| ... | @@ -0,0 +1,1577 @@ |
| 1 | /* |
| 2 | * Copyright (c) 1999-2010 Apple Inc. All Rights Reserved. |
| 3 | * |
| 4 | * @APPLE_LICENSE_HEADER_START@ |
| 5 | * |
| 6 | * This file contains Original Code and/or Modifications of Original Code |
| 7 | * as defined in and that are subject to the Apple Public Source License |
| 8 | * Version 2.0 (the 'License'). You may not use this file except in |
| 9 | * compliance with the License. Please obtain a copy of the License at |
| 10 | * http://www.opensource.apple.com/apsl/ and read it before using this |
| 11 | * file. |
| 12 | * |
| 13 | * The Original Code and all software distributed under the License are |
| 14 | * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER |
| 15 | * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, |
| 16 | * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, |
| 17 | * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT. |
| 18 | * Please see the License for the specific language governing rights and |
| 19 | * limitations under the License. |
| 20 | * |
| 21 | * @APPLE_LICENSE_HEADER_END@ |
| 22 | */ |
| 23 | #ifndef _MACHO_LOADER_H_ |
| 24 | #define _MACHO_LOADER_H_ |
| 25 | |
| 26 | /* |
| 27 | * This file describes the format of mach object files. |
| 28 | */ |
| 29 | #include <stdint.h> |
| 30 | |
| 31 | /* |
| 32 | * <mach/machine.h> is needed here for the cpu_type_t and cpu_subtype_t types |
| 33 | * and contains the constants for the possible values of these types. |
| 34 | */ |
| 35 | #include <mach/machine.h> |
| 36 | |
| 37 | /* |
| 38 | * <mach/vm_prot.h> is needed here for the vm_prot_t type and contains the |
| 39 | * constants that are or'ed together for the possible values of this type. |
| 40 | */ |
| 41 | #include <mach/vm_prot.h> |
| 42 | |
| 43 | /* |
| 44 | * <machine/thread_status.h> is expected to define the flavors of the thread |
| 45 | * states and the structures of those flavors for each machine. |
| 46 | */ |
| 47 | #include <mach/machine/thread_status.h> |
| 48 | #include <architecture/byte_order.h> |
| 49 | |
| 50 | /* |
| 51 | * The 32-bit mach header appears at the very beginning of the object file for |
| 52 | * 32-bit architectures. |
| 53 | */ |
| 54 | struct mach_header { |
| 55 | uint32_t magic; /* mach magic number identifier */ |
| 56 | cpu_type_t cputype; /* cpu specifier */ |
| 57 | cpu_subtype_t cpusubtype; /* machine specifier */ |
| 58 | uint32_t filetype; /* type of file */ |
| 59 | uint32_t ncmds; /* number of load commands */ |
| 60 | uint32_t sizeofcmds; /* the size of all the load commands */ |
| 61 | uint32_t flags; /* flags */ |
| 62 | }; |
| 63 | |
| 64 | /* Constant for the magic field of the mach_header (32-bit architectures) */ |
| 65 | #define MH_MAGIC 0xfeedface /* the mach magic number */ |
| 66 | #define MH_CIGAM 0xcefaedfe /* NXSwapInt(MH_MAGIC) */ |
| 67 | |
| 68 | /* |
| 69 | * The 64-bit mach header appears at the very beginning of object files for |
| 70 | * 64-bit architectures. |
| 71 | */ |
| 72 | struct mach_header_64 { |
| 73 | uint32_t magic; /* mach magic number identifier */ |
| 74 | cpu_type_t cputype; /* cpu specifier */ |
| 75 | cpu_subtype_t cpusubtype; /* machine specifier */ |
| 76 | uint32_t filetype; /* type of file */ |
| 77 | uint32_t ncmds; /* number of load commands */ |
| 78 | uint32_t sizeofcmds; /* the size of all the load commands */ |
| 79 | uint32_t flags; /* flags */ |
| 80 | uint32_t reserved; /* reserved */ |
| 81 | }; |
| 82 | |
| 83 | /* Constant for the magic field of the mach_header_64 (64-bit architectures) */ |
| 84 | #define MH_MAGIC_64 0xfeedfacf /* the 64-bit mach magic number */ |
| 85 | #define MH_CIGAM_64 0xcffaedfe /* NXSwapInt(MH_MAGIC_64) */ |
| 86 | |
| 87 | /* |
| 88 | * The layout of the file depends on the filetype. For all but the MH_OBJECT |
| 89 | * file type the segments are padded out and aligned on a segment alignment |
| 90 | * boundary for efficient demand pageing. The MH_EXECUTE, MH_FVMLIB, MH_DYLIB, |
| 91 | * MH_DYLINKER and MH_BUNDLE file types also have the headers included as part |
| 92 | * of their first segment. |
| 93 | * |
| 94 | * The file type MH_OBJECT is a compact format intended as output of the |
| 95 | * assembler and input (and possibly output) of the link editor (the .o |
| 96 | * format). All sections are in one unnamed segment with no segment padding. |
| 97 | * This format is used as an executable format when the file is so small the |
| 98 | * segment padding greatly increases its size. |
| 99 | * |
| 100 | * The file type MH_PRELOAD is an executable format intended for things that |
| 101 | * are not executed under the kernel (proms, stand alones, kernels, etc). The |
| 102 | * format can be executed under the kernel but may demand paged it and not |
| 103 | * preload it before execution. |
| 104 | * |
| 105 | * A core file is in MH_CORE format and can be any in an arbritray legal |
| 106 | * Mach-O file. |
| 107 | * |
| 108 | * Constants for the filetype field of the mach_header |
| 109 | */ |
| 110 | #define MH_OBJECT 0x1 /* relocatable object file */ |
| 111 | #define MH_EXECUTE 0x2 /* demand paged executable file */ |
| 112 | #define MH_FVMLIB 0x3 /* fixed VM shared library file */ |
| 113 | #define MH_CORE 0x4 /* core file */ |
| 114 | #define MH_PRELOAD 0x5 /* preloaded executable file */ |
| 115 | #define MH_DYLIB 0x6 /* dynamically bound shared library */ |
| 116 | #define MH_DYLINKER 0x7 /* dynamic link editor */ |
| 117 | #define MH_BUNDLE 0x8 /* dynamically bound bundle file */ |
| 118 | #define MH_DYLIB_STUB 0x9 /* shared library stub for static */ |
| 119 | /* linking only, no section contents */ |
| 120 | #define MH_DSYM 0xa /* companion file with only debug */ |
| 121 | /* sections */ |
| 122 | #define MH_KEXT_BUNDLE 0xb /* x86_64 kexts */ |
| 123 | |
| 124 | /* Constants for the flags field of the mach_header */ |
| 125 | #define MH_NOUNDEFS 0x1 /* the object file has no undefined |
| 126 | references */ |
| 127 | #define MH_INCRLINK 0x2 /* the object file is the output of an |
| 128 | incremental link against a base file |
| 129 | and can't be link edited again */ |
| 130 | #define MH_DYLDLINK 0x4 /* the object file is input for the |
| 131 | dynamic linker and can't be staticly |
| 132 | link edited again */ |
| 133 | #define MH_BINDATLOAD 0x8 /* the object file's undefined |
| 134 | references are bound by the dynamic |
| 135 | linker when loaded. */ |
| 136 | #define MH_PREBOUND 0x10 /* the file has its dynamic undefined |
| 137 | references prebound. */ |
| 138 | #define MH_SPLIT_SEGS 0x20 /* the file has its read-only and |
| 139 | read-write segments split */ |
| 140 | #define MH_LAZY_INIT 0x40 /* the shared library init routine is |
| 141 | to be run lazily via catching memory |
| 142 | faults to its writeable segments |
| 143 | (obsolete) */ |
| 144 | #define MH_TWOLEVEL 0x80 /* the image is using two-level name |
| 145 | space bindings */ |
| 146 | #define MH_FORCE_FLAT 0x100 /* the executable is forcing all images |
| 147 | to use flat name space bindings */ |
| 148 | #define MH_NOMULTIDEFS 0x200 /* this umbrella guarantees no multiple |
| 149 | defintions of symbols in its |
| 150 | sub-images so the two-level namespace |
| 151 | hints can always be used. */ |
| 152 | #define MH_NOFIXPREBINDING 0x400 /* do not have dyld notify the |
| 153 | prebinding agent about this |
| 154 | executable */ |
| 155 | #define MH_PREBINDABLE 0x800 /* the binary is not prebound but can |
| 156 | have its prebinding redone. only used |
| 157 | when MH_PREBOUND is not set. */ |
| 158 | #define MH_ALLMODSBOUND 0x1000 /* indicates that this binary binds to |
| 159 | all two-level namespace modules of |
| 160 | its dependent libraries. only used |
| 161 | when MH_PREBINDABLE and MH_TWOLEVEL |
| 162 | are both set. */ |
| 163 | #define MH_SUBSECTIONS_VIA_SYMBOLS 0x2000/* safe to divide up the sections into |
| 164 | sub-sections via symbols for dead |
| 165 | code stripping */ |
| 166 | #define MH_CANONICAL 0x4000 /* the binary has been canonicalized |
| 167 | via the unprebind operation */ |
| 168 | #define MH_WEAK_DEFINES 0x8000 /* the final linked image contains |
| 169 | external weak symbols */ |
| 170 | #define MH_BINDS_TO_WEAK 0x10000 /* the final linked image uses |
| 171 | weak symbols */ |
| 172 | |
| 173 | #define MH_ALLOW_STACK_EXECUTION 0x20000/* When this bit is set, all stacks |
| 174 | in the task will be given stack |
| 175 | execution privilege. Only used in |
| 176 | MH_EXECUTE filetypes. */ |
| 177 | #define MH_ROOT_SAFE 0x40000 /* When this bit is set, the binary |
| 178 | declares it is safe for use in |
| 179 | processes with uid zero */ |
| 180 | |
| 181 | #define MH_SETUID_SAFE 0x80000 /* When this bit is set, the binary |
| 182 | declares it is safe for use in |
| 183 | processes when issetugid() is true */ |
| 184 | |
| 185 | #define MH_NO_REEXPORTED_DYLIBS 0x100000 /* When this bit is set on a dylib, |
| 186 | the static linker does not need to |
| 187 | examine dependent dylibs to see |
| 188 | if any are re-exported */ |
| 189 | #define MH_PIE 0x200000 /* When this bit is set, the OS will |
| 190 | load the main executable at a |
| 191 | random address. Only used in |
| 192 | MH_EXECUTE filetypes. */ |
| 193 | #define MH_DEAD_STRIPPABLE_DYLIB 0x400000 /* Only for use on dylibs. When |
| 194 | linking against a dylib that |
| 195 | has this bit set, the static linker |
| 196 | will automatically not create a |
| 197 | LC_LOAD_DYLIB load command to the |
| 198 | dylib if no symbols are being |
| 199 | referenced from the dylib. */ |
| 200 | #define MH_HAS_TLV_DESCRIPTORS 0x800000 /* Contains a section of type |
| 201 | S_THREAD_LOCAL_VARIABLES */ |
| 202 | |
| 203 | #define MH_NO_HEAP_EXECUTION 0x1000000 /* When this bit is set, the OS will |
| 204 | run the main executable with |
| 205 | a non-executable heap even on |
| 206 | platforms (e.g. i386) that don't |
| 207 | require it. Only used in MH_EXECUTE |
| 208 | filetypes. */ |
| 209 | |
| 210 | #define MH_APP_EXTENSION_SAFE 0x02000000 /* The code was linked for use in an |
| 211 | application extension. */ |
| 212 | |
| 213 | #define MH_NLIST_OUTOFSYNC_WITH_DYLDINFO 0x04000000 /* The external symbols |
| 214 | listed in the nlist symbol table do |
| 215 | not include all the symbols listed in |
| 216 | the dyld info. */ |
| 217 | |
| 218 | #define MH_SIM_SUPPORT 0x08000000 /* Allow LC_MIN_VERSION_MACOS and |
| 219 | LC_BUILD_VERSION load commands with |
| 220 | the platforms macOS, macCatalyst, |
| 221 | iOSSimulator, tvOSSimulator and |
| 222 | watchOSSimulator. */ |
| 223 | |
| 224 | #define MH_DYLIB_IN_CACHE 0x80000000 /* Only for use on dylibs. When this bit |
| 225 | is set, the dylib is part of the dyld |
| 226 | shared cache, rather than loose in |
| 227 | the filesystem. */ |
| 228 | |
| 229 | /* |
| 230 | * The load commands directly follow the mach_header. The total size of all |
| 231 | * of the commands is given by the sizeofcmds field in the mach_header. All |
| 232 | * load commands must have as their first two fields cmd and cmdsize. The cmd |
| 233 | * field is filled in with a constant for that command type. Each command type |
| 234 | * has a structure specifically for it. The cmdsize field is the size in bytes |
| 235 | * of the particular load command structure plus anything that follows it that |
| 236 | * is a part of the load command (i.e. section structures, strings, etc.). To |
| 237 | * advance to the next load command the cmdsize can be added to the offset or |
| 238 | * pointer of the current load command. The cmdsize for 32-bit architectures |
| 239 | * MUST be a multiple of 4 bytes and for 64-bit architectures MUST be a multiple |
| 240 | * of 8 bytes (these are forever the maximum alignment of any load commands). |
| 241 | * The padded bytes must be zero. All tables in the object file must also |
| 242 | * follow these rules so the file can be memory mapped. Otherwise the pointers |
| 243 | * to these tables will not work well or at all on some machines. With all |
| 244 | * padding zeroed like objects will compare byte for byte. |
| 245 | */ |
| 246 | struct load_command { |
| 247 | uint32_t cmd; /* type of load command */ |
| 248 | uint32_t cmdsize; /* total size of command in bytes */ |
| 249 | }; |
| 250 | |
| 251 | /* |
| 252 | * After MacOS X 10.1 when a new load command is added that is required to be |
| 253 | * understood by the dynamic linker for the image to execute properly the |
| 254 | * LC_REQ_DYLD bit will be or'ed into the load command constant. If the dynamic |
| 255 | * linker sees such a load command it it does not understand will issue a |
| 256 | * "unknown load command required for execution" error and refuse to use the |
| 257 | * image. Other load commands without this bit that are not understood will |
| 258 | * simply be ignored. |
| 259 | */ |
| 260 | #define LC_REQ_DYLD 0x80000000 |
| 261 | |
| 262 | /* Constants for the cmd field of all load commands, the type */ |
| 263 | #define LC_SEGMENT 0x1 /* segment of this file to be mapped */ |
| 264 | #define LC_SYMTAB 0x2 /* link-edit stab symbol table info */ |
| 265 | #define LC_SYMSEG 0x3 /* link-edit gdb symbol table info (obsolete) */ |
| 266 | #define LC_THREAD 0x4 /* thread */ |
| 267 | #define LC_UNIXTHREAD 0x5 /* unix thread (includes a stack) */ |
| 268 | #define LC_LOADFVMLIB 0x6 /* load a specified fixed VM shared library */ |
| 269 | #define LC_IDFVMLIB 0x7 /* fixed VM shared library identification */ |
| 270 | #define LC_IDENT 0x8 /* object identification info (obsolete) */ |
| 271 | #define LC_FVMFILE 0x9 /* fixed VM file inclusion (internal use) */ |
| 272 | #define LC_PREPAGE 0xa /* prepage command (internal use) */ |
| 273 | #define LC_DYSYMTAB 0xb /* dynamic link-edit symbol table info */ |
| 274 | #define LC_LOAD_DYLIB 0xc /* load a dynamically linked shared library */ |
| 275 | #define LC_ID_DYLIB 0xd /* dynamically linked shared lib ident */ |
| 276 | #define LC_LOAD_DYLINKER 0xe /* load a dynamic linker */ |
| 277 | #define LC_ID_DYLINKER 0xf /* dynamic linker identification */ |
| 278 | #define LC_PREBOUND_DYLIB 0x10 /* modules prebound for a dynamically */ |
| 279 | /* linked shared library */ |
| 280 | #define LC_ROUTINES 0x11 /* image routines */ |
| 281 | #define LC_SUB_FRAMEWORK 0x12 /* sub framework */ |
| 282 | #define LC_SUB_UMBRELLA 0x13 /* sub umbrella */ |
| 283 | #define LC_SUB_CLIENT 0x14 /* sub client */ |
| 284 | #define LC_SUB_LIBRARY 0x15 /* sub library */ |
| 285 | #define LC_TWOLEVEL_HINTS 0x16 /* two-level namespace lookup hints */ |
| 286 | #define LC_PREBIND_CKSUM 0x17 /* prebind checksum */ |
| 287 | |
| 288 | /* |
| 289 | * load a dynamically linked shared library that is allowed to be missing |
| 290 | * (all symbols are weak imported). |
| 291 | */ |
| 292 | #define LC_LOAD_WEAK_DYLIB (0x18 | LC_REQ_DYLD) |
| 293 | |
| 294 | #define LC_SEGMENT_64 0x19 /* 64-bit segment of this file to be |
| 295 | mapped */ |
| 296 | #define LC_ROUTINES_64 0x1a /* 64-bit image routines */ |
| 297 | #define LC_UUID 0x1b /* the uuid */ |
| 298 | #define LC_RPATH (0x1c | LC_REQ_DYLD) /* runpath additions */ |
| 299 | #define LC_CODE_SIGNATURE 0x1d /* local of code signature */ |
| 300 | #define LC_SEGMENT_SPLIT_INFO 0x1e /* local of info to split segments */ |
| 301 | #define LC_REEXPORT_DYLIB (0x1f | LC_REQ_DYLD) /* load and re-export dylib */ |
| 302 | #define LC_LAZY_LOAD_DYLIB 0x20 /* delay load of dylib until first use */ |
| 303 | #define LC_ENCRYPTION_INFO 0x21 /* encrypted segment information */ |
| 304 | #define LC_DYLD_INFO 0x22 /* compressed dyld information */ |
| 305 | #define LC_DYLD_INFO_ONLY (0x22|LC_REQ_DYLD) /* compressed dyld information only */ |
| 306 | #define LC_LOAD_UPWARD_DYLIB (0x23 | LC_REQ_DYLD) /* load upward dylib */ |
| 307 | #define LC_VERSION_MIN_MACOSX 0x24 /* build for MacOSX min OS version */ |
| 308 | #define LC_VERSION_MIN_IPHONEOS 0x25 /* build for iPhoneOS min OS version */ |
| 309 | #define LC_FUNCTION_STARTS 0x26 /* compressed table of function start addresses */ |
| 310 | #define LC_DYLD_ENVIRONMENT 0x27 /* string for dyld to treat |
| 311 | like environment variable */ |
| 312 | #define LC_MAIN (0x28|LC_REQ_DYLD) /* replacement for LC_UNIXTHREAD */ |
| 313 | #define LC_DATA_IN_CODE 0x29 /* table of non-instructions in __text */ |
| 314 | #define LC_SOURCE_VERSION 0x2A /* source version used to build binary */ |
| 315 | #define LC_DYLIB_CODE_SIGN_DRS 0x2B /* Code signing DRs copied from linked dylibs */ |
| 316 | #define LC_ENCRYPTION_INFO_64 0x2C /* 64-bit encrypted segment information */ |
| 317 | #define LC_LINKER_OPTION 0x2D /* linker options in MH_OBJECT files */ |
| 318 | #define LC_LINKER_OPTIMIZATION_HINT 0x2E /* optimization hints in MH_OBJECT files */ |
| 319 | #define LC_VERSION_MIN_TVOS 0x2F /* build for AppleTV min OS version */ |
| 320 | #define LC_VERSION_MIN_WATCHOS 0x30 /* build for Watch min OS version */ |
| 321 | #define LC_NOTE 0x31 /* arbitrary data included within a Mach-O file */ |
| 322 | #define LC_BUILD_VERSION 0x32 /* build for platform min OS version */ |
| 323 | #define LC_DYLD_EXPORTS_TRIE (0x33 | LC_REQ_DYLD) /* used with linkedit_data_command, payload is trie */ |
| 324 | #define LC_DYLD_CHAINED_FIXUPS (0x34 | LC_REQ_DYLD) /* used with linkedit_data_command */ |
| 325 | |
| 326 | /* |
| 327 | * A variable length string in a load command is represented by an lc_str |
| 328 | * union. The strings are stored just after the load command structure and |
| 329 | * the offset is from the start of the load command structure. The size |
| 330 | * of the string is reflected in the cmdsize field of the load command. |
| 331 | * Once again any padded bytes to bring the cmdsize field to a multiple |
| 332 | * of 4 bytes must be zero. |
| 333 | */ |
| 334 | union lc_str { |
| 335 | uint32_t offset; /* offset to the string */ |
| 336 | #ifndef __LP64__ |
| 337 | char *ptr; /* pointer to the string */ |
| 338 | #endif |
| 339 | }; |
| 340 | |
| 341 | /* |
| 342 | * The segment load command indicates that a part of this file is to be |
| 343 | * mapped into the task's address space. The size of this segment in memory, |
| 344 | * vmsize, maybe equal to or larger than the amount to map from this file, |
| 345 | * filesize. The file is mapped starting at fileoff to the beginning of |
| 346 | * the segment in memory, vmaddr. The rest of the memory of the segment, |
| 347 | * if any, is allocated zero fill on demand. The segment's maximum virtual |
| 348 | * memory protection and initial virtual memory protection are specified |
| 349 | * by the maxprot and initprot fields. If the segment has sections then the |
| 350 | * section structures directly follow the segment command and their size is |
| 351 | * reflected in cmdsize. |
| 352 | */ |
| 353 | struct segment_command { /* for 32-bit architectures */ |
| 354 | uint32_t cmd; /* LC_SEGMENT */ |
| 355 | uint32_t cmdsize; /* includes sizeof section structs */ |
| 356 | char segname[16]; /* segment name */ |
| 357 | uint32_t vmaddr; /* memory address of this segment */ |
| 358 | uint32_t vmsize; /* memory size of this segment */ |
| 359 | uint32_t fileoff; /* file offset of this segment */ |
| 360 | uint32_t filesize; /* amount to map from the file */ |
| 361 | vm_prot_t maxprot; /* maximum VM protection */ |
| 362 | vm_prot_t initprot; /* initial VM protection */ |
| 363 | uint32_t nsects; /* number of sections in segment */ |
| 364 | uint32_t flags; /* flags */ |
| 365 | }; |
| 366 | |
| 367 | /* |
| 368 | * The 64-bit segment load command indicates that a part of this file is to be |
| 369 | * mapped into a 64-bit task's address space. If the 64-bit segment has |
| 370 | * sections then section_64 structures directly follow the 64-bit segment |
| 371 | * command and their size is reflected in cmdsize. |
| 372 | */ |
| 373 | struct segment_command_64 { /* for 64-bit architectures */ |
| 374 | uint32_t cmd; /* LC_SEGMENT_64 */ |
| 375 | uint32_t cmdsize; /* includes sizeof section_64 structs */ |
| 376 | char segname[16]; /* segment name */ |
| 377 | uint64_t vmaddr; /* memory address of this segment */ |
| 378 | uint64_t vmsize; /* memory size of this segment */ |
| 379 | uint64_t fileoff; /* file offset of this segment */ |
| 380 | uint64_t filesize; /* amount to map from the file */ |
| 381 | vm_prot_t maxprot; /* maximum VM protection */ |
| 382 | vm_prot_t initprot; /* initial VM protection */ |
| 383 | uint32_t nsects; /* number of sections in segment */ |
| 384 | uint32_t flags; /* flags */ |
| 385 | }; |
| 386 | |
| 387 | /* Constants for the flags field of the segment_command */ |
| 388 | #define SG_HIGHVM 0x1 /* the file contents for this segment is for |
| 389 | the high part of the VM space, the low part |
| 390 | is zero filled (for stacks in core files) */ |
| 391 | #define SG_FVMLIB 0x2 /* this segment is the VM that is allocated by |
| 392 | a fixed VM library, for overlap checking in |
| 393 | the link editor */ |
| 394 | #define SG_NORELOC 0x4 /* this segment has nothing that was relocated |
| 395 | in it and nothing relocated to it, that is |
| 396 | it maybe safely replaced without relocation*/ |
| 397 | #define SG_PROTECTED_VERSION_1 0x8 /* This segment is protected. If the |
| 398 | segment starts at file offset 0, the |
| 399 | first page of the segment is not |
| 400 | protected. All other pages of the |
| 401 | segment are protected. */ |
| 402 | #define SG_READ_ONLY 0x10 /* This segment is made read-only after fixups */ |
| 403 | |
| 404 | |
| 405 | |
| 406 | /* |
| 407 | * A segment is made up of zero or more sections. Non-MH_OBJECT files have |
| 408 | * all of their segments with the proper sections in each, and padded to the |
| 409 | * specified segment alignment when produced by the link editor. The first |
| 410 | * segment of a MH_EXECUTE and MH_FVMLIB format file contains the mach_header |
| 411 | * and load commands of the object file before its first section. The zero |
| 412 | * fill sections are always last in their segment (in all formats). This |
| 413 | * allows the zeroed segment padding to be mapped into memory where zero fill |
| 414 | * sections might be. The gigabyte zero fill sections, those with the section |
| 415 | * type S_GB_ZEROFILL, can only be in a segment with sections of this type. |
| 416 | * These segments are then placed after all other segments. |
| 417 | * |
| 418 | * The MH_OBJECT format has all of its sections in one segment for |
| 419 | * compactness. There is no padding to a specified segment boundary and the |
| 420 | * mach_header and load commands are not part of the segment. |
| 421 | * |
| 422 | * Sections with the same section name, sectname, going into the same segment, |
| 423 | * segname, are combined by the link editor. The resulting section is aligned |
| 424 | * to the maximum alignment of the combined sections and is the new section's |
| 425 | * alignment. The combined sections are aligned to their original alignment in |
| 426 | * the combined section. Any padded bytes to get the specified alignment are |
| 427 | * zeroed. |
| 428 | * |
| 429 | * The format of the relocation entries referenced by the reloff and nreloc |
| 430 | * fields of the section structure for mach object files is described in the |
| 431 | * header file <reloc.h>. |
| 432 | */ |
| 433 | struct section { /* for 32-bit architectures */ |
| 434 | char sectname[16]; /* name of this section */ |
| 435 | char segname[16]; /* segment this section goes in */ |
| 436 | uint32_t addr; /* memory address of this section */ |
| 437 | uint32_t size; /* size in bytes of this section */ |
| 438 | uint32_t offset; /* file offset of this section */ |
| 439 | uint32_t align; /* section alignment (power of 2) */ |
| 440 | uint32_t reloff; /* file offset of relocation entries */ |
| 441 | uint32_t nreloc; /* number of relocation entries */ |
| 442 | uint32_t flags; /* flags (section type and attributes)*/ |
| 443 | uint32_t reserved1; /* reserved (for offset or index) */ |
| 444 | uint32_t reserved2; /* reserved (for count or sizeof) */ |
| 445 | }; |
| 446 | |
| 447 | struct section_64 { /* for 64-bit architectures */ |
| 448 | char sectname[16]; /* name of this section */ |
| 449 | char segname[16]; /* segment this section goes in */ |
| 450 | uint64_t addr; /* memory address of this section */ |
| 451 | uint64_t size; /* size in bytes of this section */ |
| 452 | uint32_t offset; /* file offset of this section */ |
| 453 | uint32_t align; /* section alignment (power of 2) */ |
| 454 | uint32_t reloff; /* file offset of relocation entries */ |
| 455 | uint32_t nreloc; /* number of relocation entries */ |
| 456 | uint32_t flags; /* flags (section type and attributes)*/ |
| 457 | uint32_t reserved1; /* reserved (for offset or index) */ |
| 458 | uint32_t reserved2; /* reserved (for count or sizeof) */ |
| 459 | uint32_t reserved3; /* reserved */ |
| 460 | }; |
| 461 | |
| 462 | /* |
| 463 | * The flags field of a section structure is separated into two parts a section |
| 464 | * type and section attributes. The section types are mutually exclusive (it |
| 465 | * can only have one type) but the section attributes are not (it may have more |
| 466 | * than one attribute). |
| 467 | */ |
| 468 | #define SECTION_TYPE 0x000000ff /* 256 section types */ |
| 469 | #define SECTION_ATTRIBUTES 0xffffff00 /* 24 section attributes */ |
| 470 | |
| 471 | /* Constants for the type of a section */ |
| 472 | #define S_REGULAR 0x0 /* regular section */ |
| 473 | #define S_ZEROFILL 0x1 /* zero fill on demand section */ |
| 474 | #define S_CSTRING_LITERALS 0x2 /* section with only literal C strings*/ |
| 475 | #define S_4BYTE_LITERALS 0x3 /* section with only 4 byte literals */ |
| 476 | #define S_8BYTE_LITERALS 0x4 /* section with only 8 byte literals */ |
| 477 | #define S_LITERAL_POINTERS 0x5 /* section with only pointers to */ |
| 478 | /* literals */ |
| 479 | /* |
| 480 | * For the two types of symbol pointers sections and the symbol stubs section |
| 481 | * they have indirect symbol table entries. For each of the entries in the |
| 482 | * section the indirect symbol table entries, in corresponding order in the |
| 483 | * indirect symbol table, start at the index stored in the reserved1 field |
| 484 | * of the section structure. Since the indirect symbol table entries |
| 485 | * correspond to the entries in the section the number of indirect symbol table |
| 486 | * entries is inferred from the size of the section divided by the size of the |
| 487 | * entries in the section. For symbol pointers sections the size of the entries |
| 488 | * in the section is 4 bytes and for symbol stubs sections the byte size of the |
| 489 | * stubs is stored in the reserved2 field of the section structure. |
| 490 | */ |
| 491 | #define S_NON_LAZY_SYMBOL_POINTERS 0x6 /* section with only non-lazy |
| 492 | symbol pointers */ |
| 493 | #define S_LAZY_SYMBOL_POINTERS 0x7 /* section with only lazy symbol |
| 494 | pointers */ |
| 495 | #define S_SYMBOL_STUBS 0x8 /* section with only symbol |
| 496 | stubs, byte size of stub in |
| 497 | the reserved2 field */ |
| 498 | #define S_MOD_INIT_FUNC_POINTERS 0x9 /* section with only function |
| 499 | pointers for initialization*/ |
| 500 | #define S_MOD_TERM_FUNC_POINTERS 0xa /* section with only function |
| 501 | pointers for termination */ |
| 502 | #define S_COALESCED 0xb /* section contains symbols that |
| 503 | are to be coalesced */ |
| 504 | #define S_GB_ZEROFILL 0xc /* zero fill on demand section |
| 505 | (that can be larger than 4 |
| 506 | gigabytes) */ |
| 507 | #define S_INTERPOSING 0xd /* section with only pairs of |
| 508 | function pointers for |
| 509 | interposing */ |
| 510 | #define S_16BYTE_LITERALS 0xe /* section with only 16 byte |
| 511 | literals */ |
| 512 | #define S_DTRACE_DOF 0xf /* section contains |
| 513 | DTrace Object Format */ |
| 514 | #define S_LAZY_DYLIB_SYMBOL_POINTERS 0x10 /* section with only lazy |
| 515 | symbol pointers to lazy |
| 516 | loaded dylibs */ |
| 517 | /* |
| 518 | * Section types to support thread local variables |
| 519 | */ |
| 520 | #define S_THREAD_LOCAL_REGULAR 0x11 /* template of initial |
| 521 | values for TLVs */ |
| 522 | #define S_THREAD_LOCAL_ZEROFILL 0x12 /* template of initial |
| 523 | values for TLVs */ |
| 524 | #define S_THREAD_LOCAL_VARIABLES 0x13 /* TLV descriptors */ |
| 525 | #define S_THREAD_LOCAL_VARIABLE_POINTERS 0x14 /* pointers to TLV |
| 526 | descriptors */ |
| 527 | #define S_THREAD_LOCAL_INIT_FUNCTION_POINTERS 0x15 /* functions to call |
| 528 | to initialize TLV |
| 529 | values */ |
| 530 | #define S_INIT_FUNC_OFFSETS 0x16 /* 32-bit offsets to |
| 531 | initializers */ |
| 532 | |
| 533 | /* |
| 534 | * Constants for the section attributes part of the flags field of a section |
| 535 | * structure. |
| 536 | */ |
| 537 | #define SECTION_ATTRIBUTES_USR 0xff000000 /* User setable attributes */ |
| 538 | #define S_ATTR_PURE_INSTRUCTIONS 0x80000000 /* section contains only true |
| 539 | machine instructions */ |
| 540 | #define S_ATTR_NO_TOC 0x40000000 /* section contains coalesced |
| 541 | symbols that are not to be |
| 542 | in a ranlib table of |
| 543 | contents */ |
| 544 | #define S_ATTR_STRIP_STATIC_SYMS 0x20000000 /* ok to strip static symbols |
| 545 | in this section in files |
| 546 | with the MH_DYLDLINK flag */ |
| 547 | #define S_ATTR_NO_DEAD_STRIP 0x10000000 /* no dead stripping */ |
| 548 | #define S_ATTR_LIVE_SUPPORT 0x08000000 /* blocks are live if they |
| 549 | reference live blocks */ |
| 550 | #define S_ATTR_SELF_MODIFYING_CODE 0x04000000 /* Used with i386 code stubs |
| 551 | written on by dyld */ |
| 552 | /* |
| 553 | * If a segment contains any sections marked with S_ATTR_DEBUG then all |
| 554 | * sections in that segment must have this attribute. No section other than |
| 555 | * a section marked with this attribute may reference the contents of this |
| 556 | * section. A section with this attribute may contain no symbols and must have |
| 557 | * a section type S_REGULAR. The static linker will not copy section contents |
| 558 | * from sections with this attribute into its output file. These sections |
| 559 | * generally contain DWARF debugging info. |
| 560 | */ |
| 561 | #define S_ATTR_DEBUG 0x02000000 /* a debug section */ |
| 562 | #define SECTION_ATTRIBUTES_SYS 0x00ffff00 /* system setable attributes */ |
| 563 | #define S_ATTR_SOME_INSTRUCTIONS 0x00000400 /* section contains some |
| 564 | machine instructions */ |
| 565 | #define S_ATTR_EXT_RELOC 0x00000200 /* section has external |
| 566 | relocation entries */ |
| 567 | #define S_ATTR_LOC_RELOC 0x00000100 /* section has local |
| 568 | relocation entries */ |
| 569 | |
| 570 | |
| 571 | /* |
| 572 | * The names of segments and sections in them are mostly meaningless to the |
| 573 | * link-editor. But there are few things to support traditional UNIX |
| 574 | * executables that require the link-editor and assembler to use some names |
| 575 | * agreed upon by convention. |
| 576 | * |
| 577 | * The initial protection of the "__TEXT" segment has write protection turned |
| 578 | * off (not writeable). |
| 579 | * |
| 580 | * The link-editor will allocate common symbols at the end of the "__common" |
| 581 | * section in the "__DATA" segment. It will create the section and segment |
| 582 | * if needed. |
| 583 | */ |
| 584 | |
| 585 | /* The currently known segment names and the section names in those segments */ |
| 586 | |
| 587 | #define SEG_PAGEZERO "__PAGEZERO" /* the pagezero segment which has no */ |
| 588 | /* protections and catches NULL */ |
| 589 | /* references for MH_EXECUTE files */ |
| 590 | |
| 591 | |
| 592 | #define SEG_TEXT "__TEXT" /* the tradition UNIX text segment */ |
| 593 | #define SECT_TEXT "__text" /* the real text part of the text */ |
| 594 | /* section no headers, and no padding */ |
| 595 | #define SECT_FVMLIB_INIT0 "__fvmlib_init0" /* the fvmlib initialization */ |
| 596 | /* section */ |
| 597 | #define SECT_FVMLIB_INIT1 "__fvmlib_init1" /* the section following the */ |
| 598 | /* fvmlib initialization */ |
| 599 | /* section */ |
| 600 | |
| 601 | #define SEG_DATA "__DATA" /* the tradition UNIX data segment */ |
| 602 | #define SECT_DATA "__data" /* the real initialized data section */ |
| 603 | /* no padding, no bss overlap */ |
| 604 | #define SECT_BSS "__bss" /* the real uninitialized data section*/ |
| 605 | /* no padding */ |
| 606 | #define SECT_COMMON "__common" /* the section common symbols are */ |
| 607 | /* allocated in by the link editor */ |
| 608 | |
| 609 | #define SEG_OBJC "__OBJC" /* objective-C runtime segment */ |
| 610 | #define SECT_OBJC_SYMBOLS "__symbol_table" /* symbol table */ |
| 611 | #define SECT_OBJC_MODULES "__module_info" /* module information */ |
| 612 | #define SECT_OBJC_STRINGS "__selector_strs" /* string table */ |
| 613 | #define SECT_OBJC_REFS "__selector_refs" /* string table */ |
| 614 | |
| 615 | #define SEG_ICON "__ICON" /* the icon segment */ |
| 616 | #define SECT_ICON_HEADER "__header" /* the icon headers */ |
| 617 | #define SECT_ICON_TIFF "__tiff" /* the icons in tiff format */ |
| 618 | |
| 619 | #define SEG_LINKEDIT "__LINKEDIT" /* the segment containing all structs */ |
| 620 | /* created and maintained by the link */ |
| 621 | /* editor. Created with -seglinkedit */ |
| 622 | /* option to ld(1) for MH_EXECUTE and */ |
| 623 | /* FVMLIB file types only */ |
| 624 | |
| 625 | #define SEG_UNIXSTACK "__UNIXSTACK" /* the unix stack segment */ |
| 626 | |
| 627 | #define SEG_IMPORT "__IMPORT" /* the segment for the self (dyld) */ |
| 628 | /* modifing code stubs that has read, */ |
| 629 | /* write and execute permissions */ |
| 630 | |
| 631 | /* |
| 632 | * Fixed virtual memory shared libraries are identified by two things. The |
| 633 | * target pathname (the name of the library as found for execution), and the |
| 634 | * minor version number. The address of where the headers are loaded is in |
| 635 | * header_addr. (THIS IS OBSOLETE and no longer supported). |
| 636 | */ |
| 637 | struct fvmlib { |
| 638 | union lc_str name; /* library's target pathname */ |
| 639 | uint32_t minor_version; /* library's minor version number */ |
| 640 | uint32_t header_addr; /* library's header address */ |
| 641 | }; |
| 642 | |
| 643 | /* |
| 644 | * A fixed virtual shared library (filetype == MH_FVMLIB in the mach header) |
| 645 | * contains a fvmlib_command (cmd == LC_IDFVMLIB) to identify the library. |
| 646 | * An object that uses a fixed virtual shared library also contains a |
| 647 | * fvmlib_command (cmd == LC_LOADFVMLIB) for each library it uses. |
| 648 | * (THIS IS OBSOLETE and no longer supported). |
| 649 | */ |
| 650 | struct fvmlib_command { |
| 651 | uint32_t cmd; /* LC_IDFVMLIB or LC_LOADFVMLIB */ |
| 652 | uint32_t cmdsize; /* includes pathname string */ |
| 653 | struct fvmlib fvmlib; /* the library identification */ |
| 654 | }; |
| 655 | |
| 656 | /* |
| 657 | * Dynamicly linked shared libraries are identified by two things. The |
| 658 | * pathname (the name of the library as found for execution), and the |
| 659 | * compatibility version number. The pathname must match and the compatibility |
| 660 | * number in the user of the library must be greater than or equal to the |
| 661 | * library being used. The time stamp is used to record the time a library was |
| 662 | * built and copied into user so it can be use to determined if the library used |
| 663 | * at runtime is exactly the same as used to built the program. |
| 664 | */ |
| 665 | struct dylib { |
| 666 | union lc_str name; /* library's path name */ |
| 667 | uint32_t timestamp; /* library's build time stamp */ |
| 668 | uint32_t current_version; /* library's current version number */ |
| 669 | uint32_t compatibility_version; /* library's compatibility vers number*/ |
| 670 | }; |
| 671 | |
| 672 | /* |
| 673 | * A dynamically linked shared library (filetype == MH_DYLIB in the mach header) |
| 674 | * contains a dylib_command (cmd == LC_ID_DYLIB) to identify the library. |
| 675 | * An object that uses a dynamically linked shared library also contains a |
| 676 | * dylib_command (cmd == LC_LOAD_DYLIB, LC_LOAD_WEAK_DYLIB, or |
| 677 | * LC_REEXPORT_DYLIB) for each library it uses. |
| 678 | */ |
| 679 | struct dylib_command { |
| 680 | uint32_t cmd; /* LC_ID_DYLIB, LC_LOAD_{,WEAK_}DYLIB, |
| 681 | LC_REEXPORT_DYLIB */ |
| 682 | uint32_t cmdsize; /* includes pathname string */ |
| 683 | struct dylib dylib; /* the library identification */ |
| 684 | }; |
| 685 | |
| 686 | /* |
| 687 | * A dynamically linked shared library may be a subframework of an umbrella |
| 688 | * framework. If so it will be linked with "-umbrella umbrella_name" where |
| 689 | * Where "umbrella_name" is the name of the umbrella framework. A subframework |
| 690 | * can only be linked against by its umbrella framework or other subframeworks |
| 691 | * that are part of the same umbrella framework. Otherwise the static link |
| 692 | * editor produces an error and states to link against the umbrella framework. |
| 693 | * The name of the umbrella framework for subframeworks is recorded in the |
| 694 | * following structure. |
| 695 | */ |
| 696 | struct sub_framework_command { |
| 697 | uint32_t cmd; /* LC_SUB_FRAMEWORK */ |
| 698 | uint32_t cmdsize; /* includes umbrella string */ |
| 699 | union lc_str umbrella; /* the umbrella framework name */ |
| 700 | }; |
| 701 | |
| 702 | /* |
| 703 | * For dynamically linked shared libraries that are subframework of an umbrella |
| 704 | * framework they can allow clients other than the umbrella framework or other |
| 705 | * subframeworks in the same umbrella framework. To do this the subframework |
| 706 | * is built with "-allowable_client client_name" and an LC_SUB_CLIENT load |
| 707 | * command is created for each -allowable_client flag. The client_name is |
| 708 | * usually a framework name. It can also be a name used for bundles clients |
| 709 | * where the bundle is built with "-client_name client_name". |
| 710 | */ |
| 711 | struct sub_client_command { |
| 712 | uint32_t cmd; /* LC_SUB_CLIENT */ |
| 713 | uint32_t cmdsize; /* includes client string */ |
| 714 | union lc_str client; /* the client name */ |
| 715 | }; |
| 716 | |
| 717 | /* |
| 718 | * A dynamically linked shared library may be a sub_umbrella of an umbrella |
| 719 | * framework. If so it will be linked with "-sub_umbrella umbrella_name" where |
| 720 | * Where "umbrella_name" is the name of the sub_umbrella framework. When |
| 721 | * staticly linking when -twolevel_namespace is in effect a twolevel namespace |
| 722 | * umbrella framework will only cause its subframeworks and those frameworks |
| 723 | * listed as sub_umbrella frameworks to be implicited linked in. Any other |
| 724 | * dependent dynamic libraries will not be linked it when -twolevel_namespace |
| 725 | * is in effect. The primary library recorded by the static linker when |
| 726 | * resolving a symbol in these libraries will be the umbrella framework. |
| 727 | * Zero or more sub_umbrella frameworks may be use by an umbrella framework. |
| 728 | * The name of a sub_umbrella framework is recorded in the following structure. |
| 729 | */ |
| 730 | struct sub_umbrella_command { |
| 731 | uint32_t cmd; /* LC_SUB_UMBRELLA */ |
| 732 | uint32_t cmdsize; /* includes sub_umbrella string */ |
| 733 | union lc_str sub_umbrella; /* the sub_umbrella framework name */ |
| 734 | }; |
| 735 | |
| 736 | /* |
| 737 | * A dynamically linked shared library may be a sub_library of another shared |
| 738 | * library. If so it will be linked with "-sub_library library_name" where |
| 739 | * Where "library_name" is the name of the sub_library shared library. When |
| 740 | * staticly linking when -twolevel_namespace is in effect a twolevel namespace |
| 741 | * shared library will only cause its subframeworks and those frameworks |
| 742 | * listed as sub_umbrella frameworks and libraries listed as sub_libraries to |
| 743 | * be implicited linked in. Any other dependent dynamic libraries will not be |
| 744 | * linked it when -twolevel_namespace is in effect. The primary library |
| 745 | * recorded by the static linker when resolving a symbol in these libraries |
| 746 | * will be the umbrella framework (or dynamic library). Zero or more sub_library |
| 747 | * shared libraries may be use by an umbrella framework or (or dynamic library). |
| 748 | * The name of a sub_library framework is recorded in the following structure. |
| 749 | * For example /usr/lib/libobjc_profile.A.dylib would be recorded as "libobjc". |
| 750 | */ |
| 751 | struct sub_library_command { |
| 752 | uint32_t cmd; /* LC_SUB_LIBRARY */ |
| 753 | uint32_t cmdsize; /* includes sub_library string */ |
| 754 | union lc_str sub_library; /* the sub_library name */ |
| 755 | }; |
| 756 | |
| 757 | /* |
| 758 | * A program (filetype == MH_EXECUTE) that is |
| 759 | * prebound to its dynamic libraries has one of these for each library that |
| 760 | * the static linker used in prebinding. It contains a bit vector for the |
| 761 | * modules in the library. The bits indicate which modules are bound (1) and |
| 762 | * which are not (0) from the library. The bit for module 0 is the low bit |
| 763 | * of the first byte. So the bit for the Nth module is: |
| 764 | * (linked_modules[N/8] >> N%8) & 1 |
| 765 | */ |
| 766 | struct prebound_dylib_command { |
| 767 | uint32_t cmd; /* LC_PREBOUND_DYLIB */ |
| 768 | uint32_t cmdsize; /* includes strings */ |
| 769 | union lc_str name; /* library's path name */ |
| 770 | uint32_t nmodules; /* number of modules in library */ |
| 771 | union lc_str linked_modules; /* bit vector of linked modules */ |
| 772 | }; |
| 773 | |
| 774 | /* |
| 775 | * A program that uses a dynamic linker contains a dylinker_command to identify |
| 776 | * the name of the dynamic linker (LC_LOAD_DYLINKER). And a dynamic linker |
| 777 | * contains a dylinker_command to identify the dynamic linker (LC_ID_DYLINKER). |
| 778 | * A file can have at most one of these. |
| 779 | * This struct is also used for the LC_DYLD_ENVIRONMENT load command and |
| 780 | * contains string for dyld to treat like environment variable. |
| 781 | */ |
| 782 | struct dylinker_command { |
| 783 | uint32_t cmd; /* LC_ID_DYLINKER, LC_LOAD_DYLINKER or |
| 784 | LC_DYLD_ENVIRONMENT */ |
| 785 | uint32_t cmdsize; /* includes pathname string */ |
| 786 | union lc_str name; /* dynamic linker's path name */ |
| 787 | }; |
| 788 | |
| 789 | /* |
| 790 | * Thread commands contain machine-specific data structures suitable for |
| 791 | * use in the thread state primitives. The machine specific data structures |
| 792 | * follow the struct thread_command as follows. |
| 793 | * Each flavor of machine specific data structure is preceded by an uint32_t |
| 794 | * constant for the flavor of that data structure, an uint32_t that is the |
| 795 | * count of uint32_t's of the size of the state data structure and then |
| 796 | * the state data structure follows. This triple may be repeated for many |
| 797 | * flavors. The constants for the flavors, counts and state data structure |
| 798 | * definitions are expected to be in the header file <machine/thread_status.h>. |
| 799 | * These machine specific data structures sizes must be multiples of |
| 800 | * 4 bytes. The cmdsize reflects the total size of the thread_command |
| 801 | * and all of the sizes of the constants for the flavors, counts and state |
| 802 | * data structures. |
| 803 | * |
| 804 | * For executable objects that are unix processes there will be one |
| 805 | * thread_command (cmd == LC_UNIXTHREAD) created for it by the link-editor. |
| 806 | * This is the same as a LC_THREAD, except that a stack is automatically |
| 807 | * created (based on the shell's limit for the stack size). Command arguments |
| 808 | * and environment variables are copied onto that stack. |
| 809 | */ |
| 810 | struct thread_command { |
| 811 | uint32_t cmd; /* LC_THREAD or LC_UNIXTHREAD */ |
| 812 | uint32_t cmdsize; /* total size of this command */ |
| 813 | /* uint32_t flavor flavor of thread state */ |
| 814 | /* uint32_t count count of uint32_t's in thread state */ |
| 815 | /* struct XXX_thread_state state thread state for this flavor */ |
| 816 | /* ... */ |
| 817 | }; |
| 818 | |
| 819 | /* |
| 820 | * The routines command contains the address of the dynamic shared library |
| 821 | * initialization routine and an index into the module table for the module |
| 822 | * that defines the routine. Before any modules are used from the library the |
| 823 | * dynamic linker fully binds the module that defines the initialization routine |
| 824 | * and then calls it. This gets called before any module initialization |
| 825 | * routines (used for C++ static constructors) in the library. |
| 826 | */ |
| 827 | struct routines_command { /* for 32-bit architectures */ |
| 828 | uint32_t cmd; /* LC_ROUTINES */ |
| 829 | uint32_t cmdsize; /* total size of this command */ |
| 830 | uint32_t init_address; /* address of initialization routine */ |
| 831 | uint32_t init_module; /* index into the module table that */ |
| 832 | /* the init routine is defined in */ |
| 833 | uint32_t reserved1; |
| 834 | uint32_t reserved2; |
| 835 | uint32_t reserved3; |
| 836 | uint32_t reserved4; |
| 837 | uint32_t reserved5; |
| 838 | uint32_t reserved6; |
| 839 | }; |
| 840 | |
| 841 | /* |
| 842 | * The 64-bit routines command. Same use as above. |
| 843 | */ |
| 844 | struct routines_command_64 { /* for 64-bit architectures */ |
| 845 | uint32_t cmd; /* LC_ROUTINES_64 */ |
| 846 | uint32_t cmdsize; /* total size of this command */ |
| 847 | uint64_t init_address; /* address of initialization routine */ |
| 848 | uint64_t init_module; /* index into the module table that */ |
| 849 | /* the init routine is defined in */ |
| 850 | uint64_t reserved1; |
| 851 | uint64_t reserved2; |
| 852 | uint64_t reserved3; |
| 853 | uint64_t reserved4; |
| 854 | uint64_t reserved5; |
| 855 | uint64_t reserved6; |
| 856 | }; |
| 857 | |
| 858 | /* |
| 859 | * The symtab_command contains the offsets and sizes of the link-edit 4.3BSD |
| 860 | * "stab" style symbol table information as described in the header files |
| 861 | * <nlist.h> and <stab.h>. |
| 862 | */ |
| 863 | struct symtab_command { |
| 864 | uint32_t cmd; /* LC_SYMTAB */ |
| 865 | uint32_t cmdsize; /* sizeof(struct symtab_command) */ |
| 866 | uint32_t symoff; /* symbol table offset */ |
| 867 | uint32_t nsyms; /* number of symbol table entries */ |
| 868 | uint32_t stroff; /* string table offset */ |
| 869 | uint32_t strsize; /* string table size in bytes */ |
| 870 | }; |
| 871 | |
| 872 | /* |
| 873 | * This is the second set of the symbolic information which is used to support |
| 874 | * the data structures for the dynamically link editor. |
| 875 | * |
| 876 | * The original set of symbolic information in the symtab_command which contains |
| 877 | * the symbol and string tables must also be present when this load command is |
| 878 | * present. When this load command is present the symbol table is organized |
| 879 | * into three groups of symbols: |
| 880 | * local symbols (static and debugging symbols) - grouped by module |
| 881 | * defined external symbols - grouped by module (sorted by name if not lib) |
| 882 | * undefined external symbols (sorted by name if MH_BINDATLOAD is not set, |
| 883 | * and in order the were seen by the static |
| 884 | * linker if MH_BINDATLOAD is set) |
| 885 | * In this load command there are offsets and counts to each of the three groups |
| 886 | * of symbols. |
| 887 | * |
| 888 | * This load command contains a the offsets and sizes of the following new |
| 889 | * symbolic information tables: |
| 890 | * table of contents |
| 891 | * module table |
| 892 | * reference symbol table |
| 893 | * indirect symbol table |
| 894 | * The first three tables above (the table of contents, module table and |
| 895 | * reference symbol table) are only present if the file is a dynamically linked |
| 896 | * shared library. For executable and object modules, which are files |
| 897 | * containing only one module, the information that would be in these three |
| 898 | * tables is determined as follows: |
| 899 | * table of contents - the defined external symbols are sorted by name |
| 900 | * module table - the file contains only one module so everything in the |
| 901 | * file is part of the module. |
| 902 | * reference symbol table - is the defined and undefined external symbols |
| 903 | * |
| 904 | * For dynamically linked shared library files this load command also contains |
| 905 | * offsets and sizes to the pool of relocation entries for all sections |
| 906 | * separated into two groups: |
| 907 | * external relocation entries |
| 908 | * local relocation entries |
| 909 | * For executable and object modules the relocation entries continue to hang |
| 910 | * off the section structures. |
| 911 | */ |
| 912 | struct dysymtab_command { |
| 913 | uint32_t cmd; /* LC_DYSYMTAB */ |
| 914 | uint32_t cmdsize; /* sizeof(struct dysymtab_command) */ |
| 915 | |
| 916 | /* |
| 917 | * The symbols indicated by symoff and nsyms of the LC_SYMTAB load command |
| 918 | * are grouped into the following three groups: |
| 919 | * local symbols (further grouped by the module they are from) |
| 920 | * defined external symbols (further grouped by the module they are from) |
| 921 | * undefined symbols |
| 922 | * |
| 923 | * The local symbols are used only for debugging. The dynamic binding |
| 924 | * process may have to use them to indicate to the debugger the local |
| 925 | * symbols for a module that is being bound. |
| 926 | * |
| 927 | * The last two groups are used by the dynamic binding process to do the |
| 928 | * binding (indirectly through the module table and the reference symbol |
| 929 | * table when this is a dynamically linked shared library file). |
| 930 | */ |
| 931 | uint32_t ilocalsym; /* index to local symbols */ |
| 932 | uint32_t nlocalsym; /* number of local symbols */ |
| 933 | |
| 934 | uint32_t iextdefsym;/* index to externally defined symbols */ |
| 935 | uint32_t nextdefsym;/* number of externally defined symbols */ |
| 936 | |
| 937 | uint32_t iundefsym; /* index to undefined symbols */ |
| 938 | uint32_t nundefsym; /* number of undefined symbols */ |
| 939 | |
| 940 | /* |
| 941 | * For the for the dynamic binding process to find which module a symbol |
| 942 | * is defined in the table of contents is used (analogous to the ranlib |
| 943 | * structure in an archive) which maps defined external symbols to modules |
| 944 | * they are defined in. This exists only in a dynamically linked shared |
| 945 | * library file. For executable and object modules the defined external |
| 946 | * symbols are sorted by name and is use as the table of contents. |
| 947 | */ |
| 948 | uint32_t tocoff; /* file offset to table of contents */ |
| 949 | uint32_t ntoc; /* number of entries in table of contents */ |
| 950 | |
| 951 | /* |
| 952 | * To support dynamic binding of "modules" (whole object files) the symbol |
| 953 | * table must reflect the modules that the file was created from. This is |
| 954 | * done by having a module table that has indexes and counts into the merged |
| 955 | * tables for each module. The module structure that these two entries |
| 956 | * refer to is described below. This exists only in a dynamically linked |
| 957 | * shared library file. For executable and object modules the file only |
| 958 | * contains one module so everything in the file belongs to the module. |
| 959 | */ |
| 960 | uint32_t modtaboff; /* file offset to module table */ |
| 961 | uint32_t nmodtab; /* number of module table entries */ |
| 962 | |
| 963 | /* |
| 964 | * To support dynamic module binding the module structure for each module |
| 965 | * indicates the external references (defined and undefined) each module |
| 966 | * makes. For each module there is an offset and a count into the |
| 967 | * reference symbol table for the symbols that the module references. |
| 968 | * This exists only in a dynamically linked shared library file. For |
| 969 | * executable and object modules the defined external symbols and the |
| 970 | * undefined external symbols indicates the external references. |
| 971 | */ |
| 972 | uint32_t extrefsymoff; /* offset to referenced symbol table */ |
| 973 | uint32_t nextrefsyms; /* number of referenced symbol table entries */ |
| 974 | |
| 975 | /* |
| 976 | * The sections that contain "symbol pointers" and "routine stubs" have |
| 977 | * indexes and (implied counts based on the size of the section and fixed |
| 978 | * size of the entry) into the "indirect symbol" table for each pointer |
| 979 | * and stub. For every section of these two types the index into the |
| 980 | * indirect symbol table is stored in the section header in the field |
| 981 | * reserved1. An indirect symbol table entry is simply a 32bit index into |
| 982 | * the symbol table to the symbol that the pointer or stub is referring to. |
| 983 | * The indirect symbol table is ordered to match the entries in the section. |
| 984 | */ |
| 985 | uint32_t indirectsymoff; /* file offset to the indirect symbol table */ |
| 986 | uint32_t nindirectsyms; /* number of indirect symbol table entries */ |
| 987 | |
| 988 | /* |
| 989 | * To support relocating an individual module in a library file quickly the |
| 990 | * external relocation entries for each module in the library need to be |
| 991 | * accessed efficiently. Since the relocation entries can't be accessed |
| 992 | * through the section headers for a library file they are separated into |
| 993 | * groups of local and external entries further grouped by module. In this |
| 994 | * case the presents of this load command who's extreloff, nextrel, |
| 995 | * locreloff and nlocrel fields are non-zero indicates that the relocation |
| 996 | * entries of non-merged sections are not referenced through the section |
| 997 | * structures (and the reloff and nreloc fields in the section headers are |
| 998 | * set to zero). |
| 999 | * |
| 1000 | * Since the relocation entries are not accessed through the section headers |
| 1001 | * this requires the r_address field to be something other than a section |
| 1002 | * offset to identify the item to be relocated. In this case r_address is |
| 1003 | * set to the offset from the vmaddr of the first LC_SEGMENT command. |
| 1004 | * For MH_SPLIT_SEGS images r_address is set to the the offset from the |
| 1005 | * vmaddr of the first read-write LC_SEGMENT command. |
| 1006 | * |
| 1007 | * The relocation entries are grouped by module and the module table |
| 1008 | * entries have indexes and counts into them for the group of external |
| 1009 | * relocation entries for that the module. |
| 1010 | * |
| 1011 | * For sections that are merged across modules there must not be any |
| 1012 | * remaining external relocation entries for them (for merged sections |
| 1013 | * remaining relocation entries must be local). |
| 1014 | */ |
| 1015 | uint32_t extreloff; /* offset to external relocation entries */ |
| 1016 | uint32_t nextrel; /* number of external relocation entries */ |
| 1017 | |
| 1018 | /* |
| 1019 | * All the local relocation entries are grouped together (they are not |
| 1020 | * grouped by their module since they are only used if the object is moved |
| 1021 | * from it staticly link edited address). |
| 1022 | */ |
| 1023 | uint32_t locreloff; /* offset to local relocation entries */ |
| 1024 | uint32_t nlocrel; /* number of local relocation entries */ |
| 1025 | |
| 1026 | }; |
| 1027 | |
| 1028 | /* |
| 1029 | * An indirect symbol table entry is simply a 32bit index into the symbol table |
| 1030 | * to the symbol that the pointer or stub is refering to. Unless it is for a |
| 1031 | * non-lazy symbol pointer section for a defined symbol which strip(1) as |
| 1032 | * removed. In which case it has the value INDIRECT_SYMBOL_LOCAL. If the |
| 1033 | * symbol was also absolute INDIRECT_SYMBOL_ABS is or'ed with that. |
| 1034 | */ |
| 1035 | #define INDIRECT_SYMBOL_LOCAL 0x80000000 |
| 1036 | #define INDIRECT_SYMBOL_ABS 0x40000000 |
| 1037 | |
| 1038 | |
| 1039 | /* a table of contents entry */ |
| 1040 | struct dylib_table_of_contents { |
| 1041 | uint32_t symbol_index; /* the defined external symbol |
| 1042 | (index into the symbol table) */ |
| 1043 | uint32_t module_index; /* index into the module table this symbol |
| 1044 | is defined in */ |
| 1045 | }; |
| 1046 | |
| 1047 | /* a module table entry */ |
| 1048 | struct dylib_module { |
| 1049 | uint32_t module_name; /* the module name (index into string table) */ |
| 1050 | |
| 1051 | uint32_t iextdefsym; /* index into externally defined symbols */ |
| 1052 | uint32_t nextdefsym; /* number of externally defined symbols */ |
| 1053 | uint32_t irefsym; /* index into reference symbol table */ |
| 1054 | uint32_t nrefsym; /* number of reference symbol table entries */ |
| 1055 | uint32_t ilocalsym; /* index into symbols for local symbols */ |
| 1056 | uint32_t nlocalsym; /* number of local symbols */ |
| 1057 | |
| 1058 | uint32_t iextrel; /* index into external relocation entries */ |
| 1059 | uint32_t nextrel; /* number of external relocation entries */ |
| 1060 | |
| 1061 | uint32_t iinit_iterm; /* low 16 bits are the index into the init |
| 1062 | section, high 16 bits are the index into |
| 1063 | the term section */ |
| 1064 | uint32_t ninit_nterm; /* low 16 bits are the number of init section |
| 1065 | entries, high 16 bits are the number of |
| 1066 | term section entries */ |
| 1067 | |
| 1068 | uint32_t /* for this module address of the start of */ |
| 1069 | objc_module_info_addr; /* the (__OBJC,__module_info) section */ |
| 1070 | uint32_t /* for this module size of */ |
| 1071 | objc_module_info_size; /* the (__OBJC,__module_info) section */ |
| 1072 | }; |
| 1073 | |
| 1074 | /* a 64-bit module table entry */ |
| 1075 | struct dylib_module_64 { |
| 1076 | uint32_t module_name; /* the module name (index into string table) */ |
| 1077 | |
| 1078 | uint32_t iextdefsym; /* index into externally defined symbols */ |
| 1079 | uint32_t nextdefsym; /* number of externally defined symbols */ |
| 1080 | uint32_t irefsym; /* index into reference symbol table */ |
| 1081 | uint32_t nrefsym; /* number of reference symbol table entries */ |
| 1082 | uint32_t ilocalsym; /* index into symbols for local symbols */ |
| 1083 | uint32_t nlocalsym; /* number of local symbols */ |
| 1084 | |
| 1085 | uint32_t iextrel; /* index into external relocation entries */ |
| 1086 | uint32_t nextrel; /* number of external relocation entries */ |
| 1087 | |
| 1088 | uint32_t iinit_iterm; /* low 16 bits are the index into the init |
| 1089 | section, high 16 bits are the index into |
| 1090 | the term section */ |
| 1091 | uint32_t ninit_nterm; /* low 16 bits are the number of init section |
| 1092 | entries, high 16 bits are the number of |
| 1093 | term section entries */ |
| 1094 | |
| 1095 | uint32_t /* for this module size of */ |
| 1096 | objc_module_info_size; /* the (__OBJC,__module_info) section */ |
| 1097 | uint64_t /* for this module address of the start of */ |
| 1098 | objc_module_info_addr; /* the (__OBJC,__module_info) section */ |
| 1099 | }; |
| 1100 | |
| 1101 | /* |
| 1102 | * The entries in the reference symbol table are used when loading the module |
| 1103 | * (both by the static and dynamic link editors) and if the module is unloaded |
| 1104 | * or replaced. Therefore all external symbols (defined and undefined) are |
| 1105 | * listed in the module's reference table. The flags describe the type of |
| 1106 | * reference that is being made. The constants for the flags are defined in |
| 1107 | * <mach-o/nlist.h> as they are also used for symbol table entries. |
| 1108 | */ |
| 1109 | struct dylib_reference { |
| 1110 | uint32_t isym:24, /* index into the symbol table */ |
| 1111 | flags:8; /* flags to indicate the type of reference */ |
| 1112 | }; |
| 1113 | |
| 1114 | /* |
| 1115 | * The twolevel_hints_command contains the offset and number of hints in the |
| 1116 | * two-level namespace lookup hints table. |
| 1117 | */ |
| 1118 | struct twolevel_hints_command { |
| 1119 | uint32_t cmd; /* LC_TWOLEVEL_HINTS */ |
| 1120 | uint32_t cmdsize; /* sizeof(struct twolevel_hints_command) */ |
| 1121 | uint32_t offset; /* offset to the hint table */ |
| 1122 | uint32_t nhints; /* number of hints in the hint table */ |
| 1123 | }; |
| 1124 | |
| 1125 | /* |
| 1126 | * The entries in the two-level namespace lookup hints table are twolevel_hint |
| 1127 | * structs. These provide hints to the dynamic link editor where to start |
| 1128 | * looking for an undefined symbol in a two-level namespace image. The |
| 1129 | * isub_image field is an index into the sub-images (sub-frameworks and |
| 1130 | * sub-umbrellas list) that made up the two-level image that the undefined |
| 1131 | * symbol was found in when it was built by the static link editor. If |
| 1132 | * isub-image is 0 the the symbol is expected to be defined in library and not |
| 1133 | * in the sub-images. If isub-image is non-zero it is an index into the array |
| 1134 | * of sub-images for the umbrella with the first index in the sub-images being |
| 1135 | * 1. The array of sub-images is the ordered list of sub-images of the umbrella |
| 1136 | * that would be searched for a symbol that has the umbrella recorded as its |
| 1137 | * primary library. The table of contents index is an index into the |
| 1138 | * library's table of contents. This is used as the starting point of the |
| 1139 | * binary search or a directed linear search. |
| 1140 | */ |
| 1141 | struct twolevel_hint { |
| 1142 | uint32_t |
| 1143 | isub_image:8, /* index into the sub images */ |
| 1144 | itoc:24; /* index into the table of contents */ |
| 1145 | }; |
| 1146 | |
| 1147 | /* |
| 1148 | * The prebind_cksum_command contains the value of the original check sum for |
| 1149 | * prebound files or zero. When a prebound file is first created or modified |
| 1150 | * for other than updating its prebinding information the value of the check sum |
| 1151 | * is set to zero. When the file has it prebinding re-done and if the value of |
| 1152 | * the check sum is zero the original check sum is calculated and stored in |
| 1153 | * cksum field of this load command in the output file. If when the prebinding |
| 1154 | * is re-done and the cksum field is non-zero it is left unchanged from the |
| 1155 | * input file. |
| 1156 | */ |
| 1157 | struct prebind_cksum_command { |
| 1158 | uint32_t cmd; /* LC_PREBIND_CKSUM */ |
| 1159 | uint32_t cmdsize; /* sizeof(struct prebind_cksum_command) */ |
| 1160 | uint32_t cksum; /* the check sum or zero */ |
| 1161 | }; |
| 1162 | |
| 1163 | /* |
| 1164 | * The uuid load command contains a single 128-bit unique random number that |
| 1165 | * identifies an object produced by the static link editor. |
| 1166 | */ |
| 1167 | struct uuid_command { |
| 1168 | uint32_t cmd; /* LC_UUID */ |
| 1169 | uint32_t cmdsize; /* sizeof(struct uuid_command) */ |
| 1170 | uint8_t uuid[16]; /* the 128-bit uuid */ |
| 1171 | }; |
| 1172 | |
| 1173 | /* |
| 1174 | * The rpath_command contains a path which at runtime should be added to |
| 1175 | * the current run path used to find @rpath prefixed dylibs. |
| 1176 | */ |
| 1177 | struct rpath_command { |
| 1178 | uint32_t cmd; /* LC_RPATH */ |
| 1179 | uint32_t cmdsize; /* includes string */ |
| 1180 | union lc_str path; /* path to add to run path */ |
| 1181 | }; |
| 1182 | |
| 1183 | /* |
| 1184 | * The linkedit_data_command contains the offsets and sizes of a blob |
| 1185 | * of data in the __LINKEDIT segment. |
| 1186 | */ |
| 1187 | struct linkedit_data_command { |
| 1188 | uint32_t cmd; /* LC_CODE_SIGNATURE, LC_SEGMENT_SPLIT_INFO, |
| 1189 | LC_FUNCTION_STARTS, LC_DATA_IN_CODE, |
| 1190 | LC_DYLIB_CODE_SIGN_DRS, |
| 1191 | LC_LINKER_OPTIMIZATION_HINT, |
| 1192 | LC_DYLD_EXPORTS_TRIE, or |
| 1193 | LC_DYLD_CHAINED_FIXUPS. */ |
| 1194 | uint32_t cmdsize; /* sizeof(struct linkedit_data_command) */ |
| 1195 | uint32_t dataoff; /* file offset of data in __LINKEDIT segment */ |
| 1196 | uint32_t datasize; /* file size of data in __LINKEDIT segment */ |
| 1197 | }; |
| 1198 | |
| 1199 | /* |
| 1200 | * The encryption_info_command contains the file offset and size of an |
| 1201 | * of an encrypted segment. |
| 1202 | */ |
| 1203 | struct encryption_info_command { |
| 1204 | uint32_t cmd; /* LC_ENCRYPTION_INFO */ |
| 1205 | uint32_t cmdsize; /* sizeof(struct encryption_info_command) */ |
| 1206 | uint32_t cryptoff; /* file offset of encrypted range */ |
| 1207 | uint32_t cryptsize; /* file size of encrypted range */ |
| 1208 | uint32_t cryptid; /* which enryption system, |
| 1209 | 0 means not-encrypted yet */ |
| 1210 | }; |
| 1211 | |
| 1212 | /* |
| 1213 | * The encryption_info_command_64 contains the file offset and size of an |
| 1214 | * of an encrypted segment (for use in x86_64 targets). |
| 1215 | */ |
| 1216 | struct encryption_info_command_64 { |
| 1217 | uint32_t cmd; /* LC_ENCRYPTION_INFO_64 */ |
| 1218 | uint32_t cmdsize; /* sizeof(struct encryption_info_command_64) */ |
| 1219 | uint32_t cryptoff; /* file offset of encrypted range */ |
| 1220 | uint32_t cryptsize; /* file size of encrypted range */ |
| 1221 | uint32_t cryptid; /* which enryption system, |
| 1222 | 0 means not-encrypted yet */ |
| 1223 | uint32_t pad; /* padding to make this struct's size a multiple |
| 1224 | of 8 bytes */ |
| 1225 | }; |
| 1226 | |
| 1227 | /* |
| 1228 | * The version_min_command contains the min OS version on which this |
| 1229 | * binary was built to run. |
| 1230 | */ |
| 1231 | struct version_min_command { |
| 1232 | uint32_t cmd; /* LC_VERSION_MIN_MACOSX or |
| 1233 | LC_VERSION_MIN_IPHONEOS or |
| 1234 | LC_VERSION_MIN_WATCHOS or |
| 1235 | LC_VERSION_MIN_TVOS */ |
| 1236 | uint32_t cmdsize; /* sizeof(struct min_version_command) */ |
| 1237 | uint32_t version; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ |
| 1238 | uint32_t sdk; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ |
| 1239 | }; |
| 1240 | |
| 1241 | /* |
| 1242 | * The build_version_command contains the min OS version on which this |
| 1243 | * binary was built to run for its platform. The list of known platforms and |
| 1244 | * tool values following it. |
| 1245 | */ |
| 1246 | struct build_version_command { |
| 1247 | uint32_t cmd; /* LC_BUILD_VERSION */ |
| 1248 | uint32_t cmdsize; /* sizeof(struct build_version_command) plus */ |
| 1249 | /* ntools * sizeof(struct build_tool_version) */ |
| 1250 | uint32_t platform; /* platform */ |
| 1251 | uint32_t minos; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ |
| 1252 | uint32_t sdk; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ |
| 1253 | uint32_t ntools; /* number of tool entries following this */ |
| 1254 | }; |
| 1255 | |
| 1256 | struct build_tool_version { |
| 1257 | uint32_t tool; /* enum for the tool */ |
| 1258 | uint32_t version; /* version number of the tool */ |
| 1259 | }; |
| 1260 | |
| 1261 | /* Known values for the platform field above. */ |
| 1262 | #define PLATFORM_MACOS 1 |
| 1263 | #define PLATFORM_IOS 2 |
| 1264 | #define PLATFORM_TVOS 3 |
| 1265 | #define PLATFORM_WATCHOS 4 |
| 1266 | #define PLATFORM_BRIDGEOS 5 |
| 1267 | #define PLATFORM_MACCATALYST 6 |
| 1268 | #if (!defined(PLATFORM_MACCATALYST)) |
| 1269 | #define PLATFORM_MACCATALYST 6 |
| 1270 | #endif |
| 1271 | #define PLATFORM_IOSSIMULATOR 7 |
| 1272 | #define PLATFORM_TVOSSIMULATOR 8 |
| 1273 | #define PLATFORM_WATCHOSSIMULATOR 9 |
| 1274 | #define PLATFORM_DRIVERKIT 10 |
| 1275 | |
| 1276 | /* Known values for the tool field above. */ |
| 1277 | #define TOOL_CLANG 1 |
| 1278 | #define TOOL_SWIFT 2 |
| 1279 | #define TOOL_LD 3 |
| 1280 | |
| 1281 | /* |
| 1282 | * The dyld_info_command contains the file offsets and sizes of |
| 1283 | * the new compressed form of the information dyld needs to |
| 1284 | * load the image. This information is used by dyld on Mac OS X |
| 1285 | * 10.6 and later. All information pointed to by this command |
| 1286 | * is encoded using byte streams, so no endian swapping is needed |
| 1287 | * to interpret it. |
| 1288 | */ |
| 1289 | struct dyld_info_command { |
| 1290 | uint32_t cmd; /* LC_DYLD_INFO or LC_DYLD_INFO_ONLY */ |
| 1291 | uint32_t cmdsize; /* sizeof(struct dyld_info_command) */ |
| 1292 | |
| 1293 | /* |
| 1294 | * Dyld rebases an image whenever dyld loads it at an address different |
| 1295 | * from its preferred address. The rebase information is a stream |
| 1296 | * of byte sized opcodes whose symbolic names start with REBASE_OPCODE_. |
| 1297 | * Conceptually the rebase information is a table of tuples: |
| 1298 | * <seg-index, seg-offset, type> |
| 1299 | * The opcodes are a compressed way to encode the table by only |
| 1300 | * encoding when a column changes. In addition simple patterns |
| 1301 | * like "every n'th offset for m times" can be encoded in a few |
| 1302 | * bytes. |
| 1303 | */ |
| 1304 | uint32_t rebase_off; /* file offset to rebase info */ |
| 1305 | uint32_t rebase_size; /* size of rebase info */ |
| 1306 | |
| 1307 | /* |
| 1308 | * Dyld binds an image during the loading process, if the image |
| 1309 | * requires any pointers to be initialized to symbols in other images. |
| 1310 | * The bind information is a stream of byte sized |
| 1311 | * opcodes whose symbolic names start with BIND_OPCODE_. |
| 1312 | * Conceptually the bind information is a table of tuples: |
| 1313 | * <seg-index, seg-offset, type, symbol-library-ordinal, symbol-name, addend> |
| 1314 | * The opcodes are a compressed way to encode the table by only |
| 1315 | * encoding when a column changes. In addition simple patterns |
| 1316 | * like for runs of pointers initialzed to the same value can be |
| 1317 | * encoded in a few bytes. |
| 1318 | */ |
| 1319 | uint32_t bind_off; /* file offset to binding info */ |
| 1320 | uint32_t bind_size; /* size of binding info */ |
| 1321 | |
| 1322 | /* |
| 1323 | * Some C++ programs require dyld to unique symbols so that all |
| 1324 | * images in the process use the same copy of some code/data. |
| 1325 | * This step is done after binding. The content of the weak_bind |
| 1326 | * info is an opcode stream like the bind_info. But it is sorted |
| 1327 | * alphabetically by symbol name. This enable dyld to walk |
| 1328 | * all images with weak binding information in order and look |
| 1329 | * for collisions. If there are no collisions, dyld does |
| 1330 | * no updating. That means that some fixups are also encoded |
| 1331 | * in the bind_info. For instance, all calls to "operator new" |
| 1332 | * are first bound to libstdc++.dylib using the information |
| 1333 | * in bind_info. Then if some image overrides operator new |
| 1334 | * that is detected when the weak_bind information is processed |
| 1335 | * and the call to operator new is then rebound. |
| 1336 | */ |
| 1337 | uint32_t weak_bind_off; /* file offset to weak binding info */ |
| 1338 | uint32_t weak_bind_size; /* size of weak binding info */ |
| 1339 | |
| 1340 | /* |
| 1341 | * Some uses of external symbols do not need to be bound immediately. |
| 1342 | * Instead they can be lazily bound on first use. The lazy_bind |
| 1343 | * are contains a stream of BIND opcodes to bind all lazy symbols. |
| 1344 | * Normal use is that dyld ignores the lazy_bind section when |
| 1345 | * loading an image. Instead the static linker arranged for the |
| 1346 | * lazy pointer to initially point to a helper function which |
| 1347 | * pushes the offset into the lazy_bind area for the symbol |
| 1348 | * needing to be bound, then jumps to dyld which simply adds |
| 1349 | * the offset to lazy_bind_off to get the information on what |
| 1350 | * to bind. |
| 1351 | */ |
| 1352 | uint32_t lazy_bind_off; /* file offset to lazy binding info */ |
| 1353 | uint32_t lazy_bind_size; /* size of lazy binding infs */ |
| 1354 | |
| 1355 | /* |
| 1356 | * The symbols exported by a dylib are encoded in a trie. This |
| 1357 | * is a compact representation that factors out common prefixes. |
| 1358 | * It also reduces LINKEDIT pages in RAM because it encodes all |
| 1359 | * information (name, address, flags) in one small, contiguous range. |
| 1360 | * The export area is a stream of nodes. The first node sequentially |
| 1361 | * is the start node for the trie. |
| 1362 | * |
| 1363 | * Nodes for a symbol start with a uleb128 that is the length of |
| 1364 | * the exported symbol information for the string so far. |
| 1365 | * If there is no exported symbol, the node starts with a zero byte. |
| 1366 | * If there is exported info, it follows the length. |
| 1367 | * |
| 1368 | * First is a uleb128 containing flags. Normally, it is followed by |
| 1369 | * a uleb128 encoded offset which is location of the content named |
| 1370 | * by the symbol from the mach_header for the image. If the flags |
| 1371 | * is EXPORT_SYMBOL_FLAGS_REEXPORT, then following the flags is |
| 1372 | * a uleb128 encoded library ordinal, then a zero terminated |
| 1373 | * UTF8 string. If the string is zero length, then the symbol |
| 1374 | * is re-export from the specified dylib with the same name. |
| 1375 | * If the flags is EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER, then following |
| 1376 | * the flags is two uleb128s: the stub offset and the resolver offset. |
| 1377 | * The stub is used by non-lazy pointers. The resolver is used |
| 1378 | * by lazy pointers and must be called to get the actual address to use. |
| 1379 | * |
| 1380 | * After the optional exported symbol information is a byte of |
| 1381 | * how many edges (0-255) that this node has leaving it, |
| 1382 | * followed by each edge. |
| 1383 | * Each edge is a zero terminated UTF8 of the addition chars |
| 1384 | * in the symbol, followed by a uleb128 offset for the node that |
| 1385 | * edge points to. |
| 1386 | * |
| 1387 | */ |
| 1388 | uint32_t export_off; /* file offset to lazy binding info */ |
| 1389 | uint32_t export_size; /* size of lazy binding infs */ |
| 1390 | }; |
| 1391 | |
| 1392 | /* |
| 1393 | * The following are used to encode rebasing information |
| 1394 | */ |
| 1395 | #define REBASE_TYPE_POINTER 1 |
| 1396 | #define REBASE_TYPE_TEXT_ABSOLUTE32 2 |
| 1397 | #define REBASE_TYPE_TEXT_PCREL32 3 |
| 1398 | |
| 1399 | #define REBASE_OPCODE_MASK 0xF0 |
| 1400 | #define REBASE_IMMEDIATE_MASK 0x0F |
| 1401 | #define REBASE_OPCODE_DONE 0x00 |
| 1402 | #define REBASE_OPCODE_SET_TYPE_IMM 0x10 |
| 1403 | #define REBASE_OPCODE_SET_SEGMENT_AND_OFFSET_ULEB 0x20 |
| 1404 | #define REBASE_OPCODE_ADD_ADDR_ULEB 0x30 |
| 1405 | #define REBASE_OPCODE_ADD_ADDR_IMM_SCALED 0x40 |
| 1406 | #define REBASE_OPCODE_DO_REBASE_IMM_TIMES 0x50 |
| 1407 | #define REBASE_OPCODE_DO_REBASE_ULEB_TIMES 0x60 |
| 1408 | #define REBASE_OPCODE_DO_REBASE_ADD_ADDR_ULEB 0x70 |
| 1409 | #define REBASE_OPCODE_DO_REBASE_ULEB_TIMES_SKIPPING_ULEB 0x80 |
| 1410 | |
| 1411 | |
| 1412 | /* |
| 1413 | * The following are used to encode binding information |
| 1414 | */ |
| 1415 | #define BIND_TYPE_POINTER 1 |
| 1416 | #define BIND_TYPE_TEXT_ABSOLUTE32 2 |
| 1417 | #define BIND_TYPE_TEXT_PCREL32 3 |
| 1418 | |
| 1419 | #define BIND_SPECIAL_DYLIB_SELF 0 |
| 1420 | #define BIND_SPECIAL_DYLIB_MAIN_EXECUTABLE -1 |
| 1421 | #define BIND_SPECIAL_DYLIB_FLAT_LOOKUP -2 |
| 1422 | #define BIND_SPECIAL_DYLIB_WEAK_LOOKUP -3 |
| 1423 | |
| 1424 | #define BIND_SYMBOL_FLAGS_WEAK_IMPORT 0x1 |
| 1425 | #define BIND_SYMBOL_FLAGS_NON_WEAK_DEFINITION 0x8 |
| 1426 | |
| 1427 | #define BIND_OPCODE_MASK 0xF0 |
| 1428 | #define BIND_IMMEDIATE_MASK 0x0F |
| 1429 | #define BIND_OPCODE_DONE 0x00 |
| 1430 | #define BIND_OPCODE_SET_DYLIB_ORDINAL_IMM 0x10 |
| 1431 | #define BIND_OPCODE_SET_DYLIB_ORDINAL_ULEB 0x20 |
| 1432 | #define BIND_OPCODE_SET_DYLIB_SPECIAL_IMM 0x30 |
| 1433 | #define BIND_OPCODE_SET_SYMBOL_TRAILING_FLAGS_IMM 0x40 |
| 1434 | #define BIND_OPCODE_SET_TYPE_IMM 0x50 |
| 1435 | #define BIND_OPCODE_SET_ADDEND_SLEB 0x60 |
| 1436 | #define BIND_OPCODE_SET_SEGMENT_AND_OFFSET_ULEB 0x70 |
| 1437 | #define BIND_OPCODE_ADD_ADDR_ULEB 0x80 |
| 1438 | #define BIND_OPCODE_DO_BIND 0x90 |
| 1439 | #define BIND_OPCODE_DO_BIND_ADD_ADDR_ULEB 0xA0 |
| 1440 | #define BIND_OPCODE_DO_BIND_ADD_ADDR_IMM_SCALED 0xB0 |
| 1441 | #define BIND_OPCODE_DO_BIND_ULEB_TIMES_SKIPPING_ULEB 0xC0 |
| 1442 | #define BIND_OPCODE_THREADED 0xD0 |
| 1443 | #define BIND_SUBOPCODE_THREADED_SET_BIND_ORDINAL_TABLE_SIZE_ULEB 0x00 |
| 1444 | #define BIND_SUBOPCODE_THREADED_APPLY 0x01 |
| 1445 | |
| 1446 | |
| 1447 | /* |
| 1448 | * The following are used on the flags byte of a terminal node |
| 1449 | * in the export information. |
| 1450 | */ |
| 1451 | #define EXPORT_SYMBOL_FLAGS_KIND_MASK 0x03 |
| 1452 | #define EXPORT_SYMBOL_FLAGS_KIND_REGULAR 0x00 |
| 1453 | #define EXPORT_SYMBOL_FLAGS_KIND_THREAD_LOCAL 0x01 |
| 1454 | #define EXPORT_SYMBOL_FLAGS_KIND_ABSOLUTE 0x02 |
| 1455 | #define EXPORT_SYMBOL_FLAGS_WEAK_DEFINITION 0x04 |
| 1456 | #define EXPORT_SYMBOL_FLAGS_REEXPORT 0x08 |
| 1457 | #define EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER 0x10 |
| 1458 | |
| 1459 | /* |
| 1460 | * The linker_option_command contains linker options embedded in object files. |
| 1461 | */ |
| 1462 | struct linker_option_command { |
| 1463 | uint32_t cmd; /* LC_LINKER_OPTION only used in MH_OBJECT filetypes */ |
| 1464 | uint32_t cmdsize; |
| 1465 | uint32_t count; /* number of strings */ |
| 1466 | /* concatenation of zero terminated UTF8 strings. |
| 1467 | Zero filled at end to align */ |
| 1468 | }; |
| 1469 | |
| 1470 | /* |
| 1471 | * The symseg_command contains the offset and size of the GNU style |
| 1472 | * symbol table information as described in the header file <symseg.h>. |
| 1473 | * The symbol roots of the symbol segments must also be aligned properly |
| 1474 | * in the file. So the requirement of keeping the offsets aligned to a |
| 1475 | * multiple of a 4 bytes translates to the length field of the symbol |
| 1476 | * roots also being a multiple of a long. Also the padding must again be |
| 1477 | * zeroed. (THIS IS OBSOLETE and no longer supported). |
| 1478 | */ |
| 1479 | struct symseg_command { |
| 1480 | uint32_t cmd; /* LC_SYMSEG */ |
| 1481 | uint32_t cmdsize; /* sizeof(struct symseg_command) */ |
| 1482 | uint32_t offset; /* symbol segment offset */ |
| 1483 | uint32_t size; /* symbol segment size in bytes */ |
| 1484 | }; |
| 1485 | |
| 1486 | /* |
| 1487 | * The ident_command contains a free format string table following the |
| 1488 | * ident_command structure. The strings are null terminated and the size of |
| 1489 | * the command is padded out with zero bytes to a multiple of 4 bytes/ |
| 1490 | * (THIS IS OBSOLETE and no longer supported). |
| 1491 | */ |
| 1492 | struct ident_command { |
| 1493 | uint32_t cmd; /* LC_IDENT */ |
| 1494 | uint32_t cmdsize; /* strings that follow this command */ |
| 1495 | }; |
| 1496 | |
| 1497 | /* |
| 1498 | * The fvmfile_command contains a reference to a file to be loaded at the |
| 1499 | * specified virtual address. (Presently, this command is reserved for |
| 1500 | * internal use. The kernel ignores this command when loading a program into |
| 1501 | * memory). |
| 1502 | */ |
| 1503 | struct fvmfile_command { |
| 1504 | uint32_t cmd; /* LC_FVMFILE */ |
| 1505 | uint32_t cmdsize; /* includes pathname string */ |
| 1506 | union lc_str name; /* files pathname */ |
| 1507 | uint32_t header_addr; /* files virtual address */ |
| 1508 | }; |
| 1509 | |
| 1510 | |
| 1511 | /* |
| 1512 | * The entry_point_command is a replacement for thread_command. |
| 1513 | * It is used for main executables to specify the location (file offset) |
| 1514 | * of main(). If -stack_size was used at link time, the stacksize |
| 1515 | * field will contain the stack size need for the main thread. |
| 1516 | */ |
| 1517 | struct entry_point_command { |
| 1518 | uint32_t cmd; /* LC_MAIN only used in MH_EXECUTE filetypes */ |
| 1519 | uint32_t cmdsize; /* 24 */ |
| 1520 | uint64_t entryoff; /* file (__TEXT) offset of main() */ |
| 1521 | uint64_t stacksize;/* if not zero, initial stack size */ |
| 1522 | }; |
| 1523 | |
| 1524 | |
| 1525 | /* |
| 1526 | * The source_version_command is an optional load command containing |
| 1527 | * the version of the sources used to build the binary. |
| 1528 | */ |
| 1529 | struct source_version_command { |
| 1530 | uint32_t cmd; /* LC_SOURCE_VERSION */ |
| 1531 | uint32_t cmdsize; /* 16 */ |
| 1532 | uint64_t version; /* A.B.C.D.E packed as a24.b10.c10.d10.e10 */ |
| 1533 | }; |
| 1534 | |
| 1535 | |
| 1536 | /* |
| 1537 | * The LC_DATA_IN_CODE load commands uses a linkedit_data_command |
| 1538 | * to point to an array of data_in_code_entry entries. Each entry |
| 1539 | * describes a range of data in a code section. |
| 1540 | */ |
| 1541 | struct data_in_code_entry { |
| 1542 | uint32_t offset; /* from mach_header to start of data range*/ |
| 1543 | uint16_t length; /* number of bytes in data range */ |
| 1544 | uint16_t kind; /* a DICE_KIND_* value */ |
| 1545 | }; |
| 1546 | #define DICE_KIND_DATA 0x0001 |
| 1547 | #define DICE_KIND_JUMP_TABLE8 0x0002 |
| 1548 | #define DICE_KIND_JUMP_TABLE16 0x0003 |
| 1549 | #define DICE_KIND_JUMP_TABLE32 0x0004 |
| 1550 | #define DICE_KIND_ABS_JUMP_TABLE32 0x0005 |
| 1551 | |
| 1552 | |
| 1553 | |
| 1554 | /* |
| 1555 | * Sections of type S_THREAD_LOCAL_VARIABLES contain an array |
| 1556 | * of tlv_descriptor structures. |
| 1557 | */ |
| 1558 | struct tlv_descriptor |
| 1559 | { |
| 1560 | void* (*thunk)(struct tlv_descriptor*); |
| 1561 | unsigned long key; |
| 1562 | unsigned long offset; |
| 1563 | }; |
| 1564 | |
| 1565 | /* |
| 1566 | * LC_NOTE commands describe a region of arbitrary data included in a Mach-O |
| 1567 | * file. Its initial use is to record extra data in MH_CORE files. |
| 1568 | */ |
| 1569 | struct note_command { |
| 1570 | uint32_t cmd; /* LC_NOTE */ |
| 1571 | uint32_t cmdsize; /* sizeof(struct note_command) */ |
| 1572 | char data_owner[16]; /* owner name for this LC_NOTE */ |
| 1573 | uint64_t offset; /* file offset of this data */ |
| 1574 | uint64_t size; /* length of data region */ |
| 1575 | }; |
| 1576 | |
| 1577 | #endif /* _MACHO_LOADER_H_ */ |
Andrew Kelley <andrew@ziglang.org> 2020-11-08 19:51:30-08:00
lib