]>
Commit | Line | Data |
---|---|---|
1c79356b | 1 | /* |
316670eb | 2 | * Copyright (c) 1999-2010 Apple Inc. All Rights Reserved. |
1c79356b | 3 | * |
b0d623f7 | 4 | * @APPLE_LICENSE_HEADER_START@ |
1c79356b | 5 | * |
2d21ac55 A |
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 | |
b0d623f7 A |
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. | |
2d21ac55 A |
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 | |
8f6c56a5 A |
15 | * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, |
16 | * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, | |
2d21ac55 A |
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. | |
8f6c56a5 | 20 | * |
b0d623f7 | 21 | * @APPLE_LICENSE_HEADER_END@ |
1c79356b A |
22 | */ |
23 | #ifndef _MACHO_LOADER_H_ | |
24 | #define _MACHO_LOADER_H_ | |
25 | ||
26 | /* | |
27 | * This file describes the format of mach object files. | |
28 | */ | |
2d21ac55 | 29 | #include <stdint.h> |
1c79356b A |
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> | |
2d21ac55 | 48 | #include <architecture/byte_order.h> |
1c79356b A |
49 | |
50 | /* | |
2d21ac55 A |
51 | * The 32-bit mach header appears at the very beginning of the object file for |
52 | * 32-bit architectures. | |
1c79356b A |
53 | */ |
54 | struct mach_header { | |
91447636 | 55 | uint32_t magic; /* mach magic number identifier */ |
1c79356b A |
56 | cpu_type_t cputype; /* cpu specifier */ |
57 | cpu_subtype_t cpusubtype; /* machine specifier */ | |
91447636 A |
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 | ||
2d21ac55 A |
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 | ||
91447636 A |
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 */ | |
1c79356b A |
81 | }; |
82 | ||
91447636 | 83 | /* Constant for the magic field of the mach_header_64 (64-bit architectures) */ |
2d21ac55 A |
84 | #define MH_MAGIC_64 0xfeedfacf /* the 64-bit mach magic number */ |
85 | #define MH_CIGAM_64 0xcffaedfe /* NXSwapInt(MH_MAGIC_64) */ | |
91447636 | 86 | |
1c79356b A |
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 | |
2d21ac55 | 98 | * segment padding greatly increases its size. |
1c79356b A |
99 | * |
100 | * The file type MH_PRELOAD is an executable format intended for things that | |
2d21ac55 | 101 | * are not executed under the kernel (proms, stand alones, kernels, etc). The |
1c79356b A |
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 */ | |
2d21ac55 | 115 | #define MH_DYLIB 0x6 /* dynamically bound shared library */ |
1c79356b | 116 | #define MH_DYLINKER 0x7 /* dynamic link editor */ |
2d21ac55 A |
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 */ | |
b0d623f7 | 122 | #define MH_KEXT_BUNDLE 0xb /* x86_64 kexts */ |
1c79356b A |
123 | |
124 | /* Constants for the flags field of the mach_header */ | |
125 | #define MH_NOUNDEFS 0x1 /* the object file has no undefined | |
2d21ac55 | 126 | references */ |
1c79356b A |
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. */ | |
2d21ac55 | 136 | #define MH_PREBOUND 0x10 /* the file has its dynamic undefined |
1c79356b | 137 | references prebound. */ |
2d21ac55 A |
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 | ||
0c530ab8 A |
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. */ | |
2d21ac55 A |
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 */ | |
593a1d5f A |
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. */ | |
316670eb A |
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 | ||
6d2010ae A |
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. */ | |
1c79356b | 209 | |
3e170ce0 A |
210 | #define MH_APP_EXTENSION_SAFE 0x02000000 /* The code was linked for use in an |
211 | application extension. */ | |
212 | ||
1c79356b A |
213 | /* |
214 | * The load commands directly follow the mach_header. The total size of all | |
215 | * of the commands is given by the sizeofcmds field in the mach_header. All | |
216 | * load commands must have as their first two fields cmd and cmdsize. The cmd | |
217 | * field is filled in with a constant for that command type. Each command type | |
218 | * has a structure specifically for it. The cmdsize field is the size in bytes | |
219 | * of the particular load command structure plus anything that follows it that | |
220 | * is a part of the load command (i.e. section structures, strings, etc.). To | |
221 | * advance to the next load command the cmdsize can be added to the offset or | |
91447636 A |
222 | * pointer of the current load command. The cmdsize for 32-bit architectures |
223 | * MUST be a multiple of 4 bytes and for 64-bit architectures MUST be a multiple | |
224 | * of 8 bytes (these are forever the maximum alignment of any load commands). | |
1c79356b A |
225 | * The padded bytes must be zero. All tables in the object file must also |
226 | * follow these rules so the file can be memory mapped. Otherwise the pointers | |
227 | * to these tables will not work well or at all on some machines. With all | |
228 | * padding zeroed like objects will compare byte for byte. | |
229 | */ | |
230 | struct load_command { | |
2d21ac55 A |
231 | uint32_t cmd; /* type of load command */ |
232 | uint32_t cmdsize; /* total size of command in bytes */ | |
1c79356b A |
233 | }; |
234 | ||
2d21ac55 A |
235 | /* |
236 | * After MacOS X 10.1 when a new load command is added that is required to be | |
237 | * understood by the dynamic linker for the image to execute properly the | |
238 | * LC_REQ_DYLD bit will be or'ed into the load command constant. If the dynamic | |
239 | * linker sees such a load command it it does not understand will issue a | |
240 | * "unknown load command required for execution" error and refuse to use the | |
241 | * image. Other load commands without this bit that are not understood will | |
242 | * simply be ignored. | |
243 | */ | |
244 | #define LC_REQ_DYLD 0x80000000 | |
245 | ||
1c79356b A |
246 | /* Constants for the cmd field of all load commands, the type */ |
247 | #define LC_SEGMENT 0x1 /* segment of this file to be mapped */ | |
248 | #define LC_SYMTAB 0x2 /* link-edit stab symbol table info */ | |
249 | #define LC_SYMSEG 0x3 /* link-edit gdb symbol table info (obsolete) */ | |
250 | #define LC_THREAD 0x4 /* thread */ | |
251 | #define LC_UNIXTHREAD 0x5 /* unix thread (includes a stack) */ | |
252 | #define LC_LOADFVMLIB 0x6 /* load a specified fixed VM shared library */ | |
253 | #define LC_IDFVMLIB 0x7 /* fixed VM shared library identification */ | |
254 | #define LC_IDENT 0x8 /* object identification info (obsolete) */ | |
255 | #define LC_FVMFILE 0x9 /* fixed VM file inclusion (internal use) */ | |
256 | #define LC_PREPAGE 0xa /* prepage command (internal use) */ | |
257 | #define LC_DYSYMTAB 0xb /* dynamic link-edit symbol table info */ | |
2d21ac55 A |
258 | #define LC_LOAD_DYLIB 0xc /* load a dynamically linked shared library */ |
259 | #define LC_ID_DYLIB 0xd /* dynamically linked shared lib ident */ | |
1c79356b A |
260 | #define LC_LOAD_DYLINKER 0xe /* load a dynamic linker */ |
261 | #define LC_ID_DYLINKER 0xf /* dynamic linker identification */ | |
2d21ac55 | 262 | #define LC_PREBOUND_DYLIB 0x10 /* modules prebound for a dynamically */ |
1c79356b | 263 | /* linked shared library */ |
2d21ac55 A |
264 | #define LC_ROUTINES 0x11 /* image routines */ |
265 | #define LC_SUB_FRAMEWORK 0x12 /* sub framework */ | |
266 | #define LC_SUB_UMBRELLA 0x13 /* sub umbrella */ | |
267 | #define LC_SUB_CLIENT 0x14 /* sub client */ | |
268 | #define LC_SUB_LIBRARY 0x15 /* sub library */ | |
269 | #define LC_TWOLEVEL_HINTS 0x16 /* two-level namespace lookup hints */ | |
270 | #define LC_PREBIND_CKSUM 0x17 /* prebind checksum */ | |
271 | ||
272 | /* | |
273 | * load a dynamically linked shared library that is allowed to be missing | |
274 | * (all symbols are weak imported). | |
275 | */ | |
276 | #define LC_LOAD_WEAK_DYLIB (0x18 | LC_REQ_DYLD) | |
277 | ||
278 | #define LC_SEGMENT_64 0x19 /* 64-bit segment of this file to be | |
279 | mapped */ | |
280 | #define LC_ROUTINES_64 0x1a /* 64-bit image routines */ | |
281 | #define LC_UUID 0x1b /* the uuid */ | |
282 | #define LC_RPATH (0x1c | LC_REQ_DYLD) /* runpath additions */ | |
283 | #define LC_CODE_SIGNATURE 0x1d /* local of code signature */ | |
284 | #define LC_SEGMENT_SPLIT_INFO 0x1e /* local of info to split segments */ | |
285 | #define LC_REEXPORT_DYLIB (0x1f | LC_REQ_DYLD) /* load and re-export dylib */ | |
593a1d5f A |
286 | #define LC_LAZY_LOAD_DYLIB 0x20 /* delay load of dylib until first use */ |
287 | #define LC_ENCRYPTION_INFO 0x21 /* encrypted segment information */ | |
b0d623f7 A |
288 | #define LC_DYLD_INFO 0x22 /* compressed dyld information */ |
289 | #define LC_DYLD_INFO_ONLY (0x22|LC_REQ_DYLD) /* compressed dyld information only */ | |
316670eb A |
290 | #define LC_LOAD_UPWARD_DYLIB (0x23 | LC_REQ_DYLD) /* load upward dylib */ |
291 | #define LC_VERSION_MIN_MACOSX 0x24 /* build for MacOSX min OS version */ | |
292 | #define LC_VERSION_MIN_IPHONEOS 0x25 /* build for iPhoneOS min OS version */ | |
293 | #define LC_FUNCTION_STARTS 0x26 /* compressed table of function start addresses */ | |
294 | #define LC_DYLD_ENVIRONMENT 0x27 /* string for dyld to treat | |
295 | like environment variable */ | |
296 | #define LC_MAIN (0x28|LC_REQ_DYLD) /* replacement for LC_UNIXTHREAD */ | |
297 | #define LC_DATA_IN_CODE 0x29 /* table of non-instructions in __text */ | |
298 | #define LC_SOURCE_VERSION 0x2A /* source version used to build binary */ | |
299 | #define LC_DYLIB_CODE_SIGN_DRS 0x2B /* Code signing DRs copied from linked dylibs */ | |
39236c6e | 300 | #define LC_ENCRYPTION_INFO_64 0x2C /* 64-bit encrypted segment information */ |
3e170ce0 A |
301 | #define LC_LINKER_OPTION 0x2D /* linker options in MH_OBJECT files */ |
302 | #define LC_LINKER_OPTIMIZATION_HINT 0x2E /* optimization hints in MH_OBJECT files */ | |
39037602 | 303 | #define LC_VERSION_MIN_TVOS 0x2F /* build for AppleTV min OS version */ |
3e170ce0 | 304 | #define LC_VERSION_MIN_WATCHOS 0x30 /* build for Watch min OS version */ |
5ba3f43e A |
305 | #define LC_NOTE 0x31 /* arbitrary data included within a Mach-O file */ |
306 | #define LC_BUILD_VERSION 0x32 /* build for platform min OS version */ | |
1c79356b A |
307 | |
308 | /* | |
309 | * A variable length string in a load command is represented by an lc_str | |
310 | * union. The strings are stored just after the load command structure and | |
311 | * the offset is from the start of the load command structure. The size | |
312 | * of the string is reflected in the cmdsize field of the load command. | |
313 | * Once again any padded bytes to bring the cmdsize field to a multiple | |
2d21ac55 | 314 | * of 4 bytes must be zero. |
1c79356b A |
315 | */ |
316 | union lc_str { | |
2d21ac55 A |
317 | uint32_t offset; /* offset to the string */ |
318 | #ifndef __LP64__ | |
1c79356b | 319 | char *ptr; /* pointer to the string */ |
2d21ac55 | 320 | #endif |
1c79356b A |
321 | }; |
322 | ||
323 | /* | |
324 | * The segment load command indicates that a part of this file is to be | |
325 | * mapped into the task's address space. The size of this segment in memory, | |
326 | * vmsize, maybe equal to or larger than the amount to map from this file, | |
327 | * filesize. The file is mapped starting at fileoff to the beginning of | |
328 | * the segment in memory, vmaddr. The rest of the memory of the segment, | |
329 | * if any, is allocated zero fill on demand. The segment's maximum virtual | |
330 | * memory protection and initial virtual memory protection are specified | |
331 | * by the maxprot and initprot fields. If the segment has sections then the | |
332 | * section structures directly follow the segment command and their size is | |
333 | * reflected in cmdsize. | |
334 | */ | |
2d21ac55 A |
335 | struct segment_command { /* for 32-bit architectures */ |
336 | uint32_t cmd; /* LC_SEGMENT */ | |
337 | uint32_t cmdsize; /* includes sizeof section structs */ | |
1c79356b | 338 | char segname[16]; /* segment name */ |
2d21ac55 A |
339 | uint32_t vmaddr; /* memory address of this segment */ |
340 | uint32_t vmsize; /* memory size of this segment */ | |
341 | uint32_t fileoff; /* file offset of this segment */ | |
342 | uint32_t filesize; /* amount to map from the file */ | |
1c79356b A |
343 | vm_prot_t maxprot; /* maximum VM protection */ |
344 | vm_prot_t initprot; /* initial VM protection */ | |
2d21ac55 A |
345 | uint32_t nsects; /* number of sections in segment */ |
346 | uint32_t flags; /* flags */ | |
1c79356b A |
347 | }; |
348 | ||
91447636 A |
349 | /* |
350 | * The 64-bit segment load command indicates that a part of this file is to be | |
351 | * mapped into a 64-bit task's address space. If the 64-bit segment has | |
352 | * sections then section_64 structures directly follow the 64-bit segment | |
353 | * command and their size is reflected in cmdsize. | |
354 | */ | |
2d21ac55 | 355 | struct segment_command_64 { /* for 64-bit architectures */ |
91447636 A |
356 | uint32_t cmd; /* LC_SEGMENT_64 */ |
357 | uint32_t cmdsize; /* includes sizeof section_64 structs */ | |
358 | char segname[16]; /* segment name */ | |
359 | uint64_t vmaddr; /* memory address of this segment */ | |
360 | uint64_t vmsize; /* memory size of this segment */ | |
361 | uint64_t fileoff; /* file offset of this segment */ | |
362 | uint64_t filesize; /* amount to map from the file */ | |
363 | vm_prot_t maxprot; /* maximum VM protection */ | |
364 | vm_prot_t initprot; /* initial VM protection */ | |
365 | uint32_t nsects; /* number of sections in segment */ | |
366 | uint32_t flags; /* flags */ | |
367 | }; | |
368 | ||
1c79356b A |
369 | /* Constants for the flags field of the segment_command */ |
370 | #define SG_HIGHVM 0x1 /* the file contents for this segment is for | |
371 | the high part of the VM space, the low part | |
372 | is zero filled (for stacks in core files) */ | |
373 | #define SG_FVMLIB 0x2 /* this segment is the VM that is allocated by | |
374 | a fixed VM library, for overlap checking in | |
375 | the link editor */ | |
376 | #define SG_NORELOC 0x4 /* this segment has nothing that was relocated | |
377 | in it and nothing relocated to it, that is | |
378 | it maybe safely replaced without relocation*/ | |
2d21ac55 A |
379 | #define SG_PROTECTED_VERSION_1 0x8 /* This segment is protected. If the |
380 | segment starts at file offset 0, the | |
381 | first page of the segment is not | |
382 | protected. All other pages of the | |
383 | segment are protected. */ | |
1c79356b A |
384 | |
385 | /* | |
386 | * A segment is made up of zero or more sections. Non-MH_OBJECT files have | |
387 | * all of their segments with the proper sections in each, and padded to the | |
388 | * specified segment alignment when produced by the link editor. The first | |
389 | * segment of a MH_EXECUTE and MH_FVMLIB format file contains the mach_header | |
2d21ac55 | 390 | * and load commands of the object file before its first section. The zero |
1c79356b A |
391 | * fill sections are always last in their segment (in all formats). This |
392 | * allows the zeroed segment padding to be mapped into memory where zero fill | |
91447636 A |
393 | * sections might be. The gigabyte zero fill sections, those with the section |
394 | * type S_GB_ZEROFILL, can only be in a segment with sections of this type. | |
395 | * These segments are then placed after all other segments. | |
1c79356b | 396 | * |
2d21ac55 | 397 | * The MH_OBJECT format has all of its sections in one segment for |
1c79356b A |
398 | * compactness. There is no padding to a specified segment boundary and the |
399 | * mach_header and load commands are not part of the segment. | |
400 | * | |
401 | * Sections with the same section name, sectname, going into the same segment, | |
402 | * segname, are combined by the link editor. The resulting section is aligned | |
403 | * to the maximum alignment of the combined sections and is the new section's | |
404 | * alignment. The combined sections are aligned to their original alignment in | |
405 | * the combined section. Any padded bytes to get the specified alignment are | |
406 | * zeroed. | |
407 | * | |
408 | * The format of the relocation entries referenced by the reloff and nreloc | |
409 | * fields of the section structure for mach object files is described in the | |
410 | * header file <reloc.h>. | |
411 | */ | |
2d21ac55 | 412 | struct section { /* for 32-bit architectures */ |
1c79356b A |
413 | char sectname[16]; /* name of this section */ |
414 | char segname[16]; /* segment this section goes in */ | |
2d21ac55 A |
415 | uint32_t addr; /* memory address of this section */ |
416 | uint32_t size; /* size in bytes of this section */ | |
417 | uint32_t offset; /* file offset of this section */ | |
418 | uint32_t align; /* section alignment (power of 2) */ | |
419 | uint32_t reloff; /* file offset of relocation entries */ | |
420 | uint32_t nreloc; /* number of relocation entries */ | |
421 | uint32_t flags; /* flags (section type and attributes)*/ | |
422 | uint32_t reserved1; /* reserved (for offset or index) */ | |
423 | uint32_t reserved2; /* reserved (for count or sizeof) */ | |
1c79356b A |
424 | }; |
425 | ||
91447636 A |
426 | struct section_64 { /* for 64-bit architectures */ |
427 | char sectname[16]; /* name of this section */ | |
428 | char segname[16]; /* segment this section goes in */ | |
429 | uint64_t addr; /* memory address of this section */ | |
430 | uint64_t size; /* size in bytes of this section */ | |
431 | uint32_t offset; /* file offset of this section */ | |
432 | uint32_t align; /* section alignment (power of 2) */ | |
433 | uint32_t reloff; /* file offset of relocation entries */ | |
434 | uint32_t nreloc; /* number of relocation entries */ | |
435 | uint32_t flags; /* flags (section type and attributes)*/ | |
436 | uint32_t reserved1; /* reserved (for offset or index) */ | |
437 | uint32_t reserved2; /* reserved (for count or sizeof) */ | |
438 | uint32_t reserved3; /* reserved */ | |
439 | }; | |
440 | ||
1c79356b A |
441 | /* |
442 | * The flags field of a section structure is separated into two parts a section | |
443 | * type and section attributes. The section types are mutually exclusive (it | |
444 | * can only have one type) but the section attributes are not (it may have more | |
445 | * than one attribute). | |
446 | */ | |
447 | #define SECTION_TYPE 0x000000ff /* 256 section types */ | |
448 | #define SECTION_ATTRIBUTES 0xffffff00 /* 24 section attributes */ | |
449 | ||
450 | /* Constants for the type of a section */ | |
451 | #define S_REGULAR 0x0 /* regular section */ | |
452 | #define S_ZEROFILL 0x1 /* zero fill on demand section */ | |
453 | #define S_CSTRING_LITERALS 0x2 /* section with only literal C strings*/ | |
454 | #define S_4BYTE_LITERALS 0x3 /* section with only 4 byte literals */ | |
455 | #define S_8BYTE_LITERALS 0x4 /* section with only 8 byte literals */ | |
456 | #define S_LITERAL_POINTERS 0x5 /* section with only pointers to */ | |
457 | /* literals */ | |
458 | /* | |
459 | * For the two types of symbol pointers sections and the symbol stubs section | |
460 | * they have indirect symbol table entries. For each of the entries in the | |
461 | * section the indirect symbol table entries, in corresponding order in the | |
462 | * indirect symbol table, start at the index stored in the reserved1 field | |
463 | * of the section structure. Since the indirect symbol table entries | |
464 | * correspond to the entries in the section the number of indirect symbol table | |
465 | * entries is inferred from the size of the section divided by the size of the | |
466 | * entries in the section. For symbol pointers sections the size of the entries | |
467 | * in the section is 4 bytes and for symbol stubs sections the byte size of the | |
468 | * stubs is stored in the reserved2 field of the section structure. | |
469 | */ | |
470 | #define S_NON_LAZY_SYMBOL_POINTERS 0x6 /* section with only non-lazy | |
471 | symbol pointers */ | |
472 | #define S_LAZY_SYMBOL_POINTERS 0x7 /* section with only lazy symbol | |
473 | pointers */ | |
474 | #define S_SYMBOL_STUBS 0x8 /* section with only symbol | |
475 | stubs, byte size of stub in | |
476 | the reserved2 field */ | |
477 | #define S_MOD_INIT_FUNC_POINTERS 0x9 /* section with only function | |
478 | pointers for initialization*/ | |
2d21ac55 A |
479 | #define S_MOD_TERM_FUNC_POINTERS 0xa /* section with only function |
480 | pointers for termination */ | |
481 | #define S_COALESCED 0xb /* section contains symbols that | |
482 | are to be coalesced */ | |
483 | #define S_GB_ZEROFILL 0xc /* zero fill on demand section | |
484 | (that can be larger than 4 | |
485 | gigabytes) */ | |
486 | #define S_INTERPOSING 0xd /* section with only pairs of | |
487 | function pointers for | |
488 | interposing */ | |
593a1d5f A |
489 | #define S_16BYTE_LITERALS 0xe /* section with only 16 byte |
490 | literals */ | |
491 | #define S_DTRACE_DOF 0xf /* section contains | |
492 | DTrace Object Format */ | |
493 | #define S_LAZY_DYLIB_SYMBOL_POINTERS 0x10 /* section with only lazy | |
494 | symbol pointers to lazy | |
495 | loaded dylibs */ | |
316670eb A |
496 | /* |
497 | * Section types to support thread local variables | |
498 | */ | |
499 | #define S_THREAD_LOCAL_REGULAR 0x11 /* template of initial | |
500 | values for TLVs */ | |
501 | #define S_THREAD_LOCAL_ZEROFILL 0x12 /* template of initial | |
502 | values for TLVs */ | |
503 | #define S_THREAD_LOCAL_VARIABLES 0x13 /* TLV descriptors */ | |
504 | #define S_THREAD_LOCAL_VARIABLE_POINTERS 0x14 /* pointers to TLV | |
505 | descriptors */ | |
506 | #define S_THREAD_LOCAL_INIT_FUNCTION_POINTERS 0x15 /* functions to call | |
507 | to initialize TLV | |
508 | values */ | |
509 | ||
1c79356b A |
510 | /* |
511 | * Constants for the section attributes part of the flags field of a section | |
512 | * structure. | |
513 | */ | |
514 | #define SECTION_ATTRIBUTES_USR 0xff000000 /* User setable attributes */ | |
515 | #define S_ATTR_PURE_INSTRUCTIONS 0x80000000 /* section contains only true | |
516 | machine instructions */ | |
2d21ac55 A |
517 | #define S_ATTR_NO_TOC 0x40000000 /* section contains coalesced |
518 | symbols that are not to be | |
519 | in a ranlib table of | |
520 | contents */ | |
521 | #define S_ATTR_STRIP_STATIC_SYMS 0x20000000 /* ok to strip static symbols | |
522 | in this section in files | |
523 | with the MH_DYLDLINK flag */ | |
524 | #define S_ATTR_NO_DEAD_STRIP 0x10000000 /* no dead stripping */ | |
525 | #define S_ATTR_LIVE_SUPPORT 0x08000000 /* blocks are live if they | |
526 | reference live blocks */ | |
527 | #define S_ATTR_SELF_MODIFYING_CODE 0x04000000 /* Used with i386 code stubs | |
528 | written on by dyld */ | |
529 | /* | |
530 | * If a segment contains any sections marked with S_ATTR_DEBUG then all | |
531 | * sections in that segment must have this attribute. No section other than | |
532 | * a section marked with this attribute may reference the contents of this | |
533 | * section. A section with this attribute may contain no symbols and must have | |
534 | * a section type S_REGULAR. The static linker will not copy section contents | |
535 | * from sections with this attribute into its output file. These sections | |
536 | * generally contain DWARF debugging info. | |
537 | */ | |
538 | #define S_ATTR_DEBUG 0x02000000 /* a debug section */ | |
1c79356b A |
539 | #define SECTION_ATTRIBUTES_SYS 0x00ffff00 /* system setable attributes */ |
540 | #define S_ATTR_SOME_INSTRUCTIONS 0x00000400 /* section contains some | |
541 | machine instructions */ | |
542 | #define S_ATTR_EXT_RELOC 0x00000200 /* section has external | |
543 | relocation entries */ | |
544 | #define S_ATTR_LOC_RELOC 0x00000100 /* section has local | |
545 | relocation entries */ | |
546 | ||
547 | ||
548 | /* | |
549 | * The names of segments and sections in them are mostly meaningless to the | |
550 | * link-editor. But there are few things to support traditional UNIX | |
551 | * executables that require the link-editor and assembler to use some names | |
552 | * agreed upon by convention. | |
553 | * | |
554 | * The initial protection of the "__TEXT" segment has write protection turned | |
555 | * off (not writeable). | |
556 | * | |
557 | * The link-editor will allocate common symbols at the end of the "__common" | |
558 | * section in the "__DATA" segment. It will create the section and segment | |
559 | * if needed. | |
560 | */ | |
561 | ||
562 | /* The currently known segment names and the section names in those segments */ | |
563 | ||
564 | #define SEG_PAGEZERO "__PAGEZERO" /* the pagezero segment which has no */ | |
565 | /* protections and catches NULL */ | |
566 | /* references for MH_EXECUTE files */ | |
567 | ||
568 | ||
569 | #define SEG_TEXT "__TEXT" /* the tradition UNIX text segment */ | |
570 | #define SECT_TEXT "__text" /* the real text part of the text */ | |
571 | /* section no headers, and no padding */ | |
572 | #define SECT_FVMLIB_INIT0 "__fvmlib_init0" /* the fvmlib initialization */ | |
573 | /* section */ | |
574 | #define SECT_FVMLIB_INIT1 "__fvmlib_init1" /* the section following the */ | |
575 | /* fvmlib initialization */ | |
576 | /* section */ | |
577 | ||
578 | #define SEG_DATA "__DATA" /* the tradition UNIX data segment */ | |
579 | #define SECT_DATA "__data" /* the real initialized data section */ | |
580 | /* no padding, no bss overlap */ | |
581 | #define SECT_BSS "__bss" /* the real uninitialized data section*/ | |
582 | /* no padding */ | |
583 | #define SECT_COMMON "__common" /* the section common symbols are */ | |
584 | /* allocated in by the link editor */ | |
585 | ||
586 | #define SEG_OBJC "__OBJC" /* objective-C runtime segment */ | |
587 | #define SECT_OBJC_SYMBOLS "__symbol_table" /* symbol table */ | |
588 | #define SECT_OBJC_MODULES "__module_info" /* module information */ | |
589 | #define SECT_OBJC_STRINGS "__selector_strs" /* string table */ | |
590 | #define SECT_OBJC_REFS "__selector_refs" /* string table */ | |
591 | ||
2d21ac55 | 592 | #define SEG_ICON "__ICON" /* the icon segment */ |
1c79356b A |
593 | #define SECT_ICON_HEADER "__header" /* the icon headers */ |
594 | #define SECT_ICON_TIFF "__tiff" /* the icons in tiff format */ | |
595 | ||
596 | #define SEG_LINKEDIT "__LINKEDIT" /* the segment containing all structs */ | |
597 | /* created and maintained by the link */ | |
598 | /* editor. Created with -seglinkedit */ | |
599 | /* option to ld(1) for MH_EXECUTE and */ | |
600 | /* FVMLIB file types only */ | |
601 | ||
602 | #define SEG_UNIXSTACK "__UNIXSTACK" /* the unix stack segment */ | |
603 | ||
2d21ac55 A |
604 | #define SEG_IMPORT "__IMPORT" /* the segment for the self (dyld) */ |
605 | /* modifing code stubs that has read, */ | |
606 | /* write and execute permissions */ | |
607 | ||
1c79356b A |
608 | /* |
609 | * Fixed virtual memory shared libraries are identified by two things. The | |
610 | * target pathname (the name of the library as found for execution), and the | |
611 | * minor version number. The address of where the headers are loaded is in | |
2d21ac55 | 612 | * header_addr. (THIS IS OBSOLETE and no longer supported). |
1c79356b A |
613 | */ |
614 | struct fvmlib { | |
615 | union lc_str name; /* library's target pathname */ | |
2d21ac55 A |
616 | uint32_t minor_version; /* library's minor version number */ |
617 | uint32_t header_addr; /* library's header address */ | |
1c79356b A |
618 | }; |
619 | ||
620 | /* | |
621 | * A fixed virtual shared library (filetype == MH_FVMLIB in the mach header) | |
622 | * contains a fvmlib_command (cmd == LC_IDFVMLIB) to identify the library. | |
623 | * An object that uses a fixed virtual shared library also contains a | |
624 | * fvmlib_command (cmd == LC_LOADFVMLIB) for each library it uses. | |
2d21ac55 | 625 | * (THIS IS OBSOLETE and no longer supported). |
1c79356b A |
626 | */ |
627 | struct fvmlib_command { | |
2d21ac55 A |
628 | uint32_t cmd; /* LC_IDFVMLIB or LC_LOADFVMLIB */ |
629 | uint32_t cmdsize; /* includes pathname string */ | |
1c79356b A |
630 | struct fvmlib fvmlib; /* the library identification */ |
631 | }; | |
632 | ||
633 | /* | |
634 | * Dynamicly linked shared libraries are identified by two things. The | |
635 | * pathname (the name of the library as found for execution), and the | |
636 | * compatibility version number. The pathname must match and the compatibility | |
637 | * number in the user of the library must be greater than or equal to the | |
638 | * library being used. The time stamp is used to record the time a library was | |
639 | * built and copied into user so it can be use to determined if the library used | |
640 | * at runtime is exactly the same as used to built the program. | |
641 | */ | |
642 | struct dylib { | |
643 | union lc_str name; /* library's path name */ | |
2d21ac55 A |
644 | uint32_t timestamp; /* library's build time stamp */ |
645 | uint32_t current_version; /* library's current version number */ | |
646 | uint32_t compatibility_version; /* library's compatibility vers number*/ | |
1c79356b A |
647 | }; |
648 | ||
649 | /* | |
2d21ac55 | 650 | * A dynamically linked shared library (filetype == MH_DYLIB in the mach header) |
1c79356b | 651 | * contains a dylib_command (cmd == LC_ID_DYLIB) to identify the library. |
2d21ac55 A |
652 | * An object that uses a dynamically linked shared library also contains a |
653 | * dylib_command (cmd == LC_LOAD_DYLIB, LC_LOAD_WEAK_DYLIB, or | |
654 | * LC_REEXPORT_DYLIB) for each library it uses. | |
1c79356b A |
655 | */ |
656 | struct dylib_command { | |
2d21ac55 A |
657 | uint32_t cmd; /* LC_ID_DYLIB, LC_LOAD_{,WEAK_}DYLIB, |
658 | LC_REEXPORT_DYLIB */ | |
659 | uint32_t cmdsize; /* includes pathname string */ | |
1c79356b A |
660 | struct dylib dylib; /* the library identification */ |
661 | }; | |
662 | ||
663 | /* | |
2d21ac55 A |
664 | * A dynamically linked shared library may be a subframework of an umbrella |
665 | * framework. If so it will be linked with "-umbrella umbrella_name" where | |
666 | * Where "umbrella_name" is the name of the umbrella framework. A subframework | |
667 | * can only be linked against by its umbrella framework or other subframeworks | |
668 | * that are part of the same umbrella framework. Otherwise the static link | |
669 | * editor produces an error and states to link against the umbrella framework. | |
670 | * The name of the umbrella framework for subframeworks is recorded in the | |
671 | * following structure. | |
672 | */ | |
673 | struct sub_framework_command { | |
674 | uint32_t cmd; /* LC_SUB_FRAMEWORK */ | |
675 | uint32_t cmdsize; /* includes umbrella string */ | |
676 | union lc_str umbrella; /* the umbrella framework name */ | |
677 | }; | |
678 | ||
679 | /* | |
680 | * For dynamically linked shared libraries that are subframework of an umbrella | |
681 | * framework they can allow clients other than the umbrella framework or other | |
682 | * subframeworks in the same umbrella framework. To do this the subframework | |
683 | * is built with "-allowable_client client_name" and an LC_SUB_CLIENT load | |
684 | * command is created for each -allowable_client flag. The client_name is | |
685 | * usually a framework name. It can also be a name used for bundles clients | |
686 | * where the bundle is built with "-client_name client_name". | |
687 | */ | |
688 | struct sub_client_command { | |
689 | uint32_t cmd; /* LC_SUB_CLIENT */ | |
690 | uint32_t cmdsize; /* includes client string */ | |
691 | union lc_str client; /* the client name */ | |
692 | }; | |
693 | ||
694 | /* | |
695 | * A dynamically linked shared library may be a sub_umbrella of an umbrella | |
696 | * framework. If so it will be linked with "-sub_umbrella umbrella_name" where | |
697 | * Where "umbrella_name" is the name of the sub_umbrella framework. When | |
698 | * staticly linking when -twolevel_namespace is in effect a twolevel namespace | |
699 | * umbrella framework will only cause its subframeworks and those frameworks | |
700 | * listed as sub_umbrella frameworks to be implicited linked in. Any other | |
701 | * dependent dynamic libraries will not be linked it when -twolevel_namespace | |
702 | * is in effect. The primary library recorded by the static linker when | |
703 | * resolving a symbol in these libraries will be the umbrella framework. | |
704 | * Zero or more sub_umbrella frameworks may be use by an umbrella framework. | |
705 | * The name of a sub_umbrella framework is recorded in the following structure. | |
706 | */ | |
707 | struct sub_umbrella_command { | |
708 | uint32_t cmd; /* LC_SUB_UMBRELLA */ | |
709 | uint32_t cmdsize; /* includes sub_umbrella string */ | |
710 | union lc_str sub_umbrella; /* the sub_umbrella framework name */ | |
711 | }; | |
712 | ||
713 | /* | |
714 | * A dynamically linked shared library may be a sub_library of another shared | |
715 | * library. If so it will be linked with "-sub_library library_name" where | |
716 | * Where "library_name" is the name of the sub_library shared library. When | |
717 | * staticly linking when -twolevel_namespace is in effect a twolevel namespace | |
718 | * shared library will only cause its subframeworks and those frameworks | |
719 | * listed as sub_umbrella frameworks and libraries listed as sub_libraries to | |
720 | * be implicited linked in. Any other dependent dynamic libraries will not be | |
721 | * linked it when -twolevel_namespace is in effect. The primary library | |
722 | * recorded by the static linker when resolving a symbol in these libraries | |
723 | * will be the umbrella framework (or dynamic library). Zero or more sub_library | |
724 | * shared libraries may be use by an umbrella framework or (or dynamic library). | |
725 | * The name of a sub_library framework is recorded in the following structure. | |
726 | * For example /usr/lib/libobjc_profile.A.dylib would be recorded as "libobjc". | |
727 | */ | |
728 | struct sub_library_command { | |
729 | uint32_t cmd; /* LC_SUB_LIBRARY */ | |
730 | uint32_t cmdsize; /* includes sub_library string */ | |
731 | union lc_str sub_library; /* the sub_library name */ | |
732 | }; | |
733 | ||
734 | /* | |
735 | * A program (filetype == MH_EXECUTE) that is | |
736 | * prebound to its dynamic libraries has one of these for each library that | |
1c79356b A |
737 | * the static linker used in prebinding. It contains a bit vector for the |
738 | * modules in the library. The bits indicate which modules are bound (1) and | |
739 | * which are not (0) from the library. The bit for module 0 is the low bit | |
740 | * of the first byte. So the bit for the Nth module is: | |
741 | * (linked_modules[N/8] >> N%8) & 1 | |
742 | */ | |
743 | struct prebound_dylib_command { | |
2d21ac55 A |
744 | uint32_t cmd; /* LC_PREBOUND_DYLIB */ |
745 | uint32_t cmdsize; /* includes strings */ | |
1c79356b | 746 | union lc_str name; /* library's path name */ |
2d21ac55 | 747 | uint32_t nmodules; /* number of modules in library */ |
1c79356b A |
748 | union lc_str linked_modules; /* bit vector of linked modules */ |
749 | }; | |
750 | ||
751 | /* | |
752 | * A program that uses a dynamic linker contains a dylinker_command to identify | |
753 | * the name of the dynamic linker (LC_LOAD_DYLINKER). And a dynamic linker | |
754 | * contains a dylinker_command to identify the dynamic linker (LC_ID_DYLINKER). | |
755 | * A file can have at most one of these. | |
316670eb A |
756 | * This struct is also used for the LC_DYLD_ENVIRONMENT load command and |
757 | * contains string for dyld to treat like environment variable. | |
1c79356b A |
758 | */ |
759 | struct dylinker_command { | |
316670eb A |
760 | uint32_t cmd; /* LC_ID_DYLINKER, LC_LOAD_DYLINKER or |
761 | LC_DYLD_ENVIRONMENT */ | |
2d21ac55 | 762 | uint32_t cmdsize; /* includes pathname string */ |
1c79356b A |
763 | union lc_str name; /* dynamic linker's path name */ |
764 | }; | |
765 | ||
766 | /* | |
767 | * Thread commands contain machine-specific data structures suitable for | |
768 | * use in the thread state primitives. The machine specific data structures | |
769 | * follow the struct thread_command as follows. | |
770 | * Each flavor of machine specific data structure is preceded by an unsigned | |
2d21ac55 | 771 | * long constant for the flavor of that data structure, an uint32_t |
1c79356b A |
772 | * that is the count of longs of the size of the state data structure and then |
773 | * the state data structure follows. This triple may be repeated for many | |
774 | * flavors. The constants for the flavors, counts and state data structure | |
775 | * definitions are expected to be in the header file <machine/thread_status.h>. | |
776 | * These machine specific data structures sizes must be multiples of | |
2d21ac55 | 777 | * 4 bytes The cmdsize reflects the total size of the thread_command |
1c79356b A |
778 | * and all of the sizes of the constants for the flavors, counts and state |
779 | * data structures. | |
780 | * | |
781 | * For executable objects that are unix processes there will be one | |
782 | * thread_command (cmd == LC_UNIXTHREAD) created for it by the link-editor. | |
783 | * This is the same as a LC_THREAD, except that a stack is automatically | |
784 | * created (based on the shell's limit for the stack size). Command arguments | |
785 | * and environment variables are copied onto that stack. | |
786 | */ | |
787 | struct thread_command { | |
2d21ac55 A |
788 | uint32_t cmd; /* LC_THREAD or LC_UNIXTHREAD */ |
789 | uint32_t cmdsize; /* total size of this command */ | |
790 | /* uint32_t flavor flavor of thread state */ | |
791 | /* uint32_t count count of longs in thread state */ | |
1c79356b A |
792 | /* struct XXX_thread_state state thread state for this flavor */ |
793 | /* ... */ | |
794 | }; | |
795 | ||
2d21ac55 A |
796 | /* |
797 | * The routines command contains the address of the dynamic shared library | |
798 | * initialization routine and an index into the module table for the module | |
799 | * that defines the routine. Before any modules are used from the library the | |
800 | * dynamic linker fully binds the module that defines the initialization routine | |
801 | * and then calls it. This gets called before any module initialization | |
802 | * routines (used for C++ static constructors) in the library. | |
803 | */ | |
804 | struct routines_command { /* for 32-bit architectures */ | |
805 | uint32_t cmd; /* LC_ROUTINES */ | |
806 | uint32_t cmdsize; /* total size of this command */ | |
807 | uint32_t init_address; /* address of initialization routine */ | |
808 | uint32_t init_module; /* index into the module table that */ | |
809 | /* the init routine is defined in */ | |
810 | uint32_t reserved1; | |
811 | uint32_t reserved2; | |
812 | uint32_t reserved3; | |
813 | uint32_t reserved4; | |
814 | uint32_t reserved5; | |
815 | uint32_t reserved6; | |
816 | }; | |
817 | ||
818 | /* | |
819 | * The 64-bit routines command. Same use as above. | |
820 | */ | |
821 | struct routines_command_64 { /* for 64-bit architectures */ | |
822 | uint32_t cmd; /* LC_ROUTINES_64 */ | |
823 | uint32_t cmdsize; /* total size of this command */ | |
824 | uint64_t init_address; /* address of initialization routine */ | |
825 | uint64_t init_module; /* index into the module table that */ | |
826 | /* the init routine is defined in */ | |
827 | uint64_t reserved1; | |
828 | uint64_t reserved2; | |
829 | uint64_t reserved3; | |
830 | uint64_t reserved4; | |
831 | uint64_t reserved5; | |
832 | uint64_t reserved6; | |
833 | }; | |
834 | ||
1c79356b A |
835 | /* |
836 | * The symtab_command contains the offsets and sizes of the link-edit 4.3BSD | |
837 | * "stab" style symbol table information as described in the header files | |
838 | * <nlist.h> and <stab.h>. | |
839 | */ | |
840 | struct symtab_command { | |
2d21ac55 A |
841 | uint32_t cmd; /* LC_SYMTAB */ |
842 | uint32_t cmdsize; /* sizeof(struct symtab_command) */ | |
843 | uint32_t symoff; /* symbol table offset */ | |
844 | uint32_t nsyms; /* number of symbol table entries */ | |
845 | uint32_t stroff; /* string table offset */ | |
846 | uint32_t strsize; /* string table size in bytes */ | |
1c79356b A |
847 | }; |
848 | ||
849 | /* | |
850 | * This is the second set of the symbolic information which is used to support | |
2d21ac55 | 851 | * the data structures for the dynamically link editor. |
1c79356b A |
852 | * |
853 | * The original set of symbolic information in the symtab_command which contains | |
854 | * the symbol and string tables must also be present when this load command is | |
855 | * present. When this load command is present the symbol table is organized | |
856 | * into three groups of symbols: | |
857 | * local symbols (static and debugging symbols) - grouped by module | |
858 | * defined external symbols - grouped by module (sorted by name if not lib) | |
2d21ac55 A |
859 | * undefined external symbols (sorted by name if MH_BINDATLOAD is not set, |
860 | * and in order the were seen by the static | |
861 | * linker if MH_BINDATLOAD is set) | |
1c79356b A |
862 | * In this load command there are offsets and counts to each of the three groups |
863 | * of symbols. | |
864 | * | |
865 | * This load command contains a the offsets and sizes of the following new | |
866 | * symbolic information tables: | |
867 | * table of contents | |
868 | * module table | |
869 | * reference symbol table | |
870 | * indirect symbol table | |
871 | * The first three tables above (the table of contents, module table and | |
2d21ac55 | 872 | * reference symbol table) are only present if the file is a dynamically linked |
1c79356b A |
873 | * shared library. For executable and object modules, which are files |
874 | * containing only one module, the information that would be in these three | |
875 | * tables is determined as follows: | |
876 | * table of contents - the defined external symbols are sorted by name | |
877 | * module table - the file contains only one module so everything in the | |
878 | * file is part of the module. | |
879 | * reference symbol table - is the defined and undefined external symbols | |
880 | * | |
2d21ac55 | 881 | * For dynamically linked shared library files this load command also contains |
1c79356b A |
882 | * offsets and sizes to the pool of relocation entries for all sections |
883 | * separated into two groups: | |
884 | * external relocation entries | |
885 | * local relocation entries | |
886 | * For executable and object modules the relocation entries continue to hang | |
887 | * off the section structures. | |
888 | */ | |
889 | struct dysymtab_command { | |
2d21ac55 A |
890 | uint32_t cmd; /* LC_DYSYMTAB */ |
891 | uint32_t cmdsize; /* sizeof(struct dysymtab_command) */ | |
1c79356b A |
892 | |
893 | /* | |
894 | * The symbols indicated by symoff and nsyms of the LC_SYMTAB load command | |
895 | * are grouped into the following three groups: | |
896 | * local symbols (further grouped by the module they are from) | |
897 | * defined external symbols (further grouped by the module they are from) | |
898 | * undefined symbols | |
899 | * | |
900 | * The local symbols are used only for debugging. The dynamic binding | |
901 | * process may have to use them to indicate to the debugger the local | |
902 | * symbols for a module that is being bound. | |
903 | * | |
904 | * The last two groups are used by the dynamic binding process to do the | |
905 | * binding (indirectly through the module table and the reference symbol | |
2d21ac55 | 906 | * table when this is a dynamically linked shared library file). |
1c79356b | 907 | */ |
2d21ac55 A |
908 | uint32_t ilocalsym; /* index to local symbols */ |
909 | uint32_t nlocalsym; /* number of local symbols */ | |
1c79356b | 910 | |
2d21ac55 A |
911 | uint32_t iextdefsym;/* index to externally defined symbols */ |
912 | uint32_t nextdefsym;/* number of externally defined symbols */ | |
1c79356b | 913 | |
2d21ac55 A |
914 | uint32_t iundefsym; /* index to undefined symbols */ |
915 | uint32_t nundefsym; /* number of undefined symbols */ | |
1c79356b A |
916 | |
917 | /* | |
918 | * For the for the dynamic binding process to find which module a symbol | |
919 | * is defined in the table of contents is used (analogous to the ranlib | |
920 | * structure in an archive) which maps defined external symbols to modules | |
2d21ac55 | 921 | * they are defined in. This exists only in a dynamically linked shared |
1c79356b A |
922 | * library file. For executable and object modules the defined external |
923 | * symbols are sorted by name and is use as the table of contents. | |
924 | */ | |
2d21ac55 A |
925 | uint32_t tocoff; /* file offset to table of contents */ |
926 | uint32_t ntoc; /* number of entries in table of contents */ | |
1c79356b A |
927 | |
928 | /* | |
929 | * To support dynamic binding of "modules" (whole object files) the symbol | |
930 | * table must reflect the modules that the file was created from. This is | |
931 | * done by having a module table that has indexes and counts into the merged | |
932 | * tables for each module. The module structure that these two entries | |
2d21ac55 | 933 | * refer to is described below. This exists only in a dynamically linked |
1c79356b A |
934 | * shared library file. For executable and object modules the file only |
935 | * contains one module so everything in the file belongs to the module. | |
936 | */ | |
2d21ac55 A |
937 | uint32_t modtaboff; /* file offset to module table */ |
938 | uint32_t nmodtab; /* number of module table entries */ | |
1c79356b A |
939 | |
940 | /* | |
941 | * To support dynamic module binding the module structure for each module | |
942 | * indicates the external references (defined and undefined) each module | |
943 | * makes. For each module there is an offset and a count into the | |
944 | * reference symbol table for the symbols that the module references. | |
2d21ac55 | 945 | * This exists only in a dynamically linked shared library file. For |
1c79356b A |
946 | * executable and object modules the defined external symbols and the |
947 | * undefined external symbols indicates the external references. | |
948 | */ | |
2d21ac55 A |
949 | uint32_t extrefsymoff; /* offset to referenced symbol table */ |
950 | uint32_t nextrefsyms; /* number of referenced symbol table entries */ | |
1c79356b A |
951 | |
952 | /* | |
953 | * The sections that contain "symbol pointers" and "routine stubs" have | |
954 | * indexes and (implied counts based on the size of the section and fixed | |
955 | * size of the entry) into the "indirect symbol" table for each pointer | |
956 | * and stub. For every section of these two types the index into the | |
957 | * indirect symbol table is stored in the section header in the field | |
958 | * reserved1. An indirect symbol table entry is simply a 32bit index into | |
959 | * the symbol table to the symbol that the pointer or stub is referring to. | |
960 | * The indirect symbol table is ordered to match the entries in the section. | |
961 | */ | |
2d21ac55 A |
962 | uint32_t indirectsymoff; /* file offset to the indirect symbol table */ |
963 | uint32_t nindirectsyms; /* number of indirect symbol table entries */ | |
1c79356b A |
964 | |
965 | /* | |
966 | * To support relocating an individual module in a library file quickly the | |
967 | * external relocation entries for each module in the library need to be | |
968 | * accessed efficiently. Since the relocation entries can't be accessed | |
969 | * through the section headers for a library file they are separated into | |
970 | * groups of local and external entries further grouped by module. In this | |
971 | * case the presents of this load command who's extreloff, nextrel, | |
972 | * locreloff and nlocrel fields are non-zero indicates that the relocation | |
973 | * entries of non-merged sections are not referenced through the section | |
974 | * structures (and the reloff and nreloc fields in the section headers are | |
975 | * set to zero). | |
976 | * | |
977 | * Since the relocation entries are not accessed through the section headers | |
978 | * this requires the r_address field to be something other than a section | |
979 | * offset to identify the item to be relocated. In this case r_address is | |
980 | * set to the offset from the vmaddr of the first LC_SEGMENT command. | |
2d21ac55 A |
981 | * For MH_SPLIT_SEGS images r_address is set to the the offset from the |
982 | * vmaddr of the first read-write LC_SEGMENT command. | |
1c79356b A |
983 | * |
984 | * The relocation entries are grouped by module and the module table | |
985 | * entries have indexes and counts into them for the group of external | |
986 | * relocation entries for that the module. | |
987 | * | |
988 | * For sections that are merged across modules there must not be any | |
989 | * remaining external relocation entries for them (for merged sections | |
990 | * remaining relocation entries must be local). | |
991 | */ | |
2d21ac55 A |
992 | uint32_t extreloff; /* offset to external relocation entries */ |
993 | uint32_t nextrel; /* number of external relocation entries */ | |
1c79356b A |
994 | |
995 | /* | |
996 | * All the local relocation entries are grouped together (they are not | |
997 | * grouped by their module since they are only used if the object is moved | |
998 | * from it staticly link edited address). | |
999 | */ | |
2d21ac55 A |
1000 | uint32_t locreloff; /* offset to local relocation entries */ |
1001 | uint32_t nlocrel; /* number of local relocation entries */ | |
1c79356b A |
1002 | |
1003 | }; | |
1004 | ||
1005 | /* | |
1006 | * An indirect symbol table entry is simply a 32bit index into the symbol table | |
1007 | * to the symbol that the pointer or stub is refering to. Unless it is for a | |
1008 | * non-lazy symbol pointer section for a defined symbol which strip(1) as | |
1009 | * removed. In which case it has the value INDIRECT_SYMBOL_LOCAL. If the | |
1010 | * symbol was also absolute INDIRECT_SYMBOL_ABS is or'ed with that. | |
1011 | */ | |
1012 | #define INDIRECT_SYMBOL_LOCAL 0x80000000 | |
1013 | #define INDIRECT_SYMBOL_ABS 0x40000000 | |
1014 | ||
1015 | ||
1016 | /* a table of contents entry */ | |
1017 | struct dylib_table_of_contents { | |
2d21ac55 | 1018 | uint32_t symbol_index; /* the defined external symbol |
1c79356b | 1019 | (index into the symbol table) */ |
2d21ac55 | 1020 | uint32_t module_index; /* index into the module table this symbol |
1c79356b A |
1021 | is defined in */ |
1022 | }; | |
1023 | ||
1024 | /* a module table entry */ | |
1025 | struct dylib_module { | |
2d21ac55 | 1026 | uint32_t module_name; /* the module name (index into string table) */ |
1c79356b | 1027 | |
2d21ac55 A |
1028 | uint32_t iextdefsym; /* index into externally defined symbols */ |
1029 | uint32_t nextdefsym; /* number of externally defined symbols */ | |
1030 | uint32_t irefsym; /* index into reference symbol table */ | |
1031 | uint32_t nrefsym; /* number of reference symbol table entries */ | |
1032 | uint32_t ilocalsym; /* index into symbols for local symbols */ | |
1033 | uint32_t nlocalsym; /* number of local symbols */ | |
1c79356b | 1034 | |
2d21ac55 A |
1035 | uint32_t iextrel; /* index into external relocation entries */ |
1036 | uint32_t nextrel; /* number of external relocation entries */ | |
1c79356b | 1037 | |
2d21ac55 A |
1038 | uint32_t iinit_iterm; /* low 16 bits are the index into the init |
1039 | section, high 16 bits are the index into | |
1040 | the term section */ | |
1041 | uint32_t ninit_nterm; /* low 16 bits are the number of init section | |
1042 | entries, high 16 bits are the number of | |
1043 | term section entries */ | |
1c79356b | 1044 | |
2d21ac55 | 1045 | uint32_t /* for this module address of the start of */ |
1c79356b | 1046 | objc_module_info_addr; /* the (__OBJC,__module_info) section */ |
2d21ac55 | 1047 | uint32_t /* for this module size of */ |
1c79356b A |
1048 | objc_module_info_size; /* the (__OBJC,__module_info) section */ |
1049 | }; | |
1050 | ||
91447636 A |
1051 | /* a 64-bit module table entry */ |
1052 | struct dylib_module_64 { | |
2d21ac55 | 1053 | uint32_t module_name; /* the module name (index into string table) */ |
91447636 | 1054 | |
2d21ac55 A |
1055 | uint32_t iextdefsym; /* index into externally defined symbols */ |
1056 | uint32_t nextdefsym; /* number of externally defined symbols */ | |
1057 | uint32_t irefsym; /* index into reference symbol table */ | |
1058 | uint32_t nrefsym; /* number of reference symbol table entries */ | |
1059 | uint32_t ilocalsym; /* index into symbols for local symbols */ | |
1060 | uint32_t nlocalsym; /* number of local symbols */ | |
91447636 | 1061 | |
2d21ac55 A |
1062 | uint32_t iextrel; /* index into external relocation entries */ |
1063 | uint32_t nextrel; /* number of external relocation entries */ | |
91447636 | 1064 | |
2d21ac55 | 1065 | uint32_t iinit_iterm; /* low 16 bits are the index into the init |
91447636 A |
1066 | section, high 16 bits are the index into |
1067 | the term section */ | |
2d21ac55 A |
1068 | uint32_t ninit_nterm; /* low 16 bits are the number of init section |
1069 | entries, high 16 bits are the number of | |
1070 | term section entries */ | |
1071 | ||
1072 | uint32_t /* for this module size of */ | |
1073 | objc_module_info_size; /* the (__OBJC,__module_info) section */ | |
1074 | uint64_t /* for this module address of the start of */ | |
1075 | objc_module_info_addr; /* the (__OBJC,__module_info) section */ | |
91447636 A |
1076 | }; |
1077 | ||
1c79356b A |
1078 | /* |
1079 | * The entries in the reference symbol table are used when loading the module | |
1080 | * (both by the static and dynamic link editors) and if the module is unloaded | |
1081 | * or replaced. Therefore all external symbols (defined and undefined) are | |
1082 | * listed in the module's reference table. The flags describe the type of | |
1083 | * reference that is being made. The constants for the flags are defined in | |
1084 | * <mach-o/nlist.h> as they are also used for symbol table entries. | |
1085 | */ | |
1086 | struct dylib_reference { | |
2d21ac55 | 1087 | uint32_t isym:24, /* index into the symbol table */ |
1c79356b A |
1088 | flags:8; /* flags to indicate the type of reference */ |
1089 | }; | |
1090 | ||
2d21ac55 A |
1091 | /* |
1092 | * The twolevel_hints_command contains the offset and number of hints in the | |
1093 | * two-level namespace lookup hints table. | |
1094 | */ | |
1095 | struct twolevel_hints_command { | |
1096 | uint32_t cmd; /* LC_TWOLEVEL_HINTS */ | |
1097 | uint32_t cmdsize; /* sizeof(struct twolevel_hints_command) */ | |
1098 | uint32_t offset; /* offset to the hint table */ | |
1099 | uint32_t nhints; /* number of hints in the hint table */ | |
1100 | }; | |
1101 | ||
1102 | /* | |
1103 | * The entries in the two-level namespace lookup hints table are twolevel_hint | |
1104 | * structs. These provide hints to the dynamic link editor where to start | |
1105 | * looking for an undefined symbol in a two-level namespace image. The | |
1106 | * isub_image field is an index into the sub-images (sub-frameworks and | |
1107 | * sub-umbrellas list) that made up the two-level image that the undefined | |
1108 | * symbol was found in when it was built by the static link editor. If | |
1109 | * isub-image is 0 the the symbol is expected to be defined in library and not | |
1110 | * in the sub-images. If isub-image is non-zero it is an index into the array | |
1111 | * of sub-images for the umbrella with the first index in the sub-images being | |
1112 | * 1. The array of sub-images is the ordered list of sub-images of the umbrella | |
1113 | * that would be searched for a symbol that has the umbrella recorded as its | |
1114 | * primary library. The table of contents index is an index into the | |
1115 | * library's table of contents. This is used as the starting point of the | |
1116 | * binary search or a directed linear search. | |
1117 | */ | |
1118 | struct twolevel_hint { | |
1119 | uint32_t | |
1120 | isub_image:8, /* index into the sub images */ | |
1121 | itoc:24; /* index into the table of contents */ | |
1122 | }; | |
1123 | ||
1124 | /* | |
1125 | * The prebind_cksum_command contains the value of the original check sum for | |
1126 | * prebound files or zero. When a prebound file is first created or modified | |
1127 | * for other than updating its prebinding information the value of the check sum | |
1128 | * is set to zero. When the file has it prebinding re-done and if the value of | |
1129 | * the check sum is zero the original check sum is calculated and stored in | |
1130 | * cksum field of this load command in the output file. If when the prebinding | |
1131 | * is re-done and the cksum field is non-zero it is left unchanged from the | |
1132 | * input file. | |
1133 | */ | |
1134 | struct prebind_cksum_command { | |
1135 | uint32_t cmd; /* LC_PREBIND_CKSUM */ | |
1136 | uint32_t cmdsize; /* sizeof(struct prebind_cksum_command) */ | |
1137 | uint32_t cksum; /* the check sum or zero */ | |
1138 | }; | |
1139 | ||
1140 | /* | |
1141 | * The uuid load command contains a single 128-bit unique random number that | |
1142 | * identifies an object produced by the static link editor. | |
1143 | */ | |
1144 | struct uuid_command { | |
1145 | uint32_t cmd; /* LC_UUID */ | |
1146 | uint32_t cmdsize; /* sizeof(struct uuid_command) */ | |
1147 | uint8_t uuid[16]; /* the 128-bit uuid */ | |
1148 | }; | |
1149 | ||
1150 | /* | |
1151 | * The rpath_command contains a path which at runtime should be added to | |
1152 | * the current run path used to find @rpath prefixed dylibs. | |
1153 | */ | |
1154 | struct rpath_command { | |
1155 | uint32_t cmd; /* LC_RPATH */ | |
1156 | uint32_t cmdsize; /* includes string */ | |
1157 | union lc_str path; /* path to add to run path */ | |
1158 | }; | |
1159 | ||
1160 | /* | |
1161 | * The linkedit_data_command contains the offsets and sizes of a blob | |
1162 | * of data in the __LINKEDIT segment. | |
1163 | */ | |
1164 | struct linkedit_data_command { | |
316670eb A |
1165 | uint32_t cmd; /* LC_CODE_SIGNATURE, LC_SEGMENT_SPLIT_INFO, |
1166 | LC_FUNCTION_STARTS, LC_DATA_IN_CODE, | |
3e170ce0 A |
1167 | LC_DYLIB_CODE_SIGN_DRS or |
1168 | LC_LINKER_OPTIMIZATION_HINT. */ | |
2d21ac55 A |
1169 | uint32_t cmdsize; /* sizeof(struct linkedit_data_command) */ |
1170 | uint32_t dataoff; /* file offset of data in __LINKEDIT segment */ | |
1171 | uint32_t datasize; /* file size of data in __LINKEDIT segment */ | |
1172 | }; | |
1173 | ||
593a1d5f A |
1174 | /* |
1175 | * The encryption_info_command contains the file offset and size of an | |
1176 | * of an encrypted segment. | |
1177 | */ | |
1178 | struct encryption_info_command { | |
1179 | uint32_t cmd; /* LC_ENCRYPTION_INFO */ | |
1180 | uint32_t cmdsize; /* sizeof(struct encryption_info_command) */ | |
1181 | uint32_t cryptoff; /* file offset of encrypted range */ | |
1182 | uint32_t cryptsize; /* file size of encrypted range */ | |
1183 | uint32_t cryptid; /* which enryption system, | |
1184 | 0 means not-encrypted yet */ | |
1185 | }; | |
1186 | ||
39236c6e A |
1187 | /* |
1188 | * The encryption_info_command_64 contains the file offset and size of an | |
3e170ce0 | 1189 | * of an encrypted segment (for use in x86_64 targets). |
39236c6e A |
1190 | */ |
1191 | struct encryption_info_command_64 { | |
1192 | uint32_t cmd; /* LC_ENCRYPTION_INFO_64 */ | |
1193 | uint32_t cmdsize; /* sizeof(struct encryption_info_command_64) */ | |
1194 | uint32_t cryptoff; /* file offset of encrypted range */ | |
1195 | uint32_t cryptsize; /* file size of encrypted range */ | |
1196 | uint32_t cryptid; /* which enryption system, | |
1197 | 0 means not-encrypted yet */ | |
1198 | uint32_t pad; /* padding to make this struct's size a multiple | |
1199 | of 8 bytes */ | |
1200 | }; | |
1201 | ||
316670eb A |
1202 | /* |
1203 | * The version_min_command contains the min OS version on which this | |
1204 | * binary was built to run. | |
1205 | */ | |
1206 | struct version_min_command { | |
1207 | uint32_t cmd; /* LC_VERSION_MIN_MACOSX or | |
5ba3f43e A |
1208 | LC_VERSION_MIN_IPHONEOS or |
1209 | LC_VERSION_MIN_WATCHOS or | |
1210 | LC_VERSION_MIN_TVOS */ | |
316670eb A |
1211 | uint32_t cmdsize; /* sizeof(struct min_version_command) */ |
1212 | uint32_t version; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ | |
1213 | uint32_t sdk; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ | |
1214 | }; | |
1215 | ||
5ba3f43e A |
1216 | /* |
1217 | * The build_version_command contains the min OS version on which this | |
1218 | * binary was built to run for its platform. The list of known platforms and | |
1219 | * tool values following it. | |
1220 | */ | |
1221 | struct build_version_command { | |
1222 | uint32_t cmd; /* LC_BUILD_VERSION */ | |
1223 | uint32_t cmdsize; /* sizeof(struct build_version_command) plus */ | |
1224 | /* ntools * sizeof(struct build_tool_version) */ | |
1225 | uint32_t platform; /* platform */ | |
1226 | uint32_t minos; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ | |
1227 | uint32_t sdk; /* X.Y.Z is encoded in nibbles xxxx.yy.zz */ | |
1228 | uint32_t ntools; /* number of tool entries following this */ | |
1229 | }; | |
1230 | ||
1231 | struct build_tool_version { | |
1232 | uint32_t tool; /* enum for the tool */ | |
1233 | uint32_t version; /* version number of the tool */ | |
1234 | }; | |
1235 | ||
1236 | /* Known values for the platform field above. */ | |
1237 | #define PLATFORM_MACOS 1 | |
1238 | #define PLATFORM_IOS 2 | |
1239 | #define PLATFORM_TVOS 3 | |
1240 | #define PLATFORM_WATCHOS 4 | |
1241 | ||
1242 | /* Known values for the tool field above. */ | |
1243 | #define TOOL_CLANG 1 | |
1244 | #define TOOL_SWIFT 2 | |
1245 | #define TOOL_LD 3 | |
1246 | ||
b0d623f7 A |
1247 | /* |
1248 | * The dyld_info_command contains the file offsets and sizes of | |
1249 | * the new compressed form of the information dyld needs to | |
1250 | * load the image. This information is used by dyld on Mac OS X | |
1251 | * 10.6 and later. All information pointed to by this command | |
1252 | * is encoded using byte streams, so no endian swapping is needed | |
1253 | * to interpret it. | |
1254 | */ | |
1255 | struct dyld_info_command { | |
1256 | uint32_t cmd; /* LC_DYLD_INFO or LC_DYLD_INFO_ONLY */ | |
1257 | uint32_t cmdsize; /* sizeof(struct dyld_info_command) */ | |
1258 | ||
1259 | /* | |
1260 | * Dyld rebases an image whenever dyld loads it at an address different | |
1261 | * from its preferred address. The rebase information is a stream | |
1262 | * of byte sized opcodes whose symbolic names start with REBASE_OPCODE_. | |
1263 | * Conceptually the rebase information is a table of tuples: | |
1264 | * <seg-index, seg-offset, type> | |
1265 | * The opcodes are a compressed way to encode the table by only | |
1266 | * encoding when a column changes. In addition simple patterns | |
1267 | * like "every n'th offset for m times" can be encoded in a few | |
1268 | * bytes. | |
1269 | */ | |
1270 | uint32_t rebase_off; /* file offset to rebase info */ | |
1271 | uint32_t rebase_size; /* size of rebase info */ | |
1272 | ||
1273 | /* | |
1274 | * Dyld binds an image during the loading process, if the image | |
1275 | * requires any pointers to be initialized to symbols in other images. | |
316670eb | 1276 | * The bind information is a stream of byte sized |
b0d623f7 A |
1277 | * opcodes whose symbolic names start with BIND_OPCODE_. |
1278 | * Conceptually the bind information is a table of tuples: | |
1279 | * <seg-index, seg-offset, type, symbol-library-ordinal, symbol-name, addend> | |
1280 | * The opcodes are a compressed way to encode the table by only | |
1281 | * encoding when a column changes. In addition simple patterns | |
1282 | * like for runs of pointers initialzed to the same value can be | |
1283 | * encoded in a few bytes. | |
1284 | */ | |
1285 | uint32_t bind_off; /* file offset to binding info */ | |
1286 | uint32_t bind_size; /* size of binding info */ | |
1287 | ||
1288 | /* | |
1289 | * Some C++ programs require dyld to unique symbols so that all | |
1290 | * images in the process use the same copy of some code/data. | |
1291 | * This step is done after binding. The content of the weak_bind | |
1292 | * info is an opcode stream like the bind_info. But it is sorted | |
1293 | * alphabetically by symbol name. This enable dyld to walk | |
1294 | * all images with weak binding information in order and look | |
1295 | * for collisions. If there are no collisions, dyld does | |
1296 | * no updating. That means that some fixups are also encoded | |
1297 | * in the bind_info. For instance, all calls to "operator new" | |
1298 | * are first bound to libstdc++.dylib using the information | |
1299 | * in bind_info. Then if some image overrides operator new | |
1300 | * that is detected when the weak_bind information is processed | |
1301 | * and the call to operator new is then rebound. | |
1302 | */ | |
1303 | uint32_t weak_bind_off; /* file offset to weak binding info */ | |
1304 | uint32_t weak_bind_size; /* size of weak binding info */ | |
1305 | ||
1306 | /* | |
1307 | * Some uses of external symbols do not need to be bound immediately. | |
1308 | * Instead they can be lazily bound on first use. The lazy_bind | |
1309 | * are contains a stream of BIND opcodes to bind all lazy symbols. | |
1310 | * Normal use is that dyld ignores the lazy_bind section when | |
1311 | * loading an image. Instead the static linker arranged for the | |
1312 | * lazy pointer to initially point to a helper function which | |
1313 | * pushes the offset into the lazy_bind area for the symbol | |
1314 | * needing to be bound, then jumps to dyld which simply adds | |
1315 | * the offset to lazy_bind_off to get the information on what | |
1316 | * to bind. | |
1317 | */ | |
1318 | uint32_t lazy_bind_off; /* file offset to lazy binding info */ | |
1319 | uint32_t lazy_bind_size; /* size of lazy binding infs */ | |
1320 | ||
1321 | /* | |
1322 | * The symbols exported by a dylib are encoded in a trie. This | |
1323 | * is a compact representation that factors out common prefixes. | |
1324 | * It also reduces LINKEDIT pages in RAM because it encodes all | |
1325 | * information (name, address, flags) in one small, contiguous range. | |
1326 | * The export area is a stream of nodes. The first node sequentially | |
1327 | * is the start node for the trie. | |
1328 | * | |
316670eb | 1329 | * Nodes for a symbol start with a uleb128 that is the length of |
b0d623f7 | 1330 | * the exported symbol information for the string so far. |
316670eb A |
1331 | * If there is no exported symbol, the node starts with a zero byte. |
1332 | * If there is exported info, it follows the length. | |
1333 | * | |
1334 | * First is a uleb128 containing flags. Normally, it is followed by | |
1335 | * a uleb128 encoded offset which is location of the content named | |
1336 | * by the symbol from the mach_header for the image. If the flags | |
1337 | * is EXPORT_SYMBOL_FLAGS_REEXPORT, then following the flags is | |
1338 | * a uleb128 encoded library ordinal, then a zero terminated | |
1339 | * UTF8 string. If the string is zero length, then the symbol | |
1340 | * is re-export from the specified dylib with the same name. | |
1341 | * If the flags is EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER, then following | |
1342 | * the flags is two uleb128s: the stub offset and the resolver offset. | |
1343 | * The stub is used by non-lazy pointers. The resolver is used | |
1344 | * by lazy pointers and must be called to get the actual address to use. | |
b0d623f7 | 1345 | * |
316670eb A |
1346 | * After the optional exported symbol information is a byte of |
1347 | * how many edges (0-255) that this node has leaving it, | |
1348 | * followed by each edge. | |
1349 | * Each edge is a zero terminated UTF8 of the addition chars | |
b0d623f7 A |
1350 | * in the symbol, followed by a uleb128 offset for the node that |
1351 | * edge points to. | |
1352 | * | |
1353 | */ | |
1354 | uint32_t export_off; /* file offset to lazy binding info */ | |
1355 | uint32_t export_size; /* size of lazy binding infs */ | |
1356 | }; | |
1357 | ||
1358 | /* | |
1359 | * The following are used to encode rebasing information | |
1360 | */ | |
1361 | #define REBASE_TYPE_POINTER 1 | |
1362 | #define REBASE_TYPE_TEXT_ABSOLUTE32 2 | |
1363 | #define REBASE_TYPE_TEXT_PCREL32 3 | |
1364 | ||
1365 | #define REBASE_OPCODE_MASK 0xF0 | |
1366 | #define REBASE_IMMEDIATE_MASK 0x0F | |
1367 | #define REBASE_OPCODE_DONE 0x00 | |
1368 | #define REBASE_OPCODE_SET_TYPE_IMM 0x10 | |
1369 | #define REBASE_OPCODE_SET_SEGMENT_AND_OFFSET_ULEB 0x20 | |
1370 | #define REBASE_OPCODE_ADD_ADDR_ULEB 0x30 | |
1371 | #define REBASE_OPCODE_ADD_ADDR_IMM_SCALED 0x40 | |
1372 | #define REBASE_OPCODE_DO_REBASE_IMM_TIMES 0x50 | |
1373 | #define REBASE_OPCODE_DO_REBASE_ULEB_TIMES 0x60 | |
1374 | #define REBASE_OPCODE_DO_REBASE_ADD_ADDR_ULEB 0x70 | |
1375 | #define REBASE_OPCODE_DO_REBASE_ULEB_TIMES_SKIPPING_ULEB 0x80 | |
1376 | ||
1377 | ||
1378 | /* | |
1379 | * The following are used to encode binding information | |
1380 | */ | |
1381 | #define BIND_TYPE_POINTER 1 | |
1382 | #define BIND_TYPE_TEXT_ABSOLUTE32 2 | |
1383 | #define BIND_TYPE_TEXT_PCREL32 3 | |
1384 | ||
1385 | #define BIND_SPECIAL_DYLIB_SELF 0 | |
1386 | #define BIND_SPECIAL_DYLIB_MAIN_EXECUTABLE -1 | |
1387 | #define BIND_SPECIAL_DYLIB_FLAT_LOOKUP -2 | |
1388 | ||
1389 | #define BIND_SYMBOL_FLAGS_WEAK_IMPORT 0x1 | |
1390 | #define BIND_SYMBOL_FLAGS_NON_WEAK_DEFINITION 0x8 | |
1391 | ||
1392 | #define BIND_OPCODE_MASK 0xF0 | |
1393 | #define BIND_IMMEDIATE_MASK 0x0F | |
1394 | #define BIND_OPCODE_DONE 0x00 | |
1395 | #define BIND_OPCODE_SET_DYLIB_ORDINAL_IMM 0x10 | |
1396 | #define BIND_OPCODE_SET_DYLIB_ORDINAL_ULEB 0x20 | |
1397 | #define BIND_OPCODE_SET_DYLIB_SPECIAL_IMM 0x30 | |
1398 | #define BIND_OPCODE_SET_SYMBOL_TRAILING_FLAGS_IMM 0x40 | |
1399 | #define BIND_OPCODE_SET_TYPE_IMM 0x50 | |
1400 | #define BIND_OPCODE_SET_ADDEND_SLEB 0x60 | |
1401 | #define BIND_OPCODE_SET_SEGMENT_AND_OFFSET_ULEB 0x70 | |
1402 | #define BIND_OPCODE_ADD_ADDR_ULEB 0x80 | |
1403 | #define BIND_OPCODE_DO_BIND 0x90 | |
1404 | #define BIND_OPCODE_DO_BIND_ADD_ADDR_ULEB 0xA0 | |
1405 | #define BIND_OPCODE_DO_BIND_ADD_ADDR_IMM_SCALED 0xB0 | |
1406 | #define BIND_OPCODE_DO_BIND_ULEB_TIMES_SKIPPING_ULEB 0xC0 | |
1407 | ||
1408 | ||
1409 | /* | |
1410 | * The following are used on the flags byte of a terminal node | |
1411 | * in the export information. | |
1412 | */ | |
1413 | #define EXPORT_SYMBOL_FLAGS_KIND_MASK 0x03 | |
1414 | #define EXPORT_SYMBOL_FLAGS_KIND_REGULAR 0x00 | |
1415 | #define EXPORT_SYMBOL_FLAGS_KIND_THREAD_LOCAL 0x01 | |
1416 | #define EXPORT_SYMBOL_FLAGS_WEAK_DEFINITION 0x04 | |
316670eb A |
1417 | #define EXPORT_SYMBOL_FLAGS_REEXPORT 0x08 |
1418 | #define EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER 0x10 | |
b0d623f7 | 1419 | |
3e170ce0 A |
1420 | /* |
1421 | * The linker_option_command contains linker options embedded in object files. | |
1422 | */ | |
1423 | struct linker_option_command { | |
1424 | uint32_t cmd; /* LC_LINKER_OPTION only used in MH_OBJECT filetypes */ | |
1425 | uint32_t cmdsize; | |
1426 | uint32_t count; /* number of strings */ | |
1427 | /* concatenation of zero terminated UTF8 strings. | |
1428 | Zero filled at end to align */ | |
1429 | }; | |
1430 | ||
1c79356b A |
1431 | /* |
1432 | * The symseg_command contains the offset and size of the GNU style | |
1433 | * symbol table information as described in the header file <symseg.h>. | |
1434 | * The symbol roots of the symbol segments must also be aligned properly | |
1435 | * in the file. So the requirement of keeping the offsets aligned to a | |
2d21ac55 | 1436 | * multiple of a 4 bytes translates to the length field of the symbol |
1c79356b A |
1437 | * roots also being a multiple of a long. Also the padding must again be |
1438 | * zeroed. (THIS IS OBSOLETE and no longer supported). | |
1439 | */ | |
1440 | struct symseg_command { | |
2d21ac55 A |
1441 | uint32_t cmd; /* LC_SYMSEG */ |
1442 | uint32_t cmdsize; /* sizeof(struct symseg_command) */ | |
1443 | uint32_t offset; /* symbol segment offset */ | |
1444 | uint32_t size; /* symbol segment size in bytes */ | |
1c79356b A |
1445 | }; |
1446 | ||
1447 | /* | |
1448 | * The ident_command contains a free format string table following the | |
1449 | * ident_command structure. The strings are null terminated and the size of | |
2d21ac55 | 1450 | * the command is padded out with zero bytes to a multiple of 4 bytes/ |
1c79356b A |
1451 | * (THIS IS OBSOLETE and no longer supported). |
1452 | */ | |
1453 | struct ident_command { | |
2d21ac55 A |
1454 | uint32_t cmd; /* LC_IDENT */ |
1455 | uint32_t cmdsize; /* strings that follow this command */ | |
1c79356b A |
1456 | }; |
1457 | ||
1458 | /* | |
1459 | * The fvmfile_command contains a reference to a file to be loaded at the | |
2d21ac55 | 1460 | * specified virtual address. (Presently, this command is reserved for |
1c79356b A |
1461 | * internal use. The kernel ignores this command when loading a program into |
1462 | * memory). | |
1463 | */ | |
1464 | struct fvmfile_command { | |
2d21ac55 A |
1465 | uint32_t cmd; /* LC_FVMFILE */ |
1466 | uint32_t cmdsize; /* includes pathname string */ | |
1c79356b | 1467 | union lc_str name; /* files pathname */ |
2d21ac55 | 1468 | uint32_t header_addr; /* files virtual address */ |
1c79356b A |
1469 | }; |
1470 | ||
316670eb A |
1471 | |
1472 | /* | |
1473 | * The entry_point_command is a replacement for thread_command. | |
1474 | * It is used for main executables to specify the location (file offset) | |
1475 | * of main(). If -stack_size was used at link time, the stacksize | |
1476 | * field will contain the stack size need for the main thread. | |
1477 | */ | |
1478 | struct entry_point_command { | |
1479 | uint32_t cmd; /* LC_MAIN only used in MH_EXECUTE filetypes */ | |
1480 | uint32_t cmdsize; /* 24 */ | |
1481 | uint64_t entryoff; /* file (__TEXT) offset of main() */ | |
1482 | uint64_t stacksize;/* if not zero, initial stack size */ | |
1483 | }; | |
1484 | ||
1485 | ||
1486 | /* | |
1487 | * The source_version_command is an optional load command containing | |
1488 | * the version of the sources used to build the binary. | |
1489 | */ | |
1490 | struct source_version_command { | |
1491 | uint32_t cmd; /* LC_SOURCE_VERSION */ | |
1492 | uint32_t cmdsize; /* 16 */ | |
1493 | uint64_t version; /* A.B.C.D.E packed as a24.b10.c10.d10.e10 */ | |
1494 | }; | |
1495 | ||
1496 | ||
1497 | /* | |
1498 | * The LC_DATA_IN_CODE load commands uses a linkedit_data_command | |
1499 | * to point to an array of data_in_code_entry entries. Each entry | |
39236c6e | 1500 | * describes a range of data in a code section. |
316670eb A |
1501 | */ |
1502 | struct data_in_code_entry { | |
1503 | uint32_t offset; /* from mach_header to start of data range*/ | |
1504 | uint16_t length; /* number of bytes in data range */ | |
1505 | uint16_t kind; /* a DICE_KIND_* value */ | |
1506 | }; | |
39236c6e A |
1507 | #define DICE_KIND_DATA 0x0001 |
1508 | #define DICE_KIND_JUMP_TABLE8 0x0002 | |
1509 | #define DICE_KIND_JUMP_TABLE16 0x0003 | |
1510 | #define DICE_KIND_JUMP_TABLE32 0x0004 | |
1511 | #define DICE_KIND_ABS_JUMP_TABLE32 0x0005 | |
316670eb A |
1512 | |
1513 | ||
1514 | ||
1515 | /* | |
1516 | * Sections of type S_THREAD_LOCAL_VARIABLES contain an array | |
1517 | * of tlv_descriptor structures. | |
1518 | */ | |
1519 | struct tlv_descriptor | |
1520 | { | |
1521 | void* (*thunk)(struct tlv_descriptor*); | |
1522 | unsigned long key; | |
1523 | unsigned long offset; | |
1524 | }; | |
1525 | ||
5ba3f43e A |
1526 | /* |
1527 | * LC_NOTE commands describe a region of arbitrary data included in a Mach-O | |
1528 | * file. Its initial use is to record extra data in MH_CORE files. | |
1529 | */ | |
1530 | struct note_command { | |
1531 | uint32_t cmd; /* LC_NOTE */ | |
1532 | uint32_t cmdsize; /* sizeof(struct note_command) */ | |
1533 | char data_owner[16]; /* owner name for this LC_NOTE */ | |
1534 | uint64_t offset; /* file offset of this data */ | |
1535 | uint64_t size; /* length of data region */ | |
1536 | }; | |
1537 | ||
55e303ae | 1538 | #endif /* _MACHO_LOADER_H_ */ |