev2_176; __rev2_176 = __builtin_shufflevector(__s2_176, __s2_176, 3, 2, 1, 0); \
41448 float32x4_t __ret_176; \
4144941449float32x4_t __reint_176 = __rev2_176; \
4145041450uint64x2_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)}; \
4145141451 __ret_176 = __noswap_vcmlaq_f32(__rev0_176, __rev1_176, *(float32x4_t *) &__reint1_176); \
......@@ -41462,10 +41462,10 @@ __ai float32x4_t vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4
4146241462}
4146341463#else
4146441464__ai float32x4_t vcmlaq_rot180_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41465 float32x4_t __ret;
4146541466 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4146641467 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4146741468 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41468 float32x4_t __ret;
4146941469 __ret = (float32x4_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4147041470 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4147141471 return __ret;
......@@ -41485,10 +41485,10 @@ __ai float32x2_t vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_
4148541485}
4148641486#else
4148741487__ai float32x2_t vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41488 float32x2_t __ret;
4148841489 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4148941490 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4149041491 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41491 float32x2_t __ret;
4149241492 __ret = (float32x2_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4149341493 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4149441494 return __ret;
......@@ -41502,10 +41502,10 @@ __ai float32x2_t __noswap_vcmla_rot180_f32(float32x2_t __p0, float32x2_t __p1, f
4150241502
4150341503#ifdef __LITTLE_ENDIAN__
4150441504#define vcmla_rot180_lane_f32(__p0_177, __p1_177, __p2_177, __p3_177) __extension__ ({ \
41505 float32x2_t __ret_177; \
4150541506 float32x2_t __s0_177 = __p0_177; \
4150641507 float32x2_t __s1_177 = __p1_177; \
4150741508 float32x2_t __s2_177 = __p2_177; \
41508 float32x2_t __ret_177; \
4150941509float32x2_t __reint_177 = __s2_177; \
4151041510uint64x1_t __reint1_177 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_177, __p3_177)}; \
4151141511 __ret_177 = vcmla_rot180_f32(__s0_177, __s1_177, *(float32x2_t *) &__reint1_177); \
......@@ -41513,13 +41513,13 @@ uint64x1_t __reint1_177 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4151341513})
4151441514#else
4151541515#define vcmla_rot180_lane_f32(__p0_178, __p1_178, __p2_178, __p3_178) __extension__ ({ \
41516 float32x2_t __ret_178; \
4151641517 float32x2_t __s0_178 = __p0_178; \
4151741518 float32x2_t __s1_178 = __p1_178; \
4151841519 float32x2_t __s2_178 = __p2_178; \
4151941520 float32x2_t __rev0_178; __rev0_178 = __builtin_shufflevector(__s0_178, __s0_178, 1, 0); \
4152041521 float32x2_t __rev1_178; __rev1_178 = __builtin_shufflevector(__s1_178, __s1_178, 1, 0); \
4152141522 float32x2_t __rev2_178; __rev2_178 = __builtin_shufflevector(__s2_178, __s2_178, 1, 0); \
41522 float32x2_t __ret_178; \
4152341523float32x2_t __reint_178 = __rev2_178; \
4152441524uint64x1_t __reint1_178 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_178, __p3_178)}; \
4152541525 __ret_178 = __noswap_vcmla_rot180_f32(__rev0_178, __rev1_178, *(float32x2_t *) &__reint1_178); \
......@@ -41530,10 +41530,10 @@ uint64x1_t __reint1_178 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4153041530
4153141531#ifdef __LITTLE_ENDIAN__
4153241532#define vcmlaq_rot180_lane_f32(__p0_179, __p1_179, __p2_179, __p3_179) __extension__ ({ \
41533 float32x4_t __ret_179; \
4153341534 float32x4_t __s0_179 = __p0_179; \
4153441535 float32x4_t __s1_179 = __p1_179; \
4153541536 float32x2_t __s2_179 = __p2_179; \
41536 float32x4_t __ret_179; \
4153741537float32x2_t __reint_179 = __s2_179; \
4153841538uint64x2_t __reint1_179 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_179, __p3_179), vget_lane_u64(*(uint64x1_t *) &__reint_179, __p3_179)}; \
4153941539 __ret_179 = vcmlaq_rot180_f32(__s0_179, __s1_179, *(float32x4_t *) &__reint1_179); \
......@@ -41541,13 +41541,13 @@ uint64x2_t __reint1_179 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4154141541})
4154241542#else
4154341543#define vcmlaq_rot180_lane_f32(__p0_180, __p1_180, __p2_180, __p3_180) __extension__ ({ \
41544 float32x4_t __ret_180; \
4154441545 float32x4_t __s0_180 = __p0_180; \
4154541546 float32x4_t __s1_180 = __p1_180; \
4154641547 float32x2_t __s2_180 = __p2_180; \
4154741548 float32x4_t __rev0_180; __rev0_180 = __builtin_shufflevector(__s0_180, __s0_180, 3, 2, 1, 0); \
4154841549 float32x4_t __rev1_180; __rev1_180 = __builtin_shufflevector(__s1_180, __s1_180, 3, 2, 1, 0); \
4154941550 float32x2_t __rev2_180; __rev2_180 = __builtin_shufflevector(__s2_180, __s2_180, 1, 0); \
41550 float32x4_t __ret_180; \
4155141551float32x2_t __reint_180 = __rev2_180; \
4155241552uint64x2_t __reint1_180 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_180, __p3_180), vget_lane_u64(*(uint64x1_t *) &__reint_180, __p3_180)}; \
4155341553 __ret_180 = __noswap_vcmlaq_rot180_f32(__rev0_180, __rev1_180, *(float32x4_t *) &__reint1_180); \
......@@ -41558,10 +41558,10 @@ uint64x2_t __reint1_180 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4155841558
4155941559#ifdef __LITTLE_ENDIAN__
4156041560#define vcmla_rot180_laneq_f32(__p0_181, __p1_181, __p2_181, __p3_181) __extension__ ({ \
41561 float32x2_t __ret_181; \
4156141562 float32x2_t __s0_181 = __p0_181; \
4156241563 float32x2_t __s1_181 = __p1_181; \
4156341564 float32x4_t __s2_181 = __p2_181; \
41564 float32x2_t __ret_181; \
4156541565float32x4_t __reint_181 = __s2_181; \
4156641566uint64x1_t __reint1_181 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_181, __p3_181)}; \
4156741567 __ret_181 = vcmla_rot180_f32(__s0_181, __s1_181, *(float32x2_t *) &__reint1_181); \
......@@ -41569,13 +41569,13 @@ uint64x1_t __reint1_181 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4156941569})
4157041570#else
4157141571#define vcmla_rot180_laneq_f32(__p0_182, __p1_182, __p2_182, __p3_182) __extension__ ({ \
41572 float32x2_t __ret_182; \
4157241573 float32x2_t __s0_182 = __p0_182; \
4157341574 float32x2_t __s1_182 = __p1_182; \
4157441575 float32x4_t __s2_182 = __p2_182; \
4157541576 float32x2_t __rev0_182; __rev0_182 = __builtin_shufflevector(__s0_182, __s0_182, 1, 0); \
4157641577 float32x2_t __rev1_182; __rev1_182 = __builtin_shufflevector(__s1_182, __s1_182, 1, 0); \
4157741578 float32x4_t __rev2_182; __rev2_182 = __builtin_shufflevector(__s2_182, __s2_182, 3, 2, 1, 0); \
41578 float32x2_t __ret_182; \
4157941579float32x4_t __reint_182 = __rev2_182; \
4158041580uint64x1_t __reint1_182 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *) &__reint_182, __p3_182)}; \
4158141581 __ret_182 = __noswap_vcmla_rot180_f32(__rev0_182, __rev1_182, *(float32x2_t *) &__reint1_182); \
......@@ -41586,10 +41586,10 @@ uint64x1_t __reint1_182 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4158641586
4158741587#ifdef __LITTLE_ENDIAN__
4158841588#define vcmlaq_rot180_laneq_f32(__p0_183, __p1_183, __p2_183, __p3_183) __extension__ ({ \
41589 float32x4_t __ret_183; \
4158941590 float32x4_t __s0_183 = __p0_183; \
4159041591 float32x4_t __s1_183 = __p1_183; \
4159141592 float32x4_t __s2_183 = __p2_183; \
41592 float32x4_t __ret_183; \
4159341593float32x4_t __reint_183 = __s2_183; \
4159441594uint64x2_t __reint1_183 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_183, __p3_183), vgetq_lane_u64(*(uint64x2_t *) &__reint_183, __p3_183)}; \
4159541595 __ret_183 = vcmlaq_rot180_f32(__s0_183, __s1_183, *(float32x4_t *) &__reint1_183); \
......@@ -41597,13 +41597,13 @@ uint64x2_t __reint1_183 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4159741597})
4159841598#else
4159941599#define vcmlaq_rot180_laneq_f32(__p0_184, __p1_184, __p2_184, __p3_184) __extension__ ({ \
41600 float32x4_t __ret_184; \
4160041601 float32x4_t __s0_184 = __p0_184; \
4160141602 float32x4_t __s1_184 = __p1_184; \
4160241603 float32x4_t __s2_184 = __p2_184; \
4160341604 float32x4_t __rev0_184; __rev0_184 = __builtin_shufflevector(__s0_184, __s0_184, 3, 2, 1, 0); \
4160441605 float32x4_t __rev1_184; __rev1_184 = __builtin_shufflevector(__s1_184, __s1_184, 3, 2, 1, 0); \
4160541606 float32x4_t __rev2_184; __rev2_184 = __builtin_shufflevector(__s2_184, __s2_184, 3, 2, 1, 0); \
41606 float32x4_t __ret_184; \
4160741607float32x4_t __reint_184 = __rev2_184; \
4160841608uint64x2_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)}; \
4160941609 __ret_184 = __noswap_vcmlaq_rot180_f32(__rev0_184, __rev1_184, *(float32x4_t *) &__reint1_184); \
......@@ -41620,10 +41620,10 @@ __ai float32x4_t vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4
4162041620}
4162141621#else
4162241622__ai float32x4_t vcmlaq_rot270_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41623 float32x4_t __ret;
4162341624 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4162441625 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4162541626 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41626 float32x4_t __ret;
4162741627 __ret = (float32x4_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4162841628 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4162941629 return __ret;
......@@ -41643,10 +41643,10 @@ __ai float32x2_t vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_
4164341643}
4164441644#else
4164541645__ai float32x2_t vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41646 float32x2_t __ret;
4164641647 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4164741648 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4164841649 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41649 float32x2_t __ret;
4165041650 __ret = (float32x2_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4165141651 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4165241652 return __ret;
......@@ -41660,10 +41660,10 @@ __ai float32x2_t __noswap_vcmla_rot270_f32(float32x2_t __p0, float32x2_t __p1, f
4166041660
4166141661#ifdef __LITTLE_ENDIAN__
4166241662#define vcmla_rot270_lane_f32(__p0_185, __p1_185, __p2_185, __p3_185) __extension__ ({ \
41663 float32x2_t __ret_185; \
4166341664 float32x2_t __s0_185 = __p0_185; \
4166441665 float32x2_t __s1_185 = __p1_185; \
4166541666 float32x2_t __s2_185 = __p2_185; \
41666 float32x2_t __ret_185; \
4166741667float32x2_t __reint_185 = __s2_185; \
4166841668uint64x1_t __reint1_185 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_185, __p3_185)}; \
4166941669 __ret_185 = vcmla_rot270_f32(__s0_185, __s1_185, *(float32x2_t *) &__reint1_185); \
......@@ -41671,13 +41671,13 @@ uint64x1_t __reint1_185 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4167141671})
4167241672#else
4167341673#define vcmla_rot270_lane_f32(__p0_186, __p1_186, __p2_186, __p3_186) __extension__ ({ \
41674 float32x2_t __ret_186; \
4167441675 float32x2_t __s0_186 = __p0_186; \
4167541676 float32x2_t __s1_186 = __p1_186; \
4167641677 float32x2_t __s2_186 = __p2_186; \
4167741678 float32x2_t __rev0_186; __rev0_186 = __builtin_shufflevector(__s0_186, __s0_186, 1, 0); \
4167841679 float32x2_t __rev1_186; __rev1_186 = __builtin_shufflevector(__s1_186, __s1_186, 1, 0); \
4167941680 float32x2_t __rev2_186; __rev2_186 = __builtin_shufflevector(__s2_186, __s2_186, 1, 0); \
41680 float32x2_t __ret_186; \
4168141681float32x2_t __reint_186 = __rev2_186; \
4168241682uint64x1_t __reint1_186 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_186, __p3_186)}; \
4168341683 __ret_186 = __noswap_vcmla_rot270_f32(__rev0_186, __rev1_186, *(float32x2_t *) &__reint1_186); \
......@@ -41688,10 +41688,10 @@ uint64x1_t __reint1_186 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4168841688
4168941689#ifdef __LITTLE_ENDIAN__
4169041690#define vcmlaq_rot270_lane_f32(__p0_187, __p1_187, __p2_187, __p3_187) __extension__ ({ \
41691 float32x4_t __ret_187; \
4169141692 float32x4_t __s0_187 = __p0_187; \
4169241693 float32x4_t __s1_187 = __p1_187; \
4169341694 float32x2_t __s2_187 = __p2_187; \
41694 float32x4_t __ret_187; \
4169541695float32x2_t __reint_187 = __s2_187; \
4169641696uint64x2_t __reint1_187 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_187, __p3_187), vget_lane_u64(*(uint64x1_t *) &__reint_187, __p3_187)}; \
4169741697 __ret_187 = vcmlaq_rot270_f32(__s0_187, __s1_187, *(float32x4_t *) &__reint1_187); \
......@@ -41699,13 +41699,13 @@ uint64x2_t __reint1_187 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4169941699})
4170041700#else
4170141701#define vcmlaq_rot270_lane_f32(__p0_188, __p1_188, __p2_188, __p3_188) __extension__ ({ \
41702 float32x4_t __ret_188; \
4170241703 float32x4_t __s0_188 = __p0_188; \
4170341704 float32x4_t __s1_188 = __p1_188; \
4170441705 float32x2_t __s2_188 = __p2_188; \
4170541706 float32x4_t __rev0_188; __rev0_188 = __builtin_shufflevector(__s0_188, __s0_188, 3, 2, 1, 0); \
4170641707 float32x4_t __rev1_188; __rev1_188 = __builtin_shufflevector(__s1_188, __s1_188, 3, 2, 1, 0); \
4170741708 float32x2_t __rev2_188; __rev2_188 = __builtin_shufflevector(__s2_188, __s2_188, 1, 0); \
41708 float32x4_t __ret_188; \
4170941709float32x2_t __reint_188 = __rev2_188; \
4171041710uint64x2_t __reint1_188 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_188, __p3_188), vget_lane_u64(*(uint64x1_t *) &__reint_188, __p3_188)}; \
4171141711 __ret_188 = __noswap_vcmlaq_rot270_f32(__rev0_188, __rev1_188, *(float32x4_t *) &__reint1_188); \
......@@ -41716,10 +41716,10 @@ uint64x2_t __reint1_188 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4171641716
4171741717#ifdef __LITTLE_ENDIAN__
4171841718#define vcmla_rot270_laneq_f32(__p0_189, __p1_189, __p2_189, __p3_189) __extension__ ({ \
41719 float32x2_t __ret_189; \
4171941720 float32x2_t __s0_189 = __p0_189; \
4172041721 float32x2_t __s1_189 = __p1_189; \
4172141722 float32x4_t __s2_189 = __p2_189; \
41722 float32x2_t __ret_189; \
4172341723float32x4_t __reint_189 = __s2_189; \
4172441724uint64x1_t __reint1_189 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_189, __p3_189)}; \
4172541725 __ret_189 = vcmla_rot270_f32(__s0_189, __s1_189, *(float32x2_t *) &__reint1_189); \
......@@ -41727,13 +41727,13 @@ uint64x1_t __reint1_189 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4172741727})
4172841728#else
4172941729#define vcmla_rot270_laneq_f32(__p0_190, __p1_190, __p2_190, __p3_190) __extension__ ({ \
41730 float32x2_t __ret_190; \
4173041731 float32x2_t __s0_190 = __p0_190; \
4173141732 float32x2_t __s1_190 = __p1_190; \
4173241733 float32x4_t __s2_190 = __p2_190; \
4173341734 float32x2_t __rev0_190; __rev0_190 = __builtin_shufflevector(__s0_190, __s0_190, 1, 0); \
4173441735 float32x2_t __rev1_190; __rev1_190 = __builtin_shufflevector(__s1_190, __s1_190, 1, 0); \
4173541736 float32x4_t __rev2_190; __rev2_190 = __builtin_shufflevector(__s2_190, __s2_190, 3, 2, 1, 0); \
41736 float32x2_t __ret_190; \
4173741737float32x4_t __reint_190 = __rev2_190; \
4173841738uint64x1_t __reint1_190 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *) &__reint_190, __p3_190)}; \
4173941739 __ret_190 = __noswap_vcmla_rot270_f32(__rev0_190, __rev1_190, *(float32x2_t *) &__reint1_190); \
......@@ -41744,10 +41744,10 @@ uint64x1_t __reint1_190 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4174441744
4174541745#ifdef __LITTLE_ENDIAN__
4174641746#define vcmlaq_rot270_laneq_f32(__p0_191, __p1_191, __p2_191, __p3_191) __extension__ ({ \
41747 float32x4_t __ret_191; \
4174741748 float32x4_t __s0_191 = __p0_191; \
4174841749 float32x4_t __s1_191 = __p1_191; \
4174941750 float32x4_t __s2_191 = __p2_191; \
41750 float32x4_t __ret_191; \
4175141751float32x4_t __reint_191 = __s2_191; \
4175241752uint64x2_t __reint1_191 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_191, __p3_191), vgetq_lane_u64(*(uint64x2_t *) &__reint_191, __p3_191)}; \
4175341753 __ret_191 = vcmlaq_rot270_f32(__s0_191, __s1_191, *(float32x4_t *) &__reint1_191); \
......@@ -41755,13 +41755,13 @@ uint64x2_t __reint1_191 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4175541755})
4175641756#else
4175741757#define vcmlaq_rot270_laneq_f32(__p0_192, __p1_192, __p2_192, __p3_192) __extension__ ({ \
41758 float32x4_t __ret_192; \
4175841759 float32x4_t __s0_192 = __p0_192; \
4175941760 float32x4_t __s1_192 = __p1_192; \
4176041761 float32x4_t __s2_192 = __p2_192; \
4176141762 float32x4_t __rev0_192; __rev0_192 = __builtin_shufflevector(__s0_192, __s0_192, 3, 2, 1, 0); \
4176241763 float32x4_t __rev1_192; __rev1_192 = __builtin_shufflevector(__s1_192, __s1_192, 3, 2, 1, 0); \
4176341764 float32x4_t __rev2_192; __rev2_192 = __builtin_shufflevector(__s2_192, __s2_192, 3, 2, 1, 0); \
41764 float32x4_t __ret_192; \
4176541765float32x4_t __reint_192 = __rev2_192; \
4176641766uint64x2_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)}; \
4176741767 __ret_192 = __noswap_vcmlaq_rot270_f32(__rev0_192, __rev1_192, *(float32x4_t *) &__reint1_192); \
......@@ -41778,10 +41778,10 @@ __ai float32x4_t vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_
4177841778}
4177941779#else
4178041780__ai float32x4_t vcmlaq_rot90_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
41781 float32x4_t __ret;
4178141782 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4178241783 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4178341784 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
41784 float32x4_t __ret;
4178541785 __ret = (float32x4_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4178641786 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4178741787 return __ret;
......@@ -41801,10 +41801,10 @@ __ai float32x2_t vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t
4180141801}
4180241802#else
4180341803__ai float32x2_t vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
41804 float32x2_t __ret;
4180441805 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4180541806 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4180641807 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
41807 float32x2_t __ret;
4180841808 __ret = (float32x2_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4180941809 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4181041810 return __ret;
......@@ -41818,10 +41818,10 @@ __ai float32x2_t __noswap_vcmla_rot90_f32(float32x2_t __p0, float32x2_t __p1, fl
4181841818
4181941819#ifdef __LITTLE_ENDIAN__
4182041820#define vcmla_rot90_lane_f32(__p0_193, __p1_193, __p2_193, __p3_193) __extension__ ({ \
41821 float32x2_t __ret_193; \
4182141822 float32x2_t __s0_193 = __p0_193; \
4182241823 float32x2_t __s1_193 = __p1_193; \
4182341824 float32x2_t __s2_193 = __p2_193; \
41824 float32x2_t __ret_193; \
4182541825float32x2_t __reint_193 = __s2_193; \
4182641826uint64x1_t __reint1_193 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_193, __p3_193)}; \
4182741827 __ret_193 = vcmla_rot90_f32(__s0_193, __s1_193, *(float32x2_t *) &__reint1_193); \
......@@ -41829,13 +41829,13 @@ uint64x1_t __reint1_193 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4182941829})
4183041830#else
4183141831#define vcmla_rot90_lane_f32(__p0_194, __p1_194, __p2_194, __p3_194) __extension__ ({ \
41832 float32x2_t __ret_194; \
4183241833 float32x2_t __s0_194 = __p0_194; \
4183341834 float32x2_t __s1_194 = __p1_194; \
4183441835 float32x2_t __s2_194 = __p2_194; \
4183541836 float32x2_t __rev0_194; __rev0_194 = __builtin_shufflevector(__s0_194, __s0_194, 1, 0); \
4183641837 float32x2_t __rev1_194; __rev1_194 = __builtin_shufflevector(__s1_194, __s1_194, 1, 0); \
4183741838 float32x2_t __rev2_194; __rev2_194 = __builtin_shufflevector(__s2_194, __s2_194, 1, 0); \
41838 float32x2_t __ret_194; \
4183941839float32x2_t __reint_194 = __rev2_194; \
4184041840uint64x1_t __reint1_194 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_194, __p3_194)}; \
4184141841 __ret_194 = __noswap_vcmla_rot90_f32(__rev0_194, __rev1_194, *(float32x2_t *) &__reint1_194); \
......@@ -41846,10 +41846,10 @@ uint64x1_t __reint1_194 = (uint64x1_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4184641846
4184741847#ifdef __LITTLE_ENDIAN__
4184841848#define vcmlaq_rot90_lane_f32(__p0_195, __p1_195, __p2_195, __p3_195) __extension__ ({ \
41849 float32x4_t __ret_195; \
4184941850 float32x4_t __s0_195 = __p0_195; \
4185041851 float32x4_t __s1_195 = __p1_195; \
4185141852 float32x2_t __s2_195 = __p2_195; \
41852 float32x4_t __ret_195; \
4185341853float32x2_t __reint_195 = __s2_195; \
4185441854uint64x2_t __reint1_195 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_195, __p3_195), vget_lane_u64(*(uint64x1_t *) &__reint_195, __p3_195)}; \
4185541855 __ret_195 = vcmlaq_rot90_f32(__s0_195, __s1_195, *(float32x4_t *) &__reint1_195); \
......@@ -41857,13 +41857,13 @@ uint64x2_t __reint1_195 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4185741857})
4185841858#else
4185941859#define vcmlaq_rot90_lane_f32(__p0_196, __p1_196, __p2_196, __p3_196) __extension__ ({ \
41860 float32x4_t __ret_196; \
4186041861 float32x4_t __s0_196 = __p0_196; \
4186141862 float32x4_t __s1_196 = __p1_196; \
4186241863 float32x2_t __s2_196 = __p2_196; \
4186341864 float32x4_t __rev0_196; __rev0_196 = __builtin_shufflevector(__s0_196, __s0_196, 3, 2, 1, 0); \
4186441865 float32x4_t __rev1_196; __rev1_196 = __builtin_shufflevector(__s1_196, __s1_196, 3, 2, 1, 0); \
4186541866 float32x2_t __rev2_196; __rev2_196 = __builtin_shufflevector(__s2_196, __s2_196, 1, 0); \
41866 float32x4_t __ret_196; \
4186741867float32x2_t __reint_196 = __rev2_196; \
4186841868uint64x2_t __reint1_196 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_196, __p3_196), vget_lane_u64(*(uint64x1_t *) &__reint_196, __p3_196)}; \
4186941869 __ret_196 = __noswap_vcmlaq_rot90_f32(__rev0_196, __rev1_196, *(float32x4_t *) &__reint1_196); \
......@@ -41874,10 +41874,10 @@ uint64x2_t __reint1_196 = (uint64x2_t) {vget_lane_u64(*(uint64x1_t *) &__reint_1
4187441874
4187541875#ifdef __LITTLE_ENDIAN__
4187641876#define vcmla_rot90_laneq_f32(__p0_197, __p1_197, __p2_197, __p3_197) __extension__ ({ \
41877 float32x2_t __ret_197; \
4187741878 float32x2_t __s0_197 = __p0_197; \
4187841879 float32x2_t __s1_197 = __p1_197; \
4187941880 float32x4_t __s2_197 = __p2_197; \
41880 float32x2_t __ret_197; \
4188141881float32x4_t __reint_197 = __s2_197; \
4188241882uint64x1_t __reint1_197 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_197, __p3_197)}; \
4188341883 __ret_197 = vcmla_rot90_f32(__s0_197, __s1_197, *(float32x2_t *) &__reint1_197); \
......@@ -41885,13 +41885,13 @@ uint64x1_t __reint1_197 = (uint64x1_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4188541885})
4188641886#else
4188741887#define vcmla_rot90_laneq_f32(__p0_198, __p1_198, __p2_198, __p3_198) __extension__ ({ \
41888 float32x2_t __ret_198; \
4188841889 float32x2_t __s0_198 = __p0_198; \
4188941890 float32x2_t __s1_198 = __p1_198; \
4189041891 float32x4_t __s2_198 = __p2_198; \
4189141892 float32x2_t __rev0_198; __rev0_198 = __builtin_shufflevector(__s0_198, __s0_198, 1, 0); \
4189241893 float32x2_t __rev1_198; __rev1_198 = __builtin_shufflevector(__s1_198, __s1_198, 1, 0); \
4189341894 float32x4_t __rev2_198; __rev2_198 = __builtin_shufflevector(__s2_198, __s2_198, 3, 2, 1, 0); \
41894 float32x2_t __ret_198; \
4189541895float32x4_t __reint_198 = __rev2_198; \
4189641896uint64x1_t __reint1_198 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *) &__reint_198, __p3_198)}; \
4189741897 __ret_198 = __noswap_vcmla_rot90_f32(__rev0_198, __rev1_198, *(float32x2_t *) &__reint1_198); \
......@@ -41902,10 +41902,10 @@ uint64x1_t __reint1_198 = (uint64x1_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4190241902
4190341903#ifdef __LITTLE_ENDIAN__
4190441904#define vcmlaq_rot90_laneq_f32(__p0_199, __p1_199, __p2_199, __p3_199) __extension__ ({ \
41905 float32x4_t __ret_199; \
4190541906 float32x4_t __s0_199 = __p0_199; \
4190641907 float32x4_t __s1_199 = __p1_199; \
4190741908 float32x4_t __s2_199 = __p2_199; \
41908 float32x4_t __ret_199; \
4190941909float32x4_t __reint_199 = __s2_199; \
4191041910uint64x2_t __reint1_199 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_199, __p3_199), vgetq_lane_u64(*(uint64x2_t *) &__reint_199, __p3_199)}; \
4191141911 __ret_199 = vcmlaq_rot90_f32(__s0_199, __s1_199, *(float32x4_t *) &__reint1_199); \
......@@ -41913,13 +41913,13 @@ uint64x2_t __reint1_199 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4191341913})
4191441914#else
4191541915#define vcmlaq_rot90_laneq_f32(__p0_200, __p1_200, __p2_200, __p3_200) __extension__ ({ \
41916 float32x4_t __ret_200; \
4191641917 float32x4_t __s0_200 = __p0_200; \
4191741918 float32x4_t __s1_200 = __p1_200; \
4191841919 float32x4_t __s2_200 = __p2_200; \
4191941920 float32x4_t __rev0_200; __rev0_200 = __builtin_shufflevector(__s0_200, __s0_200, 3, 2, 1, 0); \
4192041921 float32x4_t __rev1_200; __rev1_200 = __builtin_shufflevector(__s1_200, __s1_200, 3, 2, 1, 0); \
4192141922 float32x4_t __rev2_200; __rev2_200 = __builtin_shufflevector(__s2_200, __s2_200, 3, 2, 1, 0); \
41922 float32x4_t __ret_200; \
4192341923float32x4_t __reint_200 = __rev2_200; \
4192441924uint64x2_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)}; \
4192541925 __ret_200 = __noswap_vcmlaq_rot90_f32(__rev0_200, __rev1_200, *(float32x4_t *) &__reint1_200); \
......@@ -41938,9 +41938,9 @@ __ai float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
4193841938}
4193941939#else
4194041940__ai float16x4_t vcadd_rot270_f16(float16x4_t __p0, float16x4_t __p1) {
41941 float16x4_t __ret;
4194141942 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4194241943 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41943 float16x4_t __ret;
4194441944 __ret = (float16x4_t) __builtin_neon_vcadd_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4194541945 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4194641946 return __ret;
......@@ -41955,9 +41955,9 @@ __ai float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
4195541955}
4195641956#else
4195741957__ai float16x4_t vcadd_rot90_f16(float16x4_t __p0, float16x4_t __p1) {
41958 float16x4_t __ret;
4195841959 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4195941960 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41960 float16x4_t __ret;
4196141961 __ret = (float16x4_t) __builtin_neon_vcadd_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4196241962 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4196341963 return __ret;
......@@ -41972,9 +41972,9 @@ __ai float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
4197241972}
4197341973#else
4197441974__ai float16x8_t vcaddq_rot270_f16(float16x8_t __p0, float16x8_t __p1) {
41975 float16x8_t __ret;
4197541976 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4197641977 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
41977 float16x8_t __ret;
4197841978 __ret = (float16x8_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4197941979 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4198041980 return __ret;
......@@ -41989,9 +41989,9 @@ __ai float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
4198941989}
4199041990#else
4199141991__ai float16x8_t vcaddq_rot90_f16(float16x8_t __p0, float16x8_t __p1) {
41992 float16x8_t __ret;
4199241993 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4199341994 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
41994 float16x8_t __ret;
4199541995 __ret = (float16x8_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4199641996 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4199741997 return __ret;
......@@ -42006,10 +42006,10 @@ __ai float16x8_t vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2
4200642006}
4200742007#else
4200842008__ai float16x8_t vcmlaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42009 float16x8_t __ret;
4200942010 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4201042011 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4201142012 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42012 float16x8_t __ret;
4201342013 __ret = (float16x8_t) __builtin_neon_vcmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
4201442014 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4201542015 return __ret;
......@@ -42029,10 +42029,10 @@ __ai float16x4_t vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2)
4202942029}
4203042030#else
4203142031__ai float16x4_t vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42032 float16x4_t __ret;
4203242033 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4203342034 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4203442035 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42035 float16x4_t __ret;
4203642036 __ret = (float16x4_t) __builtin_neon_vcmla_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
4203742037 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4203842038 return __ret;
......@@ -42046,10 +42046,10 @@ __ai float16x4_t __noswap_vcmla_f16(float16x4_t __p0, float16x4_t __p1, float16x
4204642046
4204742047#ifdef __LITTLE_ENDIAN__
4204842048#define vcmla_lane_f16(__p0_201, __p1_201, __p2_201, __p3_201) __extension__ ({ \
42049 float16x4_t __ret_201; \
4204942050 float16x4_t __s0_201 = __p0_201; \
4205042051 float16x4_t __s1_201 = __p1_201; \
4205142052 float16x4_t __s2_201 = __p2_201; \
42052 float16x4_t __ret_201; \
4205342053float16x4_t __reint_201 = __s2_201; \
4205442054uint32x2_t __reint1_201 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_201, __p3_201), vget_lane_u32(*(uint32x2_t *) &__reint_201, __p3_201)}; \
4205542055 __ret_201 = vcmla_f16(__s0_201, __s1_201, *(float16x4_t *) &__reint1_201); \
......@@ -42057,13 +42057,13 @@ uint32x2_t __reint1_201 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4205742057})
4205842058#else
4205942059#define vcmla_lane_f16(__p0_202, __p1_202, __p2_202, __p3_202) __extension__ ({ \
42060 float16x4_t __ret_202; \
4206042061 float16x4_t __s0_202 = __p0_202; \
4206142062 float16x4_t __s1_202 = __p1_202; \
4206242063 float16x4_t __s2_202 = __p2_202; \
4206342064 float16x4_t __rev0_202; __rev0_202 = __builtin_shufflevector(__s0_202, __s0_202, 3, 2, 1, 0); \
4206442065 float16x4_t __rev1_202; __rev1_202 = __builtin_shufflevector(__s1_202, __s1_202, 3, 2, 1, 0); \
4206542066 float16x4_t __rev2_202; __rev2_202 = __builtin_shufflevector(__s2_202, __s2_202, 3, 2, 1, 0); \
42066 float16x4_t __ret_202; \
4206742067float16x4_t __reint_202 = __rev2_202; \
4206842068uint32x2_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)}; \
4206942069 __ret_202 = __noswap_vcmla_f16(__rev0_202, __rev1_202, *(float16x4_t *) &__reint1_202); \
......@@ -42074,10 +42074,10 @@ uint32x2_t __reint1_202 = (uint32x2_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4207442074
4207542075#ifdef __LITTLE_ENDIAN__
4207642076#define vcmlaq_lane_f16(__p0_203, __p1_203, __p2_203, __p3_203) __extension__ ({ \
42077 float16x8_t __ret_203; \
4207742078 float16x8_t __s0_203 = __p0_203; \
4207842079 float16x8_t __s1_203 = __p1_203; \
4207942080 float16x4_t __s2_203 = __p2_203; \
42080 float16x8_t __ret_203; \
4208142081float16x4_t __reint_203 = __s2_203; \
4208242082uint32x4_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)}; \
4208342083 __ret_203 = vcmlaq_f16(__s0_203, __s1_203, *(float16x8_t *) &__reint1_203); \
......@@ -42085,13 +42085,13 @@ uint32x4_t __reint1_203 = (uint32x4_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4208542085})
4208642086#else
4208742087#define vcmlaq_lane_f16(__p0_204, __p1_204, __p2_204, __p3_204) __extension__ ({ \
42088 float16x8_t __ret_204; \
4208842089 float16x8_t __s0_204 = __p0_204; \
4208942090 float16x8_t __s1_204 = __p1_204; \
4209042091 float16x4_t __s2_204 = __p2_204; \
4209142092 float16x8_t __rev0_204; __rev0_204 = __builtin_shufflevector(__s0_204, __s0_204, 7, 6, 5, 4, 3, 2, 1, 0); \
4209242093 float16x8_t __rev1_204; __rev1_204 = __builtin_shufflevector(__s1_204, __s1_204, 7, 6, 5, 4, 3, 2, 1, 0); \
4209342094 float16x4_t __rev2_204; __rev2_204 = __builtin_shufflevector(__s2_204, __s2_204, 3, 2, 1, 0); \
42094 float16x8_t __ret_204; \
4209542095float16x4_t __reint_204 = __rev2_204; \
4209642096uint32x4_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)}; \
4209742097 __ret_204 = __noswap_vcmlaq_f16(__rev0_204, __rev1_204, *(float16x8_t *) &__reint1_204); \
......@@ -42102,10 +42102,10 @@ uint32x4_t __reint1_204 = (uint32x4_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4210242102
4210342103#ifdef __LITTLE_ENDIAN__
4210442104#define vcmla_laneq_f16(__p0_205, __p1_205, __p2_205, __p3_205) __extension__ ({ \
42105 float16x4_t __ret_205; \
4210542106 float16x4_t __s0_205 = __p0_205; \
4210642107 float16x4_t __s1_205 = __p1_205; \
4210742108 float16x8_t __s2_205 = __p2_205; \
42108 float16x4_t __ret_205; \
4210942109float16x8_t __reint_205 = __s2_205; \
4211042110uint32x2_t __reint1_205 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_205, __p3_205), vgetq_lane_u32(*(uint32x4_t *) &__reint_205, __p3_205)}; \
4211142111 __ret_205 = vcmla_f16(__s0_205, __s1_205, *(float16x4_t *) &__reint1_205); \
......@@ -42113,13 +42113,13 @@ uint32x2_t __reint1_205 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4211342113})
4211442114#else
4211542115#define vcmla_laneq_f16(__p0_206, __p1_206, __p2_206, __p3_206) __extension__ ({ \
42116 float16x4_t __ret_206; \
4211642117 float16x4_t __s0_206 = __p0_206; \
4211742118 float16x4_t __s1_206 = __p1_206; \
4211842119 float16x8_t __s2_206 = __p2_206; \
4211942120 float16x4_t __rev0_206; __rev0_206 = __builtin_shufflevector(__s0_206, __s0_206, 3, 2, 1, 0); \
4212042121 float16x4_t __rev1_206; __rev1_206 = __builtin_shufflevector(__s1_206, __s1_206, 3, 2, 1, 0); \
4212142122 float16x8_t __rev2_206; __rev2_206 = __builtin_shufflevector(__s2_206, __s2_206, 7, 6, 5, 4, 3, 2, 1, 0); \
42122 float16x4_t __ret_206; \
4212342123float16x8_t __reint_206 = __rev2_206; \
4212442124uint32x2_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)}; \
4212542125 __ret_206 = __noswap_vcmla_f16(__rev0_206, __rev1_206, *(float16x4_t *) &__reint1_206); \
......@@ -42130,10 +42130,10 @@ uint32x2_t __reint1_206 = (uint32x2_t) {__noswap_vgetq_lane_u32(*(uint32x4_t *)
4213042130
4213142131#ifdef __LITTLE_ENDIAN__
4213242132#define vcmlaq_laneq_f16(__p0_207, __p1_207, __p2_207, __p3_207) __extension__ ({ \
42133 float16x8_t __ret_207; \
4213342134 float16x8_t __s0_207 = __p0_207; \
4213442135 float16x8_t __s1_207 = __p1_207; \
4213542136 float16x8_t __s2_207 = __p2_207; \
42136 float16x8_t __ret_207; \
4213742137float16x8_t __reint_207 = __s2_207; \
4213842138uint32x4_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)}; \
4213942139 __ret_207 = vcmlaq_f16(__s0_207, __s1_207, *(float16x8_t *) &__reint1_207); \
......@@ -42141,13 +42141,13 @@ uint32x4_t __reint1_207 = (uint32x4_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4214142141})
4214242142#else
4214342143#define vcmlaq_laneq_f16(__p0_208, __p1_208, __p2_208, __p3_208) __extension__ ({ \
42144 float16x8_t __ret_208; \
4214442145 float16x8_t __s0_208 = __p0_208; \
4214542146 float16x8_t __s1_208 = __p1_208; \
4214642147 float16x8_t __s2_208 = __p2_208; \
4214742148 float16x8_t __rev0_208; __rev0_208 = __builtin_shufflevector(__s0_208, __s0_208, 7, 6, 5, 4, 3, 2, 1, 0); \
4214842149 float16x8_t __rev1_208; __rev1_208 = __builtin_shufflevector(__s1_208, __s1_208, 7, 6, 5, 4, 3, 2, 1, 0); \
4214942150 float16x8_t __rev2_208; __rev2_208 = __builtin_shufflevector(__s2_208, __s2_208, 7, 6, 5, 4, 3, 2, 1, 0); \
42150 float16x8_t __ret_208; \
4215142151float16x8_t __reint_208 = __rev2_208; \
4215242152uint32x4_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)}; \
4215342153 __ret_208 = __noswap_vcmlaq_f16(__rev0_208, __rev1_208, *(float16x8_t *) &__reint1_208); \
......@@ -42164,10 +42164,10 @@ __ai float16x8_t vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8
4216442164}
4216542165#else
4216642166__ai float16x8_t vcmlaq_rot180_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42167 float16x8_t __ret;
4216742168 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4216842169 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4216942170 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42170 float16x8_t __ret;
4217142171 __ret = (float16x8_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
4217242172 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4217342173 return __ret;
......@@ -42187,10 +42187,10 @@ __ai float16x4_t vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_
4218742187}
4218842188#else
4218942189__ai float16x4_t vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42190 float16x4_t __ret;
4219042191 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4219142192 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4219242193 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42193 float16x4_t __ret;
4219442194 __ret = (float16x4_t) __builtin_neon_vcmla_rot180_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
4219542195 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4219642196 return __ret;
......@@ -42204,10 +42204,10 @@ __ai float16x4_t __noswap_vcmla_rot180_f16(float16x4_t __p0, float16x4_t __p1, f
4220442204
4220542205#ifdef __LITTLE_ENDIAN__
4220642206#define vcmla_rot180_lane_f16(__p0_209, __p1_209, __p2_209, __p3_209) __extension__ ({ \
42207 float16x4_t __ret_209; \
4220742208 float16x4_t __s0_209 = __p0_209; \
4220842209 float16x4_t __s1_209 = __p1_209; \
4220942210 float16x4_t __s2_209 = __p2_209; \
42210 float16x4_t __ret_209; \
4221142211float16x4_t __reint_209 = __s2_209; \
4221242212uint32x2_t __reint1_209 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_209, __p3_209), vget_lane_u32(*(uint32x2_t *) &__reint_209, __p3_209)}; \
4221342213 __ret_209 = vcmla_rot180_f16(__s0_209, __s1_209, *(float16x4_t *) &__reint1_209); \
......@@ -42215,13 +42215,13 @@ uint32x2_t __reint1_209 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4221542215})
4221642216#else
4221742217#define vcmla_rot180_lane_f16(__p0_210, __p1_210, __p2_210, __p3_210) __extension__ ({ \
42218 float16x4_t __ret_210; \
4221842219 float16x4_t __s0_210 = __p0_210; \
4221942220 float16x4_t __s1_210 = __p1_210; \
4222042221 float16x4_t __s2_210 = __p2_210; \
4222142222 float16x4_t __rev0_210; __rev0_210 = __builtin_shufflevector(__s0_210, __s0_210, 3, 2, 1, 0); \
4222242223 float16x4_t __rev1_210; __rev1_210 = __builtin_shufflevector(__s1_210, __s1_210, 3, 2, 1, 0); \
4222342224 float16x4_t __rev2_210; __rev2_210 = __builtin_shufflevector(__s2_210, __s2_210, 3, 2, 1, 0); \
42224 float16x4_t __ret_210; \
4222542225float16x4_t __reint_210 = __rev2_210; \
4222642226uint32x2_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)}; \
4222742227 __ret_210 = __noswap_vcmla_rot180_f16(__rev0_210, __rev1_210, *(float16x4_t *) &__reint1_210); \
......@@ -42232,10 +42232,10 @@ uint32x2_t __reint1_210 = (uint32x2_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4223242232
4223342233#ifdef __LITTLE_ENDIAN__
4223442234#define vcmlaq_rot180_lane_f16(__p0_211, __p1_211, __p2_211, __p3_211) __extension__ ({ \
42235 float16x8_t __ret_211; \
4223542236 float16x8_t __s0_211 = __p0_211; \
4223642237 float16x8_t __s1_211 = __p1_211; \
4223742238 float16x4_t __s2_211 = __p2_211; \
42238 float16x8_t __ret_211; \
4223942239float16x4_t __reint_211 = __s2_211; \
4224042240uint32x4_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)}; \
4224142241 __ret_211 = vcmlaq_rot180_f16(__s0_211, __s1_211, *(float16x8_t *) &__reint1_211); \
......@@ -42243,13 +42243,13 @@ uint32x4_t __reint1_211 = (uint32x4_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4224342243})
4224442244#else
4224542245#define vcmlaq_rot180_lane_f16(__p0_212, __p1_212, __p2_212, __p3_212) __extension__ ({ \
42246 float16x8_t __ret_212; \
4224642247 float16x8_t __s0_212 = __p0_212; \
4224742248 float16x8_t __s1_212 = __p1_212; \
4224842249 float16x4_t __s2_212 = __p2_212; \
4224942250 float16x8_t __rev0_212; __rev0_212 = __builtin_shufflevector(__s0_212, __s0_212, 7, 6, 5, 4, 3, 2, 1, 0); \
4225042251 float16x8_t __rev1_212; __rev1_212 = __builtin_shufflevector(__s1_212, __s1_212, 7, 6, 5, 4, 3, 2, 1, 0); \
4225142252 float16x4_t __rev2_212; __rev2_212 = __builtin_shufflevector(__s2_212, __s2_212, 3, 2, 1, 0); \
42252 float16x8_t __ret_212; \
4225342253float16x4_t __reint_212 = __rev2_212; \
4225442254uint32x4_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)}; \
4225542255 __ret_212 = __noswap_vcmlaq_rot180_f16(__rev0_212, __rev1_212, *(float16x8_t *) &__reint1_212); \
......@@ -42260,10 +42260,10 @@ uint32x4_t __reint1_212 = (uint32x4_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4226042260
4226142261#ifdef __LITTLE_ENDIAN__
4226242262#define vcmla_rot180_laneq_f16(__p0_213, __p1_213, __p2_213, __p3_213) __extension__ ({ \
42263 float16x4_t __ret_213; \
4226342264 float16x4_t __s0_213 = __p0_213; \
4226442265 float16x4_t __s1_213 = __p1_213; \
4226542266 float16x8_t __s2_213 = __p2_213; \
42266 float16x4_t __ret_213; \
4226742267float16x8_t __reint_213 = __s2_213; \
4226842268uint32x2_t __reint1_213 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_213, __p3_213), vgetq_lane_u32(*(uint32x4_t *) &__reint_213, __p3_213)}; \
4226942269 __ret_213 = vcmla_rot180_f16(__s0_213, __s1_213, *(float16x4_t *) &__reint1_213); \
......@@ -42271,13 +42271,13 @@ uint32x2_t __reint1_213 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4227142271})
4227242272#else
4227342273#define vcmla_rot180_laneq_f16(__p0_214, __p1_214, __p2_214, __p3_214) __extension__ ({ \
42274 float16x4_t __ret_214; \
4227442275 float16x4_t __s0_214 = __p0_214; \
4227542276 float16x4_t __s1_214 = __p1_214; \
4227642277 float16x8_t __s2_214 = __p2_214; \
4227742278 float16x4_t __rev0_214; __rev0_214 = __builtin_shufflevector(__s0_214, __s0_214, 3, 2, 1, 0); \
4227842279 float16x4_t __rev1_214; __rev1_214 = __builtin_shufflevector(__s1_214, __s1_214, 3, 2, 1, 0); \
4227942280 float16x8_t __rev2_214; __rev2_214 = __builtin_shufflevector(__s2_214, __s2_214, 7, 6, 5, 4, 3, 2, 1, 0); \
42280 float16x4_t __ret_214; \
4228142281float16x8_t __reint_214 = __rev2_214; \
4228242282uint32x2_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)}; \
4228342283 __ret_214 = __noswap_vcmla_rot180_f16(__rev0_214, __rev1_214, *(float16x4_t *) &__reint1_214); \
......@@ -42288,10 +42288,10 @@ uint32x2_t __reint1_214 = (uint32x2_t) {__noswap_vgetq_lane_u32(*(uint32x4_t *)
4228842288
4228942289#ifdef __LITTLE_ENDIAN__
4229042290#define vcmlaq_rot180_laneq_f16(__p0_215, __p1_215, __p2_215, __p3_215) __extension__ ({ \
42291 float16x8_t __ret_215; \
4229142292 float16x8_t __s0_215 = __p0_215; \
4229242293 float16x8_t __s1_215 = __p1_215; \
4229342294 float16x8_t __s2_215 = __p2_215; \
42294 float16x8_t __ret_215; \
4229542295float16x8_t __reint_215 = __s2_215; \
4229642296uint32x4_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)}; \
4229742297 __ret_215 = vcmlaq_rot180_f16(__s0_215, __s1_215, *(float16x8_t *) &__reint1_215); \
......@@ -42299,13 +42299,13 @@ uint32x4_t __reint1_215 = (uint32x4_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4229942299})
4230042300#else
4230142301#define vcmlaq_rot180_laneq_f16(__p0_216, __p1_216, __p2_216, __p3_216) __extension__ ({ \
42302 float16x8_t __ret_216; \
4230242303 float16x8_t __s0_216 = __p0_216; \
4230342304 float16x8_t __s1_216 = __p1_216; \
4230442305 float16x8_t __s2_216 = __p2_216; \
4230542306 float16x8_t __rev0_216; __rev0_216 = __builtin_shufflevector(__s0_216, __s0_216, 7, 6, 5, 4, 3, 2, 1, 0); \
4230642307 float16x8_t __rev1_216; __rev1_216 = __builtin_shufflevector(__s1_216, __s1_216, 7, 6, 5, 4, 3, 2, 1, 0); \
4230742308 float16x8_t __rev2_216; __rev2_216 = __builtin_shufflevector(__s2_216, __s2_216, 7, 6, 5, 4, 3, 2, 1, 0); \
42308 float16x8_t __ret_216; \
4230942309float16x8_t __reint_216 = __rev2_216; \
4231042310uint32x4_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)}; \
4231142311 __ret_216 = __noswap_vcmlaq_rot180_f16(__rev0_216, __rev1_216, *(float16x8_t *) &__reint1_216); \
......@@ -42322,10 +42322,10 @@ __ai float16x8_t vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8
4232242322}
4232342323#else
4232442324__ai float16x8_t vcmlaq_rot270_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42325 float16x8_t __ret;
4232542326 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4232642327 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4232742328 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42328 float16x8_t __ret;
4232942329 __ret = (float16x8_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
4233042330 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4233142331 return __ret;
......@@ -42345,10 +42345,10 @@ __ai float16x4_t vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_
4234542345}
4234642346#else
4234742347__ai float16x4_t vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42348 float16x4_t __ret;
4234842349 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4234942350 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4235042351 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42351 float16x4_t __ret;
4235242352 __ret = (float16x4_t) __builtin_neon_vcmla_rot270_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
4235342353 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4235442354 return __ret;
......@@ -42362,10 +42362,10 @@ __ai float16x4_t __noswap_vcmla_rot270_f16(float16x4_t __p0, float16x4_t __p1, f
4236242362
4236342363#ifdef __LITTLE_ENDIAN__
4236442364#define vcmla_rot270_lane_f16(__p0_217, __p1_217, __p2_217, __p3_217) __extension__ ({ \
42365 float16x4_t __ret_217; \
4236542366 float16x4_t __s0_217 = __p0_217; \
4236642367 float16x4_t __s1_217 = __p1_217; \
4236742368 float16x4_t __s2_217 = __p2_217; \
42368 float16x4_t __ret_217; \
4236942369float16x4_t __reint_217 = __s2_217; \
4237042370uint32x2_t __reint1_217 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_217, __p3_217), vget_lane_u32(*(uint32x2_t *) &__reint_217, __p3_217)}; \
4237142371 __ret_217 = vcmla_rot270_f16(__s0_217, __s1_217, *(float16x4_t *) &__reint1_217); \
......@@ -42373,13 +42373,13 @@ uint32x2_t __reint1_217 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4237342373})
4237442374#else
4237542375#define vcmla_rot270_lane_f16(__p0_218, __p1_218, __p2_218, __p3_218) __extension__ ({ \
42376 float16x4_t __ret_218; \
4237642377 float16x4_t __s0_218 = __p0_218; \
4237742378 float16x4_t __s1_218 = __p1_218; \
4237842379 float16x4_t __s2_218 = __p2_218; \
4237942380 float16x4_t __rev0_218; __rev0_218 = __builtin_shufflevector(__s0_218, __s0_218, 3, 2, 1, 0); \
4238042381 float16x4_t __rev1_218; __rev1_218 = __builtin_shufflevector(__s1_218, __s1_218, 3, 2, 1, 0); \
4238142382 float16x4_t __rev2_218; __rev2_218 = __builtin_shufflevector(__s2_218, __s2_218, 3, 2, 1, 0); \
42382 float16x4_t __ret_218; \
4238342383float16x4_t __reint_218 = __rev2_218; \
4238442384uint32x2_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)}; \
4238542385 __ret_218 = __noswap_vcmla_rot270_f16(__rev0_218, __rev1_218, *(float16x4_t *) &__reint1_218); \
......@@ -42390,10 +42390,10 @@ uint32x2_t __reint1_218 = (uint32x2_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4239042390
4239142391#ifdef __LITTLE_ENDIAN__
4239242392#define vcmlaq_rot270_lane_f16(__p0_219, __p1_219, __p2_219, __p3_219) __extension__ ({ \
42393 float16x8_t __ret_219; \
4239342394 float16x8_t __s0_219 = __p0_219; \
4239442395 float16x8_t __s1_219 = __p1_219; \
4239542396 float16x4_t __s2_219 = __p2_219; \
42396 float16x8_t __ret_219; \
4239742397float16x4_t __reint_219 = __s2_219; \
4239842398uint32x4_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)}; \
4239942399 __ret_219 = vcmlaq_rot270_f16(__s0_219, __s1_219, *(float16x8_t *) &__reint1_219); \
......@@ -42401,13 +42401,13 @@ uint32x4_t __reint1_219 = (uint32x4_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4240142401})
4240242402#else
4240342403#define vcmlaq_rot270_lane_f16(__p0_220, __p1_220, __p2_220, __p3_220) __extension__ ({ \
42404 float16x8_t __ret_220; \
4240442405 float16x8_t __s0_220 = __p0_220; \
4240542406 float16x8_t __s1_220 = __p1_220; \
4240642407 float16x4_t __s2_220 = __p2_220; \
4240742408 float16x8_t __rev0_220; __rev0_220 = __builtin_shufflevector(__s0_220, __s0_220, 7, 6, 5, 4, 3, 2, 1, 0); \
4240842409 float16x8_t __rev1_220; __rev1_220 = __builtin_shufflevector(__s1_220, __s1_220, 7, 6, 5, 4, 3, 2, 1, 0); \
4240942410 float16x4_t __rev2_220; __rev2_220 = __builtin_shufflevector(__s2_220, __s2_220, 3, 2, 1, 0); \
42410 float16x8_t __ret_220; \
4241142411float16x4_t __reint_220 = __rev2_220; \
4241242412uint32x4_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)}; \
4241342413 __ret_220 = __noswap_vcmlaq_rot270_f16(__rev0_220, __rev1_220, *(float16x8_t *) &__reint1_220); \
......@@ -42418,10 +42418,10 @@ uint32x4_t __reint1_220 = (uint32x4_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4241842418
4241942419#ifdef __LITTLE_ENDIAN__
4242042420#define vcmla_rot270_laneq_f16(__p0_221, __p1_221, __p2_221, __p3_221) __extension__ ({ \
42421 float16x4_t __ret_221; \
4242142422 float16x4_t __s0_221 = __p0_221; \
4242242423 float16x4_t __s1_221 = __p1_221; \
4242342424 float16x8_t __s2_221 = __p2_221; \
42424 float16x4_t __ret_221; \
4242542425float16x8_t __reint_221 = __s2_221; \
4242642426uint32x2_t __reint1_221 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_221, __p3_221), vgetq_lane_u32(*(uint32x4_t *) &__reint_221, __p3_221)}; \
4242742427 __ret_221 = vcmla_rot270_f16(__s0_221, __s1_221, *(float16x4_t *) &__reint1_221); \
......@@ -42429,13 +42429,13 @@ uint32x2_t __reint1_221 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4242942429})
4243042430#else
4243142431#define vcmla_rot270_laneq_f16(__p0_222, __p1_222, __p2_222, __p3_222) __extension__ ({ \
42432 float16x4_t __ret_222; \
4243242433 float16x4_t __s0_222 = __p0_222; \
4243342434 float16x4_t __s1_222 = __p1_222; \
4243442435 float16x8_t __s2_222 = __p2_222; \
4243542436 float16x4_t __rev0_222; __rev0_222 = __builtin_shufflevector(__s0_222, __s0_222, 3, 2, 1, 0); \
4243642437 float16x4_t __rev1_222; __rev1_222 = __builtin_shufflevector(__s1_222, __s1_222, 3, 2, 1, 0); \
4243742438 float16x8_t __rev2_222; __rev2_222 = __builtin_shufflevector(__s2_222, __s2_222, 7, 6, 5, 4, 3, 2, 1, 0); \
42438 float16x4_t __ret_222; \
4243942439float16x8_t __reint_222 = __rev2_222; \
4244042440uint32x2_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)}; \
4244142441 __ret_222 = __noswap_vcmla_rot270_f16(__rev0_222, __rev1_222, *(float16x4_t *) &__reint1_222); \
......@@ -42446,10 +42446,10 @@ uint32x2_t __reint1_222 = (uint32x2_t) {__noswap_vgetq_lane_u32(*(uint32x4_t *)
4244642446
4244742447#ifdef __LITTLE_ENDIAN__
4244842448#define vcmlaq_rot270_laneq_f16(__p0_223, __p1_223, __p2_223, __p3_223) __extension__ ({ \
42449 float16x8_t __ret_223; \
4244942450 float16x8_t __s0_223 = __p0_223; \
4245042451 float16x8_t __s1_223 = __p1_223; \
4245142452 float16x8_t __s2_223 = __p2_223; \
42452 float16x8_t __ret_223; \
4245342453float16x8_t __reint_223 = __s2_223; \
4245442454uint32x4_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)}; \
4245542455 __ret_223 = vcmlaq_rot270_f16(__s0_223, __s1_223, *(float16x8_t *) &__reint1_223); \
......@@ -42457,13 +42457,13 @@ uint32x4_t __reint1_223 = (uint32x4_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4245742457})
4245842458#else
4245942459#define vcmlaq_rot270_laneq_f16(__p0_224, __p1_224, __p2_224, __p3_224) __extension__ ({ \
42460 float16x8_t __ret_224; \
4246042461 float16x8_t __s0_224 = __p0_224; \
4246142462 float16x8_t __s1_224 = __p1_224; \
4246242463 float16x8_t __s2_224 = __p2_224; \
4246342464 float16x8_t __rev0_224; __rev0_224 = __builtin_shufflevector(__s0_224, __s0_224, 7, 6, 5, 4, 3, 2, 1, 0); \
4246442465 float16x8_t __rev1_224; __rev1_224 = __builtin_shufflevector(__s1_224, __s1_224, 7, 6, 5, 4, 3, 2, 1, 0); \
4246542466 float16x8_t __rev2_224; __rev2_224 = __builtin_shufflevector(__s2_224, __s2_224, 7, 6, 5, 4, 3, 2, 1, 0); \
42466 float16x8_t __ret_224; \
4246742467float16x8_t __reint_224 = __rev2_224; \
4246842468uint32x4_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)}; \
4246942469 __ret_224 = __noswap_vcmlaq_rot270_f16(__rev0_224, __rev1_224, *(float16x8_t *) &__reint1_224); \
......@@ -42480,10 +42480,10 @@ __ai float16x8_t vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_
4248042480}
4248142481#else
4248242482__ai float16x8_t vcmlaq_rot90_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
42483 float16x8_t __ret;
4248342484 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4248442485 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4248542486 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
42486 float16x8_t __ret;
4248742487 __ret = (float16x8_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
4248842488 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4248942489 return __ret;
......@@ -42503,10 +42503,10 @@ __ai float16x4_t vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t
4250342503}
4250442504#else
4250542505__ai float16x4_t vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
42506 float16x4_t __ret;
4250642507 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4250742508 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4250842509 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
42509 float16x4_t __ret;
4251042510 __ret = (float16x4_t) __builtin_neon_vcmla_rot90_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
4251142511 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4251242512 return __ret;
......@@ -42520,10 +42520,10 @@ __ai float16x4_t __noswap_vcmla_rot90_f16(float16x4_t __p0, float16x4_t __p1, fl
4252042520
4252142521#ifdef __LITTLE_ENDIAN__
4252242522#define vcmla_rot90_lane_f16(__p0_225, __p1_225, __p2_225, __p3_225) __extension__ ({ \
42523 float16x4_t __ret_225; \
4252342524 float16x4_t __s0_225 = __p0_225; \
4252442525 float16x4_t __s1_225 = __p1_225; \
4252542526 float16x4_t __s2_225 = __p2_225; \
42526 float16x4_t __ret_225; \
4252742527float16x4_t __reint_225 = __s2_225; \
4252842528uint32x2_t __reint1_225 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_225, __p3_225), vget_lane_u32(*(uint32x2_t *) &__reint_225, __p3_225)}; \
4252942529 __ret_225 = vcmla_rot90_f16(__s0_225, __s1_225, *(float16x4_t *) &__reint1_225); \
......@@ -42531,13 +42531,13 @@ uint32x2_t __reint1_225 = (uint32x2_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4253142531})
4253242532#else
4253342533#define vcmla_rot90_lane_f16(__p0_226, __p1_226, __p2_226, __p3_226) __extension__ ({ \
42534 float16x4_t __ret_226; \
4253442535 float16x4_t __s0_226 = __p0_226; \
4253542536 float16x4_t __s1_226 = __p1_226; \
4253642537 float16x4_t __s2_226 = __p2_226; \
4253742538 float16x4_t __rev0_226; __rev0_226 = __builtin_shufflevector(__s0_226, __s0_226, 3, 2, 1, 0); \
4253842539 float16x4_t __rev1_226; __rev1_226 = __builtin_shufflevector(__s1_226, __s1_226, 3, 2, 1, 0); \
4253942540 float16x4_t __rev2_226; __rev2_226 = __builtin_shufflevector(__s2_226, __s2_226, 3, 2, 1, 0); \
42540 float16x4_t __ret_226; \
4254142541float16x4_t __reint_226 = __rev2_226; \
4254242542uint32x2_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)}; \
4254342543 __ret_226 = __noswap_vcmla_rot90_f16(__rev0_226, __rev1_226, *(float16x4_t *) &__reint1_226); \
......@@ -42548,10 +42548,10 @@ uint32x2_t __reint1_226 = (uint32x2_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4254842548
4254942549#ifdef __LITTLE_ENDIAN__
4255042550#define vcmlaq_rot90_lane_f16(__p0_227, __p1_227, __p2_227, __p3_227) __extension__ ({ \
42551 float16x8_t __ret_227; \
4255142552 float16x8_t __s0_227 = __p0_227; \
4255242553 float16x8_t __s1_227 = __p1_227; \
4255342554 float16x4_t __s2_227 = __p2_227; \
42554 float16x8_t __ret_227; \
4255542555float16x4_t __reint_227 = __s2_227; \
4255642556uint32x4_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)}; \
4255742557 __ret_227 = vcmlaq_rot90_f16(__s0_227, __s1_227, *(float16x8_t *) &__reint1_227); \
......@@ -42559,13 +42559,13 @@ uint32x4_t __reint1_227 = (uint32x4_t) {vget_lane_u32(*(uint32x2_t *) &__reint_2
4255942559})
4256042560#else
4256142561#define vcmlaq_rot90_lane_f16(__p0_228, __p1_228, __p2_228, __p3_228) __extension__ ({ \
42562 float16x8_t __ret_228; \
4256242563 float16x8_t __s0_228 = __p0_228; \
4256342564 float16x8_t __s1_228 = __p1_228; \
4256442565 float16x4_t __s2_228 = __p2_228; \
4256542566 float16x8_t __rev0_228; __rev0_228 = __builtin_shufflevector(__s0_228, __s0_228, 7, 6, 5, 4, 3, 2, 1, 0); \
4256642567 float16x8_t __rev1_228; __rev1_228 = __builtin_shufflevector(__s1_228, __s1_228, 7, 6, 5, 4, 3, 2, 1, 0); \
4256742568 float16x4_t __rev2_228; __rev2_228 = __builtin_shufflevector(__s2_228, __s2_228, 3, 2, 1, 0); \
42568 float16x8_t __ret_228; \
4256942569float16x4_t __reint_228 = __rev2_228; \
4257042570uint32x4_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)}; \
4257142571 __ret_228 = __noswap_vcmlaq_rot90_f16(__rev0_228, __rev1_228, *(float16x8_t *) &__reint1_228); \
......@@ -42576,10 +42576,10 @@ uint32x4_t __reint1_228 = (uint32x4_t) {__noswap_vget_lane_u32(*(uint32x2_t *) &
4257642576
4257742577#ifdef __LITTLE_ENDIAN__
4257842578#define vcmla_rot90_laneq_f16(__p0_229, __p1_229, __p2_229, __p3_229) __extension__ ({ \
42579 float16x4_t __ret_229; \
4257942580 float16x4_t __s0_229 = __p0_229; \
4258042581 float16x4_t __s1_229 = __p1_229; \
4258142582 float16x8_t __s2_229 = __p2_229; \
42582 float16x4_t __ret_229; \
4258342583float16x8_t __reint_229 = __s2_229; \
4258442584uint32x2_t __reint1_229 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_229, __p3_229), vgetq_lane_u32(*(uint32x4_t *) &__reint_229, __p3_229)}; \
4258542585 __ret_229 = vcmla_rot90_f16(__s0_229, __s1_229, *(float16x4_t *) &__reint1_229); \
......@@ -42587,13 +42587,13 @@ uint32x2_t __reint1_229 = (uint32x2_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4258742587})
4258842588#else
4258942589#define vcmla_rot90_laneq_f16(__p0_230, __p1_230, __p2_230, __p3_230) __extension__ ({ \
42590 float16x4_t __ret_230; \
4259042591 float16x4_t __s0_230 = __p0_230; \
4259142592 float16x4_t __s1_230 = __p1_230; \
4259242593 float16x8_t __s2_230 = __p2_230; \
4259342594 float16x4_t __rev0_230; __rev0_230 = __builtin_shufflevector(__s0_230, __s0_230, 3, 2, 1, 0); \
4259442595 float16x4_t __rev1_230; __rev1_230 = __builtin_shufflevector(__s1_230, __s1_230, 3, 2, 1, 0); \
4259542596 float16x8_t __rev2_230; __rev2_230 = __builtin_shufflevector(__s2_230, __s2_230, 7, 6, 5, 4, 3, 2, 1, 0); \
42596 float16x4_t __ret_230; \
4259742597float16x8_t __reint_230 = __rev2_230; \
4259842598uint32x2_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)}; \
4259942599 __ret_230 = __noswap_vcmla_rot90_f16(__rev0_230, __rev1_230, *(float16x4_t *) &__reint1_230); \
......@@ -42604,10 +42604,10 @@ uint32x2_t __reint1_230 = (uint32x2_t) {__noswap_vgetq_lane_u32(*(uint32x4_t *)
4260442604
4260542605#ifdef __LITTLE_ENDIAN__
4260642606#define vcmlaq_rot90_laneq_f16(__p0_231, __p1_231, __p2_231, __p3_231) __extension__ ({ \
42607 float16x8_t __ret_231; \
4260742608 float16x8_t __s0_231 = __p0_231; \
4260842609 float16x8_t __s1_231 = __p1_231; \
4260942610 float16x8_t __s2_231 = __p2_231; \
42610 float16x8_t __ret_231; \
4261142611float16x8_t __reint_231 = __s2_231; \
4261242612uint32x4_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)}; \
4261342613 __ret_231 = vcmlaq_rot90_f16(__s0_231, __s1_231, *(float16x8_t *) &__reint1_231); \
......@@ -42615,13 +42615,13 @@ uint32x4_t __reint1_231 = (uint32x4_t) {vgetq_lane_u32(*(uint32x4_t *) &__reint_
4261542615})
4261642616#else
4261742617#define vcmlaq_rot90_laneq_f16(__p0_232, __p1_232, __p2_232, __p3_232) __extension__ ({ \
42618 float16x8_t __ret_232; \
4261842619 float16x8_t __s0_232 = __p0_232; \
4261942620 float16x8_t __s1_232 = __p1_232; \
4262042621 float16x8_t __s2_232 = __p2_232; \
4262142622 float16x8_t __rev0_232; __rev0_232 = __builtin_shufflevector(__s0_232, __s0_232, 7, 6, 5, 4, 3, 2, 1, 0); \
4262242623 float16x8_t __rev1_232; __rev1_232 = __builtin_shufflevector(__s1_232, __s1_232, 7, 6, 5, 4, 3, 2, 1, 0); \
4262342624 float16x8_t __rev2_232; __rev2_232 = __builtin_shufflevector(__s2_232, __s2_232, 7, 6, 5, 4, 3, 2, 1, 0); \
42624 float16x8_t __ret_232; \
4262542625float16x8_t __reint_232 = __rev2_232; \
4262642626uint32x4_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)}; \
4262742627 __ret_232 = __noswap_vcmlaq_rot90_f16(__rev0_232, __rev1_232, *(float16x8_t *) &__reint1_232); \
......@@ -42640,9 +42640,9 @@ __ai float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
4264042640}
4264142641#else
4264242642__ai float64x2_t vcaddq_rot270_f64(float64x2_t __p0, float64x2_t __p1) {
42643 float64x2_t __ret;
4264342644 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4264442645 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42645 float64x2_t __ret;
4264642646 __ret = (float64x2_t) __builtin_neon_vcaddq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
4264742647 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4264842648 return __ret;
......@@ -42657,9 +42657,9 @@ __ai float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
4265742657}
4265842658#else
4265942659__ai float64x2_t vcaddq_rot90_f64(float64x2_t __p0, float64x2_t __p1) {
42660 float64x2_t __ret;
4266042661 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4266142662 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42662 float64x2_t __ret;
4266342663 __ret = (float64x2_t) __builtin_neon_vcaddq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
4266442664 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4266542665 return __ret;
......@@ -42674,10 +42674,10 @@ __ai float64x2_t vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2
4267442674}
4267542675#else
4267642676__ai float64x2_t vcmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42677 float64x2_t __ret;
4267742678 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4267842679 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4267942680 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
42680 float64x2_t __ret;
4268142681 __ret = (float64x2_t) __builtin_neon_vcmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
4268242682 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4268342683 return __ret;
......@@ -42695,10 +42695,10 @@ __ai float64x1_t vcmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
4269542695 return __ret;
4269642696}
4269742697#define vcmla_lane_f64(__p0_233, __p1_233, __p2_233, __p3_233) __extension__ ({ \
42698 float64x1_t __ret_233; \
4269842699 float64x1_t __s0_233 = __p0_233; \
4269942700 float64x1_t __s1_233 = __p1_233; \
4270042701 float64x1_t __s2_233 = __p2_233; \
42701 float64x1_t __ret_233; \
4270242702float64x1_t __reint_233 = __s2_233; \
4270342703uint64x2_t __reint1_233 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_233, __p3_233), vgetq_lane_u64(*(uint64x2_t *) &__reint_233, __p3_233)}; \
4270442704 __ret_233 = vcmla_f64(__s0_233, __s1_233, *(float64x1_t *) &__reint1_233); \
......@@ -42706,10 +42706,10 @@ uint64x2_t __reint1_233 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4270642706})
4270742707#ifdef __LITTLE_ENDIAN__
4270842708#define vcmlaq_lane_f64(__p0_234, __p1_234, __p2_234, __p3_234) __extension__ ({ \
42709 float64x2_t __ret_234; \
4270942710 float64x2_t __s0_234 = __p0_234; \
4271042711 float64x2_t __s1_234 = __p1_234; \
4271142712 float64x1_t __s2_234 = __p2_234; \
42712 float64x2_t __ret_234; \
4271342713float64x1_t __reint_234 = __s2_234; \
4271442714uint64x2_t __reint1_234 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_234, __p3_234), vgetq_lane_u64(*(uint64x2_t *) &__reint_234, __p3_234)}; \
4271542715 __ret_234 = vcmlaq_f64(__s0_234, __s1_234, *(float64x2_t *) &__reint1_234); \
......@@ -42717,12 +42717,12 @@ uint64x2_t __reint1_234 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4271742717})
4271842718#else
4271942719#define vcmlaq_lane_f64(__p0_235, __p1_235, __p2_235, __p3_235) __extension__ ({ \
42720 float64x2_t __ret_235; \
4272042721 float64x2_t __s0_235 = __p0_235; \
4272142722 float64x2_t __s1_235 = __p1_235; \
4272242723 float64x1_t __s2_235 = __p2_235; \
4272342724 float64x2_t __rev0_235; __rev0_235 = __builtin_shufflevector(__s0_235, __s0_235, 1, 0); \
4272442725 float64x2_t __rev1_235; __rev1_235 = __builtin_shufflevector(__s1_235, __s1_235, 1, 0); \
42725 float64x2_t __ret_235; \
4272642726float64x1_t __reint_235 = __s2_235; \
4272742727uint64x2_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)}; \
4272842728 __ret_235 = __noswap_vcmlaq_f64(__rev0_235, __rev1_235, *(float64x2_t *) &__reint1_235); \
......@@ -42733,10 +42733,10 @@ uint64x2_t __reint1_235 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4273342733
4273442734#ifdef __LITTLE_ENDIAN__
4273542735#define vcmla_laneq_f64(__p0_236, __p1_236, __p2_236, __p3_236) __extension__ ({ \
42736 float64x1_t __ret_236; \
4273642737 float64x1_t __s0_236 = __p0_236; \
4273742738 float64x1_t __s1_236 = __p1_236; \
4273842739 float64x2_t __s2_236 = __p2_236; \
42739 float64x1_t __ret_236; \
4274042740float64x2_t __reint_236 = __s2_236; \
4274142741uint64x2_t __reint1_236 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_236, __p3_236), vgetq_lane_u64(*(uint64x2_t *) &__reint_236, __p3_236)}; \
4274242742 __ret_236 = vcmla_f64(__s0_236, __s1_236, *(float64x1_t *) &__reint1_236); \
......@@ -42744,11 +42744,11 @@ uint64x2_t __reint1_236 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4274442744})
4274542745#else
4274642746#define vcmla_laneq_f64(__p0_237, __p1_237, __p2_237, __p3_237) __extension__ ({ \
42747 float64x1_t __ret_237; \
4274742748 float64x1_t __s0_237 = __p0_237; \
4274842749 float64x1_t __s1_237 = __p1_237; \
4274942750 float64x2_t __s2_237 = __p2_237; \
4275042751 float64x2_t __rev2_237; __rev2_237 = __builtin_shufflevector(__s2_237, __s2_237, 1, 0); \
42751 float64x1_t __ret_237; \
4275242752float64x2_t __reint_237 = __rev2_237; \
4275342753uint64x2_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)}; \
4275442754 __ret_237 = vcmla_f64(__s0_237, __s1_237, *(float64x1_t *) &__reint1_237); \
......@@ -42758,10 +42758,10 @@ uint64x2_t __reint1_237 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4275842758
4275942759#ifdef __LITTLE_ENDIAN__
4276042760#define vcmlaq_laneq_f64(__p0_238, __p1_238, __p2_238, __p3_238) __extension__ ({ \
42761 float64x2_t __ret_238; \
4276142762 float64x2_t __s0_238 = __p0_238; \
4276242763 float64x2_t __s1_238 = __p1_238; \
4276342764 float64x2_t __s2_238 = __p2_238; \
42764 float64x2_t __ret_238; \
4276542765float64x2_t __reint_238 = __s2_238; \
4276642766uint64x2_t __reint1_238 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_238, __p3_238), vgetq_lane_u64(*(uint64x2_t *) &__reint_238, __p3_238)}; \
4276742767 __ret_238 = vcmlaq_f64(__s0_238, __s1_238, *(float64x2_t *) &__reint1_238); \
......@@ -42769,13 +42769,13 @@ uint64x2_t __reint1_238 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4276942769})
4277042770#else
4277142771#define vcmlaq_laneq_f64(__p0_239, __p1_239, __p2_239, __p3_239) __extension__ ({ \
42772 float64x2_t __ret_239; \
4277242773 float64x2_t __s0_239 = __p0_239; \
4277342774 float64x2_t __s1_239 = __p1_239; \
4277442775 float64x2_t __s2_239 = __p2_239; \
4277542776 float64x2_t __rev0_239; __rev0_239 = __builtin_shufflevector(__s0_239, __s0_239, 1, 0); \
4277642777 float64x2_t __rev1_239; __rev1_239 = __builtin_shufflevector(__s1_239, __s1_239, 1, 0); \
4277742778 float64x2_t __rev2_239; __rev2_239 = __builtin_shufflevector(__s2_239, __s2_239, 1, 0); \
42778 float64x2_t __ret_239; \
4277942779float64x2_t __reint_239 = __rev2_239; \
4278042780uint64x2_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)}; \
4278142781 __ret_239 = __noswap_vcmlaq_f64(__rev0_239, __rev1_239, *(float64x2_t *) &__reint1_239); \
......@@ -42792,10 +42792,10 @@ __ai float64x2_t vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2
4279242792}
4279342793#else
4279442794__ai float64x2_t vcmlaq_rot180_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42795 float64x2_t __ret;
4279542796 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4279642797 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4279742798 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
42798 float64x2_t __ret;
4279942799 __ret = (float64x2_t) __builtin_neon_vcmlaq_rot180_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
4280042800 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4280142801 return __ret;
......@@ -42813,10 +42813,10 @@ __ai float64x1_t vcmla_rot180_f64(float64x1_t __p0, float64x1_t __p1, float64x1_
4281342813 return __ret;
4281442814}
4281542815#define vcmla_rot180_lane_f64(__p0_240, __p1_240, __p2_240, __p3_240) __extension__ ({ \
42816 float64x1_t __ret_240; \
4281642817 float64x1_t __s0_240 = __p0_240; \
4281742818 float64x1_t __s1_240 = __p1_240; \
4281842819 float64x1_t __s2_240 = __p2_240; \
42819 float64x1_t __ret_240; \
4282042820float64x1_t __reint_240 = __s2_240; \
4282142821uint64x2_t __reint1_240 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_240, __p3_240), vgetq_lane_u64(*(uint64x2_t *) &__reint_240, __p3_240)}; \
4282242822 __ret_240 = vcmla_rot180_f64(__s0_240, __s1_240, *(float64x1_t *) &__reint1_240); \
......@@ -42824,10 +42824,10 @@ uint64x2_t __reint1_240 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4282442824})
4282542825#ifdef __LITTLE_ENDIAN__
4282642826#define vcmlaq_rot180_lane_f64(__p0_241, __p1_241, __p2_241, __p3_241) __extension__ ({ \
42827 float64x2_t __ret_241; \
4282742828 float64x2_t __s0_241 = __p0_241; \
4282842829 float64x2_t __s1_241 = __p1_241; \
4282942830 float64x1_t __s2_241 = __p2_241; \
42830 float64x2_t __ret_241; \
4283142831float64x1_t __reint_241 = __s2_241; \
4283242832uint64x2_t __reint1_241 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_241, __p3_241), vgetq_lane_u64(*(uint64x2_t *) &__reint_241, __p3_241)}; \
4283342833 __ret_241 = vcmlaq_rot180_f64(__s0_241, __s1_241, *(float64x2_t *) &__reint1_241); \
......@@ -42835,12 +42835,12 @@ uint64x2_t __reint1_241 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4283542835})
4283642836#else
4283742837#define vcmlaq_rot180_lane_f64(__p0_242, __p1_242, __p2_242, __p3_242) __extension__ ({ \
42838 float64x2_t __ret_242; \
4283842839 float64x2_t __s0_242 = __p0_242; \
4283942840 float64x2_t __s1_242 = __p1_242; \
4284042841 float64x1_t __s2_242 = __p2_242; \
4284142842 float64x2_t __rev0_242; __rev0_242 = __builtin_shufflevector(__s0_242, __s0_242, 1, 0); \
4284242843 float64x2_t __rev1_242; __rev1_242 = __builtin_shufflevector(__s1_242, __s1_242, 1, 0); \
42843 float64x2_t __ret_242; \
4284442844float64x1_t __reint_242 = __s2_242; \
4284542845uint64x2_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)}; \
4284642846 __ret_242 = __noswap_vcmlaq_rot180_f64(__rev0_242, __rev1_242, *(float64x2_t *) &__reint1_242); \
......@@ -42851,10 +42851,10 @@ uint64x2_t __reint1_242 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4285142851
4285242852#ifdef __LITTLE_ENDIAN__
4285342853#define vcmla_rot180_laneq_f64(__p0_243, __p1_243, __p2_243, __p3_243) __extension__ ({ \
42854 float64x1_t __ret_243; \
4285442855 float64x1_t __s0_243 = __p0_243; \
4285542856 float64x1_t __s1_243 = __p1_243; \
4285642857 float64x2_t __s2_243 = __p2_243; \
42857 float64x1_t __ret_243; \
4285842858float64x2_t __reint_243 = __s2_243; \
4285942859uint64x2_t __reint1_243 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_243, __p3_243), vgetq_lane_u64(*(uint64x2_t *) &__reint_243, __p3_243)}; \
4286042860 __ret_243 = vcmla_rot180_f64(__s0_243, __s1_243, *(float64x1_t *) &__reint1_243); \
......@@ -42862,11 +42862,11 @@ uint64x2_t __reint1_243 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4286242862})
4286342863#else
4286442864#define vcmla_rot180_laneq_f64(__p0_244, __p1_244, __p2_244, __p3_244) __extension__ ({ \
42865 float64x1_t __ret_244; \
4286542866 float64x1_t __s0_244 = __p0_244; \
4286642867 float64x1_t __s1_244 = __p1_244; \
4286742868 float64x2_t __s2_244 = __p2_244; \
4286842869 float64x2_t __rev2_244; __rev2_244 = __builtin_shufflevector(__s2_244, __s2_244, 1, 0); \
42869 float64x1_t __ret_244; \
4287042870float64x2_t __reint_244 = __rev2_244; \
4287142871uint64x2_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)}; \
4287242872 __ret_244 = vcmla_rot180_f64(__s0_244, __s1_244, *(float64x1_t *) &__reint1_244); \
......@@ -42876,10 +42876,10 @@ uint64x2_t __reint1_244 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4287642876
4287742877#ifdef __LITTLE_ENDIAN__
4287842878#define vcmlaq_rot180_laneq_f64(__p0_245, __p1_245, __p2_245, __p3_245) __extension__ ({ \
42879 float64x2_t __ret_245; \
4287942880 float64x2_t __s0_245 = __p0_245; \
4288042881 float64x2_t __s1_245 = __p1_245; \
4288142882 float64x2_t __s2_245 = __p2_245; \
42882 float64x2_t __ret_245; \
4288342883float64x2_t __reint_245 = __s2_245; \
4288442884uint64x2_t __reint1_245 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_245, __p3_245), vgetq_lane_u64(*(uint64x2_t *) &__reint_245, __p3_245)}; \
4288542885 __ret_245 = vcmlaq_rot180_f64(__s0_245, __s1_245, *(float64x2_t *) &__reint1_245); \
......@@ -42887,13 +42887,13 @@ uint64x2_t __reint1_245 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4288742887})
4288842888#else
4288942889#define vcmlaq_rot180_laneq_f64(__p0_246, __p1_246, __p2_246, __p3_246) __extension__ ({ \
42890 float64x2_t __ret_246; \
4289042891 float64x2_t __s0_246 = __p0_246; \
4289142892 float64x2_t __s1_246 = __p1_246; \
4289242893 float64x2_t __s2_246 = __p2_246; \
4289342894 float64x2_t __rev0_246; __rev0_246 = __builtin_shufflevector(__s0_246, __s0_246, 1, 0); \
4289442895 float64x2_t __rev1_246; __rev1_246 = __builtin_shufflevector(__s1_246, __s1_246, 1, 0); \
4289542896 float64x2_t __rev2_246; __rev2_246 = __builtin_shufflevector(__s2_246, __s2_246, 1, 0); \
42896 float64x2_t __ret_246; \
4289742897float64x2_t __reint_246 = __rev2_246; \
4289842898uint64x2_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)}; \
4289942899 __ret_246 = __noswap_vcmlaq_rot180_f64(__rev0_246, __rev1_246, *(float64x2_t *) &__reint1_246); \
......@@ -42910,10 +42910,10 @@ __ai float64x2_t vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2
4291042910}
4291142911#else
4291242912__ai float64x2_t vcmlaq_rot270_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
42913 float64x2_t __ret;
4291342914 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4291442915 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4291542916 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
42916 float64x2_t __ret;
4291742917 __ret = (float64x2_t) __builtin_neon_vcmlaq_rot270_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
4291842918 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4291942919 return __ret;
......@@ -42931,10 +42931,10 @@ __ai float64x1_t vcmla_rot270_f64(float64x1_t __p0, float64x1_t __p1, float64x1_
4293142931 return __ret;
4293242932}
4293342933#define vcmla_rot270_lane_f64(__p0_247, __p1_247, __p2_247, __p3_247) __extension__ ({ \
42934 float64x1_t __ret_247; \
4293442935 float64x1_t __s0_247 = __p0_247; \
4293542936 float64x1_t __s1_247 = __p1_247; \
4293642937 float64x1_t __s2_247 = __p2_247; \
42937 float64x1_t __ret_247; \
4293842938float64x1_t __reint_247 = __s2_247; \
4293942939uint64x2_t __reint1_247 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_247, __p3_247), vgetq_lane_u64(*(uint64x2_t *) &__reint_247, __p3_247)}; \
4294042940 __ret_247 = vcmla_rot270_f64(__s0_247, __s1_247, *(float64x1_t *) &__reint1_247); \
......@@ -42942,10 +42942,10 @@ uint64x2_t __reint1_247 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4294242942})
4294342943#ifdef __LITTLE_ENDIAN__
4294442944#define vcmlaq_rot270_lane_f64(__p0_248, __p1_248, __p2_248, __p3_248) __extension__ ({ \
42945 float64x2_t __ret_248; \
4294542946 float64x2_t __s0_248 = __p0_248; \
4294642947 float64x2_t __s1_248 = __p1_248; \
4294742948 float64x1_t __s2_248 = __p2_248; \
42948 float64x2_t __ret_248; \
4294942949float64x1_t __reint_248 = __s2_248; \
4295042950uint64x2_t __reint1_248 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_248, __p3_248), vgetq_lane_u64(*(uint64x2_t *) &__reint_248, __p3_248)}; \
4295142951 __ret_248 = vcmlaq_rot270_f64(__s0_248, __s1_248, *(float64x2_t *) &__reint1_248); \
......@@ -42953,12 +42953,12 @@ uint64x2_t __reint1_248 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4295342953})
4295442954#else
4295542955#define vcmlaq_rot270_lane_f64(__p0_249, __p1_249, __p2_249, __p3_249) __extension__ ({ \
42956 float64x2_t __ret_249; \
4295642957 float64x2_t __s0_249 = __p0_249; \
4295742958 float64x2_t __s1_249 = __p1_249; \
4295842959 float64x1_t __s2_249 = __p2_249; \
4295942960 float64x2_t __rev0_249; __rev0_249 = __builtin_shufflevector(__s0_249, __s0_249, 1, 0); \
4296042961 float64x2_t __rev1_249; __rev1_249 = __builtin_shufflevector(__s1_249, __s1_249, 1, 0); \
42961 float64x2_t __ret_249; \
4296242962float64x1_t __reint_249 = __s2_249; \
4296342963uint64x2_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)}; \
4296442964 __ret_249 = __noswap_vcmlaq_rot270_f64(__rev0_249, __rev1_249, *(float64x2_t *) &__reint1_249); \
......@@ -42969,10 +42969,10 @@ uint64x2_t __reint1_249 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4296942969
4297042970#ifdef __LITTLE_ENDIAN__
4297142971#define vcmla_rot270_laneq_f64(__p0_250, __p1_250, __p2_250, __p3_250) __extension__ ({ \
42972 float64x1_t __ret_250; \
4297242973 float64x1_t __s0_250 = __p0_250; \
4297342974 float64x1_t __s1_250 = __p1_250; \
4297442975 float64x2_t __s2_250 = __p2_250; \
42975 float64x1_t __ret_250; \
4297642976float64x2_t __reint_250 = __s2_250; \
4297742977uint64x2_t __reint1_250 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_250, __p3_250), vgetq_lane_u64(*(uint64x2_t *) &__reint_250, __p3_250)}; \
4297842978 __ret_250 = vcmla_rot270_f64(__s0_250, __s1_250, *(float64x1_t *) &__reint1_250); \
......@@ -42980,11 +42980,11 @@ uint64x2_t __reint1_250 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4298042980})
4298142981#else
4298242982#define vcmla_rot270_laneq_f64(__p0_251, __p1_251, __p2_251, __p3_251) __extension__ ({ \
42983 float64x1_t __ret_251; \
4298342984 float64x1_t __s0_251 = __p0_251; \
4298442985 float64x1_t __s1_251 = __p1_251; \
4298542986 float64x2_t __s2_251 = __p2_251; \
4298642987 float64x2_t __rev2_251; __rev2_251 = __builtin_shufflevector(__s2_251, __s2_251, 1, 0); \
42987 float64x1_t __ret_251; \
4298842988float64x2_t __reint_251 = __rev2_251; \
4298942989uint64x2_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)}; \
4299042990 __ret_251 = vcmla_rot270_f64(__s0_251, __s1_251, *(float64x1_t *) &__reint1_251); \
......@@ -42994,10 +42994,10 @@ uint64x2_t __reint1_251 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4299442994
4299542995#ifdef __LITTLE_ENDIAN__
4299642996#define vcmlaq_rot270_laneq_f64(__p0_252, __p1_252, __p2_252, __p3_252) __extension__ ({ \
42997 float64x2_t __ret_252; \
4299742998 float64x2_t __s0_252 = __p0_252; \
4299842999 float64x2_t __s1_252 = __p1_252; \
4299943000 float64x2_t __s2_252 = __p2_252; \
43000 float64x2_t __ret_252; \
4300143001float64x2_t __reint_252 = __s2_252; \
4300243002uint64x2_t __reint1_252 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_252, __p3_252), vgetq_lane_u64(*(uint64x2_t *) &__reint_252, __p3_252)}; \
4300343003 __ret_252 = vcmlaq_rot270_f64(__s0_252, __s1_252, *(float64x2_t *) &__reint1_252); \
......@@ -43005,13 +43005,13 @@ uint64x2_t __reint1_252 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4300543005})
4300643006#else
4300743007#define vcmlaq_rot270_laneq_f64(__p0_253, __p1_253, __p2_253, __p3_253) __extension__ ({ \
43008 float64x2_t __ret_253; \
4300843009 float64x2_t __s0_253 = __p0_253; \
4300943010 float64x2_t __s1_253 = __p1_253; \
4301043011 float64x2_t __s2_253 = __p2_253; \
4301143012 float64x2_t __rev0_253; __rev0_253 = __builtin_shufflevector(__s0_253, __s0_253, 1, 0); \
4301243013 float64x2_t __rev1_253; __rev1_253 = __builtin_shufflevector(__s1_253, __s1_253, 1, 0); \
4301343014 float64x2_t __rev2_253; __rev2_253 = __builtin_shufflevector(__s2_253, __s2_253, 1, 0); \
43014 float64x2_t __ret_253; \
4301543015float64x2_t __reint_253 = __rev2_253; \
4301643016uint64x2_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)}; \
4301743017 __ret_253 = __noswap_vcmlaq_rot270_f64(__rev0_253, __rev1_253, *(float64x2_t *) &__reint1_253); \
......@@ -43028,10 +43028,10 @@ __ai float64x2_t vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_
4302843028}
4302943029#else
4303043030__ai float64x2_t vcmlaq_rot90_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
43031 float64x2_t __ret;
4303143032 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4303243033 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4303343034 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
43034 float64x2_t __ret;
4303543035 __ret = (float64x2_t) __builtin_neon_vcmlaq_rot90_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
4303643036 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4303743037 return __ret;
......@@ -43049,10 +43049,10 @@ __ai float64x1_t vcmla_rot90_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t
4304943049 return __ret;
4305043050}
4305143051#define vcmla_rot90_lane_f64(__p0_254, __p1_254, __p2_254, __p3_254) __extension__ ({ \
43052 float64x1_t __ret_254; \
4305243053 float64x1_t __s0_254 = __p0_254; \
4305343054 float64x1_t __s1_254 = __p1_254; \
4305443055 float64x1_t __s2_254 = __p2_254; \
43055 float64x1_t __ret_254; \
4305643056float64x1_t __reint_254 = __s2_254; \
4305743057uint64x2_t __reint1_254 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_254, __p3_254), vgetq_lane_u64(*(uint64x2_t *) &__reint_254, __p3_254)}; \
4305843058 __ret_254 = vcmla_rot90_f64(__s0_254, __s1_254, *(float64x1_t *) &__reint1_254); \
......@@ -43060,10 +43060,10 @@ uint64x2_t __reint1_254 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4306043060})
4306143061#ifdef __LITTLE_ENDIAN__
4306243062#define vcmlaq_rot90_lane_f64(__p0_255, __p1_255, __p2_255, __p3_255) __extension__ ({ \
43063 float64x2_t __ret_255; \
4306343064 float64x2_t __s0_255 = __p0_255; \
4306443065 float64x2_t __s1_255 = __p1_255; \
4306543066 float64x1_t __s2_255 = __p2_255; \
43066 float64x2_t __ret_255; \
4306743067float64x1_t __reint_255 = __s2_255; \
4306843068uint64x2_t __reint1_255 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_255, __p3_255), vgetq_lane_u64(*(uint64x2_t *) &__reint_255, __p3_255)}; \
4306943069 __ret_255 = vcmlaq_rot90_f64(__s0_255, __s1_255, *(float64x2_t *) &__reint1_255); \
......@@ -43071,12 +43071,12 @@ uint64x2_t __reint1_255 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4307143071})
4307243072#else
4307343073#define vcmlaq_rot90_lane_f64(__p0_256, __p1_256, __p2_256, __p3_256) __extension__ ({ \
43074 float64x2_t __ret_256; \
4307443075 float64x2_t __s0_256 = __p0_256; \
4307543076 float64x2_t __s1_256 = __p1_256; \
4307643077 float64x1_t __s2_256 = __p2_256; \
4307743078 float64x2_t __rev0_256; __rev0_256 = __builtin_shufflevector(__s0_256, __s0_256, 1, 0); \
4307843079 float64x2_t __rev1_256; __rev1_256 = __builtin_shufflevector(__s1_256, __s1_256, 1, 0); \
43079 float64x2_t __ret_256; \
4308043080float64x1_t __reint_256 = __s2_256; \
4308143081uint64x2_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)}; \
4308243082 __ret_256 = __noswap_vcmlaq_rot90_f64(__rev0_256, __rev1_256, *(float64x2_t *) &__reint1_256); \
......@@ -43087,10 +43087,10 @@ uint64x2_t __reint1_256 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4308743087
4308843088#ifdef __LITTLE_ENDIAN__
4308943089#define vcmla_rot90_laneq_f64(__p0_257, __p1_257, __p2_257, __p3_257) __extension__ ({ \
43090 float64x1_t __ret_257; \
4309043091 float64x1_t __s0_257 = __p0_257; \
4309143092 float64x1_t __s1_257 = __p1_257; \
4309243093 float64x2_t __s2_257 = __p2_257; \
43093 float64x1_t __ret_257; \
4309443094float64x2_t __reint_257 = __s2_257; \
4309543095uint64x2_t __reint1_257 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_257, __p3_257), vgetq_lane_u64(*(uint64x2_t *) &__reint_257, __p3_257)}; \
4309643096 __ret_257 = vcmla_rot90_f64(__s0_257, __s1_257, *(float64x1_t *) &__reint1_257); \
......@@ -43098,11 +43098,11 @@ uint64x2_t __reint1_257 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4309843098})
4309943099#else
4310043100#define vcmla_rot90_laneq_f64(__p0_258, __p1_258, __p2_258, __p3_258) __extension__ ({ \
43101 float64x1_t __ret_258; \
4310143102 float64x1_t __s0_258 = __p0_258; \
4310243103 float64x1_t __s1_258 = __p1_258; \
4310343104 float64x2_t __s2_258 = __p2_258; \
4310443105 float64x2_t __rev2_258; __rev2_258 = __builtin_shufflevector(__s2_258, __s2_258, 1, 0); \
43105 float64x1_t __ret_258; \
4310643106float64x2_t __reint_258 = __rev2_258; \
4310743107uint64x2_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)}; \
4310843108 __ret_258 = vcmla_rot90_f64(__s0_258, __s1_258, *(float64x1_t *) &__reint1_258); \
......@@ -43112,10 +43112,10 @@ uint64x2_t __reint1_258 = (uint64x2_t) {__noswap_vgetq_lane_u64(*(uint64x2_t *)
4311243112
4311343113#ifdef __LITTLE_ENDIAN__
4311443114#define vcmlaq_rot90_laneq_f64(__p0_259, __p1_259, __p2_259, __p3_259) __extension__ ({ \
43115 float64x2_t __ret_259; \
4311543116 float64x2_t __s0_259 = __p0_259; \
4311643117 float64x2_t __s1_259 = __p1_259; \
4311743118 float64x2_t __s2_259 = __p2_259; \
43118 float64x2_t __ret_259; \
4311943119float64x2_t __reint_259 = __s2_259; \
4312043120uint64x2_t __reint1_259 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_259, __p3_259), vgetq_lane_u64(*(uint64x2_t *) &__reint_259, __p3_259)}; \
4312143121 __ret_259 = vcmlaq_rot90_f64(__s0_259, __s1_259, *(float64x2_t *) &__reint1_259); \
......@@ -43123,13 +43123,13 @@ uint64x2_t __reint1_259 = (uint64x2_t) {vgetq_lane_u64(*(uint64x2_t *) &__reint_
4312343123})
4312443124#else
4312543125#define vcmlaq_rot90_laneq_f64(__p0_260, __p1_260, __p2_260, __p3_260) __extension__ ({ \
43126 float64x2_t __ret_260; \
4312643127 float64x2_t __s0_260 = __p0_260; \
4312743128 float64x2_t __s1_260 = __p1_260; \
4312843129 float64x2_t __s2_260 = __p2_260; \
4312943130 float64x2_t __rev0_260; __rev0_260 = __builtin_shufflevector(__s0_260, __s0_260, 1, 0); \
4313043131 float64x2_t __rev1_260; __rev1_260 = __builtin_shufflevector(__s1_260, __s1_260, 1, 0); \
4313143132 float64x2_t __rev2_260; __rev2_260 = __builtin_shufflevector(__s2_260, __s2_260, 1, 0); \
43132 float64x2_t __ret_260; \
4313343133float64x2_t __reint_260 = __rev2_260; \
4313443134uint64x2_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)}; \
4313543135 __ret_260 = __noswap_vcmlaq_rot90_f64(__rev0_260, __rev1_260, *(float64x2_t *) &__reint1_260); \
......@@ -43148,10 +43148,10 @@ __ai uint32x4_t vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
4314843148}
4314943149#else
4315043150__ai uint32x4_t vdotq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
43151 uint32x4_t __ret;
4315143152 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4315243153 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4315343154 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43154 uint32x4_t __ret;
4315543155 __ret = (uint32x4_t) __builtin_neon_vdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
4315643156 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4315743157 return __ret;
......@@ -43171,10 +43171,10 @@ __ai int32x4_t vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
4317143171}
4317243172#else
4317343173__ai int32x4_t vdotq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
43174 int32x4_t __ret;
4317443175 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4317543176 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4317643177 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
43177 int32x4_t __ret;
4317843178 __ret = (int32x4_t) __builtin_neon_vdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
4317943179 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4318043180 return __ret;
......@@ -43194,10 +43194,10 @@ __ai uint32x2_t vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
4319443194}
4319543195#else
4319643196__ai uint32x2_t vdot_u32(uint32x2_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
43197 uint32x2_t __ret;
4319743198 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4319843199 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4319943200 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43200 uint32x2_t __ret;
4320143201 __ret = (uint32x2_t) __builtin_neon_vdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 18);
4320243202 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4320343203 return __ret;
......@@ -43217,10 +43217,10 @@ __ai int32x2_t vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
4321743217}
4321843218#else
4321943219__ai int32x2_t vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
43220 int32x2_t __ret;
4322043221 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4322143222 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4322243223 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43223 int32x2_t __ret;
4322443224 __ret = (int32x2_t) __builtin_neon_vdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
4322543225 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4322643226 return __ret;
......@@ -43234,10 +43234,10 @@ __ai int32x2_t __noswap_vdot_s32(int32x2_t __p0, int8x8_t __p1, int8x8_t __p2) {
4323443234
4323543235#ifdef __LITTLE_ENDIAN__
4323643236#define vdotq_lane_u32(__p0_261, __p1_261, __p2_261, __p3_261) __extension__ ({ \
43237 uint32x4_t __ret_261; \
4323743238 uint32x4_t __s0_261 = __p0_261; \
4323843239 uint8x16_t __s1_261 = __p1_261; \
4323943240 uint8x8_t __s2_261 = __p2_261; \
43240 uint32x4_t __ret_261; \
4324143241uint8x8_t __reint_261 = __s2_261; \
4324243242uint32x4_t __reint1_261 = splatq_lane_u32(*(uint32x2_t *) &__reint_261, __p3_261); \
4324343243 __ret_261 = vdotq_u32(__s0_261, __s1_261, *(uint8x16_t *) &__reint1_261); \
......@@ -43245,13 +43245,13 @@ uint32x4_t __reint1_261 = splatq_lane_u32(*(uint32x2_t *) &__reint_261, __p3_261
4324543245})
4324643246#else
4324743247#define vdotq_lane_u32(__p0_262, __p1_262, __p2_262, __p3_262) __extension__ ({ \
43248 uint32x4_t __ret_262; \
4324843249 uint32x4_t __s0_262 = __p0_262; \
4324943250 uint8x16_t __s1_262 = __p1_262; \
4325043251 uint8x8_t __s2_262 = __p2_262; \
4325143252 uint32x4_t __rev0_262; __rev0_262 = __builtin_shufflevector(__s0_262, __s0_262, 3, 2, 1, 0); \
4325243253 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); \
4325343254 uint8x8_t __rev2_262; __rev2_262 = __builtin_shufflevector(__s2_262, __s2_262, 7, 6, 5, 4, 3, 2, 1, 0); \
43254 uint32x4_t __ret_262; \
4325543255uint8x8_t __reint_262 = __rev2_262; \
4325643256uint32x4_t __reint1_262 = __noswap_splatq_lane_u32(*(uint32x2_t *) &__reint_262, __p3_262); \
4325743257 __ret_262 = __noswap_vdotq_u32(__rev0_262, __rev1_262, *(uint8x16_t *) &__reint1_262); \
......@@ -43262,10 +43262,10 @@ uint32x4_t __reint1_262 = __noswap_splatq_lane_u32(*(uint32x2_t *) &__reint_262,
4326243262
4326343263#ifdef __LITTLE_ENDIAN__
4326443264#define vdotq_lane_s32(__p0_263, __p1_263, __p2_263, __p3_263) __extension__ ({ \
43265 int32x4_t __ret_263; \
4326543266 int32x4_t __s0_263 = __p0_263; \
4326643267 int8x16_t __s1_263 = __p1_263; \
4326743268 int8x8_t __s2_263 = __p2_263; \
43268 int32x4_t __ret_263; \
4326943269int8x8_t __reint_263 = __s2_263; \
4327043270int32x4_t __reint1_263 = splatq_lane_s32(*(int32x2_t *) &__reint_263, __p3_263); \
4327143271 __ret_263 = vdotq_s32(__s0_263, __s1_263, *(int8x16_t *) &__reint1_263); \
......@@ -43273,13 +43273,13 @@ int32x4_t __reint1_263 = splatq_lane_s32(*(int32x2_t *) &__reint_263, __p3_263);
4327343273})
4327443274#else
4327543275#define vdotq_lane_s32(__p0_264, __p1_264, __p2_264, __p3_264) __extension__ ({ \
43276 int32x4_t __ret_264; \
4327643277 int32x4_t __s0_264 = __p0_264; \
4327743278 int8x16_t __s1_264 = __p1_264; \
4327843279 int8x8_t __s2_264 = __p2_264; \
4327943280 int32x4_t __rev0_264; __rev0_264 = __builtin_shufflevector(__s0_264, __s0_264, 3, 2, 1, 0); \
4328043281 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); \
4328143282 int8x8_t __rev2_264; __rev2_264 = __builtin_shufflevector(__s2_264, __s2_264, 7, 6, 5, 4, 3, 2, 1, 0); \
43282 int32x4_t __ret_264; \
4328343283int8x8_t __reint_264 = __rev2_264; \
4328443284int32x4_t __reint1_264 = __noswap_splatq_lane_s32(*(int32x2_t *) &__reint_264, __p3_264); \
4328543285 __ret_264 = __noswap_vdotq_s32(__rev0_264, __rev1_264, *(int8x16_t *) &__reint1_264); \
......@@ -43290,10 +43290,10 @@ int32x4_t __reint1_264 = __noswap_splatq_lane_s32(*(int32x2_t *) &__reint_264, _
4329043290
4329143291#ifdef __LITTLE_ENDIAN__
4329243292#define vdot_lane_u32(__p0_265, __p1_265, __p2_265, __p3_265) __extension__ ({ \
43293 uint32x2_t __ret_265; \
4329343294 uint32x2_t __s0_265 = __p0_265; \
4329443295 uint8x8_t __s1_265 = __p1_265; \
4329543296 uint8x8_t __s2_265 = __p2_265; \
43296 uint32x2_t __ret_265; \
4329743297uint8x8_t __reint_265 = __s2_265; \
4329843298uint32x2_t __reint1_265 = splat_lane_u32(*(uint32x2_t *) &__reint_265, __p3_265); \
4329943299 __ret_265 = vdot_u32(__s0_265, __s1_265, *(uint8x8_t *) &__reint1_265); \
......@@ -43301,13 +43301,13 @@ uint32x2_t __reint1_265 = splat_lane_u32(*(uint32x2_t *) &__reint_265, __p3_265)
4330143301})
4330243302#else
4330343303#define vdot_lane_u32(__p0_266, __p1_266, __p2_266, __p3_266) __extension__ ({ \
43304 uint32x2_t __ret_266; \
4330443305 uint32x2_t __s0_266 = __p0_266; \
4330543306 uint8x8_t __s1_266 = __p1_266; \
4330643307 uint8x8_t __s2_266 = __p2_266; \
4330743308 uint32x2_t __rev0_266; __rev0_266 = __builtin_shufflevector(__s0_266, __s0_266, 1, 0); \
4330843309 uint8x8_t __rev1_266; __rev1_266 = __builtin_shufflevector(__s1_266, __s1_266, 7, 6, 5, 4, 3, 2, 1, 0); \
4330943310 uint8x8_t __rev2_266; __rev2_266 = __builtin_shufflevector(__s2_266, __s2_266, 7, 6, 5, 4, 3, 2, 1, 0); \
43310 uint32x2_t __ret_266; \
4331143311uint8x8_t __reint_266 = __rev2_266; \
4331243312uint32x2_t __reint1_266 = __noswap_splat_lane_u32(*(uint32x2_t *) &__reint_266, __p3_266); \
4331343313 __ret_266 = __noswap_vdot_u32(__rev0_266, __rev1_266, *(uint8x8_t *) &__reint1_266); \
......@@ -43318,10 +43318,10 @@ uint32x2_t __reint1_266 = __noswap_splat_lane_u32(*(uint32x2_t *) &__reint_266,
4331843318
4331943319#ifdef __LITTLE_ENDIAN__
4332043320#define vdot_lane_s32(__p0_267, __p1_267, __p2_267, __p3_267) __extension__ ({ \
43321 int32x2_t __ret_267; \
4332143322 int32x2_t __s0_267 = __p0_267; \
4332243323 int8x8_t __s1_267 = __p1_267; \
4332343324 int8x8_t __s2_267 = __p2_267; \
43324 int32x2_t __ret_267; \
4332543325int8x8_t __reint_267 = __s2_267; \
4332643326int32x2_t __reint1_267 = splat_lane_s32(*(int32x2_t *) &__reint_267, __p3_267); \
4332743327 __ret_267 = vdot_s32(__s0_267, __s1_267, *(int8x8_t *) &__reint1_267); \
......@@ -43329,13 +43329,13 @@ int32x2_t __reint1_267 = splat_lane_s32(*(int32x2_t *) &__reint_267, __p3_267);
4332943329})
4333043330#else
4333143331#define vdot_lane_s32(__p0_268, __p1_268, __p2_268, __p3_268) __extension__ ({ \
43332 int32x2_t __ret_268; \
4333243333 int32x2_t __s0_268 = __p0_268; \
4333343334 int8x8_t __s1_268 = __p1_268; \
4333443335 int8x8_t __s2_268 = __p2_268; \
4333543336 int32x2_t __rev0_268; __rev0_268 = __builtin_shufflevector(__s0_268, __s0_268, 1, 0); \
4333643337 int8x8_t __rev1_268; __rev1_268 = __builtin_shufflevector(__s1_268, __s1_268, 7, 6, 5, 4, 3, 2, 1, 0); \
4333743338 int8x8_t __rev2_268; __rev2_268 = __builtin_shufflevector(__s2_268, __s2_268, 7, 6, 5, 4, 3, 2, 1, 0); \
43338 int32x2_t __ret_268; \
4333943339int8x8_t __reint_268 = __rev2_268; \
4334043340int32x2_t __reint1_268 = __noswap_splat_lane_s32(*(int32x2_t *) &__reint_268, __p3_268); \
4334143341 __ret_268 = __noswap_vdot_s32(__rev0_268, __rev1_268, *(int8x8_t *) &__reint1_268); \
......@@ -43348,10 +43348,10 @@ int32x2_t __reint1_268 = __noswap_splat_lane_s32(*(int32x2_t *) &__reint_268, __
4334843348#if defined(__ARM_FEATURE_DOTPROD) && defined(__aarch64__)
4334943349#ifdef __LITTLE_ENDIAN__
4335043350#define vdotq_laneq_u32(__p0_269, __p1_269, __p2_269, __p3_269) __extension__ ({ \
43351 uint32x4_t __ret_269; \
4335143352 uint32x4_t __s0_269 = __p0_269; \
4335243353 uint8x16_t __s1_269 = __p1_269; \
4335343354 uint8x16_t __s2_269 = __p2_269; \
43354 uint32x4_t __ret_269; \
4335543355uint8x16_t __reint_269 = __s2_269; \
4335643356uint32x4_t __reint1_269 = splatq_laneq_u32(*(uint32x4_t *) &__reint_269, __p3_269); \
4335743357 __ret_269 = vdotq_u32(__s0_269, __s1_269, *(uint8x16_t *) &__reint1_269); \
......@@ -43359,13 +43359,13 @@ uint32x4_t __reint1_269 = splatq_laneq_u32(*(uint32x4_t *) &__reint_269, __p3_26
4335943359})
4336043360#else
4336143361#define vdotq_laneq_u32(__p0_270, __p1_270, __p2_270, __p3_270) __extension__ ({ \
43362 uint32x4_t __ret_270; \
4336243363 uint32x4_t __s0_270 = __p0_270; \
4336343364 uint8x16_t __s1_270 = __p1_270; \
4336443365 uint8x16_t __s2_270 = __p2_270; \
4336543366 uint32x4_t __rev0_270; __rev0_270 = __builtin_shufflevector(__s0_270, __s0_270, 3, 2, 1, 0); \
4336643367 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); \
4336743368 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); \
43368 uint32x4_t __ret_270; \
4336943369uint8x16_t __reint_270 = __rev2_270; \
4337043370uint32x4_t __reint1_270 = __noswap_splatq_laneq_u32(*(uint32x4_t *) &__reint_270, __p3_270); \
4337143371 __ret_270 = __noswap_vdotq_u32(__rev0_270, __rev1_270, *(uint8x16_t *) &__reint1_270); \
......@@ -43376,10 +43376,10 @@ uint32x4_t __reint1_270 = __noswap_splatq_laneq_u32(*(uint32x4_t *) &__reint_270
4337643376
4337743377#ifdef __LITTLE_ENDIAN__
4337843378#define vdotq_laneq_s32(__p0_271, __p1_271, __p2_271, __p3_271) __extension__ ({ \
43379 int32x4_t __ret_271; \
4337943380 int32x4_t __s0_271 = __p0_271; \
4338043381 int8x16_t __s1_271 = __p1_271; \
4338143382 int8x16_t __s2_271 = __p2_271; \
43382 int32x4_t __ret_271; \
4338343383int8x16_t __reint_271 = __s2_271; \
4338443384int32x4_t __reint1_271 = splatq_laneq_s32(*(int32x4_t *) &__reint_271, __p3_271); \
4338543385 __ret_271 = vdotq_s32(__s0_271, __s1_271, *(int8x16_t *) &__reint1_271); \
......@@ -43387,13 +43387,13 @@ int32x4_t __reint1_271 = splatq_laneq_s32(*(int32x4_t *) &__reint_271, __p3_271)
4338743387})
4338843388#else
4338943389#define vdotq_laneq_s32(__p0_272, __p1_272, __p2_272, __p3_272) __extension__ ({ \
43390 int32x4_t __ret_272; \
4339043391 int32x4_t __s0_272 = __p0_272; \
4339143392 int8x16_t __s1_272 = __p1_272; \
4339243393 int8x16_t __s2_272 = __p2_272; \
4339343394 int32x4_t __rev0_272; __rev0_272 = __builtin_shufflevector(__s0_272, __s0_272, 3, 2, 1, 0); \
4339443395 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); \
4339543396 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); \
43396 int32x4_t __ret_272; \
4339743397int8x16_t __reint_272 = __rev2_272; \
4339843398int32x4_t __reint1_272 = __noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_272, __p3_272); \
4339943399 __ret_272 = __noswap_vdotq_s32(__rev0_272, __rev1_272, *(int8x16_t *) &__reint1_272); \
......@@ -43404,10 +43404,10 @@ int32x4_t __reint1_272 = __noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_272,
4340443404
4340543405#ifdef __LITTLE_ENDIAN__
4340643406#define vdot_laneq_u32(__p0_273, __p1_273, __p2_273, __p3_273) __extension__ ({ \
43407 uint32x2_t __ret_273; \
4340743408 uint32x2_t __s0_273 = __p0_273; \
4340843409 uint8x8_t __s1_273 = __p1_273; \
4340943410 uint8x16_t __s2_273 = __p2_273; \
43410 uint32x2_t __ret_273; \
4341143411uint8x16_t __reint_273 = __s2_273; \
4341243412uint32x2_t __reint1_273 = splat_laneq_u32(*(uint32x4_t *) &__reint_273, __p3_273); \
4341343413 __ret_273 = vdot_u32(__s0_273, __s1_273, *(uint8x8_t *) &__reint1_273); \
......@@ -43415,13 +43415,13 @@ uint32x2_t __reint1_273 = splat_laneq_u32(*(uint32x4_t *) &__reint_273, __p3_273
4341543415})
4341643416#else
4341743417#define vdot_laneq_u32(__p0_274, __p1_274, __p2_274, __p3_274) __extension__ ({ \
43418 uint32x2_t __ret_274; \
4341843419 uint32x2_t __s0_274 = __p0_274; \
4341943420 uint8x8_t __s1_274 = __p1_274; \
4342043421 uint8x16_t __s2_274 = __p2_274; \
4342143422 uint32x2_t __rev0_274; __rev0_274 = __builtin_shufflevector(__s0_274, __s0_274, 1, 0); \
4342243423 uint8x8_t __rev1_274; __rev1_274 = __builtin_shufflevector(__s1_274, __s1_274, 7, 6, 5, 4, 3, 2, 1, 0); \
4342343424 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); \
43424 uint32x2_t __ret_274; \
4342543425uint8x16_t __reint_274 = __rev2_274; \
4342643426uint32x2_t __reint1_274 = __noswap_splat_laneq_u32(*(uint32x4_t *) &__reint_274, __p3_274); \
4342743427 __ret_274 = __noswap_vdot_u32(__rev0_274, __rev1_274, *(uint8x8_t *) &__reint1_274); \
......@@ -43432,10 +43432,10 @@ uint32x2_t __reint1_274 = __noswap_splat_laneq_u32(*(uint32x4_t *) &__reint_274,
4343243432
4343343433#ifdef __LITTLE_ENDIAN__
4343443434#define vdot_laneq_s32(__p0_275, __p1_275, __p2_275, __p3_275) __extension__ ({ \
43435 int32x2_t __ret_275; \
4343543436 int32x2_t __s0_275 = __p0_275; \
4343643437 int8x8_t __s1_275 = __p1_275; \
4343743438 int8x16_t __s2_275 = __p2_275; \
43438 int32x2_t __ret_275; \
4343943439int8x16_t __reint_275 = __s2_275; \
4344043440int32x2_t __reint1_275 = splat_laneq_s32(*(int32x4_t *) &__reint_275, __p3_275); \
4344143441 __ret_275 = vdot_s32(__s0_275, __s1_275, *(int8x8_t *) &__reint1_275); \
......@@ -43443,13 +43443,13 @@ int32x2_t __reint1_275 = splat_laneq_s32(*(int32x4_t *) &__reint_275, __p3_275);
4344343443})
4344443444#else
4344543445#define vdot_laneq_s32(__p0_276, __p1_276, __p2_276, __p3_276) __extension__ ({ \
43446 int32x2_t __ret_276; \
4344643447 int32x2_t __s0_276 = __p0_276; \
4344743448 int8x8_t __s1_276 = __p1_276; \
4344843449 int8x16_t __s2_276 = __p2_276; \
4344943450 int32x2_t __rev0_276; __rev0_276 = __builtin_shufflevector(__s0_276, __s0_276, 1, 0); \
4345043451 int8x8_t __rev1_276; __rev1_276 = __builtin_shufflevector(__s1_276, __s1_276, 7, 6, 5, 4, 3, 2, 1, 0); \
4345143452 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); \
43452 int32x2_t __ret_276; \
4345343453int8x16_t __reint_276 = __rev2_276; \
4345443454int32x2_t __reint1_276 = __noswap_splat_laneq_s32(*(int32x4_t *) &__reint_276, __p3_276); \
4345543455 __ret_276 = __noswap_vdot_s32(__rev0_276, __rev1_276, *(int8x8_t *) &__reint1_276); \
......@@ -43468,10 +43468,10 @@ __ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2)
4346843468}
4346943469#else
4347043470__ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43471 float32x4_t __ret;
4347143472 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4347243473 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4347343474 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43474 float32x4_t __ret;
4347543475 __ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4347643476 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4347743477 return __ret;
......@@ -43491,10 +43491,10 @@ __ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2)
4349143491}
4349243492#else
4349343493__ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43494 float32x2_t __ret;
4349443495 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4349543496 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4349643497 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
43497 float32x2_t __ret;
4349843498 __ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4349943499 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4350043500 return __ret;
......@@ -43514,9 +43514,9 @@ __ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2)
4351443514}
4351543515#else
4351643516__ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
43517 float32x4_t __ret;
4351743518 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4351843519 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43519 float32x4_t __ret;
4352043520 __ret = __noswap_vfmaq_f32(__rev0, __rev1, (float32x4_t) {__p2, __p2, __p2, __p2});
4352143521 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4352243522 return __ret;
......@@ -43531,9 +43531,9 @@ __ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2)
4353143531}
4353243532#else
4353343533__ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
43534 float32x2_t __ret;
4353443535 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4353543536 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43536 float32x2_t __ret;
4353743537 __ret = __noswap_vfma_f32(__rev0, __rev1, (float32x2_t) {__p2, __p2});
4353843538 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4353943539 return __ret;
......@@ -43548,10 +43548,10 @@ __ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2)
4354843548}
4354943549#else
4355043550__ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
43551 float32x4_t __ret;
4355143552 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4355243553 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4355343554 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43554 float32x4_t __ret;
4355543555 __ret = __noswap_vfmaq_f32(__rev0, -__rev1, __rev2);
4355643556 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4355743557 return __ret;
......@@ -43566,10 +43566,10 @@ __ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2)
4356643566}
4356743567#else
4356843568__ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
43569 float32x2_t __ret;
4356943570 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4357043571 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4357143572 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
43572 float32x2_t __ret;
4357343573 __ret = __noswap_vfma_f32(__rev0, -__rev1, __rev2);
4357443574 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4357543575 return __ret;
......@@ -43586,10 +43586,10 @@ __ai float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_
4358643586}
4358743587#else
4358843588__ai float32x4_t vfmlalq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43589 float32x4_t __ret;
4358943590 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4359043591 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4359143592 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43592 float32x4_t __ret;
4359343593 __ret = (float32x4_t) __builtin_neon_vfmlalq_high_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4359443594 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4359543595 return __ret;
......@@ -43609,10 +43609,10 @@ __ai float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t
4360943609}
4361043610#else
4361143611__ai float32x2_t vfmlal_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43612 float32x2_t __ret;
4361243613 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4361343614 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4361443615 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43615 float32x2_t __ret;
4361643616 __ret = (float32x2_t) __builtin_neon_vfmlal_high_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4361743617 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4361843618 return __ret;
......@@ -43632,10 +43632,10 @@ __ai float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t
4363243632}
4363343633#else
4363443634__ai float32x4_t vfmlalq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43635 float32x4_t __ret;
4363543636 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4363643637 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4363743638 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43638 float32x4_t __ret;
4363943639 __ret = (float32x4_t) __builtin_neon_vfmlalq_low_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4364043640 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4364143641 return __ret;
......@@ -43655,10 +43655,10 @@ __ai float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t
4365543655}
4365643656#else
4365743657__ai float32x2_t vfmlal_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43658 float32x2_t __ret;
4365843659 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4365943660 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4366043661 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43661 float32x2_t __ret;
4366243662 __ret = (float32x2_t) __builtin_neon_vfmlal_low_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4366343663 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4366443664 return __ret;
......@@ -43678,10 +43678,10 @@ __ai float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_
4367843678}
4367943679#else
4368043680__ai float32x4_t vfmlslq_high_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43681 float32x4_t __ret;
4368143682 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4368243683 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4368343684 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43684 float32x4_t __ret;
4368543685 __ret = (float32x4_t) __builtin_neon_vfmlslq_high_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4368643686 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4368743687 return __ret;
......@@ -43701,10 +43701,10 @@ __ai float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t
4370143701}
4370243702#else
4370343703__ai float32x2_t vfmlsl_high_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43704 float32x2_t __ret;
4370443705 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4370543706 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4370643707 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43707 float32x2_t __ret;
4370843708 __ret = (float32x2_t) __builtin_neon_vfmlsl_high_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4370943709 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4371043710 return __ret;
......@@ -43724,10 +43724,10 @@ __ai float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t
4372443724}
4372543725#else
4372643726__ai float32x4_t vfmlslq_low_f16(float32x4_t __p0, float16x8_t __p1, float16x8_t __p2) {
43727 float32x4_t __ret;
4372743728 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4372843729 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4372943730 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43730 float32x4_t __ret;
4373143731 __ret = (float32x4_t) __builtin_neon_vfmlslq_low_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 41);
4373243732 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4373343733 return __ret;
......@@ -43747,10 +43747,10 @@ __ai float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t
4374743747}
4374843748#else
4374943749__ai float32x2_t vfmlsl_low_f16(float32x2_t __p0, float16x4_t __p1, float16x4_t __p2) {
43750 float32x2_t __ret;
4375043751 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4375143752 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4375243753 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43753 float32x2_t __ret;
4375443754 __ret = (float32x2_t) __builtin_neon_vfmlsl_low_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 9);
4375543755 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4375643756 return __ret;
......@@ -43772,9 +43772,9 @@ __ai float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
4377243772}
4377343773#else
4377443774__ai float16x8_t vabdq_f16(float16x8_t __p0, float16x8_t __p1) {
43775 float16x8_t __ret;
4377543776 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4377643777 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43777 float16x8_t __ret;
4377843778 __ret = (float16x8_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4377943779 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4378043780 return __ret;
......@@ -43789,9 +43789,9 @@ __ai float16x4_t vabd_f16(float16x4_t __p0, float16x4_t __p1) {
4378943789}
4379043790#else
4379143791__ai float16x4_t vabd_f16(float16x4_t __p0, float16x4_t __p1) {
43792 float16x4_t __ret;
4379243793 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4379343794 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43794 float16x4_t __ret;
4379543795 __ret = (float16x4_t) __builtin_neon_vabd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4379643796 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4379743797 return __ret;
......@@ -43806,8 +43806,8 @@ __ai float16x8_t vabsq_f16(float16x8_t __p0) {
4380643806}
4380743807#else
4380843808__ai float16x8_t vabsq_f16(float16x8_t __p0) {
43809 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4381043809 float16x8_t __ret;
43810 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4381143811 __ret = (float16x8_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 40);
4381243812 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4381343813 return __ret;
......@@ -43822,8 +43822,8 @@ __ai float16x4_t vabs_f16(float16x4_t __p0) {
4382243822}
4382343823#else
4382443824__ai float16x4_t vabs_f16(float16x4_t __p0) {
43825 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4382643825 float16x4_t __ret;
43826 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4382743827 __ret = (float16x4_t) __builtin_neon_vabs_v((int8x8_t)__rev0, 8);
4382843828 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4382943829 return __ret;
......@@ -43838,9 +43838,9 @@ __ai float16x8_t vaddq_f16(float16x8_t __p0, float16x8_t __p1) {
4383843838}
4383943839#else
4384043840__ai float16x8_t vaddq_f16(float16x8_t __p0, float16x8_t __p1) {
43841 float16x8_t __ret;
4384143842 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4384243843 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43843 float16x8_t __ret;
4384443844 __ret = __rev0 + __rev1;
4384543845 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4384643846 return __ret;
......@@ -43855,9 +43855,9 @@ __ai float16x4_t vadd_f16(float16x4_t __p0, float16x4_t __p1) {
4385543855}
4385643856#else
4385743857__ai float16x4_t vadd_f16(float16x4_t __p0, float16x4_t __p1) {
43858 float16x4_t __ret;
4385843859 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4385943860 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43860 float16x4_t __ret;
4386143861 __ret = __rev0 + __rev1;
4386243862 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4386343863 return __ret;
......@@ -43872,10 +43872,10 @@ __ai float16x8_t vbslq_f16(uint16x8_t __p0, float16x8_t __p1, float16x8_t __p2)
4387243872}
4387343873#else
4387443874__ai float16x8_t vbslq_f16(uint16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
43875 float16x8_t __ret;
4387543876 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4387643877 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4387743878 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
43878 float16x8_t __ret;
4387943879 __ret = (float16x8_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
4388043880 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4388143881 return __ret;
......@@ -43890,10 +43890,10 @@ __ai float16x4_t vbsl_f16(uint16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
4389043890}
4389143891#else
4389243892__ai float16x4_t vbsl_f16(uint16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
43893 float16x4_t __ret;
4389343894 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4389443895 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4389543896 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
43896 float16x4_t __ret;
4389743897 __ret = (float16x4_t) __builtin_neon_vbsl_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
4389843898 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4389943899 return __ret;
......@@ -43908,9 +43908,9 @@ __ai uint16x8_t vcageq_f16(float16x8_t __p0, float16x8_t __p1) {
4390843908}
4390943909#else
4391043910__ai uint16x8_t vcageq_f16(float16x8_t __p0, float16x8_t __p1) {
43911 uint16x8_t __ret;
4391143912 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4391243913 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43913 uint16x8_t __ret;
4391443914 __ret = (uint16x8_t) __builtin_neon_vcageq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
4391543915 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4391643916 return __ret;
......@@ -43925,9 +43925,9 @@ __ai uint16x4_t vcage_f16(float16x4_t __p0, float16x4_t __p1) {
4392543925}
4392643926#else
4392743927__ai uint16x4_t vcage_f16(float16x4_t __p0, float16x4_t __p1) {
43928 uint16x4_t __ret;
4392843929 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4392943930 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43930 uint16x4_t __ret;
4393143931 __ret = (uint16x4_t) __builtin_neon_vcage_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
4393243932 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4393343933 return __ret;
......@@ -43942,9 +43942,9 @@ __ai uint16x8_t vcagtq_f16(float16x8_t __p0, float16x8_t __p1) {
4394243942}
4394343943#else
4394443944__ai uint16x8_t vcagtq_f16(float16x8_t __p0, float16x8_t __p1) {
43945 uint16x8_t __ret;
4394543946 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4394643947 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43947 uint16x8_t __ret;
4394843948 __ret = (uint16x8_t) __builtin_neon_vcagtq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
4394943949 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4395043950 return __ret;
......@@ -43959,9 +43959,9 @@ __ai uint16x4_t vcagt_f16(float16x4_t __p0, float16x4_t __p1) {
4395943959}
4396043960#else
4396143961__ai uint16x4_t vcagt_f16(float16x4_t __p0, float16x4_t __p1) {
43962 uint16x4_t __ret;
4396243963 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4396343964 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43964 uint16x4_t __ret;
4396543965 __ret = (uint16x4_t) __builtin_neon_vcagt_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
4396643966 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4396743967 return __ret;
......@@ -43976,9 +43976,9 @@ __ai uint16x8_t vcaleq_f16(float16x8_t __p0, float16x8_t __p1) {
4397643976}
4397743977#else
4397843978__ai uint16x8_t vcaleq_f16(float16x8_t __p0, float16x8_t __p1) {
43979 uint16x8_t __ret;
4397943980 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4398043981 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
43981 uint16x8_t __ret;
4398243982 __ret = (uint16x8_t) __builtin_neon_vcaleq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
4398343983 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4398443984 return __ret;
......@@ -43993,9 +43993,9 @@ __ai uint16x4_t vcale_f16(float16x4_t __p0, float16x4_t __p1) {
4399343993}
4399443994#else
4399543995__ai uint16x4_t vcale_f16(float16x4_t __p0, float16x4_t __p1) {
43996 uint16x4_t __ret;
4399643997 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4399743998 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
43998 uint16x4_t __ret;
4399943999 __ret = (uint16x4_t) __builtin_neon_vcale_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
4400044000 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4400144001 return __ret;
......@@ -44010,9 +44010,9 @@ __ai uint16x8_t vcaltq_f16(float16x8_t __p0, float16x8_t __p1) {
4401044010}
4401144011#else
4401244012__ai uint16x8_t vcaltq_f16(float16x8_t __p0, float16x8_t __p1) {
44013 uint16x8_t __ret;
4401344014 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4401444015 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44015 uint16x8_t __ret;
4401644016 __ret = (uint16x8_t) __builtin_neon_vcaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
4401744017 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4401844018 return __ret;
......@@ -44027,9 +44027,9 @@ __ai uint16x4_t vcalt_f16(float16x4_t __p0, float16x4_t __p1) {
4402744027}
4402844028#else
4402944029__ai uint16x4_t vcalt_f16(float16x4_t __p0, float16x4_t __p1) {
44030 uint16x4_t __ret;
4403044031 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4403144032 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44032 uint16x4_t __ret;
4403344033 __ret = (uint16x4_t) __builtin_neon_vcalt_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
4403444034 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4403544035 return __ret;
......@@ -44044,9 +44044,9 @@ __ai uint16x8_t vceqq_f16(float16x8_t __p0, float16x8_t __p1) {
4404444044}
4404544045#else
4404644046__ai uint16x8_t vceqq_f16(float16x8_t __p0, float16x8_t __p1) {
44047 uint16x8_t __ret;
4404744048 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4404844049 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44049 uint16x8_t __ret;
4405044050 __ret = (uint16x8_t)(__rev0 == __rev1);
4405144051 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4405244052 return __ret;
......@@ -44061,9 +44061,9 @@ __ai uint16x4_t vceq_f16(float16x4_t __p0, float16x4_t __p1) {
4406144061}
4406244062#else
4406344063__ai uint16x4_t vceq_f16(float16x4_t __p0, float16x4_t __p1) {
44064 uint16x4_t __ret;
4406444065 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4406544066 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44066 uint16x4_t __ret;
4406744067 __ret = (uint16x4_t)(__rev0 == __rev1);
4406844068 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4406944069 return __ret;
......@@ -44078,8 +44078,8 @@ __ai uint16x8_t vceqzq_f16(float16x8_t __p0) {
4407844078}
4407944079#else
4408044080__ai uint16x8_t vceqzq_f16(float16x8_t __p0) {
44081 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4408244081 uint16x8_t __ret;
44082 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4408344083 __ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
4408444084 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4408544085 return __ret;
......@@ -44094,8 +44094,8 @@ __ai uint16x4_t vceqz_f16(float16x4_t __p0) {
4409444094}
4409544095#else
4409644096__ai uint16x4_t vceqz_f16(float16x4_t __p0) {
44097 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4409844097 uint16x4_t __ret;
44098 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4409944099 __ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
4410044100 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4410144101 return __ret;
......@@ -44110,9 +44110,9 @@ __ai uint16x8_t vcgeq_f16(float16x8_t __p0, float16x8_t __p1) {
4411044110}
4411144111#else
4411244112__ai uint16x8_t vcgeq_f16(float16x8_t __p0, float16x8_t __p1) {
44113 uint16x8_t __ret;
4411344114 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4411444115 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44115 uint16x8_t __ret;
4411644116 __ret = (uint16x8_t)(__rev0 >= __rev1);
4411744117 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4411844118 return __ret;
......@@ -44127,9 +44127,9 @@ __ai uint16x4_t vcge_f16(float16x4_t __p0, float16x4_t __p1) {
4412744127}
4412844128#else
4412944129__ai uint16x4_t vcge_f16(float16x4_t __p0, float16x4_t __p1) {
44130 uint16x4_t __ret;
4413044131 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4413144132 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44132 uint16x4_t __ret;
4413344133 __ret = (uint16x4_t)(__rev0 >= __rev1);
4413444134 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4413544135 return __ret;
......@@ -44144,8 +44144,8 @@ __ai uint16x8_t vcgezq_f16(float16x8_t __p0) {
4414444144}
4414544145#else
4414644146__ai uint16x8_t vcgezq_f16(float16x8_t __p0) {
44147 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4414844147 uint16x8_t __ret;
44148 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4414944149 __ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 49);
4415044150 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4415144151 return __ret;
......@@ -44160,8 +44160,8 @@ __ai uint16x4_t vcgez_f16(float16x4_t __p0) {
4416044160}
4416144161#else
4416244162__ai uint16x4_t vcgez_f16(float16x4_t __p0) {
44163 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4416444163 uint16x4_t __ret;
44164 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4416544165 __ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 17);
4416644166 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4416744167 return __ret;
......@@ -44176,9 +44176,9 @@ __ai uint16x8_t vcgtq_f16(float16x8_t __p0, float16x8_t __p1) {
4417644176}
4417744177#else
4417844178__ai uint16x8_t vcgtq_f16(float16x8_t __p0, float16x8_t __p1) {
44179 uint16x8_t __ret;
4417944180 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4418044181 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44181 uint16x8_t __ret;
4418244182 __ret = (uint16x8_t)(__rev0 > __rev1);
4418344183 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4418444184 return __ret;
......@@ -44193,9 +44193,9 @@ __ai uint16x4_t vcgt_f16(float16x4_t __p0, float16x4_t __p1) {
4419344193}
4419444194#else
4419544195__ai uint16x4_t vcgt_f16(float16x4_t __p0, float16x4_t __p1) {
44196 uint16x4_t __ret;
4419644197 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4419744198 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44198 uint16x4_t __ret;
4419944199 __ret = (uint16x4_t)(__rev0 > __rev1);
4420044200 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4420144201 return __ret;
......@@ -44210,8 +44210,8 @@ __ai uint16x8_t vcgtzq_f16(float16x8_t __p0) {
4421044210}
4421144211#else
4421244212__ai uint16x8_t vcgtzq_f16(float16x8_t __p0) {
44213 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4421444213 uint16x8_t __ret;
44214 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4421544215 __ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 49);
4421644216 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4421744217 return __ret;
......@@ -44226,8 +44226,8 @@ __ai uint16x4_t vcgtz_f16(float16x4_t __p0) {
4422644226}
4422744227#else
4422844228__ai uint16x4_t vcgtz_f16(float16x4_t __p0) {
44229 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4423044229 uint16x4_t __ret;
44230 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4423144231 __ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 17);
4423244232 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4423344233 return __ret;
......@@ -44242,9 +44242,9 @@ __ai uint16x8_t vcleq_f16(float16x8_t __p0, float16x8_t __p1) {
4424244242}
4424344243#else
4424444244__ai uint16x8_t vcleq_f16(float16x8_t __p0, float16x8_t __p1) {
44245 uint16x8_t __ret;
4424544246 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4424644247 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44247 uint16x8_t __ret;
4424844248 __ret = (uint16x8_t)(__rev0 <= __rev1);
4424944249 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4425044250 return __ret;
......@@ -44259,9 +44259,9 @@ __ai uint16x4_t vcle_f16(float16x4_t __p0, float16x4_t __p1) {
4425944259}
4426044260#else
4426144261__ai uint16x4_t vcle_f16(float16x4_t __p0, float16x4_t __p1) {
44262 uint16x4_t __ret;
4426244263 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4426344264 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44264 uint16x4_t __ret;
4426544265 __ret = (uint16x4_t)(__rev0 <= __rev1);
4426644266 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4426744267 return __ret;
......@@ -44276,8 +44276,8 @@ __ai uint16x8_t vclezq_f16(float16x8_t __p0) {
4427644276}
4427744277#else
4427844278__ai uint16x8_t vclezq_f16(float16x8_t __p0) {
44279 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4428044279 uint16x8_t __ret;
44280 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4428144281 __ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 49);
4428244282 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4428344283 return __ret;
......@@ -44292,8 +44292,8 @@ __ai uint16x4_t vclez_f16(float16x4_t __p0) {
4429244292}
4429344293#else
4429444294__ai uint16x4_t vclez_f16(float16x4_t __p0) {
44295 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4429644295 uint16x4_t __ret;
44296 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4429744297 __ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 17);
4429844298 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4429944299 return __ret;
......@@ -44308,9 +44308,9 @@ __ai uint16x8_t vcltq_f16(float16x8_t __p0, float16x8_t __p1) {
4430844308}
4430944309#else
4431044310__ai uint16x8_t vcltq_f16(float16x8_t __p0, float16x8_t __p1) {
44311 uint16x8_t __ret;
4431144312 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4431244313 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
44313 uint16x8_t __ret;
4431444314 __ret = (uint16x8_t)(__rev0 < __rev1);
4431544315 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4431644316 return __ret;
......@@ -44325,9 +44325,9 @@ __ai uint16x4_t vclt_f16(float16x4_t __p0, float16x4_t __p1) {
4432544325}
4432644326#else
4432744327__ai uint16x4_t vclt_f16(float16x4_t __p0, float16x4_t __p1) {
44328 uint16x4_t __ret;
4432844329 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4432944330 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44330 uint16x4_t __ret;
4433144331 __ret = (uint16x4_t)(__rev0 < __rev1);
4433244332 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4433344333 return __ret;
......@@ -44342,8 +44342,8 @@ __ai uint16x8_t vcltzq_f16(float16x8_t __p0) {
4434244342}
4434344343#else
4434444344__ai uint16x8_t vcltzq_f16(float16x8_t __p0) {
44345 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4434644345 uint16x8_t __ret;
44346 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4434744347 __ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 49);
4434844348 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4434944349 return __ret;
......@@ -44358,8 +44358,8 @@ __ai uint16x4_t vcltz_f16(float16x4_t __p0) {
4435844358}
4435944359#else
4436044360__ai uint16x4_t vcltz_f16(float16x4_t __p0) {
44361 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4436244361 uint16x4_t __ret;
44362 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4436344363 __ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 17);
4436444364 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4436544365 return __ret;
......@@ -44374,8 +44374,8 @@ __ai float16x8_t vcvtq_f16_u16(uint16x8_t __p0) {
4437444374}
4437544375#else
4437644376__ai float16x8_t vcvtq_f16_u16(uint16x8_t __p0) {
44377 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4437844377 float16x8_t __ret;
44378 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4437944379 __ret = (float16x8_t) __builtin_neon_vcvtq_f16_v((int8x16_t)__rev0, 49);
4438044380 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4438144381 return __ret;
......@@ -44390,8 +44390,8 @@ __ai float16x8_t vcvtq_f16_s16(int16x8_t __p0) {
4439044390}
4439144391#else
4439244392__ai float16x8_t vcvtq_f16_s16(int16x8_t __p0) {
44393 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4439444393 float16x8_t __ret;
44394 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4439544395 __ret = (float16x8_t) __builtin_neon_vcvtq_f16_v((int8x16_t)__rev0, 33);
4439644396 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4439744397 return __ret;
......@@ -44406,8 +44406,8 @@ __ai float16x4_t vcvt_f16_u16(uint16x4_t __p0) {
4440644406}
4440744407#else
4440844408__ai float16x4_t vcvt_f16_u16(uint16x4_t __p0) {
44409 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4441044409 float16x4_t __ret;
44410 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4441144411 __ret = (float16x4_t) __builtin_neon_vcvt_f16_v((int8x8_t)__rev0, 17);
4441244412 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4441344413 return __ret;
......@@ -44422,8 +44422,8 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4442244422}
4442344423#else
4442444424__ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
44425 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4442644425 float16x4_t __ret;
44426 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4442744427 __ret = (float16x4_t) __builtin_neon_vcvt_f16_v((int8x8_t)__rev0, 1);
4442844428 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4442944429 return __ret;
......@@ -44432,16 +44432,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4443244432
4443344433#ifdef __LITTLE_ENDIAN__
4443444434#define vcvtq_n_f16_u16(__p0, __p1) __extension__ ({ \
44435 uint16x8_t __s0 = __p0; \
4443644435 float16x8_t __ret; \
44436 uint16x8_t __s0 = __p0; \
4443744437 __ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__s0, __p1, 49); \
4443844438 __ret; \
4443944439})
4444044440#else
4444144441#define vcvtq_n_f16_u16(__p0, __p1) __extension__ ({ \
44442 float16x8_t __ret; \
4444244443 uint16x8_t __s0 = __p0; \
4444344444 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44444 float16x8_t __ret; \
4444544445 __ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__rev0, __p1, 49); \
4444644446 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4444744447 __ret; \
......@@ -44450,16 +44450,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4445044450
4445144451#ifdef __LITTLE_ENDIAN__
4445244452#define vcvtq_n_f16_s16(__p0, __p1) __extension__ ({ \
44453 int16x8_t __s0 = __p0; \
4445444453 float16x8_t __ret; \
44454 int16x8_t __s0 = __p0; \
4445544455 __ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__s0, __p1, 33); \
4445644456 __ret; \
4445744457})
4445844458#else
4445944459#define vcvtq_n_f16_s16(__p0, __p1) __extension__ ({ \
44460 float16x8_t __ret; \
4446044461 int16x8_t __s0 = __p0; \
4446144462 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44462 float16x8_t __ret; \
4446344463 __ret = (float16x8_t) __builtin_neon_vcvtq_n_f16_v((int8x16_t)__rev0, __p1, 33); \
4446444464 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4446544465 __ret; \
......@@ -44468,16 +44468,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4446844468
4446944469#ifdef __LITTLE_ENDIAN__
4447044470#define vcvt_n_f16_u16(__p0, __p1) __extension__ ({ \
44471 uint16x4_t __s0 = __p0; \
4447244471 float16x4_t __ret; \
44472 uint16x4_t __s0 = __p0; \
4447344473 __ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__s0, __p1, 17); \
4447444474 __ret; \
4447544475})
4447644476#else
4447744477#define vcvt_n_f16_u16(__p0, __p1) __extension__ ({ \
44478 float16x4_t __ret; \
4447844479 uint16x4_t __s0 = __p0; \
4447944480 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44480 float16x4_t __ret; \
4448144481 __ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__rev0, __p1, 17); \
4448244482 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4448344483 __ret; \
......@@ -44486,16 +44486,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4448644486
4448744487#ifdef __LITTLE_ENDIAN__
4448844488#define vcvt_n_f16_s16(__p0, __p1) __extension__ ({ \
44489 int16x4_t __s0 = __p0; \
4449044489 float16x4_t __ret; \
44490 int16x4_t __s0 = __p0; \
4449144491 __ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__s0, __p1, 1); \
4449244492 __ret; \
4449344493})
4449444494#else
4449544495#define vcvt_n_f16_s16(__p0, __p1) __extension__ ({ \
44496 float16x4_t __ret; \
4449644497 int16x4_t __s0 = __p0; \
4449744498 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44498 float16x4_t __ret; \
4449944499 __ret = (float16x4_t) __builtin_neon_vcvt_n_f16_v((int8x8_t)__rev0, __p1, 1); \
4450044500 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4450144501 __ret; \
......@@ -44504,16 +44504,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4450444504
4450544505#ifdef __LITTLE_ENDIAN__
4450644506#define vcvtq_n_s16_f16(__p0, __p1) __extension__ ({ \
44507 float16x8_t __s0 = __p0; \
4450844507 int16x8_t __ret; \
44508 float16x8_t __s0 = __p0; \
4450944509 __ret = (int16x8_t) __builtin_neon_vcvtq_n_s16_v((int8x16_t)__s0, __p1, 33); \
4451044510 __ret; \
4451144511})
4451244512#else
4451344513#define vcvtq_n_s16_f16(__p0, __p1) __extension__ ({ \
44514 int16x8_t __ret; \
4451444515 float16x8_t __s0 = __p0; \
4451544516 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44516 int16x8_t __ret; \
4451744517 __ret = (int16x8_t) __builtin_neon_vcvtq_n_s16_v((int8x16_t)__rev0, __p1, 33); \
4451844518 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4451944519 __ret; \
......@@ -44522,16 +44522,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4452244522
4452344523#ifdef __LITTLE_ENDIAN__
4452444524#define vcvt_n_s16_f16(__p0, __p1) __extension__ ({ \
44525 float16x4_t __s0 = __p0; \
4452644525 int16x4_t __ret; \
44526 float16x4_t __s0 = __p0; \
4452744527 __ret = (int16x4_t) __builtin_neon_vcvt_n_s16_v((int8x8_t)__s0, __p1, 1); \
4452844528 __ret; \
4452944529})
4453044530#else
4453144531#define vcvt_n_s16_f16(__p0, __p1) __extension__ ({ \
44532 int16x4_t __ret; \
4453244533 float16x4_t __s0 = __p0; \
4453344534 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44534 int16x4_t __ret; \
4453544535 __ret = (int16x4_t) __builtin_neon_vcvt_n_s16_v((int8x8_t)__rev0, __p1, 1); \
4453644536 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4453744537 __ret; \
......@@ -44540,16 +44540,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4454044540
4454144541#ifdef __LITTLE_ENDIAN__
4454244542#define vcvtq_n_u16_f16(__p0, __p1) __extension__ ({ \
44543 float16x8_t __s0 = __p0; \
4454444543 uint16x8_t __ret; \
44544 float16x8_t __s0 = __p0; \
4454544545 __ret = (uint16x8_t) __builtin_neon_vcvtq_n_u16_v((int8x16_t)__s0, __p1, 49); \
4454644546 __ret; \
4454744547})
4454844548#else
4454944549#define vcvtq_n_u16_f16(__p0, __p1) __extension__ ({ \
44550 uint16x8_t __ret; \
4455044551 float16x8_t __s0 = __p0; \
4455144552 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
44552 uint16x8_t __ret; \
4455344553 __ret = (uint16x8_t) __builtin_neon_vcvtq_n_u16_v((int8x16_t)__rev0, __p1, 49); \
4455444554 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4455544555 __ret; \
......@@ -44558,16 +44558,16 @@ __ai float16x4_t vcvt_f16_s16(int16x4_t __p0) {
4455844558
4455944559#ifdef __LITTLE_ENDIAN__
4456044560#define vcvt_n_u16_f16(__p0, __p1) __extension__ ({ \
44561 float16x4_t __s0 = __p0; \
4456244561 uint16x4_t __ret; \
44562 float16x4_t __s0 = __p0; \
4456344563 __ret = (uint16x4_t) __builtin_neon_vcvt_n_u16_v((int8x8_t)__s0, __p1, 17); \
4456444564 __ret; \
4456544565})
4456644566#else
4456744567#define vcvt_n_u16_f16(__p0, __p1) __extension__ ({ \
44568 uint16x4_t __ret; \
4456844569 float16x4_t __s0 = __p0; \
4456944570 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
44570 uint16x4_t __ret; \
4457144571 __ret = (uint16x4_t) __builtin_neon_vcvt_n_u16_v((int8x8_t)__rev0, __p1, 17); \
4457244572 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4457344573 __ret; \
......@@ -44582,8 +44582,8 @@ __ai int16x8_t vcvtq_s16_f16(float16x8_t __p0) {
4458244582}
4458344583#else
4458444584__ai int16x8_t vcvtq_s16_f16(float16x8_t __p0) {
44585 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4458644585 int16x8_t __ret;
44586 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4458744587 __ret = (int16x8_t) __builtin_neon_vcvtq_s16_v((int8x16_t)__rev0, 33);
4458844588 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4458944589 return __ret;
......@@ -44598,8 +44598,8 @@ __ai int16x4_t vcvt_s16_f16(float16x4_t __p0) {
4459844598}
4459944599#else
4460044600__ai int16x4_t vcvt_s16_f16(float16x4_t __p0) {
44601 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4460244601 int16x4_t __ret;
44602 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4460344603 __ret = (int16x4_t) __builtin_neon_vcvt_s16_v((int8x8_t)__rev0, 1);
4460444604 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4460544605 return __ret;
......@@ -44614,8 +44614,8 @@ __ai uint16x8_t vcvtq_u16_f16(float16x8_t __p0) {
4461444614}
4461544615#else
4461644616__ai uint16x8_t vcvtq_u16_f16(float16x8_t __p0) {
44617 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4461844617 uint16x8_t __ret;
44618 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4461944619 __ret = (uint16x8_t) __builtin_neon_vcvtq_u16_v((int8x16_t)__rev0, 49);
4462044620 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4462144621 return __ret;
......@@ -44630,8 +44630,8 @@ __ai uint16x4_t vcvt_u16_f16(float16x4_t __p0) {
4463044630}
4463144631#else
4463244632__ai uint16x4_t vcvt_u16_f16(float16x4_t __p0) {
44633 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4463444633 uint16x4_t __ret;
44634 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4463544635 __ret = (uint16x4_t) __builtin_neon_vcvt_u16_v((int8x8_t)__rev0, 17);
4463644636 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4463744637 return __ret;
......@@ -44646,8 +44646,8 @@ __ai int16x8_t vcvtaq_s16_f16(float16x8_t __p0) {
4464644646}
4464744647#else
4464844648__ai int16x8_t vcvtaq_s16_f16(float16x8_t __p0) {
44649 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4465044649 int16x8_t __ret;
44650 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4465144651 __ret = (int16x8_t) __builtin_neon_vcvtaq_s16_v((int8x16_t)__rev0, 33);
4465244652 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4465344653 return __ret;
......@@ -44662,8 +44662,8 @@ __ai int16x4_t vcvta_s16_f16(float16x4_t __p0) {
4466244662}
4466344663#else
4466444664__ai int16x4_t vcvta_s16_f16(float16x4_t __p0) {
44665 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4466644665 int16x4_t __ret;
44666 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4466744667 __ret = (int16x4_t) __builtin_neon_vcvta_s16_v((int8x8_t)__rev0, 1);
4466844668 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4466944669 return __ret;
......@@ -44678,8 +44678,8 @@ __ai uint16x8_t vcvtaq_u16_f16(float16x8_t __p0) {
4467844678}
4467944679#else
4468044680__ai uint16x8_t vcvtaq_u16_f16(float16x8_t __p0) {
44681 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4468244681 uint16x8_t __ret;
44682 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4468344683 __ret = (uint16x8_t) __builtin_neon_vcvtaq_u16_v((int8x16_t)__rev0, 49);
4468444684 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4468544685 return __ret;
......@@ -44694,8 +44694,8 @@ __ai uint16x4_t vcvta_u16_f16(float16x4_t __p0) {
4469444694}
4469544695#else
4469644696__ai uint16x4_t vcvta_u16_f16(float16x4_t __p0) {
44697 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4469844697 uint16x4_t __ret;
44698 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4469944699 __ret = (uint16x4_t) __builtin_neon_vcvta_u16_v((int8x8_t)__rev0, 17);
4470044700 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4470144701 return __ret;
......@@ -44710,8 +44710,8 @@ __ai int16x8_t vcvtmq_s16_f16(float16x8_t __p0) {
4471044710}
4471144711#else
4471244712__ai int16x8_t vcvtmq_s16_f16(float16x8_t __p0) {
44713 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4471444713 int16x8_t __ret;
44714 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4471544715 __ret = (int16x8_t) __builtin_neon_vcvtmq_s16_v((int8x16_t)__rev0, 33);
4471644716 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4471744717 return __ret;
......@@ -44726,8 +44726,8 @@ __ai int16x4_t vcvtm_s16_f16(float16x4_t __p0) {
4472644726}
4472744727#else
4472844728__ai int16x4_t vcvtm_s16_f16(float16x4_t __p0) {
44729 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4473044729 int16x4_t __ret;
44730 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4473144731 __ret = (int16x4_t) __builtin_neon_vcvtm_s16_v((int8x8_t)__rev0, 1);
4473244732 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4473344733 return __ret;
......@@ -44742,8 +44742,8 @@ __ai uint16x8_t vcvtmq_u16_f16(float16x8_t __p0) {
4474244742}
4474344743#else
4474444744__ai uint16x8_t vcvtmq_u16_f16(float16x8_t __p0) {
44745 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4474644745 uint16x8_t __ret;
44746 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4474744747 __ret = (uint16x8_t) __builtin_neon_vcvtmq_u16_v((int8x16_t)__rev0, 49);
4474844748 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4474944749 return __ret;
......@@ -44758,8 +44758,8 @@ __ai uint16x4_t vcvtm_u16_f16(float16x4_t __p0) {
4475844758}
4475944759#else
4476044760__ai uint16x4_t vcvtm_u16_f16(float16x4_t __p0) {
44761 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4476244761 uint16x4_t __ret;
44762 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4476344763 __ret = (uint16x4_t) __builtin_neon_vcvtm_u16_v((int8x8_t)__rev0, 17);
4476444764 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4476544765 return __ret;
......@@ -44774,8 +44774,8 @@ __ai int16x8_t vcvtnq_s16_f16(float16x8_t __p0) {
4477444774}
4477544775#else
4477644776__ai int16x8_t vcvtnq_s16_f16(float16x8_t __p0) {
44777 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4477844777 int16x8_t __ret;
44778 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4477944779 __ret = (int16x8_t) __builtin_neon_vcvtnq_s16_v((int8x16_t)__rev0, 33);
4478044780 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4478144781 return __ret;
......@@ -44790,8 +44790,8 @@ __ai int16x4_t vcvtn_s16_f16(float16x4_t __p0) {
4479044790}
4479144791#else
4479244792__ai int16x4_t vcvtn_s16_f16(float16x4_t __p0) {
44793 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4479444793 int16x4_t __ret;
44794 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4479544795 __ret = (int16x4_t) __builtin_neon_vcvtn_s16_v((int8x8_t)__rev0, 1);
4479644796 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4479744797 return __ret;
......@@ -44806,8 +44806,8 @@ __ai uint16x8_t vcvtnq_u16_f16(float16x8_t __p0) {
4480644806}
4480744807#else
4480844808__ai uint16x8_t vcvtnq_u16_f16(float16x8_t __p0) {
44809 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4481044809 uint16x8_t __ret;
44810 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4481144811 __ret = (uint16x8_t) __builtin_neon_vcvtnq_u16_v((int8x16_t)__rev0, 49);
4481244812 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4481344813 return __ret;
......@@ -44822,8 +44822,8 @@ __ai uint16x4_t vcvtn_u16_f16(float16x4_t __p0) {
4482244822}
4482344823#else
4482444824__ai uint16x4_t vcvtn_u16_f16(float16x4_t __p0) {
44825 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4482644825 uint16x4_t __ret;
44826 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4482744827 __ret = (uint16x4_t) __builtin_neon_vcvtn_u16_v((int8x8_t)__rev0, 17);
4482844828 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4482944829 return __ret;
......@@ -44838,8 +44838,8 @@ __ai int16x8_t vcvtpq_s16_f16(float16x8_t __p0) {
4483844838}
4483944839#else
4484044840__ai int16x8_t vcvtpq_s16_f16(float16x8_t __p0) {
44841 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4484244841 int16x8_t __ret;
44842 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4484344843 __ret = (int16x8_t) __builtin_neon_vcvtpq_s16_v((int8x16_t)__rev0, 33);
4484444844 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4484544845 return __ret;
......@@ -44854,8 +44854,8 @@ __ai int16x4_t vcvtp_s16_f16(float16x4_t __p0) {
4485444854}
4485544855#else
4485644856__ai int16x4_t vcvtp_s16_f16(float16x4_t __p0) {
44857 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4485844857 int16x4_t __ret;
44858 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4485944859 __ret = (int16x4_t) __builtin_neon_vcvtp_s16_v((int8x8_t)__rev0, 1);
4486044860 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4486144861 return __ret;
......@@ -44870,8 +44870,8 @@ __ai uint16x8_t vcvtpq_u16_f16(float16x8_t __p0) {
4487044870}
4487144871#else
4487244872__ai uint16x8_t vcvtpq_u16_f16(float16x8_t __p0) {
44873 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4487444873 uint16x8_t __ret;
44874 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4487544875 __ret = (uint16x8_t) __builtin_neon_vcvtpq_u16_v((int8x16_t)__rev0, 49);
4487644876 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4487744877 return __ret;
......@@ -44886,8 +44886,8 @@ __ai uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
4488644886}
4488744887#else
4488844888__ai uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
44889 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4489044889 uint16x4_t __ret;
44890 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4489144891 __ret = (uint16x4_t) __builtin_neon_vcvtp_u16_v((int8x8_t)__rev0, 17);
4489244892 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4489344893 return __ret;
......@@ -44896,19 +44896,19 @@ __ai uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
4489644896
4489744897#ifdef __LITTLE_ENDIAN__
4489844898#define vextq_f16(__p0, __p1, __p2) __extension__ ({ \
44899 float16x8_t __ret; \
4489944900 float16x8_t __s0 = __p0; \
4490044901 float16x8_t __s1 = __p1; \
44901 float16x8_t __ret; \
4490244902 __ret = (float16x8_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 40); \
4490344903 __ret; \
4490444904})
4490544905#else
4490644906#define vextq_f16(__p0, __p1, __p2) __extension__ ({ \
44907 float16x8_t __ret; \
4490744908 float16x8_t __s0 = __p0; \
4490844909 float16x8_t __s1 = __p1; \
4490944910 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
4491044911 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
44911 float16x8_t __ret; \
4491244912 __ret = (float16x8_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 40); \
4491344913 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4491444914 __ret; \
......@@ -44917,19 +44917,19 @@ __ai uint16x4_t vcvtp_u16_f16(float16x4_t __p0) {
4491744917
4491844918#ifdef __LITTLE_ENDIAN__
4491944919#define vext_f16(__p0, __p1, __p2) __extension__ ({ \
44920 float16x4_t __ret; \
4492044921 float16x4_t __s0 = __p0; \
4492144922 float16x4_t __s1 = __p1; \
44922 float16x4_t __ret; \
4492344923 __ret = (float16x4_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 8); \
4492444924 __ret; \
4492544925})
4492644926#else
4492744927#define vext_f16(__p0, __p1, __p2) __extension__ ({ \
44928 float16x4_t __ret; \
4492844929 float16x4_t __s0 = __p0; \
4492944930 float16x4_t __s1 = __p1; \
4493044931 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
4493144932 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
44932 float16x4_t __ret; \
4493344933 __ret = (float16x4_t) __builtin_neon_vext_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 8); \
4493444934 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4493544935 __ret; \
......@@ -44944,10 +44944,10 @@ __ai float16x8_t vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2)
4494444944}
4494544945#else
4494644946__ai float16x8_t vfmaq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44947 float16x8_t __ret;
4494744948 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4494844949 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4494944950 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
44950 float16x8_t __ret;
4495144951 __ret = (float16x8_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 40);
4495244952 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4495344953 return __ret;
......@@ -44967,10 +44967,10 @@ __ai float16x4_t vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2)
4496744967}
4496844968#else
4496944969__ai float16x4_t vfma_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
44970 float16x4_t __ret;
4497044971 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4497144972 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4497244973 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
44973 float16x4_t __ret;
4497444974 __ret = (float16x4_t) __builtin_neon_vfma_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 8);
4497544975 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4497644976 return __ret;
......@@ -44990,10 +44990,10 @@ __ai float16x8_t vfmsq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2)
4499044990}
4499144991#else
4499244992__ai float16x8_t vfmsq_f16(float16x8_t __p0, float16x8_t __p1, float16x8_t __p2) {
44993 float16x8_t __ret;
4499344994 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4499444995 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4499544996 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
44996 float16x8_t __ret;
4499744997 __ret = __noswap_vfmaq_f16(__rev0, -__rev1, __rev2);
4499844998 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4499944999 return __ret;
......@@ -45008,10 +45008,10 @@ __ai float16x4_t vfms_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2)
4500845008}
4500945009#else
4501045010__ai float16x4_t vfms_f16(float16x4_t __p0, float16x4_t __p1, float16x4_t __p2) {
45011 float16x4_t __ret;
4501145012 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4501245013 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4501345014 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
45014 float16x4_t __ret;
4501545015 __ret = __noswap_vfma_f16(__rev0, -__rev1, __rev2);
4501645016 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4501745017 return __ret;
......@@ -45026,9 +45026,9 @@ __ai float16x8_t vmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
4502645026}
4502745027#else
4502845028__ai float16x8_t vmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
45029 float16x8_t __ret;
4502945030 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4503045031 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45031 float16x8_t __ret;
4503245032 __ret = (float16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4503345033 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4503445034 return __ret;
......@@ -45043,9 +45043,9 @@ __ai float16x4_t vmax_f16(float16x4_t __p0, float16x4_t __p1) {
4504345043}
4504445044#else
4504545045__ai float16x4_t vmax_f16(float16x4_t __p0, float16x4_t __p1) {
45046 float16x4_t __ret;
4504645047 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4504745048 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45048 float16x4_t __ret;
4504945049 __ret = (float16x4_t) __builtin_neon_vmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4505045050 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4505145051 return __ret;
......@@ -45060,9 +45060,9 @@ __ai float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
4506045060}
4506145061#else
4506245062__ai float16x8_t vminq_f16(float16x8_t __p0, float16x8_t __p1) {
45063 float16x8_t __ret;
4506345064 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4506445065 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45065 float16x8_t __ret;
4506645066 __ret = (float16x8_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4506745067 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4506845068 return __ret;
......@@ -45077,9 +45077,9 @@ __ai float16x4_t vmin_f16(float16x4_t __p0, float16x4_t __p1) {
4507745077}
4507845078#else
4507945079__ai float16x4_t vmin_f16(float16x4_t __p0, float16x4_t __p1) {
45080 float16x4_t __ret;
4508045081 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4508145082 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45082 float16x4_t __ret;
4508345083 __ret = (float16x4_t) __builtin_neon_vmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4508445084 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4508545085 return __ret;
......@@ -45094,9 +45094,9 @@ __ai float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
4509445094}
4509545095#else
4509645096__ai float16x8_t vmulq_f16(float16x8_t __p0, float16x8_t __p1) {
45097 float16x8_t __ret;
4509745098 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4509845099 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45099 float16x8_t __ret;
4510045100 __ret = __rev0 * __rev1;
4510145101 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4510245102 return __ret;
......@@ -45111,9 +45111,9 @@ __ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
4511145111}
4511245112#else
4511345113__ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
45114 float16x4_t __ret;
4511445115 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4511545116 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45116 float16x4_t __ret;
4511745117 __ret = __rev0 * __rev1;
4511845118 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4511945119 return __ret;
......@@ -45122,19 +45122,19 @@ __ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
4512245122
4512345123#ifdef __LITTLE_ENDIAN__
4512445124#define vmulq_lane_f16(__p0_277, __p1_277, __p2_277) __extension__ ({ \
45125 float16x8_t __ret_277; \
4512545126 float16x8_t __s0_277 = __p0_277; \
4512645127 float16x4_t __s1_277 = __p1_277; \
45127 float16x8_t __ret_277; \
4512845128 __ret_277 = __s0_277 * splatq_lane_f16(__s1_277, __p2_277); \
4512945129 __ret_277; \
4513045130})
4513145131#else
4513245132#define vmulq_lane_f16(__p0_278, __p1_278, __p2_278) __extension__ ({ \
45133 float16x8_t __ret_278; \
4513345134 float16x8_t __s0_278 = __p0_278; \
4513445135 float16x4_t __s1_278 = __p1_278; \
4513545136 float16x8_t __rev0_278; __rev0_278 = __builtin_shufflevector(__s0_278, __s0_278, 7, 6, 5, 4, 3, 2, 1, 0); \
4513645137 float16x4_t __rev1_278; __rev1_278 = __builtin_shufflevector(__s1_278, __s1_278, 3, 2, 1, 0); \
45137 float16x8_t __ret_278; \
4513845138 __ret_278 = __rev0_278 * __noswap_splatq_lane_f16(__rev1_278, __p2_278); \
4513945139 __ret_278 = __builtin_shufflevector(__ret_278, __ret_278, 7, 6, 5, 4, 3, 2, 1, 0); \
4514045140 __ret_278; \
......@@ -45143,19 +45143,19 @@ __ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
4514345143
4514445144#ifdef __LITTLE_ENDIAN__
4514545145#define vmul_lane_f16(__p0_279, __p1_279, __p2_279) __extension__ ({ \
45146 float16x4_t __ret_279; \
4514645147 float16x4_t __s0_279 = __p0_279; \
4514745148 float16x4_t __s1_279 = __p1_279; \
45148 float16x4_t __ret_279; \
4514945149 __ret_279 = __s0_279 * splat_lane_f16(__s1_279, __p2_279); \
4515045150 __ret_279; \
4515145151})
4515245152#else
4515345153#define vmul_lane_f16(__p0_280, __p1_280, __p2_280) __extension__ ({ \
45154 float16x4_t __ret_280; \
4515445155 float16x4_t __s0_280 = __p0_280; \
4515545156 float16x4_t __s1_280 = __p1_280; \
4515645157 float16x4_t __rev0_280; __rev0_280 = __builtin_shufflevector(__s0_280, __s0_280, 3, 2, 1, 0); \
4515745158 float16x4_t __rev1_280; __rev1_280 = __builtin_shufflevector(__s1_280, __s1_280, 3, 2, 1, 0); \
45158 float16x4_t __ret_280; \
4515945159 __ret_280 = __rev0_280 * __noswap_splat_lane_f16(__rev1_280, __p2_280); \
4516045160 __ret_280 = __builtin_shufflevector(__ret_280, __ret_280, 3, 2, 1, 0); \
4516145161 __ret_280; \
......@@ -45164,18 +45164,18 @@ __ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
4516445164
4516545165#ifdef __LITTLE_ENDIAN__
4516645166#define vmulq_n_f16(__p0, __p1) __extension__ ({ \
45167 float16x8_t __ret; \
4516745168 float16x8_t __s0 = __p0; \
4516845169 float16_t __s1 = __p1; \
45169 float16x8_t __ret; \
4517045170 __ret = __s0 * (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}; \
4517145171 __ret; \
4517245172})
4517345173#else
4517445174#define vmulq_n_f16(__p0, __p1) __extension__ ({ \
45175 float16x8_t __ret; \
4517545176 float16x8_t __s0 = __p0; \
4517645177 float16_t __s1 = __p1; \
4517745178 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45178 float16x8_t __ret; \
4517945179 __ret = __rev0 * (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}; \
4518045180 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4518145181 __ret; \
......@@ -45184,18 +45184,18 @@ __ai float16x4_t vmul_f16(float16x4_t __p0, float16x4_t __p1) {
4518445184
4518545185#ifdef __LITTLE_ENDIAN__
4518645186#define vmul_n_f16(__p0, __p1) __extension__ ({ \
45187 float16x4_t __ret; \
4518745188 float16x4_t __s0 = __p0; \
4518845189 float16_t __s1 = __p1; \
45189 float16x4_t __ret; \
4519045190 __ret = __s0 * (float16x4_t) {__s1, __s1, __s1, __s1}; \
4519145191 __ret; \
4519245192})
4519345193#else
4519445194#define vmul_n_f16(__p0, __p1) __extension__ ({ \
45195 float16x4_t __ret; \
4519545196 float16x4_t __s0 = __p0; \
4519645197 float16_t __s1 = __p1; \
4519745198 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45198 float16x4_t __ret; \
4519945199 __ret = __rev0 * (float16x4_t) {__s1, __s1, __s1, __s1}; \
4520045200 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4520145201 __ret; \
......@@ -45210,8 +45210,8 @@ __ai float16x8_t vnegq_f16(float16x8_t __p0) {
4521045210}
4521145211#else
4521245212__ai float16x8_t vnegq_f16(float16x8_t __p0) {
45213 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4521445213 float16x8_t __ret;
45214 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4521545215 __ret = -__rev0;
4521645216 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4521745217 return __ret;
......@@ -45226,8 +45226,8 @@ __ai float16x4_t vneg_f16(float16x4_t __p0) {
4522645226}
4522745227#else
4522845228__ai float16x4_t vneg_f16(float16x4_t __p0) {
45229 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4523045229 float16x4_t __ret;
45230 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4523145231 __ret = -__rev0;
4523245232 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4523345233 return __ret;
......@@ -45242,9 +45242,9 @@ __ai float16x4_t vpadd_f16(float16x4_t __p0, float16x4_t __p1) {
4524245242}
4524345243#else
4524445244__ai float16x4_t vpadd_f16(float16x4_t __p0, float16x4_t __p1) {
45245 float16x4_t __ret;
4524545246 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4524645247 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45247 float16x4_t __ret;
4524845248 __ret = (float16x4_t) __builtin_neon_vpadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4524945249 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4525045250 return __ret;
......@@ -45259,9 +45259,9 @@ __ai float16x4_t vpmax_f16(float16x4_t __p0, float16x4_t __p1) {
4525945259}
4526045260#else
4526145261__ai float16x4_t vpmax_f16(float16x4_t __p0, float16x4_t __p1) {
45262 float16x4_t __ret;
4526245263 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4526345264 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45264 float16x4_t __ret;
4526545265 __ret = (float16x4_t) __builtin_neon_vpmax_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4526645266 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4526745267 return __ret;
......@@ -45276,9 +45276,9 @@ __ai float16x4_t vpmin_f16(float16x4_t __p0, float16x4_t __p1) {
4527645276}
4527745277#else
4527845278__ai float16x4_t vpmin_f16(float16x4_t __p0, float16x4_t __p1) {
45279 float16x4_t __ret;
4527945280 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4528045281 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45281 float16x4_t __ret;
4528245282 __ret = (float16x4_t) __builtin_neon_vpmin_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4528345283 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4528445284 return __ret;
......@@ -45293,8 +45293,8 @@ __ai float16x8_t vrecpeq_f16(float16x8_t __p0) {
4529345293}
4529445294#else
4529545295__ai float16x8_t vrecpeq_f16(float16x8_t __p0) {
45296 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4529745296 float16x8_t __ret;
45297 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4529845298 __ret = (float16x8_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 40);
4529945299 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4530045300 return __ret;
......@@ -45309,8 +45309,8 @@ __ai float16x4_t vrecpe_f16(float16x4_t __p0) {
4530945309}
4531045310#else
4531145311__ai float16x4_t vrecpe_f16(float16x4_t __p0) {
45312 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4531345312 float16x4_t __ret;
45313 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4531445314 __ret = (float16x4_t) __builtin_neon_vrecpe_v((int8x8_t)__rev0, 8);
4531545315 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4531645316 return __ret;
......@@ -45325,9 +45325,9 @@ __ai float16x8_t vrecpsq_f16(float16x8_t __p0, float16x8_t __p1) {
4532545325}
4532645326#else
4532745327__ai float16x8_t vrecpsq_f16(float16x8_t __p0, float16x8_t __p1) {
45328 float16x8_t __ret;
4532845329 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4532945330 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45330 float16x8_t __ret;
4533145331 __ret = (float16x8_t) __builtin_neon_vrecpsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4533245332 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4533345333 return __ret;
......@@ -45342,9 +45342,9 @@ __ai float16x4_t vrecps_f16(float16x4_t __p0, float16x4_t __p1) {
4534245342}
4534345343#else
4534445344__ai float16x4_t vrecps_f16(float16x4_t __p0, float16x4_t __p1) {
45345 float16x4_t __ret;
4534545346 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4534645347 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45347 float16x4_t __ret;
4534845348 __ret = (float16x4_t) __builtin_neon_vrecps_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4534945349 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4535045350 return __ret;
......@@ -45359,8 +45359,8 @@ __ai float16x8_t vrev64q_f16(float16x8_t __p0) {
4535945359}
4536045360#else
4536145361__ai float16x8_t vrev64q_f16(float16x8_t __p0) {
45362 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4536345362 float16x8_t __ret;
45363 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4536445364 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0, 7, 6, 5, 4);
4536545365 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4536645366 return __ret;
......@@ -45375,8 +45375,8 @@ __ai float16x4_t vrev64_f16(float16x4_t __p0) {
4537545375}
4537645376#else
4537745377__ai float16x4_t vrev64_f16(float16x4_t __p0) {
45378 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4537945378 float16x4_t __ret;
45379 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4538045380 __ret = __builtin_shufflevector(__rev0, __rev0, 3, 2, 1, 0);
4538145381 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4538245382 return __ret;
......@@ -45391,8 +45391,8 @@ __ai float16x8_t vrsqrteq_f16(float16x8_t __p0) {
4539145391}
4539245392#else
4539345393__ai float16x8_t vrsqrteq_f16(float16x8_t __p0) {
45394 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4539545394 float16x8_t __ret;
45395 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4539645396 __ret = (float16x8_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 40);
4539745397 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4539845398 return __ret;
......@@ -45407,8 +45407,8 @@ __ai float16x4_t vrsqrte_f16(float16x4_t __p0) {
4540745407}
4540845408#else
4540945409__ai float16x4_t vrsqrte_f16(float16x4_t __p0) {
45410 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4541145410 float16x4_t __ret;
45411 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4541245412 __ret = (float16x4_t) __builtin_neon_vrsqrte_v((int8x8_t)__rev0, 8);
4541345413 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4541445414 return __ret;
......@@ -45423,9 +45423,9 @@ __ai float16x8_t vrsqrtsq_f16(float16x8_t __p0, float16x8_t __p1) {
4542345423}
4542445424#else
4542545425__ai float16x8_t vrsqrtsq_f16(float16x8_t __p0, float16x8_t __p1) {
45426 float16x8_t __ret;
4542645427 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4542745428 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45428 float16x8_t __ret;
4542945429 __ret = (float16x8_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4543045430 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4543145431 return __ret;
......@@ -45440,9 +45440,9 @@ __ai float16x4_t vrsqrts_f16(float16x4_t __p0, float16x4_t __p1) {
4544045440}
4544145441#else
4544245442__ai float16x4_t vrsqrts_f16(float16x4_t __p0, float16x4_t __p1) {
45443 float16x4_t __ret;
4544345444 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4544445445 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45445 float16x4_t __ret;
4544645446 __ret = (float16x4_t) __builtin_neon_vrsqrts_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4544745447 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4544845448 return __ret;
......@@ -45457,9 +45457,9 @@ __ai float16x8_t vsubq_f16(float16x8_t __p0, float16x8_t __p1) {
4545745457}
4545845458#else
4545945459__ai float16x8_t vsubq_f16(float16x8_t __p0, float16x8_t __p1) {
45460 float16x8_t __ret;
4546045461 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4546145462 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45462 float16x8_t __ret;
4546345463 __ret = __rev0 - __rev1;
4546445464 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4546545465 return __ret;
......@@ -45474,9 +45474,9 @@ __ai float16x4_t vsub_f16(float16x4_t __p0, float16x4_t __p1) {
4547445474}
4547545475#else
4547645476__ai float16x4_t vsub_f16(float16x4_t __p0, float16x4_t __p1) {
45477 float16x4_t __ret;
4547745478 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4547845479 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45479 float16x4_t __ret;
4548045480 __ret = __rev0 - __rev1;
4548145481 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4548245482 return __ret;
......@@ -45491,9 +45491,9 @@ __ai float16x8x2_t vtrnq_f16(float16x8_t __p0, float16x8_t __p1) {
4549145491}
4549245492#else
4549345493__ai float16x8x2_t vtrnq_f16(float16x8_t __p0, float16x8_t __p1) {
45494 float16x8x2_t __ret;
4549445495 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4549545496 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45496 float16x8x2_t __ret;
4549745497 __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4549845498
4549945499 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -45510,9 +45510,9 @@ __ai float16x4x2_t vtrn_f16(float16x4_t __p0, float16x4_t __p1) {
4551045510}
4551145511#else
4551245512__ai float16x4x2_t vtrn_f16(float16x4_t __p0, float16x4_t __p1) {
45513 float16x4x2_t __ret;
4551345514 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4551445515 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45515 float16x4x2_t __ret;
4551645516 __builtin_neon_vtrn_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4551745517
4551845518 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -45529,9 +45529,9 @@ __ai float16x8x2_t vuzpq_f16(float16x8_t __p0, float16x8_t __p1) {
4552945529}
4553045530#else
4553145531__ai float16x8x2_t vuzpq_f16(float16x8_t __p0, float16x8_t __p1) {
45532 float16x8x2_t __ret;
4553245533 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4553345534 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45534 float16x8x2_t __ret;
4553545535 __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4553645536
4553745537 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -45548,9 +45548,9 @@ __ai float16x4x2_t vuzp_f16(float16x4_t __p0, float16x4_t __p1) {
4554845548}
4554945549#else
4555045550__ai float16x4x2_t vuzp_f16(float16x4_t __p0, float16x4_t __p1) {
45551 float16x4x2_t __ret;
4555145552 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4555245553 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45553 float16x4x2_t __ret;
4555445554 __builtin_neon_vuzp_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4555545555
4555645556 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -45567,9 +45567,9 @@ __ai float16x8x2_t vzipq_f16(float16x8_t __p0, float16x8_t __p1) {
4556745567}
4556845568#else
4556945569__ai float16x8x2_t vzipq_f16(float16x8_t __p0, float16x8_t __p1) {
45570 float16x8x2_t __ret;
4557045571 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4557145572 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45572 float16x8x2_t __ret;
4557345573 __builtin_neon_vzipq_v(&__ret, (int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4557445574
4557545575 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -45586,9 +45586,9 @@ __ai float16x4x2_t vzip_f16(float16x4_t __p0, float16x4_t __p1) {
4558645586}
4558745587#else
4558845588__ai float16x4x2_t vzip_f16(float16x4_t __p0, float16x4_t __p1) {
45589 float16x4x2_t __ret;
4558945590 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4559045591 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45591 float16x4x2_t __ret;
4559245592 __builtin_neon_vzip_v(&__ret, (int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4559345593
4559445594 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 3, 2, 1, 0);
......@@ -45607,9 +45607,9 @@ __ai float16x8_t vdivq_f16(float16x8_t __p0, float16x8_t __p1) {
4560745607}
4560845608#else
4560945609__ai float16x8_t vdivq_f16(float16x8_t __p0, float16x8_t __p1) {
45610 float16x8_t __ret;
4561045611 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4561145612 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
45612 float16x8_t __ret;
4561345613 __ret = __rev0 / __rev1;
4561445614 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4561545615 return __ret;
......@@ -45624,9 +45624,9 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4562445624}
4562545625#else
4562645626__ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
45627 float16x4_t __ret;
4562745628 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4562845629 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45629 float16x4_t __ret;
4563045630 __ret = __rev0 / __rev1;
4563145631 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4563245632 return __ret;
......@@ -45635,16 +45635,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4563545635
4563645636#ifdef __LITTLE_ENDIAN__
4563745637#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
45638 float16x4_t __s0 = __p0; \
4563945638 float16_t __ret; \
45639 float16x4_t __s0 = __p0; \
4564045640 __ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__s0, __p1); \
4564145641 __ret; \
4564245642})
4564345643#else
4564445644#define vduph_lane_f16(__p0, __p1) __extension__ ({ \
45645 float16_t __ret; \
4564545646 float16x4_t __s0 = __p0; \
4564645647 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
45647 float16_t __ret; \
4564845648 __ret = (float16_t) __builtin_neon_vduph_lane_f16((float16x4_t)__rev0, __p1); \
4564945649 __ret; \
4565045650})
......@@ -45652,16 +45652,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4565245652
4565345653#ifdef __LITTLE_ENDIAN__
4565445654#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
45655 float16x8_t __s0 = __p0; \
4565645655 float16_t __ret; \
45656 float16x8_t __s0 = __p0; \
4565745657 __ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__s0, __p1); \
4565845658 __ret; \
4565945659})
4566045660#else
4566145661#define vduph_laneq_f16(__p0, __p1) __extension__ ({ \
45662 float16_t __ret; \
4566245663 float16x8_t __s0 = __p0; \
4566345664 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
45664 float16_t __ret; \
4566545665 __ret = (float16_t) __builtin_neon_vduph_laneq_f16((float16x8_t)__rev0, __p1); \
4566645666 __ret; \
4566745667})
......@@ -45669,28 +45669,28 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4566945669
4567045670#ifdef __LITTLE_ENDIAN__
4567145671#define vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45672 float16_t __ret; \
4567245673 float16_t __s0 = __p0; \
4567345674 float16_t __s1 = __p1; \
4567445675 float16x4_t __s2 = __p2; \
45675 float16_t __ret; \
4567645676 __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
4567745677 __ret; \
4567845678})
4567945679#else
4568045680#define vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45681 float16_t __ret; \
4568145682 float16_t __s0 = __p0; \
4568245683 float16_t __s1 = __p1; \
4568345684 float16x4_t __s2 = __p2; \
4568445685 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
45685 float16_t __ret; \
4568645686 __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__rev2, __p3); \
4568745687 __ret; \
4568845688})
4568945689#define __noswap_vfmah_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45690 float16_t __ret; \
4569045691 float16_t __s0 = __p0; \
4569145692 float16_t __s1 = __p1; \
4569245693 float16x4_t __s2 = __p2; \
45693 float16_t __ret; \
4569445694 __ret = (float16_t) __builtin_neon_vfmah_lane_f16(__s0, __s1, (float16x4_t)__s2, __p3); \
4569545695 __ret; \
4569645696})
......@@ -45698,31 +45698,31 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4569845698
4569945699#ifdef __LITTLE_ENDIAN__
4570045700#define vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45701 float16x8_t __ret; \
4570145702 float16x8_t __s0 = __p0; \
4570245703 float16x8_t __s1 = __p1; \
4570345704 float16x4_t __s2 = __p2; \
45704 float16x8_t __ret; \
4570545705 __ret = (float16x8_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 40); \
4570645706 __ret; \
4570745707})
4570845708#else
4570945709#define vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45710 float16x8_t __ret; \
4571045711 float16x8_t __s0 = __p0; \
4571145712 float16x8_t __s1 = __p1; \
4571245713 float16x4_t __s2 = __p2; \
4571345714 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
4571445715 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
4571545716 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
45716 float16x8_t __ret; \
4571745717 __ret = (float16x8_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, __p3, 40); \
4571845718 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4571945719 __ret; \
4572045720})
4572145721#define __noswap_vfmaq_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45722 float16x8_t __ret; \
4572245723 float16x8_t __s0 = __p0; \
4572345724 float16x8_t __s1 = __p1; \
4572445725 float16x4_t __s2 = __p2; \
45725 float16x8_t __ret; \
4572645726 __ret = (float16x8_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 40); \
4572745727 __ret; \
4572845728})
......@@ -45730,31 +45730,31 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4573045730
4573145731#ifdef __LITTLE_ENDIAN__
4573245732#define vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45733 float16x4_t __ret; \
4573345734 float16x4_t __s0 = __p0; \
4573445735 float16x4_t __s1 = __p1; \
4573545736 float16x4_t __s2 = __p2; \
45736 float16x4_t __ret; \
4573745737 __ret = (float16x4_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 8); \
4573845738 __ret; \
4573945739})
4574045740#else
4574145741#define vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45742 float16x4_t __ret; \
4574245743 float16x4_t __s0 = __p0; \
4574345744 float16x4_t __s1 = __p1; \
4574445745 float16x4_t __s2 = __p2; \
4574545746 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
4574645747 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
4574745748 float16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
45748 float16x4_t __ret; \
4574945749 __ret = (float16x4_t) __builtin_neon_vfma_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, __p3, 8); \
4575045750 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4575145751 __ret; \
4575245752})
4575345753#define __noswap_vfma_lane_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45754 float16x4_t __ret; \
4575445755 float16x4_t __s0 = __p0; \
4575545756 float16x4_t __s1 = __p1; \
4575645757 float16x4_t __s2 = __p2; \
45757 float16x4_t __ret; \
4575845758 __ret = (float16x4_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 8); \
4575945759 __ret; \
4576045760})
......@@ -45762,28 +45762,28 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4576245762
4576345763#ifdef __LITTLE_ENDIAN__
4576445764#define vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45765 float16_t __ret; \
4576545766 float16_t __s0 = __p0; \
4576645767 float16_t __s1 = __p1; \
4576745768 float16x8_t __s2 = __p2; \
45768 float16_t __ret; \
4576945769 __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
4577045770 __ret; \
4577145771})
4577245772#else
4577345773#define vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45774 float16_t __ret; \
4577445775 float16_t __s0 = __p0; \
4577545776 float16_t __s1 = __p1; \
4577645777 float16x8_t __s2 = __p2; \
4577745778 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
45778 float16_t __ret; \
4577945779 __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__rev2, __p3); \
4578045780 __ret; \
4578145781})
4578245782#define __noswap_vfmah_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45783 float16_t __ret; \
4578345784 float16_t __s0 = __p0; \
4578445785 float16_t __s1 = __p1; \
4578545786 float16x8_t __s2 = __p2; \
45786 float16_t __ret; \
4578745787 __ret = (float16_t) __builtin_neon_vfmah_laneq_f16(__s0, __s1, (float16x8_t)__s2, __p3); \
4578845788 __ret; \
4578945789})
......@@ -45791,31 +45791,31 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4579145791
4579245792#ifdef __LITTLE_ENDIAN__
4579345793#define vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45794 float16x8_t __ret; \
4579445795 float16x8_t __s0 = __p0; \
4579545796 float16x8_t __s1 = __p1; \
4579645797 float16x8_t __s2 = __p2; \
45797 float16x8_t __ret; \
4579845798 __ret = (float16x8_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 40); \
4579945799 __ret; \
4580045800})
4580145801#else
4580245802#define vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45803 float16x8_t __ret; \
4580345804 float16x8_t __s0 = __p0; \
4580445805 float16x8_t __s1 = __p1; \
4580545806 float16x8_t __s2 = __p2; \
4580645807 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
4580745808 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
4580845809 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
45809 float16x8_t __ret; \
4581045810 __ret = (float16x8_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 40); \
4581145811 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4581245812 __ret; \
4581345813})
4581445814#define __noswap_vfmaq_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45815 float16x8_t __ret; \
4581545816 float16x8_t __s0 = __p0; \
4581645817 float16x8_t __s1 = __p1; \
4581745818 float16x8_t __s2 = __p2; \
45818 float16x8_t __ret; \
4581945819 __ret = (float16x8_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 40); \
4582045820 __ret; \
4582145821})
......@@ -45823,31 +45823,31 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4582345823
4582445824#ifdef __LITTLE_ENDIAN__
4582545825#define vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45826 float16x4_t __ret; \
4582645827 float16x4_t __s0 = __p0; \
4582745828 float16x4_t __s1 = __p1; \
4582845829 float16x8_t __s2 = __p2; \
45829 float16x4_t __ret; \
4583045830 __ret = (float16x4_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 8); \
4583145831 __ret; \
4583245832})
4583345833#else
4583445834#define vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45835 float16x4_t __ret; \
4583545836 float16x4_t __s0 = __p0; \
4583645837 float16x4_t __s1 = __p1; \
4583745838 float16x8_t __s2 = __p2; \
4583845839 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
4583945840 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
4584045841 float16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
45841 float16x4_t __ret; \
4584245842 __ret = (float16x4_t) __builtin_neon_vfma_laneq_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x16_t)__rev2, __p3, 8); \
4584345843 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4584445844 __ret; \
4584545845})
4584645846#define __noswap_vfma_laneq_f16(__p0, __p1, __p2, __p3) __extension__ ({ \
45847 float16x4_t __ret; \
4584745848 float16x4_t __s0 = __p0; \
4584845849 float16x4_t __s1 = __p1; \
4584945850 float16x8_t __s2 = __p2; \
45850 float16x4_t __ret; \
4585145851 __ret = (float16x4_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 8); \
4585245852 __ret; \
4585345853})
......@@ -45855,21 +45855,21 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4585545855
4585645856#ifdef __LITTLE_ENDIAN__
4585745857#define vfmaq_n_f16(__p0, __p1, __p2) __extension__ ({ \
45858 float16x8_t __ret; \
4585845859 float16x8_t __s0 = __p0; \
4585945860 float16x8_t __s1 = __p1; \
4586045861 float16_t __s2 = __p2; \
45861 float16x8_t __ret; \
4586245862 __ret = vfmaq_f16(__s0, __s1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
4586345863 __ret; \
4586445864})
4586545865#else
4586645866#define vfmaq_n_f16(__p0, __p1, __p2) __extension__ ({ \
45867 float16x8_t __ret; \
4586745868 float16x8_t __s0 = __p0; \
4586845869 float16x8_t __s1 = __p1; \
4586945870 float16_t __s2 = __p2; \
4587045871 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
4587145872 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
45872 float16x8_t __ret; \
4587345873 __ret = __noswap_vfmaq_f16(__rev0, __rev1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
4587445874 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4587545875 __ret; \
......@@ -45878,21 +45878,21 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4587845878
4587945879#ifdef __LITTLE_ENDIAN__
4588045880#define vfma_n_f16(__p0, __p1, __p2) __extension__ ({ \
45881 float16x4_t __ret; \
4588145882 float16x4_t __s0 = __p0; \
4588245883 float16x4_t __s1 = __p1; \
4588345884 float16_t __s2 = __p2; \
45884 float16x4_t __ret; \
4588545885 __ret = vfma_f16(__s0, __s1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
4588645886 __ret; \
4588745887})
4588845888#else
4588945889#define vfma_n_f16(__p0, __p1, __p2) __extension__ ({ \
45890 float16x4_t __ret; \
4589045891 float16x4_t __s0 = __p0; \
4589145892 float16x4_t __s1 = __p1; \
4589245893 float16_t __s2 = __p2; \
4589345894 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
4589445895 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
45895 float16x4_t __ret; \
4589645896 __ret = __noswap_vfma_f16(__rev0, __rev1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
4589745897 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4589845898 __ret; \
......@@ -45901,20 +45901,20 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4590145901
4590245902#ifdef __LITTLE_ENDIAN__
4590345903#define vfmsh_lane_f16(__p0_281, __p1_281, __p2_281, __p3_281) __extension__ ({ \
45904 float16_t __ret_281; \
4590445905 float16_t __s0_281 = __p0_281; \
4590545906 float16_t __s1_281 = __p1_281; \
4590645907 float16x4_t __s2_281 = __p2_281; \
45907 float16_t __ret_281; \
4590845908 __ret_281 = vfmah_lane_f16(__s0_281, -__s1_281, __s2_281, __p3_281); \
4590945909 __ret_281; \
4591045910})
4591145911#else
4591245912#define vfmsh_lane_f16(__p0_282, __p1_282, __p2_282, __p3_282) __extension__ ({ \
45913 float16_t __ret_282; \
4591345914 float16_t __s0_282 = __p0_282; \
4591445915 float16_t __s1_282 = __p1_282; \
4591545916 float16x4_t __s2_282 = __p2_282; \
4591645917 float16x4_t __rev2_282; __rev2_282 = __builtin_shufflevector(__s2_282, __s2_282, 3, 2, 1, 0); \
45917 float16_t __ret_282; \
4591845918 __ret_282 = __noswap_vfmah_lane_f16(__s0_282, -__s1_282, __rev2_282, __p3_282); \
4591945919 __ret_282; \
4592045920})
......@@ -45922,22 +45922,22 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4592245922
4592345923#ifdef __LITTLE_ENDIAN__
4592445924#define vfmsq_lane_f16(__p0_283, __p1_283, __p2_283, __p3_283) __extension__ ({ \
45925 float16x8_t __ret_283; \
4592545926 float16x8_t __s0_283 = __p0_283; \
4592645927 float16x8_t __s1_283 = __p1_283; \
4592745928 float16x4_t __s2_283 = __p2_283; \
45928 float16x8_t __ret_283; \
4592945929 __ret_283 = vfmaq_lane_f16(__s0_283, -__s1_283, __s2_283, __p3_283); \
4593045930 __ret_283; \
4593145931})
4593245932#else
4593345933#define vfmsq_lane_f16(__p0_284, __p1_284, __p2_284, __p3_284) __extension__ ({ \
45934 float16x8_t __ret_284; \
4593445935 float16x8_t __s0_284 = __p0_284; \
4593545936 float16x8_t __s1_284 = __p1_284; \
4593645937 float16x4_t __s2_284 = __p2_284; \
4593745938 float16x8_t __rev0_284; __rev0_284 = __builtin_shufflevector(__s0_284, __s0_284, 7, 6, 5, 4, 3, 2, 1, 0); \
4593845939 float16x8_t __rev1_284; __rev1_284 = __builtin_shufflevector(__s1_284, __s1_284, 7, 6, 5, 4, 3, 2, 1, 0); \
4593945940 float16x4_t __rev2_284; __rev2_284 = __builtin_shufflevector(__s2_284, __s2_284, 3, 2, 1, 0); \
45940 float16x8_t __ret_284; \
4594145941 __ret_284 = __noswap_vfmaq_lane_f16(__rev0_284, -__rev1_284, __rev2_284, __p3_284); \
4594245942 __ret_284 = __builtin_shufflevector(__ret_284, __ret_284, 7, 6, 5, 4, 3, 2, 1, 0); \
4594345943 __ret_284; \
......@@ -45946,22 +45946,22 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4594645946
4594745947#ifdef __LITTLE_ENDIAN__
4594845948#define vfms_lane_f16(__p0_285, __p1_285, __p2_285, __p3_285) __extension__ ({ \
45949 float16x4_t __ret_285; \
4594945950 float16x4_t __s0_285 = __p0_285; \
4595045951 float16x4_t __s1_285 = __p1_285; \
4595145952 float16x4_t __s2_285 = __p2_285; \
45952 float16x4_t __ret_285; \
4595345953 __ret_285 = vfma_lane_f16(__s0_285, -__s1_285, __s2_285, __p3_285); \
4595445954 __ret_285; \
4595545955})
4595645956#else
4595745957#define vfms_lane_f16(__p0_286, __p1_286, __p2_286, __p3_286) __extension__ ({ \
45958 float16x4_t __ret_286; \
4595845959 float16x4_t __s0_286 = __p0_286; \
4595945960 float16x4_t __s1_286 = __p1_286; \
4596045961 float16x4_t __s2_286 = __p2_286; \
4596145962 float16x4_t __rev0_286; __rev0_286 = __builtin_shufflevector(__s0_286, __s0_286, 3, 2, 1, 0); \
4596245963 float16x4_t __rev1_286; __rev1_286 = __builtin_shufflevector(__s1_286, __s1_286, 3, 2, 1, 0); \
4596345964 float16x4_t __rev2_286; __rev2_286 = __builtin_shufflevector(__s2_286, __s2_286, 3, 2, 1, 0); \
45964 float16x4_t __ret_286; \
4596545965 __ret_286 = __noswap_vfma_lane_f16(__rev0_286, -__rev1_286, __rev2_286, __p3_286); \
4596645966 __ret_286 = __builtin_shufflevector(__ret_286, __ret_286, 3, 2, 1, 0); \
4596745967 __ret_286; \
......@@ -45970,20 +45970,20 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4597045970
4597145971#ifdef __LITTLE_ENDIAN__
4597245972#define vfmsh_laneq_f16(__p0_287, __p1_287, __p2_287, __p3_287) __extension__ ({ \
45973 float16_t __ret_287; \
4597345974 float16_t __s0_287 = __p0_287; \
4597445975 float16_t __s1_287 = __p1_287; \
4597545976 float16x8_t __s2_287 = __p2_287; \
45976 float16_t __ret_287; \
4597745977 __ret_287 = vfmah_laneq_f16(__s0_287, -__s1_287, __s2_287, __p3_287); \
4597845978 __ret_287; \
4597945979})
4598045980#else
4598145981#define vfmsh_laneq_f16(__p0_288, __p1_288, __p2_288, __p3_288) __extension__ ({ \
45982 float16_t __ret_288; \
4598245983 float16_t __s0_288 = __p0_288; \
4598345984 float16_t __s1_288 = __p1_288; \
4598445985 float16x8_t __s2_288 = __p2_288; \
4598545986 float16x8_t __rev2_288; __rev2_288 = __builtin_shufflevector(__s2_288, __s2_288, 7, 6, 5, 4, 3, 2, 1, 0); \
45986 float16_t __ret_288; \
4598745987 __ret_288 = __noswap_vfmah_laneq_f16(__s0_288, -__s1_288, __rev2_288, __p3_288); \
4598845988 __ret_288; \
4598945989})
......@@ -45991,22 +45991,22 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4599145991
4599245992#ifdef __LITTLE_ENDIAN__
4599345993#define vfmsq_laneq_f16(__p0_289, __p1_289, __p2_289, __p3_289) __extension__ ({ \
45994 float16x8_t __ret_289; \
4599445995 float16x8_t __s0_289 = __p0_289; \
4599545996 float16x8_t __s1_289 = __p1_289; \
4599645997 float16x8_t __s2_289 = __p2_289; \
45997 float16x8_t __ret_289; \
4599845998 __ret_289 = vfmaq_laneq_f16(__s0_289, -__s1_289, __s2_289, __p3_289); \
4599945999 __ret_289; \
4600046000})
4600146001#else
4600246002#define vfmsq_laneq_f16(__p0_290, __p1_290, __p2_290, __p3_290) __extension__ ({ \
46003 float16x8_t __ret_290; \
4600346004 float16x8_t __s0_290 = __p0_290; \
4600446005 float16x8_t __s1_290 = __p1_290; \
4600546006 float16x8_t __s2_290 = __p2_290; \
4600646007 float16x8_t __rev0_290; __rev0_290 = __builtin_shufflevector(__s0_290, __s0_290, 7, 6, 5, 4, 3, 2, 1, 0); \
4600746008 float16x8_t __rev1_290; __rev1_290 = __builtin_shufflevector(__s1_290, __s1_290, 7, 6, 5, 4, 3, 2, 1, 0); \
4600846009 float16x8_t __rev2_290; __rev2_290 = __builtin_shufflevector(__s2_290, __s2_290, 7, 6, 5, 4, 3, 2, 1, 0); \
46009 float16x8_t __ret_290; \
4601046010 __ret_290 = __noswap_vfmaq_laneq_f16(__rev0_290, -__rev1_290, __rev2_290, __p3_290); \
4601146011 __ret_290 = __builtin_shufflevector(__ret_290, __ret_290, 7, 6, 5, 4, 3, 2, 1, 0); \
4601246012 __ret_290; \
......@@ -46015,22 +46015,22 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4601546015
4601646016#ifdef __LITTLE_ENDIAN__
4601746017#define vfms_laneq_f16(__p0_291, __p1_291, __p2_291, __p3_291) __extension__ ({ \
46018 float16x4_t __ret_291; \
4601846019 float16x4_t __s0_291 = __p0_291; \
4601946020 float16x4_t __s1_291 = __p1_291; \
4602046021 float16x8_t __s2_291 = __p2_291; \
46021 float16x4_t __ret_291; \
4602246022 __ret_291 = vfma_laneq_f16(__s0_291, -__s1_291, __s2_291, __p3_291); \
4602346023 __ret_291; \
4602446024})
4602546025#else
4602646026#define vfms_laneq_f16(__p0_292, __p1_292, __p2_292, __p3_292) __extension__ ({ \
46027 float16x4_t __ret_292; \
4602746028 float16x4_t __s0_292 = __p0_292; \
4602846029 float16x4_t __s1_292 = __p1_292; \
4602946030 float16x8_t __s2_292 = __p2_292; \
4603046031 float16x4_t __rev0_292; __rev0_292 = __builtin_shufflevector(__s0_292, __s0_292, 3, 2, 1, 0); \
4603146032 float16x4_t __rev1_292; __rev1_292 = __builtin_shufflevector(__s1_292, __s1_292, 3, 2, 1, 0); \
4603246033 float16x8_t __rev2_292; __rev2_292 = __builtin_shufflevector(__s2_292, __s2_292, 7, 6, 5, 4, 3, 2, 1, 0); \
46033 float16x4_t __ret_292; \
4603446034 __ret_292 = __noswap_vfma_laneq_f16(__rev0_292, -__rev1_292, __rev2_292, __p3_292); \
4603546035 __ret_292 = __builtin_shufflevector(__ret_292, __ret_292, 3, 2, 1, 0); \
4603646036 __ret_292; \
......@@ -46039,21 +46039,21 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4603946039
4604046040#ifdef __LITTLE_ENDIAN__
4604146041#define vfmsq_n_f16(__p0, __p1, __p2) __extension__ ({ \
46042 float16x8_t __ret; \
4604246043 float16x8_t __s0 = __p0; \
4604346044 float16x8_t __s1 = __p1; \
4604446045 float16_t __s2 = __p2; \
46045 float16x8_t __ret; \
4604646046 __ret = vfmaq_f16(__s0, -__s1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
4604746047 __ret; \
4604846048})
4604946049#else
4605046050#define vfmsq_n_f16(__p0, __p1, __p2) __extension__ ({ \
46051 float16x8_t __ret; \
4605146052 float16x8_t __s0 = __p0; \
4605246053 float16x8_t __s1 = __p1; \
4605346054 float16_t __s2 = __p2; \
4605446055 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
4605546056 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
46056 float16x8_t __ret; \
4605746057 __ret = __noswap_vfmaq_f16(__rev0, -__rev1, (float16x8_t) {__s2, __s2, __s2, __s2, __s2, __s2, __s2, __s2}); \
4605846058 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4605946059 __ret; \
......@@ -46062,21 +46062,21 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4606246062
4606346063#ifdef __LITTLE_ENDIAN__
4606446064#define vfms_n_f16(__p0, __p1, __p2) __extension__ ({ \
46065 float16x4_t __ret; \
4606546066 float16x4_t __s0 = __p0; \
4606646067 float16x4_t __s1 = __p1; \
4606746068 float16_t __s2 = __p2; \
46068 float16x4_t __ret; \
4606946069 __ret = vfma_f16(__s0, -__s1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
4607046070 __ret; \
4607146071})
4607246072#else
4607346073#define vfms_n_f16(__p0, __p1, __p2) __extension__ ({ \
46074 float16x4_t __ret; \
4607446075 float16x4_t __s0 = __p0; \
4607546076 float16x4_t __s1 = __p1; \
4607646077 float16_t __s2 = __p2; \
4607746078 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
4607846079 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
46079 float16x4_t __ret; \
4608046080 __ret = __noswap_vfma_f16(__rev0, -__rev1, (float16x4_t) {__s2, __s2, __s2, __s2}); \
4608146081 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4608246082 __ret; \
......@@ -46085,16 +46085,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4608546085
4608646086#ifdef __LITTLE_ENDIAN__
4608746087#define vmaxnmvq_f16(__p0) __extension__ ({ \
46088 float16x8_t __s0 = __p0; \
4608946088 float16_t __ret; \
46089 float16x8_t __s0 = __p0; \
4609046090 __ret = (float16_t) __builtin_neon_vmaxnmvq_f16((int8x16_t)__s0); \
4609146091 __ret; \
4609246092})
4609346093#else
4609446094#define vmaxnmvq_f16(__p0) __extension__ ({ \
46095 float16_t __ret; \
4609546096 float16x8_t __s0 = __p0; \
4609646097 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46097 float16_t __ret; \
4609846098 __ret = (float16_t) __builtin_neon_vmaxnmvq_f16((int8x16_t)__rev0); \
4609946099 __ret; \
4610046100})
......@@ -46102,16 +46102,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4610246102
4610346103#ifdef __LITTLE_ENDIAN__
4610446104#define vmaxnmv_f16(__p0) __extension__ ({ \
46105 float16x4_t __s0 = __p0; \
4610646105 float16_t __ret; \
46106 float16x4_t __s0 = __p0; \
4610746107 __ret = (float16_t) __builtin_neon_vmaxnmv_f16((int8x8_t)__s0); \
4610846108 __ret; \
4610946109})
4611046110#else
4611146111#define vmaxnmv_f16(__p0) __extension__ ({ \
46112 float16_t __ret; \
4611246113 float16x4_t __s0 = __p0; \
4611346114 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46114 float16_t __ret; \
4611546115 __ret = (float16_t) __builtin_neon_vmaxnmv_f16((int8x8_t)__rev0); \
4611646116 __ret; \
4611746117})
......@@ -46119,16 +46119,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4611946119
4612046120#ifdef __LITTLE_ENDIAN__
4612146121#define vmaxvq_f16(__p0) __extension__ ({ \
46122 float16x8_t __s0 = __p0; \
4612346122 float16_t __ret; \
46123 float16x8_t __s0 = __p0; \
4612446124 __ret = (float16_t) __builtin_neon_vmaxvq_f16((int8x16_t)__s0); \
4612546125 __ret; \
4612646126})
4612746127#else
4612846128#define vmaxvq_f16(__p0) __extension__ ({ \
46129 float16_t __ret; \
4612946130 float16x8_t __s0 = __p0; \
4613046131 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46131 float16_t __ret; \
4613246132 __ret = (float16_t) __builtin_neon_vmaxvq_f16((int8x16_t)__rev0); \
4613346133 __ret; \
4613446134})
......@@ -46136,16 +46136,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4613646136
4613746137#ifdef __LITTLE_ENDIAN__
4613846138#define vmaxv_f16(__p0) __extension__ ({ \
46139 float16x4_t __s0 = __p0; \
4614046139 float16_t __ret; \
46140 float16x4_t __s0 = __p0; \
4614146141 __ret = (float16_t) __builtin_neon_vmaxv_f16((int8x8_t)__s0); \
4614246142 __ret; \
4614346143})
4614446144#else
4614546145#define vmaxv_f16(__p0) __extension__ ({ \
46146 float16_t __ret; \
4614646147 float16x4_t __s0 = __p0; \
4614746148 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46148 float16_t __ret; \
4614946149 __ret = (float16_t) __builtin_neon_vmaxv_f16((int8x8_t)__rev0); \
4615046150 __ret; \
4615146151})
......@@ -46153,16 +46153,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4615346153
4615446154#ifdef __LITTLE_ENDIAN__
4615546155#define vminnmvq_f16(__p0) __extension__ ({ \
46156 float16x8_t __s0 = __p0; \
4615746156 float16_t __ret; \
46157 float16x8_t __s0 = __p0; \
4615846158 __ret = (float16_t) __builtin_neon_vminnmvq_f16((int8x16_t)__s0); \
4615946159 __ret; \
4616046160})
4616146161#else
4616246162#define vminnmvq_f16(__p0) __extension__ ({ \
46163 float16_t __ret; \
4616346164 float16x8_t __s0 = __p0; \
4616446165 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46165 float16_t __ret; \
4616646166 __ret = (float16_t) __builtin_neon_vminnmvq_f16((int8x16_t)__rev0); \
4616746167 __ret; \
4616846168})
......@@ -46170,16 +46170,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4617046170
4617146171#ifdef __LITTLE_ENDIAN__
4617246172#define vminnmv_f16(__p0) __extension__ ({ \
46173 float16x4_t __s0 = __p0; \
4617446173 float16_t __ret; \
46174 float16x4_t __s0 = __p0; \
4617546175 __ret = (float16_t) __builtin_neon_vminnmv_f16((int8x8_t)__s0); \
4617646176 __ret; \
4617746177})
4617846178#else
4617946179#define vminnmv_f16(__p0) __extension__ ({ \
46180 float16_t __ret; \
4618046181 float16x4_t __s0 = __p0; \
4618146182 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46182 float16_t __ret; \
4618346183 __ret = (float16_t) __builtin_neon_vminnmv_f16((int8x8_t)__rev0); \
4618446184 __ret; \
4618546185})
......@@ -46187,16 +46187,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4618746187
4618846188#ifdef __LITTLE_ENDIAN__
4618946189#define vminvq_f16(__p0) __extension__ ({ \
46190 float16x8_t __s0 = __p0; \
4619146190 float16_t __ret; \
46191 float16x8_t __s0 = __p0; \
4619246192 __ret = (float16_t) __builtin_neon_vminvq_f16((int8x16_t)__s0); \
4619346193 __ret; \
4619446194})
4619546195#else
4619646196#define vminvq_f16(__p0) __extension__ ({ \
46197 float16_t __ret; \
4619746198 float16x8_t __s0 = __p0; \
4619846199 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46199 float16_t __ret; \
4620046200 __ret = (float16_t) __builtin_neon_vminvq_f16((int8x16_t)__rev0); \
4620146201 __ret; \
4620246202})
......@@ -46204,16 +46204,16 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4620446204
4620546205#ifdef __LITTLE_ENDIAN__
4620646206#define vminv_f16(__p0) __extension__ ({ \
46207 float16x4_t __s0 = __p0; \
4620846207 float16_t __ret; \
46208 float16x4_t __s0 = __p0; \
4620946209 __ret = (float16_t) __builtin_neon_vminv_f16((int8x8_t)__s0); \
4621046210 __ret; \
4621146211})
4621246212#else
4621346213#define vminv_f16(__p0) __extension__ ({ \
46214 float16_t __ret; \
4621446215 float16x4_t __s0 = __p0; \
4621546216 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46216 float16_t __ret; \
4621746217 __ret = (float16_t) __builtin_neon_vminv_f16((int8x8_t)__rev0); \
4621846218 __ret; \
4621946219})
......@@ -46221,19 +46221,19 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4622146221
4622246222#ifdef __LITTLE_ENDIAN__
4622346223#define vmulq_laneq_f16(__p0_293, __p1_293, __p2_293) __extension__ ({ \
46224 float16x8_t __ret_293; \
4622446225 float16x8_t __s0_293 = __p0_293; \
4622546226 float16x8_t __s1_293 = __p1_293; \
46226 float16x8_t __ret_293; \
4622746227 __ret_293 = __s0_293 * splatq_laneq_f16(__s1_293, __p2_293); \
4622846228 __ret_293; \
4622946229})
4623046230#else
4623146231#define vmulq_laneq_f16(__p0_294, __p1_294, __p2_294) __extension__ ({ \
46232 float16x8_t __ret_294; \
4623246233 float16x8_t __s0_294 = __p0_294; \
4623346234 float16x8_t __s1_294 = __p1_294; \
4623446235 float16x8_t __rev0_294; __rev0_294 = __builtin_shufflevector(__s0_294, __s0_294, 7, 6, 5, 4, 3, 2, 1, 0); \
4623546236 float16x8_t __rev1_294; __rev1_294 = __builtin_shufflevector(__s1_294, __s1_294, 7, 6, 5, 4, 3, 2, 1, 0); \
46236 float16x8_t __ret_294; \
4623746237 __ret_294 = __rev0_294 * __noswap_splatq_laneq_f16(__rev1_294, __p2_294); \
4623846238 __ret_294 = __builtin_shufflevector(__ret_294, __ret_294, 7, 6, 5, 4, 3, 2, 1, 0); \
4623946239 __ret_294; \
......@@ -46242,19 +46242,19 @@ __ai float16x4_t vdiv_f16(float16x4_t __p0, float16x4_t __p1) {
4624246242
4624346243#ifdef __LITTLE_ENDIAN__
4624446244#define vmul_laneq_f16(__p0_295, __p1_295, __p2_295) __extension__ ({ \
46245 float16x4_t __ret_295; \
4624546246 float16x4_t __s0_295 = __p0_295; \
4624646247 float16x8_t __s1_295 = __p1_295; \
46247 float16x4_t __ret_295; \
4624846248 __ret_295 = __s0_295 * splat_laneq_f16(__s1_295, __p2_295); \
4624946249 __ret_295; \
4625046250})
4625146251#else
4625246252#define vmul_laneq_f16(__p0_296, __p1_296, __p2_296) __extension__ ({ \
46253 float16x4_t __ret_296; \
4625346254 float16x4_t __s0_296 = __p0_296; \
4625446255 float16x8_t __s1_296 = __p1_296; \
4625546256 float16x4_t __rev0_296; __rev0_296 = __builtin_shufflevector(__s0_296, __s0_296, 3, 2, 1, 0); \
4625646257 float16x8_t __rev1_296; __rev1_296 = __builtin_shufflevector(__s1_296, __s1_296, 7, 6, 5, 4, 3, 2, 1, 0); \
46257 float16x4_t __ret_296; \
4625846258 __ret_296 = __rev0_296 * __noswap_splat_laneq_f16(__rev1_296, __p2_296); \
4625946259 __ret_296 = __builtin_shufflevector(__ret_296, __ret_296, 3, 2, 1, 0); \
4626046260 __ret_296; \
......@@ -46269,9 +46269,9 @@ __ai float16x8_t vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
4626946269}
4627046270#else
4627146271__ai float16x8_t vmulxq_f16(float16x8_t __p0, float16x8_t __p1) {
46272 float16x8_t __ret;
4627246273 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4627346274 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46274 float16x8_t __ret;
4627546275 __ret = (float16x8_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4627646276 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4627746277 return __ret;
......@@ -46291,9 +46291,9 @@ __ai float16x4_t vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4629146291}
4629246292#else
4629346293__ai float16x4_t vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
46294 float16x4_t __ret;
4629446295 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4629546296 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46296 float16x4_t __ret;
4629746297 __ret = (float16x4_t) __builtin_neon_vmulx_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4629846298 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4629946299 return __ret;
......@@ -46307,18 +46307,18 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4630746307
4630846308#ifdef __LITTLE_ENDIAN__
4630946309#define vmulxh_lane_f16(__p0, __p1, __p2) __extension__ ({ \
46310 float16_t __ret; \
4631046311 float16_t __s0 = __p0; \
4631146312 float16x4_t __s1 = __p1; \
46312 float16_t __ret; \
4631346313 __ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__s1, __p2); \
4631446314 __ret; \
4631546315})
4631646316#else
4631746317#define vmulxh_lane_f16(__p0, __p1, __p2) __extension__ ({ \
46318 float16_t __ret; \
4631846319 float16_t __s0 = __p0; \
4631946320 float16x4_t __s1 = __p1; \
4632046321 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
46321 float16_t __ret; \
4632246322 __ret = (float16_t) __builtin_neon_vmulxh_lane_f16(__s0, (float16x4_t)__rev1, __p2); \
4632346323 __ret; \
4632446324})
......@@ -46326,19 +46326,19 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4632646326
4632746327#ifdef __LITTLE_ENDIAN__
4632846328#define vmulxq_lane_f16(__p0_297, __p1_297, __p2_297) __extension__ ({ \
46329 float16x8_t __ret_297; \
4632946330 float16x8_t __s0_297 = __p0_297; \
4633046331 float16x4_t __s1_297 = __p1_297; \
46331 float16x8_t __ret_297; \
4633246332 __ret_297 = vmulxq_f16(__s0_297, splatq_lane_f16(__s1_297, __p2_297)); \
4633346333 __ret_297; \
4633446334})
4633546335#else
4633646336#define vmulxq_lane_f16(__p0_298, __p1_298, __p2_298) __extension__ ({ \
46337 float16x8_t __ret_298; \
4633746338 float16x8_t __s0_298 = __p0_298; \
4633846339 float16x4_t __s1_298 = __p1_298; \
4633946340 float16x8_t __rev0_298; __rev0_298 = __builtin_shufflevector(__s0_298, __s0_298, 7, 6, 5, 4, 3, 2, 1, 0); \
4634046341 float16x4_t __rev1_298; __rev1_298 = __builtin_shufflevector(__s1_298, __s1_298, 3, 2, 1, 0); \
46341 float16x8_t __ret_298; \
4634246342 __ret_298 = __noswap_vmulxq_f16(__rev0_298, __noswap_splatq_lane_f16(__rev1_298, __p2_298)); \
4634346343 __ret_298 = __builtin_shufflevector(__ret_298, __ret_298, 7, 6, 5, 4, 3, 2, 1, 0); \
4634446344 __ret_298; \
......@@ -46347,19 +46347,19 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4634746347
4634846348#ifdef __LITTLE_ENDIAN__
4634946349#define vmulx_lane_f16(__p0_299, __p1_299, __p2_299) __extension__ ({ \
46350 float16x4_t __ret_299; \
4635046351 float16x4_t __s0_299 = __p0_299; \
4635146352 float16x4_t __s1_299 = __p1_299; \
46352 float16x4_t __ret_299; \
4635346353 __ret_299 = vmulx_f16(__s0_299, splat_lane_f16(__s1_299, __p2_299)); \
4635446354 __ret_299; \
4635546355})
4635646356#else
4635746357#define vmulx_lane_f16(__p0_300, __p1_300, __p2_300) __extension__ ({ \
46358 float16x4_t __ret_300; \
4635846359 float16x4_t __s0_300 = __p0_300; \
4635946360 float16x4_t __s1_300 = __p1_300; \
4636046361 float16x4_t __rev0_300; __rev0_300 = __builtin_shufflevector(__s0_300, __s0_300, 3, 2, 1, 0); \
4636146362 float16x4_t __rev1_300; __rev1_300 = __builtin_shufflevector(__s1_300, __s1_300, 3, 2, 1, 0); \
46362 float16x4_t __ret_300; \
4636346363 __ret_300 = __noswap_vmulx_f16(__rev0_300, __noswap_splat_lane_f16(__rev1_300, __p2_300)); \
4636446364 __ret_300 = __builtin_shufflevector(__ret_300, __ret_300, 3, 2, 1, 0); \
4636546365 __ret_300; \
......@@ -46368,18 +46368,18 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4636846368
4636946369#ifdef __LITTLE_ENDIAN__
4637046370#define vmulxh_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
46371 float16_t __ret; \
4637146372 float16_t __s0 = __p0; \
4637246373 float16x8_t __s1 = __p1; \
46373 float16_t __ret; \
4637446374 __ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__s1, __p2); \
4637546375 __ret; \
4637646376})
4637746377#else
4637846378#define vmulxh_laneq_f16(__p0, __p1, __p2) __extension__ ({ \
46379 float16_t __ret; \
4637946380 float16_t __s0 = __p0; \
4638046381 float16x8_t __s1 = __p1; \
4638146382 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
46382 float16_t __ret; \
4638346383 __ret = (float16_t) __builtin_neon_vmulxh_laneq_f16(__s0, (float16x8_t)__rev1, __p2); \
4638446384 __ret; \
4638546385})
......@@ -46387,19 +46387,19 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4638746387
4638846388#ifdef __LITTLE_ENDIAN__
4638946389#define vmulxq_laneq_f16(__p0_301, __p1_301, __p2_301) __extension__ ({ \
46390 float16x8_t __ret_301; \
4639046391 float16x8_t __s0_301 = __p0_301; \
4639146392 float16x8_t __s1_301 = __p1_301; \
46392 float16x8_t __ret_301; \
4639346393 __ret_301 = vmulxq_f16(__s0_301, splatq_laneq_f16(__s1_301, __p2_301)); \
4639446394 __ret_301; \
4639546395})
4639646396#else
4639746397#define vmulxq_laneq_f16(__p0_302, __p1_302, __p2_302) __extension__ ({ \
46398 float16x8_t __ret_302; \
4639846399 float16x8_t __s0_302 = __p0_302; \
4639946400 float16x8_t __s1_302 = __p1_302; \
4640046401 float16x8_t __rev0_302; __rev0_302 = __builtin_shufflevector(__s0_302, __s0_302, 7, 6, 5, 4, 3, 2, 1, 0); \
4640146402 float16x8_t __rev1_302; __rev1_302 = __builtin_shufflevector(__s1_302, __s1_302, 7, 6, 5, 4, 3, 2, 1, 0); \
46402 float16x8_t __ret_302; \
4640346403 __ret_302 = __noswap_vmulxq_f16(__rev0_302, __noswap_splatq_laneq_f16(__rev1_302, __p2_302)); \
4640446404 __ret_302 = __builtin_shufflevector(__ret_302, __ret_302, 7, 6, 5, 4, 3, 2, 1, 0); \
4640546405 __ret_302; \
......@@ -46408,19 +46408,19 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4640846408
4640946409#ifdef __LITTLE_ENDIAN__
4641046410#define vmulx_laneq_f16(__p0_303, __p1_303, __p2_303) __extension__ ({ \
46411 float16x4_t __ret_303; \
4641146412 float16x4_t __s0_303 = __p0_303; \
4641246413 float16x8_t __s1_303 = __p1_303; \
46413 float16x4_t __ret_303; \
4641446414 __ret_303 = vmulx_f16(__s0_303, splat_laneq_f16(__s1_303, __p2_303)); \
4641546415 __ret_303; \
4641646416})
4641746417#else
4641846418#define vmulx_laneq_f16(__p0_304, __p1_304, __p2_304) __extension__ ({ \
46419 float16x4_t __ret_304; \
4641946420 float16x4_t __s0_304 = __p0_304; \
4642046421 float16x8_t __s1_304 = __p1_304; \
4642146422 float16x4_t __rev0_304; __rev0_304 = __builtin_shufflevector(__s0_304, __s0_304, 3, 2, 1, 0); \
4642246423 float16x8_t __rev1_304; __rev1_304 = __builtin_shufflevector(__s1_304, __s1_304, 7, 6, 5, 4, 3, 2, 1, 0); \
46423 float16x4_t __ret_304; \
4642446424 __ret_304 = __noswap_vmulx_f16(__rev0_304, __noswap_splat_laneq_f16(__rev1_304, __p2_304)); \
4642546425 __ret_304 = __builtin_shufflevector(__ret_304, __ret_304, 3, 2, 1, 0); \
4642646426 __ret_304; \
......@@ -46429,18 +46429,18 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4642946429
4643046430#ifdef __LITTLE_ENDIAN__
4643146431#define vmulxq_n_f16(__p0, __p1) __extension__ ({ \
46432 float16x8_t __ret; \
4643246433 float16x8_t __s0 = __p0; \
4643346434 float16_t __s1 = __p1; \
46434 float16x8_t __ret; \
4643546435 __ret = vmulxq_f16(__s0, (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}); \
4643646436 __ret; \
4643746437})
4643846438#else
4643946439#define vmulxq_n_f16(__p0, __p1) __extension__ ({ \
46440 float16x8_t __ret; \
4644046441 float16x8_t __s0 = __p0; \
4644146442 float16_t __s1 = __p1; \
4644246443 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
46443 float16x8_t __ret; \
4644446444 __ret = __noswap_vmulxq_f16(__rev0, (float16x8_t) {__s1, __s1, __s1, __s1, __s1, __s1, __s1, __s1}); \
4644546445 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
4644646446 __ret; \
......@@ -46449,18 +46449,18 @@ __ai float16x4_t __noswap_vmulx_f16(float16x4_t __p0, float16x4_t __p1) {
4644946449
4645046450#ifdef __LITTLE_ENDIAN__
4645146451#define vmulx_n_f16(__p0, __p1) __extension__ ({ \
46452 float16x4_t __ret; \
4645246453 float16x4_t __s0 = __p0; \
4645346454 float16_t __s1 = __p1; \
46454 float16x4_t __ret; \
4645546455 __ret = vmulx_f16(__s0, (float16x4_t) {__s1, __s1, __s1, __s1}); \
4645646456 __ret; \
4645746457})
4645846458#else
4645946459#define vmulx_n_f16(__p0, __p1) __extension__ ({ \
46460 float16x4_t __ret; \
4646046461 float16x4_t __s0 = __p0; \
4646146462 float16_t __s1 = __p1; \
4646246463 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
46463 float16x4_t __ret; \
4646446464 __ret = __noswap_vmulx_f16(__rev0, (float16x4_t) {__s1, __s1, __s1, __s1}); \
4646546465 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
4646646466 __ret; \
......@@ -46475,9 +46475,9 @@ __ai float16x8_t vpaddq_f16(float16x8_t __p0, float16x8_t __p1) {
4647546475}
4647646476#else
4647746477__ai float16x8_t vpaddq_f16(float16x8_t __p0, float16x8_t __p1) {
46478 float16x8_t __ret;
4647846479 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4647946480 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46480 float16x8_t __ret;
4648146481 __ret = (float16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4648246482 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4648346483 return __ret;
......@@ -46492,9 +46492,9 @@ __ai float16x8_t vpmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
4649246492}
4649346493#else
4649446494__ai float16x8_t vpmaxq_f16(float16x8_t __p0, float16x8_t __p1) {
46495 float16x8_t __ret;
4649546496 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4649646497 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46497 float16x8_t __ret;
4649846498 __ret = (float16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4649946499 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4650046500 return __ret;
......@@ -46509,9 +46509,9 @@ __ai float16x8_t vpmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
4650946509}
4651046510#else
4651146511__ai float16x8_t vpmaxnmq_f16(float16x8_t __p0, float16x8_t __p1) {
46512 float16x8_t __ret;
4651246513 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4651346514 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46514 float16x8_t __ret;
4651546515 __ret = (float16x8_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4651646516 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4651746517 return __ret;
......@@ -46526,9 +46526,9 @@ __ai float16x4_t vpmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
4652646526}
4652746527#else
4652846528__ai float16x4_t vpmaxnm_f16(float16x4_t __p0, float16x4_t __p1) {
46529 float16x4_t __ret;
4652946530 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4653046531 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46531 float16x4_t __ret;
4653246532 __ret = (float16x4_t) __builtin_neon_vpmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4653346533 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4653446534 return __ret;
......@@ -46543,9 +46543,9 @@ __ai float16x8_t vpminq_f16(float16x8_t __p0, float16x8_t __p1) {
4654346543}
4654446544#else
4654546545__ai float16x8_t vpminq_f16(float16x8_t __p0, float16x8_t __p1) {
46546 float16x8_t __ret;
4654646547 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4654746548 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46548 float16x8_t __ret;
4654946549 __ret = (float16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4655046550 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4655146551 return __ret;
......@@ -46560,9 +46560,9 @@ __ai float16x8_t vpminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
4656046560}
4656146561#else
4656246562__ai float16x8_t vpminnmq_f16(float16x8_t __p0, float16x8_t __p1) {
46563 float16x8_t __ret;
4656346564 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4656446565 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46565 float16x8_t __ret;
4656646566 __ret = (float16x8_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 40);
4656746567 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4656846568 return __ret;
......@@ -46577,9 +46577,9 @@ __ai float16x4_t vpminnm_f16(float16x4_t __p0, float16x4_t __p1) {
4657746577}
4657846578#else
4657946579__ai float16x4_t vpminnm_f16(float16x4_t __p0, float16x4_t __p1) {
46580 float16x4_t __ret;
4658046581 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4658146582 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46582 float16x4_t __ret;
4658346583 __ret = (float16x4_t) __builtin_neon_vpminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 8);
4658446584 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4658546585 return __ret;
......@@ -46594,8 +46594,8 @@ __ai float16x8_t vrndiq_f16(float16x8_t __p0) {
4659446594}
4659546595#else
4659646596__ai float16x8_t vrndiq_f16(float16x8_t __p0) {
46597 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4659846597 float16x8_t __ret;
46598 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4659946599 __ret = (float16x8_t) __builtin_neon_vrndiq_v((int8x16_t)__rev0, 40);
4660046600 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4660146601 return __ret;
......@@ -46610,8 +46610,8 @@ __ai float16x4_t vrndi_f16(float16x4_t __p0) {
4661046610}
4661146611#else
4661246612__ai float16x4_t vrndi_f16(float16x4_t __p0) {
46613 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4661446613 float16x4_t __ret;
46614 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4661546615 __ret = (float16x4_t) __builtin_neon_vrndi_v((int8x8_t)__rev0, 8);
4661646616 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4661746617 return __ret;
......@@ -46626,8 +46626,8 @@ __ai float16x8_t vsqrtq_f16(float16x8_t __p0) {
4662646626}
4662746627#else
4662846628__ai float16x8_t vsqrtq_f16(float16x8_t __p0) {
46629 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4663046629 float16x8_t __ret;
46630 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4663146631 __ret = (float16x8_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 40);
4663246632 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4663346633 return __ret;
......@@ -46642,8 +46642,8 @@ __ai float16x4_t vsqrt_f16(float16x4_t __p0) {
4664246642}
4664346643#else
4664446644__ai float16x4_t vsqrt_f16(float16x4_t __p0) {
46645 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4664646645 float16x4_t __ret;
46646 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4664746647 __ret = (float16x4_t) __builtin_neon_vsqrt_v((int8x8_t)__rev0, 8);
4664846648 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4664946649 return __ret;
......@@ -46658,9 +46658,9 @@ __ai float16x8_t vtrn1q_f16(float16x8_t __p0, float16x8_t __p1) {
4665846658}
4665946659#else
4666046660__ai float16x8_t vtrn1q_f16(float16x8_t __p0, float16x8_t __p1) {
46661 float16x8_t __ret;
4666146662 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4666246663 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46663 float16x8_t __ret;
4666446664 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
4666546665 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4666646666 return __ret;
......@@ -46675,9 +46675,9 @@ __ai float16x4_t vtrn1_f16(float16x4_t __p0, float16x4_t __p1) {
4667546675}
4667646676#else
4667746677__ai float16x4_t vtrn1_f16(float16x4_t __p0, float16x4_t __p1) {
46678 float16x4_t __ret;
4667846679 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4667946680 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46680 float16x4_t __ret;
4668146681 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
4668246682 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4668346683 return __ret;
......@@ -46692,9 +46692,9 @@ __ai float16x8_t vtrn2q_f16(float16x8_t __p0, float16x8_t __p1) {
4669246692}
4669346693#else
4669446694__ai float16x8_t vtrn2q_f16(float16x8_t __p0, float16x8_t __p1) {
46695 float16x8_t __ret;
4669546696 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4669646697 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46697 float16x8_t __ret;
4669846698 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
4669946699 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4670046700 return __ret;
......@@ -46709,9 +46709,9 @@ __ai float16x4_t vtrn2_f16(float16x4_t __p0, float16x4_t __p1) {
4670946709}
4671046710#else
4671146711__ai float16x4_t vtrn2_f16(float16x4_t __p0, float16x4_t __p1) {
46712 float16x4_t __ret;
4671246713 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4671346714 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46714 float16x4_t __ret;
4671546715 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
4671646716 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4671746717 return __ret;
......@@ -46726,9 +46726,9 @@ __ai float16x8_t vuzp1q_f16(float16x8_t __p0, float16x8_t __p1) {
4672646726}
4672746727#else
4672846728__ai float16x8_t vuzp1q_f16(float16x8_t __p0, float16x8_t __p1) {
46729 float16x8_t __ret;
4672946730 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4673046731 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46731 float16x8_t __ret;
4673246732 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
4673346733 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4673446734 return __ret;
......@@ -46743,9 +46743,9 @@ __ai float16x4_t vuzp1_f16(float16x4_t __p0, float16x4_t __p1) {
4674346743}
4674446744#else
4674546745__ai float16x4_t vuzp1_f16(float16x4_t __p0, float16x4_t __p1) {
46746 float16x4_t __ret;
4674646747 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4674746748 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46748 float16x4_t __ret;
4674946749 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
4675046750 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4675146751 return __ret;
......@@ -46760,9 +46760,9 @@ __ai float16x8_t vuzp2q_f16(float16x8_t __p0, float16x8_t __p1) {
4676046760}
4676146761#else
4676246762__ai float16x8_t vuzp2q_f16(float16x8_t __p0, float16x8_t __p1) {
46763 float16x8_t __ret;
4676346764 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4676446765 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46765 float16x8_t __ret;
4676646766 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
4676746767 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4676846768 return __ret;
......@@ -46777,9 +46777,9 @@ __ai float16x4_t vuzp2_f16(float16x4_t __p0, float16x4_t __p1) {
4677746777}
4677846778#else
4677946779__ai float16x4_t vuzp2_f16(float16x4_t __p0, float16x4_t __p1) {
46780 float16x4_t __ret;
4678046781 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4678146782 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46782 float16x4_t __ret;
4678346783 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
4678446784 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4678546785 return __ret;
......@@ -46794,9 +46794,9 @@ __ai float16x8_t vzip1q_f16(float16x8_t __p0, float16x8_t __p1) {
4679446794}
4679546795#else
4679646796__ai float16x8_t vzip1q_f16(float16x8_t __p0, float16x8_t __p1) {
46797 float16x8_t __ret;
4679746798 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4679846799 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46799 float16x8_t __ret;
4680046800 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
4680146801 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4680246802 return __ret;
......@@ -46811,9 +46811,9 @@ __ai float16x4_t vzip1_f16(float16x4_t __p0, float16x4_t __p1) {
4681146811}
4681246812#else
4681346813__ai float16x4_t vzip1_f16(float16x4_t __p0, float16x4_t __p1) {
46814 float16x4_t __ret;
4681446815 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4681546816 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46816 float16x4_t __ret;
4681746817 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
4681846818 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4681946819 return __ret;
......@@ -46828,9 +46828,9 @@ __ai float16x8_t vzip2q_f16(float16x8_t __p0, float16x8_t __p1) {
4682846828}
4682946829#else
4683046830__ai float16x8_t vzip2q_f16(float16x8_t __p0, float16x8_t __p1) {
46831 float16x8_t __ret;
4683146832 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4683246833 float16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
46833 float16x8_t __ret;
4683446834 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
4683546835 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4683646836 return __ret;
......@@ -46845,9 +46845,9 @@ __ai float16x4_t vzip2_f16(float16x4_t __p0, float16x4_t __p1) {
4684546845}
4684646846#else
4684746847__ai float16x4_t vzip2_f16(float16x4_t __p0, float16x4_t __p1) {
46848 float16x4_t __ret;
4684846849 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4684946850 float16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46850 float16x4_t __ret;
4685146851 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
4685246852 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4685346853 return __ret;
......@@ -46864,10 +46864,10 @@ __ai uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
4686446864}
4686546865#else
4686646866__ai uint32x4_t vmmlaq_u32(uint32x4_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
46867 uint32x4_t __ret;
4686746868 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4686846869 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4686946870 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46870 uint32x4_t __ret;
4687146871 __ret = (uint32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 50);
4687246872 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4687346873 return __ret;
......@@ -46882,10 +46882,10 @@ __ai int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
4688246882}
4688346883#else
4688446884__ai int32x4_t vmmlaq_s32(int32x4_t __p0, int8x16_t __p1, int8x16_t __p2) {
46885 int32x4_t __ret;
4688546886 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4688646887 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4688746888 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46888 int32x4_t __ret;
4688946889 __ret = (int32x4_t) __builtin_neon_vmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
4689046890 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4689146891 return __ret;
......@@ -46900,10 +46900,10 @@ __ai int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
4690046900}
4690146901#else
4690246902__ai int32x4_t vusdotq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
46903 int32x4_t __ret;
4690346904 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4690446905 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4690546906 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
46906 int32x4_t __ret;
4690746907 __ret = (int32x4_t) __builtin_neon_vusdotq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
4690846908 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4690946909 return __ret;
......@@ -46923,10 +46923,10 @@ __ai int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
4692346923}
4692446924#else
4692546925__ai int32x2_t vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2) {
46926 int32x2_t __ret;
4692646927 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4692746928 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4692846929 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
46929 int32x2_t __ret;
4693046930 __ret = (int32x2_t) __builtin_neon_vusdot_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
4693146931 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4693246932 return __ret;
......@@ -46940,23 +46940,23 @@ __ai int32x2_t __noswap_vusdot_s32(int32x2_t __p0, uint8x8_t __p1, int8x8_t __p2
4694046940
4694146941#ifdef __LITTLE_ENDIAN__
4694246942#define vusdotq_lane_s32(__p0_305, __p1_305, __p2_305, __p3_305) __extension__ ({ \
46943 int32x4_t __ret_305; \
4694346944 int32x4_t __s0_305 = __p0_305; \
4694446945 uint8x16_t __s1_305 = __p1_305; \
4694546946 int8x8_t __s2_305 = __p2_305; \
46946 int32x4_t __ret_305; \
4694746947int8x8_t __reint_305 = __s2_305; \
4694846948 __ret_305 = vusdotq_s32(__s0_305, __s1_305, (int8x16_t)(splatq_lane_s32(*(int32x2_t *) &__reint_305, __p3_305))); \
4694946949 __ret_305; \
4695046950})
4695146951#else
4695246952#define vusdotq_lane_s32(__p0_306, __p1_306, __p2_306, __p3_306) __extension__ ({ \
46953 int32x4_t __ret_306; \
4695346954 int32x4_t __s0_306 = __p0_306; \
4695446955 uint8x16_t __s1_306 = __p1_306; \
4695546956 int8x8_t __s2_306 = __p2_306; \
4695646957 int32x4_t __rev0_306; __rev0_306 = __builtin_shufflevector(__s0_306, __s0_306, 3, 2, 1, 0); \
4695746958 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); \
4695846959 int8x8_t __rev2_306; __rev2_306 = __builtin_shufflevector(__s2_306, __s2_306, 7, 6, 5, 4, 3, 2, 1, 0); \
46959 int32x4_t __ret_306; \
4696046960int8x8_t __reint_306 = __rev2_306; \
4696146961 __ret_306 = __noswap_vusdotq_s32(__rev0_306, __rev1_306, (int8x16_t)(__noswap_splatq_lane_s32(*(int32x2_t *) &__reint_306, __p3_306))); \
4696246962 __ret_306 = __builtin_shufflevector(__ret_306, __ret_306, 3, 2, 1, 0); \
......@@ -46966,23 +46966,23 @@ int8x8_t __reint_306 = __rev2_306; \
4696646966
4696746967#ifdef __LITTLE_ENDIAN__
4696846968#define vusdot_lane_s32(__p0_307, __p1_307, __p2_307, __p3_307) __extension__ ({ \
46969 int32x2_t __ret_307; \
4696946970 int32x2_t __s0_307 = __p0_307; \
4697046971 uint8x8_t __s1_307 = __p1_307; \
4697146972 int8x8_t __s2_307 = __p2_307; \
46972 int32x2_t __ret_307; \
4697346973int8x8_t __reint_307 = __s2_307; \
4697446974 __ret_307 = vusdot_s32(__s0_307, __s1_307, (int8x8_t)(splat_lane_s32(*(int32x2_t *) &__reint_307, __p3_307))); \
4697546975 __ret_307; \
4697646976})
4697746977#else
4697846978#define vusdot_lane_s32(__p0_308, __p1_308, __p2_308, __p3_308) __extension__ ({ \
46979 int32x2_t __ret_308; \
4697946980 int32x2_t __s0_308 = __p0_308; \
4698046981 uint8x8_t __s1_308 = __p1_308; \
4698146982 int8x8_t __s2_308 = __p2_308; \
4698246983 int32x2_t __rev0_308; __rev0_308 = __builtin_shufflevector(__s0_308, __s0_308, 1, 0); \
4698346984 uint8x8_t __rev1_308; __rev1_308 = __builtin_shufflevector(__s1_308, __s1_308, 7, 6, 5, 4, 3, 2, 1, 0); \
4698446985 int8x8_t __rev2_308; __rev2_308 = __builtin_shufflevector(__s2_308, __s2_308, 7, 6, 5, 4, 3, 2, 1, 0); \
46985 int32x2_t __ret_308; \
4698646986int8x8_t __reint_308 = __rev2_308; \
4698746987 __ret_308 = __noswap_vusdot_s32(__rev0_308, __rev1_308, (int8x8_t)(__noswap_splat_lane_s32(*(int32x2_t *) &__reint_308, __p3_308))); \
4698846988 __ret_308 = __builtin_shufflevector(__ret_308, __ret_308, 1, 0); \
......@@ -46998,10 +46998,10 @@ __ai int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
4699846998}
4699946999#else
4700047000__ai int32x4_t vusmmlaq_s32(int32x4_t __p0, uint8x16_t __p1, int8x16_t __p2) {
47001 int32x4_t __ret;
4700147002 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4700247003 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4700347004 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
47004 int32x4_t __ret;
4700547005 __ret = (int32x4_t) __builtin_neon_vusmmlaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
4700647006 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4700747007 return __ret;
......@@ -47018,10 +47018,10 @@ __ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
4701847018}
4701947019#else
4702047020__ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47021 int32x4_t __ret;
4702147022 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4702247023 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4702347024 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47024 int32x4_t __ret;
4702547025 __ret = (int32x4_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
4702647026 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4702747027 return __ret;
......@@ -47041,10 +47041,10 @@ __ai int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
4704147041}
4704247042#else
4704347043__ai int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47044 int16x8_t __ret;
4704447045 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4704547046 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4704647047 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47047 int16x8_t __ret;
4704847048 __ret = (int16x8_t) __builtin_neon_vqrdmlahq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
4704947049 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4705047050 return __ret;
......@@ -47064,10 +47064,10 @@ __ai int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
4706447064}
4706547065#else
4706647066__ai int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47067 int32x2_t __ret;
4706747068 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4706847069 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4706947070 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47070 int32x2_t __ret;
4707147071 __ret = (int32x2_t) __builtin_neon_vqrdmlah_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
4707247072 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4707347073 return __ret;
......@@ -47087,10 +47087,10 @@ __ai int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
4708747087}
4708847088#else
4708947089__ai int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47090 int16x4_t __ret;
4709047091 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4709147092 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4709247093 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47093 int16x4_t __ret;
4709447094 __ret = (int16x4_t) __builtin_neon_vqrdmlah_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
4709547095 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4709647096 return __ret;
......@@ -47104,22 +47104,22 @@ __ai int16x4_t __noswap_vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4710447104
4710547105#ifdef __LITTLE_ENDIAN__
4710647106#define vqrdmlahq_lane_s32(__p0_309, __p1_309, __p2_309, __p3_309) __extension__ ({ \
47107 int32x4_t __ret_309; \
4710747108 int32x4_t __s0_309 = __p0_309; \
4710847109 int32x4_t __s1_309 = __p1_309; \
4710947110 int32x2_t __s2_309 = __p2_309; \
47110 int32x4_t __ret_309; \
4711147111 __ret_309 = vqrdmlahq_s32(__s0_309, __s1_309, splatq_lane_s32(__s2_309, __p3_309)); \
4711247112 __ret_309; \
4711347113})
4711447114#else
4711547115#define vqrdmlahq_lane_s32(__p0_310, __p1_310, __p2_310, __p3_310) __extension__ ({ \
47116 int32x4_t __ret_310; \
4711647117 int32x4_t __s0_310 = __p0_310; \
4711747118 int32x4_t __s1_310 = __p1_310; \
4711847119 int32x2_t __s2_310 = __p2_310; \
4711947120 int32x4_t __rev0_310; __rev0_310 = __builtin_shufflevector(__s0_310, __s0_310, 3, 2, 1, 0); \
4712047121 int32x4_t __rev1_310; __rev1_310 = __builtin_shufflevector(__s1_310, __s1_310, 3, 2, 1, 0); \
4712147122 int32x2_t __rev2_310; __rev2_310 = __builtin_shufflevector(__s2_310, __s2_310, 1, 0); \
47122 int32x4_t __ret_310; \
4712347123 __ret_310 = __noswap_vqrdmlahq_s32(__rev0_310, __rev1_310, __noswap_splatq_lane_s32(__rev2_310, __p3_310)); \
4712447124 __ret_310 = __builtin_shufflevector(__ret_310, __ret_310, 3, 2, 1, 0); \
4712547125 __ret_310; \
......@@ -47128,22 +47128,22 @@ __ai int16x4_t __noswap_vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4712847128
4712947129#ifdef __LITTLE_ENDIAN__
4713047130#define vqrdmlahq_lane_s16(__p0_311, __p1_311, __p2_311, __p3_311) __extension__ ({ \
47131 int16x8_t __ret_311; \
4713147132 int16x8_t __s0_311 = __p0_311; \
4713247133 int16x8_t __s1_311 = __p1_311; \
4713347134 int16x4_t __s2_311 = __p2_311; \
47134 int16x8_t __ret_311; \
4713547135 __ret_311 = vqrdmlahq_s16(__s0_311, __s1_311, splatq_lane_s16(__s2_311, __p3_311)); \
4713647136 __ret_311; \
4713747137})
4713847138#else
4713947139#define vqrdmlahq_lane_s16(__p0_312, __p1_312, __p2_312, __p3_312) __extension__ ({ \
47140 int16x8_t __ret_312; \
4714047141 int16x8_t __s0_312 = __p0_312; \
4714147142 int16x8_t __s1_312 = __p1_312; \
4714247143 int16x4_t __s2_312 = __p2_312; \
4714347144 int16x8_t __rev0_312; __rev0_312 = __builtin_shufflevector(__s0_312, __s0_312, 7, 6, 5, 4, 3, 2, 1, 0); \
4714447145 int16x8_t __rev1_312; __rev1_312 = __builtin_shufflevector(__s1_312, __s1_312, 7, 6, 5, 4, 3, 2, 1, 0); \
4714547146 int16x4_t __rev2_312; __rev2_312 = __builtin_shufflevector(__s2_312, __s2_312, 3, 2, 1, 0); \
47146 int16x8_t __ret_312; \
4714747147 __ret_312 = __noswap_vqrdmlahq_s16(__rev0_312, __rev1_312, __noswap_splatq_lane_s16(__rev2_312, __p3_312)); \
4714847148 __ret_312 = __builtin_shufflevector(__ret_312, __ret_312, 7, 6, 5, 4, 3, 2, 1, 0); \
4714947149 __ret_312; \
......@@ -47152,22 +47152,22 @@ __ai int16x4_t __noswap_vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4715247152
4715347153#ifdef __LITTLE_ENDIAN__
4715447154#define vqrdmlah_lane_s32(__p0_313, __p1_313, __p2_313, __p3_313) __extension__ ({ \
47155 int32x2_t __ret_313; \
4715547156 int32x2_t __s0_313 = __p0_313; \
4715647157 int32x2_t __s1_313 = __p1_313; \
4715747158 int32x2_t __s2_313 = __p2_313; \
47158 int32x2_t __ret_313; \
4715947159 __ret_313 = vqrdmlah_s32(__s0_313, __s1_313, splat_lane_s32(__s2_313, __p3_313)); \
4716047160 __ret_313; \
4716147161})
4716247162#else
4716347163#define vqrdmlah_lane_s32(__p0_314, __p1_314, __p2_314, __p3_314) __extension__ ({ \
47164 int32x2_t __ret_314; \
4716447165 int32x2_t __s0_314 = __p0_314; \
4716547166 int32x2_t __s1_314 = __p1_314; \
4716647167 int32x2_t __s2_314 = __p2_314; \
4716747168 int32x2_t __rev0_314; __rev0_314 = __builtin_shufflevector(__s0_314, __s0_314, 1, 0); \
4716847169 int32x2_t __rev1_314; __rev1_314 = __builtin_shufflevector(__s1_314, __s1_314, 1, 0); \
4716947170 int32x2_t __rev2_314; __rev2_314 = __builtin_shufflevector(__s2_314, __s2_314, 1, 0); \
47170 int32x2_t __ret_314; \
4717147171 __ret_314 = __noswap_vqrdmlah_s32(__rev0_314, __rev1_314, __noswap_splat_lane_s32(__rev2_314, __p3_314)); \
4717247172 __ret_314 = __builtin_shufflevector(__ret_314, __ret_314, 1, 0); \
4717347173 __ret_314; \
......@@ -47176,22 +47176,22 @@ __ai int16x4_t __noswap_vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4717647176
4717747177#ifdef __LITTLE_ENDIAN__
4717847178#define vqrdmlah_lane_s16(__p0_315, __p1_315, __p2_315, __p3_315) __extension__ ({ \
47179 int16x4_t __ret_315; \
4717947180 int16x4_t __s0_315 = __p0_315; \
4718047181 int16x4_t __s1_315 = __p1_315; \
4718147182 int16x4_t __s2_315 = __p2_315; \
47182 int16x4_t __ret_315; \
4718347183 __ret_315 = vqrdmlah_s16(__s0_315, __s1_315, splat_lane_s16(__s2_315, __p3_315)); \
4718447184 __ret_315; \
4718547185})
4718647186#else
4718747187#define vqrdmlah_lane_s16(__p0_316, __p1_316, __p2_316, __p3_316) __extension__ ({ \
47188 int16x4_t __ret_316; \
4718847189 int16x4_t __s0_316 = __p0_316; \
4718947190 int16x4_t __s1_316 = __p1_316; \
4719047191 int16x4_t __s2_316 = __p2_316; \
4719147192 int16x4_t __rev0_316; __rev0_316 = __builtin_shufflevector(__s0_316, __s0_316, 3, 2, 1, 0); \
4719247193 int16x4_t __rev1_316; __rev1_316 = __builtin_shufflevector(__s1_316, __s1_316, 3, 2, 1, 0); \
4719347194 int16x4_t __rev2_316; __rev2_316 = __builtin_shufflevector(__s2_316, __s2_316, 3, 2, 1, 0); \
47194 int16x4_t __ret_316; \
4719547195 __ret_316 = __noswap_vqrdmlah_s16(__rev0_316, __rev1_316, __noswap_splat_lane_s16(__rev2_316, __p3_316)); \
4719647196 __ret_316 = __builtin_shufflevector(__ret_316, __ret_316, 3, 2, 1, 0); \
4719747197 __ret_316; \
......@@ -47206,10 +47206,10 @@ __ai int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
4720647206}
4720747207#else
4720847208__ai int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47209 int32x4_t __ret;
4720947210 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4721047211 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4721147212 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47212 int32x4_t __ret;
4721347213 __ret = (int32x4_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 34);
4721447214 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4721547215 return __ret;
......@@ -47229,10 +47229,10 @@ __ai int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
4722947229}
4723047230#else
4723147231__ai int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47232 int16x8_t __ret;
4723247233 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4723347234 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4723447235 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47235 int16x8_t __ret;
4723647236 __ret = (int16x8_t) __builtin_neon_vqrdmlshq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 33);
4723747237 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4723847238 return __ret;
......@@ -47252,10 +47252,10 @@ __ai int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
4725247252}
4725347253#else
4725447254__ai int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
47255 int32x2_t __ret;
4725547256 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4725647257 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4725747258 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47258 int32x2_t __ret;
4725947259 __ret = (int32x2_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 2);
4726047260 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4726147261 return __ret;
......@@ -47275,10 +47275,10 @@ __ai int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
4727547275}
4727647276#else
4727747277__ai int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
47278 int16x4_t __ret;
4727847279 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4727947280 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4728047281 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47281 int16x4_t __ret;
4728247282 __ret = (int16x4_t) __builtin_neon_vqrdmlsh_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, 1);
4728347283 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4728447284 return __ret;
......@@ -47292,22 +47292,22 @@ __ai int16x4_t __noswap_vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4729247292
4729347293#ifdef __LITTLE_ENDIAN__
4729447294#define vqrdmlshq_lane_s32(__p0_317, __p1_317, __p2_317, __p3_317) __extension__ ({ \
47295 int32x4_t __ret_317; \
4729547296 int32x4_t __s0_317 = __p0_317; \
4729647297 int32x4_t __s1_317 = __p1_317; \
4729747298 int32x2_t __s2_317 = __p2_317; \
47298 int32x4_t __ret_317; \
4729947299 __ret_317 = vqrdmlshq_s32(__s0_317, __s1_317, splatq_lane_s32(__s2_317, __p3_317)); \
4730047300 __ret_317; \
4730147301})
4730247302#else
4730347303#define vqrdmlshq_lane_s32(__p0_318, __p1_318, __p2_318, __p3_318) __extension__ ({ \
47304 int32x4_t __ret_318; \
4730447305 int32x4_t __s0_318 = __p0_318; \
4730547306 int32x4_t __s1_318 = __p1_318; \
4730647307 int32x2_t __s2_318 = __p2_318; \
4730747308 int32x4_t __rev0_318; __rev0_318 = __builtin_shufflevector(__s0_318, __s0_318, 3, 2, 1, 0); \
4730847309 int32x4_t __rev1_318; __rev1_318 = __builtin_shufflevector(__s1_318, __s1_318, 3, 2, 1, 0); \
4730947310 int32x2_t __rev2_318; __rev2_318 = __builtin_shufflevector(__s2_318, __s2_318, 1, 0); \
47310 int32x4_t __ret_318; \
4731147311 __ret_318 = __noswap_vqrdmlshq_s32(__rev0_318, __rev1_318, __noswap_splatq_lane_s32(__rev2_318, __p3_318)); \
4731247312 __ret_318 = __builtin_shufflevector(__ret_318, __ret_318, 3, 2, 1, 0); \
4731347313 __ret_318; \
......@@ -47316,22 +47316,22 @@ __ai int16x4_t __noswap_vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4731647316
4731747317#ifdef __LITTLE_ENDIAN__
4731847318#define vqrdmlshq_lane_s16(__p0_319, __p1_319, __p2_319, __p3_319) __extension__ ({ \
47319 int16x8_t __ret_319; \
4731947320 int16x8_t __s0_319 = __p0_319; \
4732047321 int16x8_t __s1_319 = __p1_319; \
4732147322 int16x4_t __s2_319 = __p2_319; \
47322 int16x8_t __ret_319; \
4732347323 __ret_319 = vqrdmlshq_s16(__s0_319, __s1_319, splatq_lane_s16(__s2_319, __p3_319)); \
4732447324 __ret_319; \
4732547325})
4732647326#else
4732747327#define vqrdmlshq_lane_s16(__p0_320, __p1_320, __p2_320, __p3_320) __extension__ ({ \
47328 int16x8_t __ret_320; \
4732847329 int16x8_t __s0_320 = __p0_320; \
4732947330 int16x8_t __s1_320 = __p1_320; \
4733047331 int16x4_t __s2_320 = __p2_320; \
4733147332 int16x8_t __rev0_320; __rev0_320 = __builtin_shufflevector(__s0_320, __s0_320, 7, 6, 5, 4, 3, 2, 1, 0); \
4733247333 int16x8_t __rev1_320; __rev1_320 = __builtin_shufflevector(__s1_320, __s1_320, 7, 6, 5, 4, 3, 2, 1, 0); \
4733347334 int16x4_t __rev2_320; __rev2_320 = __builtin_shufflevector(__s2_320, __s2_320, 3, 2, 1, 0); \
47334 int16x8_t __ret_320; \
4733547335 __ret_320 = __noswap_vqrdmlshq_s16(__rev0_320, __rev1_320, __noswap_splatq_lane_s16(__rev2_320, __p3_320)); \
4733647336 __ret_320 = __builtin_shufflevector(__ret_320, __ret_320, 7, 6, 5, 4, 3, 2, 1, 0); \
4733747337 __ret_320; \
......@@ -47340,22 +47340,22 @@ __ai int16x4_t __noswap_vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4734047340
4734147341#ifdef __LITTLE_ENDIAN__
4734247342#define vqrdmlsh_lane_s32(__p0_321, __p1_321, __p2_321, __p3_321) __extension__ ({ \
47343 int32x2_t __ret_321; \
4734347344 int32x2_t __s0_321 = __p0_321; \
4734447345 int32x2_t __s1_321 = __p1_321; \
4734547346 int32x2_t __s2_321 = __p2_321; \
47346 int32x2_t __ret_321; \
4734747347 __ret_321 = vqrdmlsh_s32(__s0_321, __s1_321, splat_lane_s32(__s2_321, __p3_321)); \
4734847348 __ret_321; \
4734947349})
4735047350#else
4735147351#define vqrdmlsh_lane_s32(__p0_322, __p1_322, __p2_322, __p3_322) __extension__ ({ \
47352 int32x2_t __ret_322; \
4735247353 int32x2_t __s0_322 = __p0_322; \
4735347354 int32x2_t __s1_322 = __p1_322; \
4735447355 int32x2_t __s2_322 = __p2_322; \
4735547356 int32x2_t __rev0_322; __rev0_322 = __builtin_shufflevector(__s0_322, __s0_322, 1, 0); \
4735647357 int32x2_t __rev1_322; __rev1_322 = __builtin_shufflevector(__s1_322, __s1_322, 1, 0); \
4735747358 int32x2_t __rev2_322; __rev2_322 = __builtin_shufflevector(__s2_322, __s2_322, 1, 0); \
47358 int32x2_t __ret_322; \
4735947359 __ret_322 = __noswap_vqrdmlsh_s32(__rev0_322, __rev1_322, __noswap_splat_lane_s32(__rev2_322, __p3_322)); \
4736047360 __ret_322 = __builtin_shufflevector(__ret_322, __ret_322, 1, 0); \
4736147361 __ret_322; \
......@@ -47364,22 +47364,22 @@ __ai int16x4_t __noswap_vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t _
4736447364
4736547365#ifdef __LITTLE_ENDIAN__
4736647366#define vqrdmlsh_lane_s16(__p0_323, __p1_323, __p2_323, __p3_323) __extension__ ({ \
47367 int16x4_t __ret_323; \
4736747368 int16x4_t __s0_323 = __p0_323; \
4736847369 int16x4_t __s1_323 = __p1_323; \
4736947370 int16x4_t __s2_323 = __p2_323; \
47370 int16x4_t __ret_323; \
4737147371 __ret_323 = vqrdmlsh_s16(__s0_323, __s1_323, splat_lane_s16(__s2_323, __p3_323)); \
4737247372 __ret_323; \
4737347373})
4737447374#else
4737547375#define vqrdmlsh_lane_s16(__p0_324, __p1_324, __p2_324, __p3_324) __extension__ ({ \
47376 int16x4_t __ret_324; \
4737647377 int16x4_t __s0_324 = __p0_324; \
4737747378 int16x4_t __s1_324 = __p1_324; \
4737847379 int16x4_t __s2_324 = __p2_324; \
4737947380 int16x4_t __rev0_324; __rev0_324 = __builtin_shufflevector(__s0_324, __s0_324, 3, 2, 1, 0); \
4738047381 int16x4_t __rev1_324; __rev1_324 = __builtin_shufflevector(__s1_324, __s1_324, 3, 2, 1, 0); \
4738147382 int16x4_t __rev2_324; __rev2_324 = __builtin_shufflevector(__s2_324, __s2_324, 3, 2, 1, 0); \
47382 int16x4_t __ret_324; \
4738347383 __ret_324 = __noswap_vqrdmlsh_s16(__rev0_324, __rev1_324, __noswap_splat_lane_s16(__rev2_324, __p3_324)); \
4738447384 __ret_324 = __builtin_shufflevector(__ret_324, __ret_324, 3, 2, 1, 0); \
4738547385 __ret_324; \
......@@ -47400,20 +47400,20 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4740047400}
4740147401#ifdef __LITTLE_ENDIAN__
4740247402#define vqrdmlahs_lane_s32(__p0_325, __p1_325, __p2_325, __p3_325) __extension__ ({ \
47403 int32_t __ret_325; \
4740347404 int32_t __s0_325 = __p0_325; \
4740447405 int32_t __s1_325 = __p1_325; \
4740547406 int32x2_t __s2_325 = __p2_325; \
47406 int32_t __ret_325; \
4740747407 __ret_325 = vqrdmlahs_s32(__s0_325, __s1_325, vget_lane_s32(__s2_325, __p3_325)); \
4740847408 __ret_325; \
4740947409})
4741047410#else
4741147411#define vqrdmlahs_lane_s32(__p0_326, __p1_326, __p2_326, __p3_326) __extension__ ({ \
47412 int32_t __ret_326; \
4741247413 int32_t __s0_326 = __p0_326; \
4741347414 int32_t __s1_326 = __p1_326; \
4741447415 int32x2_t __s2_326 = __p2_326; \
4741547416 int32x2_t __rev2_326; __rev2_326 = __builtin_shufflevector(__s2_326, __s2_326, 1, 0); \
47416 int32_t __ret_326; \
4741747417 __ret_326 = vqrdmlahs_s32(__s0_326, __s1_326, __noswap_vget_lane_s32(__rev2_326, __p3_326)); \
4741847418 __ret_326; \
4741947419})
......@@ -47421,20 +47421,20 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4742147421
4742247422#ifdef __LITTLE_ENDIAN__
4742347423#define vqrdmlahh_lane_s16(__p0_327, __p1_327, __p2_327, __p3_327) __extension__ ({ \
47424 int16_t __ret_327; \
4742447425 int16_t __s0_327 = __p0_327; \
4742547426 int16_t __s1_327 = __p1_327; \
4742647427 int16x4_t __s2_327 = __p2_327; \
47427 int16_t __ret_327; \
4742847428 __ret_327 = vqrdmlahh_s16(__s0_327, __s1_327, vget_lane_s16(__s2_327, __p3_327)); \
4742947429 __ret_327; \
4743047430})
4743147431#else
4743247432#define vqrdmlahh_lane_s16(__p0_328, __p1_328, __p2_328, __p3_328) __extension__ ({ \
47433 int16_t __ret_328; \
4743347434 int16_t __s0_328 = __p0_328; \
4743447435 int16_t __s1_328 = __p1_328; \
4743547436 int16x4_t __s2_328 = __p2_328; \
4743647437 int16x4_t __rev2_328; __rev2_328 = __builtin_shufflevector(__s2_328, __s2_328, 3, 2, 1, 0); \
47437 int16_t __ret_328; \
4743847438 __ret_328 = vqrdmlahh_s16(__s0_328, __s1_328, __noswap_vget_lane_s16(__rev2_328, __p3_328)); \
4743947439 __ret_328; \
4744047440})
......@@ -47442,20 +47442,20 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4744247442
4744347443#ifdef __LITTLE_ENDIAN__
4744447444#define vqrdmlahs_laneq_s32(__p0_329, __p1_329, __p2_329, __p3_329) __extension__ ({ \
47445 int32_t __ret_329; \
4744547446 int32_t __s0_329 = __p0_329; \
4744647447 int32_t __s1_329 = __p1_329; \
4744747448 int32x4_t __s2_329 = __p2_329; \
47448 int32_t __ret_329; \
4744947449 __ret_329 = vqrdmlahs_s32(__s0_329, __s1_329, vgetq_lane_s32(__s2_329, __p3_329)); \
4745047450 __ret_329; \
4745147451})
4745247452#else
4745347453#define vqrdmlahs_laneq_s32(__p0_330, __p1_330, __p2_330, __p3_330) __extension__ ({ \
47454 int32_t __ret_330; \
4745447455 int32_t __s0_330 = __p0_330; \
4745547456 int32_t __s1_330 = __p1_330; \
4745647457 int32x4_t __s2_330 = __p2_330; \
4745747458 int32x4_t __rev2_330; __rev2_330 = __builtin_shufflevector(__s2_330, __s2_330, 3, 2, 1, 0); \
47458 int32_t __ret_330; \
4745947459 __ret_330 = vqrdmlahs_s32(__s0_330, __s1_330, __noswap_vgetq_lane_s32(__rev2_330, __p3_330)); \
4746047460 __ret_330; \
4746147461})
......@@ -47463,20 +47463,20 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4746347463
4746447464#ifdef __LITTLE_ENDIAN__
4746547465#define vqrdmlahh_laneq_s16(__p0_331, __p1_331, __p2_331, __p3_331) __extension__ ({ \
47466 int16_t __ret_331; \
4746647467 int16_t __s0_331 = __p0_331; \
4746747468 int16_t __s1_331 = __p1_331; \
4746847469 int16x8_t __s2_331 = __p2_331; \
47469 int16_t __ret_331; \
4747047470 __ret_331 = vqrdmlahh_s16(__s0_331, __s1_331, vgetq_lane_s16(__s2_331, __p3_331)); \
4747147471 __ret_331; \
4747247472})
4747347473#else
4747447474#define vqrdmlahh_laneq_s16(__p0_332, __p1_332, __p2_332, __p3_332) __extension__ ({ \
47475 int16_t __ret_332; \
4747547476 int16_t __s0_332 = __p0_332; \
4747647477 int16_t __s1_332 = __p1_332; \
4747747478 int16x8_t __s2_332 = __p2_332; \
4747847479 int16x8_t __rev2_332; __rev2_332 = __builtin_shufflevector(__s2_332, __s2_332, 7, 6, 5, 4, 3, 2, 1, 0); \
47479 int16_t __ret_332; \
4748047480 __ret_332 = vqrdmlahh_s16(__s0_332, __s1_332, __noswap_vgetq_lane_s16(__rev2_332, __p3_332)); \
4748147481 __ret_332; \
4748247482})
......@@ -47484,22 +47484,22 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4748447484
4748547485#ifdef __LITTLE_ENDIAN__
4748647486#define vqrdmlahq_laneq_s32(__p0_333, __p1_333, __p2_333, __p3_333) __extension__ ({ \
47487 int32x4_t __ret_333; \
4748747488 int32x4_t __s0_333 = __p0_333; \
4748847489 int32x4_t __s1_333 = __p1_333; \
4748947490 int32x4_t __s2_333 = __p2_333; \
47490 int32x4_t __ret_333; \
4749147491 __ret_333 = vqrdmlahq_s32(__s0_333, __s1_333, splatq_laneq_s32(__s2_333, __p3_333)); \
4749247492 __ret_333; \
4749347493})
4749447494#else
4749547495#define vqrdmlahq_laneq_s32(__p0_334, __p1_334, __p2_334, __p3_334) __extension__ ({ \
47496 int32x4_t __ret_334; \
4749647497 int32x4_t __s0_334 = __p0_334; \
4749747498 int32x4_t __s1_334 = __p1_334; \
4749847499 int32x4_t __s2_334 = __p2_334; \
4749947500 int32x4_t __rev0_334; __rev0_334 = __builtin_shufflevector(__s0_334, __s0_334, 3, 2, 1, 0); \
4750047501 int32x4_t __rev1_334; __rev1_334 = __builtin_shufflevector(__s1_334, __s1_334, 3, 2, 1, 0); \
4750147502 int32x4_t __rev2_334; __rev2_334 = __builtin_shufflevector(__s2_334, __s2_334, 3, 2, 1, 0); \
47502 int32x4_t __ret_334; \
4750347503 __ret_334 = __noswap_vqrdmlahq_s32(__rev0_334, __rev1_334, __noswap_splatq_laneq_s32(__rev2_334, __p3_334)); \
4750447504 __ret_334 = __builtin_shufflevector(__ret_334, __ret_334, 3, 2, 1, 0); \
4750547505 __ret_334; \
......@@ -47508,22 +47508,22 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4750847508
4750947509#ifdef __LITTLE_ENDIAN__
4751047510#define vqrdmlahq_laneq_s16(__p0_335, __p1_335, __p2_335, __p3_335) __extension__ ({ \
47511 int16x8_t __ret_335; \
4751147512 int16x8_t __s0_335 = __p0_335; \
4751247513 int16x8_t __s1_335 = __p1_335; \
4751347514 int16x8_t __s2_335 = __p2_335; \
47514 int16x8_t __ret_335; \
4751547515 __ret_335 = vqrdmlahq_s16(__s0_335, __s1_335, splatq_laneq_s16(__s2_335, __p3_335)); \
4751647516 __ret_335; \
4751747517})
4751847518#else
4751947519#define vqrdmlahq_laneq_s16(__p0_336, __p1_336, __p2_336, __p3_336) __extension__ ({ \
47520 int16x8_t __ret_336; \
4752047521 int16x8_t __s0_336 = __p0_336; \
4752147522 int16x8_t __s1_336 = __p1_336; \
4752247523 int16x8_t __s2_336 = __p2_336; \
4752347524 int16x8_t __rev0_336; __rev0_336 = __builtin_shufflevector(__s0_336, __s0_336, 7, 6, 5, 4, 3, 2, 1, 0); \
4752447525 int16x8_t __rev1_336; __rev1_336 = __builtin_shufflevector(__s1_336, __s1_336, 7, 6, 5, 4, 3, 2, 1, 0); \
4752547526 int16x8_t __rev2_336; __rev2_336 = __builtin_shufflevector(__s2_336, __s2_336, 7, 6, 5, 4, 3, 2, 1, 0); \
47526 int16x8_t __ret_336; \
4752747527 __ret_336 = __noswap_vqrdmlahq_s16(__rev0_336, __rev1_336, __noswap_splatq_laneq_s16(__rev2_336, __p3_336)); \
4752847528 __ret_336 = __builtin_shufflevector(__ret_336, __ret_336, 7, 6, 5, 4, 3, 2, 1, 0); \
4752947529 __ret_336; \
......@@ -47532,22 +47532,22 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4753247532
4753347533#ifdef __LITTLE_ENDIAN__
4753447534#define vqrdmlah_laneq_s32(__p0_337, __p1_337, __p2_337, __p3_337) __extension__ ({ \
47535 int32x2_t __ret_337; \
4753547536 int32x2_t __s0_337 = __p0_337; \
4753647537 int32x2_t __s1_337 = __p1_337; \
4753747538 int32x4_t __s2_337 = __p2_337; \
47538 int32x2_t __ret_337; \
4753947539 __ret_337 = vqrdmlah_s32(__s0_337, __s1_337, splat_laneq_s32(__s2_337, __p3_337)); \
4754047540 __ret_337; \
4754147541})
4754247542#else
4754347543#define vqrdmlah_laneq_s32(__p0_338, __p1_338, __p2_338, __p3_338) __extension__ ({ \
47544 int32x2_t __ret_338; \
4754447545 int32x2_t __s0_338 = __p0_338; \
4754547546 int32x2_t __s1_338 = __p1_338; \
4754647547 int32x4_t __s2_338 = __p2_338; \
4754747548 int32x2_t __rev0_338; __rev0_338 = __builtin_shufflevector(__s0_338, __s0_338, 1, 0); \
4754847549 int32x2_t __rev1_338; __rev1_338 = __builtin_shufflevector(__s1_338, __s1_338, 1, 0); \
4754947550 int32x4_t __rev2_338; __rev2_338 = __builtin_shufflevector(__s2_338, __s2_338, 3, 2, 1, 0); \
47550 int32x2_t __ret_338; \
4755147551 __ret_338 = __noswap_vqrdmlah_s32(__rev0_338, __rev1_338, __noswap_splat_laneq_s32(__rev2_338, __p3_338)); \
4755247552 __ret_338 = __builtin_shufflevector(__ret_338, __ret_338, 1, 0); \
4755347553 __ret_338; \
......@@ -47556,22 +47556,22 @@ __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4755647556
4755747557#ifdef __LITTLE_ENDIAN__
4755847558#define vqrdmlah_laneq_s16(__p0_339, __p1_339, __p2_339, __p3_339) __extension__ ({ \
47559 int16x4_t __ret_339; \
4755947560 int16x4_t __s0_339 = __p0_339; \
4756047561 int16x4_t __s1_339 = __p1_339; \
4756147562 int16x8_t __s2_339 = __p2_339; \
47562 int16x4_t __ret_339; \
4756347563 __ret_339 = vqrdmlah_s16(__s0_339, __s1_339, splat_laneq_s16(__s2_339, __p3_339)); \
4756447564 __ret_339; \
4756547565})
4756647566#else
4756747567#define vqrdmlah_laneq_s16(__p0_340, __p1_340, __p2_340, __p3_340) __extension__ ({ \
47568 int16x4_t __ret_340; \
4756847569 int16x4_t __s0_340 = __p0_340; \
4756947570 int16x4_t __s1_340 = __p1_340; \
4757047571 int16x8_t __s2_340 = __p2_340; \
4757147572 int16x4_t __rev0_340; __rev0_340 = __builtin_shufflevector(__s0_340, __s0_340, 3, 2, 1, 0); \
4757247573 int16x4_t __rev1_340; __rev1_340 = __builtin_shufflevector(__s1_340, __s1_340, 3, 2, 1, 0); \
4757347574 int16x8_t __rev2_340; __rev2_340 = __builtin_shufflevector(__s2_340, __s2_340, 7, 6, 5, 4, 3, 2, 1, 0); \
47574 int16x4_t __ret_340; \
4757547575 __ret_340 = __noswap_vqrdmlah_s16(__rev0_340, __rev1_340, __noswap_splat_laneq_s16(__rev2_340, __p3_340)); \
4757647576 __ret_340 = __builtin_shufflevector(__ret_340, __ret_340, 3, 2, 1, 0); \
4757747577 __ret_340; \
......@@ -47590,20 +47590,20 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4759047590}
4759147591#ifdef __LITTLE_ENDIAN__
4759247592#define vqrdmlshs_lane_s32(__p0_341, __p1_341, __p2_341, __p3_341) __extension__ ({ \
47593 int32_t __ret_341; \
4759347594 int32_t __s0_341 = __p0_341; \
4759447595 int32_t __s1_341 = __p1_341; \
4759547596 int32x2_t __s2_341 = __p2_341; \
47596 int32_t __ret_341; \
4759747597 __ret_341 = vqrdmlshs_s32(__s0_341, __s1_341, vget_lane_s32(__s2_341, __p3_341)); \
4759847598 __ret_341; \
4759947599})
4760047600#else
4760147601#define vqrdmlshs_lane_s32(__p0_342, __p1_342, __p2_342, __p3_342) __extension__ ({ \
47602 int32_t __ret_342; \
4760247603 int32_t __s0_342 = __p0_342; \
4760347604 int32_t __s1_342 = __p1_342; \
4760447605 int32x2_t __s2_342 = __p2_342; \
4760547606 int32x2_t __rev2_342; __rev2_342 = __builtin_shufflevector(__s2_342, __s2_342, 1, 0); \
47606 int32_t __ret_342; \
4760747607 __ret_342 = vqrdmlshs_s32(__s0_342, __s1_342, __noswap_vget_lane_s32(__rev2_342, __p3_342)); \
4760847608 __ret_342; \
4760947609})
......@@ -47611,20 +47611,20 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4761147611
4761247612#ifdef __LITTLE_ENDIAN__
4761347613#define vqrdmlshh_lane_s16(__p0_343, __p1_343, __p2_343, __p3_343) __extension__ ({ \
47614 int16_t __ret_343; \
4761447615 int16_t __s0_343 = __p0_343; \
4761547616 int16_t __s1_343 = __p1_343; \
4761647617 int16x4_t __s2_343 = __p2_343; \
47617 int16_t __ret_343; \
4761847618 __ret_343 = vqrdmlshh_s16(__s0_343, __s1_343, vget_lane_s16(__s2_343, __p3_343)); \
4761947619 __ret_343; \
4762047620})
4762147621#else
4762247622#define vqrdmlshh_lane_s16(__p0_344, __p1_344, __p2_344, __p3_344) __extension__ ({ \
47623 int16_t __ret_344; \
4762347624 int16_t __s0_344 = __p0_344; \
4762447625 int16_t __s1_344 = __p1_344; \
4762547626 int16x4_t __s2_344 = __p2_344; \
4762647627 int16x4_t __rev2_344; __rev2_344 = __builtin_shufflevector(__s2_344, __s2_344, 3, 2, 1, 0); \
47627 int16_t __ret_344; \
4762847628 __ret_344 = vqrdmlshh_s16(__s0_344, __s1_344, __noswap_vget_lane_s16(__rev2_344, __p3_344)); \
4762947629 __ret_344; \
4763047630})
......@@ -47632,20 +47632,20 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4763247632
4763347633#ifdef __LITTLE_ENDIAN__
4763447634#define vqrdmlshs_laneq_s32(__p0_345, __p1_345, __p2_345, __p3_345) __extension__ ({ \
47635 int32_t __ret_345; \
4763547636 int32_t __s0_345 = __p0_345; \
4763647637 int32_t __s1_345 = __p1_345; \
4763747638 int32x4_t __s2_345 = __p2_345; \
47638 int32_t __ret_345; \
4763947639 __ret_345 = vqrdmlshs_s32(__s0_345, __s1_345, vgetq_lane_s32(__s2_345, __p3_345)); \
4764047640 __ret_345; \
4764147641})
4764247642#else
4764347643#define vqrdmlshs_laneq_s32(__p0_346, __p1_346, __p2_346, __p3_346) __extension__ ({ \
47644 int32_t __ret_346; \
4764447645 int32_t __s0_346 = __p0_346; \
4764547646 int32_t __s1_346 = __p1_346; \
4764647647 int32x4_t __s2_346 = __p2_346; \
4764747648 int32x4_t __rev2_346; __rev2_346 = __builtin_shufflevector(__s2_346, __s2_346, 3, 2, 1, 0); \
47648 int32_t __ret_346; \
4764947649 __ret_346 = vqrdmlshs_s32(__s0_346, __s1_346, __noswap_vgetq_lane_s32(__rev2_346, __p3_346)); \
4765047650 __ret_346; \
4765147651})
......@@ -47653,20 +47653,20 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4765347653
4765447654#ifdef __LITTLE_ENDIAN__
4765547655#define vqrdmlshh_laneq_s16(__p0_347, __p1_347, __p2_347, __p3_347) __extension__ ({ \
47656 int16_t __ret_347; \
4765647657 int16_t __s0_347 = __p0_347; \
4765747658 int16_t __s1_347 = __p1_347; \
4765847659 int16x8_t __s2_347 = __p2_347; \
47659 int16_t __ret_347; \
4766047660 __ret_347 = vqrdmlshh_s16(__s0_347, __s1_347, vgetq_lane_s16(__s2_347, __p3_347)); \
4766147661 __ret_347; \
4766247662})
4766347663#else
4766447664#define vqrdmlshh_laneq_s16(__p0_348, __p1_348, __p2_348, __p3_348) __extension__ ({ \
47665 int16_t __ret_348; \
4766547666 int16_t __s0_348 = __p0_348; \
4766647667 int16_t __s1_348 = __p1_348; \
4766747668 int16x8_t __s2_348 = __p2_348; \
4766847669 int16x8_t __rev2_348; __rev2_348 = __builtin_shufflevector(__s2_348, __s2_348, 7, 6, 5, 4, 3, 2, 1, 0); \
47669 int16_t __ret_348; \
4767047670 __ret_348 = vqrdmlshh_s16(__s0_348, __s1_348, __noswap_vgetq_lane_s16(__rev2_348, __p3_348)); \
4767147671 __ret_348; \
4767247672})
......@@ -47674,22 +47674,22 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4767447674
4767547675#ifdef __LITTLE_ENDIAN__
4767647676#define vqrdmlshq_laneq_s32(__p0_349, __p1_349, __p2_349, __p3_349) __extension__ ({ \
47677 int32x4_t __ret_349; \
4767747678 int32x4_t __s0_349 = __p0_349; \
4767847679 int32x4_t __s1_349 = __p1_349; \
4767947680 int32x4_t __s2_349 = __p2_349; \
47680 int32x4_t __ret_349; \
4768147681 __ret_349 = vqrdmlshq_s32(__s0_349, __s1_349, splatq_laneq_s32(__s2_349, __p3_349)); \
4768247682 __ret_349; \
4768347683})
4768447684#else
4768547685#define vqrdmlshq_laneq_s32(__p0_350, __p1_350, __p2_350, __p3_350) __extension__ ({ \
47686 int32x4_t __ret_350; \
4768647687 int32x4_t __s0_350 = __p0_350; \
4768747688 int32x4_t __s1_350 = __p1_350; \
4768847689 int32x4_t __s2_350 = __p2_350; \
4768947690 int32x4_t __rev0_350; __rev0_350 = __builtin_shufflevector(__s0_350, __s0_350, 3, 2, 1, 0); \
4769047691 int32x4_t __rev1_350; __rev1_350 = __builtin_shufflevector(__s1_350, __s1_350, 3, 2, 1, 0); \
4769147692 int32x4_t __rev2_350; __rev2_350 = __builtin_shufflevector(__s2_350, __s2_350, 3, 2, 1, 0); \
47692 int32x4_t __ret_350; \
4769347693 __ret_350 = __noswap_vqrdmlshq_s32(__rev0_350, __rev1_350, __noswap_splatq_laneq_s32(__rev2_350, __p3_350)); \
4769447694 __ret_350 = __builtin_shufflevector(__ret_350, __ret_350, 3, 2, 1, 0); \
4769547695 __ret_350; \
......@@ -47698,22 +47698,22 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4769847698
4769947699#ifdef __LITTLE_ENDIAN__
4770047700#define vqrdmlshq_laneq_s16(__p0_351, __p1_351, __p2_351, __p3_351) __extension__ ({ \
47701 int16x8_t __ret_351; \
4770147702 int16x8_t __s0_351 = __p0_351; \
4770247703 int16x8_t __s1_351 = __p1_351; \
4770347704 int16x8_t __s2_351 = __p2_351; \
47704 int16x8_t __ret_351; \
4770547705 __ret_351 = vqrdmlshq_s16(__s0_351, __s1_351, splatq_laneq_s16(__s2_351, __p3_351)); \
4770647706 __ret_351; \
4770747707})
4770847708#else
4770947709#define vqrdmlshq_laneq_s16(__p0_352, __p1_352, __p2_352, __p3_352) __extension__ ({ \
47710 int16x8_t __ret_352; \
4771047711 int16x8_t __s0_352 = __p0_352; \
4771147712 int16x8_t __s1_352 = __p1_352; \
4771247713 int16x8_t __s2_352 = __p2_352; \
4771347714 int16x8_t __rev0_352; __rev0_352 = __builtin_shufflevector(__s0_352, __s0_352, 7, 6, 5, 4, 3, 2, 1, 0); \
4771447715 int16x8_t __rev1_352; __rev1_352 = __builtin_shufflevector(__s1_352, __s1_352, 7, 6, 5, 4, 3, 2, 1, 0); \
4771547716 int16x8_t __rev2_352; __rev2_352 = __builtin_shufflevector(__s2_352, __s2_352, 7, 6, 5, 4, 3, 2, 1, 0); \
47716 int16x8_t __ret_352; \
4771747717 __ret_352 = __noswap_vqrdmlshq_s16(__rev0_352, __rev1_352, __noswap_splatq_laneq_s16(__rev2_352, __p3_352)); \
4771847718 __ret_352 = __builtin_shufflevector(__ret_352, __ret_352, 7, 6, 5, 4, 3, 2, 1, 0); \
4771947719 __ret_352; \
......@@ -47722,22 +47722,22 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4772247722
4772347723#ifdef __LITTLE_ENDIAN__
4772447724#define vqrdmlsh_laneq_s32(__p0_353, __p1_353, __p2_353, __p3_353) __extension__ ({ \
47725 int32x2_t __ret_353; \
4772547726 int32x2_t __s0_353 = __p0_353; \
4772647727 int32x2_t __s1_353 = __p1_353; \
4772747728 int32x4_t __s2_353 = __p2_353; \
47728 int32x2_t __ret_353; \
4772947729 __ret_353 = vqrdmlsh_s32(__s0_353, __s1_353, splat_laneq_s32(__s2_353, __p3_353)); \
4773047730 __ret_353; \
4773147731})
4773247732#else
4773347733#define vqrdmlsh_laneq_s32(__p0_354, __p1_354, __p2_354, __p3_354) __extension__ ({ \
47734 int32x2_t __ret_354; \
4773447735 int32x2_t __s0_354 = __p0_354; \
4773547736 int32x2_t __s1_354 = __p1_354; \
4773647737 int32x4_t __s2_354 = __p2_354; \
4773747738 int32x2_t __rev0_354; __rev0_354 = __builtin_shufflevector(__s0_354, __s0_354, 1, 0); \
4773847739 int32x2_t __rev1_354; __rev1_354 = __builtin_shufflevector(__s1_354, __s1_354, 1, 0); \
4773947740 int32x4_t __rev2_354; __rev2_354 = __builtin_shufflevector(__s2_354, __s2_354, 3, 2, 1, 0); \
47740 int32x2_t __ret_354; \
4774147741 __ret_354 = __noswap_vqrdmlsh_s32(__rev0_354, __rev1_354, __noswap_splat_laneq_s32(__rev2_354, __p3_354)); \
4774247742 __ret_354 = __builtin_shufflevector(__ret_354, __ret_354, 1, 0); \
4774347743 __ret_354; \
......@@ -47746,22 +47746,22 @@ __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
4774647746
4774747747#ifdef __LITTLE_ENDIAN__
4774847748#define vqrdmlsh_laneq_s16(__p0_355, __p1_355, __p2_355, __p3_355) __extension__ ({ \
47749 int16x4_t __ret_355; \
4774947750 int16x4_t __s0_355 = __p0_355; \
4775047751 int16x4_t __s1_355 = __p1_355; \
4775147752 int16x8_t __s2_355 = __p2_355; \
47752 int16x4_t __ret_355; \
4775347753 __ret_355 = vqrdmlsh_s16(__s0_355, __s1_355, splat_laneq_s16(__s2_355, __p3_355)); \
4775447754 __ret_355; \
4775547755})
4775647756#else
4775747757#define vqrdmlsh_laneq_s16(__p0_356, __p1_356, __p2_356, __p3_356) __extension__ ({ \
47758 int16x4_t __ret_356; \
4775847759 int16x4_t __s0_356 = __p0_356; \
4775947760 int16x4_t __s1_356 = __p1_356; \
4776047761 int16x8_t __s2_356 = __p2_356; \
4776147762 int16x4_t __rev0_356; __rev0_356 = __builtin_shufflevector(__s0_356, __s0_356, 3, 2, 1, 0); \
4776247763 int16x4_t __rev1_356; __rev1_356 = __builtin_shufflevector(__s1_356, __s1_356, 3, 2, 1, 0); \
4776347764 int16x8_t __rev2_356; __rev2_356 = __builtin_shufflevector(__s2_356, __s2_356, 7, 6, 5, 4, 3, 2, 1, 0); \
47764 int16x4_t __ret_356; \
4776547765 __ret_356 = __noswap_vqrdmlsh_s16(__rev0_356, __rev1_356, __noswap_splat_laneq_s16(__rev2_356, __p3_356)); \
4776647766 __ret_356 = __builtin_shufflevector(__ret_356, __ret_356, 3, 2, 1, 0); \
4776747767 __ret_356; \
......@@ -47778,9 +47778,9 @@ __ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
4777847778}
4777947779#else
4778047780__ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
47781 float64x2_t __ret;
4778147782 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4778247783 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47783 float64x2_t __ret;
4778447784 __ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
4778547785 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4778647786 return __ret;
......@@ -47810,8 +47810,8 @@ __ai float64x2_t vabsq_f64(float64x2_t __p0) {
4781047810}
4781147811#else
4781247812__ai float64x2_t vabsq_f64(float64x2_t __p0) {
47813 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4781447813 float64x2_t __ret;
47814 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4781547815 __ret = (float64x2_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 42);
4781647816 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4781747817 return __ret;
......@@ -47826,8 +47826,8 @@ __ai int64x2_t vabsq_s64(int64x2_t __p0) {
4782647826}
4782747827#else
4782847828__ai int64x2_t vabsq_s64(int64x2_t __p0) {
47829 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4783047829 int64x2_t __ret;
47830 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4783147831 __ret = (int64x2_t) __builtin_neon_vabsq_v((int8x16_t)__rev0, 35);
4783247832 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4783347833 return __ret;
......@@ -47857,9 +47857,9 @@ __ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
4785747857}
4785847858#else
4785947859__ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
47860 float64x2_t __ret;
4786047861 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4786147862 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47862 float64x2_t __ret;
4786347863 __ret = __rev0 + __rev1;
4786447864 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4786547865 return __ret;
......@@ -47894,10 +47894,10 @@ __ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p
4789447894}
4789547895#else
4789647896__ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
47897 uint16x8_t __ret;
4789747898 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4789847899 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4789947900 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47900 uint16x8_t __ret;
4790147901 __ret = __noswap_vcombine_u16(__rev0, __noswap_vaddhn_u32(__rev1, __rev2));
4790247902 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4790347903 return __ret;
......@@ -47912,10 +47912,10 @@ __ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p
4791247912}
4791347913#else
4791447914__ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
47915 uint32x4_t __ret;
4791547916 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4791647917 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4791747918 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47918 uint32x4_t __ret;
4791947919 __ret = __noswap_vcombine_u32(__rev0, __noswap_vaddhn_u64(__rev1, __rev2));
4792047920 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4792147921 return __ret;
......@@ -47930,10 +47930,10 @@ __ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2
4793047930}
4793147931#else
4793247932__ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
47933 uint8x16_t __ret;
4793347934 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4793447935 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4793547936 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47936 uint8x16_t __ret;
4793747937 __ret = __noswap_vcombine_u8(__rev0, __noswap_vaddhn_u16(__rev1, __rev2));
4793847938 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4793947939 return __ret;
......@@ -47948,10 +47948,10 @@ __ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
4794847948}
4794947949#else
4795047950__ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
47951 int16x8_t __ret;
4795147952 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4795247953 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4795347954 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
47954 int16x8_t __ret;
4795547955 __ret = __noswap_vcombine_s16(__rev0, __noswap_vaddhn_s32(__rev1, __rev2));
4795647956 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4795747957 return __ret;
......@@ -47966,10 +47966,10 @@ __ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
4796647966}
4796747967#else
4796847968__ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
47969 int32x4_t __ret;
4796947970 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4797047971 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4797147972 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
47972 int32x4_t __ret;
4797347973 __ret = __noswap_vcombine_s32(__rev0, __noswap_vaddhn_s64(__rev1, __rev2));
4797447974 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4797547975 return __ret;
......@@ -47984,10 +47984,10 @@ __ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
4798447984}
4798547985#else
4798647986__ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
47987 int8x16_t __ret;
4798747988 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4798847989 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4798947990 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
47990 int8x16_t __ret;
4799147991 __ret = __noswap_vcombine_s8(__rev0, __noswap_vaddhn_s16(__rev1, __rev2));
4799247992 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4799347993 return __ret;
......@@ -48002,8 +48002,8 @@ __ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
4800248002}
4800348003#else
4800448004__ai uint16_t vaddlvq_u8(uint8x16_t __p0) {
48005 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4800648005 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);
4800748007 __ret = (uint16_t) __builtin_neon_vaddlvq_u8(__rev0);
4800848008 return __ret;
4800948009}
......@@ -48017,8 +48017,8 @@ __ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
4801748017}
4801848018#else
4801948019__ai uint64_t vaddlvq_u32(uint32x4_t __p0) {
48020 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4802148020 uint64_t __ret;
48021 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4802248022 __ret = (uint64_t) __builtin_neon_vaddlvq_u32(__rev0);
4802348023 return __ret;
4802448024}
......@@ -48032,8 +48032,8 @@ __ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
4803248032}
4803348033#else
4803448034__ai uint32_t vaddlvq_u16(uint16x8_t __p0) {
48035 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4803648035 uint32_t __ret;
48036 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4803748037 __ret = (uint32_t) __builtin_neon_vaddlvq_u16(__rev0);
4803848038 return __ret;
4803948039}
......@@ -48047,8 +48047,8 @@ __ai int16_t vaddlvq_s8(int8x16_t __p0) {
4804748047}
4804848048#else
4804948049__ai int16_t vaddlvq_s8(int8x16_t __p0) {
48050 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4805148050 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);
4805248052 __ret = (int16_t) __builtin_neon_vaddlvq_s8(__rev0);
4805348053 return __ret;
4805448054}
......@@ -48062,8 +48062,8 @@ __ai int64_t vaddlvq_s32(int32x4_t __p0) {
4806248062}
4806348063#else
4806448064__ai int64_t vaddlvq_s32(int32x4_t __p0) {
48065 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4806648065 int64_t __ret;
48066 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4806748067 __ret = (int64_t) __builtin_neon_vaddlvq_s32(__rev0);
4806848068 return __ret;
4806948069}
......@@ -48077,8 +48077,8 @@ __ai int32_t vaddlvq_s16(int16x8_t __p0) {
4807748077}
4807848078#else
4807948079__ai int32_t vaddlvq_s16(int16x8_t __p0) {
48080 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4808148080 int32_t __ret;
48081 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4808248082 __ret = (int32_t) __builtin_neon_vaddlvq_s16(__rev0);
4808348083 return __ret;
4808448084}
......@@ -48092,8 +48092,8 @@ __ai uint16_t vaddlv_u8(uint8x8_t __p0) {
4809248092}
4809348093#else
4809448094__ai uint16_t vaddlv_u8(uint8x8_t __p0) {
48095 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4809648095 uint16_t __ret;
48096 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4809748097 __ret = (uint16_t) __builtin_neon_vaddlv_u8(__rev0);
4809848098 return __ret;
4809948099}
......@@ -48107,8 +48107,8 @@ __ai uint64_t vaddlv_u32(uint32x2_t __p0) {
4810748107}
4810848108#else
4810948109__ai uint64_t vaddlv_u32(uint32x2_t __p0) {
48110 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4811148110 uint64_t __ret;
48111 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4811248112 __ret = (uint64_t) __builtin_neon_vaddlv_u32(__rev0);
4811348113 return __ret;
4811448114}
......@@ -48122,8 +48122,8 @@ __ai uint32_t vaddlv_u16(uint16x4_t __p0) {
4812248122}
4812348123#else
4812448124__ai uint32_t vaddlv_u16(uint16x4_t __p0) {
48125 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4812648125 uint32_t __ret;
48126 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4812748127 __ret = (uint32_t) __builtin_neon_vaddlv_u16(__rev0);
4812848128 return __ret;
4812948129}
......@@ -48137,8 +48137,8 @@ __ai int16_t vaddlv_s8(int8x8_t __p0) {
4813748137}
4813848138#else
4813948139__ai int16_t vaddlv_s8(int8x8_t __p0) {
48140 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4814148140 int16_t __ret;
48141 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4814248142 __ret = (int16_t) __builtin_neon_vaddlv_s8(__rev0);
4814348143 return __ret;
4814448144}
......@@ -48152,8 +48152,8 @@ __ai int64_t vaddlv_s32(int32x2_t __p0) {
4815248152}
4815348153#else
4815448154__ai int64_t vaddlv_s32(int32x2_t __p0) {
48155 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4815648155 int64_t __ret;
48156 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4815748157 __ret = (int64_t) __builtin_neon_vaddlv_s32(__rev0);
4815848158 return __ret;
4815948159}
......@@ -48167,8 +48167,8 @@ __ai int32_t vaddlv_s16(int16x4_t __p0) {
4816748167}
4816848168#else
4816948169__ai int32_t vaddlv_s16(int16x4_t __p0) {
48170 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4817148170 int32_t __ret;
48171 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4817248172 __ret = (int32_t) __builtin_neon_vaddlv_s16(__rev0);
4817348173 return __ret;
4817448174}
......@@ -48182,8 +48182,8 @@ __ai uint8_t vaddvq_u8(uint8x16_t __p0) {
4818248182}
4818348183#else
4818448184__ai uint8_t vaddvq_u8(uint8x16_t __p0) {
48185 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4818648185 uint8_t __ret;
48186 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4818748187 __ret = (uint8_t) __builtin_neon_vaddvq_u8(__rev0);
4818848188 return __ret;
4818948189}
......@@ -48197,8 +48197,8 @@ __ai uint32_t vaddvq_u32(uint32x4_t __p0) {
4819748197}
4819848198#else
4819948199__ai uint32_t vaddvq_u32(uint32x4_t __p0) {
48200 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4820148200 uint32_t __ret;
48201 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4820248202 __ret = (uint32_t) __builtin_neon_vaddvq_u32(__rev0);
4820348203 return __ret;
4820448204}
......@@ -48212,8 +48212,8 @@ __ai uint64_t vaddvq_u64(uint64x2_t __p0) {
4821248212}
4821348213#else
4821448214__ai uint64_t vaddvq_u64(uint64x2_t __p0) {
48215 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4821648215 uint64_t __ret;
48216 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4821748217 __ret = (uint64_t) __builtin_neon_vaddvq_u64(__rev0);
4821848218 return __ret;
4821948219}
......@@ -48227,8 +48227,8 @@ __ai uint16_t vaddvq_u16(uint16x8_t __p0) {
4822748227}
4822848228#else
4822948229__ai uint16_t vaddvq_u16(uint16x8_t __p0) {
48230 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4823148230 uint16_t __ret;
48231 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4823248232 __ret = (uint16_t) __builtin_neon_vaddvq_u16(__rev0);
4823348233 return __ret;
4823448234}
......@@ -48242,8 +48242,8 @@ __ai int8_t vaddvq_s8(int8x16_t __p0) {
4824248242}
4824348243#else
4824448244__ai int8_t vaddvq_s8(int8x16_t __p0) {
48245 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4824648245 int8_t __ret;
48246 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4824748247 __ret = (int8_t) __builtin_neon_vaddvq_s8(__rev0);
4824848248 return __ret;
4824948249}
......@@ -48257,8 +48257,8 @@ __ai float64_t vaddvq_f64(float64x2_t __p0) {
4825748257}
4825848258#else
4825948259__ai float64_t vaddvq_f64(float64x2_t __p0) {
48260 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4826148260 float64_t __ret;
48261 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4826248262 __ret = (float64_t) __builtin_neon_vaddvq_f64(__rev0);
4826348263 return __ret;
4826448264}
......@@ -48272,8 +48272,8 @@ __ai float32_t vaddvq_f32(float32x4_t __p0) {
4827248272}
4827348273#else
4827448274__ai float32_t vaddvq_f32(float32x4_t __p0) {
48275 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4827648275 float32_t __ret;
48276 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4827748277 __ret = (float32_t) __builtin_neon_vaddvq_f32(__rev0);
4827848278 return __ret;
4827948279}
......@@ -48287,8 +48287,8 @@ __ai int32_t vaddvq_s32(int32x4_t __p0) {
4828748287}
4828848288#else
4828948289__ai int32_t vaddvq_s32(int32x4_t __p0) {
48290 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4829148290 int32_t __ret;
48291 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4829248292 __ret = (int32_t) __builtin_neon_vaddvq_s32(__rev0);
4829348293 return __ret;
4829448294}
......@@ -48302,8 +48302,8 @@ __ai int64_t vaddvq_s64(int64x2_t __p0) {
4830248302}
4830348303#else
4830448304__ai int64_t vaddvq_s64(int64x2_t __p0) {
48305 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4830648305 int64_t __ret;
48306 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4830748307 __ret = (int64_t) __builtin_neon_vaddvq_s64(__rev0);
4830848308 return __ret;
4830948309}
......@@ -48317,8 +48317,8 @@ __ai int16_t vaddvq_s16(int16x8_t __p0) {
4831748317}
4831848318#else
4831948319__ai int16_t vaddvq_s16(int16x8_t __p0) {
48320 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4832148320 int16_t __ret;
48321 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4832248322 __ret = (int16_t) __builtin_neon_vaddvq_s16(__rev0);
4832348323 return __ret;
4832448324}
......@@ -48332,8 +48332,8 @@ __ai uint8_t vaddv_u8(uint8x8_t __p0) {
4833248332}
4833348333#else
4833448334__ai uint8_t vaddv_u8(uint8x8_t __p0) {
48335 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4833648335 uint8_t __ret;
48336 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4833748337 __ret = (uint8_t) __builtin_neon_vaddv_u8(__rev0);
4833848338 return __ret;
4833948339}
......@@ -48347,8 +48347,8 @@ __ai uint32_t vaddv_u32(uint32x2_t __p0) {
4834748347}
4834848348#else
4834948349__ai uint32_t vaddv_u32(uint32x2_t __p0) {
48350 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4835148350 uint32_t __ret;
48351 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4835248352 __ret = (uint32_t) __builtin_neon_vaddv_u32(__rev0);
4835348353 return __ret;
4835448354}
......@@ -48362,8 +48362,8 @@ __ai uint16_t vaddv_u16(uint16x4_t __p0) {
4836248362}
4836348363#else
4836448364__ai uint16_t vaddv_u16(uint16x4_t __p0) {
48365 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4836648365 uint16_t __ret;
48366 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4836748367 __ret = (uint16_t) __builtin_neon_vaddv_u16(__rev0);
4836848368 return __ret;
4836948369}
......@@ -48377,8 +48377,8 @@ __ai int8_t vaddv_s8(int8x8_t __p0) {
4837748377}
4837848378#else
4837948379__ai int8_t vaddv_s8(int8x8_t __p0) {
48380 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4838148380 int8_t __ret;
48381 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4838248382 __ret = (int8_t) __builtin_neon_vaddv_s8(__rev0);
4838348383 return __ret;
4838448384}
......@@ -48392,8 +48392,8 @@ __ai float32_t vaddv_f32(float32x2_t __p0) {
4839248392}
4839348393#else
4839448394__ai float32_t vaddv_f32(float32x2_t __p0) {
48395 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4839648395 float32_t __ret;
48396 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4839748397 __ret = (float32_t) __builtin_neon_vaddv_f32(__rev0);
4839848398 return __ret;
4839948399}
......@@ -48407,8 +48407,8 @@ __ai int32_t vaddv_s32(int32x2_t __p0) {
4840748407}
4840848408#else
4840948409__ai int32_t vaddv_s32(int32x2_t __p0) {
48410 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4841148410 int32_t __ret;
48411 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4841248412 __ret = (int32_t) __builtin_neon_vaddv_s32(__rev0);
4841348413 return __ret;
4841448414}
......@@ -48422,8 +48422,8 @@ __ai int16_t vaddv_s16(int16x4_t __p0) {
4842248422}
4842348423#else
4842448424__ai int16_t vaddv_s16(int16x4_t __p0) {
48425 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4842648425 int16_t __ret;
48426 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4842748427 __ret = (int16_t) __builtin_neon_vaddv_s16(__rev0);
4842848428 return __ret;
4842948429}
......@@ -48442,10 +48442,10 @@ __ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
4844248442}
4844348443#else
4844448444__ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
48445 poly64x2_t __ret;
4844548446 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4844648447 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4844748448 poly64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
48448 poly64x2_t __ret;
4844948449 __ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 38);
4845048450 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4845148451 return __ret;
......@@ -48460,10 +48460,10 @@ __ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2)
4846048460}
4846148461#else
4846248462__ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
48463 float64x2_t __ret;
4846348464 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4846448465 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4846548466 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
48466 float64x2_t __ret;
4846748467 __ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
4846848468 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4846948469 return __ret;
......@@ -48483,9 +48483,9 @@ __ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
4848348483}
4848448484#else
4848548485__ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
48486 uint64x2_t __ret;
4848648487 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4848748488 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48488 uint64x2_t __ret;
4848948489 __ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
4849048490 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4849148491 return __ret;
......@@ -48515,9 +48515,9 @@ __ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
4851548515}
4851648516#else
4851748517__ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
48518 uint64x2_t __ret;
4851848519 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4851948520 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48520 uint64x2_t __ret;
4852148521 __ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
4852248522 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4852348523 return __ret;
......@@ -48547,9 +48547,9 @@ __ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
4854748547}
4854848548#else
4854948549__ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
48550 uint64x2_t __ret;
4855048551 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4855148552 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48552 uint64x2_t __ret;
4855348553 __ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
4855448554 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4855548555 return __ret;
......@@ -48579,9 +48579,9 @@ __ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
4857948579}
4858048580#else
4858148581__ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
48582 uint64x2_t __ret;
4858248583 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4858348584 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48584 uint64x2_t __ret;
4858548585 __ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
4858648586 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4858748587 return __ret;
......@@ -48616,9 +48616,9 @@ __ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
4861648616}
4861748617#else
4861848618__ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
48619 uint64x2_t __ret;
4861948620 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4862048621 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48621 uint64x2_t __ret;
4862248622 __ret = (uint64x2_t)(__rev0 == __rev1);
4862348623 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4862448624 return __ret;
......@@ -48633,9 +48633,9 @@ __ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
4863348633}
4863448634#else
4863548635__ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
48636 uint64x2_t __ret;
4863648637 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4863748638 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48638 uint64x2_t __ret;
4863948639 __ret = (uint64x2_t)(__rev0 == __rev1);
4864048640 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4864148641 return __ret;
......@@ -48650,9 +48650,9 @@ __ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
4865048650}
4865148651#else
4865248652__ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
48653 uint64x2_t __ret;
4865348654 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4865448655 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48655 uint64x2_t __ret;
4865648656 __ret = (uint64x2_t)(__rev0 == __rev1);
4865748657 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4865848658 return __ret;
......@@ -48667,9 +48667,9 @@ __ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
4866748667}
4866848668#else
4866948669__ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
48670 uint64x2_t __ret;
4867048671 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4867148672 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48672 uint64x2_t __ret;
4867348673 __ret = (uint64x2_t)(__rev0 == __rev1);
4867448674 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4867548675 return __ret;
......@@ -48719,8 +48719,8 @@ __ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
4871948719}
4872048720#else
4872148721__ai uint8x8_t vceqz_p8(poly8x8_t __p0) {
48722 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4872348722 uint8x8_t __ret;
48723 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4872448724 __ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
4872548725 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4872648726 return __ret;
......@@ -48740,8 +48740,8 @@ __ai uint8x16_t vceqzq_p8(poly8x16_t __p0) {
4874048740}
4874148741#else
4874248742__ai uint8x16_t vceqzq_p8(poly8x16_t __p0) {
48743 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4874448743 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);
4874548745 __ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
4874648746 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4874748747 return __ret;
......@@ -48756,8 +48756,8 @@ __ai uint64x2_t vceqzq_p64(poly64x2_t __p0) {
4875648756}
4875748757#else
4875848758__ai uint64x2_t vceqzq_p64(poly64x2_t __p0) {
48759 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4876048759 uint64x2_t __ret;
48760 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4876148761 __ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
4876248762 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4876348763 return __ret;
......@@ -48772,8 +48772,8 @@ __ai uint8x16_t vceqzq_u8(uint8x16_t __p0) {
4877248772}
4877348773#else
4877448774__ai uint8x16_t vceqzq_u8(uint8x16_t __p0) {
48775 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4877648775 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);
4877748777 __ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
4877848778 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4877948779 return __ret;
......@@ -48788,8 +48788,8 @@ __ai uint32x4_t vceqzq_u32(uint32x4_t __p0) {
4878848788}
4878948789#else
4879048790__ai uint32x4_t vceqzq_u32(uint32x4_t __p0) {
48791 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4879248791 uint32x4_t __ret;
48792 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4879348793 __ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
4879448794 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4879548795 return __ret;
......@@ -48804,8 +48804,8 @@ __ai uint64x2_t vceqzq_u64(uint64x2_t __p0) {
4880448804}
4880548805#else
4880648806__ai uint64x2_t vceqzq_u64(uint64x2_t __p0) {
48807 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4880848807 uint64x2_t __ret;
48808 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4880948809 __ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
4881048810 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4881148811 return __ret;
......@@ -48820,8 +48820,8 @@ __ai uint16x8_t vceqzq_u16(uint16x8_t __p0) {
4882048820}
4882148821#else
4882248822__ai uint16x8_t vceqzq_u16(uint16x8_t __p0) {
48823 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4882448823 uint16x8_t __ret;
48824 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4882548825 __ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
4882648826 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4882748827 return __ret;
......@@ -48836,8 +48836,8 @@ __ai uint8x16_t vceqzq_s8(int8x16_t __p0) {
4883648836}
4883748837#else
4883848838__ai uint8x16_t vceqzq_s8(int8x16_t __p0) {
48839 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4884048839 uint8x16_t __ret;
48840 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4884148841 __ret = (uint8x16_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 48);
4884248842 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4884348843 return __ret;
......@@ -48852,8 +48852,8 @@ __ai uint64x2_t vceqzq_f64(float64x2_t __p0) {
4885248852}
4885348853#else
4885448854__ai uint64x2_t vceqzq_f64(float64x2_t __p0) {
48855 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4885648855 uint64x2_t __ret;
48856 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4885748857 __ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
4885848858 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4885948859 return __ret;
......@@ -48868,8 +48868,8 @@ __ai uint32x4_t vceqzq_f32(float32x4_t __p0) {
4886848868}
4886948869#else
4887048870__ai uint32x4_t vceqzq_f32(float32x4_t __p0) {
48871 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4887248871 uint32x4_t __ret;
48872 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4887348873 __ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
4887448874 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4887548875 return __ret;
......@@ -48884,8 +48884,8 @@ __ai uint32x4_t vceqzq_s32(int32x4_t __p0) {
4888448884}
4888548885#else
4888648886__ai uint32x4_t vceqzq_s32(int32x4_t __p0) {
48887 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4888848887 uint32x4_t __ret;
48888 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4888948889 __ret = (uint32x4_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 50);
4889048890 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4889148891 return __ret;
......@@ -48900,8 +48900,8 @@ __ai uint64x2_t vceqzq_s64(int64x2_t __p0) {
4890048900}
4890148901#else
4890248902__ai uint64x2_t vceqzq_s64(int64x2_t __p0) {
48903 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4890448903 uint64x2_t __ret;
48904 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4890548905 __ret = (uint64x2_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 51);
4890648906 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4890748907 return __ret;
......@@ -48916,8 +48916,8 @@ __ai uint16x8_t vceqzq_s16(int16x8_t __p0) {
4891648916}
4891748917#else
4891848918__ai uint16x8_t vceqzq_s16(int16x8_t __p0) {
48919 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4892048919 uint16x8_t __ret;
48920 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4892148921 __ret = (uint16x8_t) __builtin_neon_vceqzq_v((int8x16_t)__rev0, 49);
4892248922 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4892348923 return __ret;
......@@ -48932,8 +48932,8 @@ __ai uint8x8_t vceqz_u8(uint8x8_t __p0) {
4893248932}
4893348933#else
4893448934__ai uint8x8_t vceqz_u8(uint8x8_t __p0) {
48935 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4893648935 uint8x8_t __ret;
48936 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4893748937 __ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
4893848938 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4893948939 return __ret;
......@@ -48948,8 +48948,8 @@ __ai uint32x2_t vceqz_u32(uint32x2_t __p0) {
4894848948}
4894948949#else
4895048950__ai uint32x2_t vceqz_u32(uint32x2_t __p0) {
48951 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4895248951 uint32x2_t __ret;
48952 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4895348953 __ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
4895448954 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4895548955 return __ret;
......@@ -48969,8 +48969,8 @@ __ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
4896948969}
4897048970#else
4897148971__ai uint16x4_t vceqz_u16(uint16x4_t __p0) {
48972 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4897348972 uint16x4_t __ret;
48973 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4897448974 __ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
4897548975 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4897648976 return __ret;
......@@ -48985,8 +48985,8 @@ __ai uint8x8_t vceqz_s8(int8x8_t __p0) {
4898548985}
4898648986#else
4898748987__ai uint8x8_t vceqz_s8(int8x8_t __p0) {
48988 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4898948988 uint8x8_t __ret;
48989 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4899048990 __ret = (uint8x8_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 16);
4899148991 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4899248992 return __ret;
......@@ -49006,8 +49006,8 @@ __ai uint32x2_t vceqz_f32(float32x2_t __p0) {
4900649006}
4900749007#else
4900849008__ai uint32x2_t vceqz_f32(float32x2_t __p0) {
49009 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4901049009 uint32x2_t __ret;
49010 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4901149011 __ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
4901249012 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4901349013 return __ret;
......@@ -49022,8 +49022,8 @@ __ai uint32x2_t vceqz_s32(int32x2_t __p0) {
4902249022}
4902349023#else
4902449024__ai uint32x2_t vceqz_s32(int32x2_t __p0) {
49025 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4902649025 uint32x2_t __ret;
49026 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4902749027 __ret = (uint32x2_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 18);
4902849028 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4902949029 return __ret;
......@@ -49043,8 +49043,8 @@ __ai uint16x4_t vceqz_s16(int16x4_t __p0) {
4904349043}
4904449044#else
4904549045__ai uint16x4_t vceqz_s16(int16x4_t __p0) {
49046 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4904749046 uint16x4_t __ret;
49047 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4904849048 __ret = (uint16x4_t) __builtin_neon_vceqz_v((int8x8_t)__rev0, 17);
4904949049 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4905049050 return __ret;
......@@ -49079,9 +49079,9 @@ __ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
4907949079}
4908049080#else
4908149081__ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49082 uint64x2_t __ret;
4908249083 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4908349084 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49084 uint64x2_t __ret;
4908549085 __ret = (uint64x2_t)(__rev0 >= __rev1);
4908649086 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4908749087 return __ret;
......@@ -49096,9 +49096,9 @@ __ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
4909649096}
4909749097#else
4909849098__ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
49099 uint64x2_t __ret;
4909949100 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4910049101 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49101 uint64x2_t __ret;
4910249102 __ret = (uint64x2_t)(__rev0 >= __rev1);
4910349103 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4910449104 return __ret;
......@@ -49113,9 +49113,9 @@ __ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
4911349113}
4911449114#else
4911549115__ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
49116 uint64x2_t __ret;
4911649117 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4911749118 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49118 uint64x2_t __ret;
4911949119 __ret = (uint64x2_t)(__rev0 >= __rev1);
4912049120 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4912149121 return __ret;
......@@ -49165,8 +49165,8 @@ __ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
4916549165}
4916649166#else
4916749167__ai uint8x16_t vcgezq_s8(int8x16_t __p0) {
49168 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4916949168 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);
4917049170 __ret = (uint8x16_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 48);
4917149171 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4917249172 return __ret;
......@@ -49181,8 +49181,8 @@ __ai uint64x2_t vcgezq_f64(float64x2_t __p0) {
4918149181}
4918249182#else
4918349183__ai uint64x2_t vcgezq_f64(float64x2_t __p0) {
49184 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4918549184 uint64x2_t __ret;
49185 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4918649186 __ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 51);
4918749187 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4918849188 return __ret;
......@@ -49197,8 +49197,8 @@ __ai uint32x4_t vcgezq_f32(float32x4_t __p0) {
4919749197}
4919849198#else
4919949199__ai uint32x4_t vcgezq_f32(float32x4_t __p0) {
49200 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4920149200 uint32x4_t __ret;
49201 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4920249202 __ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 50);
4920349203 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4920449204 return __ret;
......@@ -49213,8 +49213,8 @@ __ai uint32x4_t vcgezq_s32(int32x4_t __p0) {
4921349213}
4921449214#else
4921549215__ai uint32x4_t vcgezq_s32(int32x4_t __p0) {
49216 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4921749216 uint32x4_t __ret;
49217 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4921849218 __ret = (uint32x4_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 50);
4921949219 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4922049220 return __ret;
......@@ -49229,8 +49229,8 @@ __ai uint64x2_t vcgezq_s64(int64x2_t __p0) {
4922949229}
4923049230#else
4923149231__ai uint64x2_t vcgezq_s64(int64x2_t __p0) {
49232 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4923349232 uint64x2_t __ret;
49233 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4923449234 __ret = (uint64x2_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 51);
4923549235 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4923649236 return __ret;
......@@ -49245,8 +49245,8 @@ __ai uint16x8_t vcgezq_s16(int16x8_t __p0) {
4924549245}
4924649246#else
4924749247__ai uint16x8_t vcgezq_s16(int16x8_t __p0) {
49248 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4924949248 uint16x8_t __ret;
49249 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4925049250 __ret = (uint16x8_t) __builtin_neon_vcgezq_v((int8x16_t)__rev0, 49);
4925149251 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4925249252 return __ret;
......@@ -49261,8 +49261,8 @@ __ai uint8x8_t vcgez_s8(int8x8_t __p0) {
4926149261}
4926249262#else
4926349263__ai uint8x8_t vcgez_s8(int8x8_t __p0) {
49264 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4926549264 uint8x8_t __ret;
49265 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4926649266 __ret = (uint8x8_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 16);
4926749267 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4926849268 return __ret;
......@@ -49282,8 +49282,8 @@ __ai uint32x2_t vcgez_f32(float32x2_t __p0) {
4928249282}
4928349283#else
4928449284__ai uint32x2_t vcgez_f32(float32x2_t __p0) {
49285 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4928649285 uint32x2_t __ret;
49286 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4928749287 __ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 18);
4928849288 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4928949289 return __ret;
......@@ -49298,8 +49298,8 @@ __ai uint32x2_t vcgez_s32(int32x2_t __p0) {
4929849298}
4929949299#else
4930049300__ai uint32x2_t vcgez_s32(int32x2_t __p0) {
49301 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4930249301 uint32x2_t __ret;
49302 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4930349303 __ret = (uint32x2_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 18);
4930449304 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4930549305 return __ret;
......@@ -49319,8 +49319,8 @@ __ai uint16x4_t vcgez_s16(int16x4_t __p0) {
4931949319}
4932049320#else
4932149321__ai uint16x4_t vcgez_s16(int16x4_t __p0) {
49322 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4932349322 uint16x4_t __ret;
49323 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4932449324 __ret = (uint16x4_t) __builtin_neon_vcgez_v((int8x8_t)__rev0, 17);
4932549325 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4932649326 return __ret;
......@@ -49350,9 +49350,9 @@ __ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
4935049350}
4935149351#else
4935249352__ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49353 uint64x2_t __ret;
4935349354 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4935449355 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49355 uint64x2_t __ret;
4935649356 __ret = (uint64x2_t)(__rev0 > __rev1);
4935749357 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4935849358 return __ret;
......@@ -49367,9 +49367,9 @@ __ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
4936749367}
4936849368#else
4936949369__ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
49370 uint64x2_t __ret;
4937049371 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4937149372 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49372 uint64x2_t __ret;
4937349373 __ret = (uint64x2_t)(__rev0 > __rev1);
4937449374 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4937549375 return __ret;
......@@ -49384,9 +49384,9 @@ __ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
4938449384}
4938549385#else
4938649386__ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
49387 uint64x2_t __ret;
4938749388 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4938849389 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49389 uint64x2_t __ret;
4939049390 __ret = (uint64x2_t)(__rev0 > __rev1);
4939149391 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4939249392 return __ret;
......@@ -49436,8 +49436,8 @@ __ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
4943649436}
4943749437#else
4943849438__ai uint8x16_t vcgtzq_s8(int8x16_t __p0) {
49439 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4944049439 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);
4944149441 __ret = (uint8x16_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 48);
4944249442 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4944349443 return __ret;
......@@ -49452,8 +49452,8 @@ __ai uint64x2_t vcgtzq_f64(float64x2_t __p0) {
4945249452}
4945349453#else
4945449454__ai uint64x2_t vcgtzq_f64(float64x2_t __p0) {
49455 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4945649455 uint64x2_t __ret;
49456 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4945749457 __ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 51);
4945849458 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4945949459 return __ret;
......@@ -49468,8 +49468,8 @@ __ai uint32x4_t vcgtzq_f32(float32x4_t __p0) {
4946849468}
4946949469#else
4947049470__ai uint32x4_t vcgtzq_f32(float32x4_t __p0) {
49471 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4947249471 uint32x4_t __ret;
49472 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4947349473 __ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 50);
4947449474 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4947549475 return __ret;
......@@ -49484,8 +49484,8 @@ __ai uint32x4_t vcgtzq_s32(int32x4_t __p0) {
4948449484}
4948549485#else
4948649486__ai uint32x4_t vcgtzq_s32(int32x4_t __p0) {
49487 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4948849487 uint32x4_t __ret;
49488 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4948949489 __ret = (uint32x4_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 50);
4949049490 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4949149491 return __ret;
......@@ -49500,8 +49500,8 @@ __ai uint64x2_t vcgtzq_s64(int64x2_t __p0) {
4950049500}
4950149501#else
4950249502__ai uint64x2_t vcgtzq_s64(int64x2_t __p0) {
49503 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4950449503 uint64x2_t __ret;
49504 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4950549505 __ret = (uint64x2_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 51);
4950649506 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4950749507 return __ret;
......@@ -49516,8 +49516,8 @@ __ai uint16x8_t vcgtzq_s16(int16x8_t __p0) {
4951649516}
4951749517#else
4951849518__ai uint16x8_t vcgtzq_s16(int16x8_t __p0) {
49519 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4952049519 uint16x8_t __ret;
49520 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4952149521 __ret = (uint16x8_t) __builtin_neon_vcgtzq_v((int8x16_t)__rev0, 49);
4952249522 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4952349523 return __ret;
......@@ -49532,8 +49532,8 @@ __ai uint8x8_t vcgtz_s8(int8x8_t __p0) {
4953249532}
4953349533#else
4953449534__ai uint8x8_t vcgtz_s8(int8x8_t __p0) {
49535 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4953649535 uint8x8_t __ret;
49536 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4953749537 __ret = (uint8x8_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 16);
4953849538 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4953949539 return __ret;
......@@ -49553,8 +49553,8 @@ __ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
4955349553}
4955449554#else
4955549555__ai uint32x2_t vcgtz_f32(float32x2_t __p0) {
49556 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4955749556 uint32x2_t __ret;
49557 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4955849558 __ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 18);
4955949559 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4956049560 return __ret;
......@@ -49569,8 +49569,8 @@ __ai uint32x2_t vcgtz_s32(int32x2_t __p0) {
4956949569}
4957049570#else
4957149571__ai uint32x2_t vcgtz_s32(int32x2_t __p0) {
49572 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4957349572 uint32x2_t __ret;
49573 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4957449574 __ret = (uint32x2_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 18);
4957549575 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4957649576 return __ret;
......@@ -49590,8 +49590,8 @@ __ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
4959049590}
4959149591#else
4959249592__ai uint16x4_t vcgtz_s16(int16x4_t __p0) {
49593 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4959449593 uint16x4_t __ret;
49594 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4959549595 __ret = (uint16x4_t) __builtin_neon_vcgtz_v((int8x8_t)__rev0, 17);
4959649596 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4959749597 return __ret;
......@@ -49621,9 +49621,9 @@ __ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
4962149621}
4962249622#else
4962349623__ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49624 uint64x2_t __ret;
4962449625 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4962549626 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49626 uint64x2_t __ret;
4962749627 __ret = (uint64x2_t)(__rev0 <= __rev1);
4962849628 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4962949629 return __ret;
......@@ -49638,9 +49638,9 @@ __ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
4963849638}
4963949639#else
4964049640__ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
49641 uint64x2_t __ret;
4964149642 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4964249643 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49643 uint64x2_t __ret;
4964449644 __ret = (uint64x2_t)(__rev0 <= __rev1);
4964549645 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4964649646 return __ret;
......@@ -49655,9 +49655,9 @@ __ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
4965549655}
4965649656#else
4965749657__ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
49658 uint64x2_t __ret;
4965849659 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4965949660 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49660 uint64x2_t __ret;
4966149661 __ret = (uint64x2_t)(__rev0 <= __rev1);
4966249662 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4966349663 return __ret;
......@@ -49707,8 +49707,8 @@ __ai uint8x16_t vclezq_s8(int8x16_t __p0) {
4970749707}
4970849708#else
4970949709__ai uint8x16_t vclezq_s8(int8x16_t __p0) {
49710 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4971149710 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);
4971249712 __ret = (uint8x16_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 48);
4971349713 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4971449714 return __ret;
......@@ -49723,8 +49723,8 @@ __ai uint64x2_t vclezq_f64(float64x2_t __p0) {
4972349723}
4972449724#else
4972549725__ai uint64x2_t vclezq_f64(float64x2_t __p0) {
49726 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4972749726 uint64x2_t __ret;
49727 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4972849728 __ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 51);
4972949729 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4973049730 return __ret;
......@@ -49739,8 +49739,8 @@ __ai uint32x4_t vclezq_f32(float32x4_t __p0) {
4973949739}
4974049740#else
4974149741__ai uint32x4_t vclezq_f32(float32x4_t __p0) {
49742 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4974349742 uint32x4_t __ret;
49743 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4974449744 __ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 50);
4974549745 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4974649746 return __ret;
......@@ -49755,8 +49755,8 @@ __ai uint32x4_t vclezq_s32(int32x4_t __p0) {
4975549755}
4975649756#else
4975749757__ai uint32x4_t vclezq_s32(int32x4_t __p0) {
49758 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4975949758 uint32x4_t __ret;
49759 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4976049760 __ret = (uint32x4_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 50);
4976149761 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4976249762 return __ret;
......@@ -49771,8 +49771,8 @@ __ai uint64x2_t vclezq_s64(int64x2_t __p0) {
4977149771}
4977249772#else
4977349773__ai uint64x2_t vclezq_s64(int64x2_t __p0) {
49774 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4977549774 uint64x2_t __ret;
49775 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4977649776 __ret = (uint64x2_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 51);
4977749777 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4977849778 return __ret;
......@@ -49787,8 +49787,8 @@ __ai uint16x8_t vclezq_s16(int16x8_t __p0) {
4978749787}
4978849788#else
4978949789__ai uint16x8_t vclezq_s16(int16x8_t __p0) {
49790 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4979149790 uint16x8_t __ret;
49791 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4979249792 __ret = (uint16x8_t) __builtin_neon_vclezq_v((int8x16_t)__rev0, 49);
4979349793 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4979449794 return __ret;
......@@ -49803,8 +49803,8 @@ __ai uint8x8_t vclez_s8(int8x8_t __p0) {
4980349803}
4980449804#else
4980549805__ai uint8x8_t vclez_s8(int8x8_t __p0) {
49806 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4980749806 uint8x8_t __ret;
49807 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
4980849808 __ret = (uint8x8_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 16);
4980949809 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
4981049810 return __ret;
......@@ -49824,8 +49824,8 @@ __ai uint32x2_t vclez_f32(float32x2_t __p0) {
4982449824}
4982549825#else
4982649826__ai uint32x2_t vclez_f32(float32x2_t __p0) {
49827 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4982849827 uint32x2_t __ret;
49828 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4982949829 __ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 18);
4983049830 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4983149831 return __ret;
......@@ -49840,8 +49840,8 @@ __ai uint32x2_t vclez_s32(int32x2_t __p0) {
4984049840}
4984149841#else
4984249842__ai uint32x2_t vclez_s32(int32x2_t __p0) {
49843 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4984449843 uint32x2_t __ret;
49844 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4984549845 __ret = (uint32x2_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 18);
4984649846 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4984749847 return __ret;
......@@ -49861,8 +49861,8 @@ __ai uint16x4_t vclez_s16(int16x4_t __p0) {
4986149861}
4986249862#else
4986349863__ai uint16x4_t vclez_s16(int16x4_t __p0) {
49864 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4986549864 uint16x4_t __ret;
49865 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
4986649866 __ret = (uint16x4_t) __builtin_neon_vclez_v((int8x8_t)__rev0, 17);
4986749867 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
4986849868 return __ret;
......@@ -49892,9 +49892,9 @@ __ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
4989249892}
4989349893#else
4989449894__ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
49895 uint64x2_t __ret;
4989549896 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4989649897 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49897 uint64x2_t __ret;
4989849898 __ret = (uint64x2_t)(__rev0 < __rev1);
4989949899 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4990049900 return __ret;
......@@ -49909,9 +49909,9 @@ __ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
4990949909}
4991049910#else
4991149911__ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
49912 uint64x2_t __ret;
4991249913 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4991349914 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49914 uint64x2_t __ret;
4991549915 __ret = (uint64x2_t)(__rev0 < __rev1);
4991649916 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4991749917 return __ret;
......@@ -49926,9 +49926,9 @@ __ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
4992649926}
4992749927#else
4992849928__ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
49929 uint64x2_t __ret;
4992949930 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4993049931 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
49931 uint64x2_t __ret;
4993249932 __ret = (uint64x2_t)(__rev0 < __rev1);
4993349933 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
4993449934 return __ret;
......@@ -49978,8 +49978,8 @@ __ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
4997849978}
4997949979#else
4998049980__ai uint8x16_t vcltzq_s8(int8x16_t __p0) {
49981 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4998249981 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);
4998349983 __ret = (uint8x16_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 48);
4998449984 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
4998549985 return __ret;
......@@ -49994,8 +49994,8 @@ __ai uint64x2_t vcltzq_f64(float64x2_t __p0) {
4999449994}
4999549995#else
4999649996__ai uint64x2_t vcltzq_f64(float64x2_t __p0) {
49997 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4999849997 uint64x2_t __ret;
49998 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
4999949999 __ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 51);
5000050000 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5000150001 return __ret;
......@@ -50010,8 +50010,8 @@ __ai uint32x4_t vcltzq_f32(float32x4_t __p0) {
5001050010}
5001150011#else
5001250012__ai uint32x4_t vcltzq_f32(float32x4_t __p0) {
50013 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5001450013 uint32x4_t __ret;
50014 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5001550015 __ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 50);
5001650016 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5001750017 return __ret;
......@@ -50026,8 +50026,8 @@ __ai uint32x4_t vcltzq_s32(int32x4_t __p0) {
5002650026}
5002750027#else
5002850028__ai uint32x4_t vcltzq_s32(int32x4_t __p0) {
50029 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5003050029 uint32x4_t __ret;
50030 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5003150031 __ret = (uint32x4_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 50);
5003250032 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5003350033 return __ret;
......@@ -50042,8 +50042,8 @@ __ai uint64x2_t vcltzq_s64(int64x2_t __p0) {
5004250042}
5004350043#else
5004450044__ai uint64x2_t vcltzq_s64(int64x2_t __p0) {
50045 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5004650045 uint64x2_t __ret;
50046 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5004750047 __ret = (uint64x2_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 51);
5004850048 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5004950049 return __ret;
......@@ -50058,8 +50058,8 @@ __ai uint16x8_t vcltzq_s16(int16x8_t __p0) {
5005850058}
5005950059#else
5006050060__ai uint16x8_t vcltzq_s16(int16x8_t __p0) {
50061 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5006250061 uint16x8_t __ret;
50062 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5006350063 __ret = (uint16x8_t) __builtin_neon_vcltzq_v((int8x16_t)__rev0, 49);
5006450064 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5006550065 return __ret;
......@@ -50074,8 +50074,8 @@ __ai uint8x8_t vcltz_s8(int8x8_t __p0) {
5007450074}
5007550075#else
5007650076__ai uint8x8_t vcltz_s8(int8x8_t __p0) {
50077 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5007850077 uint8x8_t __ret;
50078 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5007950079 __ret = (uint8x8_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 16);
5008050080 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5008150081 return __ret;
......@@ -50095,8 +50095,8 @@ __ai uint32x2_t vcltz_f32(float32x2_t __p0) {
5009550095}
5009650096#else
5009750097__ai uint32x2_t vcltz_f32(float32x2_t __p0) {
50098 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5009950098 uint32x2_t __ret;
50099 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5010050100 __ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 18);
5010150101 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5010250102 return __ret;
......@@ -50111,8 +50111,8 @@ __ai uint32x2_t vcltz_s32(int32x2_t __p0) {
5011150111}
5011250112#else
5011350113__ai uint32x2_t vcltz_s32(int32x2_t __p0) {
50114 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5011550114 uint32x2_t __ret;
50115 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5011650116 __ret = (uint32x2_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 18);
5011750117 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5011850118 return __ret;
......@@ -50132,8 +50132,8 @@ __ai uint16x4_t vcltz_s16(int16x4_t __p0) {
5013250132}
5013350133#else
5013450134__ai uint16x4_t vcltz_s16(int16x4_t __p0) {
50135 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5013650135 uint16x4_t __ret;
50136 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5013750137 __ret = (uint16x4_t) __builtin_neon_vcltz_v((int8x8_t)__rev0, 17);
5013850138 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5013950139 return __ret;
......@@ -50187,19 +50187,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5018750187
5018850188#ifdef __LITTLE_ENDIAN__
5018950189#define vcopyq_lane_p8(__p0_357, __p1_357, __p2_357, __p3_357) __extension__ ({ \
50190 poly8x16_t __ret_357; \
5019050191 poly8x16_t __s0_357 = __p0_357; \
5019150192 poly8x8_t __s2_357 = __p2_357; \
50192 poly8x16_t __ret_357; \
5019350193 __ret_357 = vsetq_lane_p8(vget_lane_p8(__s2_357, __p3_357), __s0_357, __p1_357); \
5019450194 __ret_357; \
5019550195})
5019650196#else
5019750197#define vcopyq_lane_p8(__p0_358, __p1_358, __p2_358, __p3_358) __extension__ ({ \
50198 poly8x16_t __ret_358; \
5019850199 poly8x16_t __s0_358 = __p0_358; \
5019950200 poly8x8_t __s2_358 = __p2_358; \
5020050201 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); \
5020150202 poly8x8_t __rev2_358; __rev2_358 = __builtin_shufflevector(__s2_358, __s2_358, 7, 6, 5, 4, 3, 2, 1, 0); \
50202 poly8x16_t __ret_358; \
5020350203 __ret_358 = __noswap_vsetq_lane_p8(__noswap_vget_lane_p8(__rev2_358, __p3_358), __rev0_358, __p1_358); \
5020450204 __ret_358 = __builtin_shufflevector(__ret_358, __ret_358, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5020550205 __ret_358; \
......@@ -50208,19 +50208,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5020850208
5020950209#ifdef __LITTLE_ENDIAN__
5021050210#define vcopyq_lane_p16(__p0_359, __p1_359, __p2_359, __p3_359) __extension__ ({ \
50211 poly16x8_t __ret_359; \
5021150212 poly16x8_t __s0_359 = __p0_359; \
5021250213 poly16x4_t __s2_359 = __p2_359; \
50213 poly16x8_t __ret_359; \
5021450214 __ret_359 = vsetq_lane_p16(vget_lane_p16(__s2_359, __p3_359), __s0_359, __p1_359); \
5021550215 __ret_359; \
5021650216})
5021750217#else
5021850218#define vcopyq_lane_p16(__p0_360, __p1_360, __p2_360, __p3_360) __extension__ ({ \
50219 poly16x8_t __ret_360; \
5021950220 poly16x8_t __s0_360 = __p0_360; \
5022050221 poly16x4_t __s2_360 = __p2_360; \
5022150222 poly16x8_t __rev0_360; __rev0_360 = __builtin_shufflevector(__s0_360, __s0_360, 7, 6, 5, 4, 3, 2, 1, 0); \
5022250223 poly16x4_t __rev2_360; __rev2_360 = __builtin_shufflevector(__s2_360, __s2_360, 3, 2, 1, 0); \
50223 poly16x8_t __ret_360; \
5022450224 __ret_360 = __noswap_vsetq_lane_p16(__noswap_vget_lane_p16(__rev2_360, __p3_360), __rev0_360, __p1_360); \
5022550225 __ret_360 = __builtin_shufflevector(__ret_360, __ret_360, 7, 6, 5, 4, 3, 2, 1, 0); \
5022650226 __ret_360; \
......@@ -50229,19 +50229,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5022950229
5023050230#ifdef __LITTLE_ENDIAN__
5023150231#define vcopyq_lane_u8(__p0_361, __p1_361, __p2_361, __p3_361) __extension__ ({ \
50232 uint8x16_t __ret_361; \
5023250233 uint8x16_t __s0_361 = __p0_361; \
5023350234 uint8x8_t __s2_361 = __p2_361; \
50234 uint8x16_t __ret_361; \
5023550235 __ret_361 = vsetq_lane_u8(vget_lane_u8(__s2_361, __p3_361), __s0_361, __p1_361); \
5023650236 __ret_361; \
5023750237})
5023850238#else
5023950239#define vcopyq_lane_u8(__p0_362, __p1_362, __p2_362, __p3_362) __extension__ ({ \
50240 uint8x16_t __ret_362; \
5024050241 uint8x16_t __s0_362 = __p0_362; \
5024150242 uint8x8_t __s2_362 = __p2_362; \
5024250243 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); \
5024350244 uint8x8_t __rev2_362; __rev2_362 = __builtin_shufflevector(__s2_362, __s2_362, 7, 6, 5, 4, 3, 2, 1, 0); \
50244 uint8x16_t __ret_362; \
5024550245 __ret_362 = __noswap_vsetq_lane_u8(__noswap_vget_lane_u8(__rev2_362, __p3_362), __rev0_362, __p1_362); \
5024650246 __ret_362 = __builtin_shufflevector(__ret_362, __ret_362, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5024750247 __ret_362; \
......@@ -50250,19 +50250,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5025050250
5025150251#ifdef __LITTLE_ENDIAN__
5025250252#define vcopyq_lane_u32(__p0_363, __p1_363, __p2_363, __p3_363) __extension__ ({ \
50253 uint32x4_t __ret_363; \
5025350254 uint32x4_t __s0_363 = __p0_363; \
5025450255 uint32x2_t __s2_363 = __p2_363; \
50255 uint32x4_t __ret_363; \
5025650256 __ret_363 = vsetq_lane_u32(vget_lane_u32(__s2_363, __p3_363), __s0_363, __p1_363); \
5025750257 __ret_363; \
5025850258})
5025950259#else
5026050260#define vcopyq_lane_u32(__p0_364, __p1_364, __p2_364, __p3_364) __extension__ ({ \
50261 uint32x4_t __ret_364; \
5026150262 uint32x4_t __s0_364 = __p0_364; \
5026250263 uint32x2_t __s2_364 = __p2_364; \
5026350264 uint32x4_t __rev0_364; __rev0_364 = __builtin_shufflevector(__s0_364, __s0_364, 3, 2, 1, 0); \
5026450265 uint32x2_t __rev2_364; __rev2_364 = __builtin_shufflevector(__s2_364, __s2_364, 1, 0); \
50265 uint32x4_t __ret_364; \
5026650266 __ret_364 = __noswap_vsetq_lane_u32(__noswap_vget_lane_u32(__rev2_364, __p3_364), __rev0_364, __p1_364); \
5026750267 __ret_364 = __builtin_shufflevector(__ret_364, __ret_364, 3, 2, 1, 0); \
5026850268 __ret_364; \
......@@ -50271,18 +50271,18 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5027150271
5027250272#ifdef __LITTLE_ENDIAN__
5027350273#define vcopyq_lane_u64(__p0_365, __p1_365, __p2_365, __p3_365) __extension__ ({ \
50274 uint64x2_t __ret_365; \
5027450275 uint64x2_t __s0_365 = __p0_365; \
5027550276 uint64x1_t __s2_365 = __p2_365; \
50276 uint64x2_t __ret_365; \
5027750277 __ret_365 = vsetq_lane_u64(vget_lane_u64(__s2_365, __p3_365), __s0_365, __p1_365); \
5027850278 __ret_365; \
5027950279})
5028050280#else
5028150281#define vcopyq_lane_u64(__p0_366, __p1_366, __p2_366, __p3_366) __extension__ ({ \
50282 uint64x2_t __ret_366; \
5028250283 uint64x2_t __s0_366 = __p0_366; \
5028350284 uint64x1_t __s2_366 = __p2_366; \
5028450285 uint64x2_t __rev0_366; __rev0_366 = __builtin_shufflevector(__s0_366, __s0_366, 1, 0); \
50285 uint64x2_t __ret_366; \
5028650286 __ret_366 = __noswap_vsetq_lane_u64(vget_lane_u64(__s2_366, __p3_366), __rev0_366, __p1_366); \
5028750287 __ret_366 = __builtin_shufflevector(__ret_366, __ret_366, 1, 0); \
5028850288 __ret_366; \
......@@ -50291,19 +50291,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5029150291
5029250292#ifdef __LITTLE_ENDIAN__
5029350293#define vcopyq_lane_u16(__p0_367, __p1_367, __p2_367, __p3_367) __extension__ ({ \
50294 uint16x8_t __ret_367; \
5029450295 uint16x8_t __s0_367 = __p0_367; \
5029550296 uint16x4_t __s2_367 = __p2_367; \
50296 uint16x8_t __ret_367; \
5029750297 __ret_367 = vsetq_lane_u16(vget_lane_u16(__s2_367, __p3_367), __s0_367, __p1_367); \
5029850298 __ret_367; \
5029950299})
5030050300#else
5030150301#define vcopyq_lane_u16(__p0_368, __p1_368, __p2_368, __p3_368) __extension__ ({ \
50302 uint16x8_t __ret_368; \
5030250303 uint16x8_t __s0_368 = __p0_368; \
5030350304 uint16x4_t __s2_368 = __p2_368; \
5030450305 uint16x8_t __rev0_368; __rev0_368 = __builtin_shufflevector(__s0_368, __s0_368, 7, 6, 5, 4, 3, 2, 1, 0); \
5030550306 uint16x4_t __rev2_368; __rev2_368 = __builtin_shufflevector(__s2_368, __s2_368, 3, 2, 1, 0); \
50306 uint16x8_t __ret_368; \
5030750307 __ret_368 = __noswap_vsetq_lane_u16(__noswap_vget_lane_u16(__rev2_368, __p3_368), __rev0_368, __p1_368); \
5030850308 __ret_368 = __builtin_shufflevector(__ret_368, __ret_368, 7, 6, 5, 4, 3, 2, 1, 0); \
5030950309 __ret_368; \
......@@ -50312,19 +50312,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5031250312
5031350313#ifdef __LITTLE_ENDIAN__
5031450314#define vcopyq_lane_s8(__p0_369, __p1_369, __p2_369, __p3_369) __extension__ ({ \
50315 int8x16_t __ret_369; \
5031550316 int8x16_t __s0_369 = __p0_369; \
5031650317 int8x8_t __s2_369 = __p2_369; \
50317 int8x16_t __ret_369; \
5031850318 __ret_369 = vsetq_lane_s8(vget_lane_s8(__s2_369, __p3_369), __s0_369, __p1_369); \
5031950319 __ret_369; \
5032050320})
5032150321#else
5032250322#define vcopyq_lane_s8(__p0_370, __p1_370, __p2_370, __p3_370) __extension__ ({ \
50323 int8x16_t __ret_370; \
5032350324 int8x16_t __s0_370 = __p0_370; \
5032450325 int8x8_t __s2_370 = __p2_370; \
5032550326 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); \
5032650327 int8x8_t __rev2_370; __rev2_370 = __builtin_shufflevector(__s2_370, __s2_370, 7, 6, 5, 4, 3, 2, 1, 0); \
50327 int8x16_t __ret_370; \
5032850328 __ret_370 = __noswap_vsetq_lane_s8(__noswap_vget_lane_s8(__rev2_370, __p3_370), __rev0_370, __p1_370); \
5032950329 __ret_370 = __builtin_shufflevector(__ret_370, __ret_370, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5033050330 __ret_370; \
......@@ -50333,19 +50333,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5033350333
5033450334#ifdef __LITTLE_ENDIAN__
5033550335#define vcopyq_lane_f32(__p0_371, __p1_371, __p2_371, __p3_371) __extension__ ({ \
50336 float32x4_t __ret_371; \
5033650337 float32x4_t __s0_371 = __p0_371; \
5033750338 float32x2_t __s2_371 = __p2_371; \
50338 float32x4_t __ret_371; \
5033950339 __ret_371 = vsetq_lane_f32(vget_lane_f32(__s2_371, __p3_371), __s0_371, __p1_371); \
5034050340 __ret_371; \
5034150341})
5034250342#else
5034350343#define vcopyq_lane_f32(__p0_372, __p1_372, __p2_372, __p3_372) __extension__ ({ \
50344 float32x4_t __ret_372; \
5034450345 float32x4_t __s0_372 = __p0_372; \
5034550346 float32x2_t __s2_372 = __p2_372; \
5034650347 float32x4_t __rev0_372; __rev0_372 = __builtin_shufflevector(__s0_372, __s0_372, 3, 2, 1, 0); \
5034750348 float32x2_t __rev2_372; __rev2_372 = __builtin_shufflevector(__s2_372, __s2_372, 1, 0); \
50348 float32x4_t __ret_372; \
5034950349 __ret_372 = __noswap_vsetq_lane_f32(__noswap_vget_lane_f32(__rev2_372, __p3_372), __rev0_372, __p1_372); \
5035050350 __ret_372 = __builtin_shufflevector(__ret_372, __ret_372, 3, 2, 1, 0); \
5035150351 __ret_372; \
......@@ -50354,19 +50354,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5035450354
5035550355#ifdef __LITTLE_ENDIAN__
5035650356#define vcopyq_lane_s32(__p0_373, __p1_373, __p2_373, __p3_373) __extension__ ({ \
50357 int32x4_t __ret_373; \
5035750358 int32x4_t __s0_373 = __p0_373; \
5035850359 int32x2_t __s2_373 = __p2_373; \
50359 int32x4_t __ret_373; \
5036050360 __ret_373 = vsetq_lane_s32(vget_lane_s32(__s2_373, __p3_373), __s0_373, __p1_373); \
5036150361 __ret_373; \
5036250362})
5036350363#else
5036450364#define vcopyq_lane_s32(__p0_374, __p1_374, __p2_374, __p3_374) __extension__ ({ \
50365 int32x4_t __ret_374; \
5036550366 int32x4_t __s0_374 = __p0_374; \
5036650367 int32x2_t __s2_374 = __p2_374; \
5036750368 int32x4_t __rev0_374; __rev0_374 = __builtin_shufflevector(__s0_374, __s0_374, 3, 2, 1, 0); \
5036850369 int32x2_t __rev2_374; __rev2_374 = __builtin_shufflevector(__s2_374, __s2_374, 1, 0); \
50369 int32x4_t __ret_374; \
5037050370 __ret_374 = __noswap_vsetq_lane_s32(__noswap_vget_lane_s32(__rev2_374, __p3_374), __rev0_374, __p1_374); \
5037150371 __ret_374 = __builtin_shufflevector(__ret_374, __ret_374, 3, 2, 1, 0); \
5037250372 __ret_374; \
......@@ -50375,18 +50375,18 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5037550375
5037650376#ifdef __LITTLE_ENDIAN__
5037750377#define vcopyq_lane_s64(__p0_375, __p1_375, __p2_375, __p3_375) __extension__ ({ \
50378 int64x2_t __ret_375; \
5037850379 int64x2_t __s0_375 = __p0_375; \
5037950380 int64x1_t __s2_375 = __p2_375; \
50380 int64x2_t __ret_375; \
5038150381 __ret_375 = vsetq_lane_s64(vget_lane_s64(__s2_375, __p3_375), __s0_375, __p1_375); \
5038250382 __ret_375; \
5038350383})
5038450384#else
5038550385#define vcopyq_lane_s64(__p0_376, __p1_376, __p2_376, __p3_376) __extension__ ({ \
50386 int64x2_t __ret_376; \
5038650387 int64x2_t __s0_376 = __p0_376; \
5038750388 int64x1_t __s2_376 = __p2_376; \
5038850389 int64x2_t __rev0_376; __rev0_376 = __builtin_shufflevector(__s0_376, __s0_376, 1, 0); \
50389 int64x2_t __ret_376; \
5039050390 __ret_376 = __noswap_vsetq_lane_s64(vget_lane_s64(__s2_376, __p3_376), __rev0_376, __p1_376); \
5039150391 __ret_376 = __builtin_shufflevector(__ret_376, __ret_376, 1, 0); \
5039250392 __ret_376; \
......@@ -50395,19 +50395,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5039550395
5039650396#ifdef __LITTLE_ENDIAN__
5039750397#define vcopyq_lane_s16(__p0_377, __p1_377, __p2_377, __p3_377) __extension__ ({ \
50398 int16x8_t __ret_377; \
5039850399 int16x8_t __s0_377 = __p0_377; \
5039950400 int16x4_t __s2_377 = __p2_377; \
50400 int16x8_t __ret_377; \
5040150401 __ret_377 = vsetq_lane_s16(vget_lane_s16(__s2_377, __p3_377), __s0_377, __p1_377); \
5040250402 __ret_377; \
5040350403})
5040450404#else
5040550405#define vcopyq_lane_s16(__p0_378, __p1_378, __p2_378, __p3_378) __extension__ ({ \
50406 int16x8_t __ret_378; \
5040650407 int16x8_t __s0_378 = __p0_378; \
5040750408 int16x4_t __s2_378 = __p2_378; \
5040850409 int16x8_t __rev0_378; __rev0_378 = __builtin_shufflevector(__s0_378, __s0_378, 7, 6, 5, 4, 3, 2, 1, 0); \
5040950410 int16x4_t __rev2_378; __rev2_378 = __builtin_shufflevector(__s2_378, __s2_378, 3, 2, 1, 0); \
50410 int16x8_t __ret_378; \
5041150411 __ret_378 = __noswap_vsetq_lane_s16(__noswap_vget_lane_s16(__rev2_378, __p3_378), __rev0_378, __p1_378); \
5041250412 __ret_378 = __builtin_shufflevector(__ret_378, __ret_378, 7, 6, 5, 4, 3, 2, 1, 0); \
5041350413 __ret_378; \
......@@ -50416,19 +50416,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5041650416
5041750417#ifdef __LITTLE_ENDIAN__
5041850418#define vcopy_lane_p8(__p0_379, __p1_379, __p2_379, __p3_379) __extension__ ({ \
50419 poly8x8_t __ret_379; \
5041950420 poly8x8_t __s0_379 = __p0_379; \
5042050421 poly8x8_t __s2_379 = __p2_379; \
50421 poly8x8_t __ret_379; \
5042250422 __ret_379 = vset_lane_p8(vget_lane_p8(__s2_379, __p3_379), __s0_379, __p1_379); \
5042350423 __ret_379; \
5042450424})
5042550425#else
5042650426#define vcopy_lane_p8(__p0_380, __p1_380, __p2_380, __p3_380) __extension__ ({ \
50427 poly8x8_t __ret_380; \
5042750428 poly8x8_t __s0_380 = __p0_380; \
5042850429 poly8x8_t __s2_380 = __p2_380; \
5042950430 poly8x8_t __rev0_380; __rev0_380 = __builtin_shufflevector(__s0_380, __s0_380, 7, 6, 5, 4, 3, 2, 1, 0); \
5043050431 poly8x8_t __rev2_380; __rev2_380 = __builtin_shufflevector(__s2_380, __s2_380, 7, 6, 5, 4, 3, 2, 1, 0); \
50431 poly8x8_t __ret_380; \
5043250432 __ret_380 = __noswap_vset_lane_p8(__noswap_vget_lane_p8(__rev2_380, __p3_380), __rev0_380, __p1_380); \
5043350433 __ret_380 = __builtin_shufflevector(__ret_380, __ret_380, 7, 6, 5, 4, 3, 2, 1, 0); \
5043450434 __ret_380; \
......@@ -50437,19 +50437,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5043750437
5043850438#ifdef __LITTLE_ENDIAN__
5043950439#define vcopy_lane_p16(__p0_381, __p1_381, __p2_381, __p3_381) __extension__ ({ \
50440 poly16x4_t __ret_381; \
5044050441 poly16x4_t __s0_381 = __p0_381; \
5044150442 poly16x4_t __s2_381 = __p2_381; \
50442 poly16x4_t __ret_381; \
5044350443 __ret_381 = vset_lane_p16(vget_lane_p16(__s2_381, __p3_381), __s0_381, __p1_381); \
5044450444 __ret_381; \
5044550445})
5044650446#else
5044750447#define vcopy_lane_p16(__p0_382, __p1_382, __p2_382, __p3_382) __extension__ ({ \
50448 poly16x4_t __ret_382; \
5044850449 poly16x4_t __s0_382 = __p0_382; \
5044950450 poly16x4_t __s2_382 = __p2_382; \
5045050451 poly16x4_t __rev0_382; __rev0_382 = __builtin_shufflevector(__s0_382, __s0_382, 3, 2, 1, 0); \
5045150452 poly16x4_t __rev2_382; __rev2_382 = __builtin_shufflevector(__s2_382, __s2_382, 3, 2, 1, 0); \
50452 poly16x4_t __ret_382; \
5045350453 __ret_382 = __noswap_vset_lane_p16(__noswap_vget_lane_p16(__rev2_382, __p3_382), __rev0_382, __p1_382); \
5045450454 __ret_382 = __builtin_shufflevector(__ret_382, __ret_382, 3, 2, 1, 0); \
5045550455 __ret_382; \
......@@ -50458,19 +50458,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5045850458
5045950459#ifdef __LITTLE_ENDIAN__
5046050460#define vcopy_lane_u8(__p0_383, __p1_383, __p2_383, __p3_383) __extension__ ({ \
50461 uint8x8_t __ret_383; \
5046150462 uint8x8_t __s0_383 = __p0_383; \
5046250463 uint8x8_t __s2_383 = __p2_383; \
50463 uint8x8_t __ret_383; \
5046450464 __ret_383 = vset_lane_u8(vget_lane_u8(__s2_383, __p3_383), __s0_383, __p1_383); \
5046550465 __ret_383; \
5046650466})
5046750467#else
5046850468#define vcopy_lane_u8(__p0_384, __p1_384, __p2_384, __p3_384) __extension__ ({ \
50469 uint8x8_t __ret_384; \
5046950470 uint8x8_t __s0_384 = __p0_384; \
5047050471 uint8x8_t __s2_384 = __p2_384; \
5047150472 uint8x8_t __rev0_384; __rev0_384 = __builtin_shufflevector(__s0_384, __s0_384, 7, 6, 5, 4, 3, 2, 1, 0); \
5047250473 uint8x8_t __rev2_384; __rev2_384 = __builtin_shufflevector(__s2_384, __s2_384, 7, 6, 5, 4, 3, 2, 1, 0); \
50473 uint8x8_t __ret_384; \
5047450474 __ret_384 = __noswap_vset_lane_u8(__noswap_vget_lane_u8(__rev2_384, __p3_384), __rev0_384, __p1_384); \
5047550475 __ret_384 = __builtin_shufflevector(__ret_384, __ret_384, 7, 6, 5, 4, 3, 2, 1, 0); \
5047650476 __ret_384; \
......@@ -50479,19 +50479,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5047950479
5048050480#ifdef __LITTLE_ENDIAN__
5048150481#define vcopy_lane_u32(__p0_385, __p1_385, __p2_385, __p3_385) __extension__ ({ \
50482 uint32x2_t __ret_385; \
5048250483 uint32x2_t __s0_385 = __p0_385; \
5048350484 uint32x2_t __s2_385 = __p2_385; \
50484 uint32x2_t __ret_385; \
5048550485 __ret_385 = vset_lane_u32(vget_lane_u32(__s2_385, __p3_385), __s0_385, __p1_385); \
5048650486 __ret_385; \
5048750487})
5048850488#else
5048950489#define vcopy_lane_u32(__p0_386, __p1_386, __p2_386, __p3_386) __extension__ ({ \
50490 uint32x2_t __ret_386; \
5049050491 uint32x2_t __s0_386 = __p0_386; \
5049150492 uint32x2_t __s2_386 = __p2_386; \
5049250493 uint32x2_t __rev0_386; __rev0_386 = __builtin_shufflevector(__s0_386, __s0_386, 1, 0); \
5049350494 uint32x2_t __rev2_386; __rev2_386 = __builtin_shufflevector(__s2_386, __s2_386, 1, 0); \
50494 uint32x2_t __ret_386; \
5049550495 __ret_386 = __noswap_vset_lane_u32(__noswap_vget_lane_u32(__rev2_386, __p3_386), __rev0_386, __p1_386); \
5049650496 __ret_386 = __builtin_shufflevector(__ret_386, __ret_386, 1, 0); \
5049750497 __ret_386; \
......@@ -50499,27 +50499,27 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5049950499#endif
5050050500
5050150501#define vcopy_lane_u64(__p0_387, __p1_387, __p2_387, __p3_387) __extension__ ({ \
50502 uint64x1_t __ret_387; \
5050250503 uint64x1_t __s0_387 = __p0_387; \
5050350504 uint64x1_t __s2_387 = __p2_387; \
50504 uint64x1_t __ret_387; \
5050550505 __ret_387 = vset_lane_u64(vget_lane_u64(__s2_387, __p3_387), __s0_387, __p1_387); \
5050650506 __ret_387; \
5050750507})
5050850508#ifdef __LITTLE_ENDIAN__
5050950509#define vcopy_lane_u16(__p0_388, __p1_388, __p2_388, __p3_388) __extension__ ({ \
50510 uint16x4_t __ret_388; \
5051050511 uint16x4_t __s0_388 = __p0_388; \
5051150512 uint16x4_t __s2_388 = __p2_388; \
50512 uint16x4_t __ret_388; \
5051350513 __ret_388 = vset_lane_u16(vget_lane_u16(__s2_388, __p3_388), __s0_388, __p1_388); \
5051450514 __ret_388; \
5051550515})
5051650516#else
5051750517#define vcopy_lane_u16(__p0_389, __p1_389, __p2_389, __p3_389) __extension__ ({ \
50518 uint16x4_t __ret_389; \
5051850519 uint16x4_t __s0_389 = __p0_389; \
5051950520 uint16x4_t __s2_389 = __p2_389; \
5052050521 uint16x4_t __rev0_389; __rev0_389 = __builtin_shufflevector(__s0_389, __s0_389, 3, 2, 1, 0); \
5052150522 uint16x4_t __rev2_389; __rev2_389 = __builtin_shufflevector(__s2_389, __s2_389, 3, 2, 1, 0); \
50522 uint16x4_t __ret_389; \
5052350523 __ret_389 = __noswap_vset_lane_u16(__noswap_vget_lane_u16(__rev2_389, __p3_389), __rev0_389, __p1_389); \
5052450524 __ret_389 = __builtin_shufflevector(__ret_389, __ret_389, 3, 2, 1, 0); \
5052550525 __ret_389; \
......@@ -50528,19 +50528,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5052850528
5052950529#ifdef __LITTLE_ENDIAN__
5053050530#define vcopy_lane_s8(__p0_390, __p1_390, __p2_390, __p3_390) __extension__ ({ \
50531 int8x8_t __ret_390; \
5053150532 int8x8_t __s0_390 = __p0_390; \
5053250533 int8x8_t __s2_390 = __p2_390; \
50533 int8x8_t __ret_390; \
5053450534 __ret_390 = vset_lane_s8(vget_lane_s8(__s2_390, __p3_390), __s0_390, __p1_390); \
5053550535 __ret_390; \
5053650536})
5053750537#else
5053850538#define vcopy_lane_s8(__p0_391, __p1_391, __p2_391, __p3_391) __extension__ ({ \
50539 int8x8_t __ret_391; \
5053950540 int8x8_t __s0_391 = __p0_391; \
5054050541 int8x8_t __s2_391 = __p2_391; \
5054150542 int8x8_t __rev0_391; __rev0_391 = __builtin_shufflevector(__s0_391, __s0_391, 7, 6, 5, 4, 3, 2, 1, 0); \
5054250543 int8x8_t __rev2_391; __rev2_391 = __builtin_shufflevector(__s2_391, __s2_391, 7, 6, 5, 4, 3, 2, 1, 0); \
50543 int8x8_t __ret_391; \
5054450544 __ret_391 = __noswap_vset_lane_s8(__noswap_vget_lane_s8(__rev2_391, __p3_391), __rev0_391, __p1_391); \
5054550545 __ret_391 = __builtin_shufflevector(__ret_391, __ret_391, 7, 6, 5, 4, 3, 2, 1, 0); \
5054650546 __ret_391; \
......@@ -50549,19 +50549,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5054950549
5055050550#ifdef __LITTLE_ENDIAN__
5055150551#define vcopy_lane_f32(__p0_392, __p1_392, __p2_392, __p3_392) __extension__ ({ \
50552 float32x2_t __ret_392; \
5055250553 float32x2_t __s0_392 = __p0_392; \
5055350554 float32x2_t __s2_392 = __p2_392; \
50554 float32x2_t __ret_392; \
5055550555 __ret_392 = vset_lane_f32(vget_lane_f32(__s2_392, __p3_392), __s0_392, __p1_392); \
5055650556 __ret_392; \
5055750557})
5055850558#else
5055950559#define vcopy_lane_f32(__p0_393, __p1_393, __p2_393, __p3_393) __extension__ ({ \
50560 float32x2_t __ret_393; \
5056050561 float32x2_t __s0_393 = __p0_393; \
5056150562 float32x2_t __s2_393 = __p2_393; \
5056250563 float32x2_t __rev0_393; __rev0_393 = __builtin_shufflevector(__s0_393, __s0_393, 1, 0); \
5056350564 float32x2_t __rev2_393; __rev2_393 = __builtin_shufflevector(__s2_393, __s2_393, 1, 0); \
50564 float32x2_t __ret_393; \
5056550565 __ret_393 = __noswap_vset_lane_f32(__noswap_vget_lane_f32(__rev2_393, __p3_393), __rev0_393, __p1_393); \
5056650566 __ret_393 = __builtin_shufflevector(__ret_393, __ret_393, 1, 0); \
5056750567 __ret_393; \
......@@ -50570,19 +50570,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5057050570
5057150571#ifdef __LITTLE_ENDIAN__
5057250572#define vcopy_lane_s32(__p0_394, __p1_394, __p2_394, __p3_394) __extension__ ({ \
50573 int32x2_t __ret_394; \
5057350574 int32x2_t __s0_394 = __p0_394; \
5057450575 int32x2_t __s2_394 = __p2_394; \
50575 int32x2_t __ret_394; \
5057650576 __ret_394 = vset_lane_s32(vget_lane_s32(__s2_394, __p3_394), __s0_394, __p1_394); \
5057750577 __ret_394; \
5057850578})
5057950579#else
5058050580#define vcopy_lane_s32(__p0_395, __p1_395, __p2_395, __p3_395) __extension__ ({ \
50581 int32x2_t __ret_395; \
5058150582 int32x2_t __s0_395 = __p0_395; \
5058250583 int32x2_t __s2_395 = __p2_395; \
5058350584 int32x2_t __rev0_395; __rev0_395 = __builtin_shufflevector(__s0_395, __s0_395, 1, 0); \
5058450585 int32x2_t __rev2_395; __rev2_395 = __builtin_shufflevector(__s2_395, __s2_395, 1, 0); \
50585 int32x2_t __ret_395; \
5058650586 __ret_395 = __noswap_vset_lane_s32(__noswap_vget_lane_s32(__rev2_395, __p3_395), __rev0_395, __p1_395); \
5058750587 __ret_395 = __builtin_shufflevector(__ret_395, __ret_395, 1, 0); \
5058850588 __ret_395; \
......@@ -50590,27 +50590,27 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5059050590#endif
5059150591
5059250592#define vcopy_lane_s64(__p0_396, __p1_396, __p2_396, __p3_396) __extension__ ({ \
50593 int64x1_t __ret_396; \
5059350594 int64x1_t __s0_396 = __p0_396; \
5059450595 int64x1_t __s2_396 = __p2_396; \
50595 int64x1_t __ret_396; \
5059650596 __ret_396 = vset_lane_s64(vget_lane_s64(__s2_396, __p3_396), __s0_396, __p1_396); \
5059750597 __ret_396; \
5059850598})
5059950599#ifdef __LITTLE_ENDIAN__
5060050600#define vcopy_lane_s16(__p0_397, __p1_397, __p2_397, __p3_397) __extension__ ({ \
50601 int16x4_t __ret_397; \
5060150602 int16x4_t __s0_397 = __p0_397; \
5060250603 int16x4_t __s2_397 = __p2_397; \
50603 int16x4_t __ret_397; \
5060450604 __ret_397 = vset_lane_s16(vget_lane_s16(__s2_397, __p3_397), __s0_397, __p1_397); \
5060550605 __ret_397; \
5060650606})
5060750607#else
5060850608#define vcopy_lane_s16(__p0_398, __p1_398, __p2_398, __p3_398) __extension__ ({ \
50609 int16x4_t __ret_398; \
5060950610 int16x4_t __s0_398 = __p0_398; \
5061050611 int16x4_t __s2_398 = __p2_398; \
5061150612 int16x4_t __rev0_398; __rev0_398 = __builtin_shufflevector(__s0_398, __s0_398, 3, 2, 1, 0); \
5061250613 int16x4_t __rev2_398; __rev2_398 = __builtin_shufflevector(__s2_398, __s2_398, 3, 2, 1, 0); \
50613 int16x4_t __ret_398; \
5061450614 __ret_398 = __noswap_vset_lane_s16(__noswap_vget_lane_s16(__rev2_398, __p3_398), __rev0_398, __p1_398); \
5061550615 __ret_398 = __builtin_shufflevector(__ret_398, __ret_398, 3, 2, 1, 0); \
5061650616 __ret_398; \
......@@ -50619,19 +50619,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5061950619
5062050620#ifdef __LITTLE_ENDIAN__
5062150621#define vcopyq_laneq_p8(__p0_399, __p1_399, __p2_399, __p3_399) __extension__ ({ \
50622 poly8x16_t __ret_399; \
5062250623 poly8x16_t __s0_399 = __p0_399; \
5062350624 poly8x16_t __s2_399 = __p2_399; \
50624 poly8x16_t __ret_399; \
5062550625 __ret_399 = vsetq_lane_p8(vgetq_lane_p8(__s2_399, __p3_399), __s0_399, __p1_399); \
5062650626 __ret_399; \
5062750627})
5062850628#else
5062950629#define vcopyq_laneq_p8(__p0_400, __p1_400, __p2_400, __p3_400) __extension__ ({ \
50630 poly8x16_t __ret_400; \
5063050631 poly8x16_t __s0_400 = __p0_400; \
5063150632 poly8x16_t __s2_400 = __p2_400; \
5063250633 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); \
5063350634 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); \
50634 poly8x16_t __ret_400; \
5063550635 __ret_400 = __noswap_vsetq_lane_p8(__noswap_vgetq_lane_p8(__rev2_400, __p3_400), __rev0_400, __p1_400); \
5063650636 __ret_400 = __builtin_shufflevector(__ret_400, __ret_400, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5063750637 __ret_400; \
......@@ -50640,19 +50640,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5064050640
5064150641#ifdef __LITTLE_ENDIAN__
5064250642#define vcopyq_laneq_p16(__p0_401, __p1_401, __p2_401, __p3_401) __extension__ ({ \
50643 poly16x8_t __ret_401; \
5064350644 poly16x8_t __s0_401 = __p0_401; \
5064450645 poly16x8_t __s2_401 = __p2_401; \
50645 poly16x8_t __ret_401; \
5064650646 __ret_401 = vsetq_lane_p16(vgetq_lane_p16(__s2_401, __p3_401), __s0_401, __p1_401); \
5064750647 __ret_401; \
5064850648})
5064950649#else
5065050650#define vcopyq_laneq_p16(__p0_402, __p1_402, __p2_402, __p3_402) __extension__ ({ \
50651 poly16x8_t __ret_402; \
5065150652 poly16x8_t __s0_402 = __p0_402; \
5065250653 poly16x8_t __s2_402 = __p2_402; \
5065350654 poly16x8_t __rev0_402; __rev0_402 = __builtin_shufflevector(__s0_402, __s0_402, 7, 6, 5, 4, 3, 2, 1, 0); \
5065450655 poly16x8_t __rev2_402; __rev2_402 = __builtin_shufflevector(__s2_402, __s2_402, 7, 6, 5, 4, 3, 2, 1, 0); \
50655 poly16x8_t __ret_402; \
5065650656 __ret_402 = __noswap_vsetq_lane_p16(__noswap_vgetq_lane_p16(__rev2_402, __p3_402), __rev0_402, __p1_402); \
5065750657 __ret_402 = __builtin_shufflevector(__ret_402, __ret_402, 7, 6, 5, 4, 3, 2, 1, 0); \
5065850658 __ret_402; \
......@@ -50661,19 +50661,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5066150661
5066250662#ifdef __LITTLE_ENDIAN__
5066350663#define vcopyq_laneq_u8(__p0_403, __p1_403, __p2_403, __p3_403) __extension__ ({ \
50664 uint8x16_t __ret_403; \
5066450665 uint8x16_t __s0_403 = __p0_403; \
5066550666 uint8x16_t __s2_403 = __p2_403; \
50666 uint8x16_t __ret_403; \
5066750667 __ret_403 = vsetq_lane_u8(vgetq_lane_u8(__s2_403, __p3_403), __s0_403, __p1_403); \
5066850668 __ret_403; \
5066950669})
5067050670#else
5067150671#define vcopyq_laneq_u8(__p0_404, __p1_404, __p2_404, __p3_404) __extension__ ({ \
50672 uint8x16_t __ret_404; \
5067250673 uint8x16_t __s0_404 = __p0_404; \
5067350674 uint8x16_t __s2_404 = __p2_404; \
5067450675 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); \
5067550676 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); \
50676 uint8x16_t __ret_404; \
5067750677 __ret_404 = __noswap_vsetq_lane_u8(__noswap_vgetq_lane_u8(__rev2_404, __p3_404), __rev0_404, __p1_404); \
5067850678 __ret_404 = __builtin_shufflevector(__ret_404, __ret_404, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5067950679 __ret_404; \
......@@ -50682,19 +50682,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5068250682
5068350683#ifdef __LITTLE_ENDIAN__
5068450684#define vcopyq_laneq_u32(__p0_405, __p1_405, __p2_405, __p3_405) __extension__ ({ \
50685 uint32x4_t __ret_405; \
5068550686 uint32x4_t __s0_405 = __p0_405; \
5068650687 uint32x4_t __s2_405 = __p2_405; \
50687 uint32x4_t __ret_405; \
5068850688 __ret_405 = vsetq_lane_u32(vgetq_lane_u32(__s2_405, __p3_405), __s0_405, __p1_405); \
5068950689 __ret_405; \
5069050690})
5069150691#else
5069250692#define vcopyq_laneq_u32(__p0_406, __p1_406, __p2_406, __p3_406) __extension__ ({ \
50693 uint32x4_t __ret_406; \
5069350694 uint32x4_t __s0_406 = __p0_406; \
5069450695 uint32x4_t __s2_406 = __p2_406; \
5069550696 uint32x4_t __rev0_406; __rev0_406 = __builtin_shufflevector(__s0_406, __s0_406, 3, 2, 1, 0); \
5069650697 uint32x4_t __rev2_406; __rev2_406 = __builtin_shufflevector(__s2_406, __s2_406, 3, 2, 1, 0); \
50697 uint32x4_t __ret_406; \
5069850698 __ret_406 = __noswap_vsetq_lane_u32(__noswap_vgetq_lane_u32(__rev2_406, __p3_406), __rev0_406, __p1_406); \
5069950699 __ret_406 = __builtin_shufflevector(__ret_406, __ret_406, 3, 2, 1, 0); \
5070050700 __ret_406; \
......@@ -50703,19 +50703,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5070350703
5070450704#ifdef __LITTLE_ENDIAN__
5070550705#define vcopyq_laneq_u64(__p0_407, __p1_407, __p2_407, __p3_407) __extension__ ({ \
50706 uint64x2_t __ret_407; \
5070650707 uint64x2_t __s0_407 = __p0_407; \
5070750708 uint64x2_t __s2_407 = __p2_407; \
50708 uint64x2_t __ret_407; \
5070950709 __ret_407 = vsetq_lane_u64(vgetq_lane_u64(__s2_407, __p3_407), __s0_407, __p1_407); \
5071050710 __ret_407; \
5071150711})
5071250712#else
5071350713#define vcopyq_laneq_u64(__p0_408, __p1_408, __p2_408, __p3_408) __extension__ ({ \
50714 uint64x2_t __ret_408; \
5071450715 uint64x2_t __s0_408 = __p0_408; \
5071550716 uint64x2_t __s2_408 = __p2_408; \
5071650717 uint64x2_t __rev0_408; __rev0_408 = __builtin_shufflevector(__s0_408, __s0_408, 1, 0); \
5071750718 uint64x2_t __rev2_408; __rev2_408 = __builtin_shufflevector(__s2_408, __s2_408, 1, 0); \
50718 uint64x2_t __ret_408; \
5071950719 __ret_408 = __noswap_vsetq_lane_u64(__noswap_vgetq_lane_u64(__rev2_408, __p3_408), __rev0_408, __p1_408); \
5072050720 __ret_408 = __builtin_shufflevector(__ret_408, __ret_408, 1, 0); \
5072150721 __ret_408; \
......@@ -50724,19 +50724,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5072450724
5072550725#ifdef __LITTLE_ENDIAN__
5072650726#define vcopyq_laneq_u16(__p0_409, __p1_409, __p2_409, __p3_409) __extension__ ({ \
50727 uint16x8_t __ret_409; \
5072750728 uint16x8_t __s0_409 = __p0_409; \
5072850729 uint16x8_t __s2_409 = __p2_409; \
50729 uint16x8_t __ret_409; \
5073050730 __ret_409 = vsetq_lane_u16(vgetq_lane_u16(__s2_409, __p3_409), __s0_409, __p1_409); \
5073150731 __ret_409; \
5073250732})
5073350733#else
5073450734#define vcopyq_laneq_u16(__p0_410, __p1_410, __p2_410, __p3_410) __extension__ ({ \
50735 uint16x8_t __ret_410; \
5073550736 uint16x8_t __s0_410 = __p0_410; \
5073650737 uint16x8_t __s2_410 = __p2_410; \
5073750738 uint16x8_t __rev0_410; __rev0_410 = __builtin_shufflevector(__s0_410, __s0_410, 7, 6, 5, 4, 3, 2, 1, 0); \
5073850739 uint16x8_t __rev2_410; __rev2_410 = __builtin_shufflevector(__s2_410, __s2_410, 7, 6, 5, 4, 3, 2, 1, 0); \
50739 uint16x8_t __ret_410; \
5074050740 __ret_410 = __noswap_vsetq_lane_u16(__noswap_vgetq_lane_u16(__rev2_410, __p3_410), __rev0_410, __p1_410); \
5074150741 __ret_410 = __builtin_shufflevector(__ret_410, __ret_410, 7, 6, 5, 4, 3, 2, 1, 0); \
5074250742 __ret_410; \
......@@ -50745,19 +50745,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5074550745
5074650746#ifdef __LITTLE_ENDIAN__
5074750747#define vcopyq_laneq_s8(__p0_411, __p1_411, __p2_411, __p3_411) __extension__ ({ \
50748 int8x16_t __ret_411; \
5074850749 int8x16_t __s0_411 = __p0_411; \
5074950750 int8x16_t __s2_411 = __p2_411; \
50750 int8x16_t __ret_411; \
5075150751 __ret_411 = vsetq_lane_s8(vgetq_lane_s8(__s2_411, __p3_411), __s0_411, __p1_411); \
5075250752 __ret_411; \
5075350753})
5075450754#else
5075550755#define vcopyq_laneq_s8(__p0_412, __p1_412, __p2_412, __p3_412) __extension__ ({ \
50756 int8x16_t __ret_412; \
5075650757 int8x16_t __s0_412 = __p0_412; \
5075750758 int8x16_t __s2_412 = __p2_412; \
5075850759 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); \
5075950760 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); \
50760 int8x16_t __ret_412; \
5076150761 __ret_412 = __noswap_vsetq_lane_s8(__noswap_vgetq_lane_s8(__rev2_412, __p3_412), __rev0_412, __p1_412); \
5076250762 __ret_412 = __builtin_shufflevector(__ret_412, __ret_412, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5076350763 __ret_412; \
......@@ -50766,19 +50766,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5076650766
5076750767#ifdef __LITTLE_ENDIAN__
5076850768#define vcopyq_laneq_f32(__p0_413, __p1_413, __p2_413, __p3_413) __extension__ ({ \
50769 float32x4_t __ret_413; \
5076950770 float32x4_t __s0_413 = __p0_413; \
5077050771 float32x4_t __s2_413 = __p2_413; \
50771 float32x4_t __ret_413; \
5077250772 __ret_413 = vsetq_lane_f32(vgetq_lane_f32(__s2_413, __p3_413), __s0_413, __p1_413); \
5077350773 __ret_413; \
5077450774})
5077550775#else
5077650776#define vcopyq_laneq_f32(__p0_414, __p1_414, __p2_414, __p3_414) __extension__ ({ \
50777 float32x4_t __ret_414; \
5077750778 float32x4_t __s0_414 = __p0_414; \
5077850779 float32x4_t __s2_414 = __p2_414; \
5077950780 float32x4_t __rev0_414; __rev0_414 = __builtin_shufflevector(__s0_414, __s0_414, 3, 2, 1, 0); \
5078050781 float32x4_t __rev2_414; __rev2_414 = __builtin_shufflevector(__s2_414, __s2_414, 3, 2, 1, 0); \
50781 float32x4_t __ret_414; \
5078250782 __ret_414 = __noswap_vsetq_lane_f32(__noswap_vgetq_lane_f32(__rev2_414, __p3_414), __rev0_414, __p1_414); \
5078350783 __ret_414 = __builtin_shufflevector(__ret_414, __ret_414, 3, 2, 1, 0); \
5078450784 __ret_414; \
......@@ -50787,19 +50787,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5078750787
5078850788#ifdef __LITTLE_ENDIAN__
5078950789#define vcopyq_laneq_s32(__p0_415, __p1_415, __p2_415, __p3_415) __extension__ ({ \
50790 int32x4_t __ret_415; \
5079050791 int32x4_t __s0_415 = __p0_415; \
5079150792 int32x4_t __s2_415 = __p2_415; \
50792 int32x4_t __ret_415; \
5079350793 __ret_415 = vsetq_lane_s32(vgetq_lane_s32(__s2_415, __p3_415), __s0_415, __p1_415); \
5079450794 __ret_415; \
5079550795})
5079650796#else
5079750797#define vcopyq_laneq_s32(__p0_416, __p1_416, __p2_416, __p3_416) __extension__ ({ \
50798 int32x4_t __ret_416; \
5079850799 int32x4_t __s0_416 = __p0_416; \
5079950800 int32x4_t __s2_416 = __p2_416; \
5080050801 int32x4_t __rev0_416; __rev0_416 = __builtin_shufflevector(__s0_416, __s0_416, 3, 2, 1, 0); \
5080150802 int32x4_t __rev2_416; __rev2_416 = __builtin_shufflevector(__s2_416, __s2_416, 3, 2, 1, 0); \
50802 int32x4_t __ret_416; \
5080350803 __ret_416 = __noswap_vsetq_lane_s32(__noswap_vgetq_lane_s32(__rev2_416, __p3_416), __rev0_416, __p1_416); \
5080450804 __ret_416 = __builtin_shufflevector(__ret_416, __ret_416, 3, 2, 1, 0); \
5080550805 __ret_416; \
......@@ -50808,19 +50808,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5080850808
5080950809#ifdef __LITTLE_ENDIAN__
5081050810#define vcopyq_laneq_s64(__p0_417, __p1_417, __p2_417, __p3_417) __extension__ ({ \
50811 int64x2_t __ret_417; \
5081150812 int64x2_t __s0_417 = __p0_417; \
5081250813 int64x2_t __s2_417 = __p2_417; \
50813 int64x2_t __ret_417; \
5081450814 __ret_417 = vsetq_lane_s64(vgetq_lane_s64(__s2_417, __p3_417), __s0_417, __p1_417); \
5081550815 __ret_417; \
5081650816})
5081750817#else
5081850818#define vcopyq_laneq_s64(__p0_418, __p1_418, __p2_418, __p3_418) __extension__ ({ \
50819 int64x2_t __ret_418; \
5081950820 int64x2_t __s0_418 = __p0_418; \
5082050821 int64x2_t __s2_418 = __p2_418; \
5082150822 int64x2_t __rev0_418; __rev0_418 = __builtin_shufflevector(__s0_418, __s0_418, 1, 0); \
5082250823 int64x2_t __rev2_418; __rev2_418 = __builtin_shufflevector(__s2_418, __s2_418, 1, 0); \
50823 int64x2_t __ret_418; \
5082450824 __ret_418 = __noswap_vsetq_lane_s64(__noswap_vgetq_lane_s64(__rev2_418, __p3_418), __rev0_418, __p1_418); \
5082550825 __ret_418 = __builtin_shufflevector(__ret_418, __ret_418, 1, 0); \
5082650826 __ret_418; \
......@@ -50829,19 +50829,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5082950829
5083050830#ifdef __LITTLE_ENDIAN__
5083150831#define vcopyq_laneq_s16(__p0_419, __p1_419, __p2_419, __p3_419) __extension__ ({ \
50832 int16x8_t __ret_419; \
5083250833 int16x8_t __s0_419 = __p0_419; \
5083350834 int16x8_t __s2_419 = __p2_419; \
50834 int16x8_t __ret_419; \
5083550835 __ret_419 = vsetq_lane_s16(vgetq_lane_s16(__s2_419, __p3_419), __s0_419, __p1_419); \
5083650836 __ret_419; \
5083750837})
5083850838#else
5083950839#define vcopyq_laneq_s16(__p0_420, __p1_420, __p2_420, __p3_420) __extension__ ({ \
50840 int16x8_t __ret_420; \
5084050841 int16x8_t __s0_420 = __p0_420; \
5084150842 int16x8_t __s2_420 = __p2_420; \
5084250843 int16x8_t __rev0_420; __rev0_420 = __builtin_shufflevector(__s0_420, __s0_420, 7, 6, 5, 4, 3, 2, 1, 0); \
5084350844 int16x8_t __rev2_420; __rev2_420 = __builtin_shufflevector(__s2_420, __s2_420, 7, 6, 5, 4, 3, 2, 1, 0); \
50844 int16x8_t __ret_420; \
5084550845 __ret_420 = __noswap_vsetq_lane_s16(__noswap_vgetq_lane_s16(__rev2_420, __p3_420), __rev0_420, __p1_420); \
5084650846 __ret_420 = __builtin_shufflevector(__ret_420, __ret_420, 7, 6, 5, 4, 3, 2, 1, 0); \
5084750847 __ret_420; \
......@@ -50850,19 +50850,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5085050850
5085150851#ifdef __LITTLE_ENDIAN__
5085250852#define vcopy_laneq_p8(__p0_421, __p1_421, __p2_421, __p3_421) __extension__ ({ \
50853 poly8x8_t __ret_421; \
5085350854 poly8x8_t __s0_421 = __p0_421; \
5085450855 poly8x16_t __s2_421 = __p2_421; \
50855 poly8x8_t __ret_421; \
5085650856 __ret_421 = vset_lane_p8(vgetq_lane_p8(__s2_421, __p3_421), __s0_421, __p1_421); \
5085750857 __ret_421; \
5085850858})
5085950859#else
5086050860#define vcopy_laneq_p8(__p0_422, __p1_422, __p2_422, __p3_422) __extension__ ({ \
50861 poly8x8_t __ret_422; \
5086150862 poly8x8_t __s0_422 = __p0_422; \
5086250863 poly8x16_t __s2_422 = __p2_422; \
5086350864 poly8x8_t __rev0_422; __rev0_422 = __builtin_shufflevector(__s0_422, __s0_422, 7, 6, 5, 4, 3, 2, 1, 0); \
5086450865 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); \
50865 poly8x8_t __ret_422; \
5086650866 __ret_422 = __noswap_vset_lane_p8(__noswap_vgetq_lane_p8(__rev2_422, __p3_422), __rev0_422, __p1_422); \
5086750867 __ret_422 = __builtin_shufflevector(__ret_422, __ret_422, 7, 6, 5, 4, 3, 2, 1, 0); \
5086850868 __ret_422; \
......@@ -50871,19 +50871,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5087150871
5087250872#ifdef __LITTLE_ENDIAN__
5087350873#define vcopy_laneq_p16(__p0_423, __p1_423, __p2_423, __p3_423) __extension__ ({ \
50874 poly16x4_t __ret_423; \
5087450875 poly16x4_t __s0_423 = __p0_423; \
5087550876 poly16x8_t __s2_423 = __p2_423; \
50876 poly16x4_t __ret_423; \
5087750877 __ret_423 = vset_lane_p16(vgetq_lane_p16(__s2_423, __p3_423), __s0_423, __p1_423); \
5087850878 __ret_423; \
5087950879})
5088050880#else
5088150881#define vcopy_laneq_p16(__p0_424, __p1_424, __p2_424, __p3_424) __extension__ ({ \
50882 poly16x4_t __ret_424; \
5088250883 poly16x4_t __s0_424 = __p0_424; \
5088350884 poly16x8_t __s2_424 = __p2_424; \
5088450885 poly16x4_t __rev0_424; __rev0_424 = __builtin_shufflevector(__s0_424, __s0_424, 3, 2, 1, 0); \
5088550886 poly16x8_t __rev2_424; __rev2_424 = __builtin_shufflevector(__s2_424, __s2_424, 7, 6, 5, 4, 3, 2, 1, 0); \
50886 poly16x4_t __ret_424; \
5088750887 __ret_424 = __noswap_vset_lane_p16(__noswap_vgetq_lane_p16(__rev2_424, __p3_424), __rev0_424, __p1_424); \
5088850888 __ret_424 = __builtin_shufflevector(__ret_424, __ret_424, 3, 2, 1, 0); \
5088950889 __ret_424; \
......@@ -50892,19 +50892,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5089250892
5089350893#ifdef __LITTLE_ENDIAN__
5089450894#define vcopy_laneq_u8(__p0_425, __p1_425, __p2_425, __p3_425) __extension__ ({ \
50895 uint8x8_t __ret_425; \
5089550896 uint8x8_t __s0_425 = __p0_425; \
5089650897 uint8x16_t __s2_425 = __p2_425; \
50897 uint8x8_t __ret_425; \
5089850898 __ret_425 = vset_lane_u8(vgetq_lane_u8(__s2_425, __p3_425), __s0_425, __p1_425); \
5089950899 __ret_425; \
5090050900})
5090150901#else
5090250902#define vcopy_laneq_u8(__p0_426, __p1_426, __p2_426, __p3_426) __extension__ ({ \
50903 uint8x8_t __ret_426; \
5090350904 uint8x8_t __s0_426 = __p0_426; \
5090450905 uint8x16_t __s2_426 = __p2_426; \
5090550906 uint8x8_t __rev0_426; __rev0_426 = __builtin_shufflevector(__s0_426, __s0_426, 7, 6, 5, 4, 3, 2, 1, 0); \
5090650907 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); \
50907 uint8x8_t __ret_426; \
5090850908 __ret_426 = __noswap_vset_lane_u8(__noswap_vgetq_lane_u8(__rev2_426, __p3_426), __rev0_426, __p1_426); \
5090950909 __ret_426 = __builtin_shufflevector(__ret_426, __ret_426, 7, 6, 5, 4, 3, 2, 1, 0); \
5091050910 __ret_426; \
......@@ -50913,19 +50913,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5091350913
5091450914#ifdef __LITTLE_ENDIAN__
5091550915#define vcopy_laneq_u32(__p0_427, __p1_427, __p2_427, __p3_427) __extension__ ({ \
50916 uint32x2_t __ret_427; \
5091650917 uint32x2_t __s0_427 = __p0_427; \
5091750918 uint32x4_t __s2_427 = __p2_427; \
50918 uint32x2_t __ret_427; \
5091950919 __ret_427 = vset_lane_u32(vgetq_lane_u32(__s2_427, __p3_427), __s0_427, __p1_427); \
5092050920 __ret_427; \
5092150921})
5092250922#else
5092350923#define vcopy_laneq_u32(__p0_428, __p1_428, __p2_428, __p3_428) __extension__ ({ \
50924 uint32x2_t __ret_428; \
5092450925 uint32x2_t __s0_428 = __p0_428; \
5092550926 uint32x4_t __s2_428 = __p2_428; \
5092650927 uint32x2_t __rev0_428; __rev0_428 = __builtin_shufflevector(__s0_428, __s0_428, 1, 0); \
5092750928 uint32x4_t __rev2_428; __rev2_428 = __builtin_shufflevector(__s2_428, __s2_428, 3, 2, 1, 0); \
50928 uint32x2_t __ret_428; \
5092950929 __ret_428 = __noswap_vset_lane_u32(__noswap_vgetq_lane_u32(__rev2_428, __p3_428), __rev0_428, __p1_428); \
5093050930 __ret_428 = __builtin_shufflevector(__ret_428, __ret_428, 1, 0); \
5093150931 __ret_428; \
......@@ -50934,18 +50934,18 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5093450934
5093550935#ifdef __LITTLE_ENDIAN__
5093650936#define vcopy_laneq_u64(__p0_429, __p1_429, __p2_429, __p3_429) __extension__ ({ \
50937 uint64x1_t __ret_429; \
5093750938 uint64x1_t __s0_429 = __p0_429; \
5093850939 uint64x2_t __s2_429 = __p2_429; \
50939 uint64x1_t __ret_429; \
5094050940 __ret_429 = vset_lane_u64(vgetq_lane_u64(__s2_429, __p3_429), __s0_429, __p1_429); \
5094150941 __ret_429; \
5094250942})
5094350943#else
5094450944#define vcopy_laneq_u64(__p0_430, __p1_430, __p2_430, __p3_430) __extension__ ({ \
50945 uint64x1_t __ret_430; \
5094550946 uint64x1_t __s0_430 = __p0_430; \
5094650947 uint64x2_t __s2_430 = __p2_430; \
5094750948 uint64x2_t __rev2_430; __rev2_430 = __builtin_shufflevector(__s2_430, __s2_430, 1, 0); \
50948 uint64x1_t __ret_430; \
5094950949 __ret_430 = vset_lane_u64(__noswap_vgetq_lane_u64(__rev2_430, __p3_430), __s0_430, __p1_430); \
5095050950 __ret_430; \
5095150951})
......@@ -50953,19 +50953,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5095350953
5095450954#ifdef __LITTLE_ENDIAN__
5095550955#define vcopy_laneq_u16(__p0_431, __p1_431, __p2_431, __p3_431) __extension__ ({ \
50956 uint16x4_t __ret_431; \
5095650957 uint16x4_t __s0_431 = __p0_431; \
5095750958 uint16x8_t __s2_431 = __p2_431; \
50958 uint16x4_t __ret_431; \
5095950959 __ret_431 = vset_lane_u16(vgetq_lane_u16(__s2_431, __p3_431), __s0_431, __p1_431); \
5096050960 __ret_431; \
5096150961})
5096250962#else
5096350963#define vcopy_laneq_u16(__p0_432, __p1_432, __p2_432, __p3_432) __extension__ ({ \
50964 uint16x4_t __ret_432; \
5096450965 uint16x4_t __s0_432 = __p0_432; \
5096550966 uint16x8_t __s2_432 = __p2_432; \
5096650967 uint16x4_t __rev0_432; __rev0_432 = __builtin_shufflevector(__s0_432, __s0_432, 3, 2, 1, 0); \
5096750968 uint16x8_t __rev2_432; __rev2_432 = __builtin_shufflevector(__s2_432, __s2_432, 7, 6, 5, 4, 3, 2, 1, 0); \
50968 uint16x4_t __ret_432; \
5096950969 __ret_432 = __noswap_vset_lane_u16(__noswap_vgetq_lane_u16(__rev2_432, __p3_432), __rev0_432, __p1_432); \
5097050970 __ret_432 = __builtin_shufflevector(__ret_432, __ret_432, 3, 2, 1, 0); \
5097150971 __ret_432; \
......@@ -50974,19 +50974,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5097450974
5097550975#ifdef __LITTLE_ENDIAN__
5097650976#define vcopy_laneq_s8(__p0_433, __p1_433, __p2_433, __p3_433) __extension__ ({ \
50977 int8x8_t __ret_433; \
5097750978 int8x8_t __s0_433 = __p0_433; \
5097850979 int8x16_t __s2_433 = __p2_433; \
50979 int8x8_t __ret_433; \
5098050980 __ret_433 = vset_lane_s8(vgetq_lane_s8(__s2_433, __p3_433), __s0_433, __p1_433); \
5098150981 __ret_433; \
5098250982})
5098350983#else
5098450984#define vcopy_laneq_s8(__p0_434, __p1_434, __p2_434, __p3_434) __extension__ ({ \
50985 int8x8_t __ret_434; \
5098550986 int8x8_t __s0_434 = __p0_434; \
5098650987 int8x16_t __s2_434 = __p2_434; \
5098750988 int8x8_t __rev0_434; __rev0_434 = __builtin_shufflevector(__s0_434, __s0_434, 7, 6, 5, 4, 3, 2, 1, 0); \
5098850989 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); \
50989 int8x8_t __ret_434; \
5099050990 __ret_434 = __noswap_vset_lane_s8(__noswap_vgetq_lane_s8(__rev2_434, __p3_434), __rev0_434, __p1_434); \
5099150991 __ret_434 = __builtin_shufflevector(__ret_434, __ret_434, 7, 6, 5, 4, 3, 2, 1, 0); \
5099250992 __ret_434; \
......@@ -50995,19 +50995,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5099550995
5099650996#ifdef __LITTLE_ENDIAN__
5099750997#define vcopy_laneq_f32(__p0_435, __p1_435, __p2_435, __p3_435) __extension__ ({ \
50998 float32x2_t __ret_435; \
5099850999 float32x2_t __s0_435 = __p0_435; \
5099951000 float32x4_t __s2_435 = __p2_435; \
51000 float32x2_t __ret_435; \
5100151001 __ret_435 = vset_lane_f32(vgetq_lane_f32(__s2_435, __p3_435), __s0_435, __p1_435); \
5100251002 __ret_435; \
5100351003})
5100451004#else
5100551005#define vcopy_laneq_f32(__p0_436, __p1_436, __p2_436, __p3_436) __extension__ ({ \
51006 float32x2_t __ret_436; \
5100651007 float32x2_t __s0_436 = __p0_436; \
5100751008 float32x4_t __s2_436 = __p2_436; \
5100851009 float32x2_t __rev0_436; __rev0_436 = __builtin_shufflevector(__s0_436, __s0_436, 1, 0); \
5100951010 float32x4_t __rev2_436; __rev2_436 = __builtin_shufflevector(__s2_436, __s2_436, 3, 2, 1, 0); \
51010 float32x2_t __ret_436; \
5101151011 __ret_436 = __noswap_vset_lane_f32(__noswap_vgetq_lane_f32(__rev2_436, __p3_436), __rev0_436, __p1_436); \
5101251012 __ret_436 = __builtin_shufflevector(__ret_436, __ret_436, 1, 0); \
5101351013 __ret_436; \
......@@ -51016,19 +51016,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5101651016
5101751017#ifdef __LITTLE_ENDIAN__
5101851018#define vcopy_laneq_s32(__p0_437, __p1_437, __p2_437, __p3_437) __extension__ ({ \
51019 int32x2_t __ret_437; \
5101951020 int32x2_t __s0_437 = __p0_437; \
5102051021 int32x4_t __s2_437 = __p2_437; \
51021 int32x2_t __ret_437; \
5102251022 __ret_437 = vset_lane_s32(vgetq_lane_s32(__s2_437, __p3_437), __s0_437, __p1_437); \
5102351023 __ret_437; \
5102451024})
5102551025#else
5102651026#define vcopy_laneq_s32(__p0_438, __p1_438, __p2_438, __p3_438) __extension__ ({ \
51027 int32x2_t __ret_438; \
5102751028 int32x2_t __s0_438 = __p0_438; \
5102851029 int32x4_t __s2_438 = __p2_438; \
5102951030 int32x2_t __rev0_438; __rev0_438 = __builtin_shufflevector(__s0_438, __s0_438, 1, 0); \
5103051031 int32x4_t __rev2_438; __rev2_438 = __builtin_shufflevector(__s2_438, __s2_438, 3, 2, 1, 0); \
51031 int32x2_t __ret_438; \
5103251032 __ret_438 = __noswap_vset_lane_s32(__noswap_vgetq_lane_s32(__rev2_438, __p3_438), __rev0_438, __p1_438); \
5103351033 __ret_438 = __builtin_shufflevector(__ret_438, __ret_438, 1, 0); \
5103451034 __ret_438; \
......@@ -51037,18 +51037,18 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5103751037
5103851038#ifdef __LITTLE_ENDIAN__
5103951039#define vcopy_laneq_s64(__p0_439, __p1_439, __p2_439, __p3_439) __extension__ ({ \
51040 int64x1_t __ret_439; \
5104051041 int64x1_t __s0_439 = __p0_439; \
5104151042 int64x2_t __s2_439 = __p2_439; \
51042 int64x1_t __ret_439; \
5104351043 __ret_439 = vset_lane_s64(vgetq_lane_s64(__s2_439, __p3_439), __s0_439, __p1_439); \
5104451044 __ret_439; \
5104551045})
5104651046#else
5104751047#define vcopy_laneq_s64(__p0_440, __p1_440, __p2_440, __p3_440) __extension__ ({ \
51048 int64x1_t __ret_440; \
5104851049 int64x1_t __s0_440 = __p0_440; \
5104951050 int64x2_t __s2_440 = __p2_440; \
5105051051 int64x2_t __rev2_440; __rev2_440 = __builtin_shufflevector(__s2_440, __s2_440, 1, 0); \
51051 int64x1_t __ret_440; \
5105251052 __ret_440 = vset_lane_s64(__noswap_vgetq_lane_s64(__rev2_440, __p3_440), __s0_440, __p1_440); \
5105351053 __ret_440; \
5105451054})
......@@ -51056,19 +51056,19 @@ __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
5105651056
5105751057#ifdef __LITTLE_ENDIAN__
5105851058#define vcopy_laneq_s16(__p0_441, __p1_441, __p2_441, __p3_441) __extension__ ({ \
51059 int16x4_t __ret_441; \
5105951060 int16x4_t __s0_441 = __p0_441; \
5106051061 int16x8_t __s2_441 = __p2_441; \
51061 int16x4_t __ret_441; \
5106251062 __ret_441 = vset_lane_s16(vgetq_lane_s16(__s2_441, __p3_441), __s0_441, __p1_441); \
5106351063 __ret_441; \
5106451064})
5106551065#else
5106651066#define vcopy_laneq_s16(__p0_442, __p1_442, __p2_442, __p3_442) __extension__ ({ \
51067 int16x4_t __ret_442; \
5106751068 int16x4_t __s0_442 = __p0_442; \
5106851069 int16x8_t __s2_442 = __p2_442; \
5106951070 int16x4_t __rev0_442; __rev0_442 = __builtin_shufflevector(__s0_442, __s0_442, 3, 2, 1, 0); \
5107051071 int16x8_t __rev2_442; __rev2_442 = __builtin_shufflevector(__s2_442, __s2_442, 7, 6, 5, 4, 3, 2, 1, 0); \
51071 int16x4_t __ret_442; \
5107251072 __ret_442 = __noswap_vset_lane_s16(__noswap_vgetq_lane_s16(__rev2_442, __p3_442), __rev0_442, __p1_442); \
5107351073 __ret_442 = __builtin_shufflevector(__ret_442, __ret_442, 3, 2, 1, 0); \
5107451074 __ret_442; \
......@@ -51105,8 +51105,8 @@ __ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
5110551105}
5110651106#else
5110751107__ai float32x2_t vcvt_f32_f64(float64x2_t __p0) {
51108 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5110951108 float32x2_t __ret;
51109 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5111051110 __ret = (float32x2_t) __builtin_neon_vcvt_f32_f64((int8x16_t)__rev0, 9);
5111151111 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5111251112 return __ret;
......@@ -51136,8 +51136,8 @@ __ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
5113651136}
5113751137#else
5113851138__ai float64x2_t vcvtq_f64_u64(uint64x2_t __p0) {
51139 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5114051139 float64x2_t __ret;
51140 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5114151141 __ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__rev0, 51);
5114251142 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5114351143 return __ret;
......@@ -51152,8 +51152,8 @@ __ai float64x2_t vcvtq_f64_s64(int64x2_t __p0) {
5115251152}
5115351153#else
5115451154__ai float64x2_t vcvtq_f64_s64(int64x2_t __p0) {
51155 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5115651155 float64x2_t __ret;
51156 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5115751157 __ret = (float64x2_t) __builtin_neon_vcvtq_f64_v((int8x16_t)__rev0, 35);
5115851158 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5115951159 return __ret;
......@@ -51178,8 +51178,8 @@ __ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
5117851178}
5117951179#else
5118051180__ai float64x2_t vcvt_f64_f32(float32x2_t __p0) {
51181 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5118251181 float64x2_t __ret;
51182 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5118351183 __ret = (float64x2_t) __builtin_neon_vcvt_f64_f32((int8x8_t)__rev0, 42);
5118451184 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5118551185 return __ret;
......@@ -51199,9 +51199,9 @@ __ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
5119951199}
5120051200#else
5120151201__ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
51202 float16x8_t __ret;
5120251203 float16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5120351204 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
51204 float16x8_t __ret;
5120551205 __ret = __noswap_vcombine_f16(__rev0, __noswap_vcvt_f16_f32(__rev1));
5120651206 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5120751207 return __ret;
......@@ -51216,8 +51216,8 @@ __ai float32x4_t vcvt_high_f32_f16(float16x8_t __p0) {
5121651216}
5121751217#else
5121851218__ai float32x4_t vcvt_high_f32_f16(float16x8_t __p0) {
51219 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5122051219 float32x4_t __ret;
51220 float16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5122151221 __ret = __noswap_vcvt_f32_f16(__noswap_vget_high_f16(__rev0));
5122251222 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5122351223 return __ret;
......@@ -51232,9 +51232,9 @@ __ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
5123251232}
5123351233#else
5123451234__ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
51235 float32x4_t __ret;
5123551236 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5123651237 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51237 float32x4_t __ret;
5123851238 __ret = __noswap_vcombine_f32(__rev0, __noswap_vcvt_f32_f64(__rev1));
5123951239 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5124051240 return __ret;
......@@ -51249,8 +51249,8 @@ __ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
5124951249}
5125051250#else
5125151251__ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
51252 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5125351252 float64x2_t __ret;
51253 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5125451254 __ret = __noswap_vcvt_f64_f32(__noswap_vget_high_f32(__rev0));
5125551255 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5125651256 return __ret;
......@@ -51258,29 +51258,29 @@ __ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
5125851258#endif
5125951259
5126051260#define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
51261 uint32_t __s0 = __p0; \
5126251261 float32_t __ret; \
51262 uint32_t __s0 = __p0; \
5126351263 __ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
5126451264 __ret; \
5126551265})
5126651266#define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
51267 int32_t __s0 = __p0; \
5126851267 float32_t __ret; \
51268 int32_t __s0 = __p0; \
5126951269 __ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
5127051270 __ret; \
5127151271})
5127251272#ifdef __LITTLE_ENDIAN__
5127351273#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
51274 uint64x2_t __s0 = __p0; \
5127551274 float64x2_t __ret; \
51275 uint64x2_t __s0 = __p0; \
5127651276 __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 51); \
5127751277 __ret; \
5127851278})
5127951279#else
5128051280#define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
51281 float64x2_t __ret; \
5128151282 uint64x2_t __s0 = __p0; \
5128251283 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51283 float64x2_t __ret; \
5128451284 __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 51); \
5128551285 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5128651286 __ret; \
......@@ -51289,16 +51289,16 @@ __ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
5128951289
5129051290#ifdef __LITTLE_ENDIAN__
5129151291#define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
51292 int64x2_t __s0 = __p0; \
5129351292 float64x2_t __ret; \
51293 int64x2_t __s0 = __p0; \
5129451294 __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 35); \
5129551295 __ret; \
5129651296})
5129751297#else
5129851298#define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
51299 float64x2_t __ret; \
5129951300 int64x2_t __s0 = __p0; \
5130051301 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51301 float64x2_t __ret; \
5130251302 __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 35); \
5130351303 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5130451304 __ret; \
......@@ -51306,47 +51306,47 @@ __ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
5130651306#endif
5130751307
5130851308#define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
51309 uint64x1_t __s0 = __p0; \
5131051309 float64x1_t __ret; \
51310 uint64x1_t __s0 = __p0; \
5131151311 __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
5131251312 __ret; \
5131351313})
5131451314#define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
51315 int64x1_t __s0 = __p0; \
5131651315 float64x1_t __ret; \
51316 int64x1_t __s0 = __p0; \
5131751317 __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
5131851318 __ret; \
5131951319})
5132051320#define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
51321 uint64_t __s0 = __p0; \
5132251321 float64_t __ret; \
51322 uint64_t __s0 = __p0; \
5132351323 __ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
5132451324 __ret; \
5132551325})
5132651326#define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
51327 int64_t __s0 = __p0; \
5132851327 float64_t __ret; \
51328 int64_t __s0 = __p0; \
5132951329 __ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
5133051330 __ret; \
5133151331})
5133251332#define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
51333 float32_t __s0 = __p0; \
5133451333 int32_t __ret; \
51334 float32_t __s0 = __p0; \
5133551335 __ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
5133651336 __ret; \
5133751337})
5133851338#ifdef __LITTLE_ENDIAN__
5133951339#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
51340 float64x2_t __s0 = __p0; \
5134151340 int64x2_t __ret; \
51341 float64x2_t __s0 = __p0; \
5134251342 __ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__s0, __p1, 35); \
5134351343 __ret; \
5134451344})
5134551345#else
5134651346#define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
51347 int64x2_t __ret; \
5134751348 float64x2_t __s0 = __p0; \
5134851349 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51349 int64x2_t __ret; \
5135051350 __ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__rev0, __p1, 35); \
5135151351 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5135251352 __ret; \
......@@ -51354,35 +51354,35 @@ __ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
5135451354#endif
5135551355
5135651356#define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
51357 float64x1_t __s0 = __p0; \
5135851357 int64x1_t __ret; \
51358 float64x1_t __s0 = __p0; \
5135951359 __ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
5136051360 __ret; \
5136151361})
5136251362#define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
51363 float64_t __s0 = __p0; \
5136451363 int64_t __ret; \
51364 float64_t __s0 = __p0; \
5136551365 __ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
5136651366 __ret; \
5136751367})
5136851368#define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
51369 float32_t __s0 = __p0; \
5137051369 uint32_t __ret; \
51370 float32_t __s0 = __p0; \
5137151371 __ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
5137251372 __ret; \
5137351373})
5137451374#ifdef __LITTLE_ENDIAN__
5137551375#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
51376 float64x2_t __s0 = __p0; \
5137751376 uint64x2_t __ret; \
51377 float64x2_t __s0 = __p0; \
5137851378 __ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__s0, __p1, 51); \
5137951379 __ret; \
5138051380})
5138151381#else
5138251382#define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
51383 uint64x2_t __ret; \
5138351384 float64x2_t __s0 = __p0; \
5138451385 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51385 uint64x2_t __ret; \
5138651386 __ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__rev0, __p1, 51); \
5138751387 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5138851388 __ret; \
......@@ -51390,14 +51390,14 @@ __ai float64x2_t vcvt_high_f64_f32(float32x4_t __p0) {
5139051390#endif
5139151391
5139251392#define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
51393 float64x1_t __s0 = __p0; \
5139451393 uint64x1_t __ret; \
51394 float64x1_t __s0 = __p0; \
5139551395 __ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
5139651396 __ret; \
5139751397})
5139851398#define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
51399 float64_t __s0 = __p0; \
5140051399 uint64_t __ret; \
51400 float64_t __s0 = __p0; \
5140151401 __ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
5140251402 __ret; \
5140351403})
......@@ -51419,8 +51419,8 @@ __ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
5141951419}
5142051420#else
5142151421__ai int64x2_t vcvtq_s64_f64(float64x2_t __p0) {
51422 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5142351422 int64x2_t __ret;
51423 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5142451424 __ret = (int64x2_t) __builtin_neon_vcvtq_s64_v((int8x16_t)__rev0, 35);
5142551425 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5142651426 return __ret;
......@@ -51450,8 +51450,8 @@ __ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
5145051450}
5145151451#else
5145251452__ai uint64x2_t vcvtq_u64_f64(float64x2_t __p0) {
51453 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5145451453 uint64x2_t __ret;
51454 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5145551455 __ret = (uint64x2_t) __builtin_neon_vcvtq_u64_v((int8x16_t)__rev0, 51);
5145651456 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5145751457 return __ret;
......@@ -51556,8 +51556,8 @@ __ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
5155651556}
5155751557#else
5155851558__ai float32x2_t vcvtx_f32_f64(float64x2_t __p0) {
51559 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5156051559 float32x2_t __ret;
51560 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5156151561 __ret = (float32x2_t) __builtin_neon_vcvtx_f32_v((int8x16_t)__rev0, 42);
5156251562 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5156351563 return __ret;
......@@ -51577,9 +51577,9 @@ __ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
5157751577}
5157851578#else
5157951579__ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
51580 float32x4_t __ret;
5158051581 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5158151582 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51582 float32x4_t __ret;
5158351583 __ret = __noswap_vcombine_f32(__rev0, __noswap_vcvtx_f32_f64(__rev1));
5158451584 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5158551585 return __ret;
......@@ -51594,9 +51594,9 @@ __ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
5159451594}
5159551595#else
5159651596__ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
51597 float64x2_t __ret;
5159751598 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5159851599 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51599 float64x2_t __ret;
5160051600 __ret = __rev0 / __rev1;
5160151601 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5160251602 return __ret;
......@@ -51611,9 +51611,9 @@ __ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
5161151611}
5161251612#else
5161351613__ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
51614 float32x4_t __ret;
5161451615 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5161551616 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
51616 float32x4_t __ret;
5161751617 __ret = __rev0 / __rev1;
5161851618 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5161951619 return __ret;
......@@ -51633,9 +51633,9 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5163351633}
5163451634#else
5163551635__ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
51636 float32x2_t __ret;
5163651637 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5163751638 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51638 float32x2_t __ret;
5163951639 __ret = __rev0 / __rev1;
5164051640 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5164151641 return __ret;
......@@ -51644,16 +51644,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5164451644
5164551645#ifdef __LITTLE_ENDIAN__
5164651646#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
51647 poly8x8_t __s0 = __p0; \
5164851647 poly8_t __ret; \
51648 poly8x8_t __s0 = __p0; \
5164951649 __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__s0, __p1); \
5165051650 __ret; \
5165151651})
5165251652#else
5165351653#define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
51654 poly8_t __ret; \
5165451655 poly8x8_t __s0 = __p0; \
5165551656 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51656 poly8_t __ret; \
5165751657 __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((poly8x8_t)__rev0, __p1); \
5165851658 __ret; \
5165951659})
......@@ -51661,16 +51661,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5166151661
5166251662#ifdef __LITTLE_ENDIAN__
5166351663#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
51664 poly16x4_t __s0 = __p0; \
5166551664 poly16_t __ret; \
51665 poly16x4_t __s0 = __p0; \
5166651666 __ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__s0, __p1); \
5166751667 __ret; \
5166851668})
5166951669#else
5167051670#define vduph_lane_p16(__p0, __p1) __extension__ ({ \
51671 poly16_t __ret; \
5167151672 poly16x4_t __s0 = __p0; \
5167251673 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51673 poly16_t __ret; \
5167451674 __ret = (poly16_t) __builtin_neon_vduph_lane_i16((poly16x4_t)__rev0, __p1); \
5167551675 __ret; \
5167651676})
......@@ -51678,16 +51678,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5167851678
5167951679#ifdef __LITTLE_ENDIAN__
5168051680#define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
51681 uint8x8_t __s0 = __p0; \
5168251681 uint8_t __ret; \
51682 uint8x8_t __s0 = __p0; \
5168351683 __ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
5168451684 __ret; \
5168551685})
5168651686#else
5168751687#define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
51688 uint8_t __ret; \
5168851689 uint8x8_t __s0 = __p0; \
5168951690 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51690 uint8_t __ret; \
5169151691 __ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
5169251692 __ret; \
5169351693})
......@@ -51695,39 +51695,39 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5169551695
5169651696#ifdef __LITTLE_ENDIAN__
5169751697#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
51698 uint32x2_t __s0 = __p0; \
5169951698 uint32_t __ret; \
51699 uint32x2_t __s0 = __p0; \
5170051700 __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
5170151701 __ret; \
5170251702})
5170351703#else
5170451704#define vdups_lane_u32(__p0, __p1) __extension__ ({ \
51705 uint32_t __ret; \
5170551706 uint32x2_t __s0 = __p0; \
5170651707 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51707 uint32_t __ret; \
5170851708 __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
5170951709 __ret; \
5171051710})
5171151711#endif
5171251712
5171351713#define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
51714 uint64x1_t __s0 = __p0; \
5171551714 uint64_t __ret; \
51715 uint64x1_t __s0 = __p0; \
5171651716 __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
5171751717 __ret; \
5171851718})
5171951719#ifdef __LITTLE_ENDIAN__
5172051720#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
51721 uint16x4_t __s0 = __p0; \
5172251721 uint16_t __ret; \
51722 uint16x4_t __s0 = __p0; \
5172351723 __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
5172451724 __ret; \
5172551725})
5172651726#else
5172751727#define vduph_lane_u16(__p0, __p1) __extension__ ({ \
51728 uint16_t __ret; \
5172851729 uint16x4_t __s0 = __p0; \
5172951730 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51730 uint16_t __ret; \
5173151731 __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
5173251732 __ret; \
5173351733})
......@@ -51735,39 +51735,39 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5173551735
5173651736#ifdef __LITTLE_ENDIAN__
5173751737#define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
51738 int8x8_t __s0 = __p0; \
5173951738 int8_t __ret; \
51739 int8x8_t __s0 = __p0; \
5174051740 __ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
5174151741 __ret; \
5174251742})
5174351743#else
5174451744#define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
51745 int8_t __ret; \
5174551746 int8x8_t __s0 = __p0; \
5174651747 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51747 int8_t __ret; \
5174851748 __ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
5174951749 __ret; \
5175051750})
5175151751#endif
5175251752
5175351753#define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
51754 float64x1_t __s0 = __p0; \
5175551754 float64_t __ret; \
51755 float64x1_t __s0 = __p0; \
5175651756 __ret = (float64_t) __builtin_neon_vdupd_lane_f64((float64x1_t)__s0, __p1); \
5175751757 __ret; \
5175851758})
5175951759#ifdef __LITTLE_ENDIAN__
5176051760#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
51761 float32x2_t __s0 = __p0; \
5176251761 float32_t __ret; \
51762 float32x2_t __s0 = __p0; \
5176351763 __ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__s0, __p1); \
5176451764 __ret; \
5176551765})
5176651766#else
5176751767#define vdups_lane_f32(__p0, __p1) __extension__ ({ \
51768 float32_t __ret; \
5176851769 float32x2_t __s0 = __p0; \
5176951770 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51770 float32_t __ret; \
5177151771 __ret = (float32_t) __builtin_neon_vdups_lane_f32((float32x2_t)__rev0, __p1); \
5177251772 __ret; \
5177351773})
......@@ -51775,61 +51775,61 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5177551775
5177651776#ifdef __LITTLE_ENDIAN__
5177751777#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
51778 int32x2_t __s0 = __p0; \
5177951778 int32_t __ret; \
51779 int32x2_t __s0 = __p0; \
5178051780 __ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__s0, __p1); \
5178151781 __ret; \
5178251782})
5178351783#else
5178451784#define vdups_lane_s32(__p0, __p1) __extension__ ({ \
51785 int32_t __ret; \
5178551786 int32x2_t __s0 = __p0; \
5178651787 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51787 int32_t __ret; \
5178851788 __ret = (int32_t) __builtin_neon_vdups_lane_i32((int32x2_t)__rev0, __p1); \
5178951789 __ret; \
5179051790})
5179151791#endif
5179251792
5179351793#define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
51794 int64x1_t __s0 = __p0; \
5179551794 int64_t __ret; \
51795 int64x1_t __s0 = __p0; \
5179651796 __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int64x1_t)__s0, __p1); \
5179751797 __ret; \
5179851798})
5179951799#ifdef __LITTLE_ENDIAN__
5180051800#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
51801 int16x4_t __s0 = __p0; \
5180251801 int16_t __ret; \
51802 int16x4_t __s0 = __p0; \
5180351803 __ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__s0, __p1); \
5180451804 __ret; \
5180551805})
5180651806#else
5180751807#define vduph_lane_s16(__p0, __p1) __extension__ ({ \
51808 int16_t __ret; \
5180851809 int16x4_t __s0 = __p0; \
5180951810 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51810 int16_t __ret; \
5181151811 __ret = (int16_t) __builtin_neon_vduph_lane_i16((int16x4_t)__rev0, __p1); \
5181251812 __ret; \
5181351813})
5181451814#endif
5181551815
5181651816#define vdup_lane_p64(__p0_443, __p1_443) __extension__ ({ \
51817 poly64x1_t __s0_443 = __p0_443; \
5181851817 poly64x1_t __ret_443; \
51818 poly64x1_t __s0_443 = __p0_443; \
5181951819 __ret_443 = splat_lane_p64(__s0_443, __p1_443); \
5182051820 __ret_443; \
5182151821})
5182251822#ifdef __LITTLE_ENDIAN__
5182351823#define vdupq_lane_p64(__p0_444, __p1_444) __extension__ ({ \
51824 poly64x1_t __s0_444 = __p0_444; \
5182551824 poly64x2_t __ret_444; \
51825 poly64x1_t __s0_444 = __p0_444; \
5182651826 __ret_444 = splatq_lane_p64(__s0_444, __p1_444); \
5182751827 __ret_444; \
5182851828})
5182951829#else
5183051830#define vdupq_lane_p64(__p0_445, __p1_445) __extension__ ({ \
51831 poly64x1_t __s0_445 = __p0_445; \
5183251831 poly64x2_t __ret_445; \
51832 poly64x1_t __s0_445 = __p0_445; \
5183351833 __ret_445 = __noswap_splatq_lane_p64(__s0_445, __p1_445); \
5183451834 __ret_445 = __builtin_shufflevector(__ret_445, __ret_445, 1, 0); \
5183551835 __ret_445; \
......@@ -51838,15 +51838,15 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5183851838
5183951839#ifdef __LITTLE_ENDIAN__
5184051840#define vdupq_lane_f64(__p0_446, __p1_446) __extension__ ({ \
51841 float64x1_t __s0_446 = __p0_446; \
5184251841 float64x2_t __ret_446; \
51842 float64x1_t __s0_446 = __p0_446; \
5184351843 __ret_446 = splatq_lane_f64(__s0_446, __p1_446); \
5184451844 __ret_446; \
5184551845})
5184651846#else
5184751847#define vdupq_lane_f64(__p0_447, __p1_447) __extension__ ({ \
51848 float64x1_t __s0_447 = __p0_447; \
5184951848 float64x2_t __ret_447; \
51849 float64x1_t __s0_447 = __p0_447; \
5185051850 __ret_447 = __noswap_splatq_lane_f64(__s0_447, __p1_447); \
5185151851 __ret_447 = __builtin_shufflevector(__ret_447, __ret_447, 1, 0); \
5185251852 __ret_447; \
......@@ -51855,16 +51855,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5185551855
5185651856#ifdef __LITTLE_ENDIAN__
5185751857#define vdupq_lane_f16(__p0_448, __p1_448) __extension__ ({ \
51858 float16x4_t __s0_448 = __p0_448; \
5185951858 float16x8_t __ret_448; \
51859 float16x4_t __s0_448 = __p0_448; \
5186051860 __ret_448 = splatq_lane_f16(__s0_448, __p1_448); \
5186151861 __ret_448; \
5186251862})
5186351863#else
5186451864#define vdupq_lane_f16(__p0_449, __p1_449) __extension__ ({ \
51865 float16x8_t __ret_449; \
5186551866 float16x4_t __s0_449 = __p0_449; \
5186651867 float16x4_t __rev0_449; __rev0_449 = __builtin_shufflevector(__s0_449, __s0_449, 3, 2, 1, 0); \
51867 float16x8_t __ret_449; \
5186851868 __ret_449 = __noswap_splatq_lane_f16(__rev0_449, __p1_449); \
5186951869 __ret_449 = __builtin_shufflevector(__ret_449, __ret_449, 7, 6, 5, 4, 3, 2, 1, 0); \
5187051870 __ret_449; \
......@@ -51872,23 +51872,23 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5187251872#endif
5187351873
5187451874#define vdup_lane_f64(__p0_450, __p1_450) __extension__ ({ \
51875 float64x1_t __s0_450 = __p0_450; \
5187651875 float64x1_t __ret_450; \
51876 float64x1_t __s0_450 = __p0_450; \
5187751877 __ret_450 = splat_lane_f64(__s0_450, __p1_450); \
5187851878 __ret_450; \
5187951879})
5188051880#ifdef __LITTLE_ENDIAN__
5188151881#define vdup_lane_f16(__p0_451, __p1_451) __extension__ ({ \
51882 float16x4_t __s0_451 = __p0_451; \
5188351882 float16x4_t __ret_451; \
51883 float16x4_t __s0_451 = __p0_451; \
5188451884 __ret_451 = splat_lane_f16(__s0_451, __p1_451); \
5188551885 __ret_451; \
5188651886})
5188751887#else
5188851888#define vdup_lane_f16(__p0_452, __p1_452) __extension__ ({ \
51889 float16x4_t __ret_452; \
5188951890 float16x4_t __s0_452 = __p0_452; \
5189051891 float16x4_t __rev0_452; __rev0_452 = __builtin_shufflevector(__s0_452, __s0_452, 3, 2, 1, 0); \
51891 float16x4_t __ret_452; \
5189251892 __ret_452 = __noswap_splat_lane_f16(__rev0_452, __p1_452); \
5189351893 __ret_452 = __builtin_shufflevector(__ret_452, __ret_452, 3, 2, 1, 0); \
5189451894 __ret_452; \
......@@ -51897,16 +51897,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5189751897
5189851898#ifdef __LITTLE_ENDIAN__
5189951899#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
51900 poly8x16_t __s0 = __p0; \
5190151900 poly8_t __ret; \
51901 poly8x16_t __s0 = __p0; \
5190251902 __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__s0, __p1); \
5190351903 __ret; \
5190451904})
5190551905#else
5190651906#define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
51907 poly8_t __ret; \
5190751908 poly8x16_t __s0 = __p0; \
5190851909 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
51909 poly8_t __ret; \
5191051910 __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((poly8x16_t)__rev0, __p1); \
5191151911 __ret; \
5191251912})
......@@ -51914,16 +51914,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5191451914
5191551915#ifdef __LITTLE_ENDIAN__
5191651916#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
51917 poly16x8_t __s0 = __p0; \
5191851917 poly16_t __ret; \
51918 poly16x8_t __s0 = __p0; \
5191951919 __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__s0, __p1); \
5192051920 __ret; \
5192151921})
5192251922#else
5192351923#define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
51924 poly16_t __ret; \
5192451925 poly16x8_t __s0 = __p0; \
5192551926 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51926 poly16_t __ret; \
5192751927 __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((poly16x8_t)__rev0, __p1); \
5192851928 __ret; \
5192951929})
......@@ -51931,16 +51931,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5193151931
5193251932#ifdef __LITTLE_ENDIAN__
5193351933#define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
51934 uint8x16_t __s0 = __p0; \
5193551934 uint8_t __ret; \
51935 uint8x16_t __s0 = __p0; \
5193651936 __ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
5193751937 __ret; \
5193851938})
5193951939#else
5194051940#define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
51941 uint8_t __ret; \
5194151942 uint8x16_t __s0 = __p0; \
5194251943 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
51943 uint8_t __ret; \
5194451944 __ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
5194551945 __ret; \
5194651946})
......@@ -51948,16 +51948,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5194851948
5194951949#ifdef __LITTLE_ENDIAN__
5195051950#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
51951 uint32x4_t __s0 = __p0; \
5195251951 uint32_t __ret; \
51952 uint32x4_t __s0 = __p0; \
5195351953 __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
5195451954 __ret; \
5195551955})
5195651956#else
5195751957#define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
51958 uint32_t __ret; \
5195851959 uint32x4_t __s0 = __p0; \
5195951960 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
51960 uint32_t __ret; \
5196151961 __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
5196251962 __ret; \
5196351963})
......@@ -51965,16 +51965,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5196551965
5196651966#ifdef __LITTLE_ENDIAN__
5196751967#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
51968 uint64x2_t __s0 = __p0; \
5196951968 uint64_t __ret; \
51969 uint64x2_t __s0 = __p0; \
5197051970 __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
5197151971 __ret; \
5197251972})
5197351973#else
5197451974#define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
51975 uint64_t __ret; \
5197551976 uint64x2_t __s0 = __p0; \
5197651977 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
51977 uint64_t __ret; \
5197851978 __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
5197951979 __ret; \
5198051980})
......@@ -51982,16 +51982,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5198251982
5198351983#ifdef __LITTLE_ENDIAN__
5198451984#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
51985 uint16x8_t __s0 = __p0; \
5198651985 uint16_t __ret; \
51986 uint16x8_t __s0 = __p0; \
5198751987 __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
5198851988 __ret; \
5198951989})
5199051990#else
5199151991#define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
51992 uint16_t __ret; \
5199251993 uint16x8_t __s0 = __p0; \
5199351994 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
51994 uint16_t __ret; \
5199551995 __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
5199651996 __ret; \
5199751997})
......@@ -51999,16 +51999,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5199951999
5200052000#ifdef __LITTLE_ENDIAN__
5200152001#define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
52002 int8x16_t __s0 = __p0; \
5200352002 int8_t __ret; \
52003 int8x16_t __s0 = __p0; \
5200452004 __ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
5200552005 __ret; \
5200652006})
5200752007#else
5200852008#define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
52009 int8_t __ret; \
5200952010 int8x16_t __s0 = __p0; \
5201052011 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
52011 int8_t __ret; \
5201252012 __ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
5201352013 __ret; \
5201452014})
......@@ -52016,16 +52016,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5201652016
5201752017#ifdef __LITTLE_ENDIAN__
5201852018#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
52019 float64x2_t __s0 = __p0; \
5202052019 float64_t __ret; \
52020 float64x2_t __s0 = __p0; \
5202152021 __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__s0, __p1); \
5202252022 __ret; \
5202352023})
5202452024#else
5202552025#define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
52026 float64_t __ret; \
5202652027 float64x2_t __s0 = __p0; \
5202752028 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52028 float64_t __ret; \
5202952029 __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((float64x2_t)__rev0, __p1); \
5203052030 __ret; \
5203152031})
......@@ -52033,16 +52033,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5203352033
5203452034#ifdef __LITTLE_ENDIAN__
5203552035#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
52036 float32x4_t __s0 = __p0; \
5203752036 float32_t __ret; \
52037 float32x4_t __s0 = __p0; \
5203852038 __ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__s0, __p1); \
5203952039 __ret; \
5204052040})
5204152041#else
5204252042#define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
52043 float32_t __ret; \
5204352044 float32x4_t __s0 = __p0; \
5204452045 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
52045 float32_t __ret; \
5204652046 __ret = (float32_t) __builtin_neon_vdups_laneq_f32((float32x4_t)__rev0, __p1); \
5204752047 __ret; \
5204852048})
......@@ -52050,16 +52050,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5205052050
5205152051#ifdef __LITTLE_ENDIAN__
5205252052#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
52053 int32x4_t __s0 = __p0; \
5205452053 int32_t __ret; \
52054 int32x4_t __s0 = __p0; \
5205552055 __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__s0, __p1); \
5205652056 __ret; \
5205752057})
5205852058#else
5205952059#define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
52060 int32_t __ret; \
5206052061 int32x4_t __s0 = __p0; \
5206152062 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
52062 int32_t __ret; \
5206352063 __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int32x4_t)__rev0, __p1); \
5206452064 __ret; \
5206552065})
......@@ -52067,16 +52067,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5206752067
5206852068#ifdef __LITTLE_ENDIAN__
5206952069#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
52070 int64x2_t __s0 = __p0; \
5207152070 int64_t __ret; \
52071 int64x2_t __s0 = __p0; \
5207252072 __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__s0, __p1); \
5207352073 __ret; \
5207452074})
5207552075#else
5207652076#define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
52077 int64_t __ret; \
5207752078 int64x2_t __s0 = __p0; \
5207852079 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
52079 int64_t __ret; \
5208052080 __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int64x2_t)__rev0, __p1); \
5208152081 __ret; \
5208252082})
......@@ -52084,16 +52084,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5208452084
5208552085#ifdef __LITTLE_ENDIAN__
5208652086#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
52087 int16x8_t __s0 = __p0; \
5208852087 int16_t __ret; \
52088 int16x8_t __s0 = __p0; \
5208952089 __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__s0, __p1); \
5209052090 __ret; \
5209152091})
5209252092#else
5209352093#define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
52094 int16_t __ret; \
5209452095 int16x8_t __s0 = __p0; \
5209552096 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
52096 int16_t __ret; \
5209752097 __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int16x8_t)__rev0, __p1); \
5209852098 __ret; \
5209952099})
......@@ -52101,16 +52101,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5210152101
5210252102#ifdef __LITTLE_ENDIAN__
5210352103#define vdup_laneq_p8(__p0_453, __p1_453) __extension__ ({ \
52104 poly8x16_t __s0_453 = __p0_453; \
5210552104 poly8x8_t __ret_453; \
52105 poly8x16_t __s0_453 = __p0_453; \
5210652106 __ret_453 = splat_laneq_p8(__s0_453, __p1_453); \
5210752107 __ret_453; \
5210852108})
5210952109#else
5211052110#define vdup_laneq_p8(__p0_454, __p1_454) __extension__ ({ \
52111 poly8x8_t __ret_454; \
5211152112 poly8x16_t __s0_454 = __p0_454; \
5211252113 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); \
52113 poly8x8_t __ret_454; \
5211452114 __ret_454 = __noswap_splat_laneq_p8(__rev0_454, __p1_454); \
5211552115 __ret_454 = __builtin_shufflevector(__ret_454, __ret_454, 7, 6, 5, 4, 3, 2, 1, 0); \
5211652116 __ret_454; \
......@@ -52119,16 +52119,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5211952119
5212052120#ifdef __LITTLE_ENDIAN__
5212152121#define vdup_laneq_p64(__p0_455, __p1_455) __extension__ ({ \
52122 poly64x2_t __s0_455 = __p0_455; \
5212352122 poly64x1_t __ret_455; \
52123 poly64x2_t __s0_455 = __p0_455; \
5212452124 __ret_455 = splat_laneq_p64(__s0_455, __p1_455); \
5212552125 __ret_455; \
5212652126})
5212752127#else
5212852128#define vdup_laneq_p64(__p0_456, __p1_456) __extension__ ({ \
52129 poly64x1_t __ret_456; \
5212952130 poly64x2_t __s0_456 = __p0_456; \
5213052131 poly64x2_t __rev0_456; __rev0_456 = __builtin_shufflevector(__s0_456, __s0_456, 1, 0); \
52131 poly64x1_t __ret_456; \
5213252132 __ret_456 = __noswap_splat_laneq_p64(__rev0_456, __p1_456); \
5213352133 __ret_456; \
5213452134})
......@@ -52136,16 +52136,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5213652136
5213752137#ifdef __LITTLE_ENDIAN__
5213852138#define vdup_laneq_p16(__p0_457, __p1_457) __extension__ ({ \
52139 poly16x8_t __s0_457 = __p0_457; \
5214052139 poly16x4_t __ret_457; \
52140 poly16x8_t __s0_457 = __p0_457; \
5214152141 __ret_457 = splat_laneq_p16(__s0_457, __p1_457); \
5214252142 __ret_457; \
5214352143})
5214452144#else
5214552145#define vdup_laneq_p16(__p0_458, __p1_458) __extension__ ({ \
52146 poly16x4_t __ret_458; \
5214652147 poly16x8_t __s0_458 = __p0_458; \
5214752148 poly16x8_t __rev0_458; __rev0_458 = __builtin_shufflevector(__s0_458, __s0_458, 7, 6, 5, 4, 3, 2, 1, 0); \
52148 poly16x4_t __ret_458; \
5214952149 __ret_458 = __noswap_splat_laneq_p16(__rev0_458, __p1_458); \
5215052150 __ret_458 = __builtin_shufflevector(__ret_458, __ret_458, 3, 2, 1, 0); \
5215152151 __ret_458; \
......@@ -52154,16 +52154,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5215452154
5215552155#ifdef __LITTLE_ENDIAN__
5215652156#define vdupq_laneq_p8(__p0_459, __p1_459) __extension__ ({ \
52157 poly8x16_t __s0_459 = __p0_459; \
5215852157 poly8x16_t __ret_459; \
52158 poly8x16_t __s0_459 = __p0_459; \
5215952159 __ret_459 = splatq_laneq_p8(__s0_459, __p1_459); \
5216052160 __ret_459; \
5216152161})
5216252162#else
5216352163#define vdupq_laneq_p8(__p0_460, __p1_460) __extension__ ({ \
52164 poly8x16_t __ret_460; \
5216452165 poly8x16_t __s0_460 = __p0_460; \
5216552166 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); \
52166 poly8x16_t __ret_460; \
5216752167 __ret_460 = __noswap_splatq_laneq_p8(__rev0_460, __p1_460); \
5216852168 __ret_460 = __builtin_shufflevector(__ret_460, __ret_460, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5216952169 __ret_460; \
......@@ -52172,16 +52172,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5217252172
5217352173#ifdef __LITTLE_ENDIAN__
5217452174#define vdupq_laneq_p64(__p0_461, __p1_461) __extension__ ({ \
52175 poly64x2_t __s0_461 = __p0_461; \
5217652175 poly64x2_t __ret_461; \
52176 poly64x2_t __s0_461 = __p0_461; \
5217752177 __ret_461 = splatq_laneq_p64(__s0_461, __p1_461); \
5217852178 __ret_461; \
5217952179})
5218052180#else
5218152181#define vdupq_laneq_p64(__p0_462, __p1_462) __extension__ ({ \
52182 poly64x2_t __ret_462; \
5218252183 poly64x2_t __s0_462 = __p0_462; \
5218352184 poly64x2_t __rev0_462; __rev0_462 = __builtin_shufflevector(__s0_462, __s0_462, 1, 0); \
52184 poly64x2_t __ret_462; \
5218552185 __ret_462 = __noswap_splatq_laneq_p64(__rev0_462, __p1_462); \
5218652186 __ret_462 = __builtin_shufflevector(__ret_462, __ret_462, 1, 0); \
5218752187 __ret_462; \
......@@ -52190,16 +52190,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5219052190
5219152191#ifdef __LITTLE_ENDIAN__
5219252192#define vdupq_laneq_p16(__p0_463, __p1_463) __extension__ ({ \
52193 poly16x8_t __s0_463 = __p0_463; \
5219452193 poly16x8_t __ret_463; \
52194 poly16x8_t __s0_463 = __p0_463; \
5219552195 __ret_463 = splatq_laneq_p16(__s0_463, __p1_463); \
5219652196 __ret_463; \
5219752197})
5219852198#else
5219952199#define vdupq_laneq_p16(__p0_464, __p1_464) __extension__ ({ \
52200 poly16x8_t __ret_464; \
5220052201 poly16x8_t __s0_464 = __p0_464; \
5220152202 poly16x8_t __rev0_464; __rev0_464 = __builtin_shufflevector(__s0_464, __s0_464, 7, 6, 5, 4, 3, 2, 1, 0); \
52202 poly16x8_t __ret_464; \
5220352203 __ret_464 = __noswap_splatq_laneq_p16(__rev0_464, __p1_464); \
5220452204 __ret_464 = __builtin_shufflevector(__ret_464, __ret_464, 7, 6, 5, 4, 3, 2, 1, 0); \
5220552205 __ret_464; \
......@@ -52208,16 +52208,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5220852208
5220952209#ifdef __LITTLE_ENDIAN__
5221052210#define vdupq_laneq_u8(__p0_465, __p1_465) __extension__ ({ \
52211 uint8x16_t __s0_465 = __p0_465; \
5221252211 uint8x16_t __ret_465; \
52212 uint8x16_t __s0_465 = __p0_465; \
5221352213 __ret_465 = splatq_laneq_u8(__s0_465, __p1_465); \
5221452214 __ret_465; \
5221552215})
5221652216#else
5221752217#define vdupq_laneq_u8(__p0_466, __p1_466) __extension__ ({ \
52218 uint8x16_t __ret_466; \
5221852219 uint8x16_t __s0_466 = __p0_466; \
5221952220 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); \
52220 uint8x16_t __ret_466; \
5222152221 __ret_466 = __noswap_splatq_laneq_u8(__rev0_466, __p1_466); \
5222252222 __ret_466 = __builtin_shufflevector(__ret_466, __ret_466, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5222352223 __ret_466; \
......@@ -52226,16 +52226,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5222652226
5222752227#ifdef __LITTLE_ENDIAN__
5222852228#define vdupq_laneq_u32(__p0_467, __p1_467) __extension__ ({ \
52229 uint32x4_t __s0_467 = __p0_467; \
5223052229 uint32x4_t __ret_467; \
52230 uint32x4_t __s0_467 = __p0_467; \
5223152231 __ret_467 = splatq_laneq_u32(__s0_467, __p1_467); \
5223252232 __ret_467; \
5223352233})
5223452234#else
5223552235#define vdupq_laneq_u32(__p0_468, __p1_468) __extension__ ({ \
52236 uint32x4_t __ret_468; \
5223652237 uint32x4_t __s0_468 = __p0_468; \
5223752238 uint32x4_t __rev0_468; __rev0_468 = __builtin_shufflevector(__s0_468, __s0_468, 3, 2, 1, 0); \
52238 uint32x4_t __ret_468; \
5223952239 __ret_468 = __noswap_splatq_laneq_u32(__rev0_468, __p1_468); \
5224052240 __ret_468 = __builtin_shufflevector(__ret_468, __ret_468, 3, 2, 1, 0); \
5224152241 __ret_468; \
......@@ -52244,16 +52244,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5224452244
5224552245#ifdef __LITTLE_ENDIAN__
5224652246#define vdupq_laneq_u64(__p0_469, __p1_469) __extension__ ({ \
52247 uint64x2_t __s0_469 = __p0_469; \
5224852247 uint64x2_t __ret_469; \
52248 uint64x2_t __s0_469 = __p0_469; \
5224952249 __ret_469 = splatq_laneq_u64(__s0_469, __p1_469); \
5225052250 __ret_469; \
5225152251})
5225252252#else
5225352253#define vdupq_laneq_u64(__p0_470, __p1_470) __extension__ ({ \
52254 uint64x2_t __ret_470; \
5225452255 uint64x2_t __s0_470 = __p0_470; \
5225552256 uint64x2_t __rev0_470; __rev0_470 = __builtin_shufflevector(__s0_470, __s0_470, 1, 0); \
52256 uint64x2_t __ret_470; \
5225752257 __ret_470 = __noswap_splatq_laneq_u64(__rev0_470, __p1_470); \
5225852258 __ret_470 = __builtin_shufflevector(__ret_470, __ret_470, 1, 0); \
5225952259 __ret_470; \
......@@ -52262,16 +52262,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5226252262
5226352263#ifdef __LITTLE_ENDIAN__
5226452264#define vdupq_laneq_u16(__p0_471, __p1_471) __extension__ ({ \
52265 uint16x8_t __s0_471 = __p0_471; \
5226652265 uint16x8_t __ret_471; \
52266 uint16x8_t __s0_471 = __p0_471; \
5226752267 __ret_471 = splatq_laneq_u16(__s0_471, __p1_471); \
5226852268 __ret_471; \
5226952269})
5227052270#else
5227152271#define vdupq_laneq_u16(__p0_472, __p1_472) __extension__ ({ \
52272 uint16x8_t __ret_472; \
5227252273 uint16x8_t __s0_472 = __p0_472; \
5227352274 uint16x8_t __rev0_472; __rev0_472 = __builtin_shufflevector(__s0_472, __s0_472, 7, 6, 5, 4, 3, 2, 1, 0); \
52274 uint16x8_t __ret_472; \
5227552275 __ret_472 = __noswap_splatq_laneq_u16(__rev0_472, __p1_472); \
5227652276 __ret_472 = __builtin_shufflevector(__ret_472, __ret_472, 7, 6, 5, 4, 3, 2, 1, 0); \
5227752277 __ret_472; \
......@@ -52280,16 +52280,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5228052280
5228152281#ifdef __LITTLE_ENDIAN__
5228252282#define vdupq_laneq_s8(__p0_473, __p1_473) __extension__ ({ \
52283 int8x16_t __s0_473 = __p0_473; \
5228452283 int8x16_t __ret_473; \
52284 int8x16_t __s0_473 = __p0_473; \
5228552285 __ret_473 = splatq_laneq_s8(__s0_473, __p1_473); \
5228652286 __ret_473; \
5228752287})
5228852288#else
5228952289#define vdupq_laneq_s8(__p0_474, __p1_474) __extension__ ({ \
52290 int8x16_t __ret_474; \
5229052291 int8x16_t __s0_474 = __p0_474; \
5229152292 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); \
52292 int8x16_t __ret_474; \
5229352293 __ret_474 = __noswap_splatq_laneq_s8(__rev0_474, __p1_474); \
5229452294 __ret_474 = __builtin_shufflevector(__ret_474, __ret_474, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
5229552295 __ret_474; \
......@@ -52298,16 +52298,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5229852298
5229952299#ifdef __LITTLE_ENDIAN__
5230052300#define vdupq_laneq_f64(__p0_475, __p1_475) __extension__ ({ \
52301 float64x2_t __s0_475 = __p0_475; \
5230252301 float64x2_t __ret_475; \
52302 float64x2_t __s0_475 = __p0_475; \
5230352303 __ret_475 = splatq_laneq_f64(__s0_475, __p1_475); \
5230452304 __ret_475; \
5230552305})
5230652306#else
5230752307#define vdupq_laneq_f64(__p0_476, __p1_476) __extension__ ({ \
52308 float64x2_t __ret_476; \
5230852309 float64x2_t __s0_476 = __p0_476; \
5230952310 float64x2_t __rev0_476; __rev0_476 = __builtin_shufflevector(__s0_476, __s0_476, 1, 0); \
52310 float64x2_t __ret_476; \
5231152311 __ret_476 = __noswap_splatq_laneq_f64(__rev0_476, __p1_476); \
5231252312 __ret_476 = __builtin_shufflevector(__ret_476, __ret_476, 1, 0); \
5231352313 __ret_476; \
......@@ -52316,16 +52316,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5231652316
5231752317#ifdef __LITTLE_ENDIAN__
5231852318#define vdupq_laneq_f32(__p0_477, __p1_477) __extension__ ({ \
52319 float32x4_t __s0_477 = __p0_477; \
5232052319 float32x4_t __ret_477; \
52320 float32x4_t __s0_477 = __p0_477; \
5232152321 __ret_477 = splatq_laneq_f32(__s0_477, __p1_477); \
5232252322 __ret_477; \
5232352323})
5232452324#else
5232552325#define vdupq_laneq_f32(__p0_478, __p1_478) __extension__ ({ \
52326 float32x4_t __ret_478; \
5232652327 float32x4_t __s0_478 = __p0_478; \
5232752328 float32x4_t __rev0_478; __rev0_478 = __builtin_shufflevector(__s0_478, __s0_478, 3, 2, 1, 0); \
52328 float32x4_t __ret_478; \
5232952329 __ret_478 = __noswap_splatq_laneq_f32(__rev0_478, __p1_478); \
5233052330 __ret_478 = __builtin_shufflevector(__ret_478, __ret_478, 3, 2, 1, 0); \
5233152331 __ret_478; \
......@@ -52334,16 +52334,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5233452334
5233552335#ifdef __LITTLE_ENDIAN__
5233652336#define vdupq_laneq_f16(__p0_479, __p1_479) __extension__ ({ \
52337 float16x8_t __s0_479 = __p0_479; \
5233852337 float16x8_t __ret_479; \
52338 float16x8_t __s0_479 = __p0_479; \
5233952339 __ret_479 = splatq_laneq_f16(__s0_479, __p1_479); \
5234052340 __ret_479; \
5234152341})
5234252342#else
5234352343#define vdupq_laneq_f16(__p0_480, __p1_480) __extension__ ({ \
52344 float16x8_t __ret_480; \
5234452345 float16x8_t __s0_480 = __p0_480; \
5234552346 float16x8_t __rev0_480; __rev0_480 = __builtin_shufflevector(__s0_480, __s0_480, 7, 6, 5, 4, 3, 2, 1, 0); \
52346 float16x8_t __ret_480; \
5234752347 __ret_480 = __noswap_splatq_laneq_f16(__rev0_480, __p1_480); \
5234852348 __ret_480 = __builtin_shufflevector(__ret_480, __ret_480, 7, 6, 5, 4, 3, 2, 1, 0); \
5234952349 __ret_480; \
......@@ -52352,16 +52352,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5235252352
5235352353#ifdef __LITTLE_ENDIAN__
5235452354#define vdupq_laneq_s32(__p0_481, __p1_481) __extension__ ({ \
52355 int32x4_t __s0_481 = __p0_481; \
5235652355 int32x4_t __ret_481; \
52356 int32x4_t __s0_481 = __p0_481; \
5235752357 __ret_481 = splatq_laneq_s32(__s0_481, __p1_481); \
5235852358 __ret_481; \
5235952359})
5236052360#else
5236152361#define vdupq_laneq_s32(__p0_482, __p1_482) __extension__ ({ \
52362 int32x4_t __ret_482; \
5236252363 int32x4_t __s0_482 = __p0_482; \
5236352364 int32x4_t __rev0_482; __rev0_482 = __builtin_shufflevector(__s0_482, __s0_482, 3, 2, 1, 0); \
52364 int32x4_t __ret_482; \
5236552365 __ret_482 = __noswap_splatq_laneq_s32(__rev0_482, __p1_482); \
5236652366 __ret_482 = __builtin_shufflevector(__ret_482, __ret_482, 3, 2, 1, 0); \
5236752367 __ret_482; \
......@@ -52370,16 +52370,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5237052370
5237152371#ifdef __LITTLE_ENDIAN__
5237252372#define vdupq_laneq_s64(__p0_483, __p1_483) __extension__ ({ \
52373 int64x2_t __s0_483 = __p0_483; \
5237452373 int64x2_t __ret_483; \
52374 int64x2_t __s0_483 = __p0_483; \
5237552375 __ret_483 = splatq_laneq_s64(__s0_483, __p1_483); \
5237652376 __ret_483; \
5237752377})
5237852378#else
5237952379#define vdupq_laneq_s64(__p0_484, __p1_484) __extension__ ({ \
52380 int64x2_t __ret_484; \
5238052381 int64x2_t __s0_484 = __p0_484; \
5238152382 int64x2_t __rev0_484; __rev0_484 = __builtin_shufflevector(__s0_484, __s0_484, 1, 0); \
52382 int64x2_t __ret_484; \
5238352383 __ret_484 = __noswap_splatq_laneq_s64(__rev0_484, __p1_484); \
5238452384 __ret_484 = __builtin_shufflevector(__ret_484, __ret_484, 1, 0); \
5238552385 __ret_484; \
......@@ -52388,16 +52388,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5238852388
5238952389#ifdef __LITTLE_ENDIAN__
5239052390#define vdupq_laneq_s16(__p0_485, __p1_485) __extension__ ({ \
52391 int16x8_t __s0_485 = __p0_485; \
5239252391 int16x8_t __ret_485; \
52392 int16x8_t __s0_485 = __p0_485; \
5239352393 __ret_485 = splatq_laneq_s16(__s0_485, __p1_485); \
5239452394 __ret_485; \
5239552395})
5239652396#else
5239752397#define vdupq_laneq_s16(__p0_486, __p1_486) __extension__ ({ \
52398 int16x8_t __ret_486; \
5239852399 int16x8_t __s0_486 = __p0_486; \
5239952400 int16x8_t __rev0_486; __rev0_486 = __builtin_shufflevector(__s0_486, __s0_486, 7, 6, 5, 4, 3, 2, 1, 0); \
52400 int16x8_t __ret_486; \
5240152401 __ret_486 = __noswap_splatq_laneq_s16(__rev0_486, __p1_486); \
5240252402 __ret_486 = __builtin_shufflevector(__ret_486, __ret_486, 7, 6, 5, 4, 3, 2, 1, 0); \
5240352403 __ret_486; \
......@@ -52406,16 +52406,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5240652406
5240752407#ifdef __LITTLE_ENDIAN__
5240852408#define vdup_laneq_u8(__p0_487, __p1_487) __extension__ ({ \
52409 uint8x16_t __s0_487 = __p0_487; \
5241052409 uint8x8_t __ret_487; \
52410 uint8x16_t __s0_487 = __p0_487; \
5241152411 __ret_487 = splat_laneq_u8(__s0_487, __p1_487); \
5241252412 __ret_487; \
5241352413})
5241452414#else
5241552415#define vdup_laneq_u8(__p0_488, __p1_488) __extension__ ({ \
52416 uint8x8_t __ret_488; \
5241652417 uint8x16_t __s0_488 = __p0_488; \
5241752418 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); \
52418 uint8x8_t __ret_488; \
5241952419 __ret_488 = __noswap_splat_laneq_u8(__rev0_488, __p1_488); \
5242052420 __ret_488 = __builtin_shufflevector(__ret_488, __ret_488, 7, 6, 5, 4, 3, 2, 1, 0); \
5242152421 __ret_488; \
......@@ -52424,16 +52424,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5242452424
5242552425#ifdef __LITTLE_ENDIAN__
5242652426#define vdup_laneq_u32(__p0_489, __p1_489) __extension__ ({ \
52427 uint32x4_t __s0_489 = __p0_489; \
5242852427 uint32x2_t __ret_489; \
52428 uint32x4_t __s0_489 = __p0_489; \
5242952429 __ret_489 = splat_laneq_u32(__s0_489, __p1_489); \
5243052430 __ret_489; \
5243152431})
5243252432#else
5243352433#define vdup_laneq_u32(__p0_490, __p1_490) __extension__ ({ \
52434 uint32x2_t __ret_490; \
5243452435 uint32x4_t __s0_490 = __p0_490; \
5243552436 uint32x4_t __rev0_490; __rev0_490 = __builtin_shufflevector(__s0_490, __s0_490, 3, 2, 1, 0); \
52436 uint32x2_t __ret_490; \
5243752437 __ret_490 = __noswap_splat_laneq_u32(__rev0_490, __p1_490); \
5243852438 __ret_490 = __builtin_shufflevector(__ret_490, __ret_490, 1, 0); \
5243952439 __ret_490; \
......@@ -52442,16 +52442,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5244252442
5244352443#ifdef __LITTLE_ENDIAN__
5244452444#define vdup_laneq_u64(__p0_491, __p1_491) __extension__ ({ \
52445 uint64x2_t __s0_491 = __p0_491; \
5244652445 uint64x1_t __ret_491; \
52446 uint64x2_t __s0_491 = __p0_491; \
5244752447 __ret_491 = splat_laneq_u64(__s0_491, __p1_491); \
5244852448 __ret_491; \
5244952449})
5245052450#else
5245152451#define vdup_laneq_u64(__p0_492, __p1_492) __extension__ ({ \
52452 uint64x1_t __ret_492; \
5245252453 uint64x2_t __s0_492 = __p0_492; \
5245352454 uint64x2_t __rev0_492; __rev0_492 = __builtin_shufflevector(__s0_492, __s0_492, 1, 0); \
52454 uint64x1_t __ret_492; \
5245552455 __ret_492 = __noswap_splat_laneq_u64(__rev0_492, __p1_492); \
5245652456 __ret_492; \
5245752457})
......@@ -52459,16 +52459,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5245952459
5246052460#ifdef __LITTLE_ENDIAN__
5246152461#define vdup_laneq_u16(__p0_493, __p1_493) __extension__ ({ \
52462 uint16x8_t __s0_493 = __p0_493; \
5246352462 uint16x4_t __ret_493; \
52463 uint16x8_t __s0_493 = __p0_493; \
5246452464 __ret_493 = splat_laneq_u16(__s0_493, __p1_493); \
5246552465 __ret_493; \
5246652466})
5246752467#else
5246852468#define vdup_laneq_u16(__p0_494, __p1_494) __extension__ ({ \
52469 uint16x4_t __ret_494; \
5246952470 uint16x8_t __s0_494 = __p0_494; \
5247052471 uint16x8_t __rev0_494; __rev0_494 = __builtin_shufflevector(__s0_494, __s0_494, 7, 6, 5, 4, 3, 2, 1, 0); \
52471 uint16x4_t __ret_494; \
5247252472 __ret_494 = __noswap_splat_laneq_u16(__rev0_494, __p1_494); \
5247352473 __ret_494 = __builtin_shufflevector(__ret_494, __ret_494, 3, 2, 1, 0); \
5247452474 __ret_494; \
......@@ -52477,16 +52477,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5247752477
5247852478#ifdef __LITTLE_ENDIAN__
5247952479#define vdup_laneq_s8(__p0_495, __p1_495) __extension__ ({ \
52480 int8x16_t __s0_495 = __p0_495; \
5248152480 int8x8_t __ret_495; \
52481 int8x16_t __s0_495 = __p0_495; \
5248252482 __ret_495 = splat_laneq_s8(__s0_495, __p1_495); \
5248352483 __ret_495; \
5248452484})
5248552485#else
5248652486#define vdup_laneq_s8(__p0_496, __p1_496) __extension__ ({ \
52487 int8x8_t __ret_496; \
5248752488 int8x16_t __s0_496 = __p0_496; \
5248852489 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); \
52489 int8x8_t __ret_496; \
5249052490 __ret_496 = __noswap_splat_laneq_s8(__rev0_496, __p1_496); \
5249152491 __ret_496 = __builtin_shufflevector(__ret_496, __ret_496, 7, 6, 5, 4, 3, 2, 1, 0); \
5249252492 __ret_496; \
......@@ -52495,16 +52495,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5249552495
5249652496#ifdef __LITTLE_ENDIAN__
5249752497#define vdup_laneq_f64(__p0_497, __p1_497) __extension__ ({ \
52498 float64x2_t __s0_497 = __p0_497; \
5249952498 float64x1_t __ret_497; \
52499 float64x2_t __s0_497 = __p0_497; \
5250052500 __ret_497 = splat_laneq_f64(__s0_497, __p1_497); \
5250152501 __ret_497; \
5250252502})
5250352503#else
5250452504#define vdup_laneq_f64(__p0_498, __p1_498) __extension__ ({ \
52505 float64x1_t __ret_498; \
5250552506 float64x2_t __s0_498 = __p0_498; \
5250652507 float64x2_t __rev0_498; __rev0_498 = __builtin_shufflevector(__s0_498, __s0_498, 1, 0); \
52507 float64x1_t __ret_498; \
5250852508 __ret_498 = __noswap_splat_laneq_f64(__rev0_498, __p1_498); \
5250952509 __ret_498; \
5251052510})
......@@ -52512,16 +52512,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5251252512
5251352513#ifdef __LITTLE_ENDIAN__
5251452514#define vdup_laneq_f32(__p0_499, __p1_499) __extension__ ({ \
52515 float32x4_t __s0_499 = __p0_499; \
5251652515 float32x2_t __ret_499; \
52516 float32x4_t __s0_499 = __p0_499; \
5251752517 __ret_499 = splat_laneq_f32(__s0_499, __p1_499); \
5251852518 __ret_499; \
5251952519})
5252052520#else
5252152521#define vdup_laneq_f32(__p0_500, __p1_500) __extension__ ({ \
52522 float32x2_t __ret_500; \
5252252523 float32x4_t __s0_500 = __p0_500; \
5252352524 float32x4_t __rev0_500; __rev0_500 = __builtin_shufflevector(__s0_500, __s0_500, 3, 2, 1, 0); \
52524 float32x2_t __ret_500; \
5252552525 __ret_500 = __noswap_splat_laneq_f32(__rev0_500, __p1_500); \
5252652526 __ret_500 = __builtin_shufflevector(__ret_500, __ret_500, 1, 0); \
5252752527 __ret_500; \
......@@ -52530,16 +52530,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5253052530
5253152531#ifdef __LITTLE_ENDIAN__
5253252532#define vdup_laneq_f16(__p0_501, __p1_501) __extension__ ({ \
52533 float16x8_t __s0_501 = __p0_501; \
5253452533 float16x4_t __ret_501; \
52534 float16x8_t __s0_501 = __p0_501; \
5253552535 __ret_501 = splat_laneq_f16(__s0_501, __p1_501); \
5253652536 __ret_501; \
5253752537})
5253852538#else
5253952539#define vdup_laneq_f16(__p0_502, __p1_502) __extension__ ({ \
52540 float16x4_t __ret_502; \
5254052541 float16x8_t __s0_502 = __p0_502; \
5254152542 float16x8_t __rev0_502; __rev0_502 = __builtin_shufflevector(__s0_502, __s0_502, 7, 6, 5, 4, 3, 2, 1, 0); \
52542 float16x4_t __ret_502; \
5254352543 __ret_502 = __noswap_splat_laneq_f16(__rev0_502, __p1_502); \
5254452544 __ret_502 = __builtin_shufflevector(__ret_502, __ret_502, 3, 2, 1, 0); \
5254552545 __ret_502; \
......@@ -52548,16 +52548,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5254852548
5254952549#ifdef __LITTLE_ENDIAN__
5255052550#define vdup_laneq_s32(__p0_503, __p1_503) __extension__ ({ \
52551 int32x4_t __s0_503 = __p0_503; \
5255252551 int32x2_t __ret_503; \
52552 int32x4_t __s0_503 = __p0_503; \
5255352553 __ret_503 = splat_laneq_s32(__s0_503, __p1_503); \
5255452554 __ret_503; \
5255552555})
5255652556#else
5255752557#define vdup_laneq_s32(__p0_504, __p1_504) __extension__ ({ \
52558 int32x2_t __ret_504; \
5255852559 int32x4_t __s0_504 = __p0_504; \
5255952560 int32x4_t __rev0_504; __rev0_504 = __builtin_shufflevector(__s0_504, __s0_504, 3, 2, 1, 0); \
52560 int32x2_t __ret_504; \
5256152561 __ret_504 = __noswap_splat_laneq_s32(__rev0_504, __p1_504); \
5256252562 __ret_504 = __builtin_shufflevector(__ret_504, __ret_504, 1, 0); \
5256352563 __ret_504; \
......@@ -52566,16 +52566,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5256652566
5256752567#ifdef __LITTLE_ENDIAN__
5256852568#define vdup_laneq_s64(__p0_505, __p1_505) __extension__ ({ \
52569 int64x2_t __s0_505 = __p0_505; \
5257052569 int64x1_t __ret_505; \
52570 int64x2_t __s0_505 = __p0_505; \
5257152571 __ret_505 = splat_laneq_s64(__s0_505, __p1_505); \
5257252572 __ret_505; \
5257352573})
5257452574#else
5257552575#define vdup_laneq_s64(__p0_506, __p1_506) __extension__ ({ \
52576 int64x1_t __ret_506; \
5257652577 int64x2_t __s0_506 = __p0_506; \
5257752578 int64x2_t __rev0_506; __rev0_506 = __builtin_shufflevector(__s0_506, __s0_506, 1, 0); \
52578 int64x1_t __ret_506; \
5257952579 __ret_506 = __noswap_splat_laneq_s64(__rev0_506, __p1_506); \
5258052580 __ret_506; \
5258152581})
......@@ -52583,16 +52583,16 @@ __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
5258352583
5258452584#ifdef __LITTLE_ENDIAN__
5258552585#define vdup_laneq_s16(__p0_507, __p1_507) __extension__ ({ \
52586 int16x8_t __s0_507 = __p0_507; \
5258752586 int16x4_t __ret_507; \
52587 int16x8_t __s0_507 = __p0_507; \
5258852588 __ret_507 = splat_laneq_s16(__s0_507, __p1_507); \
5258952589 __ret_507; \
5259052590})
5259152591#else
5259252592#define vdup_laneq_s16(__p0_508, __p1_508) __extension__ ({ \
52593 int16x4_t __ret_508; \
5259352594 int16x8_t __s0_508 = __p0_508; \
5259452595 int16x8_t __rev0_508; __rev0_508 = __builtin_shufflevector(__s0_508, __s0_508, 7, 6, 5, 4, 3, 2, 1, 0); \
52595 int16x4_t __ret_508; \
5259652596 __ret_508 = __noswap_splat_laneq_s16(__rev0_508, __p1_508); \
5259752597 __ret_508 = __builtin_shufflevector(__ret_508, __ret_508, 3, 2, 1, 0); \
5259852598 __ret_508; \
......@@ -52640,27 +52640,27 @@ __ai float64x1_t vdup_n_f64(float64_t __p0) {
5264052640 return __ret;
5264152641}
5264252642#define vext_p64(__p0, __p1, __p2) __extension__ ({ \
52643 poly64x1_t __ret; \
5264352644 poly64x1_t __s0 = __p0; \
5264452645 poly64x1_t __s1 = __p1; \
52645 poly64x1_t __ret; \
5264652646 __ret = (poly64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
5264752647 __ret; \
5264852648})
5264952649#ifdef __LITTLE_ENDIAN__
5265052650#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
52651 poly64x2_t __ret; \
5265152652 poly64x2_t __s0 = __p0; \
5265252653 poly64x2_t __s1 = __p1; \
52653 poly64x2_t __ret; \
5265452654 __ret = (poly64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
5265552655 __ret; \
5265652656})
5265752657#else
5265852658#define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
52659 poly64x2_t __ret; \
5265952660 poly64x2_t __s0 = __p0; \
5266052661 poly64x2_t __s1 = __p1; \
5266152662 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5266252663 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52663 poly64x2_t __ret; \
5266452664 __ret = (poly64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
5266552665 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5266652666 __ret; \
......@@ -52669,19 +52669,19 @@ __ai float64x1_t vdup_n_f64(float64_t __p0) {
5266952669
5267052670#ifdef __LITTLE_ENDIAN__
5267152671#define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
52672 float64x2_t __ret; \
5267252673 float64x2_t __s0 = __p0; \
5267352674 float64x2_t __s1 = __p1; \
52674 float64x2_t __ret; \
5267552675 __ret = (float64x2_t) __builtin_neon_vextq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 42); \
5267652676 __ret; \
5267752677})
5267852678#else
5267952679#define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
52680 float64x2_t __ret; \
5268052681 float64x2_t __s0 = __p0; \
5268152682 float64x2_t __s1 = __p1; \
5268252683 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5268352684 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52684 float64x2_t __ret; \
5268552685 __ret = (float64x2_t) __builtin_neon_vextq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 42); \
5268652686 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5268752687 __ret; \
......@@ -52689,9 +52689,9 @@ __ai float64x1_t vdup_n_f64(float64_t __p0) {
5268952689#endif
5269052690
5269152691#define vext_f64(__p0, __p1, __p2) __extension__ ({ \
52692 float64x1_t __ret; \
5269252693 float64x1_t __s0 = __p0; \
5269352694 float64x1_t __s1 = __p1; \
52694 float64x1_t __ret; \
5269552695 __ret = (float64x1_t) __builtin_neon_vext_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
5269652696 __ret; \
5269752697})
......@@ -52703,10 +52703,10 @@ __ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2)
5270352703}
5270452704#else
5270552705__ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52706 float64x2_t __ret;
5270652707 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5270752708 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5270852709 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
52709 float64x2_t __ret;
5271052710 __ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 42);
5271152711 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5271252712 return __ret;
......@@ -52724,37 +52724,37 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5272452724 return __ret;
5272552725}
5272652726#define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52727 float64_t __ret; \
5272752728 float64_t __s0 = __p0; \
5272852729 float64_t __s1 = __p1; \
5272952730 float64x1_t __s2 = __p2; \
52730 float64_t __ret; \
5273152731 __ret = (float64_t) __builtin_neon_vfmad_lane_f64(__s0, __s1, (float64x1_t)__s2, __p3); \
5273252732 __ret; \
5273352733})
5273452734#ifdef __LITTLE_ENDIAN__
5273552735#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52736 float32_t __ret; \
5273652737 float32_t __s0 = __p0; \
5273752738 float32_t __s1 = __p1; \
5273852739 float32x2_t __s2 = __p2; \
52739 float32_t __ret; \
5274052740 __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
5274152741 __ret; \
5274252742})
5274352743#else
5274452744#define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52745 float32_t __ret; \
5274552746 float32_t __s0 = __p0; \
5274652747 float32_t __s1 = __p1; \
5274752748 float32x2_t __s2 = __p2; \
5274852749 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52749 float32_t __ret; \
5275052750 __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__rev2, __p3); \
5275152751 __ret; \
5275252752})
5275352753#define __noswap_vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52754 float32_t __ret; \
5275452755 float32_t __s0 = __p0; \
5275552756 float32_t __s1 = __p1; \
5275652757 float32x2_t __s2 = __p2; \
52757 float32_t __ret; \
5275852758 __ret = (float32_t) __builtin_neon_vfmas_lane_f32(__s0, __s1, (float32x2_t)__s2, __p3); \
5275952759 __ret; \
5276052760})
......@@ -52762,30 +52762,30 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5276252762
5276352763#ifdef __LITTLE_ENDIAN__
5276452764#define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52765 float64x2_t __ret; \
5276552766 float64x2_t __s0 = __p0; \
5276652767 float64x2_t __s1 = __p1; \
5276752768 float64x1_t __s2 = __p2; \
52768 float64x2_t __ret; \
5276952769 __ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 42); \
5277052770 __ret; \
5277152771})
5277252772#else
5277352773#define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52774 float64x2_t __ret; \
5277452775 float64x2_t __s0 = __p0; \
5277552776 float64x2_t __s1 = __p1; \
5277652777 float64x1_t __s2 = __p2; \
5277752778 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5277852779 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
52779 float64x2_t __ret; \
5278052780 __ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__s2, __p3, 42); \
5278152781 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5278252782 __ret; \
5278352783})
5278452784#define __noswap_vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52785 float64x2_t __ret; \
5278552786 float64x2_t __s0 = __p0; \
5278652787 float64x2_t __s1 = __p1; \
5278752788 float64x1_t __s2 = __p2; \
52788 float64x2_t __ret; \
5278952789 __ret = (float64x2_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 42); \
5279052790 __ret; \
5279152791})
......@@ -52793,71 +52793,71 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5279352793
5279452794#ifdef __LITTLE_ENDIAN__
5279552795#define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52796 float32x4_t __ret; \
5279652797 float32x4_t __s0 = __p0; \
5279752798 float32x4_t __s1 = __p1; \
5279852799 float32x2_t __s2 = __p2; \
52799 float32x4_t __ret; \
5280052800 __ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 41); \
5280152801 __ret; \
5280252802})
5280352803#else
5280452804#define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52805 float32x4_t __ret; \
5280552806 float32x4_t __s0 = __p0; \
5280652807 float32x4_t __s1 = __p1; \
5280752808 float32x2_t __s2 = __p2; \
5280852809 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5280952810 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
5281052811 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52811 float32x4_t __ret; \
5281252812 __ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, __p3, 41); \
5281352813 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
5281452814 __ret; \
5281552815})
5281652816#define __noswap_vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52817 float32x4_t __ret; \
5281752818 float32x4_t __s0 = __p0; \
5281852819 float32x4_t __s1 = __p1; \
5281952820 float32x2_t __s2 = __p2; \
52820 float32x4_t __ret; \
5282152821 __ret = (float32x4_t) __builtin_neon_vfmaq_lane_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x8_t)__s2, __p3, 41); \
5282252822 __ret; \
5282352823})
5282452824#endif
5282552825
5282652826#define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52827 float64x1_t __ret; \
5282752828 float64x1_t __s0 = __p0; \
5282852829 float64x1_t __s1 = __p1; \
5282952830 float64x1_t __s2 = __p2; \
52830 float64x1_t __ret; \
5283152831 __ret = (float64x1_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 10); \
5283252832 __ret; \
5283352833})
5283452834#ifdef __LITTLE_ENDIAN__
5283552835#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52836 float32x2_t __ret; \
5283652837 float32x2_t __s0 = __p0; \
5283752838 float32x2_t __s1 = __p1; \
5283852839 float32x2_t __s2 = __p2; \
52839 float32x2_t __ret; \
5284052840 __ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 9); \
5284152841 __ret; \
5284252842})
5284352843#else
5284452844#define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52845 float32x2_t __ret; \
5284552846 float32x2_t __s0 = __p0; \
5284652847 float32x2_t __s1 = __p1; \
5284752848 float32x2_t __s2 = __p2; \
5284852849 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5284952850 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
5285052851 float32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52851 float32x2_t __ret; \
5285252852 __ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x8_t)__rev2, __p3, 9); \
5285352853 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5285452854 __ret; \
5285552855})
5285652856#define __noswap_vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52857 float32x2_t __ret; \
5285752858 float32x2_t __s0 = __p0; \
5285852859 float32x2_t __s1 = __p1; \
5285952860 float32x2_t __s2 = __p2; \
52860 float32x2_t __ret; \
5286152861 __ret = (float32x2_t) __builtin_neon_vfma_lane_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x8_t)__s2, __p3, 9); \
5286252862 __ret; \
5286352863})
......@@ -52865,28 +52865,28 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5286552865
5286652866#ifdef __LITTLE_ENDIAN__
5286752867#define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52868 float64_t __ret; \
5286852869 float64_t __s0 = __p0; \
5286952870 float64_t __s1 = __p1; \
5287052871 float64x2_t __s2 = __p2; \
52871 float64_t __ret; \
5287252872 __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
5287352873 __ret; \
5287452874})
5287552875#else
5287652876#define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52877 float64_t __ret; \
5287752878 float64_t __s0 = __p0; \
5287852879 float64_t __s1 = __p1; \
5287952880 float64x2_t __s2 = __p2; \
5288052881 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52881 float64_t __ret; \
5288252882 __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__rev2, __p3); \
5288352883 __ret; \
5288452884})
5288552885#define __noswap_vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52886 float64_t __ret; \
5288652887 float64_t __s0 = __p0; \
5288752888 float64_t __s1 = __p1; \
5288852889 float64x2_t __s2 = __p2; \
52889 float64_t __ret; \
5289052890 __ret = (float64_t) __builtin_neon_vfmad_laneq_f64(__s0, __s1, (float64x2_t)__s2, __p3); \
5289152891 __ret; \
5289252892})
......@@ -52894,28 +52894,28 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5289452894
5289552895#ifdef __LITTLE_ENDIAN__
5289652896#define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52897 float32_t __ret; \
5289752898 float32_t __s0 = __p0; \
5289852899 float32_t __s1 = __p1; \
5289952900 float32x4_t __s2 = __p2; \
52900 float32_t __ret; \
5290152901 __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
5290252902 __ret; \
5290352903})
5290452904#else
5290552905#define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52906 float32_t __ret; \
5290652907 float32_t __s0 = __p0; \
5290752908 float32_t __s1 = __p1; \
5290852909 float32x4_t __s2 = __p2; \
5290952910 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
52910 float32_t __ret; \
5291152911 __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__rev2, __p3); \
5291252912 __ret; \
5291352913})
5291452914#define __noswap_vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52915 float32_t __ret; \
5291552916 float32_t __s0 = __p0; \
5291652917 float32_t __s1 = __p1; \
5291752918 float32x4_t __s2 = __p2; \
52918 float32_t __ret; \
5291952919 __ret = (float32_t) __builtin_neon_vfmas_laneq_f32(__s0, __s1, (float32x4_t)__s2, __p3); \
5292052920 __ret; \
5292152921})
......@@ -52923,31 +52923,31 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5292352923
5292452924#ifdef __LITTLE_ENDIAN__
5292552925#define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52926 float64x2_t __ret; \
5292652927 float64x2_t __s0 = __p0; \
5292752928 float64x2_t __s1 = __p1; \
5292852929 float64x2_t __s2 = __p2; \
52929 float64x2_t __ret; \
5293052930 __ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 42); \
5293152931 __ret; \
5293252932})
5293352933#else
5293452934#define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52935 float64x2_t __ret; \
5293552936 float64x2_t __s0 = __p0; \
5293652937 float64x2_t __s1 = __p1; \
5293752938 float64x2_t __s2 = __p2; \
5293852939 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5293952940 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
5294052941 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
52941 float64x2_t __ret; \
5294252942 __ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 42); \
5294352943 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5294452944 __ret; \
5294552945})
5294652946#define __noswap_vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52947 float64x2_t __ret; \
5294752948 float64x2_t __s0 = __p0; \
5294852949 float64x2_t __s1 = __p1; \
5294952950 float64x2_t __s2 = __p2; \
52950 float64x2_t __ret; \
5295152951 __ret = (float64x2_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 42); \
5295252952 __ret; \
5295352953})
......@@ -52955,31 +52955,31 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5295552955
5295652956#ifdef __LITTLE_ENDIAN__
5295752957#define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52958 float32x4_t __ret; \
5295852959 float32x4_t __s0 = __p0; \
5295952960 float32x4_t __s1 = __p1; \
5296052961 float32x4_t __s2 = __p2; \
52961 float32x4_t __ret; \
5296252962 __ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 41); \
5296352963 __ret; \
5296452964})
5296552965#else
5296652966#define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52967 float32x4_t __ret; \
5296752968 float32x4_t __s0 = __p0; \
5296852969 float32x4_t __s1 = __p1; \
5296952970 float32x4_t __s2 = __p2; \
5297052971 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5297152972 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
5297252973 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
52973 float32x4_t __ret; \
5297452974 __ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, __p3, 41); \
5297552975 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
5297652976 __ret; \
5297752977})
5297852978#define __noswap_vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52979 float32x4_t __ret; \
5297952980 float32x4_t __s0 = __p0; \
5298052981 float32x4_t __s1 = __p1; \
5298152982 float32x4_t __s2 = __p2; \
52982 float32x4_t __ret; \
5298352983 __ret = (float32x4_t) __builtin_neon_vfmaq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, (int8x16_t)__s2, __p3, 41); \
5298452984 __ret; \
5298552985})
......@@ -52987,28 +52987,28 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5298752987
5298852988#ifdef __LITTLE_ENDIAN__
5298952989#define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52990 float64x1_t __ret; \
5299052991 float64x1_t __s0 = __p0; \
5299152992 float64x1_t __s1 = __p1; \
5299252993 float64x2_t __s2 = __p2; \
52993 float64x1_t __ret; \
5299452994 __ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 10); \
5299552995 __ret; \
5299652996})
5299752997#else
5299852998#define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
52999 float64x1_t __ret; \
5299953000 float64x1_t __s0 = __p0; \
5300053001 float64x1_t __s1 = __p1; \
5300153002 float64x2_t __s2 = __p2; \
5300253003 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
53003 float64x1_t __ret; \
5300453004 __ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__rev2, __p3, 10); \
5300553005 __ret; \
5300653006})
5300753007#define __noswap_vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
53008 float64x1_t __ret; \
5300853009 float64x1_t __s0 = __p0; \
5300953010 float64x1_t __s1 = __p1; \
5301053011 float64x2_t __s2 = __p2; \
53011 float64x1_t __ret; \
5301253012 __ret = (float64x1_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 10); \
5301353013 __ret; \
5301453014})
......@@ -53016,31 +53016,31 @@ __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5301653016
5301753017#ifdef __LITTLE_ENDIAN__
5301853018#define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53019 float32x2_t __ret; \
5301953020 float32x2_t __s0 = __p0; \
5302053021 float32x2_t __s1 = __p1; \
5302153022 float32x4_t __s2 = __p2; \
53022 float32x2_t __ret; \
5302353023 __ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 9); \
5302453024 __ret; \
5302553025})
5302653026#else
5302753027#define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53028 float32x2_t __ret; \
5302853029 float32x2_t __s0 = __p0; \
5302953030 float32x2_t __s1 = __p1; \
5303053031 float32x4_t __s2 = __p2; \
5303153032 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5303253033 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
5303353034 float32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
53034 float32x2_t __ret; \
5303553035 __ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__rev0, (int8x8_t)__rev1, (int8x16_t)__rev2, __p3, 9); \
5303653036 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5303753037 __ret; \
5303853038})
5303953039#define __noswap_vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53040 float32x2_t __ret; \
5304053041 float32x2_t __s0 = __p0; \
5304153042 float32x2_t __s1 = __p1; \
5304253043 float32x4_t __s2 = __p2; \
53043 float32x2_t __ret; \
5304453044 __ret = (float32x2_t) __builtin_neon_vfma_laneq_v((int8x8_t)__s0, (int8x8_t)__s1, (int8x16_t)__s2, __p3, 9); \
5304553045 __ret; \
5304653046})
......@@ -53054,9 +53054,9 @@ __ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2)
5305453054}
5305553055#else
5305653056__ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53057 float64x2_t __ret;
5305753058 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5305853059 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53059 float64x2_t __ret;
5306053060 __ret = __noswap_vfmaq_f64(__rev0, __rev1, (float64x2_t) {__p2, __p2});
5306153061 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5306253062 return __ret;
......@@ -53076,10 +53076,10 @@ __ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2)
5307653076}
5307753077#else
5307853078__ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
53079 float64x2_t __ret;
5307953080 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5308053081 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5308153082 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
53082 float64x2_t __ret;
5308353083 __ret = __noswap_vfmaq_f64(__rev0, -__rev1, __rev2);
5308453084 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5308553085 return __ret;
......@@ -53092,29 +53092,29 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5309253092 return __ret;
5309353093}
5309453094#define vfmsd_lane_f64(__p0_509, __p1_509, __p2_509, __p3_509) __extension__ ({ \
53095 float64_t __ret_509; \
5309553096 float64_t __s0_509 = __p0_509; \
5309653097 float64_t __s1_509 = __p1_509; \
5309753098 float64x1_t __s2_509 = __p2_509; \
53098 float64_t __ret_509; \
5309953099 __ret_509 = vfmad_lane_f64(__s0_509, -__s1_509, __s2_509, __p3_509); \
5310053100 __ret_509; \
5310153101})
5310253102#ifdef __LITTLE_ENDIAN__
5310353103#define vfmss_lane_f32(__p0_510, __p1_510, __p2_510, __p3_510) __extension__ ({ \
53104 float32_t __ret_510; \
5310453105 float32_t __s0_510 = __p0_510; \
5310553106 float32_t __s1_510 = __p1_510; \
5310653107 float32x2_t __s2_510 = __p2_510; \
53107 float32_t __ret_510; \
5310853108 __ret_510 = vfmas_lane_f32(__s0_510, -__s1_510, __s2_510, __p3_510); \
5310953109 __ret_510; \
5311053110})
5311153111#else
5311253112#define vfmss_lane_f32(__p0_511, __p1_511, __p2_511, __p3_511) __extension__ ({ \
53113 float32_t __ret_511; \
5311353114 float32_t __s0_511 = __p0_511; \
5311453115 float32_t __s1_511 = __p1_511; \
5311553116 float32x2_t __s2_511 = __p2_511; \
5311653117 float32x2_t __rev2_511; __rev2_511 = __builtin_shufflevector(__s2_511, __s2_511, 1, 0); \
53117 float32_t __ret_511; \
5311853118 __ret_511 = __noswap_vfmas_lane_f32(__s0_511, -__s1_511, __rev2_511, __p3_511); \
5311953119 __ret_511; \
5312053120})
......@@ -53122,21 +53122,21 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5312253122
5312353123#ifdef __LITTLE_ENDIAN__
5312453124#define vfmsq_lane_f64(__p0_512, __p1_512, __p2_512, __p3_512) __extension__ ({ \
53125 float64x2_t __ret_512; \
5312553126 float64x2_t __s0_512 = __p0_512; \
5312653127 float64x2_t __s1_512 = __p1_512; \
5312753128 float64x1_t __s2_512 = __p2_512; \
53128 float64x2_t __ret_512; \
5312953129 __ret_512 = vfmaq_lane_f64(__s0_512, -__s1_512, __s2_512, __p3_512); \
5313053130 __ret_512; \
5313153131})
5313253132#else
5313353133#define vfmsq_lane_f64(__p0_513, __p1_513, __p2_513, __p3_513) __extension__ ({ \
53134 float64x2_t __ret_513; \
5313453135 float64x2_t __s0_513 = __p0_513; \
5313553136 float64x2_t __s1_513 = __p1_513; \
5313653137 float64x1_t __s2_513 = __p2_513; \
5313753138 float64x2_t __rev0_513; __rev0_513 = __builtin_shufflevector(__s0_513, __s0_513, 1, 0); \
5313853139 float64x2_t __rev1_513; __rev1_513 = __builtin_shufflevector(__s1_513, __s1_513, 1, 0); \
53139 float64x2_t __ret_513; \
5314053140 __ret_513 = __noswap_vfmaq_lane_f64(__rev0_513, -__rev1_513, __s2_513, __p3_513); \
5314153141 __ret_513 = __builtin_shufflevector(__ret_513, __ret_513, 1, 0); \
5314253142 __ret_513; \
......@@ -53145,22 +53145,22 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5314553145
5314653146#ifdef __LITTLE_ENDIAN__
5314753147#define vfmsq_lane_f32(__p0_514, __p1_514, __p2_514, __p3_514) __extension__ ({ \
53148 float32x4_t __ret_514; \
5314853149 float32x4_t __s0_514 = __p0_514; \
5314953150 float32x4_t __s1_514 = __p1_514; \
5315053151 float32x2_t __s2_514 = __p2_514; \
53151 float32x4_t __ret_514; \
5315253152 __ret_514 = vfmaq_lane_f32(__s0_514, -__s1_514, __s2_514, __p3_514); \
5315353153 __ret_514; \
5315453154})
5315553155#else
5315653156#define vfmsq_lane_f32(__p0_515, __p1_515, __p2_515, __p3_515) __extension__ ({ \
53157 float32x4_t __ret_515; \
5315753158 float32x4_t __s0_515 = __p0_515; \
5315853159 float32x4_t __s1_515 = __p1_515; \
5315953160 float32x2_t __s2_515 = __p2_515; \
5316053161 float32x4_t __rev0_515; __rev0_515 = __builtin_shufflevector(__s0_515, __s0_515, 3, 2, 1, 0); \
5316153162 float32x4_t __rev1_515; __rev1_515 = __builtin_shufflevector(__s1_515, __s1_515, 3, 2, 1, 0); \
5316253163 float32x2_t __rev2_515; __rev2_515 = __builtin_shufflevector(__s2_515, __s2_515, 1, 0); \
53163 float32x4_t __ret_515; \
5316453164 __ret_515 = __noswap_vfmaq_lane_f32(__rev0_515, -__rev1_515, __rev2_515, __p3_515); \
5316553165 __ret_515 = __builtin_shufflevector(__ret_515, __ret_515, 3, 2, 1, 0); \
5316653166 __ret_515; \
......@@ -53168,31 +53168,31 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5316853168#endif
5316953169
5317053170#define vfms_lane_f64(__p0_516, __p1_516, __p2_516, __p3_516) __extension__ ({ \
53171 float64x1_t __ret_516; \
5317153172 float64x1_t __s0_516 = __p0_516; \
5317253173 float64x1_t __s1_516 = __p1_516; \
5317353174 float64x1_t __s2_516 = __p2_516; \
53174 float64x1_t __ret_516; \
5317553175 __ret_516 = vfma_lane_f64(__s0_516, -__s1_516, __s2_516, __p3_516); \
5317653176 __ret_516; \
5317753177})
5317853178#ifdef __LITTLE_ENDIAN__
5317953179#define vfms_lane_f32(__p0_517, __p1_517, __p2_517, __p3_517) __extension__ ({ \
53180 float32x2_t __ret_517; \
5318053181 float32x2_t __s0_517 = __p0_517; \
5318153182 float32x2_t __s1_517 = __p1_517; \
5318253183 float32x2_t __s2_517 = __p2_517; \
53183 float32x2_t __ret_517; \
5318453184 __ret_517 = vfma_lane_f32(__s0_517, -__s1_517, __s2_517, __p3_517); \
5318553185 __ret_517; \
5318653186})
5318753187#else
5318853188#define vfms_lane_f32(__p0_518, __p1_518, __p2_518, __p3_518) __extension__ ({ \
53189 float32x2_t __ret_518; \
5318953190 float32x2_t __s0_518 = __p0_518; \
5319053191 float32x2_t __s1_518 = __p1_518; \
5319153192 float32x2_t __s2_518 = __p2_518; \
5319253193 float32x2_t __rev0_518; __rev0_518 = __builtin_shufflevector(__s0_518, __s0_518, 1, 0); \
5319353194 float32x2_t __rev1_518; __rev1_518 = __builtin_shufflevector(__s1_518, __s1_518, 1, 0); \
5319453195 float32x2_t __rev2_518; __rev2_518 = __builtin_shufflevector(__s2_518, __s2_518, 1, 0); \
53195 float32x2_t __ret_518; \
5319653196 __ret_518 = __noswap_vfma_lane_f32(__rev0_518, -__rev1_518, __rev2_518, __p3_518); \
5319753197 __ret_518 = __builtin_shufflevector(__ret_518, __ret_518, 1, 0); \
5319853198 __ret_518; \
......@@ -53201,20 +53201,20 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5320153201
5320253202#ifdef __LITTLE_ENDIAN__
5320353203#define vfmsd_laneq_f64(__p0_519, __p1_519, __p2_519, __p3_519) __extension__ ({ \
53204 float64_t __ret_519; \
5320453205 float64_t __s0_519 = __p0_519; \
5320553206 float64_t __s1_519 = __p1_519; \
5320653207 float64x2_t __s2_519 = __p2_519; \
53207 float64_t __ret_519; \
5320853208 __ret_519 = vfmad_laneq_f64(__s0_519, -__s1_519, __s2_519, __p3_519); \
5320953209 __ret_519; \
5321053210})
5321153211#else
5321253212#define vfmsd_laneq_f64(__p0_520, __p1_520, __p2_520, __p3_520) __extension__ ({ \
53213 float64_t __ret_520; \
5321353214 float64_t __s0_520 = __p0_520; \
5321453215 float64_t __s1_520 = __p1_520; \
5321553216 float64x2_t __s2_520 = __p2_520; \
5321653217 float64x2_t __rev2_520; __rev2_520 = __builtin_shufflevector(__s2_520, __s2_520, 1, 0); \
53217 float64_t __ret_520; \
5321853218 __ret_520 = __noswap_vfmad_laneq_f64(__s0_520, -__s1_520, __rev2_520, __p3_520); \
5321953219 __ret_520; \
5322053220})
......@@ -53222,20 +53222,20 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5322253222
5322353223#ifdef __LITTLE_ENDIAN__
5322453224#define vfmss_laneq_f32(__p0_521, __p1_521, __p2_521, __p3_521) __extension__ ({ \
53225 float32_t __ret_521; \
5322553226 float32_t __s0_521 = __p0_521; \
5322653227 float32_t __s1_521 = __p1_521; \
5322753228 float32x4_t __s2_521 = __p2_521; \
53228 float32_t __ret_521; \
5322953229 __ret_521 = vfmas_laneq_f32(__s0_521, -__s1_521, __s2_521, __p3_521); \
5323053230 __ret_521; \
5323153231})
5323253232#else
5323353233#define vfmss_laneq_f32(__p0_522, __p1_522, __p2_522, __p3_522) __extension__ ({ \
53234 float32_t __ret_522; \
5323453235 float32_t __s0_522 = __p0_522; \
5323553236 float32_t __s1_522 = __p1_522; \
5323653237 float32x4_t __s2_522 = __p2_522; \
5323753238 float32x4_t __rev2_522; __rev2_522 = __builtin_shufflevector(__s2_522, __s2_522, 3, 2, 1, 0); \
53238 float32_t __ret_522; \
5323953239 __ret_522 = __noswap_vfmas_laneq_f32(__s0_522, -__s1_522, __rev2_522, __p3_522); \
5324053240 __ret_522; \
5324153241})
......@@ -53243,22 +53243,22 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5324353243
5324453244#ifdef __LITTLE_ENDIAN__
5324553245#define vfmsq_laneq_f64(__p0_523, __p1_523, __p2_523, __p3_523) __extension__ ({ \
53246 float64x2_t __ret_523; \
5324653247 float64x2_t __s0_523 = __p0_523; \
5324753248 float64x2_t __s1_523 = __p1_523; \
5324853249 float64x2_t __s2_523 = __p2_523; \
53249 float64x2_t __ret_523; \
5325053250 __ret_523 = vfmaq_laneq_f64(__s0_523, -__s1_523, __s2_523, __p3_523); \
5325153251 __ret_523; \
5325253252})
5325353253#else
5325453254#define vfmsq_laneq_f64(__p0_524, __p1_524, __p2_524, __p3_524) __extension__ ({ \
53255 float64x2_t __ret_524; \
5325553256 float64x2_t __s0_524 = __p0_524; \
5325653257 float64x2_t __s1_524 = __p1_524; \
5325753258 float64x2_t __s2_524 = __p2_524; \
5325853259 float64x2_t __rev0_524; __rev0_524 = __builtin_shufflevector(__s0_524, __s0_524, 1, 0); \
5325953260 float64x2_t __rev1_524; __rev1_524 = __builtin_shufflevector(__s1_524, __s1_524, 1, 0); \
5326053261 float64x2_t __rev2_524; __rev2_524 = __builtin_shufflevector(__s2_524, __s2_524, 1, 0); \
53261 float64x2_t __ret_524; \
5326253262 __ret_524 = __noswap_vfmaq_laneq_f64(__rev0_524, -__rev1_524, __rev2_524, __p3_524); \
5326353263 __ret_524 = __builtin_shufflevector(__ret_524, __ret_524, 1, 0); \
5326453264 __ret_524; \
......@@ -53267,22 +53267,22 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5326753267
5326853268#ifdef __LITTLE_ENDIAN__
5326953269#define vfmsq_laneq_f32(__p0_525, __p1_525, __p2_525, __p3_525) __extension__ ({ \
53270 float32x4_t __ret_525; \
5327053271 float32x4_t __s0_525 = __p0_525; \
5327153272 float32x4_t __s1_525 = __p1_525; \
5327253273 float32x4_t __s2_525 = __p2_525; \
53273 float32x4_t __ret_525; \
5327453274 __ret_525 = vfmaq_laneq_f32(__s0_525, -__s1_525, __s2_525, __p3_525); \
5327553275 __ret_525; \
5327653276})
5327753277#else
5327853278#define vfmsq_laneq_f32(__p0_526, __p1_526, __p2_526, __p3_526) __extension__ ({ \
53279 float32x4_t __ret_526; \
5327953280 float32x4_t __s0_526 = __p0_526; \
5328053281 float32x4_t __s1_526 = __p1_526; \
5328153282 float32x4_t __s2_526 = __p2_526; \
5328253283 float32x4_t __rev0_526; __rev0_526 = __builtin_shufflevector(__s0_526, __s0_526, 3, 2, 1, 0); \
5328353284 float32x4_t __rev1_526; __rev1_526 = __builtin_shufflevector(__s1_526, __s1_526, 3, 2, 1, 0); \
5328453285 float32x4_t __rev2_526; __rev2_526 = __builtin_shufflevector(__s2_526, __s2_526, 3, 2, 1, 0); \
53285 float32x4_t __ret_526; \
5328653286 __ret_526 = __noswap_vfmaq_laneq_f32(__rev0_526, -__rev1_526, __rev2_526, __p3_526); \
5328753287 __ret_526 = __builtin_shufflevector(__ret_526, __ret_526, 3, 2, 1, 0); \
5328853288 __ret_526; \
......@@ -53291,20 +53291,20 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5329153291
5329253292#ifdef __LITTLE_ENDIAN__
5329353293#define vfms_laneq_f64(__p0_527, __p1_527, __p2_527, __p3_527) __extension__ ({ \
53294 float64x1_t __ret_527; \
5329453295 float64x1_t __s0_527 = __p0_527; \
5329553296 float64x1_t __s1_527 = __p1_527; \
5329653297 float64x2_t __s2_527 = __p2_527; \
53297 float64x1_t __ret_527; \
5329853298 __ret_527 = vfma_laneq_f64(__s0_527, -__s1_527, __s2_527, __p3_527); \
5329953299 __ret_527; \
5330053300})
5330153301#else
5330253302#define vfms_laneq_f64(__p0_528, __p1_528, __p2_528, __p3_528) __extension__ ({ \
53303 float64x1_t __ret_528; \
5330353304 float64x1_t __s0_528 = __p0_528; \
5330453305 float64x1_t __s1_528 = __p1_528; \
5330553306 float64x2_t __s2_528 = __p2_528; \
5330653307 float64x2_t __rev2_528; __rev2_528 = __builtin_shufflevector(__s2_528, __s2_528, 1, 0); \
53307 float64x1_t __ret_528; \
5330853308 __ret_528 = __noswap_vfma_laneq_f64(__s0_528, -__s1_528, __rev2_528, __p3_528); \
5330953309 __ret_528; \
5331053310})
......@@ -53312,22 +53312,22 @@ __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5331253312
5331353313#ifdef __LITTLE_ENDIAN__
5331453314#define vfms_laneq_f32(__p0_529, __p1_529, __p2_529, __p3_529) __extension__ ({ \
53315 float32x2_t __ret_529; \
5331553316 float32x2_t __s0_529 = __p0_529; \
5331653317 float32x2_t __s1_529 = __p1_529; \
5331753318 float32x4_t __s2_529 = __p2_529; \
53318 float32x2_t __ret_529; \
5331953319 __ret_529 = vfma_laneq_f32(__s0_529, -__s1_529, __s2_529, __p3_529); \
5332053320 __ret_529; \
5332153321})
5332253322#else
5332353323#define vfms_laneq_f32(__p0_530, __p1_530, __p2_530, __p3_530) __extension__ ({ \
53324 float32x2_t __ret_530; \
5332453325 float32x2_t __s0_530 = __p0_530; \
5332553326 float32x2_t __s1_530 = __p1_530; \
5332653327 float32x4_t __s2_530 = __p2_530; \
5332753328 float32x2_t __rev0_530; __rev0_530 = __builtin_shufflevector(__s0_530, __s0_530, 1, 0); \
5332853329 float32x2_t __rev1_530; __rev1_530 = __builtin_shufflevector(__s1_530, __s1_530, 1, 0); \
5332953330 float32x4_t __rev2_530; __rev2_530 = __builtin_shufflevector(__s2_530, __s2_530, 3, 2, 1, 0); \
53330 float32x2_t __ret_530; \
5333153331 __ret_530 = __noswap_vfma_laneq_f32(__rev0_530, -__rev1_530, __rev2_530, __p3_530); \
5333253332 __ret_530 = __builtin_shufflevector(__ret_530, __ret_530, 1, 0); \
5333353333 __ret_530; \
......@@ -53342,9 +53342,9 @@ __ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2)
5334253342}
5334353343#else
5334453344__ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53345 float64x2_t __ret;
5334553346 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5334653347 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53347 float64x2_t __ret;
5334853348 __ret = __noswap_vfmaq_f64(__rev0, -__rev1, (float64x2_t) {__p2, __p2});
5334953349 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5335053350 return __ret;
......@@ -53359,9 +53359,9 @@ __ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2)
5335953359}
5336053360#else
5336153361__ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
53362 float32x4_t __ret;
5336253363 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5336353364 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53364 float32x4_t __ret;
5336553365 __ret = __noswap_vfmaq_f32(__rev0, -__rev1, (float32x4_t) {__p2, __p2, __p2, __p2});
5336653366 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5336753367 return __ret;
......@@ -53381,9 +53381,9 @@ __ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2)
5338153381}
5338253382#else
5338353383__ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
53384 float32x2_t __ret;
5338453385 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5338553386 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53386 float32x2_t __ret;
5338753387 __ret = __noswap_vfma_f32(__rev0, -__rev1, (float32x2_t) {__p2, __p2});
5338853388 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5338953389 return __ret;
......@@ -53398,8 +53398,8 @@ __ai poly64x1_t vget_high_p64(poly64x2_t __p0) {
5339853398}
5339953399#else
5340053400__ai poly64x1_t vget_high_p64(poly64x2_t __p0) {
53401 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5340253401 poly64x1_t __ret;
53402 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5340353403 __ret = __builtin_shufflevector(__rev0, __rev0, 1);
5340453404 return __ret;
5340553405}
......@@ -53418,37 +53418,37 @@ __ai float64x1_t vget_high_f64(float64x2_t __p0) {
5341853418}
5341953419#else
5342053420__ai float64x1_t vget_high_f64(float64x2_t __p0) {
53421 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5342253421 float64x1_t __ret;
53422 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5342353423 __ret = __builtin_shufflevector(__rev0, __rev0, 1);
5342453424 return __ret;
5342553425}
5342653426#endif
5342753427
5342853428#define vget_lane_p64(__p0, __p1) __extension__ ({ \
53429 poly64x1_t __s0 = __p0; \
5343053429 poly64_t __ret; \
53430 poly64x1_t __s0 = __p0; \
5343153431 __ret = (poly64_t) __builtin_neon_vget_lane_i64((poly64x1_t)__s0, __p1); \
5343253432 __ret; \
5343353433})
5343453434#ifdef __LITTLE_ENDIAN__
5343553435#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
53436 poly64x2_t __s0 = __p0; \
5343753436 poly64_t __ret; \
53437 poly64x2_t __s0 = __p0; \
5343853438 __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
5343953439 __ret; \
5344053440})
5344153441#else
5344253442#define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
53443 poly64_t __ret; \
5344353444 poly64x2_t __s0 = __p0; \
5344453445 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
53445 poly64_t __ret; \
5344653446 __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__rev0, __p1); \
5344753447 __ret; \
5344853448})
5344953449#define __noswap_vgetq_lane_p64(__p0, __p1) __extension__ ({ \
53450 poly64x2_t __s0 = __p0; \
5345153450 poly64_t __ret; \
53451 poly64x2_t __s0 = __p0; \
5345253452 __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((poly64x2_t)__s0, __p1); \
5345353453 __ret; \
5345453454})
......@@ -53456,30 +53456,30 @@ __ai float64x1_t vget_high_f64(float64x2_t __p0) {
5345653456
5345753457#ifdef __LITTLE_ENDIAN__
5345853458#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
53459 float64x2_t __s0 = __p0; \
5346053459 float64_t __ret; \
53460 float64x2_t __s0 = __p0; \
5346153461 __ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
5346253462 __ret; \
5346353463})
5346453464#else
5346553465#define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
53466 float64_t __ret; \
5346653467 float64x2_t __s0 = __p0; \
5346753468 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
53468 float64_t __ret; \
5346953469 __ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__rev0, __p1); \
5347053470 __ret; \
5347153471})
5347253472#define __noswap_vgetq_lane_f64(__p0, __p1) __extension__ ({ \
53473 float64x2_t __s0 = __p0; \
5347453473 float64_t __ret; \
53474 float64x2_t __s0 = __p0; \
5347553475 __ret = (float64_t) __builtin_neon_vgetq_lane_f64((float64x2_t)__s0, __p1); \
5347653476 __ret; \
5347753477})
5347853478#endif
5347953479
5348053480#define vget_lane_f64(__p0, __p1) __extension__ ({ \
53481 float64x1_t __s0 = __p0; \
5348253481 float64_t __ret; \
53482 float64x1_t __s0 = __p0; \
5348353483 __ret = (float64_t) __builtin_neon_vget_lane_f64((float64x1_t)__s0, __p1); \
5348453484 __ret; \
5348553485})
......@@ -53491,8 +53491,8 @@ __ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
5349153491}
5349253492#else
5349353493__ai poly64x1_t vget_low_p64(poly64x2_t __p0) {
53494 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5349553494 poly64x1_t __ret;
53495 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5349653496 __ret = __builtin_shufflevector(__rev0, __rev0, 0);
5349753497 return __ret;
5349853498}
......@@ -53506,8 +53506,8 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5350653506}
5350753507#else
5350853508__ai float64x1_t vget_low_f64(float64x2_t __p0) {
53509 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5351053509 float64x1_t __ret;
53510 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5351153511 __ret = __builtin_shufflevector(__rev0, __rev0, 0);
5351253512 return __ret;
5351353513}
......@@ -53594,23 +53594,23 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5359453594 __ret; \
5359553595})
5359653596#define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53597 poly64x1_t __s1 = __p1; \
5359853597 poly64x1_t __ret; \
53598 poly64x1_t __s1 = __p1; \
5359953599 __ret = (poly64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 6); \
5360053600 __ret; \
5360153601})
5360253602#ifdef __LITTLE_ENDIAN__
5360353603#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53604 poly64x2_t __s1 = __p1; \
5360553604 poly64x2_t __ret; \
53605 poly64x2_t __s1 = __p1; \
5360653606 __ret = (poly64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 38); \
5360753607 __ret; \
5360853608})
5360953609#else
5361053610#define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53611 poly64x2_t __ret; \
5361153612 poly64x2_t __s1 = __p1; \
5361253613 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
53613 poly64x2_t __ret; \
5361453614 __ret = (poly64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 38); \
5361553615 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5361653616 __ret; \
......@@ -53619,16 +53619,16 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5361953619
5362053620#ifdef __LITTLE_ENDIAN__
5362153621#define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53622 float64x2_t __s1 = __p1; \
5362353622 float64x2_t __ret; \
53623 float64x2_t __s1 = __p1; \
5362453624 __ret = (float64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__s1, __p2, 42); \
5362553625 __ret; \
5362653626})
5362753627#else
5362853628#define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53629 float64x2_t __ret; \
5362953630 float64x2_t __s1 = __p1; \
5363053631 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
53631 float64x2_t __ret; \
5363253632 __ret = (float64x2_t) __builtin_neon_vld1q_lane_v(__p0, (int8x16_t)__rev1, __p2, 42); \
5363353633 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5363453634 __ret; \
......@@ -53636,8 +53636,8 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5363653636#endif
5363753637
5363853638#define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53639 float64x1_t __s1 = __p1; \
5364053639 float64x1_t __ret; \
53640 float64x1_t __s1 = __p1; \
5364153641 __ret = (float64x1_t) __builtin_neon_vld1_lane_v(__p0, (int8x8_t)__s1, __p2, 10); \
5364253642 __ret; \
5364353643})
......@@ -53902,25 +53902,25 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5390253902 __ret; \
5390353903})
5390453904#define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53905 poly64x1x2_t __s1 = __p1; \
5390653905 poly64x1x2_t __ret; \
53906 poly64x1x2_t __s1 = __p1; \
5390753907 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 6); \
5390853908 __ret; \
5390953909})
5391053910#ifdef __LITTLE_ENDIAN__
5391153911#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
53912 poly8x16x2_t __s1 = __p1; \
5391353912 poly8x16x2_t __ret; \
53913 poly8x16x2_t __s1 = __p1; \
5391453914 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 36); \
5391553915 __ret; \
5391653916})
5391753917#else
5391853918#define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
53919 poly8x16x2_t __ret; \
5391953920 poly8x16x2_t __s1 = __p1; \
5392053921 poly8x16x2_t __rev1; \
5392153922 __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); \
5392253923 __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); \
53923 poly8x16x2_t __ret; \
5392453924 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 36); \
5392553925 \
5392653926 __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); \
......@@ -53931,18 +53931,18 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5393153931
5393253932#ifdef __LITTLE_ENDIAN__
5393353933#define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53934 poly64x2x2_t __s1 = __p1; \
5393553934 poly64x2x2_t __ret; \
53935 poly64x2x2_t __s1 = __p1; \
5393653936 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 38); \
5393753937 __ret; \
5393853938})
5393953939#else
5394053940#define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
53941 poly64x2x2_t __ret; \
5394153942 poly64x2x2_t __s1 = __p1; \
5394253943 poly64x2x2_t __rev1; \
5394353944 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5394453945 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
53945 poly64x2x2_t __ret; \
5394653946 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 38); \
5394753947 \
5394853948 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -53953,18 +53953,18 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5395353953
5395453954#ifdef __LITTLE_ENDIAN__
5395553955#define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
53956 uint8x16x2_t __s1 = __p1; \
5395753956 uint8x16x2_t __ret; \
53957 uint8x16x2_t __s1 = __p1; \
5395853958 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 48); \
5395953959 __ret; \
5396053960})
5396153961#else
5396253962#define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
53963 uint8x16x2_t __ret; \
5396353964 uint8x16x2_t __s1 = __p1; \
5396453965 uint8x16x2_t __rev1; \
5396553966 __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); \
5396653967 __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); \
53967 uint8x16x2_t __ret; \
5396853968 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 48); \
5396953969 \
5397053970 __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); \
......@@ -53975,18 +53975,18 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5397553975
5397653976#ifdef __LITTLE_ENDIAN__
5397753977#define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
53978 uint64x2x2_t __s1 = __p1; \
5397953978 uint64x2x2_t __ret; \
53979 uint64x2x2_t __s1 = __p1; \
5398053980 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 51); \
5398153981 __ret; \
5398253982})
5398353983#else
5398453984#define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
53985 uint64x2x2_t __ret; \
5398553986 uint64x2x2_t __s1 = __p1; \
5398653987 uint64x2x2_t __rev1; \
5398753988 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5398853989 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
53989 uint64x2x2_t __ret; \
5399053990 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 51); \
5399153991 \
5399253992 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -53997,18 +53997,18 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5399753997
5399853998#ifdef __LITTLE_ENDIAN__
5399953999#define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54000 int8x16x2_t __s1 = __p1; \
5400154000 int8x16x2_t __ret; \
54001 int8x16x2_t __s1 = __p1; \
5400254002 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 32); \
5400354003 __ret; \
5400454004})
5400554005#else
5400654006#define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54007 int8x16x2_t __ret; \
5400754008 int8x16x2_t __s1 = __p1; \
5400854009 int8x16x2_t __rev1; \
5400954010 __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); \
5401054011 __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); \
54011 int8x16x2_t __ret; \
5401254012 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 32); \
5401354013 \
5401454014 __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); \
......@@ -54019,18 +54019,18 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5401954019
5402054020#ifdef __LITTLE_ENDIAN__
5402154021#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54022 float64x2x2_t __s1 = __p1; \
5402354022 float64x2x2_t __ret; \
54023 float64x2x2_t __s1 = __p1; \
5402454024 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 42); \
5402554025 __ret; \
5402654026})
5402754027#else
5402854028#define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54029 float64x2x2_t __ret; \
5402954030 float64x2x2_t __s1 = __p1; \
5403054031 float64x2x2_t __rev1; \
5403154032 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5403254033 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54033 float64x2x2_t __ret; \
5403454034 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 42); \
5403554035 \
5403654036 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54041,18 +54041,18 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5404154041
5404254042#ifdef __LITTLE_ENDIAN__
5404354043#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54044 int64x2x2_t __s1 = __p1; \
5404554044 int64x2x2_t __ret; \
54045 int64x2x2_t __s1 = __p1; \
5404654046 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], __p2, 35); \
5404754047 __ret; \
5404854048})
5404954049#else
5405054050#define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54051 int64x2x2_t __ret; \
5405154052 int64x2x2_t __s1 = __p1; \
5405254053 int64x2x2_t __rev1; \
5405354054 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5405454055 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
54055 int64x2x2_t __ret; \
5405654056 __builtin_neon_vld2q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], __p2, 35); \
5405754057 \
5405854058 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54062,20 +54062,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5406254062#endif
5406354063
5406454064#define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54065 uint64x1x2_t __s1 = __p1; \
5406654065 uint64x1x2_t __ret; \
54066 uint64x1x2_t __s1 = __p1; \
5406754067 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 19); \
5406854068 __ret; \
5406954069})
5407054070#define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54071 float64x1x2_t __s1 = __p1; \
5407254071 float64x1x2_t __ret; \
54072 float64x1x2_t __s1 = __p1; \
5407354073 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 10); \
5407454074 __ret; \
5407554075})
5407654076#define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54077 int64x1x2_t __s1 = __p1; \
5407854077 int64x1x2_t __ret; \
54078 int64x1x2_t __s1 = __p1; \
5407954079 __builtin_neon_vld2_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], __p2, 3); \
5408054080 __ret; \
5408154081})
......@@ -54208,26 +54208,26 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5420854208 __ret; \
5420954209})
5421054210#define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54211 poly64x1x3_t __s1 = __p1; \
5421254211 poly64x1x3_t __ret; \
54212 poly64x1x3_t __s1 = __p1; \
5421354213 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 6); \
5421454214 __ret; \
5421554215})
5421654216#ifdef __LITTLE_ENDIAN__
5421754217#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54218 poly8x16x3_t __s1 = __p1; \
5421954218 poly8x16x3_t __ret; \
54219 poly8x16x3_t __s1 = __p1; \
5422054220 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 36); \
5422154221 __ret; \
5422254222})
5422354223#else
5422454224#define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54225 poly8x16x3_t __ret; \
5422554226 poly8x16x3_t __s1 = __p1; \
5422654227 poly8x16x3_t __rev1; \
5422754228 __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); \
5422854229 __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); \
5422954230 __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); \
54230 poly8x16x3_t __ret; \
5423154231 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 36); \
5423254232 \
5423354233 __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); \
......@@ -54239,19 +54239,19 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5423954239
5424054240#ifdef __LITTLE_ENDIAN__
5424154241#define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54242 poly64x2x3_t __s1 = __p1; \
5424354242 poly64x2x3_t __ret; \
54243 poly64x2x3_t __s1 = __p1; \
5424454244 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 38); \
5424554245 __ret; \
5424654246})
5424754247#else
5424854248#define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54249 poly64x2x3_t __ret; \
5424954250 poly64x2x3_t __s1 = __p1; \
5425054251 poly64x2x3_t __rev1; \
5425154252 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5425254253 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5425354254 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54254 poly64x2x3_t __ret; \
5425554255 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 38); \
5425654256 \
5425754257 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54263,19 +54263,19 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5426354263
5426454264#ifdef __LITTLE_ENDIAN__
5426554265#define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54266 uint8x16x3_t __s1 = __p1; \
5426754266 uint8x16x3_t __ret; \
54267 uint8x16x3_t __s1 = __p1; \
5426854268 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 48); \
5426954269 __ret; \
5427054270})
5427154271#else
5427254272#define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54273 uint8x16x3_t __ret; \
5427354274 uint8x16x3_t __s1 = __p1; \
5427454275 uint8x16x3_t __rev1; \
5427554276 __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); \
5427654277 __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); \
5427754278 __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); \
54278 uint8x16x3_t __ret; \
5427954279 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 48); \
5428054280 \
5428154281 __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); \
......@@ -54287,19 +54287,19 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5428754287
5428854288#ifdef __LITTLE_ENDIAN__
5428954289#define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54290 uint64x2x3_t __s1 = __p1; \
5429154290 uint64x2x3_t __ret; \
54291 uint64x2x3_t __s1 = __p1; \
5429254292 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 51); \
5429354293 __ret; \
5429454294})
5429554295#else
5429654296#define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54297 uint64x2x3_t __ret; \
5429754298 uint64x2x3_t __s1 = __p1; \
5429854299 uint64x2x3_t __rev1; \
5429954300 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5430054301 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5430154302 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54302 uint64x2x3_t __ret; \
5430354303 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 51); \
5430454304 \
5430554305 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54311,19 +54311,19 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5431154311
5431254312#ifdef __LITTLE_ENDIAN__
5431354313#define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54314 int8x16x3_t __s1 = __p1; \
5431554314 int8x16x3_t __ret; \
54315 int8x16x3_t __s1 = __p1; \
5431654316 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 32); \
5431754317 __ret; \
5431854318})
5431954319#else
5432054320#define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54321 int8x16x3_t __ret; \
5432154322 int8x16x3_t __s1 = __p1; \
5432254323 int8x16x3_t __rev1; \
5432354324 __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); \
5432454325 __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); \
5432554326 __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); \
54326 int8x16x3_t __ret; \
5432754327 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 32); \
5432854328 \
5432954329 __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); \
......@@ -54335,19 +54335,19 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5433554335
5433654336#ifdef __LITTLE_ENDIAN__
5433754337#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54338 float64x2x3_t __s1 = __p1; \
5433954338 float64x2x3_t __ret; \
54339 float64x2x3_t __s1 = __p1; \
5434054340 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 42); \
5434154341 __ret; \
5434254342})
5434354343#else
5434454344#define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54345 float64x2x3_t __ret; \
5434554346 float64x2x3_t __s1 = __p1; \
5434654347 float64x2x3_t __rev1; \
5434754348 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5434854349 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5434954350 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54350 float64x2x3_t __ret; \
5435154351 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 42); \
5435254352 \
5435354353 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54359,19 +54359,19 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5435954359
5436054360#ifdef __LITTLE_ENDIAN__
5436154361#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54362 int64x2x3_t __s1 = __p1; \
5436354362 int64x2x3_t __ret; \
54363 int64x2x3_t __s1 = __p1; \
5436454364 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__s1.val[0], (int8x16_t)__s1.val[1], (int8x16_t)__s1.val[2], __p2, 35); \
5436554365 __ret; \
5436654366})
5436754367#else
5436854368#define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54369 int64x2x3_t __ret; \
5436954370 int64x2x3_t __s1 = __p1; \
5437054371 int64x2x3_t __rev1; \
5437154372 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5437254373 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5437354374 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
54374 int64x2x3_t __ret; \
5437554375 __builtin_neon_vld3q_lane_v(&__ret, __p0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev1.val[2], __p2, 35); \
5437654376 \
5437754377 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54382,20 +54382,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5438254382#endif
5438354383
5438454384#define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54385 uint64x1x3_t __s1 = __p1; \
5438654385 uint64x1x3_t __ret; \
54386 uint64x1x3_t __s1 = __p1; \
5438754387 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 19); \
5438854388 __ret; \
5438954389})
5439054390#define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54391 float64x1x3_t __s1 = __p1; \
5439254391 float64x1x3_t __ret; \
54392 float64x1x3_t __s1 = __p1; \
5439354393 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 10); \
5439454394 __ret; \
5439554395})
5439654396#define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54397 int64x1x3_t __s1 = __p1; \
5439854397 int64x1x3_t __ret; \
54398 int64x1x3_t __s1 = __p1; \
5439954399 __builtin_neon_vld3_lane_v(&__ret, __p0, (int8x8_t)__s1.val[0], (int8x8_t)__s1.val[1], (int8x8_t)__s1.val[2], __p2, 3); \
5440054400 __ret; \
5440154401})
......@@ -54534,27 +54534,27 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5453454534 __ret; \
5453554535})
5453654536#define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54537 poly64x1x4_t __s1 = __p1; \
5453854537 poly64x1x4_t __ret; \
54538 poly64x1x4_t __s1 = __p1; \
5453954539 __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); \
5454054540 __ret; \
5454154541})
5454254542#ifdef __LITTLE_ENDIAN__
5454354543#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54544 poly8x16x4_t __s1 = __p1; \
5454554544 poly8x16x4_t __ret; \
54545 poly8x16x4_t __s1 = __p1; \
5454654546 __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); \
5454754547 __ret; \
5454854548})
5454954549#else
5455054550#define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
54551 poly8x16x4_t __ret; \
5455154552 poly8x16x4_t __s1 = __p1; \
5455254553 poly8x16x4_t __rev1; \
5455354554 __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); \
5455454555 __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); \
5455554556 __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); \
5455654557 __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); \
54557 poly8x16x4_t __ret; \
5455854558 __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); \
5455954559 \
5456054560 __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); \
......@@ -54567,20 +54567,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5456754567
5456854568#ifdef __LITTLE_ENDIAN__
5456954569#define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54570 poly64x2x4_t __s1 = __p1; \
5457154570 poly64x2x4_t __ret; \
54571 poly64x2x4_t __s1 = __p1; \
5457254572 __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); \
5457354573 __ret; \
5457454574})
5457554575#else
5457654576#define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
54577 poly64x2x4_t __ret; \
5457754578 poly64x2x4_t __s1 = __p1; \
5457854579 poly64x2x4_t __rev1; \
5457954580 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5458054581 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5458154582 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
5458254583 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54583 poly64x2x4_t __ret; \
5458454584 __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); \
5458554585 \
5458654586 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54593,20 +54593,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5459354593
5459454594#ifdef __LITTLE_ENDIAN__
5459554595#define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54596 uint8x16x4_t __s1 = __p1; \
5459754596 uint8x16x4_t __ret; \
54597 uint8x16x4_t __s1 = __p1; \
5459854598 __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); \
5459954599 __ret; \
5460054600})
5460154601#else
5460254602#define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
54603 uint8x16x4_t __ret; \
5460354604 uint8x16x4_t __s1 = __p1; \
5460454605 uint8x16x4_t __rev1; \
5460554606 __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); \
5460654607 __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); \
5460754608 __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); \
5460854609 __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); \
54609 uint8x16x4_t __ret; \
5461054610 __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); \
5461154611 \
5461254612 __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); \
......@@ -54619,20 +54619,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5461954619
5462054620#ifdef __LITTLE_ENDIAN__
5462154621#define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54622 uint64x2x4_t __s1 = __p1; \
5462354622 uint64x2x4_t __ret; \
54623 uint64x2x4_t __s1 = __p1; \
5462454624 __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); \
5462554625 __ret; \
5462654626})
5462754627#else
5462854628#define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54629 uint64x2x4_t __ret; \
5462954630 uint64x2x4_t __s1 = __p1; \
5463054631 uint64x2x4_t __rev1; \
5463154632 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5463254633 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5463354634 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
5463454635 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54635 uint64x2x4_t __ret; \
5463654636 __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); \
5463754637 \
5463854638 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54645,20 +54645,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5464554645
5464654646#ifdef __LITTLE_ENDIAN__
5464754647#define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54648 int8x16x4_t __s1 = __p1; \
5464954648 int8x16x4_t __ret; \
54649 int8x16x4_t __s1 = __p1; \
5465054650 __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); \
5465154651 __ret; \
5465254652})
5465354653#else
5465454654#define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
54655 int8x16x4_t __ret; \
5465554656 int8x16x4_t __s1 = __p1; \
5465654657 int8x16x4_t __rev1; \
5465754658 __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); \
5465854659 __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); \
5465954660 __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); \
5466054661 __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); \
54661 int8x16x4_t __ret; \
5466254662 __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); \
5466354663 \
5466454664 __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); \
......@@ -54671,20 +54671,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5467154671
5467254672#ifdef __LITTLE_ENDIAN__
5467354673#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54674 float64x2x4_t __s1 = __p1; \
5467554674 float64x2x4_t __ret; \
54675 float64x2x4_t __s1 = __p1; \
5467654676 __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); \
5467754677 __ret; \
5467854678})
5467954679#else
5468054680#define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54681 float64x2x4_t __ret; \
5468154682 float64x2x4_t __s1 = __p1; \
5468254683 float64x2x4_t __rev1; \
5468354684 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5468454685 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5468554686 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
5468654687 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54687 float64x2x4_t __ret; \
5468854688 __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); \
5468954689 \
5469054690 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54697,20 +54697,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5469754697
5469854698#ifdef __LITTLE_ENDIAN__
5469954699#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54700 int64x2x4_t __s1 = __p1; \
5470154700 int64x2x4_t __ret; \
54701 int64x2x4_t __s1 = __p1; \
5470254702 __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); \
5470354703 __ret; \
5470454704})
5470554705#else
5470654706#define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54707 int64x2x4_t __ret; \
5470754708 int64x2x4_t __s1 = __p1; \
5470854709 int64x2x4_t __rev1; \
5470954710 __rev1.val[0] = __builtin_shufflevector(__s1.val[0], __s1.val[0], 1, 0); \
5471054711 __rev1.val[1] = __builtin_shufflevector(__s1.val[1], __s1.val[1], 1, 0); \
5471154712 __rev1.val[2] = __builtin_shufflevector(__s1.val[2], __s1.val[2], 1, 0); \
5471254713 __rev1.val[3] = __builtin_shufflevector(__s1.val[3], __s1.val[3], 1, 0); \
54713 int64x2x4_t __ret; \
5471454714 __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); \
5471554715 \
5471654716 __ret.val[0] = __builtin_shufflevector(__ret.val[0], __ret.val[0], 1, 0); \
......@@ -54722,20 +54722,20 @@ __ai float64x1_t vget_low_f64(float64x2_t __p0) {
5472254722#endif
5472354723
5472454724#define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
54725 uint64x1x4_t __s1 = __p1; \
5472654725 uint64x1x4_t __ret; \
54726 uint64x1x4_t __s1 = __p1; \
5472754727 __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); \
5472854728 __ret; \
5472954729})
5473054730#define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54731 float64x1x4_t __s1 = __p1; \
5473254731 float64x1x4_t __ret; \
54732 float64x1x4_t __s1 = __p1; \
5473354733 __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); \
5473454734 __ret; \
5473554735})
5473654736#define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
54737 int64x1x4_t __s1 = __p1; \
5473854737 int64x1x4_t __ret; \
54738 int64x1x4_t __s1 = __p1; \
5473954739 __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); \
5474054740 __ret; \
5474154741})
......@@ -54752,9 +54752,9 @@ __ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
5475254752}
5475354753#else
5475454754__ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
54755 float64x2_t __ret;
5475554756 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5475654757 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54757 float64x2_t __ret;
5475854758 __ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5475954759 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5476054760 return __ret;
......@@ -54774,8 +54774,8 @@ __ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
5477454774}
5477554775#else
5477654776__ai float64_t vmaxnmvq_f64(float64x2_t __p0) {
54777 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5477854777 float64_t __ret;
54778 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5477954779 __ret = (float64_t) __builtin_neon_vmaxnmvq_f64(__rev0);
5478054780 return __ret;
5478154781}
......@@ -54789,8 +54789,8 @@ __ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
5478954789}
5479054790#else
5479154791__ai float32_t vmaxnmvq_f32(float32x4_t __p0) {
54792 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5479354792 float32_t __ret;
54793 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5479454794 __ret = (float32_t) __builtin_neon_vmaxnmvq_f32(__rev0);
5479554795 return __ret;
5479654796}
......@@ -54804,8 +54804,8 @@ __ai float32_t vmaxnmv_f32(float32x2_t __p0) {
5480454804}
5480554805#else
5480654806__ai float32_t vmaxnmv_f32(float32x2_t __p0) {
54807 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5480854807 float32_t __ret;
54808 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5480954809 __ret = (float32_t) __builtin_neon_vmaxnmv_f32(__rev0);
5481054810 return __ret;
5481154811}
......@@ -54819,8 +54819,8 @@ __ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
5481954819}
5482054820#else
5482154821__ai uint8_t vmaxvq_u8(uint8x16_t __p0) {
54822 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5482354822 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);
5482454824 __ret = (uint8_t) __builtin_neon_vmaxvq_u8(__rev0);
5482554825 return __ret;
5482654826}
......@@ -54834,8 +54834,8 @@ __ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
5483454834}
5483554835#else
5483654836__ai uint32_t vmaxvq_u32(uint32x4_t __p0) {
54837 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5483854837 uint32_t __ret;
54838 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5483954839 __ret = (uint32_t) __builtin_neon_vmaxvq_u32(__rev0);
5484054840 return __ret;
5484154841}
......@@ -54849,8 +54849,8 @@ __ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
5484954849}
5485054850#else
5485154851__ai uint16_t vmaxvq_u16(uint16x8_t __p0) {
54852 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5485354852 uint16_t __ret;
54853 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5485454854 __ret = (uint16_t) __builtin_neon_vmaxvq_u16(__rev0);
5485554855 return __ret;
5485654856}
......@@ -54864,8 +54864,8 @@ __ai int8_t vmaxvq_s8(int8x16_t __p0) {
5486454864}
5486554865#else
5486654866__ai int8_t vmaxvq_s8(int8x16_t __p0) {
54867 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5486854867 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);
5486954869 __ret = (int8_t) __builtin_neon_vmaxvq_s8(__rev0);
5487054870 return __ret;
5487154871}
......@@ -54879,8 +54879,8 @@ __ai float64_t vmaxvq_f64(float64x2_t __p0) {
5487954879}
5488054880#else
5488154881__ai float64_t vmaxvq_f64(float64x2_t __p0) {
54882 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5488354882 float64_t __ret;
54883 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5488454884 __ret = (float64_t) __builtin_neon_vmaxvq_f64(__rev0);
5488554885 return __ret;
5488654886}
......@@ -54894,8 +54894,8 @@ __ai float32_t vmaxvq_f32(float32x4_t __p0) {
5489454894}
5489554895#else
5489654896__ai float32_t vmaxvq_f32(float32x4_t __p0) {
54897 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5489854897 float32_t __ret;
54898 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5489954899 __ret = (float32_t) __builtin_neon_vmaxvq_f32(__rev0);
5490054900 return __ret;
5490154901}
......@@ -54909,8 +54909,8 @@ __ai int32_t vmaxvq_s32(int32x4_t __p0) {
5490954909}
5491054910#else
5491154911__ai int32_t vmaxvq_s32(int32x4_t __p0) {
54912 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5491354912 int32_t __ret;
54913 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5491454914 __ret = (int32_t) __builtin_neon_vmaxvq_s32(__rev0);
5491554915 return __ret;
5491654916}
......@@ -54924,8 +54924,8 @@ __ai int16_t vmaxvq_s16(int16x8_t __p0) {
5492454924}
5492554925#else
5492654926__ai int16_t vmaxvq_s16(int16x8_t __p0) {
54927 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5492854927 int16_t __ret;
54928 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5492954929 __ret = (int16_t) __builtin_neon_vmaxvq_s16(__rev0);
5493054930 return __ret;
5493154931}
......@@ -54939,8 +54939,8 @@ __ai uint8_t vmaxv_u8(uint8x8_t __p0) {
5493954939}
5494054940#else
5494154941__ai uint8_t vmaxv_u8(uint8x8_t __p0) {
54942 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5494354942 uint8_t __ret;
54943 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5494454944 __ret = (uint8_t) __builtin_neon_vmaxv_u8(__rev0);
5494554945 return __ret;
5494654946}
......@@ -54954,8 +54954,8 @@ __ai uint32_t vmaxv_u32(uint32x2_t __p0) {
5495454954}
5495554955#else
5495654956__ai uint32_t vmaxv_u32(uint32x2_t __p0) {
54957 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5495854957 uint32_t __ret;
54958 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5495954959 __ret = (uint32_t) __builtin_neon_vmaxv_u32(__rev0);
5496054960 return __ret;
5496154961}
......@@ -54969,8 +54969,8 @@ __ai uint16_t vmaxv_u16(uint16x4_t __p0) {
5496954969}
5497054970#else
5497154971__ai uint16_t vmaxv_u16(uint16x4_t __p0) {
54972 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5497354972 uint16_t __ret;
54973 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5497454974 __ret = (uint16_t) __builtin_neon_vmaxv_u16(__rev0);
5497554975 return __ret;
5497654976}
......@@ -54984,8 +54984,8 @@ __ai int8_t vmaxv_s8(int8x8_t __p0) {
5498454984}
5498554985#else
5498654986__ai int8_t vmaxv_s8(int8x8_t __p0) {
54987 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5498854987 int8_t __ret;
54988 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5498954989 __ret = (int8_t) __builtin_neon_vmaxv_s8(__rev0);
5499054990 return __ret;
5499154991}
......@@ -54999,8 +54999,8 @@ __ai float32_t vmaxv_f32(float32x2_t __p0) {
5499954999}
5500055000#else
5500155001__ai float32_t vmaxv_f32(float32x2_t __p0) {
55002 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5500355002 float32_t __ret;
55003 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5500455004 __ret = (float32_t) __builtin_neon_vmaxv_f32(__rev0);
5500555005 return __ret;
5500655006}
......@@ -55014,8 +55014,8 @@ __ai int32_t vmaxv_s32(int32x2_t __p0) {
5501455014}
5501555015#else
5501655016__ai int32_t vmaxv_s32(int32x2_t __p0) {
55017 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5501855017 int32_t __ret;
55018 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5501955019 __ret = (int32_t) __builtin_neon_vmaxv_s32(__rev0);
5502055020 return __ret;
5502155021}
......@@ -55029,8 +55029,8 @@ __ai int16_t vmaxv_s16(int16x4_t __p0) {
5502955029}
5503055030#else
5503155031__ai int16_t vmaxv_s16(int16x4_t __p0) {
55032 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5503355032 int16_t __ret;
55033 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5503455034 __ret = (int16_t) __builtin_neon_vmaxv_s16(__rev0);
5503555035 return __ret;
5503655036}
......@@ -55044,9 +55044,9 @@ __ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
5504455044}
5504555045#else
5504655046__ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
55047 float64x2_t __ret;
5504755048 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5504855049 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55049 float64x2_t __ret;
5505055050 __ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5505155051 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5505255052 return __ret;
......@@ -55066,8 +55066,8 @@ __ai float64_t vminnmvq_f64(float64x2_t __p0) {
5506655066}
5506755067#else
5506855068__ai float64_t vminnmvq_f64(float64x2_t __p0) {
55069 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5507055069 float64_t __ret;
55070 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5507155071 __ret = (float64_t) __builtin_neon_vminnmvq_f64(__rev0);
5507255072 return __ret;
5507355073}
......@@ -55081,8 +55081,8 @@ __ai float32_t vminnmvq_f32(float32x4_t __p0) {
5508155081}
5508255082#else
5508355083__ai float32_t vminnmvq_f32(float32x4_t __p0) {
55084 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5508555084 float32_t __ret;
55085 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5508655086 __ret = (float32_t) __builtin_neon_vminnmvq_f32(__rev0);
5508755087 return __ret;
5508855088}
......@@ -55096,8 +55096,8 @@ __ai float32_t vminnmv_f32(float32x2_t __p0) {
5509655096}
5509755097#else
5509855098__ai float32_t vminnmv_f32(float32x2_t __p0) {
55099 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5510055099 float32_t __ret;
55100 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5510155101 __ret = (float32_t) __builtin_neon_vminnmv_f32(__rev0);
5510255102 return __ret;
5510355103}
......@@ -55111,8 +55111,8 @@ __ai uint8_t vminvq_u8(uint8x16_t __p0) {
5511155111}
5511255112#else
5511355113__ai uint8_t vminvq_u8(uint8x16_t __p0) {
55114 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5511555114 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);
5511655116 __ret = (uint8_t) __builtin_neon_vminvq_u8(__rev0);
5511755117 return __ret;
5511855118}
......@@ -55126,8 +55126,8 @@ __ai uint32_t vminvq_u32(uint32x4_t __p0) {
5512655126}
5512755127#else
5512855128__ai uint32_t vminvq_u32(uint32x4_t __p0) {
55129 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5513055129 uint32_t __ret;
55130 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5513155131 __ret = (uint32_t) __builtin_neon_vminvq_u32(__rev0);
5513255132 return __ret;
5513355133}
......@@ -55141,8 +55141,8 @@ __ai uint16_t vminvq_u16(uint16x8_t __p0) {
5514155141}
5514255142#else
5514355143__ai uint16_t vminvq_u16(uint16x8_t __p0) {
55144 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5514555144 uint16_t __ret;
55145 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5514655146 __ret = (uint16_t) __builtin_neon_vminvq_u16(__rev0);
5514755147 return __ret;
5514855148}
......@@ -55156,8 +55156,8 @@ __ai int8_t vminvq_s8(int8x16_t __p0) {
5515655156}
5515755157#else
5515855158__ai int8_t vminvq_s8(int8x16_t __p0) {
55159 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5516055159 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);
5516155161 __ret = (int8_t) __builtin_neon_vminvq_s8(__rev0);
5516255162 return __ret;
5516355163}
......@@ -55171,8 +55171,8 @@ __ai float64_t vminvq_f64(float64x2_t __p0) {
5517155171}
5517255172#else
5517355173__ai float64_t vminvq_f64(float64x2_t __p0) {
55174 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5517555174 float64_t __ret;
55175 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5517655176 __ret = (float64_t) __builtin_neon_vminvq_f64(__rev0);
5517755177 return __ret;
5517855178}
......@@ -55186,8 +55186,8 @@ __ai float32_t vminvq_f32(float32x4_t __p0) {
5518655186}
5518755187#else
5518855188__ai float32_t vminvq_f32(float32x4_t __p0) {
55189 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5519055189 float32_t __ret;
55190 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5519155191 __ret = (float32_t) __builtin_neon_vminvq_f32(__rev0);
5519255192 return __ret;
5519355193}
......@@ -55201,8 +55201,8 @@ __ai int32_t vminvq_s32(int32x4_t __p0) {
5520155201}
5520255202#else
5520355203__ai int32_t vminvq_s32(int32x4_t __p0) {
55204 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5520555204 int32_t __ret;
55205 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5520655206 __ret = (int32_t) __builtin_neon_vminvq_s32(__rev0);
5520755207 return __ret;
5520855208}
......@@ -55216,8 +55216,8 @@ __ai int16_t vminvq_s16(int16x8_t __p0) {
5521655216}
5521755217#else
5521855218__ai int16_t vminvq_s16(int16x8_t __p0) {
55219 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5522055219 int16_t __ret;
55220 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5522155221 __ret = (int16_t) __builtin_neon_vminvq_s16(__rev0);
5522255222 return __ret;
5522355223}
......@@ -55231,8 +55231,8 @@ __ai uint8_t vminv_u8(uint8x8_t __p0) {
5523155231}
5523255232#else
5523355233__ai uint8_t vminv_u8(uint8x8_t __p0) {
55234 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5523555234 uint8_t __ret;
55235 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5523655236 __ret = (uint8_t) __builtin_neon_vminv_u8(__rev0);
5523755237 return __ret;
5523855238}
......@@ -55246,8 +55246,8 @@ __ai uint32_t vminv_u32(uint32x2_t __p0) {
5524655246}
5524755247#else
5524855248__ai uint32_t vminv_u32(uint32x2_t __p0) {
55249 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5525055249 uint32_t __ret;
55250 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5525155251 __ret = (uint32_t) __builtin_neon_vminv_u32(__rev0);
5525255252 return __ret;
5525355253}
......@@ -55261,8 +55261,8 @@ __ai uint16_t vminv_u16(uint16x4_t __p0) {
5526155261}
5526255262#else
5526355263__ai uint16_t vminv_u16(uint16x4_t __p0) {
55264 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5526555264 uint16_t __ret;
55265 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5526655266 __ret = (uint16_t) __builtin_neon_vminv_u16(__rev0);
5526755267 return __ret;
5526855268}
......@@ -55276,8 +55276,8 @@ __ai int8_t vminv_s8(int8x8_t __p0) {
5527655276}
5527755277#else
5527855278__ai int8_t vminv_s8(int8x8_t __p0) {
55279 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5528055279 int8_t __ret;
55280 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5528155281 __ret = (int8_t) __builtin_neon_vminv_s8(__rev0);
5528255282 return __ret;
5528355283}
......@@ -55291,8 +55291,8 @@ __ai float32_t vminv_f32(float32x2_t __p0) {
5529155291}
5529255292#else
5529355293__ai float32_t vminv_f32(float32x2_t __p0) {
55294 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5529555294 float32_t __ret;
55295 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5529655296 __ret = (float32_t) __builtin_neon_vminv_f32(__rev0);
5529755297 return __ret;
5529855298}
......@@ -55306,8 +55306,8 @@ __ai int32_t vminv_s32(int32x2_t __p0) {
5530655306}
5530755307#else
5530855308__ai int32_t vminv_s32(int32x2_t __p0) {
55309 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5531055309 int32_t __ret;
55310 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5531155311 __ret = (int32_t) __builtin_neon_vminv_s32(__rev0);
5531255312 return __ret;
5531355313}
......@@ -55321,8 +55321,8 @@ __ai int16_t vminv_s16(int16x4_t __p0) {
5532155321}
5532255322#else
5532355323__ai int16_t vminv_s16(int16x4_t __p0) {
55324 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5532555324 int16_t __ret;
55325 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5532655326 __ret = (int16_t) __builtin_neon_vminv_s16(__rev0);
5532755327 return __ret;
5532855328}
......@@ -55336,10 +55336,10 @@ __ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2)
5533655336}
5533755337#else
5533855338__ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
55339 float64x2_t __ret;
5533955340 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5534055341 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5534155342 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
55342 float64x2_t __ret;
5534355343 __ret = __rev0 + __rev1 * __rev2;
5534455344 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5534555345 return __ret;
......@@ -55353,22 +55353,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5535355353}
5535455354#ifdef __LITTLE_ENDIAN__
5535555355#define vmlaq_laneq_u32(__p0_531, __p1_531, __p2_531, __p3_531) __extension__ ({ \
55356 uint32x4_t __ret_531; \
5535655357 uint32x4_t __s0_531 = __p0_531; \
5535755358 uint32x4_t __s1_531 = __p1_531; \
5535855359 uint32x4_t __s2_531 = __p2_531; \
55359 uint32x4_t __ret_531; \
5536055360 __ret_531 = __s0_531 + __s1_531 * splatq_laneq_u32(__s2_531, __p3_531); \
5536155361 __ret_531; \
5536255362})
5536355363#else
5536455364#define vmlaq_laneq_u32(__p0_532, __p1_532, __p2_532, __p3_532) __extension__ ({ \
55365 uint32x4_t __ret_532; \
5536555366 uint32x4_t __s0_532 = __p0_532; \
5536655367 uint32x4_t __s1_532 = __p1_532; \
5536755368 uint32x4_t __s2_532 = __p2_532; \
5536855369 uint32x4_t __rev0_532; __rev0_532 = __builtin_shufflevector(__s0_532, __s0_532, 3, 2, 1, 0); \
5536955370 uint32x4_t __rev1_532; __rev1_532 = __builtin_shufflevector(__s1_532, __s1_532, 3, 2, 1, 0); \
5537055371 uint32x4_t __rev2_532; __rev2_532 = __builtin_shufflevector(__s2_532, __s2_532, 3, 2, 1, 0); \
55371 uint32x4_t __ret_532; \
5537255372 __ret_532 = __rev0_532 + __rev1_532 * __noswap_splatq_laneq_u32(__rev2_532, __p3_532); \
5537355373 __ret_532 = __builtin_shufflevector(__ret_532, __ret_532, 3, 2, 1, 0); \
5537455374 __ret_532; \
......@@ -55377,22 +55377,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5537755377
5537855378#ifdef __LITTLE_ENDIAN__
5537955379#define vmlaq_laneq_u16(__p0_533, __p1_533, __p2_533, __p3_533) __extension__ ({ \
55380 uint16x8_t __ret_533; \
5538055381 uint16x8_t __s0_533 = __p0_533; \
5538155382 uint16x8_t __s1_533 = __p1_533; \
5538255383 uint16x8_t __s2_533 = __p2_533; \
55383 uint16x8_t __ret_533; \
5538455384 __ret_533 = __s0_533 + __s1_533 * splatq_laneq_u16(__s2_533, __p3_533); \
5538555385 __ret_533; \
5538655386})
5538755387#else
5538855388#define vmlaq_laneq_u16(__p0_534, __p1_534, __p2_534, __p3_534) __extension__ ({ \
55389 uint16x8_t __ret_534; \
5538955390 uint16x8_t __s0_534 = __p0_534; \
5539055391 uint16x8_t __s1_534 = __p1_534; \
5539155392 uint16x8_t __s2_534 = __p2_534; \
5539255393 uint16x8_t __rev0_534; __rev0_534 = __builtin_shufflevector(__s0_534, __s0_534, 7, 6, 5, 4, 3, 2, 1, 0); \
5539355394 uint16x8_t __rev1_534; __rev1_534 = __builtin_shufflevector(__s1_534, __s1_534, 7, 6, 5, 4, 3, 2, 1, 0); \
5539455395 uint16x8_t __rev2_534; __rev2_534 = __builtin_shufflevector(__s2_534, __s2_534, 7, 6, 5, 4, 3, 2, 1, 0); \
55395 uint16x8_t __ret_534; \
5539655396 __ret_534 = __rev0_534 + __rev1_534 * __noswap_splatq_laneq_u16(__rev2_534, __p3_534); \
5539755397 __ret_534 = __builtin_shufflevector(__ret_534, __ret_534, 7, 6, 5, 4, 3, 2, 1, 0); \
5539855398 __ret_534; \
......@@ -55401,22 +55401,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5540155401
5540255402#ifdef __LITTLE_ENDIAN__
5540355403#define vmlaq_laneq_f32(__p0_535, __p1_535, __p2_535, __p3_535) __extension__ ({ \
55404 float32x4_t __ret_535; \
5540455405 float32x4_t __s0_535 = __p0_535; \
5540555406 float32x4_t __s1_535 = __p1_535; \
5540655407 float32x4_t __s2_535 = __p2_535; \
55407 float32x4_t __ret_535; \
5540855408 __ret_535 = __s0_535 + __s1_535 * splatq_laneq_f32(__s2_535, __p3_535); \
5540955409 __ret_535; \
5541055410})
5541155411#else
5541255412#define vmlaq_laneq_f32(__p0_536, __p1_536, __p2_536, __p3_536) __extension__ ({ \
55413 float32x4_t __ret_536; \
5541355414 float32x4_t __s0_536 = __p0_536; \
5541455415 float32x4_t __s1_536 = __p1_536; \
5541555416 float32x4_t __s2_536 = __p2_536; \
5541655417 float32x4_t __rev0_536; __rev0_536 = __builtin_shufflevector(__s0_536, __s0_536, 3, 2, 1, 0); \
5541755418 float32x4_t __rev1_536; __rev1_536 = __builtin_shufflevector(__s1_536, __s1_536, 3, 2, 1, 0); \
5541855419 float32x4_t __rev2_536; __rev2_536 = __builtin_shufflevector(__s2_536, __s2_536, 3, 2, 1, 0); \
55419 float32x4_t __ret_536; \
5542055420 __ret_536 = __rev0_536 + __rev1_536 * __noswap_splatq_laneq_f32(__rev2_536, __p3_536); \
5542155421 __ret_536 = __builtin_shufflevector(__ret_536, __ret_536, 3, 2, 1, 0); \
5542255422 __ret_536; \
......@@ -55425,22 +55425,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5542555425
5542655426#ifdef __LITTLE_ENDIAN__
5542755427#define vmlaq_laneq_s32(__p0_537, __p1_537, __p2_537, __p3_537) __extension__ ({ \
55428 int32x4_t __ret_537; \
5542855429 int32x4_t __s0_537 = __p0_537; \
5542955430 int32x4_t __s1_537 = __p1_537; \
5543055431 int32x4_t __s2_537 = __p2_537; \
55431 int32x4_t __ret_537; \
5543255432 __ret_537 = __s0_537 + __s1_537 * splatq_laneq_s32(__s2_537, __p3_537); \
5543355433 __ret_537; \
5543455434})
5543555435#else
5543655436#define vmlaq_laneq_s32(__p0_538, __p1_538, __p2_538, __p3_538) __extension__ ({ \
55437 int32x4_t __ret_538; \
5543755438 int32x4_t __s0_538 = __p0_538; \
5543855439 int32x4_t __s1_538 = __p1_538; \
5543955440 int32x4_t __s2_538 = __p2_538; \
5544055441 int32x4_t __rev0_538; __rev0_538 = __builtin_shufflevector(__s0_538, __s0_538, 3, 2, 1, 0); \
5544155442 int32x4_t __rev1_538; __rev1_538 = __builtin_shufflevector(__s1_538, __s1_538, 3, 2, 1, 0); \
5544255443 int32x4_t __rev2_538; __rev2_538 = __builtin_shufflevector(__s2_538, __s2_538, 3, 2, 1, 0); \
55443 int32x4_t __ret_538; \
5544455444 __ret_538 = __rev0_538 + __rev1_538 * __noswap_splatq_laneq_s32(__rev2_538, __p3_538); \
5544555445 __ret_538 = __builtin_shufflevector(__ret_538, __ret_538, 3, 2, 1, 0); \
5544655446 __ret_538; \
......@@ -55449,22 +55449,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5544955449
5545055450#ifdef __LITTLE_ENDIAN__
5545155451#define vmlaq_laneq_s16(__p0_539, __p1_539, __p2_539, __p3_539) __extension__ ({ \
55452 int16x8_t __ret_539; \
5545255453 int16x8_t __s0_539 = __p0_539; \
5545355454 int16x8_t __s1_539 = __p1_539; \
5545455455 int16x8_t __s2_539 = __p2_539; \
55455 int16x8_t __ret_539; \
5545655456 __ret_539 = __s0_539 + __s1_539 * splatq_laneq_s16(__s2_539, __p3_539); \
5545755457 __ret_539; \
5545855458})
5545955459#else
5546055460#define vmlaq_laneq_s16(__p0_540, __p1_540, __p2_540, __p3_540) __extension__ ({ \
55461 int16x8_t __ret_540; \
5546155462 int16x8_t __s0_540 = __p0_540; \
5546255463 int16x8_t __s1_540 = __p1_540; \
5546355464 int16x8_t __s2_540 = __p2_540; \
5546455465 int16x8_t __rev0_540; __rev0_540 = __builtin_shufflevector(__s0_540, __s0_540, 7, 6, 5, 4, 3, 2, 1, 0); \
5546555466 int16x8_t __rev1_540; __rev1_540 = __builtin_shufflevector(__s1_540, __s1_540, 7, 6, 5, 4, 3, 2, 1, 0); \
5546655467 int16x8_t __rev2_540; __rev2_540 = __builtin_shufflevector(__s2_540, __s2_540, 7, 6, 5, 4, 3, 2, 1, 0); \
55467 int16x8_t __ret_540; \
5546855468 __ret_540 = __rev0_540 + __rev1_540 * __noswap_splatq_laneq_s16(__rev2_540, __p3_540); \
5546955469 __ret_540 = __builtin_shufflevector(__ret_540, __ret_540, 7, 6, 5, 4, 3, 2, 1, 0); \
5547055470 __ret_540; \
......@@ -55473,22 +55473,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5547355473
5547455474#ifdef __LITTLE_ENDIAN__
5547555475#define vmla_laneq_u32(__p0_541, __p1_541, __p2_541, __p3_541) __extension__ ({ \
55476 uint32x2_t __ret_541; \
5547655477 uint32x2_t __s0_541 = __p0_541; \
5547755478 uint32x2_t __s1_541 = __p1_541; \
5547855479 uint32x4_t __s2_541 = __p2_541; \
55479 uint32x2_t __ret_541; \
5548055480 __ret_541 = __s0_541 + __s1_541 * splat_laneq_u32(__s2_541, __p3_541); \
5548155481 __ret_541; \
5548255482})
5548355483#else
5548455484#define vmla_laneq_u32(__p0_542, __p1_542, __p2_542, __p3_542) __extension__ ({ \
55485 uint32x2_t __ret_542; \
5548555486 uint32x2_t __s0_542 = __p0_542; \
5548655487 uint32x2_t __s1_542 = __p1_542; \
5548755488 uint32x4_t __s2_542 = __p2_542; \
5548855489 uint32x2_t __rev0_542; __rev0_542 = __builtin_shufflevector(__s0_542, __s0_542, 1, 0); \
5548955490 uint32x2_t __rev1_542; __rev1_542 = __builtin_shufflevector(__s1_542, __s1_542, 1, 0); \
5549055491 uint32x4_t __rev2_542; __rev2_542 = __builtin_shufflevector(__s2_542, __s2_542, 3, 2, 1, 0); \
55491 uint32x2_t __ret_542; \
5549255492 __ret_542 = __rev0_542 + __rev1_542 * __noswap_splat_laneq_u32(__rev2_542, __p3_542); \
5549355493 __ret_542 = __builtin_shufflevector(__ret_542, __ret_542, 1, 0); \
5549455494 __ret_542; \
......@@ -55497,22 +55497,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5549755497
5549855498#ifdef __LITTLE_ENDIAN__
5549955499#define vmla_laneq_u16(__p0_543, __p1_543, __p2_543, __p3_543) __extension__ ({ \
55500 uint16x4_t __ret_543; \
5550055501 uint16x4_t __s0_543 = __p0_543; \
5550155502 uint16x4_t __s1_543 = __p1_543; \
5550255503 uint16x8_t __s2_543 = __p2_543; \
55503 uint16x4_t __ret_543; \
5550455504 __ret_543 = __s0_543 + __s1_543 * splat_laneq_u16(__s2_543, __p3_543); \
5550555505 __ret_543; \
5550655506})
5550755507#else
5550855508#define vmla_laneq_u16(__p0_544, __p1_544, __p2_544, __p3_544) __extension__ ({ \
55509 uint16x4_t __ret_544; \
5550955510 uint16x4_t __s0_544 = __p0_544; \
5551055511 uint16x4_t __s1_544 = __p1_544; \
5551155512 uint16x8_t __s2_544 = __p2_544; \
5551255513 uint16x4_t __rev0_544; __rev0_544 = __builtin_shufflevector(__s0_544, __s0_544, 3, 2, 1, 0); \
5551355514 uint16x4_t __rev1_544; __rev1_544 = __builtin_shufflevector(__s1_544, __s1_544, 3, 2, 1, 0); \
5551455515 uint16x8_t __rev2_544; __rev2_544 = __builtin_shufflevector(__s2_544, __s2_544, 7, 6, 5, 4, 3, 2, 1, 0); \
55515 uint16x4_t __ret_544; \
5551655516 __ret_544 = __rev0_544 + __rev1_544 * __noswap_splat_laneq_u16(__rev2_544, __p3_544); \
5551755517 __ret_544 = __builtin_shufflevector(__ret_544, __ret_544, 3, 2, 1, 0); \
5551855518 __ret_544; \
......@@ -55521,22 +55521,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5552155521
5552255522#ifdef __LITTLE_ENDIAN__
5552355523#define vmla_laneq_f32(__p0_545, __p1_545, __p2_545, __p3_545) __extension__ ({ \
55524 float32x2_t __ret_545; \
5552455525 float32x2_t __s0_545 = __p0_545; \
5552555526 float32x2_t __s1_545 = __p1_545; \
5552655527 float32x4_t __s2_545 = __p2_545; \
55527 float32x2_t __ret_545; \
5552855528 __ret_545 = __s0_545 + __s1_545 * splat_laneq_f32(__s2_545, __p3_545); \
5552955529 __ret_545; \
5553055530})
5553155531#else
5553255532#define vmla_laneq_f32(__p0_546, __p1_546, __p2_546, __p3_546) __extension__ ({ \
55533 float32x2_t __ret_546; \
5553355534 float32x2_t __s0_546 = __p0_546; \
5553455535 float32x2_t __s1_546 = __p1_546; \
5553555536 float32x4_t __s2_546 = __p2_546; \
5553655537 float32x2_t __rev0_546; __rev0_546 = __builtin_shufflevector(__s0_546, __s0_546, 1, 0); \
5553755538 float32x2_t __rev1_546; __rev1_546 = __builtin_shufflevector(__s1_546, __s1_546, 1, 0); \
5553855539 float32x4_t __rev2_546; __rev2_546 = __builtin_shufflevector(__s2_546, __s2_546, 3, 2, 1, 0); \
55539 float32x2_t __ret_546; \
5554055540 __ret_546 = __rev0_546 + __rev1_546 * __noswap_splat_laneq_f32(__rev2_546, __p3_546); \
5554155541 __ret_546 = __builtin_shufflevector(__ret_546, __ret_546, 1, 0); \
5554255542 __ret_546; \
......@@ -55545,22 +55545,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5554555545
5554655546#ifdef __LITTLE_ENDIAN__
5554755547#define vmla_laneq_s32(__p0_547, __p1_547, __p2_547, __p3_547) __extension__ ({ \
55548 int32x2_t __ret_547; \
5554855549 int32x2_t __s0_547 = __p0_547; \
5554955550 int32x2_t __s1_547 = __p1_547; \
5555055551 int32x4_t __s2_547 = __p2_547; \
55551 int32x2_t __ret_547; \
5555255552 __ret_547 = __s0_547 + __s1_547 * splat_laneq_s32(__s2_547, __p3_547); \
5555355553 __ret_547; \
5555455554})
5555555555#else
5555655556#define vmla_laneq_s32(__p0_548, __p1_548, __p2_548, __p3_548) __extension__ ({ \
55557 int32x2_t __ret_548; \
5555755558 int32x2_t __s0_548 = __p0_548; \
5555855559 int32x2_t __s1_548 = __p1_548; \
5555955560 int32x4_t __s2_548 = __p2_548; \
5556055561 int32x2_t __rev0_548; __rev0_548 = __builtin_shufflevector(__s0_548, __s0_548, 1, 0); \
5556155562 int32x2_t __rev1_548; __rev1_548 = __builtin_shufflevector(__s1_548, __s1_548, 1, 0); \
5556255563 int32x4_t __rev2_548; __rev2_548 = __builtin_shufflevector(__s2_548, __s2_548, 3, 2, 1, 0); \
55563 int32x2_t __ret_548; \
5556455564 __ret_548 = __rev0_548 + __rev1_548 * __noswap_splat_laneq_s32(__rev2_548, __p3_548); \
5556555565 __ret_548 = __builtin_shufflevector(__ret_548, __ret_548, 1, 0); \
5556655566 __ret_548; \
......@@ -55569,22 +55569,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5556955569
5557055570#ifdef __LITTLE_ENDIAN__
5557155571#define vmla_laneq_s16(__p0_549, __p1_549, __p2_549, __p3_549) __extension__ ({ \
55572 int16x4_t __ret_549; \
5557255573 int16x4_t __s0_549 = __p0_549; \
5557355574 int16x4_t __s1_549 = __p1_549; \
5557455575 int16x8_t __s2_549 = __p2_549; \
55575 int16x4_t __ret_549; \
5557655576 __ret_549 = __s0_549 + __s1_549 * splat_laneq_s16(__s2_549, __p3_549); \
5557755577 __ret_549; \
5557855578})
5557955579#else
5558055580#define vmla_laneq_s16(__p0_550, __p1_550, __p2_550, __p3_550) __extension__ ({ \
55581 int16x4_t __ret_550; \
5558155582 int16x4_t __s0_550 = __p0_550; \
5558255583 int16x4_t __s1_550 = __p1_550; \
5558355584 int16x8_t __s2_550 = __p2_550; \
5558455585 int16x4_t __rev0_550; __rev0_550 = __builtin_shufflevector(__s0_550, __s0_550, 3, 2, 1, 0); \
5558555586 int16x4_t __rev1_550; __rev1_550 = __builtin_shufflevector(__s1_550, __s1_550, 3, 2, 1, 0); \
5558655587 int16x8_t __rev2_550; __rev2_550 = __builtin_shufflevector(__s2_550, __s2_550, 7, 6, 5, 4, 3, 2, 1, 0); \
55587 int16x4_t __ret_550; \
5558855588 __ret_550 = __rev0_550 + __rev1_550 * __noswap_splat_laneq_s16(__rev2_550, __p3_550); \
5558955589 __ret_550 = __builtin_shufflevector(__ret_550, __ret_550, 3, 2, 1, 0); \
5559055590 __ret_550; \
......@@ -55593,22 +55593,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5559355593
5559455594#ifdef __LITTLE_ENDIAN__
5559555595#define vmlal_high_lane_u32(__p0_551, __p1_551, __p2_551, __p3_551) __extension__ ({ \
55596 uint64x2_t __ret_551; \
5559655597 uint64x2_t __s0_551 = __p0_551; \
5559755598 uint32x4_t __s1_551 = __p1_551; \
5559855599 uint32x2_t __s2_551 = __p2_551; \
55599 uint64x2_t __ret_551; \
5560055600 __ret_551 = __s0_551 + vmull_u32(vget_high_u32(__s1_551), splat_lane_u32(__s2_551, __p3_551)); \
5560155601 __ret_551; \
5560255602})
5560355603#else
5560455604#define vmlal_high_lane_u32(__p0_552, __p1_552, __p2_552, __p3_552) __extension__ ({ \
55605 uint64x2_t __ret_552; \
5560555606 uint64x2_t __s0_552 = __p0_552; \
5560655607 uint32x4_t __s1_552 = __p1_552; \
5560755608 uint32x2_t __s2_552 = __p2_552; \
5560855609 uint64x2_t __rev0_552; __rev0_552 = __builtin_shufflevector(__s0_552, __s0_552, 1, 0); \
5560955610 uint32x4_t __rev1_552; __rev1_552 = __builtin_shufflevector(__s1_552, __s1_552, 3, 2, 1, 0); \
5561055611 uint32x2_t __rev2_552; __rev2_552 = __builtin_shufflevector(__s2_552, __s2_552, 1, 0); \
55611 uint64x2_t __ret_552; \
5561255612 __ret_552 = __rev0_552 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_552), __noswap_splat_lane_u32(__rev2_552, __p3_552)); \
5561355613 __ret_552 = __builtin_shufflevector(__ret_552, __ret_552, 1, 0); \
5561455614 __ret_552; \
......@@ -55617,22 +55617,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5561755617
5561855618#ifdef __LITTLE_ENDIAN__
5561955619#define vmlal_high_lane_u16(__p0_553, __p1_553, __p2_553, __p3_553) __extension__ ({ \
55620 uint32x4_t __ret_553; \
5562055621 uint32x4_t __s0_553 = __p0_553; \
5562155622 uint16x8_t __s1_553 = __p1_553; \
5562255623 uint16x4_t __s2_553 = __p2_553; \
55623 uint32x4_t __ret_553; \
5562455624 __ret_553 = __s0_553 + vmull_u16(vget_high_u16(__s1_553), splat_lane_u16(__s2_553, __p3_553)); \
5562555625 __ret_553; \
5562655626})
5562755627#else
5562855628#define vmlal_high_lane_u16(__p0_554, __p1_554, __p2_554, __p3_554) __extension__ ({ \
55629 uint32x4_t __ret_554; \
5562955630 uint32x4_t __s0_554 = __p0_554; \
5563055631 uint16x8_t __s1_554 = __p1_554; \
5563155632 uint16x4_t __s2_554 = __p2_554; \
5563255633 uint32x4_t __rev0_554; __rev0_554 = __builtin_shufflevector(__s0_554, __s0_554, 3, 2, 1, 0); \
5563355634 uint16x8_t __rev1_554; __rev1_554 = __builtin_shufflevector(__s1_554, __s1_554, 7, 6, 5, 4, 3, 2, 1, 0); \
5563455635 uint16x4_t __rev2_554; __rev2_554 = __builtin_shufflevector(__s2_554, __s2_554, 3, 2, 1, 0); \
55635 uint32x4_t __ret_554; \
5563655636 __ret_554 = __rev0_554 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_554), __noswap_splat_lane_u16(__rev2_554, __p3_554)); \
5563755637 __ret_554 = __builtin_shufflevector(__ret_554, __ret_554, 3, 2, 1, 0); \
5563855638 __ret_554; \
......@@ -55641,22 +55641,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5564155641
5564255642#ifdef __LITTLE_ENDIAN__
5564355643#define vmlal_high_lane_s32(__p0_555, __p1_555, __p2_555, __p3_555) __extension__ ({ \
55644 int64x2_t __ret_555; \
5564455645 int64x2_t __s0_555 = __p0_555; \
5564555646 int32x4_t __s1_555 = __p1_555; \
5564655647 int32x2_t __s2_555 = __p2_555; \
55647 int64x2_t __ret_555; \
5564855648 __ret_555 = __s0_555 + vmull_s32(vget_high_s32(__s1_555), splat_lane_s32(__s2_555, __p3_555)); \
5564955649 __ret_555; \
5565055650})
5565155651#else
5565255652#define vmlal_high_lane_s32(__p0_556, __p1_556, __p2_556, __p3_556) __extension__ ({ \
55653 int64x2_t __ret_556; \
5565355654 int64x2_t __s0_556 = __p0_556; \
5565455655 int32x4_t __s1_556 = __p1_556; \
5565555656 int32x2_t __s2_556 = __p2_556; \
5565655657 int64x2_t __rev0_556; __rev0_556 = __builtin_shufflevector(__s0_556, __s0_556, 1, 0); \
5565755658 int32x4_t __rev1_556; __rev1_556 = __builtin_shufflevector(__s1_556, __s1_556, 3, 2, 1, 0); \
5565855659 int32x2_t __rev2_556; __rev2_556 = __builtin_shufflevector(__s2_556, __s2_556, 1, 0); \
55659 int64x2_t __ret_556; \
5566055660 __ret_556 = __rev0_556 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_556), __noswap_splat_lane_s32(__rev2_556, __p3_556)); \
5566155661 __ret_556 = __builtin_shufflevector(__ret_556, __ret_556, 1, 0); \
5566255662 __ret_556; \
......@@ -55665,22 +55665,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5566555665
5566655666#ifdef __LITTLE_ENDIAN__
5566755667#define vmlal_high_lane_s16(__p0_557, __p1_557, __p2_557, __p3_557) __extension__ ({ \
55668 int32x4_t __ret_557; \
5566855669 int32x4_t __s0_557 = __p0_557; \
5566955670 int16x8_t __s1_557 = __p1_557; \
5567055671 int16x4_t __s2_557 = __p2_557; \
55671 int32x4_t __ret_557; \
5567255672 __ret_557 = __s0_557 + vmull_s16(vget_high_s16(__s1_557), splat_lane_s16(__s2_557, __p3_557)); \
5567355673 __ret_557; \
5567455674})
5567555675#else
5567655676#define vmlal_high_lane_s16(__p0_558, __p1_558, __p2_558, __p3_558) __extension__ ({ \
55677 int32x4_t __ret_558; \
5567755678 int32x4_t __s0_558 = __p0_558; \
5567855679 int16x8_t __s1_558 = __p1_558; \
5567955680 int16x4_t __s2_558 = __p2_558; \
5568055681 int32x4_t __rev0_558; __rev0_558 = __builtin_shufflevector(__s0_558, __s0_558, 3, 2, 1, 0); \
5568155682 int16x8_t __rev1_558; __rev1_558 = __builtin_shufflevector(__s1_558, __s1_558, 7, 6, 5, 4, 3, 2, 1, 0); \
5568255683 int16x4_t __rev2_558; __rev2_558 = __builtin_shufflevector(__s2_558, __s2_558, 3, 2, 1, 0); \
55683 int32x4_t __ret_558; \
5568455684 __ret_558 = __rev0_558 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_558), __noswap_splat_lane_s16(__rev2_558, __p3_558)); \
5568555685 __ret_558 = __builtin_shufflevector(__ret_558, __ret_558, 3, 2, 1, 0); \
5568655686 __ret_558; \
......@@ -55689,22 +55689,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5568955689
5569055690#ifdef __LITTLE_ENDIAN__
5569155691#define vmlal_high_laneq_u32(__p0_559, __p1_559, __p2_559, __p3_559) __extension__ ({ \
55692 uint64x2_t __ret_559; \
5569255693 uint64x2_t __s0_559 = __p0_559; \
5569355694 uint32x4_t __s1_559 = __p1_559; \
5569455695 uint32x4_t __s2_559 = __p2_559; \
55695 uint64x2_t __ret_559; \
5569655696 __ret_559 = __s0_559 + vmull_u32(vget_high_u32(__s1_559), splat_laneq_u32(__s2_559, __p3_559)); \
5569755697 __ret_559; \
5569855698})
5569955699#else
5570055700#define vmlal_high_laneq_u32(__p0_560, __p1_560, __p2_560, __p3_560) __extension__ ({ \
55701 uint64x2_t __ret_560; \
5570155702 uint64x2_t __s0_560 = __p0_560; \
5570255703 uint32x4_t __s1_560 = __p1_560; \
5570355704 uint32x4_t __s2_560 = __p2_560; \
5570455705 uint64x2_t __rev0_560; __rev0_560 = __builtin_shufflevector(__s0_560, __s0_560, 1, 0); \
5570555706 uint32x4_t __rev1_560; __rev1_560 = __builtin_shufflevector(__s1_560, __s1_560, 3, 2, 1, 0); \
5570655707 uint32x4_t __rev2_560; __rev2_560 = __builtin_shufflevector(__s2_560, __s2_560, 3, 2, 1, 0); \
55707 uint64x2_t __ret_560; \
5570855708 __ret_560 = __rev0_560 + __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_560), __noswap_splat_laneq_u32(__rev2_560, __p3_560)); \
5570955709 __ret_560 = __builtin_shufflevector(__ret_560, __ret_560, 1, 0); \
5571055710 __ret_560; \
......@@ -55713,22 +55713,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5571355713
5571455714#ifdef __LITTLE_ENDIAN__
5571555715#define vmlal_high_laneq_u16(__p0_561, __p1_561, __p2_561, __p3_561) __extension__ ({ \
55716 uint32x4_t __ret_561; \
5571655717 uint32x4_t __s0_561 = __p0_561; \
5571755718 uint16x8_t __s1_561 = __p1_561; \
5571855719 uint16x8_t __s2_561 = __p2_561; \
55719 uint32x4_t __ret_561; \
5572055720 __ret_561 = __s0_561 + vmull_u16(vget_high_u16(__s1_561), splat_laneq_u16(__s2_561, __p3_561)); \
5572155721 __ret_561; \
5572255722})
5572355723#else
5572455724#define vmlal_high_laneq_u16(__p0_562, __p1_562, __p2_562, __p3_562) __extension__ ({ \
55725 uint32x4_t __ret_562; \
5572555726 uint32x4_t __s0_562 = __p0_562; \
5572655727 uint16x8_t __s1_562 = __p1_562; \
5572755728 uint16x8_t __s2_562 = __p2_562; \
5572855729 uint32x4_t __rev0_562; __rev0_562 = __builtin_shufflevector(__s0_562, __s0_562, 3, 2, 1, 0); \
5572955730 uint16x8_t __rev1_562; __rev1_562 = __builtin_shufflevector(__s1_562, __s1_562, 7, 6, 5, 4, 3, 2, 1, 0); \
5573055731 uint16x8_t __rev2_562; __rev2_562 = __builtin_shufflevector(__s2_562, __s2_562, 7, 6, 5, 4, 3, 2, 1, 0); \
55731 uint32x4_t __ret_562; \
5573255732 __ret_562 = __rev0_562 + __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_562), __noswap_splat_laneq_u16(__rev2_562, __p3_562)); \
5573355733 __ret_562 = __builtin_shufflevector(__ret_562, __ret_562, 3, 2, 1, 0); \
5573455734 __ret_562; \
......@@ -55737,22 +55737,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5573755737
5573855738#ifdef __LITTLE_ENDIAN__
5573955739#define vmlal_high_laneq_s32(__p0_563, __p1_563, __p2_563, __p3_563) __extension__ ({ \
55740 int64x2_t __ret_563; \
5574055741 int64x2_t __s0_563 = __p0_563; \
5574155742 int32x4_t __s1_563 = __p1_563; \
5574255743 int32x4_t __s2_563 = __p2_563; \
55743 int64x2_t __ret_563; \
5574455744 __ret_563 = __s0_563 + vmull_s32(vget_high_s32(__s1_563), splat_laneq_s32(__s2_563, __p3_563)); \
5574555745 __ret_563; \
5574655746})
5574755747#else
5574855748#define vmlal_high_laneq_s32(__p0_564, __p1_564, __p2_564, __p3_564) __extension__ ({ \
55749 int64x2_t __ret_564; \
5574955750 int64x2_t __s0_564 = __p0_564; \
5575055751 int32x4_t __s1_564 = __p1_564; \
5575155752 int32x4_t __s2_564 = __p2_564; \
5575255753 int64x2_t __rev0_564; __rev0_564 = __builtin_shufflevector(__s0_564, __s0_564, 1, 0); \
5575355754 int32x4_t __rev1_564; __rev1_564 = __builtin_shufflevector(__s1_564, __s1_564, 3, 2, 1, 0); \
5575455755 int32x4_t __rev2_564; __rev2_564 = __builtin_shufflevector(__s2_564, __s2_564, 3, 2, 1, 0); \
55755 int64x2_t __ret_564; \
5575655756 __ret_564 = __rev0_564 + __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_564), __noswap_splat_laneq_s32(__rev2_564, __p3_564)); \
5575755757 __ret_564 = __builtin_shufflevector(__ret_564, __ret_564, 1, 0); \
5575855758 __ret_564; \
......@@ -55761,22 +55761,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5576155761
5576255762#ifdef __LITTLE_ENDIAN__
5576355763#define vmlal_high_laneq_s16(__p0_565, __p1_565, __p2_565, __p3_565) __extension__ ({ \
55764 int32x4_t __ret_565; \
5576455765 int32x4_t __s0_565 = __p0_565; \
5576555766 int16x8_t __s1_565 = __p1_565; \
5576655767 int16x8_t __s2_565 = __p2_565; \
55767 int32x4_t __ret_565; \
5576855768 __ret_565 = __s0_565 + vmull_s16(vget_high_s16(__s1_565), splat_laneq_s16(__s2_565, __p3_565)); \
5576955769 __ret_565; \
5577055770})
5577155771#else
5577255772#define vmlal_high_laneq_s16(__p0_566, __p1_566, __p2_566, __p3_566) __extension__ ({ \
55773 int32x4_t __ret_566; \
5577355774 int32x4_t __s0_566 = __p0_566; \
5577455775 int16x8_t __s1_566 = __p1_566; \
5577555776 int16x8_t __s2_566 = __p2_566; \
5577655777 int32x4_t __rev0_566; __rev0_566 = __builtin_shufflevector(__s0_566, __s0_566, 3, 2, 1, 0); \
5577755778 int16x8_t __rev1_566; __rev1_566 = __builtin_shufflevector(__s1_566, __s1_566, 7, 6, 5, 4, 3, 2, 1, 0); \
5577855779 int16x8_t __rev2_566; __rev2_566 = __builtin_shufflevector(__s2_566, __s2_566, 7, 6, 5, 4, 3, 2, 1, 0); \
55779 int32x4_t __ret_566; \
5578055780 __ret_566 = __rev0_566 + __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_566), __noswap_splat_laneq_s16(__rev2_566, __p3_566)); \
5578155781 __ret_566 = __builtin_shufflevector(__ret_566, __ret_566, 3, 2, 1, 0); \
5578255782 __ret_566; \
......@@ -55785,22 +55785,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5578555785
5578655786#ifdef __LITTLE_ENDIAN__
5578755787#define vmlal_laneq_u32(__p0_567, __p1_567, __p2_567, __p3_567) __extension__ ({ \
55788 uint64x2_t __ret_567; \
5578855789 uint64x2_t __s0_567 = __p0_567; \
5578955790 uint32x2_t __s1_567 = __p1_567; \
5579055791 uint32x4_t __s2_567 = __p2_567; \
55791 uint64x2_t __ret_567; \
5579255792 __ret_567 = __s0_567 + vmull_u32(__s1_567, splat_laneq_u32(__s2_567, __p3_567)); \
5579355793 __ret_567; \
5579455794})
5579555795#else
5579655796#define vmlal_laneq_u32(__p0_568, __p1_568, __p2_568, __p3_568) __extension__ ({ \
55797 uint64x2_t __ret_568; \
5579755798 uint64x2_t __s0_568 = __p0_568; \
5579855799 uint32x2_t __s1_568 = __p1_568; \
5579955800 uint32x4_t __s2_568 = __p2_568; \
5580055801 uint64x2_t __rev0_568; __rev0_568 = __builtin_shufflevector(__s0_568, __s0_568, 1, 0); \
5580155802 uint32x2_t __rev1_568; __rev1_568 = __builtin_shufflevector(__s1_568, __s1_568, 1, 0); \
5580255803 uint32x4_t __rev2_568; __rev2_568 = __builtin_shufflevector(__s2_568, __s2_568, 3, 2, 1, 0); \
55803 uint64x2_t __ret_568; \
5580455804 __ret_568 = __rev0_568 + __noswap_vmull_u32(__rev1_568, __noswap_splat_laneq_u32(__rev2_568, __p3_568)); \
5580555805 __ret_568 = __builtin_shufflevector(__ret_568, __ret_568, 1, 0); \
5580655806 __ret_568; \
......@@ -55809,22 +55809,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5580955809
5581055810#ifdef __LITTLE_ENDIAN__
5581155811#define vmlal_laneq_u16(__p0_569, __p1_569, __p2_569, __p3_569) __extension__ ({ \
55812 uint32x4_t __ret_569; \
5581255813 uint32x4_t __s0_569 = __p0_569; \
5581355814 uint16x4_t __s1_569 = __p1_569; \
5581455815 uint16x8_t __s2_569 = __p2_569; \
55815 uint32x4_t __ret_569; \
5581655816 __ret_569 = __s0_569 + vmull_u16(__s1_569, splat_laneq_u16(__s2_569, __p3_569)); \
5581755817 __ret_569; \
5581855818})
5581955819#else
5582055820#define vmlal_laneq_u16(__p0_570, __p1_570, __p2_570, __p3_570) __extension__ ({ \
55821 uint32x4_t __ret_570; \
5582155822 uint32x4_t __s0_570 = __p0_570; \
5582255823 uint16x4_t __s1_570 = __p1_570; \
5582355824 uint16x8_t __s2_570 = __p2_570; \
5582455825 uint32x4_t __rev0_570; __rev0_570 = __builtin_shufflevector(__s0_570, __s0_570, 3, 2, 1, 0); \
5582555826 uint16x4_t __rev1_570; __rev1_570 = __builtin_shufflevector(__s1_570, __s1_570, 3, 2, 1, 0); \
5582655827 uint16x8_t __rev2_570; __rev2_570 = __builtin_shufflevector(__s2_570, __s2_570, 7, 6, 5, 4, 3, 2, 1, 0); \
55827 uint32x4_t __ret_570; \
5582855828 __ret_570 = __rev0_570 + __noswap_vmull_u16(__rev1_570, __noswap_splat_laneq_u16(__rev2_570, __p3_570)); \
5582955829 __ret_570 = __builtin_shufflevector(__ret_570, __ret_570, 3, 2, 1, 0); \
5583055830 __ret_570; \
......@@ -55833,22 +55833,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5583355833
5583455834#ifdef __LITTLE_ENDIAN__
5583555835#define vmlal_laneq_s32(__p0_571, __p1_571, __p2_571, __p3_571) __extension__ ({ \
55836 int64x2_t __ret_571; \
5583655837 int64x2_t __s0_571 = __p0_571; \
5583755838 int32x2_t __s1_571 = __p1_571; \
5583855839 int32x4_t __s2_571 = __p2_571; \
55839 int64x2_t __ret_571; \
5584055840 __ret_571 = __s0_571 + vmull_s32(__s1_571, splat_laneq_s32(__s2_571, __p3_571)); \
5584155841 __ret_571; \
5584255842})
5584355843#else
5584455844#define vmlal_laneq_s32(__p0_572, __p1_572, __p2_572, __p3_572) __extension__ ({ \
55845 int64x2_t __ret_572; \
5584555846 int64x2_t __s0_572 = __p0_572; \
5584655847 int32x2_t __s1_572 = __p1_572; \
5584755848 int32x4_t __s2_572 = __p2_572; \
5584855849 int64x2_t __rev0_572; __rev0_572 = __builtin_shufflevector(__s0_572, __s0_572, 1, 0); \
5584955850 int32x2_t __rev1_572; __rev1_572 = __builtin_shufflevector(__s1_572, __s1_572, 1, 0); \
5585055851 int32x4_t __rev2_572; __rev2_572 = __builtin_shufflevector(__s2_572, __s2_572, 3, 2, 1, 0); \
55851 int64x2_t __ret_572; \
5585255852 __ret_572 = __rev0_572 + __noswap_vmull_s32(__rev1_572, __noswap_splat_laneq_s32(__rev2_572, __p3_572)); \
5585355853 __ret_572 = __builtin_shufflevector(__ret_572, __ret_572, 1, 0); \
5585455854 __ret_572; \
......@@ -55857,22 +55857,22 @@ __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5585755857
5585855858#ifdef __LITTLE_ENDIAN__
5585955859#define vmlal_laneq_s16(__p0_573, __p1_573, __p2_573, __p3_573) __extension__ ({ \
55860 int32x4_t __ret_573; \
5586055861 int32x4_t __s0_573 = __p0_573; \
5586155862 int16x4_t __s1_573 = __p1_573; \
5586255863 int16x8_t __s2_573 = __p2_573; \
55863 int32x4_t __ret_573; \
5586455864 __ret_573 = __s0_573 + vmull_s16(__s1_573, splat_laneq_s16(__s2_573, __p3_573)); \
5586555865 __ret_573; \
5586655866})
5586755867#else
5586855868#define vmlal_laneq_s16(__p0_574, __p1_574, __p2_574, __p3_574) __extension__ ({ \
55869 int32x4_t __ret_574; \
5586955870 int32x4_t __s0_574 = __p0_574; \
5587055871 int16x4_t __s1_574 = __p1_574; \
5587155872 int16x8_t __s2_574 = __p2_574; \
5587255873 int32x4_t __rev0_574; __rev0_574 = __builtin_shufflevector(__s0_574, __s0_574, 3, 2, 1, 0); \
5587355874 int16x4_t __rev1_574; __rev1_574 = __builtin_shufflevector(__s1_574, __s1_574, 3, 2, 1, 0); \
5587455875 int16x8_t __rev2_574; __rev2_574 = __builtin_shufflevector(__s2_574, __s2_574, 7, 6, 5, 4, 3, 2, 1, 0); \
55875 int32x4_t __ret_574; \
5587655876 __ret_574 = __rev0_574 + __noswap_vmull_s16(__rev1_574, __noswap_splat_laneq_s16(__rev2_574, __p3_574)); \
5587755877 __ret_574 = __builtin_shufflevector(__ret_574, __ret_574, 3, 2, 1, 0); \
5587855878 __ret_574; \
......@@ -55887,10 +55887,10 @@ __ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2)
5588755887}
5588855888#else
5588955889__ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
55890 float64x2_t __ret;
5589055891 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5589155892 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5589255893 float64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
55893 float64x2_t __ret;
5589455894 __ret = __rev0 - __rev1 * __rev2;
5589555895 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5589655896 return __ret;
......@@ -55904,22 +55904,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5590455904}
5590555905#ifdef __LITTLE_ENDIAN__
5590655906#define vmlsq_laneq_u32(__p0_575, __p1_575, __p2_575, __p3_575) __extension__ ({ \
55907 uint32x4_t __ret_575; \
5590755908 uint32x4_t __s0_575 = __p0_575; \
5590855909 uint32x4_t __s1_575 = __p1_575; \
5590955910 uint32x4_t __s2_575 = __p2_575; \
55910 uint32x4_t __ret_575; \
5591155911 __ret_575 = __s0_575 - __s1_575 * splatq_laneq_u32(__s2_575, __p3_575); \
5591255912 __ret_575; \
5591355913})
5591455914#else
5591555915#define vmlsq_laneq_u32(__p0_576, __p1_576, __p2_576, __p3_576) __extension__ ({ \
55916 uint32x4_t __ret_576; \
5591655917 uint32x4_t __s0_576 = __p0_576; \
5591755918 uint32x4_t __s1_576 = __p1_576; \
5591855919 uint32x4_t __s2_576 = __p2_576; \
5591955920 uint32x4_t __rev0_576; __rev0_576 = __builtin_shufflevector(__s0_576, __s0_576, 3, 2, 1, 0); \
5592055921 uint32x4_t __rev1_576; __rev1_576 = __builtin_shufflevector(__s1_576, __s1_576, 3, 2, 1, 0); \
5592155922 uint32x4_t __rev2_576; __rev2_576 = __builtin_shufflevector(__s2_576, __s2_576, 3, 2, 1, 0); \
55922 uint32x4_t __ret_576; \
5592355923 __ret_576 = __rev0_576 - __rev1_576 * __noswap_splatq_laneq_u32(__rev2_576, __p3_576); \
5592455924 __ret_576 = __builtin_shufflevector(__ret_576, __ret_576, 3, 2, 1, 0); \
5592555925 __ret_576; \
......@@ -55928,22 +55928,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5592855928
5592955929#ifdef __LITTLE_ENDIAN__
5593055930#define vmlsq_laneq_u16(__p0_577, __p1_577, __p2_577, __p3_577) __extension__ ({ \
55931 uint16x8_t __ret_577; \
5593155932 uint16x8_t __s0_577 = __p0_577; \
5593255933 uint16x8_t __s1_577 = __p1_577; \
5593355934 uint16x8_t __s2_577 = __p2_577; \
55934 uint16x8_t __ret_577; \
5593555935 __ret_577 = __s0_577 - __s1_577 * splatq_laneq_u16(__s2_577, __p3_577); \
5593655936 __ret_577; \
5593755937})
5593855938#else
5593955939#define vmlsq_laneq_u16(__p0_578, __p1_578, __p2_578, __p3_578) __extension__ ({ \
55940 uint16x8_t __ret_578; \
5594055941 uint16x8_t __s0_578 = __p0_578; \
5594155942 uint16x8_t __s1_578 = __p1_578; \
5594255943 uint16x8_t __s2_578 = __p2_578; \
5594355944 uint16x8_t __rev0_578; __rev0_578 = __builtin_shufflevector(__s0_578, __s0_578, 7, 6, 5, 4, 3, 2, 1, 0); \
5594455945 uint16x8_t __rev1_578; __rev1_578 = __builtin_shufflevector(__s1_578, __s1_578, 7, 6, 5, 4, 3, 2, 1, 0); \
5594555946 uint16x8_t __rev2_578; __rev2_578 = __builtin_shufflevector(__s2_578, __s2_578, 7, 6, 5, 4, 3, 2, 1, 0); \
55946 uint16x8_t __ret_578; \
5594755947 __ret_578 = __rev0_578 - __rev1_578 * __noswap_splatq_laneq_u16(__rev2_578, __p3_578); \
5594855948 __ret_578 = __builtin_shufflevector(__ret_578, __ret_578, 7, 6, 5, 4, 3, 2, 1, 0); \
5594955949 __ret_578; \
......@@ -55952,22 +55952,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5595255952
5595355953#ifdef __LITTLE_ENDIAN__
5595455954#define vmlsq_laneq_f32(__p0_579, __p1_579, __p2_579, __p3_579) __extension__ ({ \
55955 float32x4_t __ret_579; \
5595555956 float32x4_t __s0_579 = __p0_579; \
5595655957 float32x4_t __s1_579 = __p1_579; \
5595755958 float32x4_t __s2_579 = __p2_579; \
55958 float32x4_t __ret_579; \
5595955959 __ret_579 = __s0_579 - __s1_579 * splatq_laneq_f32(__s2_579, __p3_579); \
5596055960 __ret_579; \
5596155961})
5596255962#else
5596355963#define vmlsq_laneq_f32(__p0_580, __p1_580, __p2_580, __p3_580) __extension__ ({ \
55964 float32x4_t __ret_580; \
5596455965 float32x4_t __s0_580 = __p0_580; \
5596555966 float32x4_t __s1_580 = __p1_580; \
5596655967 float32x4_t __s2_580 = __p2_580; \
5596755968 float32x4_t __rev0_580; __rev0_580 = __builtin_shufflevector(__s0_580, __s0_580, 3, 2, 1, 0); \
5596855969 float32x4_t __rev1_580; __rev1_580 = __builtin_shufflevector(__s1_580, __s1_580, 3, 2, 1, 0); \
5596955970 float32x4_t __rev2_580; __rev2_580 = __builtin_shufflevector(__s2_580, __s2_580, 3, 2, 1, 0); \
55970 float32x4_t __ret_580; \
5597155971 __ret_580 = __rev0_580 - __rev1_580 * __noswap_splatq_laneq_f32(__rev2_580, __p3_580); \
5597255972 __ret_580 = __builtin_shufflevector(__ret_580, __ret_580, 3, 2, 1, 0); \
5597355973 __ret_580; \
......@@ -55976,22 +55976,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5597655976
5597755977#ifdef __LITTLE_ENDIAN__
5597855978#define vmlsq_laneq_s32(__p0_581, __p1_581, __p2_581, __p3_581) __extension__ ({ \
55979 int32x4_t __ret_581; \
5597955980 int32x4_t __s0_581 = __p0_581; \
5598055981 int32x4_t __s1_581 = __p1_581; \
5598155982 int32x4_t __s2_581 = __p2_581; \
55982 int32x4_t __ret_581; \
5598355983 __ret_581 = __s0_581 - __s1_581 * splatq_laneq_s32(__s2_581, __p3_581); \
5598455984 __ret_581; \
5598555985})
5598655986#else
5598755987#define vmlsq_laneq_s32(__p0_582, __p1_582, __p2_582, __p3_582) __extension__ ({ \
55988 int32x4_t __ret_582; \
5598855989 int32x4_t __s0_582 = __p0_582; \
5598955990 int32x4_t __s1_582 = __p1_582; \
5599055991 int32x4_t __s2_582 = __p2_582; \
5599155992 int32x4_t __rev0_582; __rev0_582 = __builtin_shufflevector(__s0_582, __s0_582, 3, 2, 1, 0); \
5599255993 int32x4_t __rev1_582; __rev1_582 = __builtin_shufflevector(__s1_582, __s1_582, 3, 2, 1, 0); \
5599355994 int32x4_t __rev2_582; __rev2_582 = __builtin_shufflevector(__s2_582, __s2_582, 3, 2, 1, 0); \
55994 int32x4_t __ret_582; \
5599555995 __ret_582 = __rev0_582 - __rev1_582 * __noswap_splatq_laneq_s32(__rev2_582, __p3_582); \
5599655996 __ret_582 = __builtin_shufflevector(__ret_582, __ret_582, 3, 2, 1, 0); \
5599755997 __ret_582; \
......@@ -56000,22 +56000,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5600056000
5600156001#ifdef __LITTLE_ENDIAN__
5600256002#define vmlsq_laneq_s16(__p0_583, __p1_583, __p2_583, __p3_583) __extension__ ({ \
56003 int16x8_t __ret_583; \
5600356004 int16x8_t __s0_583 = __p0_583; \
5600456005 int16x8_t __s1_583 = __p1_583; \
5600556006 int16x8_t __s2_583 = __p2_583; \
56006 int16x8_t __ret_583; \
5600756007 __ret_583 = __s0_583 - __s1_583 * splatq_laneq_s16(__s2_583, __p3_583); \
5600856008 __ret_583; \
5600956009})
5601056010#else
5601156011#define vmlsq_laneq_s16(__p0_584, __p1_584, __p2_584, __p3_584) __extension__ ({ \
56012 int16x8_t __ret_584; \
5601256013 int16x8_t __s0_584 = __p0_584; \
5601356014 int16x8_t __s1_584 = __p1_584; \
5601456015 int16x8_t __s2_584 = __p2_584; \
5601556016 int16x8_t __rev0_584; __rev0_584 = __builtin_shufflevector(__s0_584, __s0_584, 7, 6, 5, 4, 3, 2, 1, 0); \
5601656017 int16x8_t __rev1_584; __rev1_584 = __builtin_shufflevector(__s1_584, __s1_584, 7, 6, 5, 4, 3, 2, 1, 0); \
5601756018 int16x8_t __rev2_584; __rev2_584 = __builtin_shufflevector(__s2_584, __s2_584, 7, 6, 5, 4, 3, 2, 1, 0); \
56018 int16x8_t __ret_584; \
5601956019 __ret_584 = __rev0_584 - __rev1_584 * __noswap_splatq_laneq_s16(__rev2_584, __p3_584); \
5602056020 __ret_584 = __builtin_shufflevector(__ret_584, __ret_584, 7, 6, 5, 4, 3, 2, 1, 0); \
5602156021 __ret_584; \
......@@ -56024,22 +56024,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5602456024
5602556025#ifdef __LITTLE_ENDIAN__
5602656026#define vmls_laneq_u32(__p0_585, __p1_585, __p2_585, __p3_585) __extension__ ({ \
56027 uint32x2_t __ret_585; \
5602756028 uint32x2_t __s0_585 = __p0_585; \
5602856029 uint32x2_t __s1_585 = __p1_585; \
5602956030 uint32x4_t __s2_585 = __p2_585; \
56030 uint32x2_t __ret_585; \
5603156031 __ret_585 = __s0_585 - __s1_585 * splat_laneq_u32(__s2_585, __p3_585); \
5603256032 __ret_585; \
5603356033})
5603456034#else
5603556035#define vmls_laneq_u32(__p0_586, __p1_586, __p2_586, __p3_586) __extension__ ({ \
56036 uint32x2_t __ret_586; \
5603656037 uint32x2_t __s0_586 = __p0_586; \
5603756038 uint32x2_t __s1_586 = __p1_586; \
5603856039 uint32x4_t __s2_586 = __p2_586; \
5603956040 uint32x2_t __rev0_586; __rev0_586 = __builtin_shufflevector(__s0_586, __s0_586, 1, 0); \
5604056041 uint32x2_t __rev1_586; __rev1_586 = __builtin_shufflevector(__s1_586, __s1_586, 1, 0); \
5604156042 uint32x4_t __rev2_586; __rev2_586 = __builtin_shufflevector(__s2_586, __s2_586, 3, 2, 1, 0); \
56042 uint32x2_t __ret_586; \
5604356043 __ret_586 = __rev0_586 - __rev1_586 * __noswap_splat_laneq_u32(__rev2_586, __p3_586); \
5604456044 __ret_586 = __builtin_shufflevector(__ret_586, __ret_586, 1, 0); \
5604556045 __ret_586; \
......@@ -56048,22 +56048,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5604856048
5604956049#ifdef __LITTLE_ENDIAN__
5605056050#define vmls_laneq_u16(__p0_587, __p1_587, __p2_587, __p3_587) __extension__ ({ \
56051 uint16x4_t __ret_587; \
5605156052 uint16x4_t __s0_587 = __p0_587; \
5605256053 uint16x4_t __s1_587 = __p1_587; \
5605356054 uint16x8_t __s2_587 = __p2_587; \
56054 uint16x4_t __ret_587; \
5605556055 __ret_587 = __s0_587 - __s1_587 * splat_laneq_u16(__s2_587, __p3_587); \
5605656056 __ret_587; \
5605756057})
5605856058#else
5605956059#define vmls_laneq_u16(__p0_588, __p1_588, __p2_588, __p3_588) __extension__ ({ \
56060 uint16x4_t __ret_588; \
5606056061 uint16x4_t __s0_588 = __p0_588; \
5606156062 uint16x4_t __s1_588 = __p1_588; \
5606256063 uint16x8_t __s2_588 = __p2_588; \
5606356064 uint16x4_t __rev0_588; __rev0_588 = __builtin_shufflevector(__s0_588, __s0_588, 3, 2, 1, 0); \
5606456065 uint16x4_t __rev1_588; __rev1_588 = __builtin_shufflevector(__s1_588, __s1_588, 3, 2, 1, 0); \
5606556066 uint16x8_t __rev2_588; __rev2_588 = __builtin_shufflevector(__s2_588, __s2_588, 7, 6, 5, 4, 3, 2, 1, 0); \
56066 uint16x4_t __ret_588; \
5606756067 __ret_588 = __rev0_588 - __rev1_588 * __noswap_splat_laneq_u16(__rev2_588, __p3_588); \
5606856068 __ret_588 = __builtin_shufflevector(__ret_588, __ret_588, 3, 2, 1, 0); \
5606956069 __ret_588; \
......@@ -56072,22 +56072,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5607256072
5607356073#ifdef __LITTLE_ENDIAN__
5607456074#define vmls_laneq_f32(__p0_589, __p1_589, __p2_589, __p3_589) __extension__ ({ \
56075 float32x2_t __ret_589; \
5607556076 float32x2_t __s0_589 = __p0_589; \
5607656077 float32x2_t __s1_589 = __p1_589; \
5607756078 float32x4_t __s2_589 = __p2_589; \
56078 float32x2_t __ret_589; \
5607956079 __ret_589 = __s0_589 - __s1_589 * splat_laneq_f32(__s2_589, __p3_589); \
5608056080 __ret_589; \
5608156081})
5608256082#else
5608356083#define vmls_laneq_f32(__p0_590, __p1_590, __p2_590, __p3_590) __extension__ ({ \
56084 float32x2_t __ret_590; \
5608456085 float32x2_t __s0_590 = __p0_590; \
5608556086 float32x2_t __s1_590 = __p1_590; \
5608656087 float32x4_t __s2_590 = __p2_590; \
5608756088 float32x2_t __rev0_590; __rev0_590 = __builtin_shufflevector(__s0_590, __s0_590, 1, 0); \
5608856089 float32x2_t __rev1_590; __rev1_590 = __builtin_shufflevector(__s1_590, __s1_590, 1, 0); \
5608956090 float32x4_t __rev2_590; __rev2_590 = __builtin_shufflevector(__s2_590, __s2_590, 3, 2, 1, 0); \
56090 float32x2_t __ret_590; \
5609156091 __ret_590 = __rev0_590 - __rev1_590 * __noswap_splat_laneq_f32(__rev2_590, __p3_590); \
5609256092 __ret_590 = __builtin_shufflevector(__ret_590, __ret_590, 1, 0); \
5609356093 __ret_590; \
......@@ -56096,22 +56096,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5609656096
5609756097#ifdef __LITTLE_ENDIAN__
5609856098#define vmls_laneq_s32(__p0_591, __p1_591, __p2_591, __p3_591) __extension__ ({ \
56099 int32x2_t __ret_591; \
5609956100 int32x2_t __s0_591 = __p0_591; \
5610056101 int32x2_t __s1_591 = __p1_591; \
5610156102 int32x4_t __s2_591 = __p2_591; \
56102 int32x2_t __ret_591; \
5610356103 __ret_591 = __s0_591 - __s1_591 * splat_laneq_s32(__s2_591, __p3_591); \
5610456104 __ret_591; \
5610556105})
5610656106#else
5610756107#define vmls_laneq_s32(__p0_592, __p1_592, __p2_592, __p3_592) __extension__ ({ \
56108 int32x2_t __ret_592; \
5610856109 int32x2_t __s0_592 = __p0_592; \
5610956110 int32x2_t __s1_592 = __p1_592; \
5611056111 int32x4_t __s2_592 = __p2_592; \
5611156112 int32x2_t __rev0_592; __rev0_592 = __builtin_shufflevector(__s0_592, __s0_592, 1, 0); \
5611256113 int32x2_t __rev1_592; __rev1_592 = __builtin_shufflevector(__s1_592, __s1_592, 1, 0); \
5611356114 int32x4_t __rev2_592; __rev2_592 = __builtin_shufflevector(__s2_592, __s2_592, 3, 2, 1, 0); \
56114 int32x2_t __ret_592; \
5611556115 __ret_592 = __rev0_592 - __rev1_592 * __noswap_splat_laneq_s32(__rev2_592, __p3_592); \
5611656116 __ret_592 = __builtin_shufflevector(__ret_592, __ret_592, 1, 0); \
5611756117 __ret_592; \
......@@ -56120,22 +56120,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5612056120
5612156121#ifdef __LITTLE_ENDIAN__
5612256122#define vmls_laneq_s16(__p0_593, __p1_593, __p2_593, __p3_593) __extension__ ({ \
56123 int16x4_t __ret_593; \
5612356124 int16x4_t __s0_593 = __p0_593; \
5612456125 int16x4_t __s1_593 = __p1_593; \
5612556126 int16x8_t __s2_593 = __p2_593; \
56126 int16x4_t __ret_593; \
5612756127 __ret_593 = __s0_593 - __s1_593 * splat_laneq_s16(__s2_593, __p3_593); \
5612856128 __ret_593; \
5612956129})
5613056130#else
5613156131#define vmls_laneq_s16(__p0_594, __p1_594, __p2_594, __p3_594) __extension__ ({ \
56132 int16x4_t __ret_594; \
5613256133 int16x4_t __s0_594 = __p0_594; \
5613356134 int16x4_t __s1_594 = __p1_594; \
5613456135 int16x8_t __s2_594 = __p2_594; \
5613556136 int16x4_t __rev0_594; __rev0_594 = __builtin_shufflevector(__s0_594, __s0_594, 3, 2, 1, 0); \
5613656137 int16x4_t __rev1_594; __rev1_594 = __builtin_shufflevector(__s1_594, __s1_594, 3, 2, 1, 0); \
5613756138 int16x8_t __rev2_594; __rev2_594 = __builtin_shufflevector(__s2_594, __s2_594, 7, 6, 5, 4, 3, 2, 1, 0); \
56138 int16x4_t __ret_594; \
5613956139 __ret_594 = __rev0_594 - __rev1_594 * __noswap_splat_laneq_s16(__rev2_594, __p3_594); \
5614056140 __ret_594 = __builtin_shufflevector(__ret_594, __ret_594, 3, 2, 1, 0); \
5614156141 __ret_594; \
......@@ -56144,22 +56144,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5614456144
5614556145#ifdef __LITTLE_ENDIAN__
5614656146#define vmlsl_high_lane_u32(__p0_595, __p1_595, __p2_595, __p3_595) __extension__ ({ \
56147 uint64x2_t __ret_595; \
5614756148 uint64x2_t __s0_595 = __p0_595; \
5614856149 uint32x4_t __s1_595 = __p1_595; \
5614956150 uint32x2_t __s2_595 = __p2_595; \
56150 uint64x2_t __ret_595; \
5615156151 __ret_595 = __s0_595 - vmull_u32(vget_high_u32(__s1_595), splat_lane_u32(__s2_595, __p3_595)); \
5615256152 __ret_595; \
5615356153})
5615456154#else
5615556155#define vmlsl_high_lane_u32(__p0_596, __p1_596, __p2_596, __p3_596) __extension__ ({ \
56156 uint64x2_t __ret_596; \
5615656157 uint64x2_t __s0_596 = __p0_596; \
5615756158 uint32x4_t __s1_596 = __p1_596; \
5615856159 uint32x2_t __s2_596 = __p2_596; \
5615956160 uint64x2_t __rev0_596; __rev0_596 = __builtin_shufflevector(__s0_596, __s0_596, 1, 0); \
5616056161 uint32x4_t __rev1_596; __rev1_596 = __builtin_shufflevector(__s1_596, __s1_596, 3, 2, 1, 0); \
5616156162 uint32x2_t __rev2_596; __rev2_596 = __builtin_shufflevector(__s2_596, __s2_596, 1, 0); \
56162 uint64x2_t __ret_596; \
5616356163 __ret_596 = __rev0_596 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_596), __noswap_splat_lane_u32(__rev2_596, __p3_596)); \
5616456164 __ret_596 = __builtin_shufflevector(__ret_596, __ret_596, 1, 0); \
5616556165 __ret_596; \
......@@ -56168,22 +56168,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5616856168
5616956169#ifdef __LITTLE_ENDIAN__
5617056170#define vmlsl_high_lane_u16(__p0_597, __p1_597, __p2_597, __p3_597) __extension__ ({ \
56171 uint32x4_t __ret_597; \
5617156172 uint32x4_t __s0_597 = __p0_597; \
5617256173 uint16x8_t __s1_597 = __p1_597; \
5617356174 uint16x4_t __s2_597 = __p2_597; \
56174 uint32x4_t __ret_597; \
5617556175 __ret_597 = __s0_597 - vmull_u16(vget_high_u16(__s1_597), splat_lane_u16(__s2_597, __p3_597)); \
5617656176 __ret_597; \
5617756177})
5617856178#else
5617956179#define vmlsl_high_lane_u16(__p0_598, __p1_598, __p2_598, __p3_598) __extension__ ({ \
56180 uint32x4_t __ret_598; \
5618056181 uint32x4_t __s0_598 = __p0_598; \
5618156182 uint16x8_t __s1_598 = __p1_598; \
5618256183 uint16x4_t __s2_598 = __p2_598; \
5618356184 uint32x4_t __rev0_598; __rev0_598 = __builtin_shufflevector(__s0_598, __s0_598, 3, 2, 1, 0); \
5618456185 uint16x8_t __rev1_598; __rev1_598 = __builtin_shufflevector(__s1_598, __s1_598, 7, 6, 5, 4, 3, 2, 1, 0); \
5618556186 uint16x4_t __rev2_598; __rev2_598 = __builtin_shufflevector(__s2_598, __s2_598, 3, 2, 1, 0); \
56186 uint32x4_t __ret_598; \
5618756187 __ret_598 = __rev0_598 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_598), __noswap_splat_lane_u16(__rev2_598, __p3_598)); \
5618856188 __ret_598 = __builtin_shufflevector(__ret_598, __ret_598, 3, 2, 1, 0); \
5618956189 __ret_598; \
......@@ -56192,22 +56192,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5619256192
5619356193#ifdef __LITTLE_ENDIAN__
5619456194#define vmlsl_high_lane_s32(__p0_599, __p1_599, __p2_599, __p3_599) __extension__ ({ \
56195 int64x2_t __ret_599; \
5619556196 int64x2_t __s0_599 = __p0_599; \
5619656197 int32x4_t __s1_599 = __p1_599; \
5619756198 int32x2_t __s2_599 = __p2_599; \
56198 int64x2_t __ret_599; \
5619956199 __ret_599 = __s0_599 - vmull_s32(vget_high_s32(__s1_599), splat_lane_s32(__s2_599, __p3_599)); \
5620056200 __ret_599; \
5620156201})
5620256202#else
5620356203#define vmlsl_high_lane_s32(__p0_600, __p1_600, __p2_600, __p3_600) __extension__ ({ \
56204 int64x2_t __ret_600; \
5620456205 int64x2_t __s0_600 = __p0_600; \
5620556206 int32x4_t __s1_600 = __p1_600; \
5620656207 int32x2_t __s2_600 = __p2_600; \
5620756208 int64x2_t __rev0_600; __rev0_600 = __builtin_shufflevector(__s0_600, __s0_600, 1, 0); \
5620856209 int32x4_t __rev1_600; __rev1_600 = __builtin_shufflevector(__s1_600, __s1_600, 3, 2, 1, 0); \
5620956210 int32x2_t __rev2_600; __rev2_600 = __builtin_shufflevector(__s2_600, __s2_600, 1, 0); \
56210 int64x2_t __ret_600; \
5621156211 __ret_600 = __rev0_600 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_600), __noswap_splat_lane_s32(__rev2_600, __p3_600)); \
5621256212 __ret_600 = __builtin_shufflevector(__ret_600, __ret_600, 1, 0); \
5621356213 __ret_600; \
......@@ -56216,22 +56216,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5621656216
5621756217#ifdef __LITTLE_ENDIAN__
5621856218#define vmlsl_high_lane_s16(__p0_601, __p1_601, __p2_601, __p3_601) __extension__ ({ \
56219 int32x4_t __ret_601; \
5621956220 int32x4_t __s0_601 = __p0_601; \
5622056221 int16x8_t __s1_601 = __p1_601; \
5622156222 int16x4_t __s2_601 = __p2_601; \
56222 int32x4_t __ret_601; \
5622356223 __ret_601 = __s0_601 - vmull_s16(vget_high_s16(__s1_601), splat_lane_s16(__s2_601, __p3_601)); \
5622456224 __ret_601; \
5622556225})
5622656226#else
5622756227#define vmlsl_high_lane_s16(__p0_602, __p1_602, __p2_602, __p3_602) __extension__ ({ \
56228 int32x4_t __ret_602; \
5622856229 int32x4_t __s0_602 = __p0_602; \
5622956230 int16x8_t __s1_602 = __p1_602; \
5623056231 int16x4_t __s2_602 = __p2_602; \
5623156232 int32x4_t __rev0_602; __rev0_602 = __builtin_shufflevector(__s0_602, __s0_602, 3, 2, 1, 0); \
5623256233 int16x8_t __rev1_602; __rev1_602 = __builtin_shufflevector(__s1_602, __s1_602, 7, 6, 5, 4, 3, 2, 1, 0); \
5623356234 int16x4_t __rev2_602; __rev2_602 = __builtin_shufflevector(__s2_602, __s2_602, 3, 2, 1, 0); \
56234 int32x4_t __ret_602; \
5623556235 __ret_602 = __rev0_602 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_602), __noswap_splat_lane_s16(__rev2_602, __p3_602)); \
5623656236 __ret_602 = __builtin_shufflevector(__ret_602, __ret_602, 3, 2, 1, 0); \
5623756237 __ret_602; \
......@@ -56240,22 +56240,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5624056240
5624156241#ifdef __LITTLE_ENDIAN__
5624256242#define vmlsl_high_laneq_u32(__p0_603, __p1_603, __p2_603, __p3_603) __extension__ ({ \
56243 uint64x2_t __ret_603; \
5624356244 uint64x2_t __s0_603 = __p0_603; \
5624456245 uint32x4_t __s1_603 = __p1_603; \
5624556246 uint32x4_t __s2_603 = __p2_603; \
56246 uint64x2_t __ret_603; \
5624756247 __ret_603 = __s0_603 - vmull_u32(vget_high_u32(__s1_603), splat_laneq_u32(__s2_603, __p3_603)); \
5624856248 __ret_603; \
5624956249})
5625056250#else
5625156251#define vmlsl_high_laneq_u32(__p0_604, __p1_604, __p2_604, __p3_604) __extension__ ({ \
56252 uint64x2_t __ret_604; \
5625256253 uint64x2_t __s0_604 = __p0_604; \
5625356254 uint32x4_t __s1_604 = __p1_604; \
5625456255 uint32x4_t __s2_604 = __p2_604; \
5625556256 uint64x2_t __rev0_604; __rev0_604 = __builtin_shufflevector(__s0_604, __s0_604, 1, 0); \
5625656257 uint32x4_t __rev1_604; __rev1_604 = __builtin_shufflevector(__s1_604, __s1_604, 3, 2, 1, 0); \
5625756258 uint32x4_t __rev2_604; __rev2_604 = __builtin_shufflevector(__s2_604, __s2_604, 3, 2, 1, 0); \
56258 uint64x2_t __ret_604; \
5625956259 __ret_604 = __rev0_604 - __noswap_vmull_u32(__noswap_vget_high_u32(__rev1_604), __noswap_splat_laneq_u32(__rev2_604, __p3_604)); \
5626056260 __ret_604 = __builtin_shufflevector(__ret_604, __ret_604, 1, 0); \
5626156261 __ret_604; \
......@@ -56264,22 +56264,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5626456264
5626556265#ifdef __LITTLE_ENDIAN__
5626656266#define vmlsl_high_laneq_u16(__p0_605, __p1_605, __p2_605, __p3_605) __extension__ ({ \
56267 uint32x4_t __ret_605; \
5626756268 uint32x4_t __s0_605 = __p0_605; \
5626856269 uint16x8_t __s1_605 = __p1_605; \
5626956270 uint16x8_t __s2_605 = __p2_605; \
56270 uint32x4_t __ret_605; \
5627156271 __ret_605 = __s0_605 - vmull_u16(vget_high_u16(__s1_605), splat_laneq_u16(__s2_605, __p3_605)); \
5627256272 __ret_605; \
5627356273})
5627456274#else
5627556275#define vmlsl_high_laneq_u16(__p0_606, __p1_606, __p2_606, __p3_606) __extension__ ({ \
56276 uint32x4_t __ret_606; \
5627656277 uint32x4_t __s0_606 = __p0_606; \
5627756278 uint16x8_t __s1_606 = __p1_606; \
5627856279 uint16x8_t __s2_606 = __p2_606; \
5627956280 uint32x4_t __rev0_606; __rev0_606 = __builtin_shufflevector(__s0_606, __s0_606, 3, 2, 1, 0); \
5628056281 uint16x8_t __rev1_606; __rev1_606 = __builtin_shufflevector(__s1_606, __s1_606, 7, 6, 5, 4, 3, 2, 1, 0); \
5628156282 uint16x8_t __rev2_606; __rev2_606 = __builtin_shufflevector(__s2_606, __s2_606, 7, 6, 5, 4, 3, 2, 1, 0); \
56282 uint32x4_t __ret_606; \
5628356283 __ret_606 = __rev0_606 - __noswap_vmull_u16(__noswap_vget_high_u16(__rev1_606), __noswap_splat_laneq_u16(__rev2_606, __p3_606)); \
5628456284 __ret_606 = __builtin_shufflevector(__ret_606, __ret_606, 3, 2, 1, 0); \
5628556285 __ret_606; \
......@@ -56288,22 +56288,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5628856288
5628956289#ifdef __LITTLE_ENDIAN__
5629056290#define vmlsl_high_laneq_s32(__p0_607, __p1_607, __p2_607, __p3_607) __extension__ ({ \
56291 int64x2_t __ret_607; \
5629156292 int64x2_t __s0_607 = __p0_607; \
5629256293 int32x4_t __s1_607 = __p1_607; \
5629356294 int32x4_t __s2_607 = __p2_607; \
56294 int64x2_t __ret_607; \
5629556295 __ret_607 = __s0_607 - vmull_s32(vget_high_s32(__s1_607), splat_laneq_s32(__s2_607, __p3_607)); \
5629656296 __ret_607; \
5629756297})
5629856298#else
5629956299#define vmlsl_high_laneq_s32(__p0_608, __p1_608, __p2_608, __p3_608) __extension__ ({ \
56300 int64x2_t __ret_608; \
5630056301 int64x2_t __s0_608 = __p0_608; \
5630156302 int32x4_t __s1_608 = __p1_608; \
5630256303 int32x4_t __s2_608 = __p2_608; \
5630356304 int64x2_t __rev0_608; __rev0_608 = __builtin_shufflevector(__s0_608, __s0_608, 1, 0); \
5630456305 int32x4_t __rev1_608; __rev1_608 = __builtin_shufflevector(__s1_608, __s1_608, 3, 2, 1, 0); \
5630556306 int32x4_t __rev2_608; __rev2_608 = __builtin_shufflevector(__s2_608, __s2_608, 3, 2, 1, 0); \
56306 int64x2_t __ret_608; \
5630756307 __ret_608 = __rev0_608 - __noswap_vmull_s32(__noswap_vget_high_s32(__rev1_608), __noswap_splat_laneq_s32(__rev2_608, __p3_608)); \
5630856308 __ret_608 = __builtin_shufflevector(__ret_608, __ret_608, 1, 0); \
5630956309 __ret_608; \
......@@ -56312,22 +56312,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5631256312
5631356313#ifdef __LITTLE_ENDIAN__
5631456314#define vmlsl_high_laneq_s16(__p0_609, __p1_609, __p2_609, __p3_609) __extension__ ({ \
56315 int32x4_t __ret_609; \
5631556316 int32x4_t __s0_609 = __p0_609; \
5631656317 int16x8_t __s1_609 = __p1_609; \
5631756318 int16x8_t __s2_609 = __p2_609; \
56318 int32x4_t __ret_609; \
5631956319 __ret_609 = __s0_609 - vmull_s16(vget_high_s16(__s1_609), splat_laneq_s16(__s2_609, __p3_609)); \
5632056320 __ret_609; \
5632156321})
5632256322#else
5632356323#define vmlsl_high_laneq_s16(__p0_610, __p1_610, __p2_610, __p3_610) __extension__ ({ \
56324 int32x4_t __ret_610; \
5632456325 int32x4_t __s0_610 = __p0_610; \
5632556326 int16x8_t __s1_610 = __p1_610; \
5632656327 int16x8_t __s2_610 = __p2_610; \
5632756328 int32x4_t __rev0_610; __rev0_610 = __builtin_shufflevector(__s0_610, __s0_610, 3, 2, 1, 0); \
5632856329 int16x8_t __rev1_610; __rev1_610 = __builtin_shufflevector(__s1_610, __s1_610, 7, 6, 5, 4, 3, 2, 1, 0); \
5632956330 int16x8_t __rev2_610; __rev2_610 = __builtin_shufflevector(__s2_610, __s2_610, 7, 6, 5, 4, 3, 2, 1, 0); \
56330 int32x4_t __ret_610; \
5633156331 __ret_610 = __rev0_610 - __noswap_vmull_s16(__noswap_vget_high_s16(__rev1_610), __noswap_splat_laneq_s16(__rev2_610, __p3_610)); \
5633256332 __ret_610 = __builtin_shufflevector(__ret_610, __ret_610, 3, 2, 1, 0); \
5633356333 __ret_610; \
......@@ -56336,22 +56336,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5633656336
5633756337#ifdef __LITTLE_ENDIAN__
5633856338#define vmlsl_laneq_u32(__p0_611, __p1_611, __p2_611, __p3_611) __extension__ ({ \
56339 uint64x2_t __ret_611; \
5633956340 uint64x2_t __s0_611 = __p0_611; \
5634056341 uint32x2_t __s1_611 = __p1_611; \
5634156342 uint32x4_t __s2_611 = __p2_611; \
56342 uint64x2_t __ret_611; \
5634356343 __ret_611 = __s0_611 - vmull_u32(__s1_611, splat_laneq_u32(__s2_611, __p3_611)); \
5634456344 __ret_611; \
5634556345})
5634656346#else
5634756347#define vmlsl_laneq_u32(__p0_612, __p1_612, __p2_612, __p3_612) __extension__ ({ \
56348 uint64x2_t __ret_612; \
5634856349 uint64x2_t __s0_612 = __p0_612; \
5634956350 uint32x2_t __s1_612 = __p1_612; \
5635056351 uint32x4_t __s2_612 = __p2_612; \
5635156352 uint64x2_t __rev0_612; __rev0_612 = __builtin_shufflevector(__s0_612, __s0_612, 1, 0); \
5635256353 uint32x2_t __rev1_612; __rev1_612 = __builtin_shufflevector(__s1_612, __s1_612, 1, 0); \
5635356354 uint32x4_t __rev2_612; __rev2_612 = __builtin_shufflevector(__s2_612, __s2_612, 3, 2, 1, 0); \
56354 uint64x2_t __ret_612; \
5635556355 __ret_612 = __rev0_612 - __noswap_vmull_u32(__rev1_612, __noswap_splat_laneq_u32(__rev2_612, __p3_612)); \
5635656356 __ret_612 = __builtin_shufflevector(__ret_612, __ret_612, 1, 0); \
5635756357 __ret_612; \
......@@ -56360,22 +56360,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5636056360
5636156361#ifdef __LITTLE_ENDIAN__
5636256362#define vmlsl_laneq_u16(__p0_613, __p1_613, __p2_613, __p3_613) __extension__ ({ \
56363 uint32x4_t __ret_613; \
5636356364 uint32x4_t __s0_613 = __p0_613; \
5636456365 uint16x4_t __s1_613 = __p1_613; \
5636556366 uint16x8_t __s2_613 = __p2_613; \
56366 uint32x4_t __ret_613; \
5636756367 __ret_613 = __s0_613 - vmull_u16(__s1_613, splat_laneq_u16(__s2_613, __p3_613)); \
5636856368 __ret_613; \
5636956369})
5637056370#else
5637156371#define vmlsl_laneq_u16(__p0_614, __p1_614, __p2_614, __p3_614) __extension__ ({ \
56372 uint32x4_t __ret_614; \
5637256373 uint32x4_t __s0_614 = __p0_614; \
5637356374 uint16x4_t __s1_614 = __p1_614; \
5637456375 uint16x8_t __s2_614 = __p2_614; \
5637556376 uint32x4_t __rev0_614; __rev0_614 = __builtin_shufflevector(__s0_614, __s0_614, 3, 2, 1, 0); \
5637656377 uint16x4_t __rev1_614; __rev1_614 = __builtin_shufflevector(__s1_614, __s1_614, 3, 2, 1, 0); \
5637756378 uint16x8_t __rev2_614; __rev2_614 = __builtin_shufflevector(__s2_614, __s2_614, 7, 6, 5, 4, 3, 2, 1, 0); \
56378 uint32x4_t __ret_614; \
5637956379 __ret_614 = __rev0_614 - __noswap_vmull_u16(__rev1_614, __noswap_splat_laneq_u16(__rev2_614, __p3_614)); \
5638056380 __ret_614 = __builtin_shufflevector(__ret_614, __ret_614, 3, 2, 1, 0); \
5638156381 __ret_614; \
......@@ -56384,22 +56384,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5638456384
5638556385#ifdef __LITTLE_ENDIAN__
5638656386#define vmlsl_laneq_s32(__p0_615, __p1_615, __p2_615, __p3_615) __extension__ ({ \
56387 int64x2_t __ret_615; \
5638756388 int64x2_t __s0_615 = __p0_615; \
5638856389 int32x2_t __s1_615 = __p1_615; \
5638956390 int32x4_t __s2_615 = __p2_615; \
56390 int64x2_t __ret_615; \
5639156391 __ret_615 = __s0_615 - vmull_s32(__s1_615, splat_laneq_s32(__s2_615, __p3_615)); \
5639256392 __ret_615; \
5639356393})
5639456394#else
5639556395#define vmlsl_laneq_s32(__p0_616, __p1_616, __p2_616, __p3_616) __extension__ ({ \
56396 int64x2_t __ret_616; \
5639656397 int64x2_t __s0_616 = __p0_616; \
5639756398 int32x2_t __s1_616 = __p1_616; \
5639856399 int32x4_t __s2_616 = __p2_616; \
5639956400 int64x2_t __rev0_616; __rev0_616 = __builtin_shufflevector(__s0_616, __s0_616, 1, 0); \
5640056401 int32x2_t __rev1_616; __rev1_616 = __builtin_shufflevector(__s1_616, __s1_616, 1, 0); \
5640156402 int32x4_t __rev2_616; __rev2_616 = __builtin_shufflevector(__s2_616, __s2_616, 3, 2, 1, 0); \
56402 int64x2_t __ret_616; \
5640356403 __ret_616 = __rev0_616 - __noswap_vmull_s32(__rev1_616, __noswap_splat_laneq_s32(__rev2_616, __p3_616)); \
5640456404 __ret_616 = __builtin_shufflevector(__ret_616, __ret_616, 1, 0); \
5640556405 __ret_616; \
......@@ -56408,22 +56408,22 @@ __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2)
5640856408
5640956409#ifdef __LITTLE_ENDIAN__
5641056410#define vmlsl_laneq_s16(__p0_617, __p1_617, __p2_617, __p3_617) __extension__ ({ \
56411 int32x4_t __ret_617; \
5641156412 int32x4_t __s0_617 = __p0_617; \
5641256413 int16x4_t __s1_617 = __p1_617; \
5641356414 int16x8_t __s2_617 = __p2_617; \
56414 int32x4_t __ret_617; \
5641556415 __ret_617 = __s0_617 - vmull_s16(__s1_617, splat_laneq_s16(__s2_617, __p3_617)); \
5641656416 __ret_617; \
5641756417})
5641856418#else
5641956419#define vmlsl_laneq_s16(__p0_618, __p1_618, __p2_618, __p3_618) __extension__ ({ \
56420 int32x4_t __ret_618; \
5642056421 int32x4_t __s0_618 = __p0_618; \
5642156422 int16x4_t __s1_618 = __p1_618; \
5642256423 int16x8_t __s2_618 = __p2_618; \
5642356424 int32x4_t __rev0_618; __rev0_618 = __builtin_shufflevector(__s0_618, __s0_618, 3, 2, 1, 0); \
5642456425 int16x4_t __rev1_618; __rev1_618 = __builtin_shufflevector(__s1_618, __s1_618, 3, 2, 1, 0); \
5642556426 int16x8_t __rev2_618; __rev2_618 = __builtin_shufflevector(__s2_618, __s2_618, 7, 6, 5, 4, 3, 2, 1, 0); \
56426 int32x4_t __ret_618; \
5642756427 __ret_618 = __rev0_618 - __noswap_vmull_s16(__rev1_618, __noswap_splat_laneq_s16(__rev2_618, __p3_618)); \
5642856428 __ret_618 = __builtin_shufflevector(__ret_618, __ret_618, 3, 2, 1, 0); \
5642956429 __ret_618; \
......@@ -56479,8 +56479,8 @@ __ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_619) {
5647956479}
5648056480#else
5648156481__ai uint16x8_t vmovl_high_u8(uint8x16_t __p0_620) {
56482 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);
5648356482 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);
5648456484 uint8x8_t __a1_620 = __noswap_vget_high_u8(__rev0_620);
5648556485 __ret_620 = (uint16x8_t)(__noswap_vshll_n_u8(__a1_620, 0));
5648656486 __ret_620 = __builtin_shufflevector(__ret_620, __ret_620, 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -56503,8 +56503,8 @@ __ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_622) {
5650356503}
5650456504#else
5650556505__ai uint64x2_t vmovl_high_u32(uint32x4_t __p0_623) {
56506 uint32x4_t __rev0_623; __rev0_623 = __builtin_shufflevector(__p0_623, __p0_623, 3, 2, 1, 0);
5650756506 uint64x2_t __ret_623;
56507 uint32x4_t __rev0_623; __rev0_623 = __builtin_shufflevector(__p0_623, __p0_623, 3, 2, 1, 0);
5650856508 uint32x2_t __a1_623 = __noswap_vget_high_u32(__rev0_623);
5650956509 __ret_623 = (uint64x2_t)(__noswap_vshll_n_u32(__a1_623, 0));
5651056510 __ret_623 = __builtin_shufflevector(__ret_623, __ret_623, 1, 0);
......@@ -56527,8 +56527,8 @@ __ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_625) {
5652756527}
5652856528#else
5652956529__ai uint32x4_t vmovl_high_u16(uint16x8_t __p0_626) {
56530 uint16x8_t __rev0_626; __rev0_626 = __builtin_shufflevector(__p0_626, __p0_626, 7, 6, 5, 4, 3, 2, 1, 0);
5653156530 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);
5653256532 uint16x4_t __a1_626 = __noswap_vget_high_u16(__rev0_626);
5653356533 __ret_626 = (uint32x4_t)(__noswap_vshll_n_u16(__a1_626, 0));
5653456534 __ret_626 = __builtin_shufflevector(__ret_626, __ret_626, 3, 2, 1, 0);
......@@ -56551,8 +56551,8 @@ __ai int16x8_t vmovl_high_s8(int8x16_t __p0_628) {
5655156551}
5655256552#else
5655356553__ai int16x8_t vmovl_high_s8(int8x16_t __p0_629) {
56554 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);
5655556554 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);
5655656556 int8x8_t __a1_629 = __noswap_vget_high_s8(__rev0_629);
5655756557 __ret_629 = (int16x8_t)(__noswap_vshll_n_s8(__a1_629, 0));
5655856558 __ret_629 = __builtin_shufflevector(__ret_629, __ret_629, 7, 6, 5, 4, 3, 2, 1, 0);
......@@ -56575,8 +56575,8 @@ __ai int64x2_t vmovl_high_s32(int32x4_t __p0_631) {
5657556575}
5657656576#else
5657756577__ai int64x2_t vmovl_high_s32(int32x4_t __p0_632) {
56578 int32x4_t __rev0_632; __rev0_632 = __builtin_shufflevector(__p0_632, __p0_632, 3, 2, 1, 0);
5657956578 int64x2_t __ret_632;
56579 int32x4_t __rev0_632; __rev0_632 = __builtin_shufflevector(__p0_632, __p0_632, 3, 2, 1, 0);
5658056580 int32x2_t __a1_632 = __noswap_vget_high_s32(__rev0_632);
5658156581 __ret_632 = (int64x2_t)(__noswap_vshll_n_s32(__a1_632, 0));
5658256582 __ret_632 = __builtin_shufflevector(__ret_632, __ret_632, 1, 0);
......@@ -56599,8 +56599,8 @@ __ai int32x4_t vmovl_high_s16(int16x8_t __p0_634) {
5659956599}
5660056600#else
5660156601__ai int32x4_t vmovl_high_s16(int16x8_t __p0_635) {
56602 int16x8_t __rev0_635; __rev0_635 = __builtin_shufflevector(__p0_635, __p0_635, 7, 6, 5, 4, 3, 2, 1, 0);
5660356602 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);
5660456604 int16x4_t __a1_635 = __noswap_vget_high_s16(__rev0_635);
5660556605 __ret_635 = (int32x4_t)(__noswap_vshll_n_s16(__a1_635, 0));
5660656606 __ret_635 = __builtin_shufflevector(__ret_635, __ret_635, 3, 2, 1, 0);
......@@ -56622,9 +56622,9 @@ __ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
5662256622}
5662356623#else
5662456624__ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
56625 uint16x8_t __ret;
5662556626 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5662656627 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56627 uint16x8_t __ret;
5662856628 __ret = __noswap_vcombine_u16(__rev0, __noswap_vmovn_u32(__rev1));
5662956629 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5663056630 return __ret;
......@@ -56639,9 +56639,9 @@ __ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
5663956639}
5664056640#else
5664156641__ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
56642 uint32x4_t __ret;
5664256643 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5664356644 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56644 uint32x4_t __ret;
5664556645 __ret = __noswap_vcombine_u32(__rev0, __noswap_vmovn_u64(__rev1));
5664656646 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5664756647 return __ret;
......@@ -56656,9 +56656,9 @@ __ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
5665656656}
5665756657#else
5665856658__ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
56659 uint8x16_t __ret;
5665956660 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5666056661 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56661 uint8x16_t __ret;
5666256662 __ret = __noswap_vcombine_u8(__rev0, __noswap_vmovn_u16(__rev1));
5666356663 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5666456664 return __ret;
......@@ -56673,9 +56673,9 @@ __ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
5667356673}
5667456674#else
5667556675__ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
56676 int16x8_t __ret;
5667656677 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5667756678 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56678 int16x8_t __ret;
5667956679 __ret = __noswap_vcombine_s16(__rev0, __noswap_vmovn_s32(__rev1));
5668056680 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5668156681 return __ret;
......@@ -56690,9 +56690,9 @@ __ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
5669056690}
5669156691#else
5669256692__ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
56693 int32x4_t __ret;
5669356694 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5669456695 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56695 int32x4_t __ret;
5669656696 __ret = __noswap_vcombine_s32(__rev0, __noswap_vmovn_s64(__rev1));
5669756697 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5669856698 return __ret;
......@@ -56707,9 +56707,9 @@ __ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
5670756707}
5670856708#else
5670956709__ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
56710 int8x16_t __ret;
5671056711 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5671156712 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56712 int8x16_t __ret;
5671356713 __ret = __noswap_vcombine_s8(__rev0, __noswap_vmovn_s16(__rev1));
5671456714 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5671556715 return __ret;
......@@ -56724,9 +56724,9 @@ __ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
5672456724}
5672556725#else
5672656726__ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
56727 float64x2_t __ret;
5672756728 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5672856729 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56729 float64x2_t __ret;
5673056730 __ret = __rev0 * __rev1;
5673156731 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5673256732 return __ret;
......@@ -56739,52 +56739,52 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5673956739 return __ret;
5674056740}
5674156741#define vmuld_lane_f64(__p0_637, __p1_637, __p2_637) __extension__ ({ \
56742 float64_t __ret_637; \
5674256743 float64_t __s0_637 = __p0_637; \
5674356744 float64x1_t __s1_637 = __p1_637; \
56744 float64_t __ret_637; \
5674556745 __ret_637 = __s0_637 * vget_lane_f64(__s1_637, __p2_637); \
5674656746 __ret_637; \
5674756747})
5674856748#ifdef __LITTLE_ENDIAN__
5674956749#define vmuls_lane_f32(__p0_638, __p1_638, __p2_638) __extension__ ({ \
56750 float32_t __ret_638; \
5675056751 float32_t __s0_638 = __p0_638; \
5675156752 float32x2_t __s1_638 = __p1_638; \
56752 float32_t __ret_638; \
5675356753 __ret_638 = __s0_638 * vget_lane_f32(__s1_638, __p2_638); \
5675456754 __ret_638; \
5675556755})
5675656756#else
5675756757#define vmuls_lane_f32(__p0_639, __p1_639, __p2_639) __extension__ ({ \
56758 float32_t __ret_639; \
5675856759 float32_t __s0_639 = __p0_639; \
5675956760 float32x2_t __s1_639 = __p1_639; \
5676056761 float32x2_t __rev1_639; __rev1_639 = __builtin_shufflevector(__s1_639, __s1_639, 1, 0); \
56761 float32_t __ret_639; \
5676256762 __ret_639 = __s0_639 * __noswap_vget_lane_f32(__rev1_639, __p2_639); \
5676356763 __ret_639; \
5676456764})
5676556765#endif
5676656766
5676756767#define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
56768 float64x1_t __ret; \
5676856769 float64x1_t __s0 = __p0; \
5676956770 float64x1_t __s1 = __p1; \
56770 float64x1_t __ret; \
5677156771 __ret = (float64x1_t) __builtin_neon_vmul_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 10); \
5677256772 __ret; \
5677356773})
5677456774#ifdef __LITTLE_ENDIAN__
5677556775#define vmulq_lane_f64(__p0_640, __p1_640, __p2_640) __extension__ ({ \
56776 float64x2_t __ret_640; \
5677656777 float64x2_t __s0_640 = __p0_640; \
5677756778 float64x1_t __s1_640 = __p1_640; \
56778 float64x2_t __ret_640; \
5677956779 __ret_640 = __s0_640 * splatq_lane_f64(__s1_640, __p2_640); \
5678056780 __ret_640; \
5678156781})
5678256782#else
5678356783#define vmulq_lane_f64(__p0_641, __p1_641, __p2_641) __extension__ ({ \
56784 float64x2_t __ret_641; \
5678456785 float64x2_t __s0_641 = __p0_641; \
5678556786 float64x1_t __s1_641 = __p1_641; \
5678656787 float64x2_t __rev0_641; __rev0_641 = __builtin_shufflevector(__s0_641, __s0_641, 1, 0); \
56787 float64x2_t __ret_641; \
5678856788 __ret_641 = __rev0_641 * __noswap_splatq_lane_f64(__s1_641, __p2_641); \
5678956789 __ret_641 = __builtin_shufflevector(__ret_641, __ret_641, 1, 0); \
5679056790 __ret_641; \
......@@ -56793,18 +56793,18 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5679356793
5679456794#ifdef __LITTLE_ENDIAN__
5679556795#define vmuld_laneq_f64(__p0_642, __p1_642, __p2_642) __extension__ ({ \
56796 float64_t __ret_642; \
5679656797 float64_t __s0_642 = __p0_642; \
5679756798 float64x2_t __s1_642 = __p1_642; \
56798 float64_t __ret_642; \
5679956799 __ret_642 = __s0_642 * vgetq_lane_f64(__s1_642, __p2_642); \
5680056800 __ret_642; \
5680156801})
5680256802#else
5680356803#define vmuld_laneq_f64(__p0_643, __p1_643, __p2_643) __extension__ ({ \
56804 float64_t __ret_643; \
5680456805 float64_t __s0_643 = __p0_643; \
5680556806 float64x2_t __s1_643 = __p1_643; \
5680656807 float64x2_t __rev1_643; __rev1_643 = __builtin_shufflevector(__s1_643, __s1_643, 1, 0); \
56807 float64_t __ret_643; \
5680856808 __ret_643 = __s0_643 * __noswap_vgetq_lane_f64(__rev1_643, __p2_643); \
5680956809 __ret_643; \
5681056810})
......@@ -56812,18 +56812,18 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5681256812
5681356813#ifdef __LITTLE_ENDIAN__
5681456814#define vmuls_laneq_f32(__p0_644, __p1_644, __p2_644) __extension__ ({ \
56815 float32_t __ret_644; \
5681556816 float32_t __s0_644 = __p0_644; \
5681656817 float32x4_t __s1_644 = __p1_644; \
56817 float32_t __ret_644; \
5681856818 __ret_644 = __s0_644 * vgetq_lane_f32(__s1_644, __p2_644); \
5681956819 __ret_644; \
5682056820})
5682156821#else
5682256822#define vmuls_laneq_f32(__p0_645, __p1_645, __p2_645) __extension__ ({ \
56823 float32_t __ret_645; \
5682356824 float32_t __s0_645 = __p0_645; \
5682456825 float32x4_t __s1_645 = __p1_645; \
5682556826 float32x4_t __rev1_645; __rev1_645 = __builtin_shufflevector(__s1_645, __s1_645, 3, 2, 1, 0); \
56826 float32_t __ret_645; \
5682756827 __ret_645 = __s0_645 * __noswap_vgetq_lane_f32(__rev1_645, __p2_645); \
5682856828 __ret_645; \
5682956829})
......@@ -56831,18 +56831,18 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5683156831
5683256832#ifdef __LITTLE_ENDIAN__
5683356833#define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
56834 float64x1_t __ret; \
5683456835 float64x1_t __s0 = __p0; \
5683556836 float64x2_t __s1 = __p1; \
56836 float64x1_t __ret; \
5683756837 __ret = (float64x1_t) __builtin_neon_vmul_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 10); \
5683856838 __ret; \
5683956839})
5684056840#else
5684156841#define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
56842 float64x1_t __ret; \
5684256843 float64x1_t __s0 = __p0; \
5684356844 float64x2_t __s1 = __p1; \
5684456845 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
56845 float64x1_t __ret; \
5684656846 __ret = (float64x1_t) __builtin_neon_vmul_laneq_v((int8x8_t)__s0, (int8x16_t)__rev1, __p2, 10); \
5684756847 __ret; \
5684856848})
......@@ -56850,19 +56850,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5685056850
5685156851#ifdef __LITTLE_ENDIAN__
5685256852#define vmulq_laneq_u32(__p0_646, __p1_646, __p2_646) __extension__ ({ \
56853 uint32x4_t __ret_646; \
5685356854 uint32x4_t __s0_646 = __p0_646; \
5685456855 uint32x4_t __s1_646 = __p1_646; \
56855 uint32x4_t __ret_646; \
5685656856 __ret_646 = __s0_646 * splatq_laneq_u32(__s1_646, __p2_646); \
5685756857 __ret_646; \
5685856858})
5685956859#else
5686056860#define vmulq_laneq_u32(__p0_647, __p1_647, __p2_647) __extension__ ({ \
56861 uint32x4_t __ret_647; \
5686156862 uint32x4_t __s0_647 = __p0_647; \
5686256863 uint32x4_t __s1_647 = __p1_647; \
5686356864 uint32x4_t __rev0_647; __rev0_647 = __builtin_shufflevector(__s0_647, __s0_647, 3, 2, 1, 0); \
5686456865 uint32x4_t __rev1_647; __rev1_647 = __builtin_shufflevector(__s1_647, __s1_647, 3, 2, 1, 0); \
56865 uint32x4_t __ret_647; \
5686656866 __ret_647 = __rev0_647 * __noswap_splatq_laneq_u32(__rev1_647, __p2_647); \
5686756867 __ret_647 = __builtin_shufflevector(__ret_647, __ret_647, 3, 2, 1, 0); \
5686856868 __ret_647; \
......@@ -56871,19 +56871,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5687156871
5687256872#ifdef __LITTLE_ENDIAN__
5687356873#define vmulq_laneq_u16(__p0_648, __p1_648, __p2_648) __extension__ ({ \
56874 uint16x8_t __ret_648; \
5687456875 uint16x8_t __s0_648 = __p0_648; \
5687556876 uint16x8_t __s1_648 = __p1_648; \
56876 uint16x8_t __ret_648; \
5687756877 __ret_648 = __s0_648 * splatq_laneq_u16(__s1_648, __p2_648); \
5687856878 __ret_648; \
5687956879})
5688056880#else
5688156881#define vmulq_laneq_u16(__p0_649, __p1_649, __p2_649) __extension__ ({ \
56882 uint16x8_t __ret_649; \
5688256883 uint16x8_t __s0_649 = __p0_649; \
5688356884 uint16x8_t __s1_649 = __p1_649; \
5688456885 uint16x8_t __rev0_649; __rev0_649 = __builtin_shufflevector(__s0_649, __s0_649, 7, 6, 5, 4, 3, 2, 1, 0); \
5688556886 uint16x8_t __rev1_649; __rev1_649 = __builtin_shufflevector(__s1_649, __s1_649, 7, 6, 5, 4, 3, 2, 1, 0); \
56886 uint16x8_t __ret_649; \
5688756887 __ret_649 = __rev0_649 * __noswap_splatq_laneq_u16(__rev1_649, __p2_649); \
5688856888 __ret_649 = __builtin_shufflevector(__ret_649, __ret_649, 7, 6, 5, 4, 3, 2, 1, 0); \
5688956889 __ret_649; \
......@@ -56892,19 +56892,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5689256892
5689356893#ifdef __LITTLE_ENDIAN__
5689456894#define vmulq_laneq_f64(__p0_650, __p1_650, __p2_650) __extension__ ({ \
56895 float64x2_t __ret_650; \
5689556896 float64x2_t __s0_650 = __p0_650; \
5689656897 float64x2_t __s1_650 = __p1_650; \
56897 float64x2_t __ret_650; \
5689856898 __ret_650 = __s0_650 * splatq_laneq_f64(__s1_650, __p2_650); \
5689956899 __ret_650; \
5690056900})
5690156901#else
5690256902#define vmulq_laneq_f64(__p0_651, __p1_651, __p2_651) __extension__ ({ \
56903 float64x2_t __ret_651; \
5690356904 float64x2_t __s0_651 = __p0_651; \
5690456905 float64x2_t __s1_651 = __p1_651; \
5690556906 float64x2_t __rev0_651; __rev0_651 = __builtin_shufflevector(__s0_651, __s0_651, 1, 0); \
5690656907 float64x2_t __rev1_651; __rev1_651 = __builtin_shufflevector(__s1_651, __s1_651, 1, 0); \
56907 float64x2_t __ret_651; \
5690856908 __ret_651 = __rev0_651 * __noswap_splatq_laneq_f64(__rev1_651, __p2_651); \
5690956909 __ret_651 = __builtin_shufflevector(__ret_651, __ret_651, 1, 0); \
5691056910 __ret_651; \
......@@ -56913,19 +56913,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5691356913
5691456914#ifdef __LITTLE_ENDIAN__
5691556915#define vmulq_laneq_f32(__p0_652, __p1_652, __p2_652) __extension__ ({ \
56916 float32x4_t __ret_652; \
5691656917 float32x4_t __s0_652 = __p0_652; \
5691756918 float32x4_t __s1_652 = __p1_652; \
56918 float32x4_t __ret_652; \
5691956919 __ret_652 = __s0_652 * splatq_laneq_f32(__s1_652, __p2_652); \
5692056920 __ret_652; \
5692156921})
5692256922#else
5692356923#define vmulq_laneq_f32(__p0_653, __p1_653, __p2_653) __extension__ ({ \
56924 float32x4_t __ret_653; \
5692456925 float32x4_t __s0_653 = __p0_653; \
5692556926 float32x4_t __s1_653 = __p1_653; \
5692656927 float32x4_t __rev0_653; __rev0_653 = __builtin_shufflevector(__s0_653, __s0_653, 3, 2, 1, 0); \
5692756928 float32x4_t __rev1_653; __rev1_653 = __builtin_shufflevector(__s1_653, __s1_653, 3, 2, 1, 0); \
56928 float32x4_t __ret_653; \
5692956929 __ret_653 = __rev0_653 * __noswap_splatq_laneq_f32(__rev1_653, __p2_653); \
5693056930 __ret_653 = __builtin_shufflevector(__ret_653, __ret_653, 3, 2, 1, 0); \
5693156931 __ret_653; \
......@@ -56934,19 +56934,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5693456934
5693556935#ifdef __LITTLE_ENDIAN__
5693656936#define vmulq_laneq_s32(__p0_654, __p1_654, __p2_654) __extension__ ({ \
56937 int32x4_t __ret_654; \
5693756938 int32x4_t __s0_654 = __p0_654; \
5693856939 int32x4_t __s1_654 = __p1_654; \
56939 int32x4_t __ret_654; \
5694056940 __ret_654 = __s0_654 * splatq_laneq_s32(__s1_654, __p2_654); \
5694156941 __ret_654; \
5694256942})
5694356943#else
5694456944#define vmulq_laneq_s32(__p0_655, __p1_655, __p2_655) __extension__ ({ \
56945 int32x4_t __ret_655; \
5694556946 int32x4_t __s0_655 = __p0_655; \
5694656947 int32x4_t __s1_655 = __p1_655; \
5694756948 int32x4_t __rev0_655; __rev0_655 = __builtin_shufflevector(__s0_655, __s0_655, 3, 2, 1, 0); \
5694856949 int32x4_t __rev1_655; __rev1_655 = __builtin_shufflevector(__s1_655, __s1_655, 3, 2, 1, 0); \
56949 int32x4_t __ret_655; \
5695056950 __ret_655 = __rev0_655 * __noswap_splatq_laneq_s32(__rev1_655, __p2_655); \
5695156951 __ret_655 = __builtin_shufflevector(__ret_655, __ret_655, 3, 2, 1, 0); \
5695256952 __ret_655; \
......@@ -56955,19 +56955,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5695556955
5695656956#ifdef __LITTLE_ENDIAN__
5695756957#define vmulq_laneq_s16(__p0_656, __p1_656, __p2_656) __extension__ ({ \
56958 int16x8_t __ret_656; \
5695856959 int16x8_t __s0_656 = __p0_656; \
5695956960 int16x8_t __s1_656 = __p1_656; \
56960 int16x8_t __ret_656; \
5696156961 __ret_656 = __s0_656 * splatq_laneq_s16(__s1_656, __p2_656); \
5696256962 __ret_656; \
5696356963})
5696456964#else
5696556965#define vmulq_laneq_s16(__p0_657, __p1_657, __p2_657) __extension__ ({ \
56966 int16x8_t __ret_657; \
5696656967 int16x8_t __s0_657 = __p0_657; \
5696756968 int16x8_t __s1_657 = __p1_657; \
5696856969 int16x8_t __rev0_657; __rev0_657 = __builtin_shufflevector(__s0_657, __s0_657, 7, 6, 5, 4, 3, 2, 1, 0); \
5696956970 int16x8_t __rev1_657; __rev1_657 = __builtin_shufflevector(__s1_657, __s1_657, 7, 6, 5, 4, 3, 2, 1, 0); \
56970 int16x8_t __ret_657; \
5697156971 __ret_657 = __rev0_657 * __noswap_splatq_laneq_s16(__rev1_657, __p2_657); \
5697256972 __ret_657 = __builtin_shufflevector(__ret_657, __ret_657, 7, 6, 5, 4, 3, 2, 1, 0); \
5697356973 __ret_657; \
......@@ -56976,19 +56976,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5697656976
5697756977#ifdef __LITTLE_ENDIAN__
5697856978#define vmul_laneq_u32(__p0_658, __p1_658, __p2_658) __extension__ ({ \
56979 uint32x2_t __ret_658; \
5697956980 uint32x2_t __s0_658 = __p0_658; \
5698056981 uint32x4_t __s1_658 = __p1_658; \
56981 uint32x2_t __ret_658; \
5698256982 __ret_658 = __s0_658 * splat_laneq_u32(__s1_658, __p2_658); \
5698356983 __ret_658; \
5698456984})
5698556985#else
5698656986#define vmul_laneq_u32(__p0_659, __p1_659, __p2_659) __extension__ ({ \
56987 uint32x2_t __ret_659; \
5698756988 uint32x2_t __s0_659 = __p0_659; \
5698856989 uint32x4_t __s1_659 = __p1_659; \
5698956990 uint32x2_t __rev0_659; __rev0_659 = __builtin_shufflevector(__s0_659, __s0_659, 1, 0); \
5699056991 uint32x4_t __rev1_659; __rev1_659 = __builtin_shufflevector(__s1_659, __s1_659, 3, 2, 1, 0); \
56991 uint32x2_t __ret_659; \
5699256992 __ret_659 = __rev0_659 * __noswap_splat_laneq_u32(__rev1_659, __p2_659); \
5699356993 __ret_659 = __builtin_shufflevector(__ret_659, __ret_659, 1, 0); \
5699456994 __ret_659; \
......@@ -56997,19 +56997,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5699756997
5699856998#ifdef __LITTLE_ENDIAN__
5699956999#define vmul_laneq_u16(__p0_660, __p1_660, __p2_660) __extension__ ({ \
57000 uint16x4_t __ret_660; \
5700057001 uint16x4_t __s0_660 = __p0_660; \
5700157002 uint16x8_t __s1_660 = __p1_660; \
57002 uint16x4_t __ret_660; \
5700357003 __ret_660 = __s0_660 * splat_laneq_u16(__s1_660, __p2_660); \
5700457004 __ret_660; \
5700557005})
5700657006#else
5700757007#define vmul_laneq_u16(__p0_661, __p1_661, __p2_661) __extension__ ({ \
57008 uint16x4_t __ret_661; \
5700857009 uint16x4_t __s0_661 = __p0_661; \
5700957010 uint16x8_t __s1_661 = __p1_661; \
5701057011 uint16x4_t __rev0_661; __rev0_661 = __builtin_shufflevector(__s0_661, __s0_661, 3, 2, 1, 0); \
5701157012 uint16x8_t __rev1_661; __rev1_661 = __builtin_shufflevector(__s1_661, __s1_661, 7, 6, 5, 4, 3, 2, 1, 0); \
57012 uint16x4_t __ret_661; \
5701357013 __ret_661 = __rev0_661 * __noswap_splat_laneq_u16(__rev1_661, __p2_661); \
5701457014 __ret_661 = __builtin_shufflevector(__ret_661, __ret_661, 3, 2, 1, 0); \
5701557015 __ret_661; \
......@@ -57018,19 +57018,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5701857018
5701957019#ifdef __LITTLE_ENDIAN__
5702057020#define vmul_laneq_f32(__p0_662, __p1_662, __p2_662) __extension__ ({ \
57021 float32x2_t __ret_662; \
5702157022 float32x2_t __s0_662 = __p0_662; \
5702257023 float32x4_t __s1_662 = __p1_662; \
57023 float32x2_t __ret_662; \
5702457024 __ret_662 = __s0_662 * splat_laneq_f32(__s1_662, __p2_662); \
5702557025 __ret_662; \
5702657026})
5702757027#else
5702857028#define vmul_laneq_f32(__p0_663, __p1_663, __p2_663) __extension__ ({ \
57029 float32x2_t __ret_663; \
5702957030 float32x2_t __s0_663 = __p0_663; \
5703057031 float32x4_t __s1_663 = __p1_663; \
5703157032 float32x2_t __rev0_663; __rev0_663 = __builtin_shufflevector(__s0_663, __s0_663, 1, 0); \
5703257033 float32x4_t __rev1_663; __rev1_663 = __builtin_shufflevector(__s1_663, __s1_663, 3, 2, 1, 0); \
57033 float32x2_t __ret_663; \
5703457034 __ret_663 = __rev0_663 * __noswap_splat_laneq_f32(__rev1_663, __p2_663); \
5703557035 __ret_663 = __builtin_shufflevector(__ret_663, __ret_663, 1, 0); \
5703657036 __ret_663; \
......@@ -57039,19 +57039,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5703957039
5704057040#ifdef __LITTLE_ENDIAN__
5704157041#define vmul_laneq_s32(__p0_664, __p1_664, __p2_664) __extension__ ({ \
57042 int32x2_t __ret_664; \
5704257043 int32x2_t __s0_664 = __p0_664; \
5704357044 int32x4_t __s1_664 = __p1_664; \
57044 int32x2_t __ret_664; \
5704557045 __ret_664 = __s0_664 * splat_laneq_s32(__s1_664, __p2_664); \
5704657046 __ret_664; \
5704757047})
5704857048#else
5704957049#define vmul_laneq_s32(__p0_665, __p1_665, __p2_665) __extension__ ({ \
57050 int32x2_t __ret_665; \
5705057051 int32x2_t __s0_665 = __p0_665; \
5705157052 int32x4_t __s1_665 = __p1_665; \
5705257053 int32x2_t __rev0_665; __rev0_665 = __builtin_shufflevector(__s0_665, __s0_665, 1, 0); \
5705357054 int32x4_t __rev1_665; __rev1_665 = __builtin_shufflevector(__s1_665, __s1_665, 3, 2, 1, 0); \
57054 int32x2_t __ret_665; \
5705557055 __ret_665 = __rev0_665 * __noswap_splat_laneq_s32(__rev1_665, __p2_665); \
5705657056 __ret_665 = __builtin_shufflevector(__ret_665, __ret_665, 1, 0); \
5705757057 __ret_665; \
......@@ -57060,19 +57060,19 @@ __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
5706057060
5706157061#ifdef __LITTLE_ENDIAN__
5706257062#define vmul_laneq_s16(__p0_666, __p1_666, __p2_666) __extension__ ({ \
57063 int16x4_t __ret_666; \
5706357064 int16x4_t __s0_666 = __p0_666; \
5706457065 int16x8_t __s1_666 = __p1_666; \
57065 int16x4_t __ret_666; \
5706657066 __ret_666 = __s0_666 * splat_laneq_s16(__s1_666, __p2_666); \
5706757067 __ret_666; \
5706857068})
5706957069#else
5707057070#define vmul_laneq_s16(__p0_667, __p1_667, __p2_667) __extension__ ({ \
57071 int16x4_t __ret_667; \
5707157072 int16x4_t __s0_667 = __p0_667; \
5707257073 int16x8_t __s1_667 = __p1_667; \
5707357074 int16x4_t __rev0_667; __rev0_667 = __builtin_shufflevector(__s0_667, __s0_667, 3, 2, 1, 0); \
5707457075 int16x8_t __rev1_667; __rev1_667 = __builtin_shufflevector(__s1_667, __s1_667, 7, 6, 5, 4, 3, 2, 1, 0); \
57075 int16x4_t __ret_667; \
5707657076 __ret_667 = __rev0_667 * __noswap_splat_laneq_s16(__rev1_667, __p2_667); \
5707757077 __ret_667 = __builtin_shufflevector(__ret_667, __ret_667, 3, 2, 1, 0); \
5707857078 __ret_667; \
......@@ -57092,8 +57092,8 @@ __ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
5709257092}
5709357093#else
5709457094__ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
57095 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5709657095 float64x2_t __ret;
57096 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5709757097 __ret = __rev0 * (float64x2_t) {__p1, __p1};
5709857098 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5709957099 return __ret;
......@@ -57113,9 +57113,9 @@ __ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
5711357113}
5711457114#else
5711557115__ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
57116 poly16x8_t __ret;
5711657117 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5711757118 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57118 poly16x8_t __ret;
5711957119 __ret = __noswap_vmull_p8(__noswap_vget_high_p8(__rev0), __noswap_vget_high_p8(__rev1));
5712057120 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5712157121 return __ret;
......@@ -57130,9 +57130,9 @@ __ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
5713057130}
5713157131#else
5713257132__ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
57133 uint16x8_t __ret;
5713357134 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5713457135 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57135 uint16x8_t __ret;
5713657136 __ret = __noswap_vmull_u8(__noswap_vget_high_u8(__rev0), __noswap_vget_high_u8(__rev1));
5713757137 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5713857138 return __ret;
......@@ -57147,9 +57147,9 @@ __ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
5714757147}
5714857148#else
5714957149__ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
57150 uint64x2_t __ret;
5715057151 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5715157152 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57152 uint64x2_t __ret;
5715357153 __ret = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0), __noswap_vget_high_u32(__rev1));
5715457154 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5715557155 return __ret;
......@@ -57164,9 +57164,9 @@ __ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
5716457164}
5716557165#else
5716657166__ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
57167 uint32x4_t __ret;
5716757168 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5716857169 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57169 uint32x4_t __ret;
5717057170 __ret = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0), __noswap_vget_high_u16(__rev1));
5717157171 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5717257172 return __ret;
......@@ -57181,9 +57181,9 @@ __ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
5718157181}
5718257182#else
5718357183__ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
57184 int16x8_t __ret;
5718457185 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5718557186 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57186 int16x8_t __ret;
5718757187 __ret = __noswap_vmull_s8(__noswap_vget_high_s8(__rev0), __noswap_vget_high_s8(__rev1));
5718857188 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5718957189 return __ret;
......@@ -57198,9 +57198,9 @@ __ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
5719857198}
5719957199#else
5720057200__ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
57201 int64x2_t __ret;
5720157202 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5720257203 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57203 int64x2_t __ret;
5720457204 __ret = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
5720557205 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5720657206 return __ret;
......@@ -57215,9 +57215,9 @@ __ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
5721557215}
5721657216#else
5721757217__ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
57218 int32x4_t __ret;
5721857219 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5721957220 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57220 int32x4_t __ret;
5722157221 __ret = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
5722257222 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5722357223 return __ret;
......@@ -57232,9 +57232,9 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5723257232}
5723357233#else
5723457234__ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
57235 poly128_t __ret;
5723557236 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5723657237 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57237 poly128_t __ret;
5723857238 __ret = vmull_p64((poly64_t)(__noswap_vget_high_p64(__rev0)), (poly64_t)(__noswap_vget_high_p64(__rev1)));
5723957239 return __ret;
5724057240}
......@@ -57242,19 +57242,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5724257242
5724357243#ifdef __LITTLE_ENDIAN__
5724457244#define vmull_high_lane_u32(__p0_668, __p1_668, __p2_668) __extension__ ({ \
57245 uint64x2_t __ret_668; \
5724557246 uint32x4_t __s0_668 = __p0_668; \
5724657247 uint32x2_t __s1_668 = __p1_668; \
57247 uint64x2_t __ret_668; \
5724857248 __ret_668 = vmull_u32(vget_high_u32(__s0_668), splat_lane_u32(__s1_668, __p2_668)); \
5724957249 __ret_668; \
5725057250})
5725157251#else
5725257252#define vmull_high_lane_u32(__p0_669, __p1_669, __p2_669) __extension__ ({ \
57253 uint64x2_t __ret_669; \
5725357254 uint32x4_t __s0_669 = __p0_669; \
5725457255 uint32x2_t __s1_669 = __p1_669; \
5725557256 uint32x4_t __rev0_669; __rev0_669 = __builtin_shufflevector(__s0_669, __s0_669, 3, 2, 1, 0); \
5725657257 uint32x2_t __rev1_669; __rev1_669 = __builtin_shufflevector(__s1_669, __s1_669, 1, 0); \
57257 uint64x2_t __ret_669; \
5725857258 __ret_669 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_669), __noswap_splat_lane_u32(__rev1_669, __p2_669)); \
5725957259 __ret_669 = __builtin_shufflevector(__ret_669, __ret_669, 1, 0); \
5726057260 __ret_669; \
......@@ -57263,19 +57263,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5726357263
5726457264#ifdef __LITTLE_ENDIAN__
5726557265#define vmull_high_lane_u16(__p0_670, __p1_670, __p2_670) __extension__ ({ \
57266 uint32x4_t __ret_670; \
5726657267 uint16x8_t __s0_670 = __p0_670; \
5726757268 uint16x4_t __s1_670 = __p1_670; \
57268 uint32x4_t __ret_670; \
5726957269 __ret_670 = vmull_u16(vget_high_u16(__s0_670), splat_lane_u16(__s1_670, __p2_670)); \
5727057270 __ret_670; \
5727157271})
5727257272#else
5727357273#define vmull_high_lane_u16(__p0_671, __p1_671, __p2_671) __extension__ ({ \
57274 uint32x4_t __ret_671; \
5727457275 uint16x8_t __s0_671 = __p0_671; \
5727557276 uint16x4_t __s1_671 = __p1_671; \
5727657277 uint16x8_t __rev0_671; __rev0_671 = __builtin_shufflevector(__s0_671, __s0_671, 7, 6, 5, 4, 3, 2, 1, 0); \
5727757278 uint16x4_t __rev1_671; __rev1_671 = __builtin_shufflevector(__s1_671, __s1_671, 3, 2, 1, 0); \
57278 uint32x4_t __ret_671; \
5727957279 __ret_671 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_671), __noswap_splat_lane_u16(__rev1_671, __p2_671)); \
5728057280 __ret_671 = __builtin_shufflevector(__ret_671, __ret_671, 3, 2, 1, 0); \
5728157281 __ret_671; \
......@@ -57284,19 +57284,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5728457284
5728557285#ifdef __LITTLE_ENDIAN__
5728657286#define vmull_high_lane_s32(__p0_672, __p1_672, __p2_672) __extension__ ({ \
57287 int64x2_t __ret_672; \
5728757288 int32x4_t __s0_672 = __p0_672; \
5728857289 int32x2_t __s1_672 = __p1_672; \
57289 int64x2_t __ret_672; \
5729057290 __ret_672 = vmull_s32(vget_high_s32(__s0_672), splat_lane_s32(__s1_672, __p2_672)); \
5729157291 __ret_672; \
5729257292})
5729357293#else
5729457294#define vmull_high_lane_s32(__p0_673, __p1_673, __p2_673) __extension__ ({ \
57295 int64x2_t __ret_673; \
5729557296 int32x4_t __s0_673 = __p0_673; \
5729657297 int32x2_t __s1_673 = __p1_673; \
5729757298 int32x4_t __rev0_673; __rev0_673 = __builtin_shufflevector(__s0_673, __s0_673, 3, 2, 1, 0); \
5729857299 int32x2_t __rev1_673; __rev1_673 = __builtin_shufflevector(__s1_673, __s1_673, 1, 0); \
57299 int64x2_t __ret_673; \
5730057300 __ret_673 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_673), __noswap_splat_lane_s32(__rev1_673, __p2_673)); \
5730157301 __ret_673 = __builtin_shufflevector(__ret_673, __ret_673, 1, 0); \
5730257302 __ret_673; \
......@@ -57305,19 +57305,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5730557305
5730657306#ifdef __LITTLE_ENDIAN__
5730757307#define vmull_high_lane_s16(__p0_674, __p1_674, __p2_674) __extension__ ({ \
57308 int32x4_t __ret_674; \
5730857309 int16x8_t __s0_674 = __p0_674; \
5730957310 int16x4_t __s1_674 = __p1_674; \
57310 int32x4_t __ret_674; \
5731157311 __ret_674 = vmull_s16(vget_high_s16(__s0_674), splat_lane_s16(__s1_674, __p2_674)); \
5731257312 __ret_674; \
5731357313})
5731457314#else
5731557315#define vmull_high_lane_s16(__p0_675, __p1_675, __p2_675) __extension__ ({ \
57316 int32x4_t __ret_675; \
5731657317 int16x8_t __s0_675 = __p0_675; \
5731757318 int16x4_t __s1_675 = __p1_675; \
5731857319 int16x8_t __rev0_675; __rev0_675 = __builtin_shufflevector(__s0_675, __s0_675, 7, 6, 5, 4, 3, 2, 1, 0); \
5731957320 int16x4_t __rev1_675; __rev1_675 = __builtin_shufflevector(__s1_675, __s1_675, 3, 2, 1, 0); \
57320 int32x4_t __ret_675; \
5732157321 __ret_675 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_675), __noswap_splat_lane_s16(__rev1_675, __p2_675)); \
5732257322 __ret_675 = __builtin_shufflevector(__ret_675, __ret_675, 3, 2, 1, 0); \
5732357323 __ret_675; \
......@@ -57326,19 +57326,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5732657326
5732757327#ifdef __LITTLE_ENDIAN__
5732857328#define vmull_high_laneq_u32(__p0_676, __p1_676, __p2_676) __extension__ ({ \
57329 uint64x2_t __ret_676; \
5732957330 uint32x4_t __s0_676 = __p0_676; \
5733057331 uint32x4_t __s1_676 = __p1_676; \
57331 uint64x2_t __ret_676; \
5733257332 __ret_676 = vmull_u32(vget_high_u32(__s0_676), splat_laneq_u32(__s1_676, __p2_676)); \
5733357333 __ret_676; \
5733457334})
5733557335#else
5733657336#define vmull_high_laneq_u32(__p0_677, __p1_677, __p2_677) __extension__ ({ \
57337 uint64x2_t __ret_677; \
5733757338 uint32x4_t __s0_677 = __p0_677; \
5733857339 uint32x4_t __s1_677 = __p1_677; \
5733957340 uint32x4_t __rev0_677; __rev0_677 = __builtin_shufflevector(__s0_677, __s0_677, 3, 2, 1, 0); \
5734057341 uint32x4_t __rev1_677; __rev1_677 = __builtin_shufflevector(__s1_677, __s1_677, 3, 2, 1, 0); \
57341 uint64x2_t __ret_677; \
5734257342 __ret_677 = __noswap_vmull_u32(__noswap_vget_high_u32(__rev0_677), __noswap_splat_laneq_u32(__rev1_677, __p2_677)); \
5734357343 __ret_677 = __builtin_shufflevector(__ret_677, __ret_677, 1, 0); \
5734457344 __ret_677; \
......@@ -57347,19 +57347,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5734757347
5734857348#ifdef __LITTLE_ENDIAN__
5734957349#define vmull_high_laneq_u16(__p0_678, __p1_678, __p2_678) __extension__ ({ \
57350 uint32x4_t __ret_678; \
5735057351 uint16x8_t __s0_678 = __p0_678; \
5735157352 uint16x8_t __s1_678 = __p1_678; \
57352 uint32x4_t __ret_678; \
5735357353 __ret_678 = vmull_u16(vget_high_u16(__s0_678), splat_laneq_u16(__s1_678, __p2_678)); \
5735457354 __ret_678; \
5735557355})
5735657356#else
5735757357#define vmull_high_laneq_u16(__p0_679, __p1_679, __p2_679) __extension__ ({ \
57358 uint32x4_t __ret_679; \
5735857359 uint16x8_t __s0_679 = __p0_679; \
5735957360 uint16x8_t __s1_679 = __p1_679; \
5736057361 uint16x8_t __rev0_679; __rev0_679 = __builtin_shufflevector(__s0_679, __s0_679, 7, 6, 5, 4, 3, 2, 1, 0); \
5736157362 uint16x8_t __rev1_679; __rev1_679 = __builtin_shufflevector(__s1_679, __s1_679, 7, 6, 5, 4, 3, 2, 1, 0); \
57362 uint32x4_t __ret_679; \
5736357363 __ret_679 = __noswap_vmull_u16(__noswap_vget_high_u16(__rev0_679), __noswap_splat_laneq_u16(__rev1_679, __p2_679)); \
5736457364 __ret_679 = __builtin_shufflevector(__ret_679, __ret_679, 3, 2, 1, 0); \
5736557365 __ret_679; \
......@@ -57368,19 +57368,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5736857368
5736957369#ifdef __LITTLE_ENDIAN__
5737057370#define vmull_high_laneq_s32(__p0_680, __p1_680, __p2_680) __extension__ ({ \
57371 int64x2_t __ret_680; \
5737157372 int32x4_t __s0_680 = __p0_680; \
5737257373 int32x4_t __s1_680 = __p1_680; \
57373 int64x2_t __ret_680; \
5737457374 __ret_680 = vmull_s32(vget_high_s32(__s0_680), splat_laneq_s32(__s1_680, __p2_680)); \
5737557375 __ret_680; \
5737657376})
5737757377#else
5737857378#define vmull_high_laneq_s32(__p0_681, __p1_681, __p2_681) __extension__ ({ \
57379 int64x2_t __ret_681; \
5737957380 int32x4_t __s0_681 = __p0_681; \
5738057381 int32x4_t __s1_681 = __p1_681; \
5738157382 int32x4_t __rev0_681; __rev0_681 = __builtin_shufflevector(__s0_681, __s0_681, 3, 2, 1, 0); \
5738257383 int32x4_t __rev1_681; __rev1_681 = __builtin_shufflevector(__s1_681, __s1_681, 3, 2, 1, 0); \
57383 int64x2_t __ret_681; \
5738457384 __ret_681 = __noswap_vmull_s32(__noswap_vget_high_s32(__rev0_681), __noswap_splat_laneq_s32(__rev1_681, __p2_681)); \
5738557385 __ret_681 = __builtin_shufflevector(__ret_681, __ret_681, 1, 0); \
5738657386 __ret_681; \
......@@ -57389,19 +57389,19 @@ __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
5738957389
5739057390#ifdef __LITTLE_ENDIAN__
5739157391#define vmull_high_laneq_s16(__p0_682, __p1_682, __p2_682) __extension__ ({ \
57392 int32x4_t __ret_682; \
5739257393 int16x8_t __s0_682 = __p0_682; \
5739357394 int16x8_t __s1_682 = __p1_682; \
57394 int32x4_t __ret_682; \
5739557395 __ret_682 = vmull_s16(vget_high_s16(__s0_682), splat_laneq_s16(__s1_682, __p2_682)); \
5739657396 __ret_682; \
5739757397})
5739857398#else
5739957399#define vmull_high_laneq_s16(__p0_683, __p1_683, __p2_683) __extension__ ({ \
57400 int32x4_t __ret_683; \
5740057401 int16x8_t __s0_683 = __p0_683; \
5740157402 int16x8_t __s1_683 = __p1_683; \
5740257403 int16x8_t __rev0_683; __rev0_683 = __builtin_shufflevector(__s0_683, __s0_683, 7, 6, 5, 4, 3, 2, 1, 0); \
5740357404 int16x8_t __rev1_683; __rev1_683 = __builtin_shufflevector(__s1_683, __s1_683, 7, 6, 5, 4, 3, 2, 1, 0); \
57404 int32x4_t __ret_683; \
5740557405 __ret_683 = __noswap_vmull_s16(__noswap_vget_high_s16(__rev0_683), __noswap_splat_laneq_s16(__rev1_683, __p2_683)); \
5740657406 __ret_683 = __builtin_shufflevector(__ret_683, __ret_683, 3, 2, 1, 0); \
5740757407 __ret_683; \
......@@ -57416,8 +57416,8 @@ __ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
5741657416}
5741757417#else
5741857418__ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
57419 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5742057419 uint64x2_t __ret;
57420 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5742157421 __ret = __noswap_vmull_n_u32(__noswap_vget_high_u32(__rev0), __p1);
5742257422 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5742357423 return __ret;
......@@ -57432,8 +57432,8 @@ __ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
5743257432}
5743357433#else
5743457434__ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
57435 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5743657435 uint32x4_t __ret;
57436 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5743757437 __ret = __noswap_vmull_n_u16(__noswap_vget_high_u16(__rev0), __p1);
5743857438 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5743957439 return __ret;
......@@ -57448,8 +57448,8 @@ __ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
5744857448}
5744957449#else
5745057450__ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
57451 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5745257451 int64x2_t __ret;
57452 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5745357453 __ret = __noswap_vmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
5745457454 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5745557455 return __ret;
......@@ -57464,8 +57464,8 @@ __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5746457464}
5746557465#else
5746657466__ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
57467 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5746857467 int32x4_t __ret;
57468 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5746957469 __ret = __noswap_vmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
5747057470 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5747157471 return __ret;
......@@ -57474,19 +57474,19 @@ __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5747457474
5747557475#ifdef __LITTLE_ENDIAN__
5747657476#define vmull_laneq_u32(__p0_684, __p1_684, __p2_684) __extension__ ({ \
57477 uint64x2_t __ret_684; \
5747757478 uint32x2_t __s0_684 = __p0_684; \
5747857479 uint32x4_t __s1_684 = __p1_684; \
57479 uint64x2_t __ret_684; \
5748057480 __ret_684 = vmull_u32(__s0_684, splat_laneq_u32(__s1_684, __p2_684)); \
5748157481 __ret_684; \
5748257482})
5748357483#else
5748457484#define vmull_laneq_u32(__p0_685, __p1_685, __p2_685) __extension__ ({ \
57485 uint64x2_t __ret_685; \
5748557486 uint32x2_t __s0_685 = __p0_685; \
5748657487 uint32x4_t __s1_685 = __p1_685; \
5748757488 uint32x2_t __rev0_685; __rev0_685 = __builtin_shufflevector(__s0_685, __s0_685, 1, 0); \
5748857489 uint32x4_t __rev1_685; __rev1_685 = __builtin_shufflevector(__s1_685, __s1_685, 3, 2, 1, 0); \
57489 uint64x2_t __ret_685; \
5749057490 __ret_685 = __noswap_vmull_u32(__rev0_685, __noswap_splat_laneq_u32(__rev1_685, __p2_685)); \
5749157491 __ret_685 = __builtin_shufflevector(__ret_685, __ret_685, 1, 0); \
5749257492 __ret_685; \
......@@ -57495,19 +57495,19 @@ __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5749557495
5749657496#ifdef __LITTLE_ENDIAN__
5749757497#define vmull_laneq_u16(__p0_686, __p1_686, __p2_686) __extension__ ({ \
57498 uint32x4_t __ret_686; \
5749857499 uint16x4_t __s0_686 = __p0_686; \
5749957500 uint16x8_t __s1_686 = __p1_686; \
57500 uint32x4_t __ret_686; \
5750157501 __ret_686 = vmull_u16(__s0_686, splat_laneq_u16(__s1_686, __p2_686)); \
5750257502 __ret_686; \
5750357503})
5750457504#else
5750557505#define vmull_laneq_u16(__p0_687, __p1_687, __p2_687) __extension__ ({ \
57506 uint32x4_t __ret_687; \
5750657507 uint16x4_t __s0_687 = __p0_687; \
5750757508 uint16x8_t __s1_687 = __p1_687; \
5750857509 uint16x4_t __rev0_687; __rev0_687 = __builtin_shufflevector(__s0_687, __s0_687, 3, 2, 1, 0); \
5750957510 uint16x8_t __rev1_687; __rev1_687 = __builtin_shufflevector(__s1_687, __s1_687, 7, 6, 5, 4, 3, 2, 1, 0); \
57510 uint32x4_t __ret_687; \
5751157511 __ret_687 = __noswap_vmull_u16(__rev0_687, __noswap_splat_laneq_u16(__rev1_687, __p2_687)); \
5751257512 __ret_687 = __builtin_shufflevector(__ret_687, __ret_687, 3, 2, 1, 0); \
5751357513 __ret_687; \
......@@ -57516,19 +57516,19 @@ __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5751657516
5751757517#ifdef __LITTLE_ENDIAN__
5751857518#define vmull_laneq_s32(__p0_688, __p1_688, __p2_688) __extension__ ({ \
57519 int64x2_t __ret_688; \
5751957520 int32x2_t __s0_688 = __p0_688; \
5752057521 int32x4_t __s1_688 = __p1_688; \
57521 int64x2_t __ret_688; \
5752257522 __ret_688 = vmull_s32(__s0_688, splat_laneq_s32(__s1_688, __p2_688)); \
5752357523 __ret_688; \
5752457524})
5752557525#else
5752657526#define vmull_laneq_s32(__p0_689, __p1_689, __p2_689) __extension__ ({ \
57527 int64x2_t __ret_689; \
5752757528 int32x2_t __s0_689 = __p0_689; \
5752857529 int32x4_t __s1_689 = __p1_689; \
5752957530 int32x2_t __rev0_689; __rev0_689 = __builtin_shufflevector(__s0_689, __s0_689, 1, 0); \
5753057531 int32x4_t __rev1_689; __rev1_689 = __builtin_shufflevector(__s1_689, __s1_689, 3, 2, 1, 0); \
57531 int64x2_t __ret_689; \
5753257532 __ret_689 = __noswap_vmull_s32(__rev0_689, __noswap_splat_laneq_s32(__rev1_689, __p2_689)); \
5753357533 __ret_689 = __builtin_shufflevector(__ret_689, __ret_689, 1, 0); \
5753457534 __ret_689; \
......@@ -57537,19 +57537,19 @@ __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5753757537
5753857538#ifdef __LITTLE_ENDIAN__
5753957539#define vmull_laneq_s16(__p0_690, __p1_690, __p2_690) __extension__ ({ \
57540 int32x4_t __ret_690; \
5754057541 int16x4_t __s0_690 = __p0_690; \
5754157542 int16x8_t __s1_690 = __p1_690; \
57542 int32x4_t __ret_690; \
5754357543 __ret_690 = vmull_s16(__s0_690, splat_laneq_s16(__s1_690, __p2_690)); \
5754457544 __ret_690; \
5754557545})
5754657546#else
5754757547#define vmull_laneq_s16(__p0_691, __p1_691, __p2_691) __extension__ ({ \
57548 int32x4_t __ret_691; \
5754857549 int16x4_t __s0_691 = __p0_691; \
5754957550 int16x8_t __s1_691 = __p1_691; \
5755057551 int16x4_t __rev0_691; __rev0_691 = __builtin_shufflevector(__s0_691, __s0_691, 3, 2, 1, 0); \
5755157552 int16x8_t __rev1_691; __rev1_691 = __builtin_shufflevector(__s1_691, __s1_691, 7, 6, 5, 4, 3, 2, 1, 0); \
57552 int32x4_t __ret_691; \
5755357553 __ret_691 = __noswap_vmull_s16(__rev0_691, __noswap_splat_laneq_s16(__rev1_691, __p2_691)); \
5755457554 __ret_691 = __builtin_shufflevector(__ret_691, __ret_691, 3, 2, 1, 0); \
5755557555 __ret_691; \
......@@ -57564,9 +57564,9 @@ __ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
5756457564}
5756557565#else
5756657566__ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
57567 float64x2_t __ret;
5756757568 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5756857569 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57569 float64x2_t __ret;
5757057570 __ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5757157571 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5757257572 return __ret;
......@@ -57586,9 +57586,9 @@ __ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
5758657586}
5758757587#else
5758857588__ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
57589 float32x4_t __ret;
5758957590 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5759057591 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57591 float32x4_t __ret;
5759257592 __ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
5759357593 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5759457594 return __ret;
......@@ -57613,9 +57613,9 @@ __ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
5761357613}
5761457614#else
5761557615__ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
57616 float32x2_t __ret;
5761657617 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5761757618 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57618 float32x2_t __ret;
5761957619 __ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
5762057620 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5762157621 return __ret;
......@@ -57638,26 +57638,26 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5763857638 return __ret;
5763957639}
5764057640#define vmulxd_lane_f64(__p0_692, __p1_692, __p2_692) __extension__ ({ \
57641 float64_t __ret_692; \
5764157642 float64_t __s0_692 = __p0_692; \
5764257643 float64x1_t __s1_692 = __p1_692; \
57643 float64_t __ret_692; \
5764457644 __ret_692 = vmulxd_f64(__s0_692, vget_lane_f64(__s1_692, __p2_692)); \
5764557645 __ret_692; \
5764657646})
5764757647#ifdef __LITTLE_ENDIAN__
5764857648#define vmulxs_lane_f32(__p0_693, __p1_693, __p2_693) __extension__ ({ \
57649 float32_t __ret_693; \
5764957650 float32_t __s0_693 = __p0_693; \
5765057651 float32x2_t __s1_693 = __p1_693; \
57651 float32_t __ret_693; \
5765257652 __ret_693 = vmulxs_f32(__s0_693, vget_lane_f32(__s1_693, __p2_693)); \
5765357653 __ret_693; \
5765457654})
5765557655#else
5765657656#define vmulxs_lane_f32(__p0_694, __p1_694, __p2_694) __extension__ ({ \
57657 float32_t __ret_694; \
5765757658 float32_t __s0_694 = __p0_694; \
5765857659 float32x2_t __s1_694 = __p1_694; \
5765957660 float32x2_t __rev1_694; __rev1_694 = __builtin_shufflevector(__s1_694, __s1_694, 1, 0); \
57660 float32_t __ret_694; \
5766157661 __ret_694 = vmulxs_f32(__s0_694, __noswap_vget_lane_f32(__rev1_694, __p2_694)); \
5766257662 __ret_694; \
5766357663})
......@@ -57665,18 +57665,18 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5766557665
5766657666#ifdef __LITTLE_ENDIAN__
5766757667#define vmulxq_lane_f64(__p0_695, __p1_695, __p2_695) __extension__ ({ \
57668 float64x2_t __ret_695; \
5766857669 float64x2_t __s0_695 = __p0_695; \
5766957670 float64x1_t __s1_695 = __p1_695; \
57670 float64x2_t __ret_695; \
5767157671 __ret_695 = vmulxq_f64(__s0_695, splatq_lane_f64(__s1_695, __p2_695)); \
5767257672 __ret_695; \
5767357673})
5767457674#else
5767557675#define vmulxq_lane_f64(__p0_696, __p1_696, __p2_696) __extension__ ({ \
57676 float64x2_t __ret_696; \
5767657677 float64x2_t __s0_696 = __p0_696; \
5767757678 float64x1_t __s1_696 = __p1_696; \
5767857679 float64x2_t __rev0_696; __rev0_696 = __builtin_shufflevector(__s0_696, __s0_696, 1, 0); \
57679 float64x2_t __ret_696; \
5768057680 __ret_696 = __noswap_vmulxq_f64(__rev0_696, __noswap_splatq_lane_f64(__s1_696, __p2_696)); \
5768157681 __ret_696 = __builtin_shufflevector(__ret_696, __ret_696, 1, 0); \
5768257682 __ret_696; \
......@@ -57685,19 +57685,19 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5768557685
5768657686#ifdef __LITTLE_ENDIAN__
5768757687#define vmulxq_lane_f32(__p0_697, __p1_697, __p2_697) __extension__ ({ \
57688 float32x4_t __ret_697; \
5768857689 float32x4_t __s0_697 = __p0_697; \
5768957690 float32x2_t __s1_697 = __p1_697; \
57690 float32x4_t __ret_697; \
5769157691 __ret_697 = vmulxq_f32(__s0_697, splatq_lane_f32(__s1_697, __p2_697)); \
5769257692 __ret_697; \
5769357693})
5769457694#else
5769557695#define vmulxq_lane_f32(__p0_698, __p1_698, __p2_698) __extension__ ({ \
57696 float32x4_t __ret_698; \
5769657697 float32x4_t __s0_698 = __p0_698; \
5769757698 float32x2_t __s1_698 = __p1_698; \
5769857699 float32x4_t __rev0_698; __rev0_698 = __builtin_shufflevector(__s0_698, __s0_698, 3, 2, 1, 0); \
5769957700 float32x2_t __rev1_698; __rev1_698 = __builtin_shufflevector(__s1_698, __s1_698, 1, 0); \
57700 float32x4_t __ret_698; \
5770157701 __ret_698 = __noswap_vmulxq_f32(__rev0_698, __noswap_splatq_lane_f32(__rev1_698, __p2_698)); \
5770257702 __ret_698 = __builtin_shufflevector(__ret_698, __ret_698, 3, 2, 1, 0); \
5770357703 __ret_698; \
......@@ -57706,19 +57706,19 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5770657706
5770757707#ifdef __LITTLE_ENDIAN__
5770857708#define vmulx_lane_f32(__p0_699, __p1_699, __p2_699) __extension__ ({ \
57709 float32x2_t __ret_699; \
5770957710 float32x2_t __s0_699 = __p0_699; \
5771057711 float32x2_t __s1_699 = __p1_699; \
57711 float32x2_t __ret_699; \
5771257712 __ret_699 = vmulx_f32(__s0_699, splat_lane_f32(__s1_699, __p2_699)); \
5771357713 __ret_699; \
5771457714})
5771557715#else
5771657716#define vmulx_lane_f32(__p0_700, __p1_700, __p2_700) __extension__ ({ \
57717 float32x2_t __ret_700; \
5771757718 float32x2_t __s0_700 = __p0_700; \
5771857719 float32x2_t __s1_700 = __p1_700; \
5771957720 float32x2_t __rev0_700; __rev0_700 = __builtin_shufflevector(__s0_700, __s0_700, 1, 0); \
5772057721 float32x2_t __rev1_700; __rev1_700 = __builtin_shufflevector(__s1_700, __s1_700, 1, 0); \
57721 float32x2_t __ret_700; \
5772257722 __ret_700 = __noswap_vmulx_f32(__rev0_700, __noswap_splat_lane_f32(__rev1_700, __p2_700)); \
5772357723 __ret_700 = __builtin_shufflevector(__ret_700, __ret_700, 1, 0); \
5772457724 __ret_700; \
......@@ -57727,18 +57727,18 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5772757727
5772857728#ifdef __LITTLE_ENDIAN__
5772957729#define vmulxd_laneq_f64(__p0_701, __p1_701, __p2_701) __extension__ ({ \
57730 float64_t __ret_701; \
5773057731 float64_t __s0_701 = __p0_701; \
5773157732 float64x2_t __s1_701 = __p1_701; \
57732 float64_t __ret_701; \
5773357733 __ret_701 = vmulxd_f64(__s0_701, vgetq_lane_f64(__s1_701, __p2_701)); \
5773457734 __ret_701; \
5773557735})
5773657736#else
5773757737#define vmulxd_laneq_f64(__p0_702, __p1_702, __p2_702) __extension__ ({ \
57738 float64_t __ret_702; \
5773857739 float64_t __s0_702 = __p0_702; \
5773957740 float64x2_t __s1_702 = __p1_702; \
5774057741 float64x2_t __rev1_702; __rev1_702 = __builtin_shufflevector(__s1_702, __s1_702, 1, 0); \
57741 float64_t __ret_702; \
5774257742 __ret_702 = vmulxd_f64(__s0_702, __noswap_vgetq_lane_f64(__rev1_702, __p2_702)); \
5774357743 __ret_702; \
5774457744})
......@@ -57746,18 +57746,18 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5774657746
5774757747#ifdef __LITTLE_ENDIAN__
5774857748#define vmulxs_laneq_f32(__p0_703, __p1_703, __p2_703) __extension__ ({ \
57749 float32_t __ret_703; \
5774957750 float32_t __s0_703 = __p0_703; \
5775057751 float32x4_t __s1_703 = __p1_703; \
57751 float32_t __ret_703; \
5775257752 __ret_703 = vmulxs_f32(__s0_703, vgetq_lane_f32(__s1_703, __p2_703)); \
5775357753 __ret_703; \
5775457754})
5775557755#else
5775657756#define vmulxs_laneq_f32(__p0_704, __p1_704, __p2_704) __extension__ ({ \
57757 float32_t __ret_704; \
5775757758 float32_t __s0_704 = __p0_704; \
5775857759 float32x4_t __s1_704 = __p1_704; \
5775957760 float32x4_t __rev1_704; __rev1_704 = __builtin_shufflevector(__s1_704, __s1_704, 3, 2, 1, 0); \
57760 float32_t __ret_704; \
5776157761 __ret_704 = vmulxs_f32(__s0_704, __noswap_vgetq_lane_f32(__rev1_704, __p2_704)); \
5776257762 __ret_704; \
5776357763})
......@@ -57765,19 +57765,19 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5776557765
5776657766#ifdef __LITTLE_ENDIAN__
5776757767#define vmulxq_laneq_f64(__p0_705, __p1_705, __p2_705) __extension__ ({ \
57768 float64x2_t __ret_705; \
5776857769 float64x2_t __s0_705 = __p0_705; \
5776957770 float64x2_t __s1_705 = __p1_705; \
57770 float64x2_t __ret_705; \
5777157771 __ret_705 = vmulxq_f64(__s0_705, splatq_laneq_f64(__s1_705, __p2_705)); \
5777257772 __ret_705; \
5777357773})
5777457774#else
5777557775#define vmulxq_laneq_f64(__p0_706, __p1_706, __p2_706) __extension__ ({ \
57776 float64x2_t __ret_706; \
5777657777 float64x2_t __s0_706 = __p0_706; \
5777757778 float64x2_t __s1_706 = __p1_706; \
5777857779 float64x2_t __rev0_706; __rev0_706 = __builtin_shufflevector(__s0_706, __s0_706, 1, 0); \
5777957780 float64x2_t __rev1_706; __rev1_706 = __builtin_shufflevector(__s1_706, __s1_706, 1, 0); \
57780 float64x2_t __ret_706; \
5778157781 __ret_706 = __noswap_vmulxq_f64(__rev0_706, __noswap_splatq_laneq_f64(__rev1_706, __p2_706)); \
5778257782 __ret_706 = __builtin_shufflevector(__ret_706, __ret_706, 1, 0); \
5778357783 __ret_706; \
......@@ -57786,19 +57786,19 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5778657786
5778757787#ifdef __LITTLE_ENDIAN__
5778857788#define vmulxq_laneq_f32(__p0_707, __p1_707, __p2_707) __extension__ ({ \
57789 float32x4_t __ret_707; \
5778957790 float32x4_t __s0_707 = __p0_707; \
5779057791 float32x4_t __s1_707 = __p1_707; \
57791 float32x4_t __ret_707; \
5779257792 __ret_707 = vmulxq_f32(__s0_707, splatq_laneq_f32(__s1_707, __p2_707)); \
5779357793 __ret_707; \
5779457794})
5779557795#else
5779657796#define vmulxq_laneq_f32(__p0_708, __p1_708, __p2_708) __extension__ ({ \
57797 float32x4_t __ret_708; \
5779757798 float32x4_t __s0_708 = __p0_708; \
5779857799 float32x4_t __s1_708 = __p1_708; \
5779957800 float32x4_t __rev0_708; __rev0_708 = __builtin_shufflevector(__s0_708, __s0_708, 3, 2, 1, 0); \
5780057801 float32x4_t __rev1_708; __rev1_708 = __builtin_shufflevector(__s1_708, __s1_708, 3, 2, 1, 0); \
57801 float32x4_t __ret_708; \
5780257802 __ret_708 = __noswap_vmulxq_f32(__rev0_708, __noswap_splatq_laneq_f32(__rev1_708, __p2_708)); \
5780357803 __ret_708 = __builtin_shufflevector(__ret_708, __ret_708, 3, 2, 1, 0); \
5780457804 __ret_708; \
......@@ -57807,19 +57807,19 @@ __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
5780757807
5780857808#ifdef __LITTLE_ENDIAN__
5780957809#define vmulx_laneq_f32(__p0_709, __p1_709, __p2_709) __extension__ ({ \
57810 float32x2_t __ret_709; \
5781057811 float32x2_t __s0_709 = __p0_709; \
5781157812 float32x4_t __s1_709 = __p1_709; \
57812 float32x2_t __ret_709; \
5781357813 __ret_709 = vmulx_f32(__s0_709, splat_laneq_f32(__s1_709, __p2_709)); \
5781457814 __ret_709; \
5781557815})
5781657816#else
5781757817#define vmulx_laneq_f32(__p0_710, __p1_710, __p2_710) __extension__ ({ \
57818 float32x2_t __ret_710; \
5781857819 float32x2_t __s0_710 = __p0_710; \
5781957820 float32x4_t __s1_710 = __p1_710; \
5782057821 float32x2_t __rev0_710; __rev0_710 = __builtin_shufflevector(__s0_710, __s0_710, 1, 0); \
5782157822 float32x4_t __rev1_710; __rev1_710 = __builtin_shufflevector(__s1_710, __s1_710, 3, 2, 1, 0); \
57822 float32x2_t __ret_710; \
5782357823 __ret_710 = __noswap_vmulx_f32(__rev0_710, __noswap_splat_laneq_f32(__rev1_710, __p2_710)); \
5782457824 __ret_710 = __builtin_shufflevector(__ret_710, __ret_710, 1, 0); \
5782557825 __ret_710; \
......@@ -57834,8 +57834,8 @@ __ai float64x2_t vnegq_f64(float64x2_t __p0) {
5783457834}
5783557835#else
5783657836__ai float64x2_t vnegq_f64(float64x2_t __p0) {
57837 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5783857837 float64x2_t __ret;
57838 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5783957839 __ret = -__rev0;
5784057840 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5784157841 return __ret;
......@@ -57850,8 +57850,8 @@ __ai int64x2_t vnegq_s64(int64x2_t __p0) {
5785057850}
5785157851#else
5785257852__ai int64x2_t vnegq_s64(int64x2_t __p0) {
57853 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5785457853 int64x2_t __ret;
57854 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5785557855 __ret = -__rev0;
5785657856 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5785757857 return __ret;
......@@ -57881,9 +57881,9 @@ __ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
5788157881}
5788257882#else
5788357883__ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
57884 uint8x16_t __ret;
5788457885 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5788557886 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57886 uint8x16_t __ret;
5788757887 __ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
5788857888 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5788957889 return __ret;
......@@ -57898,9 +57898,9 @@ __ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
5789857898}
5789957899#else
5790057900__ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
57901 uint32x4_t __ret;
5790157902 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5790257903 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57903 uint32x4_t __ret;
5790457904 __ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
5790557905 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5790657906 return __ret;
......@@ -57915,9 +57915,9 @@ __ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
5791557915}
5791657916#else
5791757917__ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
57918 uint64x2_t __ret;
5791857919 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5791957920 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57920 uint64x2_t __ret;
5792157921 __ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
5792257922 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5792357923 return __ret;
......@@ -57932,9 +57932,9 @@ __ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
5793257932}
5793357933#else
5793457934__ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
57935 uint16x8_t __ret;
5793557936 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5793657937 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57937 uint16x8_t __ret;
5793857938 __ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
5793957939 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5794057940 return __ret;
......@@ -57949,9 +57949,9 @@ __ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
5794957949}
5795057950#else
5795157951__ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
57952 int8x16_t __ret;
5795257953 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5795357954 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57954 int8x16_t __ret;
5795557955 __ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
5795657956 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5795757957 return __ret;
......@@ -57966,9 +57966,9 @@ __ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
5796657966}
5796757967#else
5796857968__ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
57969 float64x2_t __ret;
5796957970 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5797057971 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57971 float64x2_t __ret;
5797257972 __ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5797357973 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5797457974 return __ret;
......@@ -57983,9 +57983,9 @@ __ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
5798357983}
5798457984#else
5798557985__ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
57986 float32x4_t __ret;
5798657987 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5798757988 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57988 float32x4_t __ret;
5798957989 __ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
5799057990 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5799157991 return __ret;
......@@ -58000,9 +58000,9 @@ __ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
5800058000}
5800158001#else
5800258002__ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
58003 int32x4_t __ret;
5800358004 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5800458005 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58005 int32x4_t __ret;
5800658006 __ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
5800758007 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5800858008 return __ret;
......@@ -58017,9 +58017,9 @@ __ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
5801758017}
5801858018#else
5801958019__ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
58020 int64x2_t __ret;
5802058021 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5802158022 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58022 int64x2_t __ret;
5802358023 __ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
5802458024 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5802558025 return __ret;
......@@ -58034,9 +58034,9 @@ __ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
5803458034}
5803558035#else
5803658036__ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
58037 int16x8_t __ret;
5803758038 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5803858039 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58039 int16x8_t __ret;
5804058040 __ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
5804158041 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5804258042 return __ret;
......@@ -58051,8 +58051,8 @@ __ai uint64_t vpaddd_u64(uint64x2_t __p0) {
5805158051}
5805258052#else
5805358053__ai uint64_t vpaddd_u64(uint64x2_t __p0) {
58054 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5805558054 uint64_t __ret;
58055 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5805658056 __ret = (uint64_t) __builtin_neon_vpaddd_u64(__rev0);
5805758057 return __ret;
5805858058}
......@@ -58066,8 +58066,8 @@ __ai float64_t vpaddd_f64(float64x2_t __p0) {
5806658066}
5806758067#else
5806858068__ai float64_t vpaddd_f64(float64x2_t __p0) {
58069 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5807058069 float64_t __ret;
58070 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5807158071 __ret = (float64_t) __builtin_neon_vpaddd_f64(__rev0);
5807258072 return __ret;
5807358073}
......@@ -58081,8 +58081,8 @@ __ai int64_t vpaddd_s64(int64x2_t __p0) {
5808158081}
5808258082#else
5808358083__ai int64_t vpaddd_s64(int64x2_t __p0) {
58084 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5808558084 int64_t __ret;
58085 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5808658086 __ret = (int64_t) __builtin_neon_vpaddd_s64(__rev0);
5808758087 return __ret;
5808858088}
......@@ -58096,8 +58096,8 @@ __ai float32_t vpadds_f32(float32x2_t __p0) {
5809658096}
5809758097#else
5809858098__ai float32_t vpadds_f32(float32x2_t __p0) {
58099 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5810058099 float32_t __ret;
58100 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5810158101 __ret = (float32_t) __builtin_neon_vpadds_f32(__rev0);
5810258102 return __ret;
5810358103}
......@@ -58111,9 +58111,9 @@ __ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
5811158111}
5811258112#else
5811358113__ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
58114 uint8x16_t __ret;
5811458115 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5811558116 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58116 uint8x16_t __ret;
5811758117 __ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
5811858118 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5811958119 return __ret;
......@@ -58128,9 +58128,9 @@ __ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
5812858128}
5812958129#else
5813058130__ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
58131 uint32x4_t __ret;
5813158132 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5813258133 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58133 uint32x4_t __ret;
5813458134 __ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
5813558135 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5813658136 return __ret;
......@@ -58145,9 +58145,9 @@ __ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
5814558145}
5814658146#else
5814758147__ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
58148 uint16x8_t __ret;
5814858149 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5814958150 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58150 uint16x8_t __ret;
5815158151 __ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
5815258152 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5815358153 return __ret;
......@@ -58162,9 +58162,9 @@ __ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
5816258162}
5816358163#else
5816458164__ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
58165 int8x16_t __ret;
5816558166 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5816658167 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58167 int8x16_t __ret;
5816858168 __ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
5816958169 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5817058170 return __ret;
......@@ -58179,9 +58179,9 @@ __ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
5817958179}
5818058180#else
5818158181__ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
58182 float64x2_t __ret;
5818258183 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5818358184 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58184 float64x2_t __ret;
5818558185 __ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5818658186 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5818758187 return __ret;
......@@ -58196,9 +58196,9 @@ __ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
5819658196}
5819758197#else
5819858198__ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
58199 float32x4_t __ret;
5819958200 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5820058201 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58201 float32x4_t __ret;
5820258202 __ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
5820358203 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5820458204 return __ret;
......@@ -58213,9 +58213,9 @@ __ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
5821358213}
5821458214#else
5821558215__ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
58216 int32x4_t __ret;
5821658217 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5821758218 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58218 int32x4_t __ret;
5821958219 __ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
5822058220 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5822158221 return __ret;
......@@ -58230,9 +58230,9 @@ __ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
5823058230}
5823158231#else
5823258232__ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
58233 int16x8_t __ret;
5823358234 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5823458235 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58235 int16x8_t __ret;
5823658236 __ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
5823758237 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5823858238 return __ret;
......@@ -58247,8 +58247,8 @@ __ai float64_t vpmaxqd_f64(float64x2_t __p0) {
5824758247}
5824858248#else
5824958249__ai float64_t vpmaxqd_f64(float64x2_t __p0) {
58250 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5825158250 float64_t __ret;
58251 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5825258252 __ret = (float64_t) __builtin_neon_vpmaxqd_f64(__rev0);
5825358253 return __ret;
5825458254}
......@@ -58262,8 +58262,8 @@ __ai float32_t vpmaxs_f32(float32x2_t __p0) {
5826258262}
5826358263#else
5826458264__ai float32_t vpmaxs_f32(float32x2_t __p0) {
58265 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5826658265 float32_t __ret;
58266 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5826758267 __ret = (float32_t) __builtin_neon_vpmaxs_f32(__rev0);
5826858268 return __ret;
5826958269}
......@@ -58277,9 +58277,9 @@ __ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
5827758277}
5827858278#else
5827958279__ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
58280 float64x2_t __ret;
5828058281 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5828158282 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58282 float64x2_t __ret;
5828358283 __ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5828458284 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5828558285 return __ret;
......@@ -58294,9 +58294,9 @@ __ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
5829458294}
5829558295#else
5829658296__ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
58297 float32x4_t __ret;
5829758298 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5829858299 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58299 float32x4_t __ret;
5830058300 __ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
5830158301 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5830258302 return __ret;
......@@ -58311,9 +58311,9 @@ __ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
5831158311}
5831258312#else
5831358313__ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
58314 float32x2_t __ret;
5831458315 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5831558316 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58316 float32x2_t __ret;
5831758317 __ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
5831858318 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5831958319 return __ret;
......@@ -58328,8 +58328,8 @@ __ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
5832858328}
5832958329#else
5833058330__ai float64_t vpmaxnmqd_f64(float64x2_t __p0) {
58331 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5833258331 float64_t __ret;
58332 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5833358333 __ret = (float64_t) __builtin_neon_vpmaxnmqd_f64(__rev0);
5833458334 return __ret;
5833558335}
......@@ -58343,8 +58343,8 @@ __ai float32_t vpmaxnms_f32(float32x2_t __p0) {
5834358343}
5834458344#else
5834558345__ai float32_t vpmaxnms_f32(float32x2_t __p0) {
58346 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5834758346 float32_t __ret;
58347 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5834858348 __ret = (float32_t) __builtin_neon_vpmaxnms_f32(__rev0);
5834958349 return __ret;
5835058350}
......@@ -58358,9 +58358,9 @@ __ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
5835858358}
5835958359#else
5836058360__ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
58361 uint8x16_t __ret;
5836158362 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5836258363 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58363 uint8x16_t __ret;
5836458364 __ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
5836558365 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5836658366 return __ret;
......@@ -58375,9 +58375,9 @@ __ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
5837558375}
5837658376#else
5837758377__ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
58378 uint32x4_t __ret;
5837858379 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5837958380 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58380 uint32x4_t __ret;
5838158381 __ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
5838258382 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5838358383 return __ret;
......@@ -58392,9 +58392,9 @@ __ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
5839258392}
5839358393#else
5839458394__ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
58395 uint16x8_t __ret;
5839558396 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5839658397 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58397 uint16x8_t __ret;
5839858398 __ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
5839958399 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5840058400 return __ret;
......@@ -58409,9 +58409,9 @@ __ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
5840958409}
5841058410#else
5841158411__ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
58412 int8x16_t __ret;
5841258413 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5841358414 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58414 int8x16_t __ret;
5841558415 __ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
5841658416 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5841758417 return __ret;
......@@ -58426,9 +58426,9 @@ __ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
5842658426}
5842758427#else
5842858428__ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
58429 float64x2_t __ret;
5842958430 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5843058431 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58431 float64x2_t __ret;
5843258432 __ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5843358433 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5843458434 return __ret;
......@@ -58443,9 +58443,9 @@ __ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
5844358443}
5844458444#else
5844558445__ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
58446 float32x4_t __ret;
5844658447 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5844758448 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58448 float32x4_t __ret;
5844958449 __ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
5845058450 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5845158451 return __ret;
......@@ -58460,9 +58460,9 @@ __ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
5846058460}
5846158461#else
5846258462__ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
58463 int32x4_t __ret;
5846358464 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5846458465 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58465 int32x4_t __ret;
5846658466 __ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
5846758467 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5846858468 return __ret;
......@@ -58477,9 +58477,9 @@ __ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
5847758477}
5847858478#else
5847958479__ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
58480 int16x8_t __ret;
5848058481 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5848158482 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58482 int16x8_t __ret;
5848358483 __ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
5848458484 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5848558485 return __ret;
......@@ -58494,8 +58494,8 @@ __ai float64_t vpminqd_f64(float64x2_t __p0) {
5849458494}
5849558495#else
5849658496__ai float64_t vpminqd_f64(float64x2_t __p0) {
58497 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5849858497 float64_t __ret;
58498 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5849958499 __ret = (float64_t) __builtin_neon_vpminqd_f64(__rev0);
5850058500 return __ret;
5850158501}
......@@ -58509,8 +58509,8 @@ __ai float32_t vpmins_f32(float32x2_t __p0) {
5850958509}
5851058510#else
5851158511__ai float32_t vpmins_f32(float32x2_t __p0) {
58512 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5851358512 float32_t __ret;
58513 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5851458514 __ret = (float32_t) __builtin_neon_vpmins_f32(__rev0);
5851558515 return __ret;
5851658516}
......@@ -58524,9 +58524,9 @@ __ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
5852458524}
5852558525#else
5852658526__ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
58527 float64x2_t __ret;
5852758528 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5852858529 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58529 float64x2_t __ret;
5853058530 __ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
5853158531 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5853258532 return __ret;
......@@ -58541,9 +58541,9 @@ __ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
5854158541}
5854258542#else
5854358543__ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
58544 float32x4_t __ret;
5854458545 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5854558546 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58546 float32x4_t __ret;
5854758547 __ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 41);
5854858548 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5854958549 return __ret;
......@@ -58558,9 +58558,9 @@ __ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
5855858558}
5855958559#else
5856058560__ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
58561 float32x2_t __ret;
5856158562 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5856258563 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58563 float32x2_t __ret;
5856458564 __ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__rev0, (int8x8_t)__rev1, 9);
5856558565 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5856658566 return __ret;
......@@ -58575,8 +58575,8 @@ __ai float64_t vpminnmqd_f64(float64x2_t __p0) {
5857558575}
5857658576#else
5857758577__ai float64_t vpminnmqd_f64(float64x2_t __p0) {
58578 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5857958578 float64_t __ret;
58579 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5858058580 __ret = (float64_t) __builtin_neon_vpminnmqd_f64(__rev0);
5858158581 return __ret;
5858258582}
......@@ -58590,8 +58590,8 @@ __ai float32_t vpminnms_f32(float32x2_t __p0) {
5859058590}
5859158591#else
5859258592__ai float32_t vpminnms_f32(float32x2_t __p0) {
58593 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5859458593 float32_t __ret;
58594 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5859558595 __ret = (float32_t) __builtin_neon_vpminnms_f32(__rev0);
5859658596 return __ret;
5859758597}
......@@ -58605,8 +58605,8 @@ __ai int64x2_t vqabsq_s64(int64x2_t __p0) {
5860558605}
5860658606#else
5860758607__ai int64x2_t vqabsq_s64(int64x2_t __p0) {
58608 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5860958608 int64x2_t __ret;
58609 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5861058610 __ret = (int64x2_t) __builtin_neon_vqabsq_v((int8x16_t)__rev0, 35);
5861158611 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5861258612 return __ret;
......@@ -58696,10 +58696,10 @@ __ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2)
5869658696}
5869758697#else
5869858698__ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
58699 int64x2_t __ret;
5869958700 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5870058701 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5870158702 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
58702 int64x2_t __ret;
5870358703 __ret = __noswap_vqdmlal_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
5870458704 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5870558705 return __ret;
......@@ -58714,10 +58714,10 @@ __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5871458714}
5871558715#else
5871658716__ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
58717 int32x4_t __ret;
5871758718 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5871858719 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5871958720 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
58720 int32x4_t __ret;
5872158721 __ret = __noswap_vqdmlal_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
5872258722 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5872358723 return __ret;
......@@ -58726,22 +58726,22 @@ __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5872658726
5872758727#ifdef __LITTLE_ENDIAN__
5872858728#define vqdmlal_high_lane_s32(__p0_711, __p1_711, __p2_711, __p3_711) __extension__ ({ \
58729 int64x2_t __ret_711; \
5872958730 int64x2_t __s0_711 = __p0_711; \
5873058731 int32x4_t __s1_711 = __p1_711; \
5873158732 int32x2_t __s2_711 = __p2_711; \
58732 int64x2_t __ret_711; \
5873358733 __ret_711 = vqdmlal_s32(__s0_711, vget_high_s32(__s1_711), splat_lane_s32(__s2_711, __p3_711)); \
5873458734 __ret_711; \
5873558735})
5873658736#else
5873758737#define vqdmlal_high_lane_s32(__p0_712, __p1_712, __p2_712, __p3_712) __extension__ ({ \
58738 int64x2_t __ret_712; \
5873858739 int64x2_t __s0_712 = __p0_712; \
5873958740 int32x4_t __s1_712 = __p1_712; \
5874058741 int32x2_t __s2_712 = __p2_712; \
5874158742 int64x2_t __rev0_712; __rev0_712 = __builtin_shufflevector(__s0_712, __s0_712, 1, 0); \
5874258743 int32x4_t __rev1_712; __rev1_712 = __builtin_shufflevector(__s1_712, __s1_712, 3, 2, 1, 0); \
5874358744 int32x2_t __rev2_712; __rev2_712 = __builtin_shufflevector(__s2_712, __s2_712, 1, 0); \
58744 int64x2_t __ret_712; \
5874558745 __ret_712 = __noswap_vqdmlal_s32(__rev0_712, __noswap_vget_high_s32(__rev1_712), __noswap_splat_lane_s32(__rev2_712, __p3_712)); \
5874658746 __ret_712 = __builtin_shufflevector(__ret_712, __ret_712, 1, 0); \
5874758747 __ret_712; \
......@@ -58750,22 +58750,22 @@ __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5875058750
5875158751#ifdef __LITTLE_ENDIAN__
5875258752#define vqdmlal_high_lane_s16(__p0_713, __p1_713, __p2_713, __p3_713) __extension__ ({ \
58753 int32x4_t __ret_713; \
5875358754 int32x4_t __s0_713 = __p0_713; \
5875458755 int16x8_t __s1_713 = __p1_713; \
5875558756 int16x4_t __s2_713 = __p2_713; \
58756 int32x4_t __ret_713; \
5875758757 __ret_713 = vqdmlal_s16(__s0_713, vget_high_s16(__s1_713), splat_lane_s16(__s2_713, __p3_713)); \
5875858758 __ret_713; \
5875958759})
5876058760#else
5876158761#define vqdmlal_high_lane_s16(__p0_714, __p1_714, __p2_714, __p3_714) __extension__ ({ \
58762 int32x4_t __ret_714; \
5876258763 int32x4_t __s0_714 = __p0_714; \
5876358764 int16x8_t __s1_714 = __p1_714; \
5876458765 int16x4_t __s2_714 = __p2_714; \
5876558766 int32x4_t __rev0_714; __rev0_714 = __builtin_shufflevector(__s0_714, __s0_714, 3, 2, 1, 0); \
5876658767 int16x8_t __rev1_714; __rev1_714 = __builtin_shufflevector(__s1_714, __s1_714, 7, 6, 5, 4, 3, 2, 1, 0); \
5876758768 int16x4_t __rev2_714; __rev2_714 = __builtin_shufflevector(__s2_714, __s2_714, 3, 2, 1, 0); \
58768 int32x4_t __ret_714; \
5876958769 __ret_714 = __noswap_vqdmlal_s16(__rev0_714, __noswap_vget_high_s16(__rev1_714), __noswap_splat_lane_s16(__rev2_714, __p3_714)); \
5877058770 __ret_714 = __builtin_shufflevector(__ret_714, __ret_714, 3, 2, 1, 0); \
5877158771 __ret_714; \
......@@ -58774,22 +58774,22 @@ __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5877458774
5877558775#ifdef __LITTLE_ENDIAN__
5877658776#define vqdmlal_high_laneq_s32(__p0_715, __p1_715, __p2_715, __p3_715) __extension__ ({ \
58777 int64x2_t __ret_715; \
5877758778 int64x2_t __s0_715 = __p0_715; \
5877858779 int32x4_t __s1_715 = __p1_715; \
5877958780 int32x4_t __s2_715 = __p2_715; \
58780 int64x2_t __ret_715; \
5878158781 __ret_715 = vqdmlal_s32(__s0_715, vget_high_s32(__s1_715), splat_laneq_s32(__s2_715, __p3_715)); \
5878258782 __ret_715; \
5878358783})
5878458784#else
5878558785#define vqdmlal_high_laneq_s32(__p0_716, __p1_716, __p2_716, __p3_716) __extension__ ({ \
58786 int64x2_t __ret_716; \
5878658787 int64x2_t __s0_716 = __p0_716; \
5878758788 int32x4_t __s1_716 = __p1_716; \
5878858789 int32x4_t __s2_716 = __p2_716; \
5878958790 int64x2_t __rev0_716; __rev0_716 = __builtin_shufflevector(__s0_716, __s0_716, 1, 0); \
5879058791 int32x4_t __rev1_716; __rev1_716 = __builtin_shufflevector(__s1_716, __s1_716, 3, 2, 1, 0); \
5879158792 int32x4_t __rev2_716; __rev2_716 = __builtin_shufflevector(__s2_716, __s2_716, 3, 2, 1, 0); \
58792 int64x2_t __ret_716; \
5879358793 __ret_716 = __noswap_vqdmlal_s32(__rev0_716, __noswap_vget_high_s32(__rev1_716), __noswap_splat_laneq_s32(__rev2_716, __p3_716)); \
5879458794 __ret_716 = __builtin_shufflevector(__ret_716, __ret_716, 1, 0); \
5879558795 __ret_716; \
......@@ -58798,22 +58798,22 @@ __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5879858798
5879958799#ifdef __LITTLE_ENDIAN__
5880058800#define vqdmlal_high_laneq_s16(__p0_717, __p1_717, __p2_717, __p3_717) __extension__ ({ \
58801 int32x4_t __ret_717; \
5880158802 int32x4_t __s0_717 = __p0_717; \
5880258803 int16x8_t __s1_717 = __p1_717; \
5880358804 int16x8_t __s2_717 = __p2_717; \
58804 int32x4_t __ret_717; \
5880558805 __ret_717 = vqdmlal_s16(__s0_717, vget_high_s16(__s1_717), splat_laneq_s16(__s2_717, __p3_717)); \
5880658806 __ret_717; \
5880758807})
5880858808#else
5880958809#define vqdmlal_high_laneq_s16(__p0_718, __p1_718, __p2_718, __p3_718) __extension__ ({ \
58810 int32x4_t __ret_718; \
5881058811 int32x4_t __s0_718 = __p0_718; \
5881158812 int16x8_t __s1_718 = __p1_718; \
5881258813 int16x8_t __s2_718 = __p2_718; \
5881358814 int32x4_t __rev0_718; __rev0_718 = __builtin_shufflevector(__s0_718, __s0_718, 3, 2, 1, 0); \
5881458815 int16x8_t __rev1_718; __rev1_718 = __builtin_shufflevector(__s1_718, __s1_718, 7, 6, 5, 4, 3, 2, 1, 0); \
5881558816 int16x8_t __rev2_718; __rev2_718 = __builtin_shufflevector(__s2_718, __s2_718, 7, 6, 5, 4, 3, 2, 1, 0); \
58816 int32x4_t __ret_718; \
5881758817 __ret_718 = __noswap_vqdmlal_s16(__rev0_718, __noswap_vget_high_s16(__rev1_718), __noswap_splat_laneq_s16(__rev2_718, __p3_718)); \
5881858818 __ret_718 = __builtin_shufflevector(__ret_718, __ret_718, 3, 2, 1, 0); \
5881958819 __ret_718; \
......@@ -58828,9 +58828,9 @@ __ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2)
5882858828}
5882958829#else
5883058830__ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
58831 int64x2_t __ret;
5883158832 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5883258833 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58833 int64x2_t __ret;
5883458834 __ret = __noswap_vqdmlal_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
5883558835 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5883658836 return __ret;
......@@ -58845,9 +58845,9 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5884558845}
5884658846#else
5884758847__ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
58848 int32x4_t __ret;
5884858849 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5884958850 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58850 int32x4_t __ret;
5885158851 __ret = __noswap_vqdmlal_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
5885258852 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5885358853 return __ret;
......@@ -58856,20 +58856,20 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5885658856
5885758857#ifdef __LITTLE_ENDIAN__
5885858858#define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58859 int64_t __ret; \
5885958860 int64_t __s0 = __p0; \
5886058861 int32_t __s1 = __p1; \
5886158862 int32x2_t __s2 = __p2; \
58862 int64_t __ret; \
5886358863 __ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __s2, __p3); \
5886458864 __ret; \
5886558865})
5886658866#else
5886758867#define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58868 int64_t __ret; \
5886858869 int64_t __s0 = __p0; \
5886958870 int32_t __s1 = __p1; \
5887058871 int32x2_t __s2 = __p2; \
5887158872 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
58872 int64_t __ret; \
5887358873 __ret = (int64_t) __builtin_neon_vqdmlals_lane_s32(__s0, __s1, __rev2, __p3); \
5887458874 __ret; \
5887558875})
......@@ -58877,20 +58877,20 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5887758877
5887858878#ifdef __LITTLE_ENDIAN__
5887958879#define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58880 int32_t __ret; \
5888058881 int32_t __s0 = __p0; \
5888158882 int16_t __s1 = __p1; \
5888258883 int16x4_t __s2 = __p2; \
58883 int32_t __ret; \
5888458884 __ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __s2, __p3); \
5888558885 __ret; \
5888658886})
5888758887#else
5888858888#define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58889 int32_t __ret; \
5888958890 int32_t __s0 = __p0; \
5889058891 int16_t __s1 = __p1; \
5889158892 int16x4_t __s2 = __p2; \
5889258893 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
58893 int32_t __ret; \
5889458894 __ret = (int32_t) __builtin_neon_vqdmlalh_lane_s16(__s0, __s1, __rev2, __p3); \
5889558895 __ret; \
5889658896})
......@@ -58898,20 +58898,20 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5889858898
5889958899#ifdef __LITTLE_ENDIAN__
5890058900#define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58901 int64_t __ret; \
5890158902 int64_t __s0 = __p0; \
5890258903 int32_t __s1 = __p1; \
5890358904 int32x4_t __s2 = __p2; \
58904 int64_t __ret; \
5890558905 __ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __s2, __p3); \
5890658906 __ret; \
5890758907})
5890858908#else
5890958909#define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
58910 int64_t __ret; \
5891058911 int64_t __s0 = __p0; \
5891158912 int32_t __s1 = __p1; \
5891258913 int32x4_t __s2 = __p2; \
5891358914 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
58914 int64_t __ret; \
5891558915 __ret = (int64_t) __builtin_neon_vqdmlals_laneq_s32(__s0, __s1, __rev2, __p3); \
5891658916 __ret; \
5891758917})
......@@ -58919,20 +58919,20 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5891958919
5892058920#ifdef __LITTLE_ENDIAN__
5892158921#define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58922 int32_t __ret; \
5892258923 int32_t __s0 = __p0; \
5892358924 int16_t __s1 = __p1; \
5892458925 int16x8_t __s2 = __p2; \
58925 int32_t __ret; \
5892658926 __ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __s2, __p3); \
5892758927 __ret; \
5892858928})
5892958929#else
5893058930#define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
58931 int32_t __ret; \
5893158932 int32_t __s0 = __p0; \
5893258933 int16_t __s1 = __p1; \
5893358934 int16x8_t __s2 = __p2; \
5893458935 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
58935 int32_t __ret; \
5893658936 __ret = (int32_t) __builtin_neon_vqdmlalh_laneq_s16(__s0, __s1, __rev2, __p3); \
5893758937 __ret; \
5893858938})
......@@ -58940,22 +58940,22 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5894058940
5894158941#ifdef __LITTLE_ENDIAN__
5894258942#define vqdmlal_laneq_s32(__p0_719, __p1_719, __p2_719, __p3_719) __extension__ ({ \
58943 int64x2_t __ret_719; \
5894358944 int64x2_t __s0_719 = __p0_719; \
5894458945 int32x2_t __s1_719 = __p1_719; \
5894558946 int32x4_t __s2_719 = __p2_719; \
58946 int64x2_t __ret_719; \
5894758947 __ret_719 = vqdmlal_s32(__s0_719, __s1_719, splat_laneq_s32(__s2_719, __p3_719)); \
5894858948 __ret_719; \
5894958949})
5895058950#else
5895158951#define vqdmlal_laneq_s32(__p0_720, __p1_720, __p2_720, __p3_720) __extension__ ({ \
58952 int64x2_t __ret_720; \
5895258953 int64x2_t __s0_720 = __p0_720; \
5895358954 int32x2_t __s1_720 = __p1_720; \
5895458955 int32x4_t __s2_720 = __p2_720; \
5895558956 int64x2_t __rev0_720; __rev0_720 = __builtin_shufflevector(__s0_720, __s0_720, 1, 0); \
5895658957 int32x2_t __rev1_720; __rev1_720 = __builtin_shufflevector(__s1_720, __s1_720, 1, 0); \
5895758958 int32x4_t __rev2_720; __rev2_720 = __builtin_shufflevector(__s2_720, __s2_720, 3, 2, 1, 0); \
58958 int64x2_t __ret_720; \
5895958959 __ret_720 = __noswap_vqdmlal_s32(__rev0_720, __rev1_720, __noswap_splat_laneq_s32(__rev2_720, __p3_720)); \
5896058960 __ret_720 = __builtin_shufflevector(__ret_720, __ret_720, 1, 0); \
5896158961 __ret_720; \
......@@ -58964,22 +58964,22 @@ __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5896458964
5896558965#ifdef __LITTLE_ENDIAN__
5896658966#define vqdmlal_laneq_s16(__p0_721, __p1_721, __p2_721, __p3_721) __extension__ ({ \
58967 int32x4_t __ret_721; \
5896758968 int32x4_t __s0_721 = __p0_721; \
5896858969 int16x4_t __s1_721 = __p1_721; \
5896958970 int16x8_t __s2_721 = __p2_721; \
58970 int32x4_t __ret_721; \
5897158971 __ret_721 = vqdmlal_s16(__s0_721, __s1_721, splat_laneq_s16(__s2_721, __p3_721)); \
5897258972 __ret_721; \
5897358973})
5897458974#else
5897558975#define vqdmlal_laneq_s16(__p0_722, __p1_722, __p2_722, __p3_722) __extension__ ({ \
58976 int32x4_t __ret_722; \
5897658977 int32x4_t __s0_722 = __p0_722; \
5897758978 int16x4_t __s1_722 = __p1_722; \
5897858979 int16x8_t __s2_722 = __p2_722; \
5897958980 int32x4_t __rev0_722; __rev0_722 = __builtin_shufflevector(__s0_722, __s0_722, 3, 2, 1, 0); \
5898058981 int16x4_t __rev1_722; __rev1_722 = __builtin_shufflevector(__s1_722, __s1_722, 3, 2, 1, 0); \
5898158982 int16x8_t __rev2_722; __rev2_722 = __builtin_shufflevector(__s2_722, __s2_722, 7, 6, 5, 4, 3, 2, 1, 0); \
58982 int32x4_t __ret_722; \
5898358983 __ret_722 = __noswap_vqdmlal_s16(__rev0_722, __rev1_722, __noswap_splat_laneq_s16(__rev2_722, __p3_722)); \
5898458984 __ret_722 = __builtin_shufflevector(__ret_722, __ret_722, 3, 2, 1, 0); \
5898558985 __ret_722; \
......@@ -59004,10 +59004,10 @@ __ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2)
5900459004}
5900559005#else
5900659006__ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
59007 int64x2_t __ret;
5900759008 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5900859009 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5900959010 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
59010 int64x2_t __ret;
5901159011 __ret = __noswap_vqdmlsl_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
5901259012 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5901359013 return __ret;
......@@ -59022,10 +59022,10 @@ __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5902259022}
5902359023#else
5902459024__ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
59025 int32x4_t __ret;
5902559026 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5902659027 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5902759028 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
59028 int32x4_t __ret;
5902959029 __ret = __noswap_vqdmlsl_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
5903059030 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5903159031 return __ret;
......@@ -59034,22 +59034,22 @@ __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5903459034
5903559035#ifdef __LITTLE_ENDIAN__
5903659036#define vqdmlsl_high_lane_s32(__p0_723, __p1_723, __p2_723, __p3_723) __extension__ ({ \
59037 int64x2_t __ret_723; \
5903759038 int64x2_t __s0_723 = __p0_723; \
5903859039 int32x4_t __s1_723 = __p1_723; \
5903959040 int32x2_t __s2_723 = __p2_723; \
59040 int64x2_t __ret_723; \
5904159041 __ret_723 = vqdmlsl_s32(__s0_723, vget_high_s32(__s1_723), splat_lane_s32(__s2_723, __p3_723)); \
5904259042 __ret_723; \
5904359043})
5904459044#else
5904559045#define vqdmlsl_high_lane_s32(__p0_724, __p1_724, __p2_724, __p3_724) __extension__ ({ \
59046 int64x2_t __ret_724; \
5904659047 int64x2_t __s0_724 = __p0_724; \
5904759048 int32x4_t __s1_724 = __p1_724; \
5904859049 int32x2_t __s2_724 = __p2_724; \
5904959050 int64x2_t __rev0_724; __rev0_724 = __builtin_shufflevector(__s0_724, __s0_724, 1, 0); \
5905059051 int32x4_t __rev1_724; __rev1_724 = __builtin_shufflevector(__s1_724, __s1_724, 3, 2, 1, 0); \
5905159052 int32x2_t __rev2_724; __rev2_724 = __builtin_shufflevector(__s2_724, __s2_724, 1, 0); \
59052 int64x2_t __ret_724; \
5905359053 __ret_724 = __noswap_vqdmlsl_s32(__rev0_724, __noswap_vget_high_s32(__rev1_724), __noswap_splat_lane_s32(__rev2_724, __p3_724)); \
5905459054 __ret_724 = __builtin_shufflevector(__ret_724, __ret_724, 1, 0); \
5905559055 __ret_724; \
......@@ -59058,22 +59058,22 @@ __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5905859058
5905959059#ifdef __LITTLE_ENDIAN__
5906059060#define vqdmlsl_high_lane_s16(__p0_725, __p1_725, __p2_725, __p3_725) __extension__ ({ \
59061 int32x4_t __ret_725; \
5906159062 int32x4_t __s0_725 = __p0_725; \
5906259063 int16x8_t __s1_725 = __p1_725; \
5906359064 int16x4_t __s2_725 = __p2_725; \
59064 int32x4_t __ret_725; \
5906559065 __ret_725 = vqdmlsl_s16(__s0_725, vget_high_s16(__s1_725), splat_lane_s16(__s2_725, __p3_725)); \
5906659066 __ret_725; \
5906759067})
5906859068#else
5906959069#define vqdmlsl_high_lane_s16(__p0_726, __p1_726, __p2_726, __p3_726) __extension__ ({ \
59070 int32x4_t __ret_726; \
5907059071 int32x4_t __s0_726 = __p0_726; \
5907159072 int16x8_t __s1_726 = __p1_726; \
5907259073 int16x4_t __s2_726 = __p2_726; \
5907359074 int32x4_t __rev0_726; __rev0_726 = __builtin_shufflevector(__s0_726, __s0_726, 3, 2, 1, 0); \
5907459075 int16x8_t __rev1_726; __rev1_726 = __builtin_shufflevector(__s1_726, __s1_726, 7, 6, 5, 4, 3, 2, 1, 0); \
5907559076 int16x4_t __rev2_726; __rev2_726 = __builtin_shufflevector(__s2_726, __s2_726, 3, 2, 1, 0); \
59076 int32x4_t __ret_726; \
5907759077 __ret_726 = __noswap_vqdmlsl_s16(__rev0_726, __noswap_vget_high_s16(__rev1_726), __noswap_splat_lane_s16(__rev2_726, __p3_726)); \
5907859078 __ret_726 = __builtin_shufflevector(__ret_726, __ret_726, 3, 2, 1, 0); \
5907959079 __ret_726; \
......@@ -59082,22 +59082,22 @@ __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5908259082
5908359083#ifdef __LITTLE_ENDIAN__
5908459084#define vqdmlsl_high_laneq_s32(__p0_727, __p1_727, __p2_727, __p3_727) __extension__ ({ \
59085 int64x2_t __ret_727; \
5908559086 int64x2_t __s0_727 = __p0_727; \
5908659087 int32x4_t __s1_727 = __p1_727; \
5908759088 int32x4_t __s2_727 = __p2_727; \
59088 int64x2_t __ret_727; \
5908959089 __ret_727 = vqdmlsl_s32(__s0_727, vget_high_s32(__s1_727), splat_laneq_s32(__s2_727, __p3_727)); \
5909059090 __ret_727; \
5909159091})
5909259092#else
5909359093#define vqdmlsl_high_laneq_s32(__p0_728, __p1_728, __p2_728, __p3_728) __extension__ ({ \
59094 int64x2_t __ret_728; \
5909459095 int64x2_t __s0_728 = __p0_728; \
5909559096 int32x4_t __s1_728 = __p1_728; \
5909659097 int32x4_t __s2_728 = __p2_728; \
5909759098 int64x2_t __rev0_728; __rev0_728 = __builtin_shufflevector(__s0_728, __s0_728, 1, 0); \
5909859099 int32x4_t __rev1_728; __rev1_728 = __builtin_shufflevector(__s1_728, __s1_728, 3, 2, 1, 0); \
5909959100 int32x4_t __rev2_728; __rev2_728 = __builtin_shufflevector(__s2_728, __s2_728, 3, 2, 1, 0); \
59100 int64x2_t __ret_728; \
5910159101 __ret_728 = __noswap_vqdmlsl_s32(__rev0_728, __noswap_vget_high_s32(__rev1_728), __noswap_splat_laneq_s32(__rev2_728, __p3_728)); \
5910259102 __ret_728 = __builtin_shufflevector(__ret_728, __ret_728, 1, 0); \
5910359103 __ret_728; \
......@@ -59106,22 +59106,22 @@ __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2)
5910659106
5910759107#ifdef __LITTLE_ENDIAN__
5910859108#define vqdmlsl_high_laneq_s16(__p0_729, __p1_729, __p2_729, __p3_729) __extension__ ({ \
59109 int32x4_t __ret_729; \
5910959110 int32x4_t __s0_729 = __p0_729; \
5911059111 int16x8_t __s1_729 = __p1_729; \
5911159112 int16x8_t __s2_729 = __p2_729; \
59112 int32x4_t __ret_729; \
5911359113 __ret_729 = vqdmlsl_s16(__s0_729, vget_high_s16(__s1_729), splat_laneq_s16(__s2_729, __p3_729)); \
5911459114 __ret_729; \
5911559115})
5911659116#else
5911759117#define vqdmlsl_high_laneq_s16(__p0_730, __p1_730, __p2_730, __p3_730) __extension__ ({ \
59118 int32x4_t __ret_730; \
5911859119 int32x4_t __s0_730 = __p0_730; \
5911959120 int16x8_t __s1_730 = __p1_730; \
5912059121 int16x8_t __s2_730 = __p2_730; \
5912159122 int32x4_t __rev0_730; __rev0_730 = __builtin_shufflevector(__s0_730, __s0_730, 3, 2, 1, 0); \
5912259123 int16x8_t __rev1_730; __rev1_730 = __builtin_shufflevector(__s1_730, __s1_730, 7, 6, 5, 4, 3, 2, 1, 0); \
5912359124 int16x8_t __rev2_730; __rev2_730 = __builtin_shufflevector(__s2_730, __s2_730, 7, 6, 5, 4, 3, 2, 1, 0); \
59124 int32x4_t __ret_730; \
5912559125 __ret_730 = __noswap_vqdmlsl_s16(__rev0_730, __noswap_vget_high_s16(__rev1_730), __noswap_splat_laneq_s16(__rev2_730, __p3_730)); \
5912659126 __ret_730 = __builtin_shufflevector(__ret_730, __ret_730, 3, 2, 1, 0); \
5912759127 __ret_730; \
......@@ -59136,9 +59136,9 @@ __ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2)
5913659136}
5913759137#else
5913859138__ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
59139 int64x2_t __ret;
5913959140 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5914059141 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59141 int64x2_t __ret;
5914259142 __ret = __noswap_vqdmlsl_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
5914359143 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5914459144 return __ret;
......@@ -59153,9 +59153,9 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5915359153}
5915459154#else
5915559155__ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
59156 int32x4_t __ret;
5915659157 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5915759158 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59158 int32x4_t __ret;
5915959159 __ret = __noswap_vqdmlsl_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
5916059160 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5916159161 return __ret;
......@@ -59164,20 +59164,20 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5916459164
5916559165#ifdef __LITTLE_ENDIAN__
5916659166#define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
59167 int64_t __ret; \
5916759168 int64_t __s0 = __p0; \
5916859169 int32_t __s1 = __p1; \
5916959170 int32x2_t __s2 = __p2; \
59170 int64_t __ret; \
5917159171 __ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __s2, __p3); \
5917259172 __ret; \
5917359173})
5917459174#else
5917559175#define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
59176 int64_t __ret; \
5917659177 int64_t __s0 = __p0; \
5917759178 int32_t __s1 = __p1; \
5917859179 int32x2_t __s2 = __p2; \
5917959180 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 1, 0); \
59180 int64_t __ret; \
5918159181 __ret = (int64_t) __builtin_neon_vqdmlsls_lane_s32(__s0, __s1, __rev2, __p3); \
5918259182 __ret; \
5918359183})
......@@ -59185,20 +59185,20 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5918559185
5918659186#ifdef __LITTLE_ENDIAN__
5918759187#define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59188 int32_t __ret; \
5918859189 int32_t __s0 = __p0; \
5918959190 int16_t __s1 = __p1; \
5919059191 int16x4_t __s2 = __p2; \
59191 int32_t __ret; \
5919259192 __ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __s2, __p3); \
5919359193 __ret; \
5919459194})
5919559195#else
5919659196#define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59197 int32_t __ret; \
5919759198 int32_t __s0 = __p0; \
5919859199 int16_t __s1 = __p1; \
5919959200 int16x4_t __s2 = __p2; \
5920059201 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
59201 int32_t __ret; \
5920259202 __ret = (int32_t) __builtin_neon_vqdmlslh_lane_s16(__s0, __s1, __rev2, __p3); \
5920359203 __ret; \
5920459204})
......@@ -59206,20 +59206,20 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5920659206
5920759207#ifdef __LITTLE_ENDIAN__
5920859208#define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
59209 int64_t __ret; \
5920959210 int64_t __s0 = __p0; \
5921059211 int32_t __s1 = __p1; \
5921159212 int32x4_t __s2 = __p2; \
59212 int64_t __ret; \
5921359213 __ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __s2, __p3); \
5921459214 __ret; \
5921559215})
5921659216#else
5921759217#define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
59218 int64_t __ret; \
5921859219 int64_t __s0 = __p0; \
5921959220 int32_t __s1 = __p1; \
5922059221 int32x4_t __s2 = __p2; \
5922159222 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 3, 2, 1, 0); \
59222 int64_t __ret; \
5922359223 __ret = (int64_t) __builtin_neon_vqdmlsls_laneq_s32(__s0, __s1, __rev2, __p3); \
5922459224 __ret; \
5922559225})
......@@ -59227,20 +59227,20 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5922759227
5922859228#ifdef __LITTLE_ENDIAN__
5922959229#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59230 int32_t __ret; \
5923059231 int32_t __s0 = __p0; \
5923159232 int16_t __s1 = __p1; \
5923259233 int16x8_t __s2 = __p2; \
59233 int32_t __ret; \
5923459234 __ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __s2, __p3); \
5923559235 __ret; \
5923659236})
5923759237#else
5923859238#define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
59239 int32_t __ret; \
5923959240 int32_t __s0 = __p0; \
5924059241 int16_t __s1 = __p1; \
5924159242 int16x8_t __s2 = __p2; \
5924259243 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__s2, __s2, 7, 6, 5, 4, 3, 2, 1, 0); \
59243 int32_t __ret; \
5924459244 __ret = (int32_t) __builtin_neon_vqdmlslh_laneq_s16(__s0, __s1, __rev2, __p3); \
5924559245 __ret; \
5924659246})
......@@ -59248,22 +59248,22 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5924859248
5924959249#ifdef __LITTLE_ENDIAN__
5925059250#define vqdmlsl_laneq_s32(__p0_731, __p1_731, __p2_731, __p3_731) __extension__ ({ \
59251 int64x2_t __ret_731; \
5925159252 int64x2_t __s0_731 = __p0_731; \
5925259253 int32x2_t __s1_731 = __p1_731; \
5925359254 int32x4_t __s2_731 = __p2_731; \
59254 int64x2_t __ret_731; \
5925559255 __ret_731 = vqdmlsl_s32(__s0_731, __s1_731, splat_laneq_s32(__s2_731, __p3_731)); \
5925659256 __ret_731; \
5925759257})
5925859258#else
5925959259#define vqdmlsl_laneq_s32(__p0_732, __p1_732, __p2_732, __p3_732) __extension__ ({ \
59260 int64x2_t __ret_732; \
5926059261 int64x2_t __s0_732 = __p0_732; \
5926159262 int32x2_t __s1_732 = __p1_732; \
5926259263 int32x4_t __s2_732 = __p2_732; \
5926359264 int64x2_t __rev0_732; __rev0_732 = __builtin_shufflevector(__s0_732, __s0_732, 1, 0); \
5926459265 int32x2_t __rev1_732; __rev1_732 = __builtin_shufflevector(__s1_732, __s1_732, 1, 0); \
5926559266 int32x4_t __rev2_732; __rev2_732 = __builtin_shufflevector(__s2_732, __s2_732, 3, 2, 1, 0); \
59266 int64x2_t __ret_732; \
5926759267 __ret_732 = __noswap_vqdmlsl_s32(__rev0_732, __rev1_732, __noswap_splat_laneq_s32(__rev2_732, __p3_732)); \
5926859268 __ret_732 = __builtin_shufflevector(__ret_732, __ret_732, 1, 0); \
5926959269 __ret_732; \
......@@ -59272,22 +59272,22 @@ __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2)
5927259272
5927359273#ifdef __LITTLE_ENDIAN__
5927459274#define vqdmlsl_laneq_s16(__p0_733, __p1_733, __p2_733, __p3_733) __extension__ ({ \
59275 int32x4_t __ret_733; \
5927559276 int32x4_t __s0_733 = __p0_733; \
5927659277 int16x4_t __s1_733 = __p1_733; \
5927759278 int16x8_t __s2_733 = __p2_733; \
59278 int32x4_t __ret_733; \
5927959279 __ret_733 = vqdmlsl_s16(__s0_733, __s1_733, splat_laneq_s16(__s2_733, __p3_733)); \
5928059280 __ret_733; \
5928159281})
5928259282#else
5928359283#define vqdmlsl_laneq_s16(__p0_734, __p1_734, __p2_734, __p3_734) __extension__ ({ \
59284 int32x4_t __ret_734; \
5928459285 int32x4_t __s0_734 = __p0_734; \
5928559286 int16x4_t __s1_734 = __p1_734; \
5928659287 int16x8_t __s2_734 = __p2_734; \
5928759288 int32x4_t __rev0_734; __rev0_734 = __builtin_shufflevector(__s0_734, __s0_734, 3, 2, 1, 0); \
5928859289 int16x4_t __rev1_734; __rev1_734 = __builtin_shufflevector(__s1_734, __s1_734, 3, 2, 1, 0); \
5928959290 int16x8_t __rev2_734; __rev2_734 = __builtin_shufflevector(__s2_734, __s2_734, 7, 6, 5, 4, 3, 2, 1, 0); \
59290 int32x4_t __ret_734; \
5929159291 __ret_734 = __noswap_vqdmlsl_s16(__rev0_734, __rev1_734, __noswap_splat_laneq_s16(__rev2_734, __p3_734)); \
5929259292 __ret_734 = __builtin_shufflevector(__ret_734, __ret_734, 3, 2, 1, 0); \
5929359293 __ret_734; \
......@@ -59306,19 +59306,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5930659306}
5930759307#ifdef __LITTLE_ENDIAN__
5930859308#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59309 int32x4_t __ret; \
5930959310 int32x4_t __s0 = __p0; \
5931059311 int32x2_t __s1 = __p1; \
59311 int32x4_t __ret; \
5931259312 __ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
5931359313 __ret; \
5931459314})
5931559315#else
5931659316#define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59317 int32x4_t __ret; \
5931759318 int32x4_t __s0 = __p0; \
5931859319 int32x2_t __s1 = __p1; \
5931959320 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5932059321 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
59321 int32x4_t __ret; \
5932259322 __ret = (int32x4_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
5932359323 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
5932459324 __ret; \
......@@ -59327,19 +59327,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5932759327
5932859328#ifdef __LITTLE_ENDIAN__
5932959329#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59330 int16x8_t __ret; \
5933059331 int16x8_t __s0 = __p0; \
5933159332 int16x4_t __s1 = __p1; \
59332 int16x8_t __ret; \
5933359333 __ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
5933459334 __ret; \
5933559335})
5933659336#else
5933759337#define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59338 int16x8_t __ret; \
5933859339 int16x8_t __s0 = __p0; \
5933959340 int16x4_t __s1 = __p1; \
5934059341 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
5934159342 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59342 int16x8_t __ret; \
5934359343 __ret = (int16x8_t) __builtin_neon_vqdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
5934459344 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
5934559345 __ret; \
......@@ -59348,19 +59348,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5934859348
5934959349#ifdef __LITTLE_ENDIAN__
5935059350#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59351 int32x2_t __ret; \
5935159352 int32x2_t __s0 = __p0; \
5935259353 int32x2_t __s1 = __p1; \
59353 int32x2_t __ret; \
5935459354 __ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
5935559355 __ret; \
5935659356})
5935759357#else
5935859358#define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
59359 int32x2_t __ret; \
5935959360 int32x2_t __s0 = __p0; \
5936059361 int32x2_t __s1 = __p1; \
5936159362 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5936259363 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
59363 int32x2_t __ret; \
5936459364 __ret = (int32x2_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
5936559365 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5936659366 __ret; \
......@@ -59369,19 +59369,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5936959369
5937059370#ifdef __LITTLE_ENDIAN__
5937159371#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59372 int16x4_t __ret; \
5937259373 int16x4_t __s0 = __p0; \
5937359374 int16x4_t __s1 = __p1; \
59374 int16x4_t __ret; \
5937559375 __ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
5937659376 __ret; \
5937759377})
5937859378#else
5937959379#define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
59380 int16x4_t __ret; \
5938059381 int16x4_t __s0 = __p0; \
5938159382 int16x4_t __s1 = __p1; \
5938259383 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5938359384 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59384 int16x4_t __ret; \
5938559385 __ret = (int16x4_t) __builtin_neon_vqdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
5938659386 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
5938759387 __ret; \
......@@ -59390,18 +59390,18 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5939059390
5939159391#ifdef __LITTLE_ENDIAN__
5939259392#define vqdmulhs_lane_s32(__p0_735, __p1_735, __p2_735) __extension__ ({ \
59393 int32_t __ret_735; \
5939359394 int32_t __s0_735 = __p0_735; \
5939459395 int32x2_t __s1_735 = __p1_735; \
59395 int32_t __ret_735; \
5939659396 __ret_735 = vqdmulhs_s32(__s0_735, vget_lane_s32(__s1_735, __p2_735)); \
5939759397 __ret_735; \
5939859398})
5939959399#else
5940059400#define vqdmulhs_lane_s32(__p0_736, __p1_736, __p2_736) __extension__ ({ \
59401 int32_t __ret_736; \
5940159402 int32_t __s0_736 = __p0_736; \
5940259403 int32x2_t __s1_736 = __p1_736; \
5940359404 int32x2_t __rev1_736; __rev1_736 = __builtin_shufflevector(__s1_736, __s1_736, 1, 0); \
59404 int32_t __ret_736; \
5940559405 __ret_736 = vqdmulhs_s32(__s0_736, __noswap_vget_lane_s32(__rev1_736, __p2_736)); \
5940659406 __ret_736; \
5940759407})
......@@ -59409,18 +59409,18 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5940959409
5941059410#ifdef __LITTLE_ENDIAN__
5941159411#define vqdmulhh_lane_s16(__p0_737, __p1_737, __p2_737) __extension__ ({ \
59412 int16_t __ret_737; \
5941259413 int16_t __s0_737 = __p0_737; \
5941359414 int16x4_t __s1_737 = __p1_737; \
59414 int16_t __ret_737; \
5941559415 __ret_737 = vqdmulhh_s16(__s0_737, vget_lane_s16(__s1_737, __p2_737)); \
5941659416 __ret_737; \
5941759417})
5941859418#else
5941959419#define vqdmulhh_lane_s16(__p0_738, __p1_738, __p2_738) __extension__ ({ \
59420 int16_t __ret_738; \
5942059421 int16_t __s0_738 = __p0_738; \
5942159422 int16x4_t __s1_738 = __p1_738; \
5942259423 int16x4_t __rev1_738; __rev1_738 = __builtin_shufflevector(__s1_738, __s1_738, 3, 2, 1, 0); \
59423 int16_t __ret_738; \
5942459424 __ret_738 = vqdmulhh_s16(__s0_738, __noswap_vget_lane_s16(__rev1_738, __p2_738)); \
5942559425 __ret_738; \
5942659426})
......@@ -59428,18 +59428,18 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5942859428
5942959429#ifdef __LITTLE_ENDIAN__
5943059430#define vqdmulhs_laneq_s32(__p0_739, __p1_739, __p2_739) __extension__ ({ \
59431 int32_t __ret_739; \
5943159432 int32_t __s0_739 = __p0_739; \
5943259433 int32x4_t __s1_739 = __p1_739; \
59433 int32_t __ret_739; \
5943459434 __ret_739 = vqdmulhs_s32(__s0_739, vgetq_lane_s32(__s1_739, __p2_739)); \
5943559435 __ret_739; \
5943659436})
5943759437#else
5943859438#define vqdmulhs_laneq_s32(__p0_740, __p1_740, __p2_740) __extension__ ({ \
59439 int32_t __ret_740; \
5943959440 int32_t __s0_740 = __p0_740; \
5944059441 int32x4_t __s1_740 = __p1_740; \
5944159442 int32x4_t __rev1_740; __rev1_740 = __builtin_shufflevector(__s1_740, __s1_740, 3, 2, 1, 0); \
59442 int32_t __ret_740; \
5944359443 __ret_740 = vqdmulhs_s32(__s0_740, __noswap_vgetq_lane_s32(__rev1_740, __p2_740)); \
5944459444 __ret_740; \
5944559445})
......@@ -59447,18 +59447,18 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5944759447
5944859448#ifdef __LITTLE_ENDIAN__
5944959449#define vqdmulhh_laneq_s16(__p0_741, __p1_741, __p2_741) __extension__ ({ \
59450 int16_t __ret_741; \
5945059451 int16_t __s0_741 = __p0_741; \
5945159452 int16x8_t __s1_741 = __p1_741; \
59452 int16_t __ret_741; \
5945359453 __ret_741 = vqdmulhh_s16(__s0_741, vgetq_lane_s16(__s1_741, __p2_741)); \
5945459454 __ret_741; \
5945559455})
5945659456#else
5945759457#define vqdmulhh_laneq_s16(__p0_742, __p1_742, __p2_742) __extension__ ({ \
59458 int16_t __ret_742; \
5945859459 int16_t __s0_742 = __p0_742; \
5945959460 int16x8_t __s1_742 = __p1_742; \
5946059461 int16x8_t __rev1_742; __rev1_742 = __builtin_shufflevector(__s1_742, __s1_742, 7, 6, 5, 4, 3, 2, 1, 0); \
59461 int16_t __ret_742; \
5946259462 __ret_742 = vqdmulhh_s16(__s0_742, __noswap_vgetq_lane_s16(__rev1_742, __p2_742)); \
5946359463 __ret_742; \
5946459464})
......@@ -59466,19 +59466,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5946659466
5946759467#ifdef __LITTLE_ENDIAN__
5946859468#define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59469 int32x4_t __ret; \
5946959470 int32x4_t __s0 = __p0; \
5947059471 int32x4_t __s1 = __p1; \
59471 int32x4_t __ret; \
5947259472 __ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
5947359473 __ret; \
5947459474})
5947559475#else
5947659476#define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59477 int32x4_t __ret; \
5947759478 int32x4_t __s0 = __p0; \
5947859479 int32x4_t __s1 = __p1; \
5947959480 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5948059481 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59481 int32x4_t __ret; \
5948259482 __ret = (int32x4_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
5948359483 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
5948459484 __ret; \
......@@ -59487,19 +59487,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5948759487
5948859488#ifdef __LITTLE_ENDIAN__
5948959489#define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59490 int16x8_t __ret; \
5949059491 int16x8_t __s0 = __p0; \
5949159492 int16x8_t __s1 = __p1; \
59492 int16x8_t __ret; \
5949359493 __ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
5949459494 __ret; \
5949559495})
5949659496#else
5949759497#define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59498 int16x8_t __ret; \
5949859499 int16x8_t __s0 = __p0; \
5949959500 int16x8_t __s1 = __p1; \
5950059501 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
5950159502 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
59502 int16x8_t __ret; \
5950359503 __ret = (int16x8_t) __builtin_neon_vqdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
5950459504 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
5950559505 __ret; \
......@@ -59508,19 +59508,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5950859508
5950959509#ifdef __LITTLE_ENDIAN__
5951059510#define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59511 int32x2_t __ret; \
5951159512 int32x2_t __s0 = __p0; \
5951259513 int32x4_t __s1 = __p1; \
59513 int32x2_t __ret; \
5951459514 __ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
5951559515 __ret; \
5951659516})
5951759517#else
5951859518#define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
59519 int32x2_t __ret; \
5951959520 int32x2_t __s0 = __p0; \
5952059521 int32x4_t __s1 = __p1; \
5952159522 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
5952259523 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
59523 int32x2_t __ret; \
5952459524 __ret = (int32x2_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
5952559525 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
5952659526 __ret; \
......@@ -59529,19 +59529,19 @@ __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
5952959529
5953059530#ifdef __LITTLE_ENDIAN__
5953159531#define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59532 int16x4_t __ret; \
5953259533 int16x4_t __s0 = __p0; \
5953359534 int16x8_t __s1 = __p1; \
59534 int16x4_t __ret; \
5953559535 __ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
5953659536 __ret; \
5953759537})
5953859538#else
5953959539#define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
59540 int16x4_t __ret; \
5954059541 int16x4_t __s0 = __p0; \
5954159542 int16x8_t __s1 = __p1; \
5954259543 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
5954359544 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
59544 int16x4_t __ret; \
5954559545 __ret = (int16x4_t) __builtin_neon_vqdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
5954659546 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
5954759547 __ret; \
......@@ -59566,9 +59566,9 @@ __ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
5956659566}
5956759567#else
5956859568__ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
59569 int64x2_t __ret;
5956959570 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5957059571 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59571 int64x2_t __ret;
5957259572 __ret = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
5957359573 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5957459574 return __ret;
......@@ -59583,9 +59583,9 @@ __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
5958359583}
5958459584#else
5958559585__ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
59586 int32x4_t __ret;
5958659587 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5958759588 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59588 int32x4_t __ret;
5958959589 __ret = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
5959059590 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5959159591 return __ret;
......@@ -59594,19 +59594,19 @@ __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
5959459594
5959559595#ifdef __LITTLE_ENDIAN__
5959659596#define vqdmull_high_lane_s32(__p0_743, __p1_743, __p2_743) __extension__ ({ \
59597 int64x2_t __ret_743; \
5959759598 int32x4_t __s0_743 = __p0_743; \
5959859599 int32x2_t __s1_743 = __p1_743; \
59599 int64x2_t __ret_743; \
5960059600 __ret_743 = vqdmull_s32(vget_high_s32(__s0_743), splat_lane_s32(__s1_743, __p2_743)); \
5960159601 __ret_743; \
5960259602})
5960359603#else
5960459604#define vqdmull_high_lane_s32(__p0_744, __p1_744, __p2_744) __extension__ ({ \
59605 int64x2_t __ret_744; \
5960559606 int32x4_t __s0_744 = __p0_744; \
5960659607 int32x2_t __s1_744 = __p1_744; \
5960759608 int32x4_t __rev0_744; __rev0_744 = __builtin_shufflevector(__s0_744, __s0_744, 3, 2, 1, 0); \
5960859609 int32x2_t __rev1_744; __rev1_744 = __builtin_shufflevector(__s1_744, __s1_744, 1, 0); \
59609 int64x2_t __ret_744; \
5961059610 __ret_744 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_744), __noswap_splat_lane_s32(__rev1_744, __p2_744)); \
5961159611 __ret_744 = __builtin_shufflevector(__ret_744, __ret_744, 1, 0); \
5961259612 __ret_744; \
......@@ -59615,19 +59615,19 @@ __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
5961559615
5961659616#ifdef __LITTLE_ENDIAN__
5961759617#define vqdmull_high_lane_s16(__p0_745, __p1_745, __p2_745) __extension__ ({ \
59618 int32x4_t __ret_745; \
5961859619 int16x8_t __s0_745 = __p0_745; \
5961959620 int16x4_t __s1_745 = __p1_745; \
59620 int32x4_t __ret_745; \
5962159621 __ret_745 = vqdmull_s16(vget_high_s16(__s0_745), splat_lane_s16(__s1_745, __p2_745)); \
5962259622 __ret_745; \
5962359623})
5962459624#else
5962559625#define vqdmull_high_lane_s16(__p0_746, __p1_746, __p2_746) __extension__ ({ \
59626 int32x4_t __ret_746; \
5962659627 int16x8_t __s0_746 = __p0_746; \
5962759628 int16x4_t __s1_746 = __p1_746; \
5962859629 int16x8_t __rev0_746; __rev0_746 = __builtin_shufflevector(__s0_746, __s0_746, 7, 6, 5, 4, 3, 2, 1, 0); \
5962959630 int16x4_t __rev1_746; __rev1_746 = __builtin_shufflevector(__s1_746, __s1_746, 3, 2, 1, 0); \
59630 int32x4_t __ret_746; \
5963159631 __ret_746 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_746), __noswap_splat_lane_s16(__rev1_746, __p2_746)); \
5963259632 __ret_746 = __builtin_shufflevector(__ret_746, __ret_746, 3, 2, 1, 0); \
5963359633 __ret_746; \
......@@ -59636,19 +59636,19 @@ __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
5963659636
5963759637#ifdef __LITTLE_ENDIAN__
5963859638#define vqdmull_high_laneq_s32(__p0_747, __p1_747, __p2_747) __extension__ ({ \
59639 int64x2_t __ret_747; \
5963959640 int32x4_t __s0_747 = __p0_747; \
5964059641 int32x4_t __s1_747 = __p1_747; \
59641 int64x2_t __ret_747; \
5964259642 __ret_747 = vqdmull_s32(vget_high_s32(__s0_747), splat_laneq_s32(__s1_747, __p2_747)); \
5964359643 __ret_747; \
5964459644})
5964559645#else
5964659646#define vqdmull_high_laneq_s32(__p0_748, __p1_748, __p2_748) __extension__ ({ \
59647 int64x2_t __ret_748; \
5964759648 int32x4_t __s0_748 = __p0_748; \
5964859649 int32x4_t __s1_748 = __p1_748; \
5964959650 int32x4_t __rev0_748; __rev0_748 = __builtin_shufflevector(__s0_748, __s0_748, 3, 2, 1, 0); \
5965059651 int32x4_t __rev1_748; __rev1_748 = __builtin_shufflevector(__s1_748, __s1_748, 3, 2, 1, 0); \
59651 int64x2_t __ret_748; \
5965259652 __ret_748 = __noswap_vqdmull_s32(__noswap_vget_high_s32(__rev0_748), __noswap_splat_laneq_s32(__rev1_748, __p2_748)); \
5965359653 __ret_748 = __builtin_shufflevector(__ret_748, __ret_748, 1, 0); \
5965459654 __ret_748; \
......@@ -59657,19 +59657,19 @@ __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
5965759657
5965859658#ifdef __LITTLE_ENDIAN__
5965959659#define vqdmull_high_laneq_s16(__p0_749, __p1_749, __p2_749) __extension__ ({ \
59660 int32x4_t __ret_749; \
5966059661 int16x8_t __s0_749 = __p0_749; \
5966159662 int16x8_t __s1_749 = __p1_749; \
59662 int32x4_t __ret_749; \
5966359663 __ret_749 = vqdmull_s16(vget_high_s16(__s0_749), splat_laneq_s16(__s1_749, __p2_749)); \
5966459664 __ret_749; \
5966559665})
5966659666#else
5966759667#define vqdmull_high_laneq_s16(__p0_750, __p1_750, __p2_750) __extension__ ({ \
59668 int32x4_t __ret_750; \
5966859669 int16x8_t __s0_750 = __p0_750; \
5966959670 int16x8_t __s1_750 = __p1_750; \
5967059671 int16x8_t __rev0_750; __rev0_750 = __builtin_shufflevector(__s0_750, __s0_750, 7, 6, 5, 4, 3, 2, 1, 0); \
5967159672 int16x8_t __rev1_750; __rev1_750 = __builtin_shufflevector(__s1_750, __s1_750, 7, 6, 5, 4, 3, 2, 1, 0); \
59672 int32x4_t __ret_750; \
5967359673 __ret_750 = __noswap_vqdmull_s16(__noswap_vget_high_s16(__rev0_750), __noswap_splat_laneq_s16(__rev1_750, __p2_750)); \
5967459674 __ret_750 = __builtin_shufflevector(__ret_750, __ret_750, 3, 2, 1, 0); \
5967559675 __ret_750; \
......@@ -59684,8 +59684,8 @@ __ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
5968459684}
5968559685#else
5968659686__ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
59687 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5968859687 int64x2_t __ret;
59688 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5968959689 __ret = __noswap_vqdmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
5969059690 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
5969159691 return __ret;
......@@ -59700,8 +59700,8 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5970059700}
5970159701#else
5970259702__ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
59703 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5970459703 int32x4_t __ret;
59704 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5970559705 __ret = __noswap_vqdmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
5970659706 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5970759707 return __ret;
......@@ -59710,18 +59710,18 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5971059710
5971159711#ifdef __LITTLE_ENDIAN__
5971259712#define vqdmulls_lane_s32(__p0_751, __p1_751, __p2_751) __extension__ ({ \
59713 int64_t __ret_751; \
5971359714 int32_t __s0_751 = __p0_751; \
5971459715 int32x2_t __s1_751 = __p1_751; \
59715 int64_t __ret_751; \
5971659716 __ret_751 = vqdmulls_s32(__s0_751, vget_lane_s32(__s1_751, __p2_751)); \
5971759717 __ret_751; \
5971859718})
5971959719#else
5972059720#define vqdmulls_lane_s32(__p0_752, __p1_752, __p2_752) __extension__ ({ \
59721 int64_t __ret_752; \
5972159722 int32_t __s0_752 = __p0_752; \
5972259723 int32x2_t __s1_752 = __p1_752; \
5972359724 int32x2_t __rev1_752; __rev1_752 = __builtin_shufflevector(__s1_752, __s1_752, 1, 0); \
59724 int64_t __ret_752; \
5972559725 __ret_752 = vqdmulls_s32(__s0_752, __noswap_vget_lane_s32(__rev1_752, __p2_752)); \
5972659726 __ret_752; \
5972759727})
......@@ -59729,18 +59729,18 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5972959729
5973059730#ifdef __LITTLE_ENDIAN__
5973159731#define vqdmullh_lane_s16(__p0_753, __p1_753, __p2_753) __extension__ ({ \
59732 int32_t __ret_753; \
5973259733 int16_t __s0_753 = __p0_753; \
5973359734 int16x4_t __s1_753 = __p1_753; \
59734 int32_t __ret_753; \
5973559735 __ret_753 = vqdmullh_s16(__s0_753, vget_lane_s16(__s1_753, __p2_753)); \
5973659736 __ret_753; \
5973759737})
5973859738#else
5973959739#define vqdmullh_lane_s16(__p0_754, __p1_754, __p2_754) __extension__ ({ \
59740 int32_t __ret_754; \
5974059741 int16_t __s0_754 = __p0_754; \
5974159742 int16x4_t __s1_754 = __p1_754; \
5974259743 int16x4_t __rev1_754; __rev1_754 = __builtin_shufflevector(__s1_754, __s1_754, 3, 2, 1, 0); \
59743 int32_t __ret_754; \
5974459744 __ret_754 = vqdmullh_s16(__s0_754, __noswap_vget_lane_s16(__rev1_754, __p2_754)); \
5974559745 __ret_754; \
5974659746})
......@@ -59748,18 +59748,18 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5974859748
5974959749#ifdef __LITTLE_ENDIAN__
5975059750#define vqdmulls_laneq_s32(__p0_755, __p1_755, __p2_755) __extension__ ({ \
59751 int64_t __ret_755; \
5975159752 int32_t __s0_755 = __p0_755; \
5975259753 int32x4_t __s1_755 = __p1_755; \
59753 int64_t __ret_755; \
5975459754 __ret_755 = vqdmulls_s32(__s0_755, vgetq_lane_s32(__s1_755, __p2_755)); \
5975559755 __ret_755; \
5975659756})
5975759757#else
5975859758#define vqdmulls_laneq_s32(__p0_756, __p1_756, __p2_756) __extension__ ({ \
59759 int64_t __ret_756; \
5975959760 int32_t __s0_756 = __p0_756; \
5976059761 int32x4_t __s1_756 = __p1_756; \
5976159762 int32x4_t __rev1_756; __rev1_756 = __builtin_shufflevector(__s1_756, __s1_756, 3, 2, 1, 0); \
59762 int64_t __ret_756; \
5976359763 __ret_756 = vqdmulls_s32(__s0_756, __noswap_vgetq_lane_s32(__rev1_756, __p2_756)); \
5976459764 __ret_756; \
5976559765})
......@@ -59767,18 +59767,18 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5976759767
5976859768#ifdef __LITTLE_ENDIAN__
5976959769#define vqdmullh_laneq_s16(__p0_757, __p1_757, __p2_757) __extension__ ({ \
59770 int32_t __ret_757; \
5977059771 int16_t __s0_757 = __p0_757; \
5977159772 int16x8_t __s1_757 = __p1_757; \
59772 int32_t __ret_757; \
5977359773 __ret_757 = vqdmullh_s16(__s0_757, vgetq_lane_s16(__s1_757, __p2_757)); \
5977459774 __ret_757; \
5977559775})
5977659776#else
5977759777#define vqdmullh_laneq_s16(__p0_758, __p1_758, __p2_758) __extension__ ({ \
59778 int32_t __ret_758; \
5977859779 int16_t __s0_758 = __p0_758; \
5977959780 int16x8_t __s1_758 = __p1_758; \
5978059781 int16x8_t __rev1_758; __rev1_758 = __builtin_shufflevector(__s1_758, __s1_758, 7, 6, 5, 4, 3, 2, 1, 0); \
59781 int32_t __ret_758; \
5978259782 __ret_758 = vqdmullh_s16(__s0_758, __noswap_vgetq_lane_s16(__rev1_758, __p2_758)); \
5978359783 __ret_758; \
5978459784})
......@@ -59786,19 +59786,19 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5978659786
5978759787#ifdef __LITTLE_ENDIAN__
5978859788#define vqdmull_laneq_s32(__p0_759, __p1_759, __p2_759) __extension__ ({ \
59789 int64x2_t __ret_759; \
5978959790 int32x2_t __s0_759 = __p0_759; \
5979059791 int32x4_t __s1_759 = __p1_759; \
59791 int64x2_t __ret_759; \
5979259792 __ret_759 = vqdmull_s32(__s0_759, splat_laneq_s32(__s1_759, __p2_759)); \
5979359793 __ret_759; \
5979459794})
5979559795#else
5979659796#define vqdmull_laneq_s32(__p0_760, __p1_760, __p2_760) __extension__ ({ \
59797 int64x2_t __ret_760; \
5979759798 int32x2_t __s0_760 = __p0_760; \
5979859799 int32x4_t __s1_760 = __p1_760; \
5979959800 int32x2_t __rev0_760; __rev0_760 = __builtin_shufflevector(__s0_760, __s0_760, 1, 0); \
5980059801 int32x4_t __rev1_760; __rev1_760 = __builtin_shufflevector(__s1_760, __s1_760, 3, 2, 1, 0); \
59801 int64x2_t __ret_760; \
5980259802 __ret_760 = __noswap_vqdmull_s32(__rev0_760, __noswap_splat_laneq_s32(__rev1_760, __p2_760)); \
5980359803 __ret_760 = __builtin_shufflevector(__ret_760, __ret_760, 1, 0); \
5980459804 __ret_760; \
......@@ -59807,19 +59807,19 @@ __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
5980759807
5980859808#ifdef __LITTLE_ENDIAN__
5980959809#define vqdmull_laneq_s16(__p0_761, __p1_761, __p2_761) __extension__ ({ \
59810 int32x4_t __ret_761; \
5981059811 int16x4_t __s0_761 = __p0_761; \
5981159812 int16x8_t __s1_761 = __p1_761; \
59812 int32x4_t __ret_761; \
5981359813 __ret_761 = vqdmull_s16(__s0_761, splat_laneq_s16(__s1_761, __p2_761)); \
5981459814 __ret_761; \
5981559815})
5981659816#else
5981759817#define vqdmull_laneq_s16(__p0_762, __p1_762, __p2_762) __extension__ ({ \
59818 int32x4_t __ret_762; \
5981859819 int16x4_t __s0_762 = __p0_762; \
5981959820 int16x8_t __s1_762 = __p1_762; \
5982059821 int16x4_t __rev0_762; __rev0_762 = __builtin_shufflevector(__s0_762, __s0_762, 3, 2, 1, 0); \
5982159822 int16x8_t __rev1_762; __rev1_762 = __builtin_shufflevector(__s1_762, __s1_762, 7, 6, 5, 4, 3, 2, 1, 0); \
59822 int32x4_t __ret_762; \
5982359823 __ret_762 = __noswap_vqdmull_s16(__rev0_762, __noswap_splat_laneq_s16(__rev1_762, __p2_762)); \
5982459824 __ret_762 = __builtin_shufflevector(__ret_762, __ret_762, 3, 2, 1, 0); \
5982559825 __ret_762; \
......@@ -59864,9 +59864,9 @@ __ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
5986459864}
5986559865#else
5986659866__ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
59867 uint16x8_t __ret;
5986759868 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5986859869 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59869 uint16x8_t __ret;
5987059870 __ret = __noswap_vcombine_u16(__rev0, __noswap_vqmovn_u32(__rev1));
5987159871 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5987259872 return __ret;
......@@ -59881,9 +59881,9 @@ __ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
5988159881}
5988259882#else
5988359883__ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
59884 uint32x4_t __ret;
5988459885 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5988559886 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59886 uint32x4_t __ret;
5988759887 __ret = __noswap_vcombine_u32(__rev0, __noswap_vqmovn_u64(__rev1));
5988859888 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5988959889 return __ret;
......@@ -59898,9 +59898,9 @@ __ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
5989859898}
5989959899#else
5990059900__ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
59901 uint8x16_t __ret;
5990159902 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5990259903 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59903 uint8x16_t __ret;
5990459904 __ret = __noswap_vcombine_u8(__rev0, __noswap_vqmovn_u16(__rev1));
5990559905 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5990659906 return __ret;
......@@ -59915,9 +59915,9 @@ __ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
5991559915}
5991659916#else
5991759917__ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
59918 int16x8_t __ret;
5991859919 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5991959920 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59920 int16x8_t __ret;
5992159921 __ret = __noswap_vcombine_s16(__rev0, __noswap_vqmovn_s32(__rev1));
5992259922 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5992359923 return __ret;
......@@ -59932,9 +59932,9 @@ __ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
5993259932}
5993359933#else
5993459934__ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
59935 int32x4_t __ret;
5993559936 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
5993659937 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59937 int32x4_t __ret;
5993859938 __ret = __noswap_vcombine_s32(__rev0, __noswap_vqmovn_s64(__rev1));
5993959939 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
5994059940 return __ret;
......@@ -59949,9 +59949,9 @@ __ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
5994959949}
5995059950#else
5995159951__ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
59952 int8x16_t __ret;
5995259953 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
5995359954 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59954 int8x16_t __ret;
5995559955 __ret = __noswap_vcombine_s8(__rev0, __noswap_vqmovn_s16(__rev1));
5995659956 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5995759957 return __ret;
......@@ -59981,9 +59981,9 @@ __ai uint16x8_t vqmovun_high_s32(uint16x4_t __p0, int32x4_t __p1) {
5998159981}
5998259982#else
5998359983__ai uint16x8_t vqmovun_high_s32(uint16x4_t __p0, int32x4_t __p1) {
59984 uint16x8_t __ret;
5998459985 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
5998559986 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59986 uint16x8_t __ret;
5998759987 __ret = __noswap_vcombine_u16((uint16x4_t)(__rev0), __noswap_vqmovun_s32(__rev1));
5998859988 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
5998959989 return __ret;
......@@ -59998,9 +59998,9 @@ __ai uint32x4_t vqmovun_high_s64(uint32x2_t __p0, int64x2_t __p1) {
5999859998}
5999959999#else
6000060000__ai uint32x4_t vqmovun_high_s64(uint32x2_t __p0, int64x2_t __p1) {
60001 uint32x4_t __ret;
6000160002 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6000260003 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60003 uint32x4_t __ret;
6000460004 __ret = __noswap_vcombine_u32((uint32x2_t)(__rev0), __noswap_vqmovun_s64(__rev1));
6000560005 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6000660006 return __ret;
......@@ -60015,9 +60015,9 @@ __ai uint8x16_t vqmovun_high_s16(uint8x8_t __p0, int16x8_t __p1) {
6001560015}
6001660016#else
6001760017__ai uint8x16_t vqmovun_high_s16(uint8x8_t __p0, int16x8_t __p1) {
60018 uint8x16_t __ret;
6001860019 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6001960020 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60020 uint8x16_t __ret;
6002160021 __ret = __noswap_vcombine_u8((uint8x8_t)(__rev0), __noswap_vqmovun_s16(__rev1));
6002260022 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6002360023 return __ret;
......@@ -60032,8 +60032,8 @@ __ai int64x2_t vqnegq_s64(int64x2_t __p0) {
6003260032}
6003360033#else
6003460034__ai int64x2_t vqnegq_s64(int64x2_t __p0) {
60035 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6003660035 int64x2_t __ret;
60036 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6003760037 __ret = (int64x2_t) __builtin_neon_vqnegq_v((int8x16_t)__rev0, 35);
6003860038 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6003960039 return __ret;
......@@ -60077,19 +60077,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6007760077}
6007860078#ifdef __LITTLE_ENDIAN__
6007960079#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60080 int32x4_t __ret; \
6008060081 int32x4_t __s0 = __p0; \
6008160082 int32x2_t __s1 = __p1; \
60082 int32x4_t __ret; \
6008360083 __ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 2); \
6008460084 __ret; \
6008560085})
6008660086#else
6008760087#define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60088 int32x4_t __ret; \
6008860089 int32x4_t __s0 = __p0; \
6008960090 int32x2_t __s1 = __p1; \
6009060091 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
6009160092 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
60092 int32x4_t __ret; \
6009360093 __ret = (int32x4_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
6009460094 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
6009560095 __ret; \
......@@ -60098,19 +60098,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6009860098
6009960099#ifdef __LITTLE_ENDIAN__
6010060100#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60101 int16x8_t __ret; \
6010160102 int16x8_t __s0 = __p0; \
6010260103 int16x4_t __s1 = __p1; \
60103 int16x8_t __ret; \
6010460104 __ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__s0, (int8x8_t)__s1, __p2, 1); \
6010560105 __ret; \
6010660106})
6010760107#else
6010860108#define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60109 int16x8_t __ret; \
6010960110 int16x8_t __s0 = __p0; \
6011060111 int16x4_t __s1 = __p1; \
6011160112 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
6011260113 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60113 int16x8_t __ret; \
6011460114 __ret = (int16x8_t) __builtin_neon_vqrdmulhq_lane_v((int8x16_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
6011560115 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
6011660116 __ret; \
......@@ -60119,19 +60119,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6011960119
6012060120#ifdef __LITTLE_ENDIAN__
6012160121#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60122 int32x2_t __ret; \
6012260123 int32x2_t __s0 = __p0; \
6012360124 int32x2_t __s1 = __p1; \
60124 int32x2_t __ret; \
6012560125 __ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 2); \
6012660126 __ret; \
6012760127})
6012860128#else
6012960129#define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
60130 int32x2_t __ret; \
6013060131 int32x2_t __s0 = __p0; \
6013160132 int32x2_t __s1 = __p1; \
6013260133 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
6013360134 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
60134 int32x2_t __ret; \
6013560135 __ret = (int32x2_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 2); \
6013660136 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
6013760137 __ret; \
......@@ -60140,19 +60140,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6014060140
6014160141#ifdef __LITTLE_ENDIAN__
6014260142#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60143 int16x4_t __ret; \
6014360144 int16x4_t __s0 = __p0; \
6014460145 int16x4_t __s1 = __p1; \
60145 int16x4_t __ret; \
6014660146 __ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 1); \
6014760147 __ret; \
6014860148})
6014960149#else
6015060150#define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
60151 int16x4_t __ret; \
6015160152 int16x4_t __s0 = __p0; \
6015260153 int16x4_t __s1 = __p1; \
6015360154 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
6015460155 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60155 int16x4_t __ret; \
6015660156 __ret = (int16x4_t) __builtin_neon_vqrdmulh_lane_v((int8x8_t)__rev0, (int8x8_t)__rev1, __p2, 1); \
6015760157 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
6015860158 __ret; \
......@@ -60161,18 +60161,18 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6016160161
6016260162#ifdef __LITTLE_ENDIAN__
6016360163#define vqrdmulhs_lane_s32(__p0_763, __p1_763, __p2_763) __extension__ ({ \
60164 int32_t __ret_763; \
6016460165 int32_t __s0_763 = __p0_763; \
6016560166 int32x2_t __s1_763 = __p1_763; \
60166 int32_t __ret_763; \
6016760167 __ret_763 = vqrdmulhs_s32(__s0_763, vget_lane_s32(__s1_763, __p2_763)); \
6016860168 __ret_763; \
6016960169})
6017060170#else
6017160171#define vqrdmulhs_lane_s32(__p0_764, __p1_764, __p2_764) __extension__ ({ \
60172 int32_t __ret_764; \
6017260173 int32_t __s0_764 = __p0_764; \
6017360174 int32x2_t __s1_764 = __p1_764; \
6017460175 int32x2_t __rev1_764; __rev1_764 = __builtin_shufflevector(__s1_764, __s1_764, 1, 0); \
60175 int32_t __ret_764; \
6017660176 __ret_764 = vqrdmulhs_s32(__s0_764, __noswap_vget_lane_s32(__rev1_764, __p2_764)); \
6017760177 __ret_764; \
6017860178})
......@@ -60180,18 +60180,18 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6018060180
6018160181#ifdef __LITTLE_ENDIAN__
6018260182#define vqrdmulhh_lane_s16(__p0_765, __p1_765, __p2_765) __extension__ ({ \
60183 int16_t __ret_765; \
6018360184 int16_t __s0_765 = __p0_765; \
6018460185 int16x4_t __s1_765 = __p1_765; \
60185 int16_t __ret_765; \
6018660186 __ret_765 = vqrdmulhh_s16(__s0_765, vget_lane_s16(__s1_765, __p2_765)); \
6018760187 __ret_765; \
6018860188})
6018960189#else
6019060190#define vqrdmulhh_lane_s16(__p0_766, __p1_766, __p2_766) __extension__ ({ \
60191 int16_t __ret_766; \
6019160192 int16_t __s0_766 = __p0_766; \
6019260193 int16x4_t __s1_766 = __p1_766; \
6019360194 int16x4_t __rev1_766; __rev1_766 = __builtin_shufflevector(__s1_766, __s1_766, 3, 2, 1, 0); \
60194 int16_t __ret_766; \
6019560195 __ret_766 = vqrdmulhh_s16(__s0_766, __noswap_vget_lane_s16(__rev1_766, __p2_766)); \
6019660196 __ret_766; \
6019760197})
......@@ -60199,18 +60199,18 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6019960199
6020060200#ifdef __LITTLE_ENDIAN__
6020160201#define vqrdmulhs_laneq_s32(__p0_767, __p1_767, __p2_767) __extension__ ({ \
60202 int32_t __ret_767; \
6020260203 int32_t __s0_767 = __p0_767; \
6020360204 int32x4_t __s1_767 = __p1_767; \
60204 int32_t __ret_767; \
6020560205 __ret_767 = vqrdmulhs_s32(__s0_767, vgetq_lane_s32(__s1_767, __p2_767)); \
6020660206 __ret_767; \
6020760207})
6020860208#else
6020960209#define vqrdmulhs_laneq_s32(__p0_768, __p1_768, __p2_768) __extension__ ({ \
60210 int32_t __ret_768; \
6021060211 int32_t __s0_768 = __p0_768; \
6021160212 int32x4_t __s1_768 = __p1_768; \
6021260213 int32x4_t __rev1_768; __rev1_768 = __builtin_shufflevector(__s1_768, __s1_768, 3, 2, 1, 0); \
60213 int32_t __ret_768; \
6021460214 __ret_768 = vqrdmulhs_s32(__s0_768, __noswap_vgetq_lane_s32(__rev1_768, __p2_768)); \
6021560215 __ret_768; \
6021660216})
......@@ -60218,18 +60218,18 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6021860218
6021960219#ifdef __LITTLE_ENDIAN__
6022060220#define vqrdmulhh_laneq_s16(__p0_769, __p1_769, __p2_769) __extension__ ({ \
60221 int16_t __ret_769; \
6022160222 int16_t __s0_769 = __p0_769; \
6022260223 int16x8_t __s1_769 = __p1_769; \
60223 int16_t __ret_769; \
6022460224 __ret_769 = vqrdmulhh_s16(__s0_769, vgetq_lane_s16(__s1_769, __p2_769)); \
6022560225 __ret_769; \
6022660226})
6022760227#else
6022860228#define vqrdmulhh_laneq_s16(__p0_770, __p1_770, __p2_770) __extension__ ({ \
60229 int16_t __ret_770; \
6022960230 int16_t __s0_770 = __p0_770; \
6023060231 int16x8_t __s1_770 = __p1_770; \
6023160232 int16x8_t __rev1_770; __rev1_770 = __builtin_shufflevector(__s1_770, __s1_770, 7, 6, 5, 4, 3, 2, 1, 0); \
60232 int16_t __ret_770; \
6023360233 __ret_770 = vqrdmulhh_s16(__s0_770, __noswap_vgetq_lane_s16(__rev1_770, __p2_770)); \
6023460234 __ret_770; \
6023560235})
......@@ -60237,19 +60237,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6023760237
6023860238#ifdef __LITTLE_ENDIAN__
6023960239#define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60240 int32x4_t __ret; \
6024060241 int32x4_t __s0 = __p0; \
6024160242 int32x4_t __s1 = __p1; \
60242 int32x4_t __ret; \
6024360243 __ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 34); \
6024460244 __ret; \
6024560245})
6024660246#else
6024760247#define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60248 int32x4_t __ret; \
6024860249 int32x4_t __s0 = __p0; \
6024960250 int32x4_t __s1 = __p1; \
6025060251 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
6025160252 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60252 int32x4_t __ret; \
6025360253 __ret = (int32x4_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 34); \
6025460254 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
6025560255 __ret; \
......@@ -60258,19 +60258,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6025860258
6025960259#ifdef __LITTLE_ENDIAN__
6026060260#define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60261 int16x8_t __ret; \
6026160262 int16x8_t __s0 = __p0; \
6026260263 int16x8_t __s1 = __p1; \
60263 int16x8_t __ret; \
6026460264 __ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 33); \
6026560265 __ret; \
6026660266})
6026760267#else
6026860268#define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60269 int16x8_t __ret; \
6026960270 int16x8_t __s0 = __p0; \
6027060271 int16x8_t __s1 = __p1; \
6027160272 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 7, 6, 5, 4, 3, 2, 1, 0); \
6027260273 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
60273 int16x8_t __ret; \
6027460274 __ret = (int16x8_t) __builtin_neon_vqrdmulhq_laneq_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 33); \
6027560275 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0); \
6027660276 __ret; \
......@@ -60279,19 +60279,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6027960279
6028060280#ifdef __LITTLE_ENDIAN__
6028160281#define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60282 int32x2_t __ret; \
6028260283 int32x2_t __s0 = __p0; \
6028360284 int32x4_t __s1 = __p1; \
60284 int32x2_t __ret; \
6028560285 __ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 2); \
6028660286 __ret; \
6028760287})
6028860288#else
6028960289#define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
60290 int32x2_t __ret; \
6029060291 int32x2_t __s0 = __p0; \
6029160292 int32x4_t __s1 = __p1; \
6029260293 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
6029360294 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 3, 2, 1, 0); \
60294 int32x2_t __ret; \
6029560295 __ret = (int32x2_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 2); \
6029660296 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
6029760297 __ret; \
......@@ -60300,19 +60300,19 @@ __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
6030060300
6030160301#ifdef __LITTLE_ENDIAN__
6030260302#define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60303 int16x4_t __ret; \
6030360304 int16x4_t __s0 = __p0; \
6030460305 int16x8_t __s1 = __p1; \
60305 int16x4_t __ret; \
6030660306 __ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__s0, (int8x16_t)__s1, __p2, 1); \
6030760307 __ret; \
6030860308})
6030960309#else
6031060310#define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
60311 int16x4_t __ret; \
6031160312 int16x4_t __s0 = __p0; \
6031260313 int16x8_t __s1 = __p1; \
6031360314 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 3, 2, 1, 0); \
6031460315 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 7, 6, 5, 4, 3, 2, 1, 0); \
60315 int16x4_t __ret; \
6031660316 __ret = (int16x4_t) __builtin_neon_vqrdmulh_laneq_v((int8x8_t)__rev0, (int8x16_t)__rev1, __p2, 1); \
6031760317 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0); \
6031860318 __ret; \
......@@ -60361,19 +60361,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6036160361}
6036260362#ifdef __LITTLE_ENDIAN__
6036360363#define vqrshrn_high_n_u32(__p0_771, __p1_771, __p2_771) __extension__ ({ \
60364 uint16x8_t __ret_771; \
6036460365 uint16x4_t __s0_771 = __p0_771; \
6036560366 uint32x4_t __s1_771 = __p1_771; \
60366 uint16x8_t __ret_771; \
6036760367 __ret_771 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_771), (uint16x4_t)(vqrshrn_n_u32(__s1_771, __p2_771)))); \
6036860368 __ret_771; \
6036960369})
6037060370#else
6037160371#define vqrshrn_high_n_u32(__p0_772, __p1_772, __p2_772) __extension__ ({ \
60372 uint16x8_t __ret_772; \
6037260373 uint16x4_t __s0_772 = __p0_772; \
6037360374 uint32x4_t __s1_772 = __p1_772; \
6037460375 uint16x4_t __rev0_772; __rev0_772 = __builtin_shufflevector(__s0_772, __s0_772, 3, 2, 1, 0); \
6037560376 uint32x4_t __rev1_772; __rev1_772 = __builtin_shufflevector(__s1_772, __s1_772, 3, 2, 1, 0); \
60376 uint16x8_t __ret_772; \
6037760377 __ret_772 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_772), (uint16x4_t)(__noswap_vqrshrn_n_u32(__rev1_772, __p2_772)))); \
6037860378 __ret_772 = __builtin_shufflevector(__ret_772, __ret_772, 7, 6, 5, 4, 3, 2, 1, 0); \
6037960379 __ret_772; \
......@@ -60382,19 +60382,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6038260382
6038360383#ifdef __LITTLE_ENDIAN__
6038460384#define vqrshrn_high_n_u64(__p0_773, __p1_773, __p2_773) __extension__ ({ \
60385 uint32x4_t __ret_773; \
6038560386 uint32x2_t __s0_773 = __p0_773; \
6038660387 uint64x2_t __s1_773 = __p1_773; \
60387 uint32x4_t __ret_773; \
6038860388 __ret_773 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_773), (uint32x2_t)(vqrshrn_n_u64(__s1_773, __p2_773)))); \
6038960389 __ret_773; \
6039060390})
6039160391#else
6039260392#define vqrshrn_high_n_u64(__p0_774, __p1_774, __p2_774) __extension__ ({ \
60393 uint32x4_t __ret_774; \
6039360394 uint32x2_t __s0_774 = __p0_774; \
6039460395 uint64x2_t __s1_774 = __p1_774; \
6039560396 uint32x2_t __rev0_774; __rev0_774 = __builtin_shufflevector(__s0_774, __s0_774, 1, 0); \
6039660397 uint64x2_t __rev1_774; __rev1_774 = __builtin_shufflevector(__s1_774, __s1_774, 1, 0); \
60397 uint32x4_t __ret_774; \
6039860398 __ret_774 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_774), (uint32x2_t)(__noswap_vqrshrn_n_u64(__rev1_774, __p2_774)))); \
6039960399 __ret_774 = __builtin_shufflevector(__ret_774, __ret_774, 3, 2, 1, 0); \
6040060400 __ret_774; \
......@@ -60403,19 +60403,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6040360403
6040460404#ifdef __LITTLE_ENDIAN__
6040560405#define vqrshrn_high_n_u16(__p0_775, __p1_775, __p2_775) __extension__ ({ \
60406 uint8x16_t __ret_775; \
6040660407 uint8x8_t __s0_775 = __p0_775; \
6040760408 uint16x8_t __s1_775 = __p1_775; \
60408 uint8x16_t __ret_775; \
6040960409 __ret_775 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_775), (uint8x8_t)(vqrshrn_n_u16(__s1_775, __p2_775)))); \
6041060410 __ret_775; \
6041160411})
6041260412#else
6041360413#define vqrshrn_high_n_u16(__p0_776, __p1_776, __p2_776) __extension__ ({ \
60414 uint8x16_t __ret_776; \
6041460415 uint8x8_t __s0_776 = __p0_776; \
6041560416 uint16x8_t __s1_776 = __p1_776; \
6041660417 uint8x8_t __rev0_776; __rev0_776 = __builtin_shufflevector(__s0_776, __s0_776, 7, 6, 5, 4, 3, 2, 1, 0); \
6041760418 uint16x8_t __rev1_776; __rev1_776 = __builtin_shufflevector(__s1_776, __s1_776, 7, 6, 5, 4, 3, 2, 1, 0); \
60418 uint8x16_t __ret_776; \
6041960419 __ret_776 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_776), (uint8x8_t)(__noswap_vqrshrn_n_u16(__rev1_776, __p2_776)))); \
6042060420 __ret_776 = __builtin_shufflevector(__ret_776, __ret_776, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6042160421 __ret_776; \
......@@ -60424,19 +60424,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6042460424
6042560425#ifdef __LITTLE_ENDIAN__
6042660426#define vqrshrn_high_n_s32(__p0_777, __p1_777, __p2_777) __extension__ ({ \
60427 int16x8_t __ret_777; \
6042760428 int16x4_t __s0_777 = __p0_777; \
6042860429 int32x4_t __s1_777 = __p1_777; \
60429 int16x8_t __ret_777; \
6043060430 __ret_777 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_777), (int16x4_t)(vqrshrn_n_s32(__s1_777, __p2_777)))); \
6043160431 __ret_777; \
6043260432})
6043360433#else
6043460434#define vqrshrn_high_n_s32(__p0_778, __p1_778, __p2_778) __extension__ ({ \
60435 int16x8_t __ret_778; \
6043560436 int16x4_t __s0_778 = __p0_778; \
6043660437 int32x4_t __s1_778 = __p1_778; \
6043760438 int16x4_t __rev0_778; __rev0_778 = __builtin_shufflevector(__s0_778, __s0_778, 3, 2, 1, 0); \
6043860439 int32x4_t __rev1_778; __rev1_778 = __builtin_shufflevector(__s1_778, __s1_778, 3, 2, 1, 0); \
60439 int16x8_t __ret_778; \
6044060440 __ret_778 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_778), (int16x4_t)(__noswap_vqrshrn_n_s32(__rev1_778, __p2_778)))); \
6044160441 __ret_778 = __builtin_shufflevector(__ret_778, __ret_778, 7, 6, 5, 4, 3, 2, 1, 0); \
6044260442 __ret_778; \
......@@ -60445,19 +60445,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6044560445
6044660446#ifdef __LITTLE_ENDIAN__
6044760447#define vqrshrn_high_n_s64(__p0_779, __p1_779, __p2_779) __extension__ ({ \
60448 int32x4_t __ret_779; \
6044860449 int32x2_t __s0_779 = __p0_779; \
6044960450 int64x2_t __s1_779 = __p1_779; \
60450 int32x4_t __ret_779; \
6045160451 __ret_779 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_779), (int32x2_t)(vqrshrn_n_s64(__s1_779, __p2_779)))); \
6045260452 __ret_779; \
6045360453})
6045460454#else
6045560455#define vqrshrn_high_n_s64(__p0_780, __p1_780, __p2_780) __extension__ ({ \
60456 int32x4_t __ret_780; \
6045660457 int32x2_t __s0_780 = __p0_780; \
6045760458 int64x2_t __s1_780 = __p1_780; \
6045860459 int32x2_t __rev0_780; __rev0_780 = __builtin_shufflevector(__s0_780, __s0_780, 1, 0); \
6045960460 int64x2_t __rev1_780; __rev1_780 = __builtin_shufflevector(__s1_780, __s1_780, 1, 0); \
60460 int32x4_t __ret_780; \
6046160461 __ret_780 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_780), (int32x2_t)(__noswap_vqrshrn_n_s64(__rev1_780, __p2_780)))); \
6046260462 __ret_780 = __builtin_shufflevector(__ret_780, __ret_780, 3, 2, 1, 0); \
6046360463 __ret_780; \
......@@ -60466,19 +60466,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6046660466
6046760467#ifdef __LITTLE_ENDIAN__
6046860468#define vqrshrn_high_n_s16(__p0_781, __p1_781, __p2_781) __extension__ ({ \
60469 int8x16_t __ret_781; \
6046960470 int8x8_t __s0_781 = __p0_781; \
6047060471 int16x8_t __s1_781 = __p1_781; \
60471 int8x16_t __ret_781; \
6047260472 __ret_781 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_781), (int8x8_t)(vqrshrn_n_s16(__s1_781, __p2_781)))); \
6047360473 __ret_781; \
6047460474})
6047560475#else
6047660476#define vqrshrn_high_n_s16(__p0_782, __p1_782, __p2_782) __extension__ ({ \
60477 int8x16_t __ret_782; \
6047760478 int8x8_t __s0_782 = __p0_782; \
6047860479 int16x8_t __s1_782 = __p1_782; \
6047960480 int8x8_t __rev0_782; __rev0_782 = __builtin_shufflevector(__s0_782, __s0_782, 7, 6, 5, 4, 3, 2, 1, 0); \
6048060481 int16x8_t __rev1_782; __rev1_782 = __builtin_shufflevector(__s1_782, __s1_782, 7, 6, 5, 4, 3, 2, 1, 0); \
60481 int8x16_t __ret_782; \
6048260482 __ret_782 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_782), (int8x8_t)(__noswap_vqrshrn_n_s16(__rev1_782, __p2_782)))); \
6048360483 __ret_782 = __builtin_shufflevector(__ret_782, __ret_782, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6048460484 __ret_782; \
......@@ -60486,56 +60486,56 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6048660486#endif
6048760487
6048860488#define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
60489 uint32_t __s0 = __p0; \
6049060489 uint16_t __ret; \
60490 uint32_t __s0 = __p0; \
6049160491 __ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
6049260492 __ret; \
6049360493})
6049460494#define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
60495 uint64_t __s0 = __p0; \
6049660495 uint32_t __ret; \
60496 uint64_t __s0 = __p0; \
6049760497 __ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
6049860498 __ret; \
6049960499})
6050060500#define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
60501 uint16_t __s0 = __p0; \
6050260501 uint8_t __ret; \
60502 uint16_t __s0 = __p0; \
6050360503 __ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
6050460504 __ret; \
6050560505})
6050660506#define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
60507 int32_t __s0 = __p0; \
6050860507 int16_t __ret; \
60508 int32_t __s0 = __p0; \
6050960509 __ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
6051060510 __ret; \
6051160511})
6051260512#define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
60513 int64_t __s0 = __p0; \
6051460513 int32_t __ret; \
60514 int64_t __s0 = __p0; \
6051560515 __ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
6051660516 __ret; \
6051760517})
6051860518#define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
60519 int16_t __s0 = __p0; \
6052060519 int8_t __ret; \
60520 int16_t __s0 = __p0; \
6052160521 __ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
6052260522 __ret; \
6052360523})
6052460524#ifdef __LITTLE_ENDIAN__
6052560525#define vqrshrun_high_n_s32(__p0_783, __p1_783, __p2_783) __extension__ ({ \
60526 int16x8_t __ret_783; \
6052660527 int16x4_t __s0_783 = __p0_783; \
6052760528 int32x4_t __s1_783 = __p1_783; \
60528 int16x8_t __ret_783; \
6052960529 __ret_783 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_783), (int16x4_t)(vqrshrun_n_s32(__s1_783, __p2_783)))); \
6053060530 __ret_783; \
6053160531})
6053260532#else
6053360533#define vqrshrun_high_n_s32(__p0_784, __p1_784, __p2_784) __extension__ ({ \
60534 int16x8_t __ret_784; \
6053460535 int16x4_t __s0_784 = __p0_784; \
6053560536 int32x4_t __s1_784 = __p1_784; \
6053660537 int16x4_t __rev0_784; __rev0_784 = __builtin_shufflevector(__s0_784, __s0_784, 3, 2, 1, 0); \
6053760538 int32x4_t __rev1_784; __rev1_784 = __builtin_shufflevector(__s1_784, __s1_784, 3, 2, 1, 0); \
60538 int16x8_t __ret_784; \
6053960539 __ret_784 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_784), (int16x4_t)(__noswap_vqrshrun_n_s32(__rev1_784, __p2_784)))); \
6054060540 __ret_784 = __builtin_shufflevector(__ret_784, __ret_784, 7, 6, 5, 4, 3, 2, 1, 0); \
6054160541 __ret_784; \
......@@ -60544,19 +60544,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6054460544
6054560545#ifdef __LITTLE_ENDIAN__
6054660546#define vqrshrun_high_n_s64(__p0_785, __p1_785, __p2_785) __extension__ ({ \
60547 int32x4_t __ret_785; \
6054760548 int32x2_t __s0_785 = __p0_785; \
6054860549 int64x2_t __s1_785 = __p1_785; \
60549 int32x4_t __ret_785; \
6055060550 __ret_785 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_785), (int32x2_t)(vqrshrun_n_s64(__s1_785, __p2_785)))); \
6055160551 __ret_785; \
6055260552})
6055360553#else
6055460554#define vqrshrun_high_n_s64(__p0_786, __p1_786, __p2_786) __extension__ ({ \
60555 int32x4_t __ret_786; \
6055560556 int32x2_t __s0_786 = __p0_786; \
6055660557 int64x2_t __s1_786 = __p1_786; \
6055760558 int32x2_t __rev0_786; __rev0_786 = __builtin_shufflevector(__s0_786, __s0_786, 1, 0); \
6055860559 int64x2_t __rev1_786; __rev1_786 = __builtin_shufflevector(__s1_786, __s1_786, 1, 0); \
60559 int32x4_t __ret_786; \
6056060560 __ret_786 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_786), (int32x2_t)(__noswap_vqrshrun_n_s64(__rev1_786, __p2_786)))); \
6056160561 __ret_786 = __builtin_shufflevector(__ret_786, __ret_786, 3, 2, 1, 0); \
6056260562 __ret_786; \
......@@ -60565,19 +60565,19 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6056560565
6056660566#ifdef __LITTLE_ENDIAN__
6056760567#define vqrshrun_high_n_s16(__p0_787, __p1_787, __p2_787) __extension__ ({ \
60568 int8x16_t __ret_787; \
6056860569 int8x8_t __s0_787 = __p0_787; \
6056960570 int16x8_t __s1_787 = __p1_787; \
60570 int8x16_t __ret_787; \
6057160571 __ret_787 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_787), (int8x8_t)(vqrshrun_n_s16(__s1_787, __p2_787)))); \
6057260572 __ret_787; \
6057360573})
6057460574#else
6057560575#define vqrshrun_high_n_s16(__p0_788, __p1_788, __p2_788) __extension__ ({ \
60576 int8x16_t __ret_788; \
6057660577 int8x8_t __s0_788 = __p0_788; \
6057760578 int16x8_t __s1_788 = __p1_788; \
6057860579 int8x8_t __rev0_788; __rev0_788 = __builtin_shufflevector(__s0_788, __s0_788, 7, 6, 5, 4, 3, 2, 1, 0); \
6057960580 int16x8_t __rev1_788; __rev1_788 = __builtin_shufflevector(__s1_788, __s1_788, 7, 6, 5, 4, 3, 2, 1, 0); \
60580 int8x16_t __ret_788; \
6058160581 __ret_788 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_788), (int8x8_t)(__noswap_vqrshrun_n_s16(__rev1_788, __p2_788)))); \
6058260582 __ret_788 = __builtin_shufflevector(__ret_788, __ret_788, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6058360583 __ret_788; \
......@@ -60585,20 +60585,20 @@ __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
6058560585#endif
6058660586
6058760587#define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
60588 int32_t __s0 = __p0; \
6058960588 int16_t __ret; \
60589 int32_t __s0 = __p0; \
6059060590 __ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
6059160591 __ret; \
6059260592})
6059360593#define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
60594 int64_t __s0 = __p0; \
6059560594 int32_t __ret; \
60595 int64_t __s0 = __p0; \
6059660596 __ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
6059760597 __ret; \
6059860598})
6059960599#define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
60600 int16_t __s0 = __p0; \
6060160600 int8_t __ret; \
60601 int16_t __s0 = __p0; \
6060260602 __ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
6060360603 __ret; \
6060460604})
......@@ -60643,92 +60643,92 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6064360643 return __ret;
6064460644}
6064560645#define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
60646 uint8_t __s0 = __p0; \
6064760646 uint8_t __ret; \
60647 uint8_t __s0 = __p0; \
6064860648 __ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
6064960649 __ret; \
6065060650})
6065160651#define vqshls_n_u32(__p0, __p1) __extension__ ({ \
60652 uint32_t __s0 = __p0; \
6065360652 uint32_t __ret; \
60653 uint32_t __s0 = __p0; \
6065460654 __ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
6065560655 __ret; \
6065660656})
6065760657#define vqshld_n_u64(__p0, __p1) __extension__ ({ \
60658 uint64_t __s0 = __p0; \
6065960658 uint64_t __ret; \
60659 uint64_t __s0 = __p0; \
6066060660 __ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
6066160661 __ret; \
6066260662})
6066360663#define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
60664 uint16_t __s0 = __p0; \
6066560664 uint16_t __ret; \
60665 uint16_t __s0 = __p0; \
6066660666 __ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
6066760667 __ret; \
6066860668})
6066960669#define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
60670 int8_t __s0 = __p0; \
6067160670 int8_t __ret; \
60671 int8_t __s0 = __p0; \
6067260672 __ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
6067360673 __ret; \
6067460674})
6067560675#define vqshls_n_s32(__p0, __p1) __extension__ ({ \
60676 int32_t __s0 = __p0; \
6067760676 int32_t __ret; \
60677 int32_t __s0 = __p0; \
6067860678 __ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
6067960679 __ret; \
6068060680})
6068160681#define vqshld_n_s64(__p0, __p1) __extension__ ({ \
60682 int64_t __s0 = __p0; \
6068360682 int64_t __ret; \
60683 int64_t __s0 = __p0; \
6068460684 __ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
6068560685 __ret; \
6068660686})
6068760687#define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
60688 int16_t __s0 = __p0; \
6068960688 int16_t __ret; \
60689 int16_t __s0 = __p0; \
6069060690 __ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
6069160691 __ret; \
6069260692})
6069360693#define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
60694 int8_t __s0 = __p0; \
6069560694 int8_t __ret; \
60695 int8_t __s0 = __p0; \
6069660696 __ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
6069760697 __ret; \
6069860698})
6069960699#define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
60700 int32_t __s0 = __p0; \
6070160700 int32_t __ret; \
60701 int32_t __s0 = __p0; \
6070260702 __ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
6070360703 __ret; \
6070460704})
6070560705#define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
60706 int64_t __s0 = __p0; \
6070760706 int64_t __ret; \
60707 int64_t __s0 = __p0; \
6070860708 __ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
6070960709 __ret; \
6071060710})
6071160711#define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
60712 int16_t __s0 = __p0; \
6071360712 int16_t __ret; \
60713 int16_t __s0 = __p0; \
6071460714 __ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
6071560715 __ret; \
6071660716})
6071760717#ifdef __LITTLE_ENDIAN__
6071860718#define vqshrn_high_n_u32(__p0_789, __p1_789, __p2_789) __extension__ ({ \
60719 uint16x8_t __ret_789; \
6071960720 uint16x4_t __s0_789 = __p0_789; \
6072060721 uint32x4_t __s1_789 = __p1_789; \
60721 uint16x8_t __ret_789; \
6072260722 __ret_789 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_789), (uint16x4_t)(vqshrn_n_u32(__s1_789, __p2_789)))); \
6072360723 __ret_789; \
6072460724})
6072560725#else
6072660726#define vqshrn_high_n_u32(__p0_790, __p1_790, __p2_790) __extension__ ({ \
60727 uint16x8_t __ret_790; \
6072760728 uint16x4_t __s0_790 = __p0_790; \
6072860729 uint32x4_t __s1_790 = __p1_790; \
6072960730 uint16x4_t __rev0_790; __rev0_790 = __builtin_shufflevector(__s0_790, __s0_790, 3, 2, 1, 0); \
6073060731 uint32x4_t __rev1_790; __rev1_790 = __builtin_shufflevector(__s1_790, __s1_790, 3, 2, 1, 0); \
60731 uint16x8_t __ret_790; \
6073260732 __ret_790 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_790), (uint16x4_t)(__noswap_vqshrn_n_u32(__rev1_790, __p2_790)))); \
6073360733 __ret_790 = __builtin_shufflevector(__ret_790, __ret_790, 7, 6, 5, 4, 3, 2, 1, 0); \
6073460734 __ret_790; \
......@@ -60737,19 +60737,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6073760737
6073860738#ifdef __LITTLE_ENDIAN__
6073960739#define vqshrn_high_n_u64(__p0_791, __p1_791, __p2_791) __extension__ ({ \
60740 uint32x4_t __ret_791; \
6074060741 uint32x2_t __s0_791 = __p0_791; \
6074160742 uint64x2_t __s1_791 = __p1_791; \
60742 uint32x4_t __ret_791; \
6074360743 __ret_791 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_791), (uint32x2_t)(vqshrn_n_u64(__s1_791, __p2_791)))); \
6074460744 __ret_791; \
6074560745})
6074660746#else
6074760747#define vqshrn_high_n_u64(__p0_792, __p1_792, __p2_792) __extension__ ({ \
60748 uint32x4_t __ret_792; \
6074860749 uint32x2_t __s0_792 = __p0_792; \
6074960750 uint64x2_t __s1_792 = __p1_792; \
6075060751 uint32x2_t __rev0_792; __rev0_792 = __builtin_shufflevector(__s0_792, __s0_792, 1, 0); \
6075160752 uint64x2_t __rev1_792; __rev1_792 = __builtin_shufflevector(__s1_792, __s1_792, 1, 0); \
60752 uint32x4_t __ret_792; \
6075360753 __ret_792 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_792), (uint32x2_t)(__noswap_vqshrn_n_u64(__rev1_792, __p2_792)))); \
6075460754 __ret_792 = __builtin_shufflevector(__ret_792, __ret_792, 3, 2, 1, 0); \
6075560755 __ret_792; \
......@@ -60758,19 +60758,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6075860758
6075960759#ifdef __LITTLE_ENDIAN__
6076060760#define vqshrn_high_n_u16(__p0_793, __p1_793, __p2_793) __extension__ ({ \
60761 uint8x16_t __ret_793; \
6076160762 uint8x8_t __s0_793 = __p0_793; \
6076260763 uint16x8_t __s1_793 = __p1_793; \
60763 uint8x16_t __ret_793; \
6076460764 __ret_793 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_793), (uint8x8_t)(vqshrn_n_u16(__s1_793, __p2_793)))); \
6076560765 __ret_793; \
6076660766})
6076760767#else
6076860768#define vqshrn_high_n_u16(__p0_794, __p1_794, __p2_794) __extension__ ({ \
60769 uint8x16_t __ret_794; \
6076960770 uint8x8_t __s0_794 = __p0_794; \
6077060771 uint16x8_t __s1_794 = __p1_794; \
6077160772 uint8x8_t __rev0_794; __rev0_794 = __builtin_shufflevector(__s0_794, __s0_794, 7, 6, 5, 4, 3, 2, 1, 0); \
6077260773 uint16x8_t __rev1_794; __rev1_794 = __builtin_shufflevector(__s1_794, __s1_794, 7, 6, 5, 4, 3, 2, 1, 0); \
60773 uint8x16_t __ret_794; \
6077460774 __ret_794 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_794), (uint8x8_t)(__noswap_vqshrn_n_u16(__rev1_794, __p2_794)))); \
6077560775 __ret_794 = __builtin_shufflevector(__ret_794, __ret_794, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6077660776 __ret_794; \
......@@ -60779,19 +60779,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6077960779
6078060780#ifdef __LITTLE_ENDIAN__
6078160781#define vqshrn_high_n_s32(__p0_795, __p1_795, __p2_795) __extension__ ({ \
60782 int16x8_t __ret_795; \
6078260783 int16x4_t __s0_795 = __p0_795; \
6078360784 int32x4_t __s1_795 = __p1_795; \
60784 int16x8_t __ret_795; \
6078560785 __ret_795 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_795), (int16x4_t)(vqshrn_n_s32(__s1_795, __p2_795)))); \
6078660786 __ret_795; \
6078760787})
6078860788#else
6078960789#define vqshrn_high_n_s32(__p0_796, __p1_796, __p2_796) __extension__ ({ \
60790 int16x8_t __ret_796; \
6079060791 int16x4_t __s0_796 = __p0_796; \
6079160792 int32x4_t __s1_796 = __p1_796; \
6079260793 int16x4_t __rev0_796; __rev0_796 = __builtin_shufflevector(__s0_796, __s0_796, 3, 2, 1, 0); \
6079360794 int32x4_t __rev1_796; __rev1_796 = __builtin_shufflevector(__s1_796, __s1_796, 3, 2, 1, 0); \
60794 int16x8_t __ret_796; \
6079560795 __ret_796 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_796), (int16x4_t)(__noswap_vqshrn_n_s32(__rev1_796, __p2_796)))); \
6079660796 __ret_796 = __builtin_shufflevector(__ret_796, __ret_796, 7, 6, 5, 4, 3, 2, 1, 0); \
6079760797 __ret_796; \
......@@ -60800,19 +60800,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6080060800
6080160801#ifdef __LITTLE_ENDIAN__
6080260802#define vqshrn_high_n_s64(__p0_797, __p1_797, __p2_797) __extension__ ({ \
60803 int32x4_t __ret_797; \
6080360804 int32x2_t __s0_797 = __p0_797; \
6080460805 int64x2_t __s1_797 = __p1_797; \
60805 int32x4_t __ret_797; \
6080660806 __ret_797 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_797), (int32x2_t)(vqshrn_n_s64(__s1_797, __p2_797)))); \
6080760807 __ret_797; \
6080860808})
6080960809#else
6081060810#define vqshrn_high_n_s64(__p0_798, __p1_798, __p2_798) __extension__ ({ \
60811 int32x4_t __ret_798; \
6081160812 int32x2_t __s0_798 = __p0_798; \
6081260813 int64x2_t __s1_798 = __p1_798; \
6081360814 int32x2_t __rev0_798; __rev0_798 = __builtin_shufflevector(__s0_798, __s0_798, 1, 0); \
6081460815 int64x2_t __rev1_798; __rev1_798 = __builtin_shufflevector(__s1_798, __s1_798, 1, 0); \
60815 int32x4_t __ret_798; \
6081660816 __ret_798 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_798), (int32x2_t)(__noswap_vqshrn_n_s64(__rev1_798, __p2_798)))); \
6081760817 __ret_798 = __builtin_shufflevector(__ret_798, __ret_798, 3, 2, 1, 0); \
6081860818 __ret_798; \
......@@ -60821,19 +60821,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6082160821
6082260822#ifdef __LITTLE_ENDIAN__
6082360823#define vqshrn_high_n_s16(__p0_799, __p1_799, __p2_799) __extension__ ({ \
60824 int8x16_t __ret_799; \
6082460825 int8x8_t __s0_799 = __p0_799; \
6082560826 int16x8_t __s1_799 = __p1_799; \
60826 int8x16_t __ret_799; \
6082760827 __ret_799 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_799), (int8x8_t)(vqshrn_n_s16(__s1_799, __p2_799)))); \
6082860828 __ret_799; \
6082960829})
6083060830#else
6083160831#define vqshrn_high_n_s16(__p0_800, __p1_800, __p2_800) __extension__ ({ \
60832 int8x16_t __ret_800; \
6083260833 int8x8_t __s0_800 = __p0_800; \
6083360834 int16x8_t __s1_800 = __p1_800; \
6083460835 int8x8_t __rev0_800; __rev0_800 = __builtin_shufflevector(__s0_800, __s0_800, 7, 6, 5, 4, 3, 2, 1, 0); \
6083560836 int16x8_t __rev1_800; __rev1_800 = __builtin_shufflevector(__s1_800, __s1_800, 7, 6, 5, 4, 3, 2, 1, 0); \
60836 int8x16_t __ret_800; \
6083760837 __ret_800 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_800), (int8x8_t)(__noswap_vqshrn_n_s16(__rev1_800, __p2_800)))); \
6083860838 __ret_800 = __builtin_shufflevector(__ret_800, __ret_800, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6083960839 __ret_800; \
......@@ -60841,56 +60841,56 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6084160841#endif
6084260842
6084360843#define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
60844 uint32_t __s0 = __p0; \
6084560844 uint16_t __ret; \
60845 uint32_t __s0 = __p0; \
6084660846 __ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
6084760847 __ret; \
6084860848})
6084960849#define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
60850 uint64_t __s0 = __p0; \
6085160850 uint32_t __ret; \
60851 uint64_t __s0 = __p0; \
6085260852 __ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
6085360853 __ret; \
6085460854})
6085560855#define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
60856 uint16_t __s0 = __p0; \
6085760856 uint8_t __ret; \
60857 uint16_t __s0 = __p0; \
6085860858 __ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
6085960859 __ret; \
6086060860})
6086160861#define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
60862 int32_t __s0 = __p0; \
6086360862 int16_t __ret; \
60863 int32_t __s0 = __p0; \
6086460864 __ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
6086560865 __ret; \
6086660866})
6086760867#define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
60868 int64_t __s0 = __p0; \
6086960868 int32_t __ret; \
60869 int64_t __s0 = __p0; \
6087060870 __ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
6087160871 __ret; \
6087260872})
6087360873#define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
60874 int16_t __s0 = __p0; \
6087560874 int8_t __ret; \
60875 int16_t __s0 = __p0; \
6087660876 __ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
6087760877 __ret; \
6087860878})
6087960879#ifdef __LITTLE_ENDIAN__
6088060880#define vqshrun_high_n_s32(__p0_801, __p1_801, __p2_801) __extension__ ({ \
60881 int16x8_t __ret_801; \
6088160882 int16x4_t __s0_801 = __p0_801; \
6088260883 int32x4_t __s1_801 = __p1_801; \
60883 int16x8_t __ret_801; \
6088460884 __ret_801 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_801), (int16x4_t)(vqshrun_n_s32(__s1_801, __p2_801)))); \
6088560885 __ret_801; \
6088660886})
6088760887#else
6088860888#define vqshrun_high_n_s32(__p0_802, __p1_802, __p2_802) __extension__ ({ \
60889 int16x8_t __ret_802; \
6088960890 int16x4_t __s0_802 = __p0_802; \
6089060891 int32x4_t __s1_802 = __p1_802; \
6089160892 int16x4_t __rev0_802; __rev0_802 = __builtin_shufflevector(__s0_802, __s0_802, 3, 2, 1, 0); \
6089260893 int32x4_t __rev1_802; __rev1_802 = __builtin_shufflevector(__s1_802, __s1_802, 3, 2, 1, 0); \
60893 int16x8_t __ret_802; \
6089460894 __ret_802 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_802), (int16x4_t)(__noswap_vqshrun_n_s32(__rev1_802, __p2_802)))); \
6089560895 __ret_802 = __builtin_shufflevector(__ret_802, __ret_802, 7, 6, 5, 4, 3, 2, 1, 0); \
6089660896 __ret_802; \
......@@ -60899,19 +60899,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6089960899
6090060900#ifdef __LITTLE_ENDIAN__
6090160901#define vqshrun_high_n_s64(__p0_803, __p1_803, __p2_803) __extension__ ({ \
60902 int32x4_t __ret_803; \
6090260903 int32x2_t __s0_803 = __p0_803; \
6090360904 int64x2_t __s1_803 = __p1_803; \
60904 int32x4_t __ret_803; \
6090560905 __ret_803 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_803), (int32x2_t)(vqshrun_n_s64(__s1_803, __p2_803)))); \
6090660906 __ret_803; \
6090760907})
6090860908#else
6090960909#define vqshrun_high_n_s64(__p0_804, __p1_804, __p2_804) __extension__ ({ \
60910 int32x4_t __ret_804; \
6091060911 int32x2_t __s0_804 = __p0_804; \
6091160912 int64x2_t __s1_804 = __p1_804; \
6091260913 int32x2_t __rev0_804; __rev0_804 = __builtin_shufflevector(__s0_804, __s0_804, 1, 0); \
6091360914 int64x2_t __rev1_804; __rev1_804 = __builtin_shufflevector(__s1_804, __s1_804, 1, 0); \
60914 int32x4_t __ret_804; \
6091560915 __ret_804 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_804), (int32x2_t)(__noswap_vqshrun_n_s64(__rev1_804, __p2_804)))); \
6091660916 __ret_804 = __builtin_shufflevector(__ret_804, __ret_804, 3, 2, 1, 0); \
6091760917 __ret_804; \
......@@ -60920,19 +60920,19 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6092060920
6092160921#ifdef __LITTLE_ENDIAN__
6092260922#define vqshrun_high_n_s16(__p0_805, __p1_805, __p2_805) __extension__ ({ \
60923 int8x16_t __ret_805; \
6092360924 int8x8_t __s0_805 = __p0_805; \
6092460925 int16x8_t __s1_805 = __p1_805; \
60925 int8x16_t __ret_805; \
6092660926 __ret_805 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_805), (int8x8_t)(vqshrun_n_s16(__s1_805, __p2_805)))); \
6092760927 __ret_805; \
6092860928})
6092960929#else
6093060930#define vqshrun_high_n_s16(__p0_806, __p1_806, __p2_806) __extension__ ({ \
60931 int8x16_t __ret_806; \
6093160932 int8x8_t __s0_806 = __p0_806; \
6093260933 int16x8_t __s1_806 = __p1_806; \
6093360934 int8x8_t __rev0_806; __rev0_806 = __builtin_shufflevector(__s0_806, __s0_806, 7, 6, 5, 4, 3, 2, 1, 0); \
6093460935 int16x8_t __rev1_806; __rev1_806 = __builtin_shufflevector(__s1_806, __s1_806, 7, 6, 5, 4, 3, 2, 1, 0); \
60935 int8x16_t __ret_806; \
6093660936 __ret_806 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_806), (int8x8_t)(__noswap_vqshrun_n_s16(__rev1_806, __p2_806)))); \
6093760937 __ret_806 = __builtin_shufflevector(__ret_806, __ret_806, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6093860938 __ret_806; \
......@@ -60940,20 +60940,20 @@ __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
6094060940#endif
6094160941
6094260942#define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
60943 int32_t __s0 = __p0; \
6094460943 int16_t __ret; \
60944 int32_t __s0 = __p0; \
6094560945 __ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
6094660946 __ret; \
6094760947})
6094860948#define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
60949 int64_t __s0 = __p0; \
6095060949 int32_t __ret; \
60950 int64_t __s0 = __p0; \
6095160951 __ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
6095260952 __ret; \
6095360953})
6095460954#define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
60955 int16_t __s0 = __p0; \
6095660955 int8_t __ret; \
60956 int16_t __s0 = __p0; \
6095760957 __ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
6095860958 __ret; \
6095960959})
......@@ -61005,9 +61005,9 @@ __ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
6100561005}
6100661006#else
6100761007__ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
61008 poly8x8_t __ret;
6100861009 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6100961010 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61010 poly8x8_t __ret;
6101161011 __ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 4);
6101261012 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6101361013 return __ret;
......@@ -61022,9 +61022,9 @@ __ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
6102261022}
6102361023#else
6102461024__ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
61025 poly8x16_t __ret;
6102561026 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6102661027 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61027 poly8x16_t __ret;
6102861028 __ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 36);
6102961029 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6103061030 return __ret;
......@@ -61039,9 +61039,9 @@ __ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6103961039}
6104061040#else
6104161041__ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
61042 uint8x16_t __ret;
6104261043 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6104361044 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61044 uint8x16_t __ret;
6104561045 __ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
6104661046 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6104761047 return __ret;
......@@ -61056,9 +61056,9 @@ __ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
6105661056}
6105761057#else
6105861058__ai int8x16_t vqtbl1q_s8(int8x16_t __p0, uint8x16_t __p1) {
61059 int8x16_t __ret;
6105961060 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6106061061 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61061 int8x16_t __ret;
6106261062 __ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
6106361063 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6106461064 return __ret;
......@@ -61073,9 +61073,9 @@ __ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
6107361073}
6107461074#else
6107561075__ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
61076 uint8x8_t __ret;
6107661077 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6107761078 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61078 uint8x8_t __ret;
6107961079 __ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 16);
6108061080 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6108161081 return __ret;
......@@ -61090,9 +61090,9 @@ __ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
6109061090}
6109161091#else
6109261092__ai int8x8_t vqtbl1_s8(int8x16_t __p0, uint8x8_t __p1) {
61093 int8x8_t __ret;
6109361094 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6109461095 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61095 int8x8_t __ret;
6109661096 __ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__rev0, (int8x8_t)__rev1, 0);
6109761097 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6109861098 return __ret;
......@@ -61107,11 +61107,11 @@ __ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
6110761107}
6110861108#else
6110961109__ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
61110 poly8x8_t __ret;
6111061111 poly8x16x2_t __rev0;
6111161112 __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);
6111261113 __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);
6111361114 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61114 poly8x8_t __ret;
6111561115 __ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 4);
6111661116 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6111761117 return __ret;
......@@ -61126,11 +61126,11 @@ __ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
6112661126}
6112761127#else
6112861128__ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
61129 poly8x16_t __ret;
6112961130 poly8x16x2_t __rev0;
6113061131 __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);
6113161132 __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);
6113261133 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61133 poly8x16_t __ret;
6113461134 __ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 36);
6113561135 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6113661136 return __ret;
......@@ -61145,11 +61145,11 @@ __ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
6114561145}
6114661146#else
6114761147__ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
61148 uint8x16_t __ret;
6114861149 uint8x16x2_t __rev0;
6114961150 __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);
6115061151 __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);
6115161152 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61152 uint8x16_t __ret;
6115361153 __ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 48);
6115461154 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6115561155 return __ret;
......@@ -61164,11 +61164,11 @@ __ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
6116461164}
6116561165#else
6116661166__ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, uint8x16_t __p1) {
61167 int8x16_t __ret;
6116761168 int8x16x2_t __rev0;
6116861169 __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);
6116961170 __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);
6117061171 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61171 int8x16_t __ret;
6117261172 __ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x16_t)__rev1, 32);
6117361173 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6117461174 return __ret;
......@@ -61183,11 +61183,11 @@ __ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
6118361183}
6118461184#else
6118561185__ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
61186 uint8x8_t __ret;
6118661187 uint8x16x2_t __rev0;
6118761188 __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);
6118861189 __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);
6118961190 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61190 uint8x8_t __ret;
6119161191 __ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 16);
6119261192 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6119361193 return __ret;
......@@ -61202,11 +61202,11 @@ __ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
6120261202}
6120361203#else
6120461204__ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, uint8x8_t __p1) {
61205 int8x8_t __ret;
6120561206 int8x16x2_t __rev0;
6120661207 __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);
6120761208 __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);
6120861209 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61209 int8x8_t __ret;
6121061210 __ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__rev0.val[0], (int8x16_t)__rev0.val[1], (int8x8_t)__rev1, 0);
6121161211 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6121261212 return __ret;
......@@ -61221,12 +61221,12 @@ __ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
6122161221}
6122261222#else
6122361223__ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
61224 poly8x8_t __ret;
6122461225 poly8x16x3_t __rev0;
6122561226 __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);
6122661227 __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);
6122761228 __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);
6122861229 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61229 poly8x8_t __ret;
6123061230 __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);
6123161231 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6123261232 return __ret;
......@@ -61241,12 +61241,12 @@ __ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
6124161241}
6124261242#else
6124361243__ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
61244 poly8x16_t __ret;
6124461245 poly8x16x3_t __rev0;
6124561246 __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);
6124661247 __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);
6124761248 __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);
6124861249 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61249 poly8x16_t __ret;
6125061250 __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);
6125161251 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6125261252 return __ret;
......@@ -61261,12 +61261,12 @@ __ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
6126161261}
6126261262#else
6126361263__ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
61264 uint8x16_t __ret;
6126461265 uint8x16x3_t __rev0;
6126561266 __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);
6126661267 __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);
6126761268 __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);
6126861269 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61269 uint8x16_t __ret;
6127061270 __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);
6127161271 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6127261272 return __ret;
......@@ -61281,12 +61281,12 @@ __ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
6128161281}
6128261282#else
6128361283__ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, uint8x16_t __p1) {
61284 int8x16_t __ret;
6128461285 int8x16x3_t __rev0;
6128561286 __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);
6128661287 __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);
6128761288 __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);
6128861289 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61289 int8x16_t __ret;
6129061290 __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);
6129161291 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6129261292 return __ret;
......@@ -61301,12 +61301,12 @@ __ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
6130161301}
6130261302#else
6130361303__ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
61304 uint8x8_t __ret;
6130461305 uint8x16x3_t __rev0;
6130561306 __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);
6130661307 __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);
6130761308 __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);
6130861309 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61309 uint8x8_t __ret;
6131061310 __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);
6131161311 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6131261312 return __ret;
......@@ -61321,12 +61321,12 @@ __ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
6132161321}
6132261322#else
6132361323__ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, uint8x8_t __p1) {
61324 int8x8_t __ret;
6132461325 int8x16x3_t __rev0;
6132561326 __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);
6132661327 __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);
6132761328 __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);
6132861329 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61329 int8x8_t __ret;
6133061330 __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);
6133161331 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6133261332 return __ret;
......@@ -61341,13 +61341,13 @@ __ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
6134161341}
6134261342#else
6134361343__ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
61344 poly8x8_t __ret;
6134461345 poly8x16x4_t __rev0;
6134561346 __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);
6134661347 __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);
6134761348 __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);
6134861349 __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);
6134961350 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61350 poly8x8_t __ret;
6135161351 __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);
6135261352 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6135361353 return __ret;
......@@ -61362,13 +61362,13 @@ __ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
6136261362}
6136361363#else
6136461364__ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
61365 poly8x16_t __ret;
6136561366 poly8x16x4_t __rev0;
6136661367 __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);
6136761368 __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);
6136861369 __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);
6136961370 __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);
6137061371 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61371 poly8x16_t __ret;
6137261372 __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);
6137361373 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6137461374 return __ret;
......@@ -61383,13 +61383,13 @@ __ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
6138361383}
6138461384#else
6138561385__ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
61386 uint8x16_t __ret;
6138661387 uint8x16x4_t __rev0;
6138761388 __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);
6138861389 __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);
6138961390 __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);
6139061391 __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);
6139161392 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61392 uint8x16_t __ret;
6139361393 __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);
6139461394 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6139561395 return __ret;
......@@ -61404,13 +61404,13 @@ __ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
6140461404}
6140561405#else
6140661406__ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, uint8x16_t __p1) {
61407 int8x16_t __ret;
6140761408 int8x16x4_t __rev0;
6140861409 __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);
6140961410 __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);
6141061411 __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);
6141161412 __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);
6141261413 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61413 int8x16_t __ret;
6141461414 __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);
6141561415 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6141661416 return __ret;
......@@ -61425,13 +61425,13 @@ __ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
6142561425}
6142661426#else
6142761427__ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
61428 uint8x8_t __ret;
6142861429 uint8x16x4_t __rev0;
6142961430 __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);
6143061431 __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);
6143161432 __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);
6143261433 __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);
6143361434 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61434 uint8x8_t __ret;
6143561435 __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);
6143661436 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6143761437 return __ret;
......@@ -61446,13 +61446,13 @@ __ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
6144661446}
6144761447#else
6144861448__ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, uint8x8_t __p1) {
61449 int8x8_t __ret;
6144961450 int8x16x4_t __rev0;
6145061451 __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);
6145161452 __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);
6145261453 __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);
6145361454 __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);
6145461455 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61455 int8x8_t __ret;
6145661456 __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);
6145761457 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6145861458 return __ret;
......@@ -61467,10 +61467,10 @@ __ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
6146761467}
6146861468#else
6146961469__ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
61470 poly8x8_t __ret;
6147061471 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6147161472 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6147261473 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61473 poly8x8_t __ret;
6147461474 __ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 4);
6147561475 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6147661476 return __ret;
......@@ -61485,10 +61485,10 @@ __ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
6148561485}
6148661486#else
6148761487__ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
61488 poly8x16_t __ret;
6148861489 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6148961490 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6149061491 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61491 poly8x16_t __ret;
6149261492 __ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 36);
6149361493 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6149461494 return __ret;
......@@ -61503,10 +61503,10 @@ __ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
6150361503}
6150461504#else
6150561505__ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
61506 uint8x16_t __ret;
6150661507 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6150761508 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6150861509 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61509 uint8x16_t __ret;
6151061510 __ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 48);
6151161511 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6151261512 return __ret;
......@@ -61521,10 +61521,10 @@ __ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
6152161521}
6152261522#else
6152361523__ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, uint8x16_t __p2) {
61524 int8x16_t __ret;
6152461525 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6152561526 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6152661527 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61527 int8x16_t __ret;
6152861528 __ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__rev0, (int8x16_t)__rev1, (int8x16_t)__rev2, 32);
6152961529 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6153061530 return __ret;
......@@ -61539,10 +61539,10 @@ __ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
6153961539}
6154061540#else
6154161541__ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
61542 uint8x8_t __ret;
6154261543 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6154361544 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6154461545 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61545 uint8x8_t __ret;
6154661546 __ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 16);
6154761547 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6154861548 return __ret;
......@@ -61557,10 +61557,10 @@ __ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
6155761557}
6155861558#else
6155961559__ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, uint8x8_t __p2) {
61560 int8x8_t __ret;
6156061561 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6156161562 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6156261563 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61563 int8x8_t __ret;
6156461564 __ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__rev0, (int8x16_t)__rev1, (int8x8_t)__rev2, 0);
6156561565 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6156661566 return __ret;
......@@ -61575,12 +61575,12 @@ __ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
6157561575}
6157661576#else
6157761577__ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
61578 poly8x8_t __ret;
6157861579 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6157961580 poly8x16x2_t __rev1;
6158061581 __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);
6158161582 __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);
6158261583 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61583 poly8x8_t __ret;
6158461584 __ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 4);
6158561585 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6158661586 return __ret;
......@@ -61595,12 +61595,12 @@ __ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2)
6159561595}
6159661596#else
6159761597__ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
61598 poly8x16_t __ret;
6159861599 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6159961600 poly8x16x2_t __rev1;
6160061601 __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);
6160161602 __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);
6160261603 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61603 poly8x16_t __ret;
6160461604 __ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 36);
6160561605 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6160661606 return __ret;
......@@ -61615,12 +61615,12 @@ __ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2)
6161561615}
6161661616#else
6161761617__ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
61618 uint8x16_t __ret;
6161861619 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6161961620 uint8x16x2_t __rev1;
6162061621 __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);
6162161622 __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);
6162261623 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61623 uint8x16_t __ret;
6162461624 __ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 48);
6162561625 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6162661626 return __ret;
......@@ -61635,12 +61635,12 @@ __ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
6163561635}
6163661636#else
6163761637__ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, uint8x16_t __p2) {
61638 int8x16_t __ret;
6163861639 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6163961640 int8x16x2_t __rev1;
6164061641 __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);
6164161642 __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);
6164261643 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61643 int8x16_t __ret;
6164461644 __ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x16_t)__rev2, 32);
6164561645 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6164661646 return __ret;
......@@ -61655,12 +61655,12 @@ __ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
6165561655}
6165661656#else
6165761657__ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
61658 uint8x8_t __ret;
6165861659 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6165961660 uint8x16x2_t __rev1;
6166061661 __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);
6166161662 __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);
6166261663 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61663 uint8x8_t __ret;
6166461664 __ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 16);
6166561665 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6166661666 return __ret;
......@@ -61675,12 +61675,12 @@ __ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
6167561675}
6167661676#else
6167761677__ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, uint8x8_t __p2) {
61678 int8x8_t __ret;
6167861679 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6167961680 int8x16x2_t __rev1;
6168061681 __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);
6168161682 __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);
6168261683 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61683 int8x8_t __ret;
6168461684 __ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__rev0, (int8x16_t)__rev1.val[0], (int8x16_t)__rev1.val[1], (int8x8_t)__rev2, 0);
6168561685 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6168661686 return __ret;
......@@ -61695,13 +61695,13 @@ __ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
6169561695}
6169661696#else
6169761697__ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
61698 poly8x8_t __ret;
6169861699 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6169961700 poly8x16x3_t __rev1;
6170061701 __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);
6170161702 __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);
6170261703 __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);
6170361704 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61704 poly8x8_t __ret;
6170561705 __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);
6170661706 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6170761707 return __ret;
......@@ -61716,13 +61716,13 @@ __ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2)
6171661716}
6171761717#else
6171861718__ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
61719 poly8x16_t __ret;
6171961720 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6172061721 poly8x16x3_t __rev1;
6172161722 __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);
6172261723 __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);
6172361724 __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);
6172461725 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61725 poly8x16_t __ret;
6172661726 __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);
6172761727 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6172861728 return __ret;
......@@ -61737,13 +61737,13 @@ __ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2)
6173761737}
6173861738#else
6173961739__ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
61740 uint8x16_t __ret;
6174061741 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6174161742 uint8x16x3_t __rev1;
6174261743 __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);
6174361744 __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);
6174461745 __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);
6174561746 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61746 uint8x16_t __ret;
6174761747 __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);
6174861748 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6174961749 return __ret;
......@@ -61758,13 +61758,13 @@ __ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
6175861758}
6175961759#else
6176061760__ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, uint8x16_t __p2) {
61761 int8x16_t __ret;
6176161762 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6176261763 int8x16x3_t __rev1;
6176361764 __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);
6176461765 __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);
6176561766 __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);
6176661767 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61767 int8x16_t __ret;
6176861768 __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);
6176961769 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6177061770 return __ret;
......@@ -61779,13 +61779,13 @@ __ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
6177961779}
6178061780#else
6178161781__ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
61782 uint8x8_t __ret;
6178261783 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6178361784 uint8x16x3_t __rev1;
6178461785 __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);
6178561786 __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);
6178661787 __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);
6178761788 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61788 uint8x8_t __ret;
6178961789 __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);
6179061790 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6179161791 return __ret;
......@@ -61800,13 +61800,13 @@ __ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
6180061800}
6180161801#else
6180261802__ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, uint8x8_t __p2) {
61803 int8x8_t __ret;
6180361804 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6180461805 int8x16x3_t __rev1;
6180561806 __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);
6180661807 __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);
6180761808 __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);
6180861809 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61809 int8x8_t __ret;
6181061810 __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);
6181161811 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6181261812 return __ret;
......@@ -61821,6 +61821,7 @@ __ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
6182161821}
6182261822#else
6182361823__ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
61824 poly8x8_t __ret;
6182461825 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6182561826 poly8x16x4_t __rev1;
6182661827 __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);
......@@ -61828,7 +61829,6 @@ __ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
6182861829 __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);
6182961830 __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);
6183061831 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61831 poly8x8_t __ret;
6183261832 __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);
6183361833 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6183461834 return __ret;
......@@ -61843,6 +61843,7 @@ __ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2)
6184361843}
6184461844#else
6184561845__ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
61846 poly8x16_t __ret;
6184661847 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6184761848 poly8x16x4_t __rev1;
6184861849 __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);
......@@ -61850,7 +61851,6 @@ __ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2)
6185061851 __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);
6185161852 __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);
6185261853 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61853 poly8x16_t __ret;
6185461854 __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);
6185561855 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6185661856 return __ret;
......@@ -61865,6 +61865,7 @@ __ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2)
6186561865}
6186661866#else
6186761867__ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
61868 uint8x16_t __ret;
6186861869 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6186961870 uint8x16x4_t __rev1;
6187061871 __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);
......@@ -61872,7 +61873,6 @@ __ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2)
6187261873 __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);
6187361874 __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);
6187461875 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61875 uint8x16_t __ret;
6187661876 __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);
6187761877 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6187861878 return __ret;
......@@ -61887,6 +61887,7 @@ __ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
6188761887}
6188861888#else
6188961889__ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
61890 int8x16_t __ret;
6189061891 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6189161892 int8x16x4_t __rev1;
6189261893 __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);
......@@ -61894,7 +61895,6 @@ __ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, uint8x16_t __p2) {
6189461895 __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);
6189561896 __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);
6189661897 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61897 int8x16_t __ret;
6189861898 __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);
6189961899 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6190061900 return __ret;
......@@ -61909,6 +61909,7 @@ __ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
6190961909}
6191061910#else
6191161911__ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
61912 uint8x8_t __ret;
6191261913 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6191361914 uint8x16x4_t __rev1;
6191461915 __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);
......@@ -61916,7 +61917,6 @@ __ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
6191661917 __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);
6191761918 __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);
6191861919 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61919 uint8x8_t __ret;
6192061920 __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);
6192161921 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6192261922 return __ret;
......@@ -61931,6 +61931,7 @@ __ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
6193161931}
6193261932#else
6193361933__ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
61934 int8x8_t __ret;
6193461935 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6193561936 int8x16x4_t __rev1;
6193661937 __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);
......@@ -61938,7 +61939,6 @@ __ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, uint8x8_t __p2) {
6193861939 __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);
6193961940 __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);
6194061941 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61941 int8x8_t __ret;
6194261942 __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);
6194361943 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6194461944 return __ret;
......@@ -61953,10 +61953,10 @@ __ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __
6195361953}
6195461954#else
6195561955__ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
61956 uint16x8_t __ret;
6195661957 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6195761958 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6195861959 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
61959 uint16x8_t __ret;
6196061960 __ret = __noswap_vcombine_u16(__rev0, __noswap_vraddhn_u32(__rev1, __rev2));
6196161961 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6196261962 return __ret;
......@@ -61971,10 +61971,10 @@ __ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __
6197161971}
6197261972#else
6197361973__ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
61974 uint32x4_t __ret;
6197461975 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6197561976 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6197661977 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
61977 uint32x4_t __ret;
6197861978 __ret = __noswap_vcombine_u32(__rev0, __noswap_vraddhn_u64(__rev1, __rev2));
6197961979 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6198061980 return __ret;
......@@ -61989,10 +61989,10 @@ __ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p
6198961989}
6199061990#else
6199161991__ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
61992 uint8x16_t __ret;
6199261993 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6199361994 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6199461995 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
61995 uint8x16_t __ret;
6199661996 __ret = __noswap_vcombine_u8(__rev0, __noswap_vraddhn_u16(__rev1, __rev2));
6199761997 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6199861998 return __ret;
......@@ -62007,10 +62007,10 @@ __ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2)
6200762007}
6200862008#else
6200962009__ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
62010 int16x8_t __ret;
6201062011 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6201162012 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6201262013 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
62013 int16x8_t __ret;
6201462014 __ret = __noswap_vcombine_s16(__rev0, __noswap_vraddhn_s32(__rev1, __rev2));
6201562015 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6201662016 return __ret;
......@@ -62025,10 +62025,10 @@ __ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2)
6202562025}
6202662026#else
6202762027__ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
62028 int32x4_t __ret;
6202862029 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6202962030 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6203062031 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
62031 int32x4_t __ret;
6203262032 __ret = __noswap_vcombine_s32(__rev0, __noswap_vraddhn_s64(__rev1, __rev2));
6203362033 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6203462034 return __ret;
......@@ -62043,10 +62043,10 @@ __ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
6204362043}
6204462044#else
6204562045__ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
62046 int8x16_t __ret;
6204662047 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6204762048 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6204862049 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
62049 int8x16_t __ret;
6205062050 __ret = __noswap_vcombine_s8(__rev0, __noswap_vraddhn_s16(__rev1, __rev2));
6205162051 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6205262052 return __ret;
......@@ -62061,8 +62061,8 @@ __ai poly8x8_t vrbit_p8(poly8x8_t __p0) {
6206162061}
6206262062#else
6206362063__ai poly8x8_t vrbit_p8(poly8x8_t __p0) {
62064 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6206562064 poly8x8_t __ret;
62065 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6206662066 __ret = (poly8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 4);
6206762067 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6206862068 return __ret;
......@@ -62077,8 +62077,8 @@ __ai poly8x16_t vrbitq_p8(poly8x16_t __p0) {
6207762077}
6207862078#else
6207962079__ai poly8x16_t vrbitq_p8(poly8x16_t __p0) {
62080 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6208162080 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);
6208262082 __ret = (poly8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 36);
6208362083 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6208462084 return __ret;
......@@ -62093,8 +62093,8 @@ __ai uint8x16_t vrbitq_u8(uint8x16_t __p0) {
6209362093}
6209462094#else
6209562095__ai uint8x16_t vrbitq_u8(uint8x16_t __p0) {
62096 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6209762096 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);
6209862098 __ret = (uint8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 48);
6209962099 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6210062100 return __ret;
......@@ -62109,8 +62109,8 @@ __ai int8x16_t vrbitq_s8(int8x16_t __p0) {
6210962109}
6211062110#else
6211162111__ai int8x16_t vrbitq_s8(int8x16_t __p0) {
62112 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6211362112 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);
6211462114 __ret = (int8x16_t) __builtin_neon_vrbitq_v((int8x16_t)__rev0, 32);
6211562115 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6211662116 return __ret;
......@@ -62125,8 +62125,8 @@ __ai uint8x8_t vrbit_u8(uint8x8_t __p0) {
6212562125}
6212662126#else
6212762127__ai uint8x8_t vrbit_u8(uint8x8_t __p0) {
62128 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6212962128 uint8x8_t __ret;
62129 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6213062130 __ret = (uint8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 16);
6213162131 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6213262132 return __ret;
......@@ -62141,8 +62141,8 @@ __ai int8x8_t vrbit_s8(int8x8_t __p0) {
6214162141}
6214262142#else
6214362143__ai int8x8_t vrbit_s8(int8x8_t __p0) {
62144 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6214562144 int8x8_t __ret;
62145 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6214662146 __ret = (int8x8_t) __builtin_neon_vrbit_v((int8x8_t)__rev0, 0);
6214762147 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6214862148 return __ret;
......@@ -62157,8 +62157,8 @@ __ai float64x2_t vrecpeq_f64(float64x2_t __p0) {
6215762157}
6215862158#else
6215962159__ai float64x2_t vrecpeq_f64(float64x2_t __p0) {
62160 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6216162160 float64x2_t __ret;
62161 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6216262162 __ret = (float64x2_t) __builtin_neon_vrecpeq_v((int8x16_t)__rev0, 42);
6216362163 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6216462164 return __ret;
......@@ -62188,9 +62188,9 @@ __ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
6218862188}
6218962189#else
6219062190__ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
62191 float64x2_t __ret;
6219162192 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6219262193 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62193 float64x2_t __ret;
6219462194 __ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
6219562195 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6219662196 return __ret;
......@@ -62233,32 +62233,32 @@ __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
6223362233 return __ret;
6223462234}
6223562235#define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
62236 uint64_t __s0 = __p0; \
6223762236 uint64_t __ret; \
62237 uint64_t __s0 = __p0; \
6223862238 __ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
6223962239 __ret; \
6224062240})
6224162241#define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
62242 int64_t __s0 = __p0; \
6224362242 int64_t __ret; \
62243 int64_t __s0 = __p0; \
6224462244 __ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
6224562245 __ret; \
6224662246})
6224762247#ifdef __LITTLE_ENDIAN__
6224862248#define vrshrn_high_n_u32(__p0_807, __p1_807, __p2_807) __extension__ ({ \
62249 uint16x8_t __ret_807; \
6224962250 uint16x4_t __s0_807 = __p0_807; \
6225062251 uint32x4_t __s1_807 = __p1_807; \
62251 uint16x8_t __ret_807; \
6225262252 __ret_807 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_807), (uint16x4_t)(vrshrn_n_u32(__s1_807, __p2_807)))); \
6225362253 __ret_807; \
6225462254})
6225562255#else
6225662256#define vrshrn_high_n_u32(__p0_808, __p1_808, __p2_808) __extension__ ({ \
62257 uint16x8_t __ret_808; \
6225762258 uint16x4_t __s0_808 = __p0_808; \
6225862259 uint32x4_t __s1_808 = __p1_808; \
6225962260 uint16x4_t __rev0_808; __rev0_808 = __builtin_shufflevector(__s0_808, __s0_808, 3, 2, 1, 0); \
6226062261 uint32x4_t __rev1_808; __rev1_808 = __builtin_shufflevector(__s1_808, __s1_808, 3, 2, 1, 0); \
62261 uint16x8_t __ret_808; \
6226262262 __ret_808 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_808), (uint16x4_t)(__noswap_vrshrn_n_u32(__rev1_808, __p2_808)))); \
6226362263 __ret_808 = __builtin_shufflevector(__ret_808, __ret_808, 7, 6, 5, 4, 3, 2, 1, 0); \
6226462264 __ret_808; \
......@@ -62267,19 +62267,19 @@ __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
6226762267
6226862268#ifdef __LITTLE_ENDIAN__
6226962269#define vrshrn_high_n_u64(__p0_809, __p1_809, __p2_809) __extension__ ({ \
62270 uint32x4_t __ret_809; \
6227062271 uint32x2_t __s0_809 = __p0_809; \
6227162272 uint64x2_t __s1_809 = __p1_809; \
62272 uint32x4_t __ret_809; \
6227362273 __ret_809 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_809), (uint32x2_t)(vrshrn_n_u64(__s1_809, __p2_809)))); \
6227462274 __ret_809; \
6227562275})
6227662276#else
6227762277#define vrshrn_high_n_u64(__p0_810, __p1_810, __p2_810) __extension__ ({ \
62278 uint32x4_t __ret_810; \
6227862279 uint32x2_t __s0_810 = __p0_810; \
6227962280 uint64x2_t __s1_810 = __p1_810; \
6228062281 uint32x2_t __rev0_810; __rev0_810 = __builtin_shufflevector(__s0_810, __s0_810, 1, 0); \
6228162282 uint64x2_t __rev1_810; __rev1_810 = __builtin_shufflevector(__s1_810, __s1_810, 1, 0); \
62282 uint32x4_t __ret_810; \
6228362283 __ret_810 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_810), (uint32x2_t)(__noswap_vrshrn_n_u64(__rev1_810, __p2_810)))); \
6228462284 __ret_810 = __builtin_shufflevector(__ret_810, __ret_810, 3, 2, 1, 0); \
6228562285 __ret_810; \
......@@ -62288,19 +62288,19 @@ __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
6228862288
6228962289#ifdef __LITTLE_ENDIAN__
6229062290#define vrshrn_high_n_u16(__p0_811, __p1_811, __p2_811) __extension__ ({ \
62291 uint8x16_t __ret_811; \
6229162292 uint8x8_t __s0_811 = __p0_811; \
6229262293 uint16x8_t __s1_811 = __p1_811; \
62293 uint8x16_t __ret_811; \
6229462294 __ret_811 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_811), (uint8x8_t)(vrshrn_n_u16(__s1_811, __p2_811)))); \
6229562295 __ret_811; \
6229662296})
6229762297#else
6229862298#define vrshrn_high_n_u16(__p0_812, __p1_812, __p2_812) __extension__ ({ \
62299 uint8x16_t __ret_812; \
6229962300 uint8x8_t __s0_812 = __p0_812; \
6230062301 uint16x8_t __s1_812 = __p1_812; \
6230162302 uint8x8_t __rev0_812; __rev0_812 = __builtin_shufflevector(__s0_812, __s0_812, 7, 6, 5, 4, 3, 2, 1, 0); \
6230262303 uint16x8_t __rev1_812; __rev1_812 = __builtin_shufflevector(__s1_812, __s1_812, 7, 6, 5, 4, 3, 2, 1, 0); \
62303 uint8x16_t __ret_812; \
6230462304 __ret_812 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_812), (uint8x8_t)(__noswap_vrshrn_n_u16(__rev1_812, __p2_812)))); \
6230562305 __ret_812 = __builtin_shufflevector(__ret_812, __ret_812, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6230662306 __ret_812; \
......@@ -62309,19 +62309,19 @@ __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
6230962309
6231062310#ifdef __LITTLE_ENDIAN__
6231162311#define vrshrn_high_n_s32(__p0_813, __p1_813, __p2_813) __extension__ ({ \
62312 int16x8_t __ret_813; \
6231262313 int16x4_t __s0_813 = __p0_813; \
6231362314 int32x4_t __s1_813 = __p1_813; \
62314 int16x8_t __ret_813; \
6231562315 __ret_813 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_813), (int16x4_t)(vrshrn_n_s32(__s1_813, __p2_813)))); \
6231662316 __ret_813; \
6231762317})
6231862318#else
6231962319#define vrshrn_high_n_s32(__p0_814, __p1_814, __p2_814) __extension__ ({ \
62320 int16x8_t __ret_814; \
6232062321 int16x4_t __s0_814 = __p0_814; \
6232162322 int32x4_t __s1_814 = __p1_814; \
6232262323 int16x4_t __rev0_814; __rev0_814 = __builtin_shufflevector(__s0_814, __s0_814, 3, 2, 1, 0); \
6232362324 int32x4_t __rev1_814; __rev1_814 = __builtin_shufflevector(__s1_814, __s1_814, 3, 2, 1, 0); \
62324 int16x8_t __ret_814; \
6232562325 __ret_814 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_814), (int16x4_t)(__noswap_vrshrn_n_s32(__rev1_814, __p2_814)))); \
6232662326 __ret_814 = __builtin_shufflevector(__ret_814, __ret_814, 7, 6, 5, 4, 3, 2, 1, 0); \
6232762327 __ret_814; \
......@@ -62330,19 +62330,19 @@ __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
6233062330
6233162331#ifdef __LITTLE_ENDIAN__
6233262332#define vrshrn_high_n_s64(__p0_815, __p1_815, __p2_815) __extension__ ({ \
62333 int32x4_t __ret_815; \
6233362334 int32x2_t __s0_815 = __p0_815; \
6233462335 int64x2_t __s1_815 = __p1_815; \
62335 int32x4_t __ret_815; \
6233662336 __ret_815 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_815), (int32x2_t)(vrshrn_n_s64(__s1_815, __p2_815)))); \
6233762337 __ret_815; \
6233862338})
6233962339#else
6234062340#define vrshrn_high_n_s64(__p0_816, __p1_816, __p2_816) __extension__ ({ \
62341 int32x4_t __ret_816; \
6234162342 int32x2_t __s0_816 = __p0_816; \
6234262343 int64x2_t __s1_816 = __p1_816; \
6234362344 int32x2_t __rev0_816; __rev0_816 = __builtin_shufflevector(__s0_816, __s0_816, 1, 0); \
6234462345 int64x2_t __rev1_816; __rev1_816 = __builtin_shufflevector(__s1_816, __s1_816, 1, 0); \
62345 int32x4_t __ret_816; \
6234662346 __ret_816 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_816), (int32x2_t)(__noswap_vrshrn_n_s64(__rev1_816, __p2_816)))); \
6234762347 __ret_816 = __builtin_shufflevector(__ret_816, __ret_816, 3, 2, 1, 0); \
6234862348 __ret_816; \
......@@ -62351,19 +62351,19 @@ __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
6235162351
6235262352#ifdef __LITTLE_ENDIAN__
6235362353#define vrshrn_high_n_s16(__p0_817, __p1_817, __p2_817) __extension__ ({ \
62354 int8x16_t __ret_817; \
6235462355 int8x8_t __s0_817 = __p0_817; \
6235562356 int16x8_t __s1_817 = __p1_817; \
62356 int8x16_t __ret_817; \
6235762357 __ret_817 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_817), (int8x8_t)(vrshrn_n_s16(__s1_817, __p2_817)))); \
6235862358 __ret_817; \
6235962359})
6236062360#else
6236162361#define vrshrn_high_n_s16(__p0_818, __p1_818, __p2_818) __extension__ ({ \
62362 int8x16_t __ret_818; \
6236262363 int8x8_t __s0_818 = __p0_818; \
6236362364 int16x8_t __s1_818 = __p1_818; \
6236462365 int8x8_t __rev0_818; __rev0_818 = __builtin_shufflevector(__s0_818, __s0_818, 7, 6, 5, 4, 3, 2, 1, 0); \
6236562366 int16x8_t __rev1_818; __rev1_818 = __builtin_shufflevector(__s1_818, __s1_818, 7, 6, 5, 4, 3, 2, 1, 0); \
62366 int8x16_t __ret_818; \
6236762367 __ret_818 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_818), (int8x8_t)(__noswap_vrshrn_n_s16(__rev1_818, __p2_818)))); \
6236862368 __ret_818 = __builtin_shufflevector(__ret_818, __ret_818, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6236962369 __ret_818; \
......@@ -62378,8 +62378,8 @@ __ai float64x2_t vrsqrteq_f64(float64x2_t __p0) {
6237862378}
6237962379#else
6238062380__ai float64x2_t vrsqrteq_f64(float64x2_t __p0) {
62381 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6238262381 float64x2_t __ret;
62382 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6238362383 __ret = (float64x2_t) __builtin_neon_vrsqrteq_v((int8x16_t)__rev0, 42);
6238462384 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6238562385 return __ret;
......@@ -62409,9 +62409,9 @@ __ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
6240962409}
6241062410#else
6241162411__ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
62412 float64x2_t __ret;
6241262413 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6241362414 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
62414 float64x2_t __ret;
6241562415 __ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 42);
6241662416 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6241762417 return __ret;
......@@ -62434,16 +62434,16 @@ __ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
6243462434 return __ret;
6243562435}
6243662436#define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
62437 uint64_t __ret; \
6243762438 uint64_t __s0 = __p0; \
6243862439 uint64_t __s1 = __p1; \
62439 uint64_t __ret; \
6244062440 __ret = (uint64_t) __builtin_neon_vrsrad_n_u64(__s0, __s1, __p2); \
6244162441 __ret; \
6244262442})
6244362443#define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
62444 int64_t __ret; \
6244462445 int64_t __s0 = __p0; \
6244562446 int64_t __s1 = __p1; \
62446 int64_t __ret; \
6244762447 __ret = (int64_t) __builtin_neon_vrsrad_n_s64(__s0, __s1, __p2); \
6244862448 __ret; \
6244962449})
......@@ -62455,10 +62455,10 @@ __ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __
6245562455}
6245662456#else
6245762457__ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
62458 uint16x8_t __ret;
6245862459 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6245962460 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6246062461 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
62461 uint16x8_t __ret;
6246262462 __ret = __noswap_vcombine_u16(__rev0, __noswap_vrsubhn_u32(__rev1, __rev2));
6246362463 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6246462464 return __ret;
......@@ -62473,10 +62473,10 @@ __ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __
6247362473}
6247462474#else
6247562475__ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
62476 uint32x4_t __ret;
6247662477 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6247762478 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6247862479 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
62479 uint32x4_t __ret;
6248062480 __ret = __noswap_vcombine_u32(__rev0, __noswap_vrsubhn_u64(__rev1, __rev2));
6248162481 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6248262482 return __ret;
......@@ -62491,10 +62491,10 @@ __ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p
6249162491}
6249262492#else
6249362493__ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
62494 uint8x16_t __ret;
6249462495 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6249562496 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6249662497 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
62497 uint8x16_t __ret;
6249862498 __ret = __noswap_vcombine_u8(__rev0, __noswap_vrsubhn_u16(__rev1, __rev2));
6249962499 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6250062500 return __ret;
......@@ -62509,10 +62509,10 @@ __ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2)
6250962509}
6251062510#else
6251162511__ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
62512 int16x8_t __ret;
6251262513 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6251362514 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6251462515 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
62515 int16x8_t __ret;
6251662516 __ret = __noswap_vcombine_s16(__rev0, __noswap_vrsubhn_s32(__rev1, __rev2));
6251762517 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6251862518 return __ret;
......@@ -62527,10 +62527,10 @@ __ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2)
6252762527}
6252862528#else
6252962529__ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
62530 int32x4_t __ret;
6253062531 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6253162532 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6253262533 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
62533 int32x4_t __ret;
6253462534 __ret = __noswap_vcombine_s32(__rev0, __noswap_vrsubhn_s64(__rev1, __rev2));
6253562535 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6253662536 return __ret;
......@@ -62545,10 +62545,10 @@ __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
6254562545}
6254662546#else
6254762547__ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
62548 int8x16_t __ret;
6254862549 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6254962550 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6255062551 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
62551 int8x16_t __ret;
6255262552 __ret = __noswap_vcombine_s8(__rev0, __noswap_vrsubhn_s16(__rev1, __rev2));
6255362553 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6255462554 return __ret;
......@@ -62556,34 +62556,34 @@ __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
6255662556#endif
6255762557
6255862558#define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62559 poly64x1_t __ret; \
6255962560 poly64_t __s0 = __p0; \
6256062561 poly64x1_t __s1 = __p1; \
62561 poly64x1_t __ret; \
6256262562 __ret = (poly64x1_t) __builtin_neon_vset_lane_i64(__s0, (poly64x1_t)__s1, __p2); \
6256362563 __ret; \
6256462564})
6256562565#ifdef __LITTLE_ENDIAN__
6256662566#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62567 poly64x2_t __ret; \
6256762568 poly64_t __s0 = __p0; \
6256862569 poly64x2_t __s1 = __p1; \
62569 poly64x2_t __ret; \
6257062570 __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
6257162571 __ret; \
6257262572})
6257362573#else
6257462574#define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62575 poly64x2_t __ret; \
6257562576 poly64_t __s0 = __p0; \
6257662577 poly64x2_t __s1 = __p1; \
6257762578 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
62578 poly64x2_t __ret; \
6257962579 __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__rev1, __p2); \
6258062580 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
6258162581 __ret; \
6258262582})
6258362583#define __noswap_vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62584 poly64x2_t __ret; \
6258462585 poly64_t __s0 = __p0; \
6258562586 poly64x2_t __s1 = __p1; \
62586 poly64x2_t __ret; \
6258762587 __ret = (poly64x2_t) __builtin_neon_vsetq_lane_i64(__s0, (poly64x2_t)__s1, __p2); \
6258862588 __ret; \
6258962589})
......@@ -62591,35 +62591,35 @@ __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
6259162591
6259262592#ifdef __LITTLE_ENDIAN__
6259362593#define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62594 float64x2_t __ret; \
6259462595 float64_t __s0 = __p0; \
6259562596 float64x2_t __s1 = __p1; \
62596 float64x2_t __ret; \
6259762597 __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
6259862598 __ret; \
6259962599})
6260062600#else
6260162601#define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62602 float64x2_t __ret; \
6260262603 float64_t __s0 = __p0; \
6260362604 float64x2_t __s1 = __p1; \
6260462605 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
62605 float64x2_t __ret; \
6260662606 __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__rev1, __p2); \
6260762607 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
6260862608 __ret; \
6260962609})
6261062610#define __noswap_vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62611 float64x2_t __ret; \
6261162612 float64_t __s0 = __p0; \
6261262613 float64x2_t __s1 = __p1; \
62613 float64x2_t __ret; \
6261462614 __ret = (float64x2_t) __builtin_neon_vsetq_lane_f64(__s0, (float64x2_t)__s1, __p2); \
6261562615 __ret; \
6261662616})
6261762617#endif
6261862618
6261962619#define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62620 float64x1_t __ret; \
6262062621 float64_t __s0 = __p0; \
6262162622 float64x1_t __s1 = __p1; \
62622 float64x1_t __ret; \
6262362623 __ret = (float64x1_t) __builtin_neon_vset_lane_f64(__s0, (float64x1_t)__s1, __p2); \
6262462624 __ret; \
6262562625})
......@@ -62634,29 +62634,29 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6263462634 return __ret;
6263562635}
6263662636#define vshld_n_u64(__p0, __p1) __extension__ ({ \
62637 uint64_t __s0 = __p0; \
6263862637 uint64_t __ret; \
62638 uint64_t __s0 = __p0; \
6263962639 __ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
6264062640 __ret; \
6264162641})
6264262642#define vshld_n_s64(__p0, __p1) __extension__ ({ \
62643 int64_t __s0 = __p0; \
6264462643 int64_t __ret; \
62644 int64_t __s0 = __p0; \
6264562645 __ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
6264662646 __ret; \
6264762647})
6264862648#ifdef __LITTLE_ENDIAN__
6264962649#define vshll_high_n_u8(__p0_819, __p1_819) __extension__ ({ \
62650 uint8x16_t __s0_819 = __p0_819; \
6265162650 uint16x8_t __ret_819; \
62651 uint8x16_t __s0_819 = __p0_819; \
6265262652 __ret_819 = (uint16x8_t)(vshll_n_u8(vget_high_u8(__s0_819), __p1_819)); \
6265362653 __ret_819; \
6265462654})
6265562655#else
6265662656#define vshll_high_n_u8(__p0_820, __p1_820) __extension__ ({ \
62657 uint16x8_t __ret_820; \
6265762658 uint8x16_t __s0_820 = __p0_820; \
6265862659 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); \
62659 uint16x8_t __ret_820; \
6266062660 __ret_820 = (uint16x8_t)(__noswap_vshll_n_u8(__noswap_vget_high_u8(__rev0_820), __p1_820)); \
6266162661 __ret_820 = __builtin_shufflevector(__ret_820, __ret_820, 7, 6, 5, 4, 3, 2, 1, 0); \
6266262662 __ret_820; \
......@@ -62665,16 +62665,16 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6266562665
6266662666#ifdef __LITTLE_ENDIAN__
6266762667#define vshll_high_n_u32(__p0_821, __p1_821) __extension__ ({ \
62668 uint32x4_t __s0_821 = __p0_821; \
6266962668 uint64x2_t __ret_821; \
62669 uint32x4_t __s0_821 = __p0_821; \
6267062670 __ret_821 = (uint64x2_t)(vshll_n_u32(vget_high_u32(__s0_821), __p1_821)); \
6267162671 __ret_821; \
6267262672})
6267362673#else
6267462674#define vshll_high_n_u32(__p0_822, __p1_822) __extension__ ({ \
62675 uint64x2_t __ret_822; \
6267562676 uint32x4_t __s0_822 = __p0_822; \
6267662677 uint32x4_t __rev0_822; __rev0_822 = __builtin_shufflevector(__s0_822, __s0_822, 3, 2, 1, 0); \
62677 uint64x2_t __ret_822; \
6267862678 __ret_822 = (uint64x2_t)(__noswap_vshll_n_u32(__noswap_vget_high_u32(__rev0_822), __p1_822)); \
6267962679 __ret_822 = __builtin_shufflevector(__ret_822, __ret_822, 1, 0); \
6268062680 __ret_822; \
......@@ -62683,16 +62683,16 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6268362683
6268462684#ifdef __LITTLE_ENDIAN__
6268562685#define vshll_high_n_u16(__p0_823, __p1_823) __extension__ ({ \
62686 uint16x8_t __s0_823 = __p0_823; \
6268762686 uint32x4_t __ret_823; \
62687 uint16x8_t __s0_823 = __p0_823; \
6268862688 __ret_823 = (uint32x4_t)(vshll_n_u16(vget_high_u16(__s0_823), __p1_823)); \
6268962689 __ret_823; \
6269062690})
6269162691#else
6269262692#define vshll_high_n_u16(__p0_824, __p1_824) __extension__ ({ \
62693 uint32x4_t __ret_824; \
6269362694 uint16x8_t __s0_824 = __p0_824; \
6269462695 uint16x8_t __rev0_824; __rev0_824 = __builtin_shufflevector(__s0_824, __s0_824, 7, 6, 5, 4, 3, 2, 1, 0); \
62695 uint32x4_t __ret_824; \
6269662696 __ret_824 = (uint32x4_t)(__noswap_vshll_n_u16(__noswap_vget_high_u16(__rev0_824), __p1_824)); \
6269762697 __ret_824 = __builtin_shufflevector(__ret_824, __ret_824, 3, 2, 1, 0); \
6269862698 __ret_824; \
......@@ -62701,16 +62701,16 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6270162701
6270262702#ifdef __LITTLE_ENDIAN__
6270362703#define vshll_high_n_s8(__p0_825, __p1_825) __extension__ ({ \
62704 int8x16_t __s0_825 = __p0_825; \
6270562704 int16x8_t __ret_825; \
62705 int8x16_t __s0_825 = __p0_825; \
6270662706 __ret_825 = (int16x8_t)(vshll_n_s8(vget_high_s8(__s0_825), __p1_825)); \
6270762707 __ret_825; \
6270862708})
6270962709#else
6271062710#define vshll_high_n_s8(__p0_826, __p1_826) __extension__ ({ \
62711 int16x8_t __ret_826; \
6271162712 int8x16_t __s0_826 = __p0_826; \
6271262713 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); \
62713 int16x8_t __ret_826; \
6271462714 __ret_826 = (int16x8_t)(__noswap_vshll_n_s8(__noswap_vget_high_s8(__rev0_826), __p1_826)); \
6271562715 __ret_826 = __builtin_shufflevector(__ret_826, __ret_826, 7, 6, 5, 4, 3, 2, 1, 0); \
6271662716 __ret_826; \
......@@ -62719,16 +62719,16 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6271962719
6272062720#ifdef __LITTLE_ENDIAN__
6272162721#define vshll_high_n_s32(__p0_827, __p1_827) __extension__ ({ \
62722 int32x4_t __s0_827 = __p0_827; \
6272362722 int64x2_t __ret_827; \
62723 int32x4_t __s0_827 = __p0_827; \
6272462724 __ret_827 = (int64x2_t)(vshll_n_s32(vget_high_s32(__s0_827), __p1_827)); \
6272562725 __ret_827; \
6272662726})
6272762727#else
6272862728#define vshll_high_n_s32(__p0_828, __p1_828) __extension__ ({ \
62729 int64x2_t __ret_828; \
6272962730 int32x4_t __s0_828 = __p0_828; \
6273062731 int32x4_t __rev0_828; __rev0_828 = __builtin_shufflevector(__s0_828, __s0_828, 3, 2, 1, 0); \
62731 int64x2_t __ret_828; \
6273262732 __ret_828 = (int64x2_t)(__noswap_vshll_n_s32(__noswap_vget_high_s32(__rev0_828), __p1_828)); \
6273362733 __ret_828 = __builtin_shufflevector(__ret_828, __ret_828, 1, 0); \
6273462734 __ret_828; \
......@@ -62737,16 +62737,16 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6273762737
6273862738#ifdef __LITTLE_ENDIAN__
6273962739#define vshll_high_n_s16(__p0_829, __p1_829) __extension__ ({ \
62740 int16x8_t __s0_829 = __p0_829; \
6274162740 int32x4_t __ret_829; \
62741 int16x8_t __s0_829 = __p0_829; \
6274262742 __ret_829 = (int32x4_t)(vshll_n_s16(vget_high_s16(__s0_829), __p1_829)); \
6274362743 __ret_829; \
6274462744})
6274562745#else
6274662746#define vshll_high_n_s16(__p0_830, __p1_830) __extension__ ({ \
62747 int32x4_t __ret_830; \
6274762748 int16x8_t __s0_830 = __p0_830; \
6274862749 int16x8_t __rev0_830; __rev0_830 = __builtin_shufflevector(__s0_830, __s0_830, 7, 6, 5, 4, 3, 2, 1, 0); \
62749 int32x4_t __ret_830; \
6275062750 __ret_830 = (int32x4_t)(__noswap_vshll_n_s16(__noswap_vget_high_s16(__rev0_830), __p1_830)); \
6275162751 __ret_830 = __builtin_shufflevector(__ret_830, __ret_830, 3, 2, 1, 0); \
6275262752 __ret_830; \
......@@ -62754,32 +62754,32 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6275462754#endif
6275562755
6275662756#define vshrd_n_u64(__p0, __p1) __extension__ ({ \
62757 uint64_t __s0 = __p0; \
6275862757 uint64_t __ret; \
62758 uint64_t __s0 = __p0; \
6275962759 __ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
6276062760 __ret; \
6276162761})
6276262762#define vshrd_n_s64(__p0, __p1) __extension__ ({ \
62763 int64_t __s0 = __p0; \
6276462763 int64_t __ret; \
62764 int64_t __s0 = __p0; \
6276562765 __ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
6276662766 __ret; \
6276762767})
6276862768#ifdef __LITTLE_ENDIAN__
6276962769#define vshrn_high_n_u32(__p0_831, __p1_831, __p2_831) __extension__ ({ \
62770 uint16x8_t __ret_831; \
6277062771 uint16x4_t __s0_831 = __p0_831; \
6277162772 uint32x4_t __s1_831 = __p1_831; \
62772 uint16x8_t __ret_831; \
6277362773 __ret_831 = (uint16x8_t)(vcombine_u16((uint16x4_t)(__s0_831), (uint16x4_t)(vshrn_n_u32(__s1_831, __p2_831)))); \
6277462774 __ret_831; \
6277562775})
6277662776#else
6277762777#define vshrn_high_n_u32(__p0_832, __p1_832, __p2_832) __extension__ ({ \
62778 uint16x8_t __ret_832; \
6277862779 uint16x4_t __s0_832 = __p0_832; \
6277962780 uint32x4_t __s1_832 = __p1_832; \
6278062781 uint16x4_t __rev0_832; __rev0_832 = __builtin_shufflevector(__s0_832, __s0_832, 3, 2, 1, 0); \
6278162782 uint32x4_t __rev1_832; __rev1_832 = __builtin_shufflevector(__s1_832, __s1_832, 3, 2, 1, 0); \
62782 uint16x8_t __ret_832; \
6278362783 __ret_832 = (uint16x8_t)(__noswap_vcombine_u16((uint16x4_t)(__rev0_832), (uint16x4_t)(__noswap_vshrn_n_u32(__rev1_832, __p2_832)))); \
6278462784 __ret_832 = __builtin_shufflevector(__ret_832, __ret_832, 7, 6, 5, 4, 3, 2, 1, 0); \
6278562785 __ret_832; \
......@@ -62788,19 +62788,19 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6278862788
6278962789#ifdef __LITTLE_ENDIAN__
6279062790#define vshrn_high_n_u64(__p0_833, __p1_833, __p2_833) __extension__ ({ \
62791 uint32x4_t __ret_833; \
6279162792 uint32x2_t __s0_833 = __p0_833; \
6279262793 uint64x2_t __s1_833 = __p1_833; \
62793 uint32x4_t __ret_833; \
6279462794 __ret_833 = (uint32x4_t)(vcombine_u32((uint32x2_t)(__s0_833), (uint32x2_t)(vshrn_n_u64(__s1_833, __p2_833)))); \
6279562795 __ret_833; \
6279662796})
6279762797#else
6279862798#define vshrn_high_n_u64(__p0_834, __p1_834, __p2_834) __extension__ ({ \
62799 uint32x4_t __ret_834; \
6279962800 uint32x2_t __s0_834 = __p0_834; \
6280062801 uint64x2_t __s1_834 = __p1_834; \
6280162802 uint32x2_t __rev0_834; __rev0_834 = __builtin_shufflevector(__s0_834, __s0_834, 1, 0); \
6280262803 uint64x2_t __rev1_834; __rev1_834 = __builtin_shufflevector(__s1_834, __s1_834, 1, 0); \
62803 uint32x4_t __ret_834; \
6280462804 __ret_834 = (uint32x4_t)(__noswap_vcombine_u32((uint32x2_t)(__rev0_834), (uint32x2_t)(__noswap_vshrn_n_u64(__rev1_834, __p2_834)))); \
6280562805 __ret_834 = __builtin_shufflevector(__ret_834, __ret_834, 3, 2, 1, 0); \
6280662806 __ret_834; \
......@@ -62809,19 +62809,19 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6280962809
6281062810#ifdef __LITTLE_ENDIAN__
6281162811#define vshrn_high_n_u16(__p0_835, __p1_835, __p2_835) __extension__ ({ \
62812 uint8x16_t __ret_835; \
6281262813 uint8x8_t __s0_835 = __p0_835; \
6281362814 uint16x8_t __s1_835 = __p1_835; \
62814 uint8x16_t __ret_835; \
6281562815 __ret_835 = (uint8x16_t)(vcombine_u8((uint8x8_t)(__s0_835), (uint8x8_t)(vshrn_n_u16(__s1_835, __p2_835)))); \
6281662816 __ret_835; \
6281762817})
6281862818#else
6281962819#define vshrn_high_n_u16(__p0_836, __p1_836, __p2_836) __extension__ ({ \
62820 uint8x16_t __ret_836; \
6282062821 uint8x8_t __s0_836 = __p0_836; \
6282162822 uint16x8_t __s1_836 = __p1_836; \
6282262823 uint8x8_t __rev0_836; __rev0_836 = __builtin_shufflevector(__s0_836, __s0_836, 7, 6, 5, 4, 3, 2, 1, 0); \
6282362824 uint16x8_t __rev1_836; __rev1_836 = __builtin_shufflevector(__s1_836, __s1_836, 7, 6, 5, 4, 3, 2, 1, 0); \
62824 uint8x16_t __ret_836; \
6282562825 __ret_836 = (uint8x16_t)(__noswap_vcombine_u8((uint8x8_t)(__rev0_836), (uint8x8_t)(__noswap_vshrn_n_u16(__rev1_836, __p2_836)))); \
6282662826 __ret_836 = __builtin_shufflevector(__ret_836, __ret_836, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6282762827 __ret_836; \
......@@ -62830,19 +62830,19 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6283062830
6283162831#ifdef __LITTLE_ENDIAN__
6283262832#define vshrn_high_n_s32(__p0_837, __p1_837, __p2_837) __extension__ ({ \
62833 int16x8_t __ret_837; \
6283362834 int16x4_t __s0_837 = __p0_837; \
6283462835 int32x4_t __s1_837 = __p1_837; \
62835 int16x8_t __ret_837; \
6283662836 __ret_837 = (int16x8_t)(vcombine_s16((int16x4_t)(__s0_837), (int16x4_t)(vshrn_n_s32(__s1_837, __p2_837)))); \
6283762837 __ret_837; \
6283862838})
6283962839#else
6284062840#define vshrn_high_n_s32(__p0_838, __p1_838, __p2_838) __extension__ ({ \
62841 int16x8_t __ret_838; \
6284162842 int16x4_t __s0_838 = __p0_838; \
6284262843 int32x4_t __s1_838 = __p1_838; \
6284362844 int16x4_t __rev0_838; __rev0_838 = __builtin_shufflevector(__s0_838, __s0_838, 3, 2, 1, 0); \
6284462845 int32x4_t __rev1_838; __rev1_838 = __builtin_shufflevector(__s1_838, __s1_838, 3, 2, 1, 0); \
62845 int16x8_t __ret_838; \
6284662846 __ret_838 = (int16x8_t)(__noswap_vcombine_s16((int16x4_t)(__rev0_838), (int16x4_t)(__noswap_vshrn_n_s32(__rev1_838, __p2_838)))); \
6284762847 __ret_838 = __builtin_shufflevector(__ret_838, __ret_838, 7, 6, 5, 4, 3, 2, 1, 0); \
6284862848 __ret_838; \
......@@ -62851,19 +62851,19 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6285162851
6285262852#ifdef __LITTLE_ENDIAN__
6285362853#define vshrn_high_n_s64(__p0_839, __p1_839, __p2_839) __extension__ ({ \
62854 int32x4_t __ret_839; \
6285462855 int32x2_t __s0_839 = __p0_839; \
6285562856 int64x2_t __s1_839 = __p1_839; \
62856 int32x4_t __ret_839; \
6285762857 __ret_839 = (int32x4_t)(vcombine_s32((int32x2_t)(__s0_839), (int32x2_t)(vshrn_n_s64(__s1_839, __p2_839)))); \
6285862858 __ret_839; \
6285962859})
6286062860#else
6286162861#define vshrn_high_n_s64(__p0_840, __p1_840, __p2_840) __extension__ ({ \
62862 int32x4_t __ret_840; \
6286262863 int32x2_t __s0_840 = __p0_840; \
6286362864 int64x2_t __s1_840 = __p1_840; \
6286462865 int32x2_t __rev0_840; __rev0_840 = __builtin_shufflevector(__s0_840, __s0_840, 1, 0); \
6286562866 int64x2_t __rev1_840; __rev1_840 = __builtin_shufflevector(__s1_840, __s1_840, 1, 0); \
62866 int32x4_t __ret_840; \
6286762867 __ret_840 = (int32x4_t)(__noswap_vcombine_s32((int32x2_t)(__rev0_840), (int32x2_t)(__noswap_vshrn_n_s64(__rev1_840, __p2_840)))); \
6286862868 __ret_840 = __builtin_shufflevector(__ret_840, __ret_840, 3, 2, 1, 0); \
6286962869 __ret_840; \
......@@ -62872,19 +62872,19 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6287262872
6287362873#ifdef __LITTLE_ENDIAN__
6287462874#define vshrn_high_n_s16(__p0_841, __p1_841, __p2_841) __extension__ ({ \
62875 int8x16_t __ret_841; \
6287562876 int8x8_t __s0_841 = __p0_841; \
6287662877 int16x8_t __s1_841 = __p1_841; \
62877 int8x16_t __ret_841; \
6287862878 __ret_841 = (int8x16_t)(vcombine_s8((int8x8_t)(__s0_841), (int8x8_t)(vshrn_n_s16(__s1_841, __p2_841)))); \
6287962879 __ret_841; \
6288062880})
6288162881#else
6288262882#define vshrn_high_n_s16(__p0_842, __p1_842, __p2_842) __extension__ ({ \
62883 int8x16_t __ret_842; \
6288362884 int8x8_t __s0_842 = __p0_842; \
6288462885 int16x8_t __s1_842 = __p1_842; \
6288562886 int8x8_t __rev0_842; __rev0_842 = __builtin_shufflevector(__s0_842, __s0_842, 7, 6, 5, 4, 3, 2, 1, 0); \
6288662887 int16x8_t __rev1_842; __rev1_842 = __builtin_shufflevector(__s1_842, __s1_842, 7, 6, 5, 4, 3, 2, 1, 0); \
62887 int8x16_t __ret_842; \
6288862888 __ret_842 = (int8x16_t)(__noswap_vcombine_s8((int8x8_t)(__rev0_842), (int8x8_t)(__noswap_vshrn_n_s16(__rev1_842, __p2_842)))); \
6288962889 __ret_842 = __builtin_shufflevector(__ret_842, __ret_842, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0); \
6289062890 __ret_842; \
......@@ -62892,41 +62892,41 @@ __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
6289262892#endif
6289362893
6289462894#define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
62895 uint64_t __ret; \
6289562896 uint64_t __s0 = __p0; \
6289662897 uint64_t __s1 = __p1; \
62897 uint64_t __ret; \
6289862898 __ret = (uint64_t) __builtin_neon_vslid_n_u64(__s0, __s1, __p2); \
6289962899 __ret; \
6290062900})
6290162901#define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
62902 int64_t __ret; \
6290262903 int64_t __s0 = __p0; \
6290362904 int64_t __s1 = __p1; \
62904 int64_t __ret; \
6290562905 __ret = (int64_t) __builtin_neon_vslid_n_s64(__s0, __s1, __p2); \
6290662906 __ret; \
6290762907})
6290862908#define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
62909 poly64x1_t __ret; \
6290962910 poly64x1_t __s0 = __p0; \
6291062911 poly64x1_t __s1 = __p1; \
62911 poly64x1_t __ret; \
6291262912 __ret = (poly64x1_t) __builtin_neon_vsli_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
6291362913 __ret; \
6291462914})
6291562915#ifdef __LITTLE_ENDIAN__
6291662916#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
62917 poly64x2_t __ret; \
6291762918 poly64x2_t __s0 = __p0; \
6291862919 poly64x2_t __s1 = __p1; \
62919 poly64x2_t __ret; \
6292062920 __ret = (poly64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
6292162921 __ret; \
6292262922})
6292362923#else
6292462924#define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
62925 poly64x2_t __ret; \
6292562926 poly64x2_t __s0 = __p0; \
6292662927 poly64x2_t __s1 = __p1; \
6292762928 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
6292862929 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
62929 poly64x2_t __ret; \
6293062930 __ret = (poly64x2_t) __builtin_neon_vsliq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
6293162931 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
6293262932 __ret; \
......@@ -62961,9 +62961,9 @@ __ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
6296162961}
6296262962#else
6296362963__ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, int8x16_t __p1) {
62964 uint8x16_t __ret;
6296462965 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6296562966 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
62966 uint8x16_t __ret;
6296762967 __ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 48);
6296862968 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6296962969 return __ret;
......@@ -62978,9 +62978,9 @@ __ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
6297862978}
6297962979#else
6298062980__ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, int32x4_t __p1) {
62981 uint32x4_t __ret;
6298162982 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6298262983 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
62983 uint32x4_t __ret;
6298462984 __ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 50);
6298562985 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6298662986 return __ret;
......@@ -62995,9 +62995,9 @@ __ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
6299562995}
6299662996#else
6299762997__ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, int64x2_t __p1) {
62998 uint64x2_t __ret;
6299862999 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6299963000 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63000 uint64x2_t __ret;
6300163001 __ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
6300263002 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6300363003 return __ret;
......@@ -63012,9 +63012,9 @@ __ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
6301263012}
6301363013#else
6301463014__ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, int16x8_t __p1) {
63015 uint16x8_t __ret;
6301563016 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6301663017 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63017 uint16x8_t __ret;
6301863018 __ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 49);
6301963019 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6302063020 return __ret;
......@@ -63029,9 +63029,9 @@ __ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
6302963029}
6303063030#else
6303163031__ai uint8x8_t vsqadd_u8(uint8x8_t __p0, int8x8_t __p1) {
63032 uint8x8_t __ret;
6303263033 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6303363034 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63034 uint8x8_t __ret;
6303563035 __ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 16);
6303663036 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6303763037 return __ret;
......@@ -63046,9 +63046,9 @@ __ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
6304663046}
6304763047#else
6304863048__ai uint32x2_t vsqadd_u32(uint32x2_t __p0, int32x2_t __p1) {
63049 uint32x2_t __ret;
6304963050 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6305063051 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63051 uint32x2_t __ret;
6305263052 __ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 18);
6305363053 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6305463054 return __ret;
......@@ -63068,9 +63068,9 @@ __ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
6306863068}
6306963069#else
6307063070__ai uint16x4_t vsqadd_u16(uint16x4_t __p0, int16x4_t __p1) {
63071 uint16x4_t __ret;
6307163072 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6307263073 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63073 uint16x4_t __ret;
6307463074 __ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 17);
6307563075 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6307663076 return __ret;
......@@ -63085,8 +63085,8 @@ __ai float64x2_t vsqrtq_f64(float64x2_t __p0) {
6308563085}
6308663086#else
6308763087__ai float64x2_t vsqrtq_f64(float64x2_t __p0) {
63088 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6308963088 float64x2_t __ret;
63089 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6309063090 __ret = (float64x2_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 42);
6309163091 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6309263092 return __ret;
......@@ -63101,8 +63101,8 @@ __ai float32x4_t vsqrtq_f32(float32x4_t __p0) {
6310163101}
6310263102#else
6310363103__ai float32x4_t vsqrtq_f32(float32x4_t __p0) {
63104 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6310563104 float32x4_t __ret;
63105 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6310663106 __ret = (float32x4_t) __builtin_neon_vsqrtq_v((int8x16_t)__rev0, 41);
6310763107 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6310863108 return __ret;
......@@ -63122,8 +63122,8 @@ __ai float32x2_t vsqrt_f32(float32x2_t __p0) {
6312263122}
6312363123#else
6312463124__ai float32x2_t vsqrt_f32(float32x2_t __p0) {
63125 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6312663125 float32x2_t __ret;
63126 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6312763127 __ret = (float32x2_t) __builtin_neon_vsqrt_v((int8x8_t)__rev0, 9);
6312863128 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6312963129 return __ret;
......@@ -63131,55 +63131,55 @@ __ai float32x2_t vsqrt_f32(float32x2_t __p0) {
6313163131#endif
6313263132
6313363133#define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
63134 uint64_t __ret; \
6313463135 uint64_t __s0 = __p0; \
6313563136 uint64_t __s1 = __p1; \
63136 uint64_t __ret; \
6313763137 __ret = (uint64_t) __builtin_neon_vsrad_n_u64(__s0, __s1, __p2); \
6313863138 __ret; \
6313963139})
6314063140#define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
63141 int64_t __ret; \
6314163142 int64_t __s0 = __p0; \
6314263143 int64_t __s1 = __p1; \
63143 int64_t __ret; \
6314463144 __ret = (int64_t) __builtin_neon_vsrad_n_s64(__s0, __s1, __p2); \
6314563145 __ret; \
6314663146})
6314763147#define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
63148 uint64_t __ret; \
6314863149 uint64_t __s0 = __p0; \
6314963150 uint64_t __s1 = __p1; \
63150 uint64_t __ret; \
6315163151 __ret = (uint64_t) __builtin_neon_vsrid_n_u64(__s0, __s1, __p2); \
6315263152 __ret; \
6315363153})
6315463154#define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
63155 int64_t __ret; \
6315563156 int64_t __s0 = __p0; \
6315663157 int64_t __s1 = __p1; \
63157 int64_t __ret; \
6315863158 __ret = (int64_t) __builtin_neon_vsrid_n_s64(__s0, __s1, __p2); \
6315963159 __ret; \
6316063160})
6316163161#define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
63162 poly64x1_t __ret; \
6316263163 poly64x1_t __s0 = __p0; \
6316363164 poly64x1_t __s1 = __p1; \
63164 poly64x1_t __ret; \
6316563165 __ret = (poly64x1_t) __builtin_neon_vsri_n_v((int8x8_t)__s0, (int8x8_t)__s1, __p2, 6); \
6316663166 __ret; \
6316763167})
6316863168#ifdef __LITTLE_ENDIAN__
6316963169#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
63170 poly64x2_t __ret; \
6317063171 poly64x2_t __s0 = __p0; \
6317163172 poly64x2_t __s1 = __p1; \
63172 poly64x2_t __ret; \
6317363173 __ret = (poly64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__s0, (int8x16_t)__s1, __p2, 38); \
6317463174 __ret; \
6317563175})
6317663176#else
6317763177#define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
63178 poly64x2_t __ret; \
6317863179 poly64x2_t __s0 = __p0; \
6317963180 poly64x2_t __s1 = __p1; \
6318063181 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__s0, __s0, 1, 0); \
6318163182 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__s1, __s1, 1, 0); \
63182 poly64x2_t __ret; \
6318363183 __ret = (poly64x2_t) __builtin_neon_vsriq_n_v((int8x16_t)__rev0, (int8x16_t)__rev1, __p2, 38); \
6318463184 __ret = __builtin_shufflevector(__ret, __ret, 1, 0); \
6318563185 __ret; \
......@@ -63996,9 +63996,9 @@ __ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
6399663996}
6399763997#else
6399863998__ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
63999 float64x2_t __ret;
6399964000 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6400064001 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64001 float64x2_t __ret;
6400264002 __ret = __rev0 - __rev1;
6400364003 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6400464004 return __ret;
......@@ -64018,10 +64018,10 @@ __ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p
6401864018}
6401964019#else
6402064020__ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
64021 uint16x8_t __ret;
6402164022 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6402264023 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6402364024 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
64024 uint16x8_t __ret;
6402564025 __ret = __noswap_vcombine_u16(__rev0, __noswap_vsubhn_u32(__rev1, __rev2));
6402664026 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6402764027 return __ret;
......@@ -64036,10 +64036,10 @@ __ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p
6403664036}
6403764037#else
6403864038__ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
64039 uint32x4_t __ret;
6403964040 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6404064041 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6404164042 uint64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
64042 uint32x4_t __ret;
6404364043 __ret = __noswap_vcombine_u32(__rev0, __noswap_vsubhn_u64(__rev1, __rev2));
6404464044 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6404564045 return __ret;
......@@ -64054,10 +64054,10 @@ __ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2
6405464054}
6405564055#else
6405664056__ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
64057 uint8x16_t __ret;
6405764058 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6405864059 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6405964060 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
64060 uint8x16_t __ret;
6406164061 __ret = __noswap_vcombine_u8(__rev0, __noswap_vsubhn_u16(__rev1, __rev2));
6406264062 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6406364063 return __ret;
......@@ -64072,10 +64072,10 @@ __ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
6407264072}
6407364073#else
6407464074__ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
64075 int16x8_t __ret;
6407564076 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6407664077 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6407764078 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
64078 int16x8_t __ret;
6407964079 __ret = __noswap_vcombine_s16(__rev0, __noswap_vsubhn_s32(__rev1, __rev2));
6408064080 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6408164081 return __ret;
......@@ -64090,10 +64090,10 @@ __ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
6409064090}
6409164091#else
6409264092__ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
64093 int32x4_t __ret;
6409364094 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6409464095 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6409564096 int64x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
64096 int32x4_t __ret;
6409764097 __ret = __noswap_vcombine_s32(__rev0, __noswap_vsubhn_s64(__rev1, __rev2));
6409864098 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6409964099 return __ret;
......@@ -64108,10 +64108,10 @@ __ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
6410864108}
6410964109#else
6411064110__ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
64111 int8x16_t __ret;
6411164112 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6411264113 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6411364114 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
64114 int8x16_t __ret;
6411564115 __ret = __noswap_vcombine_s8(__rev0, __noswap_vsubhn_s16(__rev1, __rev2));
6411664116 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6411764117 return __ret;
......@@ -64126,9 +64126,9 @@ __ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
6412664126}
6412764127#else
6412864128__ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
64129 uint16x8_t __ret;
6412964130 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6413064131 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64131 uint16x8_t __ret;
6413264132 __ret = __noswap_vmovl_high_u8(__rev0) - __noswap_vmovl_high_u8(__rev1);
6413364133 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6413464134 return __ret;
......@@ -64143,9 +64143,9 @@ __ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
6414364143}
6414464144#else
6414564145__ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
64146 uint64x2_t __ret;
6414664147 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6414764148 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64148 uint64x2_t __ret;
6414964149 __ret = __noswap_vmovl_high_u32(__rev0) - __noswap_vmovl_high_u32(__rev1);
6415064150 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6415164151 return __ret;
......@@ -64160,9 +64160,9 @@ __ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
6416064160}
6416164161#else
6416264162__ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
64163 uint32x4_t __ret;
6416364164 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6416464165 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64165 uint32x4_t __ret;
6416664166 __ret = __noswap_vmovl_high_u16(__rev0) - __noswap_vmovl_high_u16(__rev1);
6416764167 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6416864168 return __ret;
......@@ -64177,9 +64177,9 @@ __ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
6417764177}
6417864178#else
6417964179__ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
64180 int16x8_t __ret;
6418064181 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6418164182 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64182 int16x8_t __ret;
6418364183 __ret = __noswap_vmovl_high_s8(__rev0) - __noswap_vmovl_high_s8(__rev1);
6418464184 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6418564185 return __ret;
......@@ -64194,9 +64194,9 @@ __ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
6419464194}
6419564195#else
6419664196__ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
64197 int64x2_t __ret;
6419764198 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6419864199 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64199 int64x2_t __ret;
6420064200 __ret = __noswap_vmovl_high_s32(__rev0) - __noswap_vmovl_high_s32(__rev1);
6420164201 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6420264202 return __ret;
......@@ -64211,9 +64211,9 @@ __ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
6421164211}
6421264212#else
6421364213__ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
64214 int32x4_t __ret;
6421464215 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6421564216 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64216 int32x4_t __ret;
6421764217 __ret = __noswap_vmovl_high_s16(__rev0) - __noswap_vmovl_high_s16(__rev1);
6421864218 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6421964219 return __ret;
......@@ -64228,9 +64228,9 @@ __ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
6422864228}
6422964229#else
6423064230__ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
64231 uint16x8_t __ret;
6423164232 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6423264233 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64233 uint16x8_t __ret;
6423464234 __ret = __rev0 - __noswap_vmovl_high_u8(__rev1);
6423564235 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6423664236 return __ret;
......@@ -64245,9 +64245,9 @@ __ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
6424564245}
6424664246#else
6424764247__ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
64248 uint64x2_t __ret;
6424864249 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6424964250 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64250 uint64x2_t __ret;
6425164251 __ret = __rev0 - __noswap_vmovl_high_u32(__rev1);
6425264252 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6425364253 return __ret;
......@@ -64262,9 +64262,9 @@ __ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
6426264262}
6426364263#else
6426464264__ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
64265 uint32x4_t __ret;
6426564266 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6426664267 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64267 uint32x4_t __ret;
6426864268 __ret = __rev0 - __noswap_vmovl_high_u16(__rev1);
6426964269 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6427064270 return __ret;
......@@ -64279,9 +64279,9 @@ __ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
6427964279}
6428064280#else
6428164281__ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
64282 int16x8_t __ret;
6428264283 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6428364284 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64284 int16x8_t __ret;
6428564285 __ret = __rev0 - __noswap_vmovl_high_s8(__rev1);
6428664286 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6428764287 return __ret;
......@@ -64296,9 +64296,9 @@ __ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
6429664296}
6429764297#else
6429864298__ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
64299 int64x2_t __ret;
6429964300 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6430064301 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64301 int64x2_t __ret;
6430264302 __ret = __rev0 - __noswap_vmovl_high_s32(__rev1);
6430364303 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6430464304 return __ret;
......@@ -64313,9 +64313,9 @@ __ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6431364313}
6431464314#else
6431564315__ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
64316 int32x4_t __ret;
6431664317 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6431764318 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64318 int32x4_t __ret;
6431964319 __ret = __rev0 - __noswap_vmovl_high_s16(__rev1);
6432064320 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6432164321 return __ret;
......@@ -64324,23 +64324,23 @@ __ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6432464324
6432564325#ifdef __LITTLE_ENDIAN__
6432664326#define vsudotq_laneq_s32(__p0_843, __p1_843, __p2_843, __p3_843) __extension__ ({ \
64327 int32x4_t __ret_843; \
6432764328 int32x4_t __s0_843 = __p0_843; \
6432864329 int8x16_t __s1_843 = __p1_843; \
6432964330 uint8x16_t __s2_843 = __p2_843; \
64330 int32x4_t __ret_843; \
6433164331uint8x16_t __reint_843 = __s2_843; \
6433264332 __ret_843 = vusdotq_s32(__s0_843, (uint8x16_t)(splatq_laneq_s32(*(int32x4_t *) &__reint_843, __p3_843)), __s1_843); \
6433364333 __ret_843; \
6433464334})
6433564335#else
6433664336#define vsudotq_laneq_s32(__p0_844, __p1_844, __p2_844, __p3_844) __extension__ ({ \
64337 int32x4_t __ret_844; \
6433764338 int32x4_t __s0_844 = __p0_844; \
6433864339 int8x16_t __s1_844 = __p1_844; \
6433964340 uint8x16_t __s2_844 = __p2_844; \
6434064341 int32x4_t __rev0_844; __rev0_844 = __builtin_shufflevector(__s0_844, __s0_844, 3, 2, 1, 0); \
6434164342 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); \
6434264343 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); \
64343 int32x4_t __ret_844; \
6434464344uint8x16_t __reint_844 = __rev2_844; \
6434564345 __ret_844 = __noswap_vusdotq_s32(__rev0_844, (uint8x16_t)(__noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_844, __p3_844)), __rev1_844); \
6434664346 __ret_844 = __builtin_shufflevector(__ret_844, __ret_844, 3, 2, 1, 0); \
......@@ -64350,23 +64350,23 @@ uint8x16_t __reint_844 = __rev2_844; \
6435064350
6435164351#ifdef __LITTLE_ENDIAN__
6435264352#define vsudot_laneq_s32(__p0_845, __p1_845, __p2_845, __p3_845) __extension__ ({ \
64353 int32x2_t __ret_845; \
6435364354 int32x2_t __s0_845 = __p0_845; \
6435464355 int8x8_t __s1_845 = __p1_845; \
6435564356 uint8x16_t __s2_845 = __p2_845; \
64356 int32x2_t __ret_845; \
6435764357uint8x16_t __reint_845 = __s2_845; \
6435864358 __ret_845 = vusdot_s32(__s0_845, (uint8x8_t)(splat_laneq_s32(*(int32x4_t *) &__reint_845, __p3_845)), __s1_845); \
6435964359 __ret_845; \
6436064360})
6436164361#else
6436264362#define vsudot_laneq_s32(__p0_846, __p1_846, __p2_846, __p3_846) __extension__ ({ \
64363 int32x2_t __ret_846; \
6436364364 int32x2_t __s0_846 = __p0_846; \
6436464365 int8x8_t __s1_846 = __p1_846; \
6436564366 uint8x16_t __s2_846 = __p2_846; \
6436664367 int32x2_t __rev0_846; __rev0_846 = __builtin_shufflevector(__s0_846, __s0_846, 1, 0); \
6436764368 int8x8_t __rev1_846; __rev1_846 = __builtin_shufflevector(__s1_846, __s1_846, 7, 6, 5, 4, 3, 2, 1, 0); \
6436864369 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); \
64369 int32x2_t __ret_846; \
6437064370uint8x16_t __reint_846 = __rev2_846; \
6437164371 __ret_846 = __noswap_vusdot_s32(__rev0_846, (uint8x8_t)(__noswap_splat_laneq_s32(*(int32x4_t *) &__reint_846, __p3_846)), __rev1_846); \
6437264372 __ret_846 = __builtin_shufflevector(__ret_846, __ret_846, 1, 0); \
......@@ -64382,9 +64382,9 @@ __ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
6438264382}
6438364383#else
6438464384__ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64385 poly8x8_t __ret;
6438564386 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6438664387 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64387 poly8x8_t __ret;
6438864388 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
6438964389 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6439064390 return __ret;
......@@ -64399,9 +64399,9 @@ __ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
6439964399}
6440064400#else
6440164401__ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64402 poly16x4_t __ret;
6440264403 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6440364404 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64404 poly16x4_t __ret;
6440564405 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
6440664406 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6440764407 return __ret;
......@@ -64416,9 +64416,9 @@ __ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
6441664416}
6441764417#else
6441864418__ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64419 poly8x16_t __ret;
6441964420 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6442064421 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64421 poly8x16_t __ret;
6442264422 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
6442364423 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6442464424 return __ret;
......@@ -64433,9 +64433,9 @@ __ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
6443364433}
6443464434#else
6443564435__ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64436 poly64x2_t __ret;
6443664437 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6443764438 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64438 poly64x2_t __ret;
6443964439 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6444064440 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6444164441 return __ret;
......@@ -64450,9 +64450,9 @@ __ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
6445064450}
6445164451#else
6445264452__ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64453 poly16x8_t __ret;
6445364454 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6445464455 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64455 poly16x8_t __ret;
6445664456 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
6445764457 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6445864458 return __ret;
......@@ -64467,9 +64467,9 @@ __ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6446764467}
6446864468#else
6446964469__ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64470 uint8x16_t __ret;
6447064471 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6447164472 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64472 uint8x16_t __ret;
6447364473 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
6447464474 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6447564475 return __ret;
......@@ -64484,9 +64484,9 @@ __ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
6448464484}
6448564485#else
6448664486__ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64487 uint32x4_t __ret;
6448764488 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6448864489 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64489 uint32x4_t __ret;
6449064490 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
6449164491 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6449264492 return __ret;
......@@ -64501,9 +64501,9 @@ __ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
6450164501}
6450264502#else
6450364503__ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64504 uint64x2_t __ret;
6450464505 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6450564506 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64506 uint64x2_t __ret;
6450764507 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6450864508 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6450964509 return __ret;
......@@ -64518,9 +64518,9 @@ __ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
6451864518}
6451964519#else
6452064520__ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64521 uint16x8_t __ret;
6452164522 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6452264523 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64523 uint16x8_t __ret;
6452464524 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
6452564525 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6452664526 return __ret;
......@@ -64535,9 +64535,9 @@ __ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
6453564535}
6453664536#else
6453764537__ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
64538 int8x16_t __ret;
6453864539 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6453964540 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64540 int8x16_t __ret;
6454164541 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
6454264542 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6454364543 return __ret;
......@@ -64552,9 +64552,9 @@ __ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
6455264552}
6455364553#else
6455464554__ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
64555 float64x2_t __ret;
6455564556 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6455664557 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64557 float64x2_t __ret;
6455864558 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6455964559 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6456064560 return __ret;
......@@ -64569,9 +64569,9 @@ __ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
6456964569}
6457064570#else
6457164571__ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
64572 float32x4_t __ret;
6457264573 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6457364574 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64574 float32x4_t __ret;
6457564575 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
6457664576 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6457764577 return __ret;
......@@ -64586,9 +64586,9 @@ __ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
6458664586}
6458764587#else
6458864588__ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
64589 int32x4_t __ret;
6458964590 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6459064591 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64591 int32x4_t __ret;
6459264592 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
6459364593 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6459464594 return __ret;
......@@ -64603,9 +64603,9 @@ __ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
6460364603}
6460464604#else
6460564605__ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
64606 int64x2_t __ret;
6460664607 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6460764608 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64608 int64x2_t __ret;
6460964609 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6461064610 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6461164611 return __ret;
......@@ -64620,9 +64620,9 @@ __ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
6462064620}
6462164621#else
6462264622__ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
64623 int16x8_t __ret;
6462364624 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6462464625 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64625 int16x8_t __ret;
6462664626 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
6462764627 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6462864628 return __ret;
......@@ -64637,9 +64637,9 @@ __ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
6463764637}
6463864638#else
6463964639__ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64640 uint8x8_t __ret;
6464064641 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6464164642 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64642 uint8x8_t __ret;
6464364643 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
6464464644 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6464564645 return __ret;
......@@ -64654,9 +64654,9 @@ __ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
6465464654}
6465564655#else
6465664656__ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64657 uint32x2_t __ret;
6465764658 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6465864659 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64659 uint32x2_t __ret;
6466064660 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6466164661 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6466264662 return __ret;
......@@ -64671,9 +64671,9 @@ __ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
6467164671}
6467264672#else
6467364673__ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64674 uint16x4_t __ret;
6467464675 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6467564676 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64676 uint16x4_t __ret;
6467764677 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
6467864678 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6467964679 return __ret;
......@@ -64688,9 +64688,9 @@ __ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
6468864688}
6468964689#else
6469064690__ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
64691 int8x8_t __ret;
6469164692 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6469264693 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64693 int8x8_t __ret;
6469464694 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 2, 10, 4, 12, 6, 14);
6469564695 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6469664696 return __ret;
......@@ -64705,9 +64705,9 @@ __ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
6470564705}
6470664706#else
6470764707__ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
64708 float32x2_t __ret;
6470864709 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6470964710 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64710 float32x2_t __ret;
6471164711 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6471264712 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6471364713 return __ret;
......@@ -64722,9 +64722,9 @@ __ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
6472264722}
6472364723#else
6472464724__ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
64725 int32x2_t __ret;
6472564726 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6472664727 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64727 int32x2_t __ret;
6472864728 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6472964729 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6473064730 return __ret;
......@@ -64739,9 +64739,9 @@ __ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
6473964739}
6474064740#else
6474164741__ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
64742 int16x4_t __ret;
6474264743 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6474364744 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64744 int16x4_t __ret;
6474564745 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 2, 6);
6474664746 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6474764747 return __ret;
......@@ -64756,9 +64756,9 @@ __ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
6475664756}
6475764757#else
6475864758__ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64759 poly8x8_t __ret;
6475964760 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6476064761 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64761 poly8x8_t __ret;
6476264762 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
6476364763 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6476464764 return __ret;
......@@ -64773,9 +64773,9 @@ __ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
6477364773}
6477464774#else
6477564775__ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64776 poly16x4_t __ret;
6477664777 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6477764778 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64778 poly16x4_t __ret;
6477964779 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
6478064780 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6478164781 return __ret;
......@@ -64790,9 +64790,9 @@ __ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
6479064790}
6479164791#else
6479264792__ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64793 poly8x16_t __ret;
6479364794 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6479464795 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64795 poly8x16_t __ret;
6479664796 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
6479764797 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6479864798 return __ret;
......@@ -64807,9 +64807,9 @@ __ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
6480764807}
6480864808#else
6480964809__ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64810 poly64x2_t __ret;
6481064811 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6481164812 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64812 poly64x2_t __ret;
6481364813 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6481464814 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6481564815 return __ret;
......@@ -64824,9 +64824,9 @@ __ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
6482464824}
6482564825#else
6482664826__ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64827 poly16x8_t __ret;
6482764828 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6482864829 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64829 poly16x8_t __ret;
6483064830 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
6483164831 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6483264832 return __ret;
......@@ -64841,9 +64841,9 @@ __ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6484164841}
6484264842#else
6484364843__ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64844 uint8x16_t __ret;
6484464845 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6484564846 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64846 uint8x16_t __ret;
6484764847 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
6484864848 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6484964849 return __ret;
......@@ -64858,9 +64858,9 @@ __ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
6485864858}
6485964859#else
6486064860__ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64861 uint32x4_t __ret;
6486164862 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6486264863 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64863 uint32x4_t __ret;
6486464864 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
6486564865 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6486664866 return __ret;
......@@ -64875,9 +64875,9 @@ __ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
6487564875}
6487664876#else
6487764877__ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64878 uint64x2_t __ret;
6487864879 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6487964880 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64880 uint64x2_t __ret;
6488164881 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6488264882 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6488364883 return __ret;
......@@ -64892,9 +64892,9 @@ __ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
6489264892}
6489364893#else
6489464894__ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64895 uint16x8_t __ret;
6489564896 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6489664897 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64897 uint16x8_t __ret;
6489864898 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
6489964899 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6490064900 return __ret;
......@@ -64909,9 +64909,9 @@ __ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
6490964909}
6491064910#else
6491164911__ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64912 int8x16_t __ret;
6491264913 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6491364914 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64914 int8x16_t __ret;
6491564915 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
6491664916 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6491764917 return __ret;
......@@ -64926,9 +64926,9 @@ __ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
6492664926}
6492764927#else
6492864928__ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64929 float64x2_t __ret;
6492964930 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6493064931 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64931 float64x2_t __ret;
6493264932 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6493364933 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6493464934 return __ret;
......@@ -64943,9 +64943,9 @@ __ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
6494364943}
6494464944#else
6494564945__ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64946 float32x4_t __ret;
6494664947 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6494764948 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64948 float32x4_t __ret;
6494964949 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
6495064950 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6495164951 return __ret;
......@@ -64960,9 +64960,9 @@ __ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
6496064960}
6496164961#else
6496264962__ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64963 int32x4_t __ret;
6496364964 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6496464965 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64965 int32x4_t __ret;
6496664966 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
6496764967 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6496864968 return __ret;
......@@ -64977,9 +64977,9 @@ __ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
6497764977}
6497864978#else
6497964979__ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64980 int64x2_t __ret;
6498064981 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6498164982 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64982 int64x2_t __ret;
6498364983 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6498464984 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6498564985 return __ret;
......@@ -64994,9 +64994,9 @@ __ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
6499464994}
6499564995#else
6499664996__ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64997 int16x8_t __ret;
6499764998 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6499864999 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64999 int16x8_t __ret;
6500065000 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
6500165001 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6500265002 return __ret;
......@@ -65011,9 +65011,9 @@ __ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
6501165011}
6501265012#else
6501365013__ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65014 uint8x8_t __ret;
6501465015 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6501565016 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65016 uint8x8_t __ret;
6501765017 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
6501865018 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6501965019 return __ret;
......@@ -65028,9 +65028,9 @@ __ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
6502865028}
6502965029#else
6503065030__ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65031 uint32x2_t __ret;
6503165032 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6503265033 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65033 uint32x2_t __ret;
6503465034 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6503565035 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6503665036 return __ret;
......@@ -65045,9 +65045,9 @@ __ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
6504565045}
6504665046#else
6504765047__ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65048 uint16x4_t __ret;
6504865049 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6504965050 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65050 uint16x4_t __ret;
6505165051 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
6505265052 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6505365053 return __ret;
......@@ -65062,9 +65062,9 @@ __ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
6506265062}
6506365063#else
6506465064__ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
65065 int8x8_t __ret;
6506565066 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6506665067 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65067 int8x8_t __ret;
6506865068 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 9, 3, 11, 5, 13, 7, 15);
6506965069 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6507065070 return __ret;
......@@ -65079,9 +65079,9 @@ __ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
6507965079}
6508065080#else
6508165081__ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
65082 float32x2_t __ret;
6508265083 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6508365084 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65084 float32x2_t __ret;
6508565085 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6508665086 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6508765087 return __ret;
......@@ -65096,9 +65096,9 @@ __ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
6509665096}
6509765097#else
6509865098__ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
65099 int32x2_t __ret;
6509965100 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6510065101 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65101 int32x2_t __ret;
6510265102 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6510365103 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6510465104 return __ret;
......@@ -65113,9 +65113,9 @@ __ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
6511365113}
6511465114#else
6511565115__ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
65116 int16x4_t __ret;
6511665117 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6511765118 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65118 int16x4_t __ret;
6511965119 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 5, 3, 7);
6512065120 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6512165121 return __ret;
......@@ -65135,9 +65135,9 @@ __ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
6513565135}
6513665136#else
6513765137__ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
65138 uint64x2_t __ret;
6513865139 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6513965140 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65140 uint64x2_t __ret;
6514165141 __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
6514265142 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6514365143 return __ret;
......@@ -65152,9 +65152,9 @@ __ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
6515265152}
6515365153#else
6515465154__ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
65155 uint64x2_t __ret;
6515565156 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6515665157 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65157 uint64x2_t __ret;
6515865158 __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
6515965159 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6516065160 return __ret;
......@@ -65169,9 +65169,9 @@ __ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
6516965169}
6517065170#else
6517165171__ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
65172 uint64x2_t __ret;
6517265173 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6517365174 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65174 uint64x2_t __ret;
6517565175 __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 51);
6517665176 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6517765177 return __ret;
......@@ -65226,9 +65226,9 @@ __ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
6522665226}
6522765227#else
6522865228__ai int8x16_t vuqaddq_s8(int8x16_t __p0, uint8x16_t __p1) {
65229 int8x16_t __ret;
6522965230 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6523065231 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65231 int8x16_t __ret;
6523265232 __ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 32);
6523365233 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6523465234 return __ret;
......@@ -65243,9 +65243,9 @@ __ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
6524365243}
6524465244#else
6524565245__ai int32x4_t vuqaddq_s32(int32x4_t __p0, uint32x4_t __p1) {
65246 int32x4_t __ret;
6524665247 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6524765248 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65248 int32x4_t __ret;
6524965249 __ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 34);
6525065250 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6525165251 return __ret;
......@@ -65260,9 +65260,9 @@ __ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
6526065260}
6526165261#else
6526265262__ai int64x2_t vuqaddq_s64(int64x2_t __p0, uint64x2_t __p1) {
65263 int64x2_t __ret;
6526365264 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6526465265 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65265 int64x2_t __ret;
6526665266 __ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 35);
6526765267 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6526865268 return __ret;
......@@ -65277,9 +65277,9 @@ __ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
6527765277}
6527865278#else
6527965279__ai int16x8_t vuqaddq_s16(int16x8_t __p0, uint16x8_t __p1) {
65280 int16x8_t __ret;
6528065281 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6528165282 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65282 int16x8_t __ret;
6528365283 __ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__rev0, (int8x16_t)__rev1, 33);
6528465284 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6528565285 return __ret;
......@@ -65294,9 +65294,9 @@ __ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
6529465294}
6529565295#else
6529665296__ai int8x8_t vuqadd_s8(int8x8_t __p0, uint8x8_t __p1) {
65297 int8x8_t __ret;
6529765298 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6529865299 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65299 int8x8_t __ret;
6530065300 __ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 0);
6530165301 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6530265302 return __ret;
......@@ -65311,9 +65311,9 @@ __ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
6531165311}
6531265312#else
6531365313__ai int32x2_t vuqadd_s32(int32x2_t __p0, uint32x2_t __p1) {
65314 int32x2_t __ret;
6531465315 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6531565316 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65316 int32x2_t __ret;
6531765317 __ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 2);
6531865318 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6531965319 return __ret;
......@@ -65333,9 +65333,9 @@ __ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
6533365333}
6533465334#else
6533565335__ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
65336 int16x4_t __ret;
6533665337 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6533765338 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65338 int16x4_t __ret;
6533965339 __ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__rev0, (int8x8_t)__rev1, 1);
6534065340 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6534165341 return __ret;
......@@ -65344,23 +65344,23 @@ __ai int16x4_t vuqadd_s16(int16x4_t __p0, uint16x4_t __p1) {
6534465344
6534565345#ifdef __LITTLE_ENDIAN__
6534665346#define vusdotq_laneq_s32(__p0_847, __p1_847, __p2_847, __p3_847) __extension__ ({ \
65347 int32x4_t __ret_847; \
6534765348 int32x4_t __s0_847 = __p0_847; \
6534865349 uint8x16_t __s1_847 = __p1_847; \
6534965350 int8x16_t __s2_847 = __p2_847; \
65350 int32x4_t __ret_847; \
6535165351int8x16_t __reint_847 = __s2_847; \
6535265352 __ret_847 = vusdotq_s32(__s0_847, __s1_847, (int8x16_t)(splatq_laneq_s32(*(int32x4_t *) &__reint_847, __p3_847))); \
6535365353 __ret_847; \
6535465354})
6535565355#else
6535665356#define vusdotq_laneq_s32(__p0_848, __p1_848, __p2_848, __p3_848) __extension__ ({ \
65357 int32x4_t __ret_848; \
6535765358 int32x4_t __s0_848 = __p0_848; \
6535865359 uint8x16_t __s1_848 = __p1_848; \
6535965360 int8x16_t __s2_848 = __p2_848; \
6536065361 int32x4_t __rev0_848; __rev0_848 = __builtin_shufflevector(__s0_848, __s0_848, 3, 2, 1, 0); \
6536165362 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); \
6536265363 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); \
65363 int32x4_t __ret_848; \
6536465364int8x16_t __reint_848 = __rev2_848; \
6536565365 __ret_848 = __noswap_vusdotq_s32(__rev0_848, __rev1_848, (int8x16_t)(__noswap_splatq_laneq_s32(*(int32x4_t *) &__reint_848, __p3_848))); \
6536665366 __ret_848 = __builtin_shufflevector(__ret_848, __ret_848, 3, 2, 1, 0); \
......@@ -65370,23 +65370,23 @@ int8x16_t __reint_848 = __rev2_848; \
6537065370
6537165371#ifdef __LITTLE_ENDIAN__
6537265372#define vusdot_laneq_s32(__p0_849, __p1_849, __p2_849, __p3_849) __extension__ ({ \
65373 int32x2_t __ret_849; \
6537365374 int32x2_t __s0_849 = __p0_849; \
6537465375 uint8x8_t __s1_849 = __p1_849; \
6537565376 int8x16_t __s2_849 = __p2_849; \
65376 int32x2_t __ret_849; \
6537765377int8x16_t __reint_849 = __s2_849; \
6537865378 __ret_849 = vusdot_s32(__s0_849, __s1_849, (int8x8_t)(splat_laneq_s32(*(int32x4_t *) &__reint_849, __p3_849))); \
6537965379 __ret_849; \
6538065380})
6538165381#else
6538265382#define vusdot_laneq_s32(__p0_850, __p1_850, __p2_850, __p3_850) __extension__ ({ \
65383 int32x2_t __ret_850; \
6538365384 int32x2_t __s0_850 = __p0_850; \
6538465385 uint8x8_t __s1_850 = __p1_850; \
6538565386 int8x16_t __s2_850 = __p2_850; \
6538665387 int32x2_t __rev0_850; __rev0_850 = __builtin_shufflevector(__s0_850, __s0_850, 1, 0); \
6538765388 uint8x8_t __rev1_850; __rev1_850 = __builtin_shufflevector(__s1_850, __s1_850, 7, 6, 5, 4, 3, 2, 1, 0); \
6538865389 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); \
65389 int32x2_t __ret_850; \
6539065390int8x16_t __reint_850 = __rev2_850; \
6539165391 __ret_850 = __noswap_vusdot_s32(__rev0_850, __rev1_850, (int8x8_t)(__noswap_splat_laneq_s32(*(int32x4_t *) &__reint_850, __p3_850))); \
6539265392 __ret_850 = __builtin_shufflevector(__ret_850, __ret_850, 1, 0); \
......@@ -65402,9 +65402,9 @@ __ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
6540265402}
6540365403#else
6540465404__ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65405 poly8x8_t __ret;
6540565406 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6540665407 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65407 poly8x8_t __ret;
6540865408 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
6540965409 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6541065410 return __ret;
......@@ -65419,9 +65419,9 @@ __ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
6541965419}
6542065420#else
6542165421__ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65422 poly16x4_t __ret;
6542265423 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6542365424 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65424 poly16x4_t __ret;
6542565425 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
6542665426 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6542765427 return __ret;
......@@ -65436,9 +65436,9 @@ __ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
6543665436}
6543765437#else
6543865438__ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65439 poly8x16_t __ret;
6543965440 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6544065441 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65441 poly8x16_t __ret;
6544265442 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
6544365443 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6544465444 return __ret;
......@@ -65453,9 +65453,9 @@ __ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
6545365453}
6545465454#else
6545565455__ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65456 poly64x2_t __ret;
6545665457 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6545765458 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65458 poly64x2_t __ret;
6545965459 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6546065460 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6546165461 return __ret;
......@@ -65470,9 +65470,9 @@ __ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
6547065470}
6547165471#else
6547265472__ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65473 poly16x8_t __ret;
6547365474 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6547465475 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65475 poly16x8_t __ret;
6547665476 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
6547765477 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6547865478 return __ret;
......@@ -65487,9 +65487,9 @@ __ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6548765487}
6548865488#else
6548965489__ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65490 uint8x16_t __ret;
6549065491 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6549165492 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65492 uint8x16_t __ret;
6549365493 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
6549465494 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6549565495 return __ret;
......@@ -65504,9 +65504,9 @@ __ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
6550465504}
6550565505#else
6550665506__ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65507 uint32x4_t __ret;
6550765508 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6550865509 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65509 uint32x4_t __ret;
6551065510 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
6551165511 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6551265512 return __ret;
......@@ -65521,9 +65521,9 @@ __ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
6552165521}
6552265522#else
6552365523__ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65524 uint64x2_t __ret;
6552465525 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6552565526 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65526 uint64x2_t __ret;
6552765527 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6552865528 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6552965529 return __ret;
......@@ -65538,9 +65538,9 @@ __ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
6553865538}
6553965539#else
6554065540__ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65541 uint16x8_t __ret;
6554165542 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6554265543 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65543 uint16x8_t __ret;
6554465544 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
6554565545 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6554665546 return __ret;
......@@ -65555,9 +65555,9 @@ __ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
6555565555}
6555665556#else
6555765557__ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
65558 int8x16_t __ret;
6555865559 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6555965560 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65560 int8x16_t __ret;
6556165561 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
6556265562 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6556365563 return __ret;
......@@ -65572,9 +65572,9 @@ __ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
6557265572}
6557365573#else
6557465574__ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
65575 float64x2_t __ret;
6557565576 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6557665577 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65577 float64x2_t __ret;
6557865578 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6557965579 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6558065580 return __ret;
......@@ -65589,9 +65589,9 @@ __ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
6558965589}
6559065590#else
6559165591__ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
65592 float32x4_t __ret;
6559265593 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6559365594 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65594 float32x4_t __ret;
6559565595 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
6559665596 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6559765597 return __ret;
......@@ -65606,9 +65606,9 @@ __ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
6560665606}
6560765607#else
6560865608__ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
65609 int32x4_t __ret;
6560965610 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6561065611 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65611 int32x4_t __ret;
6561265612 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
6561365613 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6561465614 return __ret;
......@@ -65623,9 +65623,9 @@ __ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
6562365623}
6562465624#else
6562565625__ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
65626 int64x2_t __ret;
6562665627 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6562765628 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65628 int64x2_t __ret;
6562965629 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6563065630 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6563165631 return __ret;
......@@ -65640,9 +65640,9 @@ __ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
6564065640}
6564165641#else
6564265642__ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
65643 int16x8_t __ret;
6564365644 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6564465645 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65645 int16x8_t __ret;
6564665646 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
6564765647 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6564865648 return __ret;
......@@ -65657,9 +65657,9 @@ __ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
6565765657}
6565865658#else
6565965659__ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65660 uint8x8_t __ret;
6566065661 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6566165662 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65662 uint8x8_t __ret;
6566365663 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
6566465664 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6566565665 return __ret;
......@@ -65674,9 +65674,9 @@ __ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
6567465674}
6567565675#else
6567665676__ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65677 uint32x2_t __ret;
6567765678 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6567865679 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65679 uint32x2_t __ret;
6568065680 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6568165681 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6568265682 return __ret;
......@@ -65691,9 +65691,9 @@ __ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
6569165691}
6569265692#else
6569365693__ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65694 uint16x4_t __ret;
6569465695 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6569565696 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65696 uint16x4_t __ret;
6569765697 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
6569865698 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6569965699 return __ret;
......@@ -65708,9 +65708,9 @@ __ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
6570865708}
6570965709#else
6571065710__ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
65711 int8x8_t __ret;
6571165712 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6571265713 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65713 int8x8_t __ret;
6571465714 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6, 8, 10, 12, 14);
6571565715 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6571665716 return __ret;
......@@ -65725,9 +65725,9 @@ __ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
6572565725}
6572665726#else
6572765727__ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
65728 float32x2_t __ret;
6572865729 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6572965730 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65730 float32x2_t __ret;
6573165731 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6573265732 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6573365733 return __ret;
......@@ -65742,9 +65742,9 @@ __ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
6574265742}
6574365743#else
6574465744__ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
65745 int32x2_t __ret;
6574565746 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6574665747 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65747 int32x2_t __ret;
6574865748 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6574965749 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6575065750 return __ret;
......@@ -65759,9 +65759,9 @@ __ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
6575965759}
6576065760#else
6576165761__ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
65762 int16x4_t __ret;
6576265763 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6576365764 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65764 int16x4_t __ret;
6576565765 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2, 4, 6);
6576665766 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6576765767 return __ret;
......@@ -65776,9 +65776,9 @@ __ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
6577665776}
6577765777#else
6577865778__ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65779 poly8x8_t __ret;
6577965780 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6578065781 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65781 poly8x8_t __ret;
6578265782 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
6578365783 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6578465784 return __ret;
......@@ -65793,9 +65793,9 @@ __ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
6579365793}
6579465794#else
6579565795__ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65796 poly16x4_t __ret;
6579665797 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6579765798 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65798 poly16x4_t __ret;
6579965799 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
6580065800 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6580165801 return __ret;
......@@ -65810,9 +65810,9 @@ __ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
6581065810}
6581165811#else
6581265812__ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65813 poly8x16_t __ret;
6581365814 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6581465815 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65815 poly8x16_t __ret;
6581665816 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
6581765817 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6581865818 return __ret;
......@@ -65827,9 +65827,9 @@ __ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
6582765827}
6582865828#else
6582965829__ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65830 poly64x2_t __ret;
6583065831 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6583165832 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65832 poly64x2_t __ret;
6583365833 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6583465834 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6583565835 return __ret;
......@@ -65844,9 +65844,9 @@ __ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
6584465844}
6584565845#else
6584665846__ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65847 poly16x8_t __ret;
6584765848 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6584865849 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65849 poly16x8_t __ret;
6585065850 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
6585165851 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6585265852 return __ret;
......@@ -65861,9 +65861,9 @@ __ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6586165861}
6586265862#else
6586365863__ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65864 uint8x16_t __ret;
6586465865 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6586565866 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65866 uint8x16_t __ret;
6586765867 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
6586865868 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6586965869 return __ret;
......@@ -65878,9 +65878,9 @@ __ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
6587865878}
6587965879#else
6588065880__ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65881 uint32x4_t __ret;
6588165882 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6588265883 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65883 uint32x4_t __ret;
6588465884 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
6588565885 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6588665886 return __ret;
......@@ -65895,9 +65895,9 @@ __ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
6589565895}
6589665896#else
6589765897__ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65898 uint64x2_t __ret;
6589865899 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6589965900 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65900 uint64x2_t __ret;
6590165901 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6590265902 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6590365903 return __ret;
......@@ -65912,9 +65912,9 @@ __ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
6591265912}
6591365913#else
6591465914__ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65915 uint16x8_t __ret;
6591565916 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6591665917 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65917 uint16x8_t __ret;
6591865918 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
6591965919 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6592065920 return __ret;
......@@ -65929,9 +65929,9 @@ __ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
6592965929}
6593065930#else
6593165931__ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65932 int8x16_t __ret;
6593265933 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6593365934 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65934 int8x16_t __ret;
6593565935 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
6593665936 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6593765937 return __ret;
......@@ -65946,9 +65946,9 @@ __ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
6594665946}
6594765947#else
6594865948__ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65949 float64x2_t __ret;
6594965950 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6595065951 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65951 float64x2_t __ret;
6595265952 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6595365953 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6595465954 return __ret;
......@@ -65963,9 +65963,9 @@ __ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
6596365963}
6596465964#else
6596565965__ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65966 float32x4_t __ret;
6596665967 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6596765968 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65968 float32x4_t __ret;
6596965969 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
6597065970 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6597165971 return __ret;
......@@ -65980,9 +65980,9 @@ __ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
6598065980}
6598165981#else
6598265982__ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65983 int32x4_t __ret;
6598365984 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6598465985 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65985 int32x4_t __ret;
6598665986 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
6598765987 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6598865988 return __ret;
......@@ -65997,9 +65997,9 @@ __ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
6599765997}
6599865998#else
6599965999__ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
66000 int64x2_t __ret;
6600066001 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6600166002 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66002 int64x2_t __ret;
6600366003 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6600466004 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6600566005 return __ret;
......@@ -66014,9 +66014,9 @@ __ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
6601466014}
6601566015#else
6601666016__ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
66017 int16x8_t __ret;
6601766018 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6601866019 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66019 int16x8_t __ret;
6602066020 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
6602166021 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6602266022 return __ret;
......@@ -66031,9 +66031,9 @@ __ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
6603166031}
6603266032#else
6603366033__ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66034 uint8x8_t __ret;
6603466035 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6603566036 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66036 uint8x8_t __ret;
6603766037 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
6603866038 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6603966039 return __ret;
......@@ -66048,9 +66048,9 @@ __ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
6604866048}
6604966049#else
6605066050__ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66051 uint32x2_t __ret;
6605166052 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6605266053 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66053 uint32x2_t __ret;
6605466054 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6605566055 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6605666056 return __ret;
......@@ -66065,9 +66065,9 @@ __ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
6606566065}
6606666066#else
6606766067__ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66068 uint16x4_t __ret;
6606866069 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6606966070 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66070 uint16x4_t __ret;
6607166071 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
6607266072 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6607366073 return __ret;
......@@ -66082,9 +66082,9 @@ __ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
6608266082}
6608366083#else
6608466084__ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
66085 int8x8_t __ret;
6608566086 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6608666087 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66087 int8x8_t __ret;
6608866088 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7, 9, 11, 13, 15);
6608966089 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6609066090 return __ret;
......@@ -66099,9 +66099,9 @@ __ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
6609966099}
6610066100#else
6610166101__ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
66102 float32x2_t __ret;
6610266103 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6610366104 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66104 float32x2_t __ret;
6610566105 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6610666106 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6610766107 return __ret;
......@@ -66116,9 +66116,9 @@ __ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
6611666116}
6611766117#else
6611866118__ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
66119 int32x2_t __ret;
6611966120 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6612066121 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66121 int32x2_t __ret;
6612266122 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6612366123 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6612466124 return __ret;
......@@ -66133,9 +66133,9 @@ __ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
6613366133}
6613466134#else
6613566135__ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
66136 int16x4_t __ret;
6613666137 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6613766138 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66138 int16x4_t __ret;
6613966139 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3, 5, 7);
6614066140 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6614166141 return __ret;
......@@ -66150,9 +66150,9 @@ __ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
6615066150}
6615166151#else
6615266152__ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
66153 poly8x8_t __ret;
6615366154 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6615466155 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66155 poly8x8_t __ret;
6615666156 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
6615766157 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6615866158 return __ret;
......@@ -66167,9 +66167,9 @@ __ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
6616766167}
6616866168#else
6616966169__ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
66170 poly16x4_t __ret;
6617066171 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6617166172 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66172 poly16x4_t __ret;
6617366173 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
6617466174 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6617566175 return __ret;
......@@ -66184,9 +66184,9 @@ __ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
6618466184}
6618566185#else
6618666186__ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66187 poly8x16_t __ret;
6618766188 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6618866189 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66189 poly8x16_t __ret;
6619066190 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
6619166191 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6619266192 return __ret;
......@@ -66201,9 +66201,9 @@ __ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
6620166201}
6620266202#else
6620366203__ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66204 poly64x2_t __ret;
6620466205 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6620566206 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66206 poly64x2_t __ret;
6620766207 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6620866208 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6620966209 return __ret;
......@@ -66218,9 +66218,9 @@ __ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
6621866218}
6621966219#else
6622066220__ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66221 poly16x8_t __ret;
6622166222 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6622266223 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66223 poly16x8_t __ret;
6622466224 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
6622566225 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6622666226 return __ret;
......@@ -66235,9 +66235,9 @@ __ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6623566235}
6623666236#else
6623766237__ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66238 uint8x16_t __ret;
6623866239 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6623966240 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66240 uint8x16_t __ret;
6624166241 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
6624266242 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6624366243 return __ret;
......@@ -66252,9 +66252,9 @@ __ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
6625266252}
6625366253#else
6625466254__ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66255 uint32x4_t __ret;
6625566256 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6625666257 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66257 uint32x4_t __ret;
6625866258 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
6625966259 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6626066260 return __ret;
......@@ -66269,9 +66269,9 @@ __ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
6626966269}
6627066270#else
6627166271__ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66272 uint64x2_t __ret;
6627266273 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6627366274 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66274 uint64x2_t __ret;
6627566275 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6627666276 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6627766277 return __ret;
......@@ -66286,9 +66286,9 @@ __ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
6628666286}
6628766287#else
6628866288__ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66289 uint16x8_t __ret;
6628966290 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6629066291 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66291 uint16x8_t __ret;
6629266292 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
6629366293 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6629466294 return __ret;
......@@ -66303,9 +66303,9 @@ __ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
6630366303}
6630466304#else
6630566305__ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
66306 int8x16_t __ret;
6630666307 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6630766308 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66308 int8x16_t __ret;
6630966309 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
6631066310 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6631166311 return __ret;
......@@ -66320,9 +66320,9 @@ __ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
6632066320}
6632166321#else
6632266322__ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
66323 float64x2_t __ret;
6632366324 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6632466325 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66325 float64x2_t __ret;
6632666326 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6632766327 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6632866328 return __ret;
......@@ -66337,9 +66337,9 @@ __ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
6633766337}
6633866338#else
6633966339__ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
66340 float32x4_t __ret;
6634066341 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6634166342 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66342 float32x4_t __ret;
6634366343 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
6634466344 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6634566345 return __ret;
......@@ -66354,9 +66354,9 @@ __ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
6635466354}
6635566355#else
6635666356__ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
66357 int32x4_t __ret;
6635766358 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6635866359 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66359 int32x4_t __ret;
6636066360 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
6636166361 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6636266362 return __ret;
......@@ -66371,9 +66371,9 @@ __ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
6637166371}
6637266372#else
6637366373__ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
66374 int64x2_t __ret;
6637466375 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6637566376 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66376 int64x2_t __ret;
6637766377 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6637866378 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6637966379 return __ret;
......@@ -66388,9 +66388,9 @@ __ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
6638866388}
6638966389#else
6639066390__ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
66391 int16x8_t __ret;
6639166392 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6639266393 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66393 int16x8_t __ret;
6639466394 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
6639566395 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6639666396 return __ret;
......@@ -66405,9 +66405,9 @@ __ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
6640566405}
6640666406#else
6640766407__ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
66408 uint8x8_t __ret;
6640866409 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6640966410 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66410 uint8x8_t __ret;
6641166411 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
6641266412 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6641366413 return __ret;
......@@ -66422,9 +66422,9 @@ __ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
6642266422}
6642366423#else
6642466424__ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
66425 uint32x2_t __ret;
6642566426 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6642666427 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66427 uint32x2_t __ret;
6642866428 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6642966429 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6643066430 return __ret;
......@@ -66439,9 +66439,9 @@ __ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
6643966439}
6644066440#else
6644166441__ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
66442 uint16x4_t __ret;
6644266443 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6644366444 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66444 uint16x4_t __ret;
6644566445 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
6644666446 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6644766447 return __ret;
......@@ -66456,9 +66456,9 @@ __ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
6645666456}
6645766457#else
6645866458__ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
66459 int8x8_t __ret;
6645966460 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6646066461 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66461 int8x8_t __ret;
6646266462 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 8, 1, 9, 2, 10, 3, 11);
6646366463 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6646466464 return __ret;
......@@ -66473,9 +66473,9 @@ __ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
6647366473}
6647466474#else
6647566475__ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
66476 float32x2_t __ret;
6647666477 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6647766478 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66478 float32x2_t __ret;
6647966479 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6648066480 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6648166481 return __ret;
......@@ -66490,9 +66490,9 @@ __ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
6649066490}
6649166491#else
6649266492__ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
66493 int32x2_t __ret;
6649366494 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6649466495 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66495 int32x2_t __ret;
6649666496 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 2);
6649766497 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6649866498 return __ret;
......@@ -66507,9 +66507,9 @@ __ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
6650766507}
6650866508#else
6650966509__ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
66510 int16x4_t __ret;
6651066511 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6651166512 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66512 int16x4_t __ret;
6651366513 __ret = __builtin_shufflevector(__rev0, __rev1, 0, 4, 1, 5);
6651466514 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6651566515 return __ret;
......@@ -66524,9 +66524,9 @@ __ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
6652466524}
6652566525#else
6652666526__ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
66527 poly8x8_t __ret;
6652766528 poly8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6652866529 poly8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66529 poly8x8_t __ret;
6653066530 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
6653166531 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6653266532 return __ret;
......@@ -66541,9 +66541,9 @@ __ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
6654166541}
6654266542#else
6654366543__ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
66544 poly16x4_t __ret;
6654466545 poly16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6654566546 poly16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66546 poly16x4_t __ret;
6654766547 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
6654866548 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6654966549 return __ret;
......@@ -66558,9 +66558,9 @@ __ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
6655866558}
6655966559#else
6656066560__ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66561 poly8x16_t __ret;
6656166562 poly8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6656266563 poly8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66563 poly8x16_t __ret;
6656466564 __ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
6656566565 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6656666566 return __ret;
......@@ -66575,9 +66575,9 @@ __ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
6657566575}
6657666576#else
6657766577__ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66578 poly64x2_t __ret;
6657866579 poly64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6657966580 poly64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66580 poly64x2_t __ret;
6658166581 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6658266582 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6658366583 return __ret;
......@@ -66592,9 +66592,9 @@ __ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
6659266592}
6659366593#else
6659466594__ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66595 poly16x8_t __ret;
6659566596 poly16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6659666597 poly16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66597 poly16x8_t __ret;
6659866598 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
6659966599 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6660066600 return __ret;
......@@ -66609,9 +66609,9 @@ __ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
6660966609}
6661066610#else
6661166611__ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66612 uint8x16_t __ret;
6661266613 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6661366614 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66614 uint8x16_t __ret;
6661566615 __ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
6661666616 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6661766617 return __ret;
......@@ -66626,9 +66626,9 @@ __ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
6662666626}
6662766627#else
6662866628__ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66629 uint32x4_t __ret;
6662966630 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6663066631 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66631 uint32x4_t __ret;
6663266632 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
6663366633 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6663466634 return __ret;
......@@ -66643,9 +66643,9 @@ __ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
6664366643}
6664466644#else
6664566645__ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66646 uint64x2_t __ret;
6664666647 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6664766648 uint64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66648 uint64x2_t __ret;
6664966649 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6665066650 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6665166651 return __ret;
......@@ -66660,9 +66660,9 @@ __ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
6666066660}
6666166661#else
6666266662__ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66663 uint16x8_t __ret;
6666366664 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6666466665 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66665 uint16x8_t __ret;
6666666666 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
6666766667 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6666866668 return __ret;
......@@ -66677,9 +66677,9 @@ __ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
6667766677}
6667866678#else
6667966679__ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
66680 int8x16_t __ret;
6668066681 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6668166682 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66682 int8x16_t __ret;
6668366683 __ret = __builtin_shufflevector(__rev0, __rev1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
6668466684 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6668566685 return __ret;
......@@ -66694,9 +66694,9 @@ __ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
6669466694}
6669566695#else
6669666696__ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
66697 float64x2_t __ret;
6669766698 float64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6669866699 float64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66699 float64x2_t __ret;
6670066700 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6670166701 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6670266702 return __ret;
......@@ -66711,9 +66711,9 @@ __ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
6671166711}
6671266712#else
6671366713__ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
66714 float32x4_t __ret;
6671466715 float32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6671566716 float32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66716 float32x4_t __ret;
6671766717 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
6671866718 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6671966719 return __ret;
......@@ -66728,9 +66728,9 @@ __ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
6672866728}
6672966729#else
6673066730__ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
66731 int32x4_t __ret;
6673166732 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6673266733 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66733 int32x4_t __ret;
6673466734 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
6673566735 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6673666736 return __ret;
......@@ -66745,9 +66745,9 @@ __ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
6674566745}
6674666746#else
6674766747__ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
66748 int64x2_t __ret;
6674866749 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6674966750 int64x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66750 int64x2_t __ret;
6675166751 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6675266752 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6675366753 return __ret;
......@@ -66762,9 +66762,9 @@ __ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
6676266762}
6676366763#else
6676466764__ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
66765 int16x8_t __ret;
6676566766 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6676666767 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66767 int16x8_t __ret;
6676866768 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
6676966769 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6677066770 return __ret;
......@@ -66779,9 +66779,9 @@ __ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
6677966779}
6678066780#else
6678166781__ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66782 uint8x8_t __ret;
6678266783 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6678366784 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66784 uint8x8_t __ret;
6678566785 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
6678666786 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6678766787 return __ret;
......@@ -66796,9 +66796,9 @@ __ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
6679666796}
6679766797#else
6679866798__ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66799 uint32x2_t __ret;
6679966800 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6680066801 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66801 uint32x2_t __ret;
6680266802 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6680366803 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6680466804 return __ret;
......@@ -66813,9 +66813,9 @@ __ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
6681366813}
6681466814#else
6681566815__ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66816 uint16x4_t __ret;
6681666817 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6681766818 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66818 uint16x4_t __ret;
6681966819 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
6682066820 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6682166821 return __ret;
......@@ -66830,9 +66830,9 @@ __ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
6683066830}
6683166831#else
6683266832__ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66833 int8x8_t __ret;
6683366834 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6683466835 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66835 int8x8_t __ret;
6683666836 __ret = __builtin_shufflevector(__rev0, __rev1, 4, 12, 5, 13, 6, 14, 7, 15);
6683766837 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6683866838 return __ret;
......@@ -66847,9 +66847,9 @@ __ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
6684766847}
6684866848#else
6684966849__ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66850 float32x2_t __ret;
6685066851 float32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6685166852 float32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66852 float32x2_t __ret;
6685366853 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6685466854 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6685566855 return __ret;
......@@ -66864,9 +66864,9 @@ __ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
6686466864}
6686566865#else
6686666866__ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66867 int32x2_t __ret;
6686766868 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6686866869 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66869 int32x2_t __ret;
6687066870 __ret = __builtin_shufflevector(__rev0, __rev1, 1, 3);
6687166871 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6687266872 return __ret;
......@@ -66881,9 +66881,9 @@ __ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
6688166881}
6688266882#else
6688366883__ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66884 int16x4_t __ret;
6688466885 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6688566886 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66886 int16x4_t __ret;
6688766887 __ret = __builtin_shufflevector(__rev0, __rev1, 2, 6, 3, 7);
6688866888 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6688966889 return __ret;
......@@ -66899,10 +66899,10 @@ __ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
6689966899}
6690066900#else
6690166901__ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
66902 uint8x16_t __ret;
6690266903 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6690366904 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6690466905 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66905 uint8x16_t __ret;
6690666906 __ret = __rev0 + __noswap_vabdq_u8(__rev1, __rev2);
6690766907 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6690866908 return __ret;
......@@ -66917,10 +66917,10 @@ __ai uint32x4_t vabaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
6691766917}
6691866918#else
6691966919__ai uint32x4_t vabaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
66920 uint32x4_t __ret;
6692066921 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6692166922 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6692266923 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
66923 uint32x4_t __ret;
6692466924 __ret = __rev0 + __noswap_vabdq_u32(__rev1, __rev2);
6692566925 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6692666926 return __ret;
......@@ -66935,10 +66935,10 @@ __ai uint16x8_t vabaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
6693566935}
6693666936#else
6693766937__ai uint16x8_t vabaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
66938 uint16x8_t __ret;
6693866939 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6693966940 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6694066941 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
66941 uint16x8_t __ret;
6694266942 __ret = __rev0 + __noswap_vabdq_u16(__rev1, __rev2);
6694366943 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6694466944 return __ret;
......@@ -66953,10 +66953,10 @@ __ai int8x16_t vabaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
6695366953}
6695466954#else
6695566955__ai int8x16_t vabaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
66956 int8x16_t __ret;
6695666957 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6695766958 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6695866959 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66959 int8x16_t __ret;
6696066960 __ret = __rev0 + __noswap_vabdq_s8(__rev1, __rev2);
6696166961 __ret = __builtin_shufflevector(__ret, __ret, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6696266962 return __ret;
......@@ -66971,10 +66971,10 @@ __ai int32x4_t vabaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
6697166971}
6697266972#else
6697366973__ai int32x4_t vabaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
66974 int32x4_t __ret;
6697466975 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6697566976 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6697666977 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
66977 int32x4_t __ret;
6697866978 __ret = __rev0 + __noswap_vabdq_s32(__rev1, __rev2);
6697966979 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6698066980 return __ret;
......@@ -66989,10 +66989,10 @@ __ai int16x8_t vabaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
6698966989}
6699066990#else
6699166991__ai int16x8_t vabaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
66992 int16x8_t __ret;
6699266993 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6699366994 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6699466995 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
66995 int16x8_t __ret;
6699666996 __ret = __rev0 + __noswap_vabdq_s16(__rev1, __rev2);
6699766997 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6699866998 return __ret;
......@@ -67007,10 +67007,10 @@ __ai uint8x8_t vaba_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
6700767007}
6700867008#else
6700967009__ai uint8x8_t vaba_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67010 uint8x8_t __ret;
6701067011 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6701167012 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6701267013 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
67013 uint8x8_t __ret;
6701467014 __ret = __rev0 + __noswap_vabd_u8(__rev1, __rev2);
6701567015 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6701667016 return __ret;
......@@ -67025,10 +67025,10 @@ __ai uint32x2_t vaba_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
6702567025}
6702667026#else
6702767027__ai uint32x2_t vaba_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67028 uint32x2_t __ret;
6702867029 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6702967030 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6703067031 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
67031 uint32x2_t __ret;
6703267032 __ret = __rev0 + __noswap_vabd_u32(__rev1, __rev2);
6703367033 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6703467034 return __ret;
......@@ -67043,10 +67043,10 @@ __ai uint16x4_t vaba_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
6704367043}
6704467044#else
6704567045__ai uint16x4_t vaba_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67046 uint16x4_t __ret;
6704667047 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6704767048 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6704867049 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
67049 uint16x4_t __ret;
6705067050 __ret = __rev0 + __noswap_vabd_u16(__rev1, __rev2);
6705167051 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6705267052 return __ret;
......@@ -67061,10 +67061,10 @@ __ai int8x8_t vaba_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
6706167061}
6706267062#else
6706367063__ai int8x8_t vaba_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67064 int8x8_t __ret;
6706467065 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6706567066 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6706667067 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
67067 int8x8_t __ret;
6706867068 __ret = __rev0 + __noswap_vabd_s8(__rev1, __rev2);
6706967069 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6707067070 return __ret;
......@@ -67079,10 +67079,10 @@ __ai int32x2_t vaba_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
6707967079}
6708067080#else
6708167081__ai int32x2_t vaba_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67082 int32x2_t __ret;
6708267083 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6708367084 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6708467085 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
67085 int32x2_t __ret;
6708667086 __ret = __rev0 + __noswap_vabd_s32(__rev1, __rev2);
6708767087 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6708867088 return __ret;
......@@ -67097,10 +67097,10 @@ __ai int16x4_t vaba_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
6709767097}
6709867098#else
6709967099__ai int16x4_t vaba_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67100 int16x4_t __ret;
6710067101 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6710167102 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6710267103 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
67103 int16x4_t __ret;
6710467104 __ret = __rev0 + __noswap_vabd_s16(__rev1, __rev2);
6710567105 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6710667106 return __ret;
......@@ -67115,9 +67115,9 @@ __ai uint16x8_t vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
6711567115}
6711667116#else
6711767117__ai uint16x8_t vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
67118 uint16x8_t __ret;
6711867119 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6711967120 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67120 uint16x8_t __ret;
6712167121 __ret = (uint16x8_t)(__noswap_vmovl_u8((uint8x8_t)(__noswap_vabd_u8(__rev0, __rev1))));
6712267122 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6712367123 return __ret;
......@@ -67137,9 +67137,9 @@ __ai uint64x2_t vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
6713767137}
6713867138#else
6713967139__ai uint64x2_t vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
67140 uint64x2_t __ret;
6714067141 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6714167142 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67142 uint64x2_t __ret;
6714367143 __ret = (uint64x2_t)(__noswap_vmovl_u32((uint32x2_t)(__noswap_vabd_u32(__rev0, __rev1))));
6714467144 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6714567145 return __ret;
......@@ -67159,9 +67159,9 @@ __ai uint32x4_t vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
6715967159}
6716067160#else
6716167161__ai uint32x4_t vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
67162 uint32x4_t __ret;
6716267163 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6716367164 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67164 uint32x4_t __ret;
6716567165 __ret = (uint32x4_t)(__noswap_vmovl_u16((uint16x4_t)(__noswap_vabd_u16(__rev0, __rev1))));
6716667166 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6716767167 return __ret;
......@@ -67181,9 +67181,9 @@ __ai int16x8_t vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
6718167181}
6718267182#else
6718367183__ai int16x8_t vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
67184 int16x8_t __ret;
6718467185 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6718567186 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67186 int16x8_t __ret;
6718767187 __ret = (int16x8_t)(__noswap_vmovl_u8((uint8x8_t)(__noswap_vabd_s8(__rev0, __rev1))));
6718867188 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6718967189 return __ret;
......@@ -67203,9 +67203,9 @@ __ai int64x2_t vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
6720367203}
6720467204#else
6720567205__ai int64x2_t vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
67206 int64x2_t __ret;
6720667207 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6720767208 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67208 int64x2_t __ret;
6720967209 __ret = (int64x2_t)(__noswap_vmovl_u32((uint32x2_t)(__noswap_vabd_s32(__rev0, __rev1))));
6721067210 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6721167211 return __ret;
......@@ -67225,9 +67225,9 @@ __ai int32x4_t vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
6722567225}
6722667226#else
6722767227__ai int32x4_t vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
67228 int32x4_t __ret;
6722867229 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6722967230 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67230 int32x4_t __ret;
6723167231 __ret = (int32x4_t)(__noswap_vmovl_u16((uint16x4_t)(__noswap_vabd_s16(__rev0, __rev1))));
6723267232 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6723367233 return __ret;
......@@ -67247,9 +67247,9 @@ __ai uint16x8_t vaddl_u8(uint8x8_t __p0, uint8x8_t __p1) {
6724767247}
6724867248#else
6724967249__ai uint16x8_t vaddl_u8(uint8x8_t __p0, uint8x8_t __p1) {
67250 uint16x8_t __ret;
6725067251 uint8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6725167252 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67252 uint16x8_t __ret;
6725367253 __ret = __noswap_vmovl_u8(__rev0) + __noswap_vmovl_u8(__rev1);
6725467254 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6725567255 return __ret;
......@@ -67264,9 +67264,9 @@ __ai uint64x2_t vaddl_u32(uint32x2_t __p0, uint32x2_t __p1) {
6726467264}
6726567265#else
6726667266__ai uint64x2_t vaddl_u32(uint32x2_t __p0, uint32x2_t __p1) {
67267 uint64x2_t __ret;
6726767268 uint32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6726867269 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67269 uint64x2_t __ret;
6727067270 __ret = __noswap_vmovl_u32(__rev0) + __noswap_vmovl_u32(__rev1);
6727167271 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6727267272 return __ret;
......@@ -67281,9 +67281,9 @@ __ai uint32x4_t vaddl_u16(uint16x4_t __p0, uint16x4_t __p1) {
6728167281}
6728267282#else
6728367283__ai uint32x4_t vaddl_u16(uint16x4_t __p0, uint16x4_t __p1) {
67284 uint32x4_t __ret;
6728467285 uint16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6728567286 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67286 uint32x4_t __ret;
6728767287 __ret = __noswap_vmovl_u16(__rev0) + __noswap_vmovl_u16(__rev1);
6728867288 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6728967289 return __ret;
......@@ -67298,9 +67298,9 @@ __ai int16x8_t vaddl_s8(int8x8_t __p0, int8x8_t __p1) {
6729867298}
6729967299#else
6730067300__ai int16x8_t vaddl_s8(int8x8_t __p0, int8x8_t __p1) {
67301 int16x8_t __ret;
6730167302 int8x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6730267303 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67303 int16x8_t __ret;
6730467304 __ret = __noswap_vmovl_s8(__rev0) + __noswap_vmovl_s8(__rev1);
6730567305 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6730667306 return __ret;
......@@ -67315,9 +67315,9 @@ __ai int64x2_t vaddl_s32(int32x2_t __p0, int32x2_t __p1) {
6731567315}
6731667316#else
6731767317__ai int64x2_t vaddl_s32(int32x2_t __p0, int32x2_t __p1) {
67318 int64x2_t __ret;
6731867319 int32x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6731967320 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67320 int64x2_t __ret;
6732167321 __ret = __noswap_vmovl_s32(__rev0) + __noswap_vmovl_s32(__rev1);
6732267322 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6732367323 return __ret;
......@@ -67332,9 +67332,9 @@ __ai int32x4_t vaddl_s16(int16x4_t __p0, int16x4_t __p1) {
6733267332}
6733367333#else
6733467334__ai int32x4_t vaddl_s16(int16x4_t __p0, int16x4_t __p1) {
67335 int32x4_t __ret;
6733567336 int16x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6733667337 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67337 int32x4_t __ret;
6733867338 __ret = __noswap_vmovl_s16(__rev0) + __noswap_vmovl_s16(__rev1);
6733967339 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6734067340 return __ret;
......@@ -67349,9 +67349,9 @@ __ai uint16x8_t vaddw_u8(uint16x8_t __p0, uint8x8_t __p1) {
6734967349}
6735067350#else
6735167351__ai uint16x8_t vaddw_u8(uint16x8_t __p0, uint8x8_t __p1) {
67352 uint16x8_t __ret;
6735267353 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6735367354 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67354 uint16x8_t __ret;
6735567355 __ret = __rev0 + __noswap_vmovl_u8(__rev1);
6735667356 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6735767357 return __ret;
......@@ -67366,9 +67366,9 @@ __ai uint64x2_t vaddw_u32(uint64x2_t __p0, uint32x2_t __p1) {
6736667366}
6736767367#else
6736867368__ai uint64x2_t vaddw_u32(uint64x2_t __p0, uint32x2_t __p1) {
67369 uint64x2_t __ret;
6736967370 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6737067371 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67371 uint64x2_t __ret;
6737267372 __ret = __rev0 + __noswap_vmovl_u32(__rev1);
6737367373 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6737467374 return __ret;
......@@ -67383,9 +67383,9 @@ __ai uint32x4_t vaddw_u16(uint32x4_t __p0, uint16x4_t __p1) {
6738367383}
6738467384#else
6738567385__ai uint32x4_t vaddw_u16(uint32x4_t __p0, uint16x4_t __p1) {
67386 uint32x4_t __ret;
6738667387 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6738767388 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67388 uint32x4_t __ret;
6738967389 __ret = __rev0 + __noswap_vmovl_u16(__rev1);
6739067390 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6739167391 return __ret;
......@@ -67400,9 +67400,9 @@ __ai int16x8_t vaddw_s8(int16x8_t __p0, int8x8_t __p1) {
6740067400}
6740167401#else
6740267402__ai int16x8_t vaddw_s8(int16x8_t __p0, int8x8_t __p1) {
67403 int16x8_t __ret;
6740367404 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6740467405 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67405 int16x8_t __ret;
6740667406 __ret = __rev0 + __noswap_vmovl_s8(__rev1);
6740767407 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6740867408 return __ret;
......@@ -67417,9 +67417,9 @@ __ai int64x2_t vaddw_s32(int64x2_t __p0, int32x2_t __p1) {
6741767417}
6741867418#else
6741967419__ai int64x2_t vaddw_s32(int64x2_t __p0, int32x2_t __p1) {
67420 int64x2_t __ret;
6742067421 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6742167422 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67422 int64x2_t __ret;
6742367423 __ret = __rev0 + __noswap_vmovl_s32(__rev1);
6742467424 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6742567425 return __ret;
......@@ -67434,9 +67434,9 @@ __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
6743467434}
6743567435#else
6743667436__ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
67437 int32x4_t __ret;
6743767438 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6743867439 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67439 int32x4_t __ret;
6744067440 __ret = __rev0 + __noswap_vmovl_s16(__rev1);
6744167441 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6744267442 return __ret;
......@@ -67445,8 +67445,8 @@ __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
6744567445
6744667446#ifdef __LITTLE_ENDIAN__
6744767447#define vget_lane_f16(__p0_851, __p1_851) __extension__ ({ \
67448 float16x4_t __s0_851 = __p0_851; \
6744967448 float16_t __ret_851; \
67449 float16x4_t __s0_851 = __p0_851; \
6745067450float16x4_t __reint_851 = __s0_851; \
6745167451int16_t __reint1_851 = vget_lane_s16(*(int16x4_t *) &__reint_851, __p1_851); \
6745267452 __ret_851 = *(float16_t *) &__reint1_851; \
......@@ -67454,17 +67454,17 @@ int16_t __reint1_851 = vget_lane_s16(*(int16x4_t *) &__reint_851, __p1_851); \
6745467454})
6745567455#else
6745667456#define vget_lane_f16(__p0_852, __p1_852) __extension__ ({ \
67457 float16_t __ret_852; \
6745767458 float16x4_t __s0_852 = __p0_852; \
6745867459 float16x4_t __rev0_852; __rev0_852 = __builtin_shufflevector(__s0_852, __s0_852, 3, 2, 1, 0); \
67459 float16_t __ret_852; \
6746067460float16x4_t __reint_852 = __rev0_852; \
6746167461int16_t __reint1_852 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_852, __p1_852); \
6746267462 __ret_852 = *(float16_t *) &__reint1_852; \
6746367463 __ret_852; \
6746467464})
6746567465#define __noswap_vget_lane_f16(__p0_853, __p1_853) __extension__ ({ \
67466 float16x4_t __s0_853 = __p0_853; \
6746767466 float16_t __ret_853; \
67467 float16x4_t __s0_853 = __p0_853; \
6746867468float16x4_t __reint_853 = __s0_853; \
6746967469int16_t __reint1_853 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_853, __p1_853); \
6747067470 __ret_853 = *(float16_t *) &__reint1_853; \
......@@ -67474,8 +67474,8 @@ int16_t __reint1_853 = __noswap_vget_lane_s16(*(int16x4_t *) &__reint_853, __p1_
6747467474
6747567475#ifdef __LITTLE_ENDIAN__
6747667476#define vgetq_lane_f16(__p0_854, __p1_854) __extension__ ({ \
67477 float16x8_t __s0_854 = __p0_854; \
6747867477 float16_t __ret_854; \
67478 float16x8_t __s0_854 = __p0_854; \
6747967479float16x8_t __reint_854 = __s0_854; \
6748067480int16_t __reint1_854 = vgetq_lane_s16(*(int16x8_t *) &__reint_854, __p1_854); \
6748167481 __ret_854 = *(float16_t *) &__reint1_854; \
......@@ -67483,17 +67483,17 @@ int16_t __reint1_854 = vgetq_lane_s16(*(int16x8_t *) &__reint_854, __p1_854); \
6748367483})
6748467484#else
6748567485#define vgetq_lane_f16(__p0_855, __p1_855) __extension__ ({ \
67486 float16_t __ret_855; \
6748667487 float16x8_t __s0_855 = __p0_855; \
6748767488 float16x8_t __rev0_855; __rev0_855 = __builtin_shufflevector(__s0_855, __s0_855, 7, 6, 5, 4, 3, 2, 1, 0); \
67488 float16_t __ret_855; \
6748967489float16x8_t __reint_855 = __rev0_855; \
6749067490int16_t __reint1_855 = __noswap_vgetq_lane_s16(*(int16x8_t *) &__reint_855, __p1_855); \
6749167491 __ret_855 = *(float16_t *) &__reint1_855; \
6749267492 __ret_855; \
6749367493})
6749467494#define __noswap_vgetq_lane_f16(__p0_856, __p1_856) __extension__ ({ \
67495 float16x8_t __s0_856 = __p0_856; \
6749667495 float16_t __ret_856; \
67496 float16x8_t __s0_856 = __p0_856; \
6749767497float16x8_t __reint_856 = __s0_856; \
6749867498int16_t __reint1_856 = __noswap_vgetq_lane_s16(*(int16x8_t *) &__reint_856, __p1_856); \
6749967499 __ret_856 = *(float16_t *) &__reint1_856; \
......@@ -67509,10 +67509,10 @@ __ai uint16x8_t vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
6750967509}
6751067510#else
6751167511__ai uint16x8_t vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67512 uint16x8_t __ret;
6751267513 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6751367514 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6751467515 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
67515 uint16x8_t __ret;
6751667516 __ret = __rev0 + __noswap_vmull_u8(__rev1, __rev2);
6751767517 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6751867518 return __ret;
......@@ -67532,10 +67532,10 @@ __ai uint64x2_t vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
6753267532}
6753367533#else
6753467534__ai uint64x2_t vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67535 uint64x2_t __ret;
6753567536 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6753667537 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6753767538 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
67538 uint64x2_t __ret;
6753967539 __ret = __rev0 + __noswap_vmull_u32(__rev1, __rev2);
6754067540 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6754167541 return __ret;
......@@ -67555,10 +67555,10 @@ __ai uint32x4_t vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
6755567555}
6755667556#else
6755767557__ai uint32x4_t vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67558 uint32x4_t __ret;
6755867559 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6755967560 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6756067561 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
67561 uint32x4_t __ret;
6756267562 __ret = __rev0 + __noswap_vmull_u16(__rev1, __rev2);
6756367563 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6756467564 return __ret;
......@@ -67578,10 +67578,10 @@ __ai int16x8_t vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
6757867578}
6757967579#else
6758067580__ai int16x8_t vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67581 int16x8_t __ret;
6758167582 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6758267583 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6758367584 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
67584 int16x8_t __ret;
6758567585 __ret = __rev0 + __noswap_vmull_s8(__rev1, __rev2);
6758667586 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6758767587 return __ret;
......@@ -67601,10 +67601,10 @@ __ai int64x2_t vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
6760167601}
6760267602#else
6760367603__ai int64x2_t vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67604 int64x2_t __ret;
6760467605 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6760567606 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6760667607 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
67607 int64x2_t __ret;
6760867608 __ret = __rev0 + __noswap_vmull_s32(__rev1, __rev2);
6760967609 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6761067610 return __ret;
......@@ -67624,10 +67624,10 @@ __ai int32x4_t vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
6762467624}
6762567625#else
6762667626__ai int32x4_t vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67627 int32x4_t __ret;
6762767628 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6762867629 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6762967630 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
67630 int32x4_t __ret;
6763167631 __ret = __rev0 + __noswap_vmull_s16(__rev1, __rev2);
6763267632 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6763367633 return __ret;
......@@ -67641,22 +67641,22 @@ __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6764167641
6764267642#ifdef __LITTLE_ENDIAN__
6764367643#define vmlal_lane_u32(__p0_857, __p1_857, __p2_857, __p3_857) __extension__ ({ \
67644 uint64x2_t __ret_857; \
6764467645 uint64x2_t __s0_857 = __p0_857; \
6764567646 uint32x2_t __s1_857 = __p1_857; \
6764667647 uint32x2_t __s2_857 = __p2_857; \
67647 uint64x2_t __ret_857; \
6764867648 __ret_857 = __s0_857 + vmull_u32(__s1_857, splat_lane_u32(__s2_857, __p3_857)); \
6764967649 __ret_857; \
6765067650})
6765167651#else
6765267652#define vmlal_lane_u32(__p0_858, __p1_858, __p2_858, __p3_858) __extension__ ({ \
67653 uint64x2_t __ret_858; \
6765367654 uint64x2_t __s0_858 = __p0_858; \
6765467655 uint32x2_t __s1_858 = __p1_858; \
6765567656 uint32x2_t __s2_858 = __p2_858; \
6765667657 uint64x2_t __rev0_858; __rev0_858 = __builtin_shufflevector(__s0_858, __s0_858, 1, 0); \
6765767658 uint32x2_t __rev1_858; __rev1_858 = __builtin_shufflevector(__s1_858, __s1_858, 1, 0); \
6765867659 uint32x2_t __rev2_858; __rev2_858 = __builtin_shufflevector(__s2_858, __s2_858, 1, 0); \
67659 uint64x2_t __ret_858; \
6766067660 __ret_858 = __rev0_858 + __noswap_vmull_u32(__rev1_858, __noswap_splat_lane_u32(__rev2_858, __p3_858)); \
6766167661 __ret_858 = __builtin_shufflevector(__ret_858, __ret_858, 1, 0); \
6766267662 __ret_858; \
......@@ -67665,22 +67665,22 @@ __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6766567665
6766667666#ifdef __LITTLE_ENDIAN__
6766767667#define vmlal_lane_u16(__p0_859, __p1_859, __p2_859, __p3_859) __extension__ ({ \
67668 uint32x4_t __ret_859; \
6766867669 uint32x4_t __s0_859 = __p0_859; \
6766967670 uint16x4_t __s1_859 = __p1_859; \
6767067671 uint16x4_t __s2_859 = __p2_859; \
67671 uint32x4_t __ret_859; \
6767267672 __ret_859 = __s0_859 + vmull_u16(__s1_859, splat_lane_u16(__s2_859, __p3_859)); \
6767367673 __ret_859; \
6767467674})
6767567675#else
6767667676#define vmlal_lane_u16(__p0_860, __p1_860, __p2_860, __p3_860) __extension__ ({ \
67677 uint32x4_t __ret_860; \
6767767678 uint32x4_t __s0_860 = __p0_860; \
6767867679 uint16x4_t __s1_860 = __p1_860; \
6767967680 uint16x4_t __s2_860 = __p2_860; \
6768067681 uint32x4_t __rev0_860; __rev0_860 = __builtin_shufflevector(__s0_860, __s0_860, 3, 2, 1, 0); \
6768167682 uint16x4_t __rev1_860; __rev1_860 = __builtin_shufflevector(__s1_860, __s1_860, 3, 2, 1, 0); \
6768267683 uint16x4_t __rev2_860; __rev2_860 = __builtin_shufflevector(__s2_860, __s2_860, 3, 2, 1, 0); \
67683 uint32x4_t __ret_860; \
6768467684 __ret_860 = __rev0_860 + __noswap_vmull_u16(__rev1_860, __noswap_splat_lane_u16(__rev2_860, __p3_860)); \
6768567685 __ret_860 = __builtin_shufflevector(__ret_860, __ret_860, 3, 2, 1, 0); \
6768667686 __ret_860; \
......@@ -67689,22 +67689,22 @@ __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6768967689
6769067690#ifdef __LITTLE_ENDIAN__
6769167691#define vmlal_lane_s32(__p0_861, __p1_861, __p2_861, __p3_861) __extension__ ({ \
67692 int64x2_t __ret_861; \
6769267693 int64x2_t __s0_861 = __p0_861; \
6769367694 int32x2_t __s1_861 = __p1_861; \
6769467695 int32x2_t __s2_861 = __p2_861; \
67695 int64x2_t __ret_861; \
6769667696 __ret_861 = __s0_861 + vmull_s32(__s1_861, splat_lane_s32(__s2_861, __p3_861)); \
6769767697 __ret_861; \
6769867698})
6769967699#else
6770067700#define vmlal_lane_s32(__p0_862, __p1_862, __p2_862, __p3_862) __extension__ ({ \
67701 int64x2_t __ret_862; \
6770167702 int64x2_t __s0_862 = __p0_862; \
6770267703 int32x2_t __s1_862 = __p1_862; \
6770367704 int32x2_t __s2_862 = __p2_862; \
6770467705 int64x2_t __rev0_862; __rev0_862 = __builtin_shufflevector(__s0_862, __s0_862, 1, 0); \
6770567706 int32x2_t __rev1_862; __rev1_862 = __builtin_shufflevector(__s1_862, __s1_862, 1, 0); \
6770667707 int32x2_t __rev2_862; __rev2_862 = __builtin_shufflevector(__s2_862, __s2_862, 1, 0); \
67707 int64x2_t __ret_862; \
6770867708 __ret_862 = __rev0_862 + __noswap_vmull_s32(__rev1_862, __noswap_splat_lane_s32(__rev2_862, __p3_862)); \
6770967709 __ret_862 = __builtin_shufflevector(__ret_862, __ret_862, 1, 0); \
6771067710 __ret_862; \
......@@ -67713,22 +67713,22 @@ __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6771367713
6771467714#ifdef __LITTLE_ENDIAN__
6771567715#define vmlal_lane_s16(__p0_863, __p1_863, __p2_863, __p3_863) __extension__ ({ \
67716 int32x4_t __ret_863; \
6771667717 int32x4_t __s0_863 = __p0_863; \
6771767718 int16x4_t __s1_863 = __p1_863; \
6771867719 int16x4_t __s2_863 = __p2_863; \
67719 int32x4_t __ret_863; \
6772067720 __ret_863 = __s0_863 + vmull_s16(__s1_863, splat_lane_s16(__s2_863, __p3_863)); \
6772167721 __ret_863; \
6772267722})
6772367723#else
6772467724#define vmlal_lane_s16(__p0_864, __p1_864, __p2_864, __p3_864) __extension__ ({ \
67725 int32x4_t __ret_864; \
6772567726 int32x4_t __s0_864 = __p0_864; \
6772667727 int16x4_t __s1_864 = __p1_864; \
6772767728 int16x4_t __s2_864 = __p2_864; \
6772867729 int32x4_t __rev0_864; __rev0_864 = __builtin_shufflevector(__s0_864, __s0_864, 3, 2, 1, 0); \
6772967730 int16x4_t __rev1_864; __rev1_864 = __builtin_shufflevector(__s1_864, __s1_864, 3, 2, 1, 0); \
6773067731 int16x4_t __rev2_864; __rev2_864 = __builtin_shufflevector(__s2_864, __s2_864, 3, 2, 1, 0); \
67731 int32x4_t __ret_864; \
6773267732 __ret_864 = __rev0_864 + __noswap_vmull_s16(__rev1_864, __noswap_splat_lane_s16(__rev2_864, __p3_864)); \
6773367733 __ret_864 = __builtin_shufflevector(__ret_864, __ret_864, 3, 2, 1, 0); \
6773467734 __ret_864; \
......@@ -67743,9 +67743,9 @@ __ai uint64x2_t vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
6774367743}
6774467744#else
6774567745__ai uint64x2_t vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67746 uint64x2_t __ret;
6774667747 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6774767748 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67748 uint64x2_t __ret;
6774967749 __ret = __rev0 + __noswap_vmull_u32(__rev1, (uint32x2_t) {__p2, __p2});
6775067750 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6775167751 return __ret;
......@@ -67765,9 +67765,9 @@ __ai uint32x4_t vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
6776567765}
6776667766#else
6776767767__ai uint32x4_t vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67768 uint32x4_t __ret;
6776867769 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6776967770 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67770 uint32x4_t __ret;
6777167771 __ret = __rev0 + __noswap_vmull_u16(__rev1, (uint16x4_t) {__p2, __p2, __p2, __p2});
6777267772 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6777367773 return __ret;
......@@ -67787,9 +67787,9 @@ __ai int64x2_t vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
6778767787}
6778867788#else
6778967789__ai int64x2_t vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67790 int64x2_t __ret;
6779067791 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6779167792 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67792 int64x2_t __ret;
6779367793 __ret = __rev0 + __noswap_vmull_s32(__rev1, (int32x2_t) {__p2, __p2});
6779467794 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6779567795 return __ret;
......@@ -67809,9 +67809,9 @@ __ai int32x4_t vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
6780967809}
6781067810#else
6781167811__ai int32x4_t vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67812 int32x4_t __ret;
6781267813 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6781367814 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67814 int32x4_t __ret;
6781567815 __ret = __rev0 + __noswap_vmull_s16(__rev1, (int16x4_t) {__p2, __p2, __p2, __p2});
6781667816 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6781767817 return __ret;
......@@ -67831,10 +67831,10 @@ __ai uint16x8_t vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
6783167831}
6783267832#else
6783367833__ai uint16x8_t vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67834 uint16x8_t __ret;
6783467835 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6783567836 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6783667837 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
67837 uint16x8_t __ret;
6783867838 __ret = __rev0 - __noswap_vmull_u8(__rev1, __rev2);
6783967839 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6784067840 return __ret;
......@@ -67854,10 +67854,10 @@ __ai uint64x2_t vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
6785467854}
6785567855#else
6785667856__ai uint64x2_t vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67857 uint64x2_t __ret;
6785767858 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6785867859 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6785967860 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
67860 uint64x2_t __ret;
6786167861 __ret = __rev0 - __noswap_vmull_u32(__rev1, __rev2);
6786267862 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6786367863 return __ret;
......@@ -67877,10 +67877,10 @@ __ai uint32x4_t vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
6787767877}
6787867878#else
6787967879__ai uint32x4_t vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67880 uint32x4_t __ret;
6788067881 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6788167882 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6788267883 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
67883 uint32x4_t __ret;
6788467884 __ret = __rev0 - __noswap_vmull_u16(__rev1, __rev2);
6788567885 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6788667886 return __ret;
......@@ -67900,10 +67900,10 @@ __ai int16x8_t vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
6790067900}
6790167901#else
6790267902__ai int16x8_t vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67903 int16x8_t __ret;
6790367904 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6790467905 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6790567906 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
67906 int16x8_t __ret;
6790767907 __ret = __rev0 - __noswap_vmull_s8(__rev1, __rev2);
6790867908 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6790967909 return __ret;
......@@ -67923,10 +67923,10 @@ __ai int64x2_t vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
6792367923}
6792467924#else
6792567925__ai int64x2_t vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67926 int64x2_t __ret;
6792667927 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6792767928 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6792867929 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
67929 int64x2_t __ret;
6793067930 __ret = __rev0 - __noswap_vmull_s32(__rev1, __rev2);
6793167931 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6793267932 return __ret;
......@@ -67946,10 +67946,10 @@ __ai int32x4_t vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
6794667946}
6794767947#else
6794867948__ai int32x4_t vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67949 int32x4_t __ret;
6794967950 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6795067951 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6795167952 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
67952 int32x4_t __ret;
6795367953 __ret = __rev0 - __noswap_vmull_s16(__rev1, __rev2);
6795467954 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6795567955 return __ret;
......@@ -67963,22 +67963,22 @@ __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6796367963
6796467964#ifdef __LITTLE_ENDIAN__
6796567965#define vmlsl_lane_u32(__p0_865, __p1_865, __p2_865, __p3_865) __extension__ ({ \
67966 uint64x2_t __ret_865; \
6796667967 uint64x2_t __s0_865 = __p0_865; \
6796767968 uint32x2_t __s1_865 = __p1_865; \
6796867969 uint32x2_t __s2_865 = __p2_865; \
67969 uint64x2_t __ret_865; \
6797067970 __ret_865 = __s0_865 - vmull_u32(__s1_865, splat_lane_u32(__s2_865, __p3_865)); \
6797167971 __ret_865; \
6797267972})
6797367973#else
6797467974#define vmlsl_lane_u32(__p0_866, __p1_866, __p2_866, __p3_866) __extension__ ({ \
67975 uint64x2_t __ret_866; \
6797567976 uint64x2_t __s0_866 = __p0_866; \
6797667977 uint32x2_t __s1_866 = __p1_866; \
6797767978 uint32x2_t __s2_866 = __p2_866; \
6797867979 uint64x2_t __rev0_866; __rev0_866 = __builtin_shufflevector(__s0_866, __s0_866, 1, 0); \
6797967980 uint32x2_t __rev1_866; __rev1_866 = __builtin_shufflevector(__s1_866, __s1_866, 1, 0); \
6798067981 uint32x2_t __rev2_866; __rev2_866 = __builtin_shufflevector(__s2_866, __s2_866, 1, 0); \
67981 uint64x2_t __ret_866; \
6798267982 __ret_866 = __rev0_866 - __noswap_vmull_u32(__rev1_866, __noswap_splat_lane_u32(__rev2_866, __p3_866)); \
6798367983 __ret_866 = __builtin_shufflevector(__ret_866, __ret_866, 1, 0); \
6798467984 __ret_866; \
......@@ -67987,22 +67987,22 @@ __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6798767987
6798867988#ifdef __LITTLE_ENDIAN__
6798967989#define vmlsl_lane_u16(__p0_867, __p1_867, __p2_867, __p3_867) __extension__ ({ \
67990 uint32x4_t __ret_867; \
6799067991 uint32x4_t __s0_867 = __p0_867; \
6799167992 uint16x4_t __s1_867 = __p1_867; \
6799267993 uint16x4_t __s2_867 = __p2_867; \
67993 uint32x4_t __ret_867; \
6799467994 __ret_867 = __s0_867 - vmull_u16(__s1_867, splat_lane_u16(__s2_867, __p3_867)); \
6799567995 __ret_867; \
6799667996})
6799767997#else
6799867998#define vmlsl_lane_u16(__p0_868, __p1_868, __p2_868, __p3_868) __extension__ ({ \
67999 uint32x4_t __ret_868; \
6799968000 uint32x4_t __s0_868 = __p0_868; \
6800068001 uint16x4_t __s1_868 = __p1_868; \
6800168002 uint16x4_t __s2_868 = __p2_868; \
6800268003 uint32x4_t __rev0_868; __rev0_868 = __builtin_shufflevector(__s0_868, __s0_868, 3, 2, 1, 0); \
6800368004 uint16x4_t __rev1_868; __rev1_868 = __builtin_shufflevector(__s1_868, __s1_868, 3, 2, 1, 0); \
6800468005 uint16x4_t __rev2_868; __rev2_868 = __builtin_shufflevector(__s2_868, __s2_868, 3, 2, 1, 0); \
68005 uint32x4_t __ret_868; \
6800668006 __ret_868 = __rev0_868 - __noswap_vmull_u16(__rev1_868, __noswap_splat_lane_u16(__rev2_868, __p3_868)); \
6800768007 __ret_868 = __builtin_shufflevector(__ret_868, __ret_868, 3, 2, 1, 0); \
6800868008 __ret_868; \
......@@ -68011,22 +68011,22 @@ __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6801168011
6801268012#ifdef __LITTLE_ENDIAN__
6801368013#define vmlsl_lane_s32(__p0_869, __p1_869, __p2_869, __p3_869) __extension__ ({ \
68014 int64x2_t __ret_869; \
6801468015 int64x2_t __s0_869 = __p0_869; \
6801568016 int32x2_t __s1_869 = __p1_869; \
6801668017 int32x2_t __s2_869 = __p2_869; \
68017 int64x2_t __ret_869; \
6801868018 __ret_869 = __s0_869 - vmull_s32(__s1_869, splat_lane_s32(__s2_869, __p3_869)); \
6801968019 __ret_869; \
6802068020})
6802168021#else
6802268022#define vmlsl_lane_s32(__p0_870, __p1_870, __p2_870, __p3_870) __extension__ ({ \
68023 int64x2_t __ret_870; \
6802368024 int64x2_t __s0_870 = __p0_870; \
6802468025 int32x2_t __s1_870 = __p1_870; \
6802568026 int32x2_t __s2_870 = __p2_870; \
6802668027 int64x2_t __rev0_870; __rev0_870 = __builtin_shufflevector(__s0_870, __s0_870, 1, 0); \
6802768028 int32x2_t __rev1_870; __rev1_870 = __builtin_shufflevector(__s1_870, __s1_870, 1, 0); \
6802868029 int32x2_t __rev2_870; __rev2_870 = __builtin_shufflevector(__s2_870, __s2_870, 1, 0); \
68029 int64x2_t __ret_870; \
6803068030 __ret_870 = __rev0_870 - __noswap_vmull_s32(__rev1_870, __noswap_splat_lane_s32(__rev2_870, __p3_870)); \
6803168031 __ret_870 = __builtin_shufflevector(__ret_870, __ret_870, 1, 0); \
6803268032 __ret_870; \
......@@ -68035,22 +68035,22 @@ __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2
6803568035
6803668036#ifdef __LITTLE_ENDIAN__
6803768037#define vmlsl_lane_s16(__p0_871, __p1_871, __p2_871, __p3_871) __extension__ ({ \
68038 int32x4_t __ret_871; \
6803868039 int32x4_t __s0_871 = __p0_871; \
6803968040 int16x4_t __s1_871 = __p1_871; \
6804068041 int16x4_t __s2_871 = __p2_871; \
68041 int32x4_t __ret_871; \
6804268042 __ret_871 = __s0_871 - vmull_s16(__s1_871, splat_lane_s16(__s2_871, __p3_871)); \
6804368043 __ret_871; \
6804468044})
6804568045#else
6804668046#define vmlsl_lane_s16(__p0_872, __p1_872, __p2_872, __p3_872) __extension__ ({ \
68047 int32x4_t __ret_872; \
6804768048 int32x4_t __s0_872 = __p0_872; \
6804868049 int16x4_t __s1_872 = __p1_872; \
6804968050 int16x4_t __s2_872 = __p2_872; \
6805068051 int32x4_t __rev0_872; __rev0_872 = __builtin_shufflevector(__s0_872, __s0_872, 3, 2, 1, 0); \
6805168052 int16x4_t __rev1_872; __rev1_872 = __builtin_shufflevector(__s1_872, __s1_872, 3, 2, 1, 0); \
6805268053 int16x4_t __rev2_872; __rev2_872 = __builtin_shufflevector(__s2_872, __s2_872, 3, 2, 1, 0); \
68053 int32x4_t __ret_872; \
6805468054 __ret_872 = __rev0_872 - __noswap_vmull_s16(__rev1_872, __noswap_splat_lane_s16(__rev2_872, __p3_872)); \
6805568055 __ret_872 = __builtin_shufflevector(__ret_872, __ret_872, 3, 2, 1, 0); \
6805668056 __ret_872; \
......@@ -68065,9 +68065,9 @@ __ai uint64x2_t vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
6806568065}
6806668066#else
6806768067__ai uint64x2_t vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
68068 uint64x2_t __ret;
6806868069 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6806968070 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
68070 uint64x2_t __ret;
6807168071 __ret = __rev0 - __noswap_vmull_u32(__rev1, (uint32x2_t) {__p2, __p2});
6807268072 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6807368073 return __ret;
......@@ -68087,9 +68087,9 @@ __ai uint32x4_t vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
6808768087}
6808868088#else
6808968089__ai uint32x4_t vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
68090 uint32x4_t __ret;
6809068091 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6809168092 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68092 uint32x4_t __ret;
6809368093 __ret = __rev0 - __noswap_vmull_u16(__rev1, (uint16x4_t) {__p2, __p2, __p2, __p2});
6809468094 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6809568095 return __ret;
......@@ -68109,9 +68109,9 @@ __ai int64x2_t vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
6810968109}
6811068110#else
6811168111__ai int64x2_t vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
68112 int64x2_t __ret;
6811268113 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6811368114 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
68114 int64x2_t __ret;
6811568115 __ret = __rev0 - __noswap_vmull_s32(__rev1, (int32x2_t) {__p2, __p2});
6811668116 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6811768117 return __ret;
......@@ -68131,9 +68131,9 @@ __ai int32x4_t vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
6813168131}
6813268132#else
6813368133__ai int32x4_t vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
68134 int32x4_t __ret;
6813468135 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6813568136 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68136 int32x4_t __ret;
6813768137 __ret = __rev0 - __noswap_vmull_s16(__rev1, (int16x4_t) {__p2, __p2, __p2, __p2});
6813868138 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6813968139 return __ret;
......@@ -68147,9 +68147,9 @@ __ai int32x4_t __noswap_vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2
6814768147
6814868148#ifdef __LITTLE_ENDIAN__
6814968149#define vset_lane_f16(__p0_873, __p1_873, __p2_873) __extension__ ({ \
68150 float16x4_t __ret_873; \
6815068151 float16_t __s0_873 = __p0_873; \
6815168152 float16x4_t __s1_873 = __p1_873; \
68152 float16x4_t __ret_873; \
6815368153float16_t __reint_873 = __s0_873; \
6815468154float16x4_t __reint1_873 = __s1_873; \
6815568155int16x4_t __reint2_873 = vset_lane_s16(*(int16_t *) &__reint_873, *(int16x4_t *) &__reint1_873, __p2_873); \
......@@ -68158,10 +68158,10 @@ int16x4_t __reint2_873 = vset_lane_s16(*(int16_t *) &__reint_873, *(int16x4_t *)
6815868158})
6815968159#else
6816068160#define vset_lane_f16(__p0_874, __p1_874, __p2_874) __extension__ ({ \
68161 float16x4_t __ret_874; \
6816168162 float16_t __s0_874 = __p0_874; \
6816268163 float16x4_t __s1_874 = __p1_874; \
6816368164 float16x4_t __rev1_874; __rev1_874 = __builtin_shufflevector(__s1_874, __s1_874, 3, 2, 1, 0); \
68164 float16x4_t __ret_874; \
6816568165float16_t __reint_874 = __s0_874; \
6816668166float16x4_t __reint1_874 = __rev1_874; \
6816768167int16x4_t __reint2_874 = __noswap_vset_lane_s16(*(int16_t *) &__reint_874, *(int16x4_t *) &__reint1_874, __p2_874); \
......@@ -68173,9 +68173,9 @@ int16x4_t __reint2_874 = __noswap_vset_lane_s16(*(int16_t *) &__reint_874, *(int
6817368173
6817468174#ifdef __LITTLE_ENDIAN__
6817568175#define vsetq_lane_f16(__p0_875, __p1_875, __p2_875) __extension__ ({ \
68176 float16x8_t __ret_875; \
6817668177 float16_t __s0_875 = __p0_875; \
6817768178 float16x8_t __s1_875 = __p1_875; \
68178 float16x8_t __ret_875; \
6817968179float16_t __reint_875 = __s0_875; \
6818068180float16x8_t __reint1_875 = __s1_875; \
6818168181int16x8_t __reint2_875 = vsetq_lane_s16(*(int16_t *) &__reint_875, *(int16x8_t *) &__reint1_875, __p2_875); \
......@@ -68184,10 +68184,10 @@ int16x8_t __reint2_875 = vsetq_lane_s16(*(int16_t *) &__reint_875, *(int16x8_t *
6818468184})
6818568185#else
6818668186#define vsetq_lane_f16(__p0_876, __p1_876, __p2_876) __extension__ ({ \
68187 float16x8_t __ret_876; \
6818768188 float16_t __s0_876 = __p0_876; \
6818868189 float16x8_t __s1_876 = __p1_876; \
6818968190 float16x8_t __rev1_876; __rev1_876 = __builtin_shufflevector(__s1_876, __s1_876, 7, 6, 5, 4, 3, 2, 1, 0); \
68190 float16x8_t __ret_876; \
6819168191float16_t __reint_876 = __s0_876; \
6819268192float16x8_t __reint1_876 = __rev1_876; \
6819368193int16x8_t __reint2_876 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_876, *(int16x8_t *) &__reint1_876, __p2_876); \
......@@ -68200,22 +68200,22 @@ int16x8_t __reint2_876 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_876, *(in
6820068200#if defined(__ARM_FEATURE_BF16_VECTOR_ARITHMETIC)
6820168201#ifdef __LITTLE_ENDIAN__
6820268202#define vbfmlalbq_lane_f32(__p0_877, __p1_877, __p2_877, __p3_877) __extension__ ({ \
68203 float32x4_t __ret_877; \
6820368204 float32x4_t __s0_877 = __p0_877; \
6820468205 bfloat16x8_t __s1_877 = __p1_877; \
6820568206 bfloat16x4_t __s2_877 = __p2_877; \
68206 float32x4_t __ret_877; \
6820768207 __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)}); \
6820868208 __ret_877; \
6820968209})
6821068210#else
6821168211#define vbfmlalbq_lane_f32(__p0_878, __p1_878, __p2_878, __p3_878) __extension__ ({ \
68212 float32x4_t __ret_878; \
6821268213 float32x4_t __s0_878 = __p0_878; \
6821368214 bfloat16x8_t __s1_878 = __p1_878; \
6821468215 bfloat16x4_t __s2_878 = __p2_878; \
6821568216 float32x4_t __rev0_878; __rev0_878 = __builtin_shufflevector(__s0_878, __s0_878, 3, 2, 1, 0); \
6821668217 bfloat16x8_t __rev1_878; __rev1_878 = __builtin_shufflevector(__s1_878, __s1_878, 7, 6, 5, 4, 3, 2, 1, 0); \
6821768218 bfloat16x4_t __rev2_878; __rev2_878 = __builtin_shufflevector(__s2_878, __s2_878, 3, 2, 1, 0); \
68218 float32x4_t __ret_878; \
6821968219 __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)}); \
6822068220 __ret_878 = __builtin_shufflevector(__ret_878, __ret_878, 3, 2, 1, 0); \
6822168221 __ret_878; \
......@@ -68224,22 +68224,22 @@ int16x8_t __reint2_876 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_876, *(in
6822468224
6822568225#ifdef __LITTLE_ENDIAN__
6822668226#define vbfmlalbq_laneq_f32(__p0_879, __p1_879, __p2_879, __p3_879) __extension__ ({ \
68227 float32x4_t __ret_879; \
6822768228 float32x4_t __s0_879 = __p0_879; \
6822868229 bfloat16x8_t __s1_879 = __p1_879; \
6822968230 bfloat16x8_t __s2_879 = __p2_879; \
68230 float32x4_t __ret_879; \
6823168231 __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)}); \
6823268232 __ret_879; \
6823368233})
6823468234#else
6823568235#define vbfmlalbq_laneq_f32(__p0_880, __p1_880, __p2_880, __p3_880) __extension__ ({ \
68236 float32x4_t __ret_880; \
6823668237 float32x4_t __s0_880 = __p0_880; \
6823768238 bfloat16x8_t __s1_880 = __p1_880; \
6823868239 bfloat16x8_t __s2_880 = __p2_880; \
6823968240 float32x4_t __rev0_880; __rev0_880 = __builtin_shufflevector(__s0_880, __s0_880, 3, 2, 1, 0); \
6824068241 bfloat16x8_t __rev1_880; __rev1_880 = __builtin_shufflevector(__s1_880, __s1_880, 7, 6, 5, 4, 3, 2, 1, 0); \
6824168242 bfloat16x8_t __rev2_880; __rev2_880 = __builtin_shufflevector(__s2_880, __s2_880, 7, 6, 5, 4, 3, 2, 1, 0); \
68242 float32x4_t __ret_880; \
6824368243 __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)}); \
6824468244 __ret_880 = __builtin_shufflevector(__ret_880, __ret_880, 3, 2, 1, 0); \
6824568245 __ret_880; \
......@@ -68248,22 +68248,22 @@ int16x8_t __reint2_876 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_876, *(in
6824868248
6824968249#ifdef __LITTLE_ENDIAN__
6825068250#define vbfmlaltq_lane_f32(__p0_881, __p1_881, __p2_881, __p3_881) __extension__ ({ \
68251 float32x4_t __ret_881; \
6825168252 float32x4_t __s0_881 = __p0_881; \
6825268253 bfloat16x8_t __s1_881 = __p1_881; \
6825368254 bfloat16x4_t __s2_881 = __p2_881; \
68254 float32x4_t __ret_881; \
6825568255 __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)}); \
6825668256 __ret_881; \
6825768257})
6825868258#else
6825968259#define vbfmlaltq_lane_f32(__p0_882, __p1_882, __p2_882, __p3_882) __extension__ ({ \
68260 float32x4_t __ret_882; \
6826068261 float32x4_t __s0_882 = __p0_882; \
6826168262 bfloat16x8_t __s1_882 = __p1_882; \
6826268263 bfloat16x4_t __s2_882 = __p2_882; \
6826368264 float32x4_t __rev0_882; __rev0_882 = __builtin_shufflevector(__s0_882, __s0_882, 3, 2, 1, 0); \
6826468265 bfloat16x8_t __rev1_882; __rev1_882 = __builtin_shufflevector(__s1_882, __s1_882, 7, 6, 5, 4, 3, 2, 1, 0); \
6826568266 bfloat16x4_t __rev2_882; __rev2_882 = __builtin_shufflevector(__s2_882, __s2_882, 3, 2, 1, 0); \
68266 float32x4_t __ret_882; \
6826768267 __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)}); \
6826868268 __ret_882 = __builtin_shufflevector(__ret_882, __ret_882, 3, 2, 1, 0); \
6826968269 __ret_882; \
......@@ -68272,22 +68272,22 @@ int16x8_t __reint2_876 = __noswap_vsetq_lane_s16(*(int16_t *) &__reint_876, *(in
6827268272
6827368273#ifdef __LITTLE_ENDIAN__
6827468274#define vbfmlaltq_laneq_f32(__p0_883, __p1_883, __p2_883, __p3_883) __extension__ ({ \
68275 float32x4_t __ret_883; \
6827568276 float32x4_t __s0_883 = __p0_883; \
6827668277 bfloat16x8_t __s1_883 = __p1_883; \
6827768278 bfloat16x8_t __s2_883 = __p2_883; \
68278 float32x4_t __ret_883; \
6827968279 __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)}); \
6828068280 __ret_883; \
6828168281})
6828268282#else
6828368283#define vbfmlaltq_laneq_f32(__p0_884, __p1_884, __p2_884, __p3_884) __extension__ ({ \
68284 float32x4_t __ret_884; \
6828468285 float32x4_t __s0_884 = __p0_884; \
6828568286 bfloat16x8_t __s1_884 = __p1_884; \
6828668287 bfloat16x8_t __s2_884 = __p2_884; \
6828768288 float32x4_t __rev0_884; __rev0_884 = __builtin_shufflevector(__s0_884, __s0_884, 3, 2, 1, 0); \
6828868289 bfloat16x8_t __rev1_884; __rev1_884 = __builtin_shufflevector(__s1_884, __s1_884, 7, 6, 5, 4, 3, 2, 1, 0); \
6828968290 bfloat16x8_t __rev2_884; __rev2_884 = __builtin_shufflevector(__s2_884, __s2_884, 7, 6, 5, 4, 3, 2, 1, 0); \
68290 float32x4_t __ret_884; \
6829168291 __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)}); \
6829268292 __ret_884 = __builtin_shufflevector(__ret_884, __ret_884, 3, 2, 1, 0); \
6829368293 __ret_884; \
......@@ -68302,8 +68302,8 @@ __ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
6830268302}
6830368303#else
6830468304__ai float32x4_t vcvtq_high_f32_bf16(bfloat16x8_t __p0) {
68305 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6830668305 float32x4_t __ret;
68306 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6830768307 __ret = __noswap_vcvt_f32_bf16(__noswap_vget_high_bf16(__rev0));
6830868308 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6830968309 return __ret;
......@@ -68318,8 +68318,8 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6831868318}
6831968319#else
6832068320__ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
68321 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6832268321 float32x4_t __ret;
68322 bfloat16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6832368323 __ret = __noswap_vcvt_f32_bf16(__noswap_vget_low_bf16(__rev0));
6832468324 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6832568325 return __ret;
......@@ -68330,22 +68330,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6833068330#if defined(__ARM_FEATURE_FP16_FML) && defined(__aarch64__)
6833168331#ifdef __LITTLE_ENDIAN__
6833268332#define vfmlalq_lane_high_f16(__p0_885, __p1_885, __p2_885, __p3_885) __extension__ ({ \
68333 float32x4_t __ret_885; \
6833368334 float32x4_t __s0_885 = __p0_885; \
6833468335 float16x8_t __s1_885 = __p1_885; \
6833568336 float16x4_t __s2_885 = __p2_885; \
68336 float32x4_t __ret_885; \
6833768337 __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)}); \
6833868338 __ret_885; \
6833968339})
6834068340#else
6834168341#define vfmlalq_lane_high_f16(__p0_886, __p1_886, __p2_886, __p3_886) __extension__ ({ \
68342 float32x4_t __ret_886; \
6834268343 float32x4_t __s0_886 = __p0_886; \
6834368344 float16x8_t __s1_886 = __p1_886; \
6834468345 float16x4_t __s2_886 = __p2_886; \
6834568346 float32x4_t __rev0_886; __rev0_886 = __builtin_shufflevector(__s0_886, __s0_886, 3, 2, 1, 0); \
6834668347 float16x8_t __rev1_886; __rev1_886 = __builtin_shufflevector(__s1_886, __s1_886, 7, 6, 5, 4, 3, 2, 1, 0); \
6834768348 float16x4_t __rev2_886; __rev2_886 = __builtin_shufflevector(__s2_886, __s2_886, 3, 2, 1, 0); \
68348 float32x4_t __ret_886; \
6834968349 __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)}); \
6835068350 __ret_886 = __builtin_shufflevector(__ret_886, __ret_886, 3, 2, 1, 0); \
6835168351 __ret_886; \
......@@ -68354,22 +68354,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6835468354
6835568355#ifdef __LITTLE_ENDIAN__
6835668356#define vfmlal_lane_high_f16(__p0_887, __p1_887, __p2_887, __p3_887) __extension__ ({ \
68357 float32x2_t __ret_887; \
6835768358 float32x2_t __s0_887 = __p0_887; \
6835868359 float16x4_t __s1_887 = __p1_887; \
6835968360 float16x4_t __s2_887 = __p2_887; \
68360 float32x2_t __ret_887; \
6836168361 __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)}); \
6836268362 __ret_887; \
6836368363})
6836468364#else
6836568365#define vfmlal_lane_high_f16(__p0_888, __p1_888, __p2_888, __p3_888) __extension__ ({ \
68366 float32x2_t __ret_888; \
6836668367 float32x2_t __s0_888 = __p0_888; \
6836768368 float16x4_t __s1_888 = __p1_888; \
6836868369 float16x4_t __s2_888 = __p2_888; \
6836968370 float32x2_t __rev0_888; __rev0_888 = __builtin_shufflevector(__s0_888, __s0_888, 1, 0); \
6837068371 float16x4_t __rev1_888; __rev1_888 = __builtin_shufflevector(__s1_888, __s1_888, 3, 2, 1, 0); \
6837168372 float16x4_t __rev2_888; __rev2_888 = __builtin_shufflevector(__s2_888, __s2_888, 3, 2, 1, 0); \
68372 float32x2_t __ret_888; \
6837368373 __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)}); \
6837468374 __ret_888 = __builtin_shufflevector(__ret_888, __ret_888, 1, 0); \
6837568375 __ret_888; \
......@@ -68378,22 +68378,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6837868378
6837968379#ifdef __LITTLE_ENDIAN__
6838068380#define vfmlalq_lane_low_f16(__p0_889, __p1_889, __p2_889, __p3_889) __extension__ ({ \
68381 float32x4_t __ret_889; \
6838168382 float32x4_t __s0_889 = __p0_889; \
6838268383 float16x8_t __s1_889 = __p1_889; \
6838368384 float16x4_t __s2_889 = __p2_889; \
68384 float32x4_t __ret_889; \
6838568385 __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)}); \
6838668386 __ret_889; \
6838768387})
6838868388#else
6838968389#define vfmlalq_lane_low_f16(__p0_890, __p1_890, __p2_890, __p3_890) __extension__ ({ \
68390 float32x4_t __ret_890; \
6839068391 float32x4_t __s0_890 = __p0_890; \
6839168392 float16x8_t __s1_890 = __p1_890; \
6839268393 float16x4_t __s2_890 = __p2_890; \
6839368394 float32x4_t __rev0_890; __rev0_890 = __builtin_shufflevector(__s0_890, __s0_890, 3, 2, 1, 0); \
6839468395 float16x8_t __rev1_890; __rev1_890 = __builtin_shufflevector(__s1_890, __s1_890, 7, 6, 5, 4, 3, 2, 1, 0); \
6839568396 float16x4_t __rev2_890; __rev2_890 = __builtin_shufflevector(__s2_890, __s2_890, 3, 2, 1, 0); \
68396 float32x4_t __ret_890; \
6839768397 __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)}); \
6839868398 __ret_890 = __builtin_shufflevector(__ret_890, __ret_890, 3, 2, 1, 0); \
6839968399 __ret_890; \
......@@ -68402,22 +68402,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6840268402
6840368403#ifdef __LITTLE_ENDIAN__
6840468404#define vfmlal_lane_low_f16(__p0_891, __p1_891, __p2_891, __p3_891) __extension__ ({ \
68405 float32x2_t __ret_891; \
6840568406 float32x2_t __s0_891 = __p0_891; \
6840668407 float16x4_t __s1_891 = __p1_891; \
6840768408 float16x4_t __s2_891 = __p2_891; \
68408 float32x2_t __ret_891; \
6840968409 __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)}); \
6841068410 __ret_891; \
6841168411})
6841268412#else
6841368413#define vfmlal_lane_low_f16(__p0_892, __p1_892, __p2_892, __p3_892) __extension__ ({ \
68414 float32x2_t __ret_892; \
6841468415 float32x2_t __s0_892 = __p0_892; \
6841568416 float16x4_t __s1_892 = __p1_892; \
6841668417 float16x4_t __s2_892 = __p2_892; \
6841768418 float32x2_t __rev0_892; __rev0_892 = __builtin_shufflevector(__s0_892, __s0_892, 1, 0); \
6841868419 float16x4_t __rev1_892; __rev1_892 = __builtin_shufflevector(__s1_892, __s1_892, 3, 2, 1, 0); \
6841968420 float16x4_t __rev2_892; __rev2_892 = __builtin_shufflevector(__s2_892, __s2_892, 3, 2, 1, 0); \
68420 float32x2_t __ret_892; \
6842168421 __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)}); \
6842268422 __ret_892 = __builtin_shufflevector(__ret_892, __ret_892, 1, 0); \
6842368423 __ret_892; \
......@@ -68426,22 +68426,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6842668426
6842768427#ifdef __LITTLE_ENDIAN__
6842868428#define vfmlalq_laneq_high_f16(__p0_893, __p1_893, __p2_893, __p3_893) __extension__ ({ \
68429 float32x4_t __ret_893; \
6842968430 float32x4_t __s0_893 = __p0_893; \
6843068431 float16x8_t __s1_893 = __p1_893; \
6843168432 float16x8_t __s2_893 = __p2_893; \
68432 float32x4_t __ret_893; \
6843368433 __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)}); \
6843468434 __ret_893; \
6843568435})
6843668436#else
6843768437#define vfmlalq_laneq_high_f16(__p0_894, __p1_894, __p2_894, __p3_894) __extension__ ({ \
68438 float32x4_t __ret_894; \
6843868439 float32x4_t __s0_894 = __p0_894; \
6843968440 float16x8_t __s1_894 = __p1_894; \
6844068441 float16x8_t __s2_894 = __p2_894; \
6844168442 float32x4_t __rev0_894; __rev0_894 = __builtin_shufflevector(__s0_894, __s0_894, 3, 2, 1, 0); \
6844268443 float16x8_t __rev1_894; __rev1_894 = __builtin_shufflevector(__s1_894, __s1_894, 7, 6, 5, 4, 3, 2, 1, 0); \
6844368444 float16x8_t __rev2_894; __rev2_894 = __builtin_shufflevector(__s2_894, __s2_894, 7, 6, 5, 4, 3, 2, 1, 0); \
68444 float32x4_t __ret_894; \
6844568445 __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)}); \
6844668446 __ret_894 = __builtin_shufflevector(__ret_894, __ret_894, 3, 2, 1, 0); \
6844768447 __ret_894; \
......@@ -68450,22 +68450,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6845068450
6845168451#ifdef __LITTLE_ENDIAN__
6845268452#define vfmlal_laneq_high_f16(__p0_895, __p1_895, __p2_895, __p3_895) __extension__ ({ \
68453 float32x2_t __ret_895; \
6845368454 float32x2_t __s0_895 = __p0_895; \
6845468455 float16x4_t __s1_895 = __p1_895; \
6845568456 float16x8_t __s2_895 = __p2_895; \
68456 float32x2_t __ret_895; \
6845768457 __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)}); \
6845868458 __ret_895; \
6845968459})
6846068460#else
6846168461#define vfmlal_laneq_high_f16(__p0_896, __p1_896, __p2_896, __p3_896) __extension__ ({ \
68462 float32x2_t __ret_896; \
6846268463 float32x2_t __s0_896 = __p0_896; \
6846368464 float16x4_t __s1_896 = __p1_896; \
6846468465 float16x8_t __s2_896 = __p2_896; \
6846568466 float32x2_t __rev0_896; __rev0_896 = __builtin_shufflevector(__s0_896, __s0_896, 1, 0); \
6846668467 float16x4_t __rev1_896; __rev1_896 = __builtin_shufflevector(__s1_896, __s1_896, 3, 2, 1, 0); \
6846768468 float16x8_t __rev2_896; __rev2_896 = __builtin_shufflevector(__s2_896, __s2_896, 7, 6, 5, 4, 3, 2, 1, 0); \
68468 float32x2_t __ret_896; \
6846968469 __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)}); \
6847068470 __ret_896 = __builtin_shufflevector(__ret_896, __ret_896, 1, 0); \
6847168471 __ret_896; \
......@@ -68474,22 +68474,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6847468474
6847568475#ifdef __LITTLE_ENDIAN__
6847668476#define vfmlalq_laneq_low_f16(__p0_897, __p1_897, __p2_897, __p3_897) __extension__ ({ \
68477 float32x4_t __ret_897; \
6847768478 float32x4_t __s0_897 = __p0_897; \
6847868479 float16x8_t __s1_897 = __p1_897; \
6847968480 float16x8_t __s2_897 = __p2_897; \
68480 float32x4_t __ret_897; \
6848168481 __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)}); \
6848268482 __ret_897; \
6848368483})
6848468484#else
6848568485#define vfmlalq_laneq_low_f16(__p0_898, __p1_898, __p2_898, __p3_898) __extension__ ({ \
68486 float32x4_t __ret_898; \
6848668487 float32x4_t __s0_898 = __p0_898; \
6848768488 float16x8_t __s1_898 = __p1_898; \
6848868489 float16x8_t __s2_898 = __p2_898; \
6848968490 float32x4_t __rev0_898; __rev0_898 = __builtin_shufflevector(__s0_898, __s0_898, 3, 2, 1, 0); \
6849068491 float16x8_t __rev1_898; __rev1_898 = __builtin_shufflevector(__s1_898, __s1_898, 7, 6, 5, 4, 3, 2, 1, 0); \
6849168492 float16x8_t __rev2_898; __rev2_898 = __builtin_shufflevector(__s2_898, __s2_898, 7, 6, 5, 4, 3, 2, 1, 0); \
68492 float32x4_t __ret_898; \
6849368493 __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)}); \
6849468494 __ret_898 = __builtin_shufflevector(__ret_898, __ret_898, 3, 2, 1, 0); \
6849568495 __ret_898; \
......@@ -68498,22 +68498,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6849868498
6849968499#ifdef __LITTLE_ENDIAN__
6850068500#define vfmlal_laneq_low_f16(__p0_899, __p1_899, __p2_899, __p3_899) __extension__ ({ \
68501 float32x2_t __ret_899; \
6850168502 float32x2_t __s0_899 = __p0_899; \
6850268503 float16x4_t __s1_899 = __p1_899; \
6850368504 float16x8_t __s2_899 = __p2_899; \
68504 float32x2_t __ret_899; \
6850568505 __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)}); \
6850668506 __ret_899; \
6850768507})
6850868508#else
6850968509#define vfmlal_laneq_low_f16(__p0_900, __p1_900, __p2_900, __p3_900) __extension__ ({ \
68510 float32x2_t __ret_900; \
6851068511 float32x2_t __s0_900 = __p0_900; \
6851168512 float16x4_t __s1_900 = __p1_900; \
6851268513 float16x8_t __s2_900 = __p2_900; \
6851368514 float32x2_t __rev0_900; __rev0_900 = __builtin_shufflevector(__s0_900, __s0_900, 1, 0); \
6851468515 float16x4_t __rev1_900; __rev1_900 = __builtin_shufflevector(__s1_900, __s1_900, 3, 2, 1, 0); \
6851568516 float16x8_t __rev2_900; __rev2_900 = __builtin_shufflevector(__s2_900, __s2_900, 7, 6, 5, 4, 3, 2, 1, 0); \
68516 float32x2_t __ret_900; \
6851768517 __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)}); \
6851868518 __ret_900 = __builtin_shufflevector(__ret_900, __ret_900, 1, 0); \
6851968519 __ret_900; \
......@@ -68522,22 +68522,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6852268522
6852368523#ifdef __LITTLE_ENDIAN__
6852468524#define vfmlslq_lane_high_f16(__p0_901, __p1_901, __p2_901, __p3_901) __extension__ ({ \
68525 float32x4_t __ret_901; \
6852568526 float32x4_t __s0_901 = __p0_901; \
6852668527 float16x8_t __s1_901 = __p1_901; \
6852768528 float16x4_t __s2_901 = __p2_901; \
68528 float32x4_t __ret_901; \
6852968529 __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)}); \
6853068530 __ret_901; \
6853168531})
6853268532#else
6853368533#define vfmlslq_lane_high_f16(__p0_902, __p1_902, __p2_902, __p3_902) __extension__ ({ \
68534 float32x4_t __ret_902; \
6853468535 float32x4_t __s0_902 = __p0_902; \
6853568536 float16x8_t __s1_902 = __p1_902; \
6853668537 float16x4_t __s2_902 = __p2_902; \
6853768538 float32x4_t __rev0_902; __rev0_902 = __builtin_shufflevector(__s0_902, __s0_902, 3, 2, 1, 0); \
6853868539 float16x8_t __rev1_902; __rev1_902 = __builtin_shufflevector(__s1_902, __s1_902, 7, 6, 5, 4, 3, 2, 1, 0); \
6853968540 float16x4_t __rev2_902; __rev2_902 = __builtin_shufflevector(__s2_902, __s2_902, 3, 2, 1, 0); \
68540 float32x4_t __ret_902; \
6854168541 __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)}); \
6854268542 __ret_902 = __builtin_shufflevector(__ret_902, __ret_902, 3, 2, 1, 0); \
6854368543 __ret_902; \
......@@ -68546,22 +68546,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6854668546
6854768547#ifdef __LITTLE_ENDIAN__
6854868548#define vfmlsl_lane_high_f16(__p0_903, __p1_903, __p2_903, __p3_903) __extension__ ({ \
68549 float32x2_t __ret_903; \
6854968550 float32x2_t __s0_903 = __p0_903; \
6855068551 float16x4_t __s1_903 = __p1_903; \
6855168552 float16x4_t __s2_903 = __p2_903; \
68552 float32x2_t __ret_903; \
6855368553 __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)}); \
6855468554 __ret_903; \
6855568555})
6855668556#else
6855768557#define vfmlsl_lane_high_f16(__p0_904, __p1_904, __p2_904, __p3_904) __extension__ ({ \
68558 float32x2_t __ret_904; \
6855868559 float32x2_t __s0_904 = __p0_904; \
6855968560 float16x4_t __s1_904 = __p1_904; \
6856068561 float16x4_t __s2_904 = __p2_904; \
6856168562 float32x2_t __rev0_904; __rev0_904 = __builtin_shufflevector(__s0_904, __s0_904, 1, 0); \
6856268563 float16x4_t __rev1_904; __rev1_904 = __builtin_shufflevector(__s1_904, __s1_904, 3, 2, 1, 0); \
6856368564 float16x4_t __rev2_904; __rev2_904 = __builtin_shufflevector(__s2_904, __s2_904, 3, 2, 1, 0); \
68564 float32x2_t __ret_904; \
6856568565 __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)}); \
6856668566 __ret_904 = __builtin_shufflevector(__ret_904, __ret_904, 1, 0); \
6856768567 __ret_904; \
......@@ -68570,22 +68570,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6857068570
6857168571#ifdef __LITTLE_ENDIAN__
6857268572#define vfmlslq_lane_low_f16(__p0_905, __p1_905, __p2_905, __p3_905) __extension__ ({ \
68573 float32x4_t __ret_905; \
6857368574 float32x4_t __s0_905 = __p0_905; \
6857468575 float16x8_t __s1_905 = __p1_905; \
6857568576 float16x4_t __s2_905 = __p2_905; \
68576 float32x4_t __ret_905; \
6857768577 __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)}); \
6857868578 __ret_905; \
6857968579})
6858068580#else
6858168581#define vfmlslq_lane_low_f16(__p0_906, __p1_906, __p2_906, __p3_906) __extension__ ({ \
68582 float32x4_t __ret_906; \
6858268583 float32x4_t __s0_906 = __p0_906; \
6858368584 float16x8_t __s1_906 = __p1_906; \
6858468585 float16x4_t __s2_906 = __p2_906; \
6858568586 float32x4_t __rev0_906; __rev0_906 = __builtin_shufflevector(__s0_906, __s0_906, 3, 2, 1, 0); \
6858668587 float16x8_t __rev1_906; __rev1_906 = __builtin_shufflevector(__s1_906, __s1_906, 7, 6, 5, 4, 3, 2, 1, 0); \
6858768588 float16x4_t __rev2_906; __rev2_906 = __builtin_shufflevector(__s2_906, __s2_906, 3, 2, 1, 0); \
68588 float32x4_t __ret_906; \
6858968589 __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)}); \
6859068590 __ret_906 = __builtin_shufflevector(__ret_906, __ret_906, 3, 2, 1, 0); \
6859168591 __ret_906; \
......@@ -68594,22 +68594,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6859468594
6859568595#ifdef __LITTLE_ENDIAN__
6859668596#define vfmlsl_lane_low_f16(__p0_907, __p1_907, __p2_907, __p3_907) __extension__ ({ \
68597 float32x2_t __ret_907; \
6859768598 float32x2_t __s0_907 = __p0_907; \
6859868599 float16x4_t __s1_907 = __p1_907; \
6859968600 float16x4_t __s2_907 = __p2_907; \
68600 float32x2_t __ret_907; \
6860168601 __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)}); \
6860268602 __ret_907; \
6860368603})
6860468604#else
6860568605#define vfmlsl_lane_low_f16(__p0_908, __p1_908, __p2_908, __p3_908) __extension__ ({ \
68606 float32x2_t __ret_908; \
6860668607 float32x2_t __s0_908 = __p0_908; \
6860768608 float16x4_t __s1_908 = __p1_908; \
6860868609 float16x4_t __s2_908 = __p2_908; \
6860968610 float32x2_t __rev0_908; __rev0_908 = __builtin_shufflevector(__s0_908, __s0_908, 1, 0); \
6861068611 float16x4_t __rev1_908; __rev1_908 = __builtin_shufflevector(__s1_908, __s1_908, 3, 2, 1, 0); \
6861168612 float16x4_t __rev2_908; __rev2_908 = __builtin_shufflevector(__s2_908, __s2_908, 3, 2, 1, 0); \
68612 float32x2_t __ret_908; \
6861368613 __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)}); \
6861468614 __ret_908 = __builtin_shufflevector(__ret_908, __ret_908, 1, 0); \
6861568615 __ret_908; \
......@@ -68618,22 +68618,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6861868618
6861968619#ifdef __LITTLE_ENDIAN__
6862068620#define vfmlslq_laneq_high_f16(__p0_909, __p1_909, __p2_909, __p3_909) __extension__ ({ \
68621 float32x4_t __ret_909; \
6862168622 float32x4_t __s0_909 = __p0_909; \
6862268623 float16x8_t __s1_909 = __p1_909; \
6862368624 float16x8_t __s2_909 = __p2_909; \
68624 float32x4_t __ret_909; \
6862568625 __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)}); \
6862668626 __ret_909; \
6862768627})
6862868628#else
6862968629#define vfmlslq_laneq_high_f16(__p0_910, __p1_910, __p2_910, __p3_910) __extension__ ({ \
68630 float32x4_t __ret_910; \
6863068631 float32x4_t __s0_910 = __p0_910; \
6863168632 float16x8_t __s1_910 = __p1_910; \
6863268633 float16x8_t __s2_910 = __p2_910; \
6863368634 float32x4_t __rev0_910; __rev0_910 = __builtin_shufflevector(__s0_910, __s0_910, 3, 2, 1, 0); \
6863468635 float16x8_t __rev1_910; __rev1_910 = __builtin_shufflevector(__s1_910, __s1_910, 7, 6, 5, 4, 3, 2, 1, 0); \
6863568636 float16x8_t __rev2_910; __rev2_910 = __builtin_shufflevector(__s2_910, __s2_910, 7, 6, 5, 4, 3, 2, 1, 0); \
68636 float32x4_t __ret_910; \
6863768637 __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)}); \
6863868638 __ret_910 = __builtin_shufflevector(__ret_910, __ret_910, 3, 2, 1, 0); \
6863968639 __ret_910; \
......@@ -68642,22 +68642,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6864268642
6864368643#ifdef __LITTLE_ENDIAN__
6864468644#define vfmlsl_laneq_high_f16(__p0_911, __p1_911, __p2_911, __p3_911) __extension__ ({ \
68645 float32x2_t __ret_911; \
6864568646 float32x2_t __s0_911 = __p0_911; \
6864668647 float16x4_t __s1_911 = __p1_911; \
6864768648 float16x8_t __s2_911 = __p2_911; \
68648 float32x2_t __ret_911; \
6864968649 __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)}); \
6865068650 __ret_911; \
6865168651})
6865268652#else
6865368653#define vfmlsl_laneq_high_f16(__p0_912, __p1_912, __p2_912, __p3_912) __extension__ ({ \
68654 float32x2_t __ret_912; \
6865468655 float32x2_t __s0_912 = __p0_912; \
6865568656 float16x4_t __s1_912 = __p1_912; \
6865668657 float16x8_t __s2_912 = __p2_912; \
6865768658 float32x2_t __rev0_912; __rev0_912 = __builtin_shufflevector(__s0_912, __s0_912, 1, 0); \
6865868659 float16x4_t __rev1_912; __rev1_912 = __builtin_shufflevector(__s1_912, __s1_912, 3, 2, 1, 0); \
6865968660 float16x8_t __rev2_912; __rev2_912 = __builtin_shufflevector(__s2_912, __s2_912, 7, 6, 5, 4, 3, 2, 1, 0); \
68660 float32x2_t __ret_912; \
6866168661 __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)}); \
6866268662 __ret_912 = __builtin_shufflevector(__ret_912, __ret_912, 1, 0); \
6866368663 __ret_912; \
......@@ -68666,22 +68666,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6866668666
6866768667#ifdef __LITTLE_ENDIAN__
6866868668#define vfmlslq_laneq_low_f16(__p0_913, __p1_913, __p2_913, __p3_913) __extension__ ({ \
68669 float32x4_t __ret_913; \
6866968670 float32x4_t __s0_913 = __p0_913; \
6867068671 float16x8_t __s1_913 = __p1_913; \
6867168672 float16x8_t __s2_913 = __p2_913; \
68672 float32x4_t __ret_913; \
6867368673 __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)}); \
6867468674 __ret_913; \
6867568675})
6867668676#else
6867768677#define vfmlslq_laneq_low_f16(__p0_914, __p1_914, __p2_914, __p3_914) __extension__ ({ \
68678 float32x4_t __ret_914; \
6867868679 float32x4_t __s0_914 = __p0_914; \
6867968680 float16x8_t __s1_914 = __p1_914; \
6868068681 float16x8_t __s2_914 = __p2_914; \
6868168682 float32x4_t __rev0_914; __rev0_914 = __builtin_shufflevector(__s0_914, __s0_914, 3, 2, 1, 0); \
6868268683 float16x8_t __rev1_914; __rev1_914 = __builtin_shufflevector(__s1_914, __s1_914, 7, 6, 5, 4, 3, 2, 1, 0); \
6868368684 float16x8_t __rev2_914; __rev2_914 = __builtin_shufflevector(__s2_914, __s2_914, 7, 6, 5, 4, 3, 2, 1, 0); \
68684 float32x4_t __ret_914; \
6868568685 __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)}); \
6868668686 __ret_914 = __builtin_shufflevector(__ret_914, __ret_914, 3, 2, 1, 0); \
6868768687 __ret_914; \
......@@ -68690,22 +68690,22 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6869068690
6869168691#ifdef __LITTLE_ENDIAN__
6869268692#define vfmlsl_laneq_low_f16(__p0_915, __p1_915, __p2_915, __p3_915) __extension__ ({ \
68693 float32x2_t __ret_915; \
6869368694 float32x2_t __s0_915 = __p0_915; \
6869468695 float16x4_t __s1_915 = __p1_915; \
6869568696 float16x8_t __s2_915 = __p2_915; \
68696 float32x2_t __ret_915; \
6869768697 __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)}); \
6869868698 __ret_915; \
6869968699})
6870068700#else
6870168701#define vfmlsl_laneq_low_f16(__p0_916, __p1_916, __p2_916, __p3_916) __extension__ ({ \
68702 float32x2_t __ret_916; \
6870268703 float32x2_t __s0_916 = __p0_916; \
6870368704 float16x4_t __s1_916 = __p1_916; \
6870468705 float16x8_t __s2_916 = __p2_916; \
6870568706 float32x2_t __rev0_916; __rev0_916 = __builtin_shufflevector(__s0_916, __s0_916, 1, 0); \
6870668707 float16x4_t __rev1_916; __rev1_916 = __builtin_shufflevector(__s1_916, __s1_916, 3, 2, 1, 0); \
6870768708 float16x8_t __rev2_916; __rev2_916 = __builtin_shufflevector(__s2_916, __s2_916, 7, 6, 5, 4, 3, 2, 1, 0); \
68708 float32x2_t __ret_916; \
6870968709 __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)}); \
6871068710 __ret_916 = __builtin_shufflevector(__ret_916, __ret_916, 1, 0); \
6871168711 __ret_916; \
......@@ -68716,18 +68716,18 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6871668716#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(__aarch64__)
6871768717#ifdef __LITTLE_ENDIAN__
6871868718#define vmulh_lane_f16(__p0_917, __p1_917, __p2_917) __extension__ ({ \
68719 float16_t __ret_917; \
6871968720 float16_t __s0_917 = __p0_917; \
6872068721 float16x4_t __s1_917 = __p1_917; \
68721 float16_t __ret_917; \
6872268722 __ret_917 = __s0_917 * vget_lane_f16(__s1_917, __p2_917); \
6872368723 __ret_917; \
6872468724})
6872568725#else
6872668726#define vmulh_lane_f16(__p0_918, __p1_918, __p2_918) __extension__ ({ \
68727 float16_t __ret_918; \
6872768728 float16_t __s0_918 = __p0_918; \
6872868729 float16x4_t __s1_918 = __p1_918; \
6872968730 float16x4_t __rev1_918; __rev1_918 = __builtin_shufflevector(__s1_918, __s1_918, 3, 2, 1, 0); \
68730 float16_t __ret_918; \
6873168731 __ret_918 = __s0_918 * __noswap_vget_lane_f16(__rev1_918, __p2_918); \
6873268732 __ret_918; \
6873368733})
......@@ -68735,18 +68735,18 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6873568735
6873668736#ifdef __LITTLE_ENDIAN__
6873768737#define vmulh_laneq_f16(__p0_919, __p1_919, __p2_919) __extension__ ({ \
68738 float16_t __ret_919; \
6873868739 float16_t __s0_919 = __p0_919; \
6873968740 float16x8_t __s1_919 = __p1_919; \
68740 float16_t __ret_919; \
6874168741 __ret_919 = __s0_919 * vgetq_lane_f16(__s1_919, __p2_919); \
6874268742 __ret_919; \
6874368743})
6874468744#else
6874568745#define vmulh_laneq_f16(__p0_920, __p1_920, __p2_920) __extension__ ({ \
68746 float16_t __ret_920; \
6874668747 float16_t __s0_920 = __p0_920; \
6874768748 float16x8_t __s1_920 = __p1_920; \
6874868749 float16x8_t __rev1_920; __rev1_920 = __builtin_shufflevector(__s1_920, __s1_920, 7, 6, 5, 4, 3, 2, 1, 0); \
68749 float16_t __ret_920; \
6875068750 __ret_920 = __s0_920 * __noswap_vgetq_lane_f16(__rev1_920, __p2_920); \
6875168751 __ret_920; \
6875268752})
......@@ -68756,23 +68756,23 @@ __ai float32x4_t vcvtq_low_f32_bf16(bfloat16x8_t __p0) {
6875668756#if defined(__ARM_FEATURE_MATMUL_INT8)
6875768757#ifdef __LITTLE_ENDIAN__
6875868758#define vsudotq_lane_s32(__p0_921, __p1_921, __p2_921, __p3_921) __extension__ ({ \
68759 int32x4_t __ret_921; \
6875968760 int32x4_t __s0_921 = __p0_921; \
6876068761 int8x16_t __s1_921 = __p1_921; \
6876168762 uint8x8_t __s2_921 = __p2_921; \
68762 int32x4_t __ret_921; \
6876368763uint8x8_t __reint_921 = __s2_921; \
6876468764 __ret_921 = vusdotq_s32(__s0_921, (uint8x16_t)(splatq_lane_s32(*(int32x2_t *) &__reint_921, __p3_921)), __s1_921); \
6876568765 __ret_921; \
6876668766})
6876768767#else
6876868768#define vsudotq_lane_s32(__p0_922, __p1_922, __p2_922, __p3_922) __extension__ ({ \
68769 int32x4_t __ret_922; \
6876968770 int32x4_t __s0_922 = __p0_922; \
6877068771 int8x16_t __s1_922 = __p1_922; \
6877168772 uint8x8_t __s2_922 = __p2_922; \
6877268773 int32x4_t __rev0_922; __rev0_922 = __builtin_shufflevector(__s0_922, __s0_922, 3, 2, 1, 0); \
6877368774 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); \
6877468775 uint8x8_t __rev2_922; __rev2_922 = __builtin_shufflevector(__s2_922, __s2_922, 7, 6, 5, 4, 3, 2, 1, 0); \
68775 int32x4_t __ret_922; \
6877668776uint8x8_t __reint_922 = __rev2_922; \
6877768777 __ret_922 = __noswap_vusdotq_s32(__rev0_922, (uint8x16_t)(__noswap_splatq_lane_s32(*(int32x2_t *) &__reint_922, __p3_922)), __rev1_922); \
6877868778 __ret_922 = __builtin_shufflevector(__ret_922, __ret_922, 3, 2, 1, 0); \
......@@ -68782,23 +68782,23 @@ uint8x8_t __reint_922 = __rev2_922; \
6878268782
6878368783#ifdef __LITTLE_ENDIAN__
6878468784#define vsudot_lane_s32(__p0_923, __p1_923, __p2_923, __p3_923) __extension__ ({ \
68785 int32x2_t __ret_923; \
6878568786 int32x2_t __s0_923 = __p0_923; \
6878668787 int8x8_t __s1_923 = __p1_923; \
6878768788 uint8x8_t __s2_923 = __p2_923; \
68788 int32x2_t __ret_923; \
6878968789uint8x8_t __reint_923 = __s2_923; \
6879068790 __ret_923 = vusdot_s32(__s0_923, (uint8x8_t)(splat_lane_s32(*(int32x2_t *) &__reint_923, __p3_923)), __s1_923); \
6879168791 __ret_923; \
6879268792})
6879368793#else
6879468794#define vsudot_lane_s32(__p0_924, __p1_924, __p2_924, __p3_924) __extension__ ({ \
68795 int32x2_t __ret_924; \
6879568796 int32x2_t __s0_924 = __p0_924; \
6879668797 int8x8_t __s1_924 = __p1_924; \
6879768798 uint8x8_t __s2_924 = __p2_924; \
6879868799 int32x2_t __rev0_924; __rev0_924 = __builtin_shufflevector(__s0_924, __s0_924, 1, 0); \
6879968800 int8x8_t __rev1_924; __rev1_924 = __builtin_shufflevector(__s1_924, __s1_924, 7, 6, 5, 4, 3, 2, 1, 0); \
6880068801 uint8x8_t __rev2_924; __rev2_924 = __builtin_shufflevector(__s2_924, __s2_924, 7, 6, 5, 4, 3, 2, 1, 0); \
68801 int32x2_t __ret_924; \
6880268802uint8x8_t __reint_924 = __rev2_924; \
6880368803 __ret_924 = __noswap_vusdot_s32(__rev0_924, (uint8x8_t)(__noswap_splat_lane_s32(*(int32x2_t *) &__reint_924, __p3_924)), __rev1_924); \
6880468804 __ret_924 = __builtin_shufflevector(__ret_924, __ret_924, 1, 0); \
......@@ -68816,9 +68816,9 @@ __ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
6881668816}
6881768817#else
6881868818__ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
68819 uint16x8_t __ret;
6881968820 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6882068821 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68821 uint16x8_t __ret;
6882268822 __ret = __noswap_vabdl_u8(__noswap_vget_high_u8(__rev0), __noswap_vget_high_u8(__rev1));
6882368823 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6882468824 return __ret;
......@@ -68833,9 +68833,9 @@ __ai uint64x2_t vabdl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
6883368833}
6883468834#else
6883568835__ai uint64x2_t vabdl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
68836 uint64x2_t __ret;
6883668837 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6883768838 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68838 uint64x2_t __ret;
6883968839 __ret = __noswap_vabdl_u32(__noswap_vget_high_u32(__rev0), __noswap_vget_high_u32(__rev1));
6884068840 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6884168841 return __ret;
......@@ -68850,9 +68850,9 @@ __ai uint32x4_t vabdl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
6885068850}
6885168851#else
6885268852__ai uint32x4_t vabdl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
68853 uint32x4_t __ret;
6885368854 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6885468855 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68855 uint32x4_t __ret;
6885668856 __ret = __noswap_vabdl_u16(__noswap_vget_high_u16(__rev0), __noswap_vget_high_u16(__rev1));
6885768857 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6885868858 return __ret;
......@@ -68867,9 +68867,9 @@ __ai int16x8_t vabdl_high_s8(int8x16_t __p0, int8x16_t __p1) {
6886768867}
6886868868#else
6886968869__ai int16x8_t vabdl_high_s8(int8x16_t __p0, int8x16_t __p1) {
68870 int16x8_t __ret;
6887068871 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6887168872 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68872 int16x8_t __ret;
6887368873 __ret = __noswap_vabdl_s8(__noswap_vget_high_s8(__rev0), __noswap_vget_high_s8(__rev1));
6887468874 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6887568875 return __ret;
......@@ -68884,9 +68884,9 @@ __ai int64x2_t vabdl_high_s32(int32x4_t __p0, int32x4_t __p1) {
6888468884}
6888568885#else
6888668886__ai int64x2_t vabdl_high_s32(int32x4_t __p0, int32x4_t __p1) {
68887 int64x2_t __ret;
6888768888 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6888868889 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68889 int64x2_t __ret;
6889068890 __ret = __noswap_vabdl_s32(__noswap_vget_high_s32(__rev0), __noswap_vget_high_s32(__rev1));
6889168891 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6889268892 return __ret;
......@@ -68901,9 +68901,9 @@ __ai int32x4_t vabdl_high_s16(int16x8_t __p0, int16x8_t __p1) {
6890168901}
6890268902#else
6890368903__ai int32x4_t vabdl_high_s16(int16x8_t __p0, int16x8_t __p1) {
68904 int32x4_t __ret;
6890468905 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6890568906 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68906 int32x4_t __ret;
6890768907 __ret = __noswap_vabdl_s16(__noswap_vget_high_s16(__rev0), __noswap_vget_high_s16(__rev1));
6890868908 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6890968909 return __ret;
......@@ -68918,9 +68918,9 @@ __ai uint16x8_t vaddl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
6891868918}
6891968919#else
6892068920__ai uint16x8_t vaddl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
68921 uint16x8_t __ret;
6892168922 uint8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6892268923 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68923 uint16x8_t __ret;
6892468924 __ret = __noswap_vmovl_high_u8(__rev0) + __noswap_vmovl_high_u8(__rev1);
6892568925 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6892668926 return __ret;
......@@ -68935,9 +68935,9 @@ __ai uint64x2_t vaddl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
6893568935}
6893668936#else
6893768937__ai uint64x2_t vaddl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
68938 uint64x2_t __ret;
6893868939 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6893968940 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68940 uint64x2_t __ret;
6894168941 __ret = __noswap_vmovl_high_u32(__rev0) + __noswap_vmovl_high_u32(__rev1);
6894268942 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6894368943 return __ret;
......@@ -68952,9 +68952,9 @@ __ai uint32x4_t vaddl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
6895268952}
6895368953#else
6895468954__ai uint32x4_t vaddl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
68955 uint32x4_t __ret;
6895568956 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6895668957 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68957 uint32x4_t __ret;
6895868958 __ret = __noswap_vmovl_high_u16(__rev0) + __noswap_vmovl_high_u16(__rev1);
6895968959 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6896068960 return __ret;
......@@ -68969,9 +68969,9 @@ __ai int16x8_t vaddl_high_s8(int8x16_t __p0, int8x16_t __p1) {
6896968969}
6897068970#else
6897168971__ai int16x8_t vaddl_high_s8(int8x16_t __p0, int8x16_t __p1) {
68972 int16x8_t __ret;
6897268973 int8x16_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6897368974 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68974 int16x8_t __ret;
6897568975 __ret = __noswap_vmovl_high_s8(__rev0) + __noswap_vmovl_high_s8(__rev1);
6897668976 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6897768977 return __ret;
......@@ -68986,9 +68986,9 @@ __ai int64x2_t vaddl_high_s32(int32x4_t __p0, int32x4_t __p1) {
6898668986}
6898768987#else
6898868988__ai int64x2_t vaddl_high_s32(int32x4_t __p0, int32x4_t __p1) {
68989 int64x2_t __ret;
6898968990 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6899068991 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68991 int64x2_t __ret;
6899268992 __ret = __noswap_vmovl_high_s32(__rev0) + __noswap_vmovl_high_s32(__rev1);
6899368993 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6899468994 return __ret;
......@@ -69003,9 +69003,9 @@ __ai int32x4_t vaddl_high_s16(int16x8_t __p0, int16x8_t __p1) {
6900369003}
6900469004#else
6900569005__ai int32x4_t vaddl_high_s16(int16x8_t __p0, int16x8_t __p1) {
69006 int32x4_t __ret;
6900669007 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6900769008 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69008 int32x4_t __ret;
6900969009 __ret = __noswap_vmovl_high_s16(__rev0) + __noswap_vmovl_high_s16(__rev1);
6901069010 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6901169011 return __ret;
......@@ -69020,9 +69020,9 @@ __ai uint16x8_t vaddw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
6902069020}
6902169021#else
6902269022__ai uint16x8_t vaddw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
69023 uint16x8_t __ret;
6902369024 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6902469025 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69025 uint16x8_t __ret;
6902669026 __ret = __rev0 + __noswap_vmovl_high_u8(__rev1);
6902769027 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6902869028 return __ret;
......@@ -69037,9 +69037,9 @@ __ai uint64x2_t vaddw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
6903769037}
6903869038#else
6903969039__ai uint64x2_t vaddw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
69040 uint64x2_t __ret;
6904069041 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6904169042 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69042 uint64x2_t __ret;
6904369043 __ret = __rev0 + __noswap_vmovl_high_u32(__rev1);
6904469044 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6904569045 return __ret;
......@@ -69054,9 +69054,9 @@ __ai uint32x4_t vaddw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
6905469054}
6905569055#else
6905669056__ai uint32x4_t vaddw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
69057 uint32x4_t __ret;
6905769058 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6905869059 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69059 uint32x4_t __ret;
6906069060 __ret = __rev0 + __noswap_vmovl_high_u16(__rev1);
6906169061 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6906269062 return __ret;
......@@ -69071,9 +69071,9 @@ __ai int16x8_t vaddw_high_s8(int16x8_t __p0, int8x16_t __p1) {
6907169071}
6907269072#else
6907369073__ai int16x8_t vaddw_high_s8(int16x8_t __p0, int8x16_t __p1) {
69074 int16x8_t __ret;
6907469075 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6907569076 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69076 int16x8_t __ret;
6907769077 __ret = __rev0 + __noswap_vmovl_high_s8(__rev1);
6907869078 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6907969079 return __ret;
......@@ -69088,9 +69088,9 @@ __ai int64x2_t vaddw_high_s32(int64x2_t __p0, int32x4_t __p1) {
6908869088}
6908969089#else
6909069090__ai int64x2_t vaddw_high_s32(int64x2_t __p0, int32x4_t __p1) {
69091 int64x2_t __ret;
6909169092 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6909269093 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69093 int64x2_t __ret;
6909469094 __ret = __rev0 + __noswap_vmovl_high_s32(__rev1);
6909569095 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6909669096 return __ret;
......@@ -69105,9 +69105,9 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6910569105}
6910669106#else
6910769107__ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
69108 int32x4_t __ret;
6910869109 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6910969110 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69110 int32x4_t __ret;
6911169111 __ret = __rev0 + __noswap_vmovl_high_s16(__rev1);
6911269112 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6911369113 return __ret;
......@@ -69116,18 +69116,18 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6911669116
6911769117#ifdef __LITTLE_ENDIAN__
6911869118#define vcopyq_lane_p64(__p0_925, __p1_925, __p2_925, __p3_925) __extension__ ({ \
69119 poly64x2_t __ret_925; \
6911969120 poly64x2_t __s0_925 = __p0_925; \
6912069121 poly64x1_t __s2_925 = __p2_925; \
69121 poly64x2_t __ret_925; \
6912269122 __ret_925 = vsetq_lane_p64(vget_lane_p64(__s2_925, __p3_925), __s0_925, __p1_925); \
6912369123 __ret_925; \
6912469124})
6912569125#else
6912669126#define vcopyq_lane_p64(__p0_926, __p1_926, __p2_926, __p3_926) __extension__ ({ \
69127 poly64x2_t __ret_926; \
6912769128 poly64x2_t __s0_926 = __p0_926; \
6912869129 poly64x1_t __s2_926 = __p2_926; \
6912969130 poly64x2_t __rev0_926; __rev0_926 = __builtin_shufflevector(__s0_926, __s0_926, 1, 0); \
69130 poly64x2_t __ret_926; \
6913169131 __ret_926 = __noswap_vsetq_lane_p64(vget_lane_p64(__s2_926, __p3_926), __rev0_926, __p1_926); \
6913269132 __ret_926 = __builtin_shufflevector(__ret_926, __ret_926, 1, 0); \
6913369133 __ret_926; \
......@@ -69136,18 +69136,18 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6913669136
6913769137#ifdef __LITTLE_ENDIAN__
6913869138#define vcopyq_lane_f64(__p0_927, __p1_927, __p2_927, __p3_927) __extension__ ({ \
69139 float64x2_t __ret_927; \
6913969140 float64x2_t __s0_927 = __p0_927; \
6914069141 float64x1_t __s2_927 = __p2_927; \
69141 float64x2_t __ret_927; \
6914269142 __ret_927 = vsetq_lane_f64(vget_lane_f64(__s2_927, __p3_927), __s0_927, __p1_927); \
6914369143 __ret_927; \
6914469144})
6914569145#else
6914669146#define vcopyq_lane_f64(__p0_928, __p1_928, __p2_928, __p3_928) __extension__ ({ \
69147 float64x2_t __ret_928; \
6914769148 float64x2_t __s0_928 = __p0_928; \
6914869149 float64x1_t __s2_928 = __p2_928; \
6914969150 float64x2_t __rev0_928; __rev0_928 = __builtin_shufflevector(__s0_928, __s0_928, 1, 0); \
69150 float64x2_t __ret_928; \
6915169151 __ret_928 = __noswap_vsetq_lane_f64(vget_lane_f64(__s2_928, __p3_928), __rev0_928, __p1_928); \
6915269152 __ret_928 = __builtin_shufflevector(__ret_928, __ret_928, 1, 0); \
6915369153 __ret_928; \
......@@ -69155,34 +69155,34 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6915569155#endif
6915669156
6915769157#define vcopy_lane_p64(__p0_929, __p1_929, __p2_929, __p3_929) __extension__ ({ \
69158 poly64x1_t __ret_929; \
6915869159 poly64x1_t __s0_929 = __p0_929; \
6915969160 poly64x1_t __s2_929 = __p2_929; \
69160 poly64x1_t __ret_929; \
6916169161 __ret_929 = vset_lane_p64(vget_lane_p64(__s2_929, __p3_929), __s0_929, __p1_929); \
6916269162 __ret_929; \
6916369163})
6916469164#define vcopy_lane_f64(__p0_930, __p1_930, __p2_930, __p3_930) __extension__ ({ \
69165 float64x1_t __ret_930; \
6916569166 float64x1_t __s0_930 = __p0_930; \
6916669167 float64x1_t __s2_930 = __p2_930; \
69167 float64x1_t __ret_930; \
6916869168 __ret_930 = vset_lane_f64(vget_lane_f64(__s2_930, __p3_930), __s0_930, __p1_930); \
6916969169 __ret_930; \
6917069170})
6917169171#ifdef __LITTLE_ENDIAN__
6917269172#define vcopyq_laneq_p64(__p0_931, __p1_931, __p2_931, __p3_931) __extension__ ({ \
69173 poly64x2_t __ret_931; \
6917369174 poly64x2_t __s0_931 = __p0_931; \
6917469175 poly64x2_t __s2_931 = __p2_931; \
69175 poly64x2_t __ret_931; \
6917669176 __ret_931 = vsetq_lane_p64(vgetq_lane_p64(__s2_931, __p3_931), __s0_931, __p1_931); \
6917769177 __ret_931; \
6917869178})
6917969179#else
6918069180#define vcopyq_laneq_p64(__p0_932, __p1_932, __p2_932, __p3_932) __extension__ ({ \
69181 poly64x2_t __ret_932; \
6918169182 poly64x2_t __s0_932 = __p0_932; \
6918269183 poly64x2_t __s2_932 = __p2_932; \
6918369184 poly64x2_t __rev0_932; __rev0_932 = __builtin_shufflevector(__s0_932, __s0_932, 1, 0); \
6918469185 poly64x2_t __rev2_932; __rev2_932 = __builtin_shufflevector(__s2_932, __s2_932, 1, 0); \
69185 poly64x2_t __ret_932; \
6918669186 __ret_932 = __noswap_vsetq_lane_p64(__noswap_vgetq_lane_p64(__rev2_932, __p3_932), __rev0_932, __p1_932); \
6918769187 __ret_932 = __builtin_shufflevector(__ret_932, __ret_932, 1, 0); \
6918869188 __ret_932; \
......@@ -69191,19 +69191,19 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6919169191
6919269192#ifdef __LITTLE_ENDIAN__
6919369193#define vcopyq_laneq_f64(__p0_933, __p1_933, __p2_933, __p3_933) __extension__ ({ \
69194 float64x2_t __ret_933; \
6919469195 float64x2_t __s0_933 = __p0_933; \
6919569196 float64x2_t __s2_933 = __p2_933; \
69196 float64x2_t __ret_933; \
6919769197 __ret_933 = vsetq_lane_f64(vgetq_lane_f64(__s2_933, __p3_933), __s0_933, __p1_933); \
6919869198 __ret_933; \
6919969199})
6920069200#else
6920169201#define vcopyq_laneq_f64(__p0_934, __p1_934, __p2_934, __p3_934) __extension__ ({ \
69202 float64x2_t __ret_934; \
6920269203 float64x2_t __s0_934 = __p0_934; \
6920369204 float64x2_t __s2_934 = __p2_934; \
6920469205 float64x2_t __rev0_934; __rev0_934 = __builtin_shufflevector(__s0_934, __s0_934, 1, 0); \
6920569206 float64x2_t __rev2_934; __rev2_934 = __builtin_shufflevector(__s2_934, __s2_934, 1, 0); \
69206 float64x2_t __ret_934; \
6920769207 __ret_934 = __noswap_vsetq_lane_f64(__noswap_vgetq_lane_f64(__rev2_934, __p3_934), __rev0_934, __p1_934); \
6920869208 __ret_934 = __builtin_shufflevector(__ret_934, __ret_934, 1, 0); \
6920969209 __ret_934; \
......@@ -69212,18 +69212,18 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6921269212
6921369213#ifdef __LITTLE_ENDIAN__
6921469214#define vcopy_laneq_p64(__p0_935, __p1_935, __p2_935, __p3_935) __extension__ ({ \
69215 poly64x1_t __ret_935; \
6921569216 poly64x1_t __s0_935 = __p0_935; \
6921669217 poly64x2_t __s2_935 = __p2_935; \
69217 poly64x1_t __ret_935; \
6921869218 __ret_935 = vset_lane_p64(vgetq_lane_p64(__s2_935, __p3_935), __s0_935, __p1_935); \
6921969219 __ret_935; \
6922069220})
6922169221#else
6922269222#define vcopy_laneq_p64(__p0_936, __p1_936, __p2_936, __p3_936) __extension__ ({ \
69223 poly64x1_t __ret_936; \
6922369224 poly64x1_t __s0_936 = __p0_936; \
6922469225 poly64x2_t __s2_936 = __p2_936; \
6922569226 poly64x2_t __rev2_936; __rev2_936 = __builtin_shufflevector(__s2_936, __s2_936, 1, 0); \
69226 poly64x1_t __ret_936; \
6922769227 __ret_936 = vset_lane_p64(__noswap_vgetq_lane_p64(__rev2_936, __p3_936), __s0_936, __p1_936); \
6922869228 __ret_936; \
6922969229})
......@@ -69231,18 +69231,18 @@ __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
6923169231
6923269232#ifdef __LITTLE_ENDIAN__
6923369233#define vcopy_laneq_f64(__p0_937, __p1_937, __p2_937, __p3_937) __extension__ ({ \
69234 float64x1_t __ret_937; \
6923469235 float64x1_t __s0_937 = __p0_937; \
6923569236 float64x2_t __s2_937 = __p2_937; \
69236 float64x1_t __ret_937; \
6923769237 __ret_937 = vset_lane_f64(vgetq_lane_f64(__s2_937, __p3_937), __s0_937, __p1_937); \
6923869238 __ret_937; \
6923969239})
6924069240#else
6924169241#define vcopy_laneq_f64(__p0_938, __p1_938, __p2_938, __p3_938) __extension__ ({ \
69242 float64x1_t __ret_938; \
6924269243 float64x1_t __s0_938 = __p0_938; \
6924369244 float64x2_t __s2_938 = __p2_938; \
6924469245 float64x2_t __rev2_938; __rev2_938 = __builtin_shufflevector(__s2_938, __s2_938, 1, 0); \
69245 float64x1_t __ret_938; \
6924669246 __ret_938 = vset_lane_f64(__noswap_vgetq_lane_f64(__rev2_938, __p3_938), __s0_938, __p1_938); \
6924769247 __ret_938; \
6924869248})
......@@ -69256,10 +69256,10 @@ __ai uint16x8_t vmlal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2)
6925669256}
6925769257#else
6925869258__ai uint16x8_t vmlal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
69259 uint16x8_t __ret;
6925969260 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6926069261 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6926169262 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69262 uint16x8_t __ret;
6926369263 __ret = __noswap_vmlal_u8(__rev0, __noswap_vget_high_u8(__rev1), __noswap_vget_high_u8(__rev2));
6926469264 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6926569265 return __ret;
......@@ -69274,10 +69274,10 @@ __ai uint64x2_t vmlal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2
6927469274}
6927569275#else
6927669276__ai uint64x2_t vmlal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
69277 uint64x2_t __ret;
6927769278 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6927869279 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6927969280 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69280 uint64x2_t __ret;
6928169281 __ret = __noswap_vmlal_u32(__rev0, __noswap_vget_high_u32(__rev1), __noswap_vget_high_u32(__rev2));
6928269282 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6928369283 return __ret;
......@@ -69292,10 +69292,10 @@ __ai uint32x4_t vmlal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2
6929269292}
6929369293#else
6929469294__ai uint32x4_t vmlal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
69295 uint32x4_t __ret;
6929569296 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6929669297 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6929769298 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69298 uint32x4_t __ret;
6929969299 __ret = __noswap_vmlal_u16(__rev0, __noswap_vget_high_u16(__rev1), __noswap_vget_high_u16(__rev2));
6930069300 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6930169301 return __ret;
......@@ -69310,10 +69310,10 @@ __ai int16x8_t vmlal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
6931069310}
6931169311#else
6931269312__ai int16x8_t vmlal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
69313 int16x8_t __ret;
6931369314 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6931469315 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6931569316 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69316 int16x8_t __ret;
6931769317 __ret = __noswap_vmlal_s8(__rev0, __noswap_vget_high_s8(__rev1), __noswap_vget_high_s8(__rev2));
6931869318 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6931969319 return __ret;
......@@ -69328,10 +69328,10 @@ __ai int64x2_t vmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
6932869328}
6932969329#else
6933069330__ai int64x2_t vmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
69331 int64x2_t __ret;
6933169332 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6933269333 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6933369334 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69334 int64x2_t __ret;
6933569335 __ret = __noswap_vmlal_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
6933669336 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6933769337 return __ret;
......@@ -69346,10 +69346,10 @@ __ai int32x4_t vmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
6934669346}
6934769347#else
6934869348__ai int32x4_t vmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
69349 int32x4_t __ret;
6934969350 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6935069351 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6935169352 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69352 int32x4_t __ret;
6935369353 __ret = __noswap_vmlal_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
6935469354 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6935569355 return __ret;
......@@ -69364,9 +69364,9 @@ __ai uint64x2_t vmlal_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2
6936469364}
6936569365#else
6936669366__ai uint64x2_t vmlal_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
69367 uint64x2_t __ret;
6936769368 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6936869369 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69369 uint64x2_t __ret;
6937069370 __ret = __noswap_vmlal_n_u32(__rev0, __noswap_vget_high_u32(__rev1), __p2);
6937169371 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6937269372 return __ret;
......@@ -69381,9 +69381,9 @@ __ai uint32x4_t vmlal_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2
6938169381}
6938269382#else
6938369383__ai uint32x4_t vmlal_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
69384 uint32x4_t __ret;
6938469385 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6938569386 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69386 uint32x4_t __ret;
6938769387 __ret = __noswap_vmlal_n_u16(__rev0, __noswap_vget_high_u16(__rev1), __p2);
6938869388 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6938969389 return __ret;
......@@ -69398,9 +69398,9 @@ __ai int64x2_t vmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
6939869398}
6939969399#else
6940069400__ai int64x2_t vmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
69401 int64x2_t __ret;
6940169402 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6940269403 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69403 int64x2_t __ret;
6940469404 __ret = __noswap_vmlal_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
6940569405 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6940669406 return __ret;
......@@ -69415,9 +69415,9 @@ __ai int32x4_t vmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
6941569415}
6941669416#else
6941769417__ai int32x4_t vmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
69418 int32x4_t __ret;
6941869419 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6941969420 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69420 int32x4_t __ret;
6942169421 __ret = __noswap_vmlal_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
6942269422 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6942369423 return __ret;
......@@ -69432,10 +69432,10 @@ __ai uint16x8_t vmlsl_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2)
6943269432}
6943369433#else
6943469434__ai uint16x8_t vmlsl_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
69435 uint16x8_t __ret;
6943569436 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6943669437 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6943769438 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69438 uint16x8_t __ret;
6943969439 __ret = __noswap_vmlsl_u8(__rev0, __noswap_vget_high_u8(__rev1), __noswap_vget_high_u8(__rev2));
6944069440 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6944169441 return __ret;
......@@ -69450,10 +69450,10 @@ __ai uint64x2_t vmlsl_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2
6945069450}
6945169451#else
6945269452__ai uint64x2_t vmlsl_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
69453 uint64x2_t __ret;
6945369454 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6945469455 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6945569456 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69456 uint64x2_t __ret;
6945769457 __ret = __noswap_vmlsl_u32(__rev0, __noswap_vget_high_u32(__rev1), __noswap_vget_high_u32(__rev2));
6945869458 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6945969459 return __ret;
......@@ -69468,10 +69468,10 @@ __ai uint32x4_t vmlsl_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2
6946869468}
6946969469#else
6947069470__ai uint32x4_t vmlsl_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
69471 uint32x4_t __ret;
6947169472 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6947269473 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6947369474 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69474 uint32x4_t __ret;
6947569475 __ret = __noswap_vmlsl_u16(__rev0, __noswap_vget_high_u16(__rev1), __noswap_vget_high_u16(__rev2));
6947669476 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6947769477 return __ret;
......@@ -69486,10 +69486,10 @@ __ai int16x8_t vmlsl_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
6948669486}
6948769487#else
6948869488__ai int16x8_t vmlsl_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
69489 int16x8_t __ret;
6948969490 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6949069491 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6949169492 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69492 int16x8_t __ret;
6949369493 __ret = __noswap_vmlsl_s8(__rev0, __noswap_vget_high_s8(__rev1), __noswap_vget_high_s8(__rev2));
6949469494 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6949569495 return __ret;
......@@ -69504,10 +69504,10 @@ __ai int64x2_t vmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
6950469504}
6950569505#else
6950669506__ai int64x2_t vmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
69507 int64x2_t __ret;
6950769508 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6950869509 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6950969510 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69510 int64x2_t __ret;
6951169511 __ret = __noswap_vmlsl_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
6951269512 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6951369513 return __ret;
......@@ -69522,10 +69522,10 @@ __ai int32x4_t vmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
6952269522}
6952369523#else
6952469524__ai int32x4_t vmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
69525 int32x4_t __ret;
6952569526 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6952669527 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6952769528 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69528 int32x4_t __ret;
6952969529 __ret = __noswap_vmlsl_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
6953069530 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6953169531 return __ret;
......@@ -69540,9 +69540,9 @@ __ai uint64x2_t vmlsl_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2
6954069540}
6954169541#else
6954269542__ai uint64x2_t vmlsl_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
69543 uint64x2_t __ret;
6954369544 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6954469545 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69545 uint64x2_t __ret;
6954669546 __ret = __noswap_vmlsl_n_u32(__rev0, __noswap_vget_high_u32(__rev1), __p2);
6954769547 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6954869548 return __ret;
......@@ -69557,9 +69557,9 @@ __ai uint32x4_t vmlsl_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2
6955769557}
6955869558#else
6955969559__ai uint32x4_t vmlsl_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
69560 uint32x4_t __ret;
6956069561 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6956169562 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69562 uint32x4_t __ret;
6956369563 __ret = __noswap_vmlsl_n_u16(__rev0, __noswap_vget_high_u16(__rev1), __p2);
6956469564 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6956569565 return __ret;
......@@ -69574,9 +69574,9 @@ __ai int64x2_t vmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
6957469574}
6957569575#else
6957669576__ai int64x2_t vmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
69577 int64x2_t __ret;
6957769578 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6957869579 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69579 int64x2_t __ret;
6958069580 __ret = __noswap_vmlsl_n_s32(__rev0, __noswap_vget_high_s32(__rev1), __p2);
6958169581 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6958269582 return __ret;
......@@ -69591,9 +69591,9 @@ __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
6959169591}
6959269592#else
6959369593__ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
69594 int32x4_t __ret;
6959469595 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6959569596 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69596 int32x4_t __ret;
6959769597 __ret = __noswap_vmlsl_n_s16(__rev0, __noswap_vget_high_s16(__rev1), __p2);
6959869598 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6959969599 return __ret;
......@@ -69601,9 +69601,9 @@ __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
6960169601#endif
6960269602
6960369603#define vmulx_lane_f64(__p0_939, __p1_939, __p2_939) __extension__ ({ \
69604 float64x1_t __ret_939; \
6960469605 float64x1_t __s0_939 = __p0_939; \
6960569606 float64x1_t __s1_939 = __p1_939; \
69606 float64x1_t __ret_939; \
6960769607 float64_t __x_939 = vget_lane_f64(__s0_939, 0); \
6960869608 float64_t __y_939 = vget_lane_f64(__s1_939, __p2_939); \
6960969609 float64_t __z_939 = vmulxd_f64(__x_939, __y_939); \
......@@ -69612,9 +69612,9 @@ __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
6961269612})
6961369613#ifdef __LITTLE_ENDIAN__
6961469614#define vmulx_laneq_f64(__p0_940, __p1_940, __p2_940) __extension__ ({ \
69615 float64x1_t __ret_940; \
6961569616 float64x1_t __s0_940 = __p0_940; \
6961669617 float64x2_t __s1_940 = __p1_940; \
69617 float64x1_t __ret_940; \
6961869618 float64_t __x_940 = vget_lane_f64(__s0_940, 0); \
6961969619 float64_t __y_940 = vgetq_lane_f64(__s1_940, __p2_940); \
6962069620 float64_t __z_940 = vmulxd_f64(__x_940, __y_940); \
......@@ -69623,10 +69623,10 @@ __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
6962369623})
6962469624#else
6962569625#define vmulx_laneq_f64(__p0_941, __p1_941, __p2_941) __extension__ ({ \
69626 float64x1_t __ret_941; \
6962669627 float64x1_t __s0_941 = __p0_941; \
6962769628 float64x2_t __s1_941 = __p1_941; \
6962869629 float64x2_t __rev1_941; __rev1_941 = __builtin_shufflevector(__s1_941, __s1_941, 1, 0); \
69629 float64x1_t __ret_941; \
6963069630 float64_t __x_941 = vget_lane_f64(__s0_941, 0); \
6963169631 float64_t __y_941 = __noswap_vgetq_lane_f64(__rev1_941, __p2_941); \
6963269632 float64_t __z_941 = vmulxd_f64(__x_941, __y_941); \
......@@ -69644,10 +69644,10 @@ __ai uint16x8_t vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
6964469644}
6964569645#else
6964669646__ai uint16x8_t vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
69647 uint16x8_t __ret;
6964769648 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6964869649 uint8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6964969650 uint8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69650 uint16x8_t __ret;
6965169651 __ret = __rev0 + __noswap_vabdl_u8(__rev1, __rev2);
6965269652 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6965369653 return __ret;
......@@ -69667,10 +69667,10 @@ __ai uint64x2_t vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
6966769667}
6966869668#else
6966969669__ai uint64x2_t vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
69670 uint64x2_t __ret;
6967069671 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6967169672 uint32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6967269673 uint32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
69673 uint64x2_t __ret;
6967469674 __ret = __rev0 + __noswap_vabdl_u32(__rev1, __rev2);
6967569675 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6967669676 return __ret;
......@@ -69690,10 +69690,10 @@ __ai uint32x4_t vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
6969069690}
6969169691#else
6969269692__ai uint32x4_t vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
69693 uint32x4_t __ret;
6969369694 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6969469695 uint16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6969569696 uint16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69696 uint32x4_t __ret;
6969769697 __ret = __rev0 + __noswap_vabdl_u16(__rev1, __rev2);
6969869698 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6969969699 return __ret;
......@@ -69713,10 +69713,10 @@ __ai int16x8_t vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
6971369713}
6971469714#else
6971569715__ai int16x8_t vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
69716 int16x8_t __ret;
6971669717 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6971769718 int8x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6971869719 int8x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69719 int16x8_t __ret;
6972069720 __ret = __rev0 + __noswap_vabdl_s8(__rev1, __rev2);
6972169721 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6972269722 return __ret;
......@@ -69736,10 +69736,10 @@ __ai int64x2_t vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
6973669736}
6973769737#else
6973869738__ai int64x2_t vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
69739 int64x2_t __ret;
6973969740 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6974069741 int32x2_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6974169742 int32x2_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 1, 0);
69742 int64x2_t __ret;
6974369743 __ret = __rev0 + __noswap_vabdl_s32(__rev1, __rev2);
6974469744 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6974569745 return __ret;
......@@ -69759,10 +69759,10 @@ __ai int32x4_t vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
6975969759}
6976069760#else
6976169761__ai int32x4_t vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
69762 int32x4_t __ret;
6976269763 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6976369764 int16x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6976469765 int16x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69765 int32x4_t __ret;
6976669766 __ret = __rev0 + __noswap_vabdl_s16(__rev1, __rev2);
6976769767 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6976869768 return __ret;
......@@ -69783,10 +69783,10 @@ __ai uint16x8_t vabal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2)
6978369783}
6978469784#else
6978569785__ai uint16x8_t vabal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
69786 uint16x8_t __ret;
6978669787 uint16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6978769788 uint8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6978869789 uint8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69789 uint16x8_t __ret;
6979069790 __ret = __noswap_vabal_u8(__rev0, __noswap_vget_high_u8(__rev1), __noswap_vget_high_u8(__rev2));
6979169791 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6979269792 return __ret;
......@@ -69801,10 +69801,10 @@ __ai uint64x2_t vabal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2
6980169801}
6980269802#else
6980369803__ai uint64x2_t vabal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
69804 uint64x2_t __ret;
6980469805 uint64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6980569806 uint32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6980669807 uint32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69807 uint64x2_t __ret;
6980869808 __ret = __noswap_vabal_u32(__rev0, __noswap_vget_high_u32(__rev1), __noswap_vget_high_u32(__rev2));
6980969809 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6981069810 return __ret;
......@@ -69819,10 +69819,10 @@ __ai uint32x4_t vabal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2
6981969819}
6982069820#else
6982169821__ai uint32x4_t vabal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
69822 uint32x4_t __ret;
6982269823 uint32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6982369824 uint16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6982469825 uint16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69825 uint32x4_t __ret;
6982669826 __ret = __noswap_vabal_u16(__rev0, __noswap_vget_high_u16(__rev1), __noswap_vget_high_u16(__rev2));
6982769827 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6982869828 return __ret;
......@@ -69837,10 +69837,10 @@ __ai int16x8_t vabal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
6983769837}
6983869838#else
6983969839__ai int16x8_t vabal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
69840 int16x8_t __ret;
6984069841 int16x8_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 7, 6, 5, 4, 3, 2, 1, 0);
6984169842 int8x16_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
6984269843 int8x16_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69843 int16x8_t __ret;
6984469844 __ret = __noswap_vabal_s8(__rev0, __noswap_vget_high_s8(__rev1), __noswap_vget_high_s8(__rev2));
6984569845 __ret = __builtin_shufflevector(__ret, __ret, 7, 6, 5, 4, 3, 2, 1, 0);
6984669846 return __ret;
......@@ -69855,10 +69855,10 @@ __ai int64x2_t vabal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
6985569855}
6985669856#else
6985769857__ai int64x2_t vabal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
69858 int64x2_t __ret;
6985869859 int64x2_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 1, 0);
6985969860 int32x4_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6986069861 int32x4_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 3, 2, 1, 0);
69861 int64x2_t __ret;
6986269862 __ret = __noswap_vabal_s32(__rev0, __noswap_vget_high_s32(__rev1), __noswap_vget_high_s32(__rev2));
6986369863 __ret = __builtin_shufflevector(__ret, __ret, 1, 0);
6986469864 return __ret;
......@@ -69873,10 +69873,10 @@ __ai int32x4_t vabal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
6987369873}
6987469874#else
6987569875__ai int32x4_t vabal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
69876 int32x4_t __ret;
6987669877 int32x4_t __rev0; __rev0 = __builtin_shufflevector(__p0, __p0, 3, 2, 1, 0);
6987769878 int16x8_t __rev1; __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
6987869879 int16x8_t __rev2; __rev2 = __builtin_shufflevector(__p2, __p2, 7, 6, 5, 4, 3, 2, 1, 0);
69879 int32x4_t __ret;
6988069880 __ret = __noswap_vabal_s16(__rev0, __noswap_vget_high_s16(__rev1), __noswap_vget_high_s16(__rev2));
6988169881 __ret = __builtin_shufflevector(__ret, __ret, 3, 2, 1, 0);
6988269882 return __ret;
lib/include/opencl-c-base.h+2
......@@ -67,6 +67,8 @@
6767#if (__OPENCL_CPP_VERSION__ == 202100 || __OPENCL_C_VERSION__ == 300)
6868// For the SPIR and SPIR-V target all features are supported.
6969#if defined(__SPIR__) || defined(__SPIRV__)
70#define __opencl_c_atomic_order_seq_cst 1
71#define __opencl_c_atomic_scope_device 1
7072#define __opencl_c_atomic_scope_all_devices 1
7173#define __opencl_c_read_write_images 1
7274#endif // defined(__SPIR__)
lib/include/opencl-c.h+6
......@@ -13832,6 +13832,7 @@ float __ovld atomic_fetch_max_explicit(volatile atomic_float *object,
1383213832#endif // defined(__opencl_c_ext_fp32_global_atomic_min_max) && \
1383313833 defined(__opencl_c_ext_fp32_local_atomic_min_max)
1383413834
13835#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
1383513836#if defined(__opencl_c_ext_fp64_global_atomic_min_max)
1383613837double __ovld atomic_fetch_min(volatile __global atomic_double *object,
1383713838 double operand);
......@@ -13882,6 +13883,8 @@ double __ovld atomic_fetch_max_explicit(volatile atomic_double *object,
1388213883 memory_scope scope);
1388313884#endif // defined(__opencl_c_ext_fp64_global_atomic_min_max) && \
1388413885 defined(__opencl_c_ext_fp64_local_atomic_min_max)
13886#endif // defined(cl_khr_int64_base_atomics) && \
13887 defined(cl_khr_int64_extended_atomics)
1388513888
1388613889#if defined(__opencl_c_ext_fp16_global_atomic_add)
1388713890half __ovld atomic_fetch_add(volatile __global atomic_half *object,
......@@ -13985,6 +13988,7 @@ float __ovld atomic_fetch_sub_explicit(volatile atomic_float *object,
1398513988#endif // defined(__opencl_c_ext_fp32_global_atomic_add) && \
1398613989 defined(__opencl_c_ext_fp32_local_atomic_add)
1398713990
13991#if defined(cl_khr_int64_base_atomics) && defined(cl_khr_int64_extended_atomics)
1398813992#if defined(__opencl_c_ext_fp64_global_atomic_add)
1398913993double __ovld atomic_fetch_add(volatile __global atomic_double *object,
1399013994 double operand);
......@@ -14035,6 +14039,8 @@ double __ovld atomic_fetch_sub_explicit(volatile atomic_double *object,
1403514039 memory_scope scope);
1403614040#endif // defined(__opencl_c_ext_fp64_global_atomic_add) && \
1403714041 defined(__opencl_c_ext_fp64_local_atomic_add)
14042#endif // defined(cl_khr_int64_base_atomics) && \
14043 defined(cl_khr_int64_extended_atomics)
1403814044
1403914045#endif // cl_ext_float_atomics
1404014046
lib/include/ppc_wrappers/mm_malloc.h+1-1
......@@ -19,7 +19,7 @@
1919#ifndef __cplusplus
2020extern int posix_memalign (void **, size_t, size_t);
2121#else
22extern "C" int posix_memalign (void **, size_t, size_t) throw ();
22extern "C" int posix_memalign (void **, size_t, size_t);
2323#endif
2424
2525static __inline void *