1/* Mach-O support for BFD.
2   Copyright 1999, 2000, 2001, 2002, 2003, 2005
3   Free Software Foundation, Inc.
4
5   This file is part of BFD, the Binary File Descriptor library.
6
7   This program is free software; you can redistribute it and/or modify
8   it under the terms of the GNU General Public License as published by
9   the Free Software Foundation; either version 2 of the License, or
10   (at your option) any later version.
11
12   This program is distributed in the hope that it will be useful,
13   but WITHOUT ANY WARRANTY; without even the implied warranty of
14   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
15   GNU General Public License for more details.
16
17   You should have received a copy of the GNU General Public License
18   along with this program; if not, write to the Free Software
19   Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston, MA 02110-1301, USA.  */
20
21#ifndef _BFD_MACH_O_H_
22#define _BFD_MACH_O_H_
23
24#include "bfd.h"
25
26#define BFD_MACH_O_N_STAB  0xe0	/* If any of these bits set, a symbolic debugging entry.  */
27#define BFD_MACH_O_N_PEXT  0x10	/* Private external symbol bit.  */
28#define BFD_MACH_O_N_TYPE  0x0e	/* Mask for the type bits.  */
29#define BFD_MACH_O_N_EXT   0x01	/* External symbol bit, set for external symbols.  */
30#define BFD_MACH_O_N_UNDF  0x00	/* Undefined, n_sect == NO_SECT.  */
31#define BFD_MACH_O_N_ABS   0x02	/* Absolute, n_sect == NO_SECT.  */
32#define BFD_MACH_O_N_SECT  0x0e	/* Defined in section number n_sect.  */
33#define BFD_MACH_O_N_PBUD  0x0c /* Prebound undefined (defined in a dylib).  */
34#define BFD_MACH_O_N_INDR  0x0a	/* Indirect.  */
35
36typedef enum bfd_mach_o_ppc_thread_flavour
37{
38  BFD_MACH_O_PPC_THREAD_STATE = 1,
39  BFD_MACH_O_PPC_FLOAT_STATE = 2,
40  BFD_MACH_O_PPC_EXCEPTION_STATE = 3,
41  BFD_MACH_O_PPC_VECTOR_STATE = 4
42}
43bfd_mach_o_ppc_thread_flavour;
44
45typedef enum bfd_mach_o_i386_thread_flavour
46{
47  BFD_MACH_O_i386_NEW_THREAD_STATE = 1,
48  BFD_MACH_O_i386_FLOAT_STATE = 2,
49  BFD_MACH_O_i386_ISA_PORT_MAP_STATE = 3,
50  BFD_MACH_O_i386_V86_ASSIST_STATE = 4,
51  BFD_MACH_O_i386_REGS_SEGS_STATE = 5,
52  BFD_MACH_O_i386_THREAD_SYSCALL_STATE = 6,
53  BFD_MACH_O_i386_THREAD_STATE_NONE = 7,
54  BFD_MACH_O_i386_SAVED_STATE = 8,
55  BFD_MACH_O_i386_THREAD_STATE = -1,
56  BFD_MACH_O_i386_THREAD_FPSTATE = -2,
57  BFD_MACH_O_i386_THREAD_EXCEPTSTATE = -3,
58  BFD_MACH_O_i386_THREAD_CTHREADSTATE = -4,
59}
60bfd_mach_o_i386_thread_flavour;
61
62#define BFD_MACH_O_LC_REQ_DYLD 0x80000000
63
64typedef enum bfd_mach_o_load_command_type
65{
66  BFD_MACH_O_LC_SEGMENT = 0x1,		/* File segment to be mapped.  */
67  BFD_MACH_O_LC_SYMTAB = 0x2,		/* Link-edit stab symbol table info (obsolete).  */
68  BFD_MACH_O_LC_SYMSEG = 0x3,		/* Link-edit gdb symbol table info.  */
69  BFD_MACH_O_LC_THREAD = 0x4,		/* Thread.  */
70  BFD_MACH_O_LC_UNIXTHREAD = 0x5,	/* UNIX thread (includes a stack).  */
71  BFD_MACH_O_LC_LOADFVMLIB = 0x6,	/* Load a fixed VM shared library.  */
72  BFD_MACH_O_LC_IDFVMLIB = 0x7,		/* Fixed VM shared library id.  */
73  BFD_MACH_O_LC_IDENT = 0x8,		/* Object identification information (obsolete).  */
74  BFD_MACH_O_LC_FVMFILE = 0x9,		/* Fixed VM file inclusion.  */
75  BFD_MACH_O_LC_PREPAGE = 0xa,		/* Prepage command (internal use).  */
76  BFD_MACH_O_LC_DYSYMTAB = 0xb,		/* Dynamic link-edit symbol table info.  */
77  BFD_MACH_O_LC_LOAD_DYLIB = 0xc,	/* Load a dynamically linked shared library.  */
78  BFD_MACH_O_LC_ID_DYLIB = 0xd,		/* Dynamically linked shared lib identification.  */
79  BFD_MACH_O_LC_LOAD_DYLINKER = 0xe,	/* Load a dynamic linker.  */
80  BFD_MACH_O_LC_ID_DYLINKER = 0xf,	/* Dynamic linker identification.  */
81  BFD_MACH_O_LC_PREBOUND_DYLIB = 0x10,	/* Modules prebound for a dynamically.  */
82  BFD_MACH_O_LC_ROUTINES = 0x11,	/* Image routines.  */
83  BFD_MACH_O_LC_SUB_FRAMEWORK = 0x12,	/* Sub framework.  */
84  BFD_MACH_O_LC_SUB_UMBRELLA = 0x13,	/* Sub umbrella.  */
85  BFD_MACH_O_LC_SUB_CLIENT = 0x14,	/* Sub client.  */
86  BFD_MACH_O_LC_SUB_LIBRARY = 0x15,   	/* Sub library.  */
87  BFD_MACH_O_LC_TWOLEVEL_HINTS = 0x16,	/* Two-level namespace lookup hints.  */
88  BFD_MACH_O_LC_PREBIND_CKSUM = 0x17, 	/* Prebind checksum.  */
89  /* Load a dynamically linked shared library that is allowed to be
90       missing (weak).  */
91  BFD_MACH_O_LC_LOAD_WEAK_DYLIB = 0x18
92}
93bfd_mach_o_load_command_type;
94
95typedef enum bfd_mach_o_cpu_type
96{
97  BFD_MACH_O_CPU_TYPE_VAX = 1,
98  BFD_MACH_O_CPU_TYPE_MC680x0 = 6,
99  BFD_MACH_O_CPU_TYPE_I386 = 7,
100  BFD_MACH_O_CPU_TYPE_MIPS = 8,
101  BFD_MACH_O_CPU_TYPE_MC98000 = 10,
102  BFD_MACH_O_CPU_TYPE_HPPA = 11,
103  BFD_MACH_O_CPU_TYPE_ARM = 12,
104  BFD_MACH_O_CPU_TYPE_MC88000 = 13,
105  BFD_MACH_O_CPU_TYPE_SPARC = 14,
106  BFD_MACH_O_CPU_TYPE_I860 = 15,
107  BFD_MACH_O_CPU_TYPE_ALPHA = 16,
108  BFD_MACH_O_CPU_TYPE_POWERPC = 18
109}
110bfd_mach_o_cpu_type;
111
112typedef enum bfd_mach_o_filetype
113{
114  BFD_MACH_O_MH_OBJECT = 1,
115  BFD_MACH_O_MH_EXECUTE = 2,
116  BFD_MACH_O_MH_FVMLIB = 3,
117  BFD_MACH_O_MH_CORE = 4,
118  BFD_MACH_O_MH_PRELOAD = 5,
119  BFD_MACH_O_MH_DYLIB = 6,
120  BFD_MACH_O_MH_DYLINKER = 7,
121  BFD_MACH_O_MH_BUNDLE = 8
122}
123bfd_mach_o_filetype;
124
125/* Constants for the type of a section.  */
126
127typedef enum bfd_mach_o_section_type
128{
129  /* Regular section.  */
130  BFD_MACH_O_S_REGULAR = 0x0,
131
132  /* Zero fill on demand section.  */
133  BFD_MACH_O_S_ZEROFILL = 0x1,
134
135  /* Section with only literal C strings.  */
136  BFD_MACH_O_S_CSTRING_LITERALS = 0x2,
137
138  /* Section with only 4 byte literals.  */
139  BFD_MACH_O_S_4BYTE_LITERALS = 0x3,
140
141  /* Section with only 8 byte literals.  */
142  BFD_MACH_O_S_8BYTE_LITERALS = 0x4,
143
144  /* Section with only pointers to literals.  */
145  BFD_MACH_O_S_LITERAL_POINTERS = 0x5,
146
147  /* For the two types of symbol pointers sections and the symbol stubs
148     section they have indirect symbol table entries.  For each of the
149     entries in the section the indirect symbol table entries, in
150     corresponding order in the indirect symbol table, start at the index
151     stored in the reserved1 field of the section structure.  Since the
152     indirect symbol table entries correspond to the entries in the
153     section the number of indirect symbol table entries is inferred from
154     the size of the section divided by the size of the entries in the
155     section.  For symbol pointers sections the size of the entries in
156     the section is 4 bytes and for symbol stubs sections the byte size
157     of the stubs is stored in the reserved2 field of the section
158     structure.  */
159
160  /* Section with only non-lazy symbol pointers.  */
161  BFD_MACH_O_S_NON_LAZY_SYMBOL_POINTERS = 0x6,
162
163  /* Section with only lazy symbol pointers.  */
164  BFD_MACH_O_S_LAZY_SYMBOL_POINTERS = 0x7,
165
166  /* Section with only symbol stubs, byte size of stub in the reserved2 field.  */
167  BFD_MACH_O_S_SYMBOL_STUBS = 0x8,
168
169  /* Section with only function pointers for initialization.  */
170  BFD_MACH_O_S_MOD_INIT_FUNC_POINTERS = 0x9
171}
172bfd_mach_o_section_type;
173
174typedef unsigned long bfd_mach_o_cpu_subtype;
175
176typedef struct bfd_mach_o_header
177{
178  unsigned long magic;
179  unsigned long cputype;
180  unsigned long cpusubtype;
181  unsigned long filetype;
182  unsigned long ncmds;
183  unsigned long sizeofcmds;
184  unsigned long flags;
185  enum bfd_endian byteorder;
186}
187bfd_mach_o_header;
188
189typedef struct bfd_mach_o_section
190{
191  asection *bfdsection;
192  char sectname[16 + 1];
193  char segname[16 + 1];
194  bfd_vma addr;
195  bfd_vma size;
196  bfd_vma offset;
197  unsigned long align;
198  bfd_vma reloff;
199  unsigned long nreloc;
200  unsigned long flags;
201  unsigned long reserved1;
202  unsigned long reserved2;
203}
204bfd_mach_o_section;
205
206typedef struct bfd_mach_o_segment_command
207{
208  char segname[16];
209  bfd_vma vmaddr;
210  bfd_vma vmsize;
211  bfd_vma fileoff;
212  unsigned long filesize;
213  unsigned long nsects;
214  unsigned long flags;
215  bfd_mach_o_section *sections;
216  asection *segment;
217}
218bfd_mach_o_segment_command;
219
220typedef struct bfd_mach_o_symtab_command
221{
222  unsigned long symoff;
223  unsigned long nsyms;
224  unsigned long stroff;
225  unsigned long strsize;
226  asymbol *symbols;
227  char *strtab;
228  asection *stabs_segment;
229  asection *stabstr_segment;
230}
231bfd_mach_o_symtab_command;
232
233/* This is the second set of the symbolic information which is used to support
234   the data structures for the dynamically link editor.
235
236   The original set of symbolic information in the symtab_command which contains
237   the symbol and string tables must also be present when this load command is
238   present.  When this load command is present the symbol table is organized
239   into three groups of symbols:
240       local symbols (static and debugging symbols) - grouped by module
241       defined external symbols - grouped by module (sorted by name if not lib)
242       undefined external symbols (sorted by name)
243   In this load command there are offsets and counts to each of the three groups
244   of symbols.
245
246   This load command contains a the offsets and sizes of the following new
247   symbolic information tables:
248       table of contents
249       module table
250       reference symbol table
251       indirect symbol table
252   The first three tables above (the table of contents, module table and
253   reference symbol table) are only present if the file is a dynamically linked
254   shared library.  For executable and object modules, which are files
255   containing only one module, the information that would be in these three
256   tables is determined as follows:
257       table of contents - the defined external symbols are sorted by name
258       module table - the file contains only one module so everything in the
259                      file is part of the module.
260       reference symbol table - is the defined and undefined external symbols
261
262   For dynamically linked shared library files this load command also contains
263   offsets and sizes to the pool of relocation entries for all sections
264   separated into two groups:
265       external relocation entries
266       local relocation entries
267   For executable and object modules the relocation entries continue to hang
268   off the section structures.  */
269
270typedef struct bfd_mach_o_dysymtab_command
271{
272  /* The symbols indicated by symoff and nsyms of the LC_SYMTAB load command
273     are grouped into the following three groups:
274       local symbols (further grouped by the module they are from)
275       defined external symbols (further grouped by the module they are from)
276       undefined symbols
277
278     The local symbols are used only for debugging.  The dynamic binding
279     process may have to use them to indicate to the debugger the local
280     symbols for a module that is being bound.
281
282     The last two groups are used by the dynamic binding process to do the
283     binding (indirectly through the module table and the reference symbol
284     table when this is a dynamically linked shared library file).  */
285
286  unsigned long ilocalsym;    /* Index to local symbols.  */
287  unsigned long nlocalsym;    /* Number of local symbols.  */
288  unsigned long iextdefsym;   /* Index to externally defined symbols.  */
289  unsigned long nextdefsym;   /* Number of externally defined symbols.  */
290  unsigned long iundefsym;    /* Index to undefined symbols.  */
291  unsigned long nundefsym;    /* Number of undefined symbols.  */
292
293  /* For the for the dynamic binding process to find which module a symbol
294     is defined in the table of contents is used (analogous to the ranlib
295     structure in an archive) which maps defined external symbols to modules
296     they are defined in.  This exists only in a dynamically linked shared
297     library file.  For executable and object modules the defined external
298     symbols are sorted by name and is use as the table of contents.  */
299
300  unsigned long tocoff;       /* File offset to table of contents.  */
301  unsigned long ntoc;         /* Number of entries in table of contents.  */
302
303  /* To support dynamic binding of "modules" (whole object files) the symbol
304     table must reflect the modules that the file was created from.  This is
305     done by having a module table that has indexes and counts into the merged
306     tables for each module.  The module structure that these two entries
307     refer to is described below.  This exists only in a dynamically linked
308     shared library file.  For executable and object modules the file only
309     contains one module so everything in the file belongs to the module.  */
310
311  unsigned long modtaboff;    /* File offset to module table.  */
312  unsigned long nmodtab;      /* Number of module table entries.  */
313
314  /* To support dynamic module binding the module structure for each module
315     indicates the external references (defined and undefined) each module
316     makes.  For each module there is an offset and a count into the
317     reference symbol table for the symbols that the module references.
318     This exists only in a dynamically linked shared library file.  For
319     executable and object modules the defined external symbols and the
320     undefined external symbols indicates the external references.  */
321
322  unsigned long extrefsymoff;  /* Offset to referenced symbol table.  */
323  unsigned long nextrefsyms;   /* Number of referenced symbol table entries.  */
324
325  /* The sections that contain "symbol pointers" and "routine stubs" have
326     indexes and (implied counts based on the size of the section and fixed
327     size of the entry) into the "indirect symbol" table for each pointer
328     and stub.  For every section of these two types the index into the
329     indirect symbol table is stored in the section header in the field
330     reserved1.  An indirect symbol table entry is simply a 32bit index into
331     the symbol table to the symbol that the pointer or stub is referring to.
332     The indirect symbol table is ordered to match the entries in the section.  */
333
334  unsigned long indirectsymoff; /* File offset to the indirect symbol table.  */
335  unsigned long nindirectsyms;  /* Number of indirect symbol table entries.  */
336
337  /* To support relocating an individual module in a library file quickly the
338     external relocation entries for each module in the library need to be
339     accessed efficiently.  Since the relocation entries can't be accessed
340     through the section headers for a library file they are separated into
341     groups of local and external entries further grouped by module.  In this
342     case the presents of this load command who's extreloff, nextrel,
343     locreloff and nlocrel fields are non-zero indicates that the relocation
344     entries of non-merged sections are not referenced through the section
345     structures (and the reloff and nreloc fields in the section headers are
346     set to zero).
347
348     Since the relocation entries are not accessed through the section headers
349     this requires the r_address field to be something other than a section
350     offset to identify the item to be relocated.  In this case r_address is
351     set to the offset from the vmaddr of the first LC_SEGMENT command.
352
353     The relocation entries are grouped by module and the module table
354     entries have indexes and counts into them for the group of external
355     relocation entries for that the module.
356
357     For sections that are merged across modules there must not be any
358     remaining external relocation entries for them (for merged sections
359     remaining relocation entries must be local).  */
360
361  unsigned long extreloff;    /* Offset to external relocation entries.  */
362  unsigned long nextrel;      /* Number of external relocation entries.  */
363
364  /* All the local relocation entries are grouped together (they are not
365     grouped by their module since they are only used if the object is moved
366     from it statically link edited address).  */
367
368  unsigned long locreloff;    /* Offset to local relocation entries.  */
369  unsigned long nlocrel;      /* Number of local relocation entries.  */
370}
371bfd_mach_o_dysymtab_command;
372
373/* An indirect symbol table entry is simply a 32bit index into the symbol table
374   to the symbol that the pointer or stub is refering to.  Unless it is for a
375   non-lazy symbol pointer section for a defined symbol which strip(1) as
376   removed.  In which case it has the value INDIRECT_SYMBOL_LOCAL.  If the
377   symbol was also absolute INDIRECT_SYMBOL_ABS is or'ed with that.  */
378
379#define INDIRECT_SYMBOL_LOCAL 0x80000000
380#define INDIRECT_SYMBOL_ABS   0x40000000
381
382typedef struct bfd_mach_o_thread_flavour
383{
384  unsigned long flavour;
385  bfd_vma offset;
386  unsigned long size;
387}
388bfd_mach_o_thread_flavour;
389
390typedef struct bfd_mach_o_thread_command
391{
392  unsigned long nflavours;
393  bfd_mach_o_thread_flavour *flavours;
394  asection *section;
395}
396bfd_mach_o_thread_command;
397
398typedef struct bfd_mach_o_dylinker_command
399{
400  unsigned long cmd;                   /* LC_ID_DYLIB or LC_LOAD_DYLIB.  */
401  unsigned long cmdsize;               /* Includes pathname string.  */
402  unsigned long name_offset;           /* Offset to library's path name.  */
403  unsigned long name_len;              /* Offset to library's path name.  */
404  asection *section;
405}
406bfd_mach_o_dylinker_command;
407
408typedef struct bfd_mach_o_dylib_command
409{
410  unsigned long cmd;                   /* LC_ID_DYLIB or LC_LOAD_DYLIB.  */
411  unsigned long cmdsize;               /* Includes pathname string.  */
412  unsigned long name_offset;           /* Offset to library's path name.  */
413  unsigned long name_len;              /* Offset to library's path name.  */
414  unsigned long timestamp;	       /* Library's build time stamp.  */
415  unsigned long current_version;       /* Library's current version number.  */
416  unsigned long compatibility_version; /* Library's compatibility vers number.  */
417  asection *section;
418}
419bfd_mach_o_dylib_command;
420
421typedef struct bfd_mach_o_prebound_dylib_command
422{
423  unsigned long cmd;                 /* LC_PREBOUND_DYLIB.  */
424  unsigned long cmdsize;             /* Includes strings.  */
425  unsigned long name;                /* Library's path name.  */
426  unsigned long nmodules;            /* Number of modules in library.  */
427  unsigned long linked_modules;      /* Bit vector of linked modules.  */
428  asection *section;
429}
430bfd_mach_o_prebound_dylib_command;
431
432typedef struct bfd_mach_o_load_command
433{
434  bfd_mach_o_load_command_type type;
435  unsigned int type_required;
436  bfd_vma offset;
437  bfd_vma len;
438  union
439  {
440    bfd_mach_o_segment_command segment;
441    bfd_mach_o_symtab_command symtab;
442    bfd_mach_o_dysymtab_command dysymtab;
443    bfd_mach_o_thread_command thread;
444    bfd_mach_o_dylib_command dylib;
445    bfd_mach_o_dylinker_command dylinker;
446    bfd_mach_o_prebound_dylib_command prebound_dylib;
447  }
448  command;
449}
450bfd_mach_o_load_command;
451
452typedef struct mach_o_data_struct
453{
454  bfd_mach_o_header header;
455  bfd_mach_o_load_command *commands;
456  unsigned long nsymbols;
457  asymbol *symbols;
458  unsigned long nsects;
459  bfd_mach_o_section **sections;
460  bfd *ibfd;
461}
462mach_o_data_struct;
463
464typedef struct mach_o_data_struct bfd_mach_o_data_struct;
465
466bfd_boolean        bfd_mach_o_valid  (bfd *);
467int                bfd_mach_o_scan_read_symtab_symbol        (bfd *, bfd_mach_o_symtab_command *, asymbol *, unsigned long);
468int                bfd_mach_o_scan_read_symtab_strtab        (bfd *, bfd_mach_o_symtab_command *);
469int                bfd_mach_o_scan_read_symtab_symbols       (bfd *, bfd_mach_o_symtab_command *);
470int                bfd_mach_o_scan_read_dysymtab_symbol      (bfd *, bfd_mach_o_dysymtab_command *, bfd_mach_o_symtab_command *, asymbol *, unsigned long);
471int                bfd_mach_o_scan_start_address             (bfd *);
472int                bfd_mach_o_scan                           (bfd *, bfd_mach_o_header *, bfd_mach_o_data_struct *);
473bfd_boolean        bfd_mach_o_mkobject                       (bfd *);
474const bfd_target * bfd_mach_o_object_p                       (bfd *);
475const bfd_target * bfd_mach_o_core_p                         (bfd *);
476const bfd_target * bfd_mach_o_archive_p                      (bfd *);
477bfd *              bfd_mach_o_openr_next_archived_file       (bfd *, bfd *);
478int                bfd_mach_o_lookup_section                 (bfd *, asection *, bfd_mach_o_load_command **, bfd_mach_o_section **);
479int                bfd_mach_o_lookup_command                 (bfd *, bfd_mach_o_load_command_type, bfd_mach_o_load_command **);
480unsigned long      bfd_mach_o_stack_addr                     (enum bfd_mach_o_cpu_type);
481int                bfd_mach_o_core_fetch_environment         (bfd *, unsigned char **, unsigned int *);
482char *             bfd_mach_o_core_file_failing_command      (bfd *);
483int                bfd_mach_o_core_file_failing_signal       (bfd *);
484bfd_boolean        bfd_mach_o_core_file_matches_executable_p (bfd *, bfd *);
485
486extern const bfd_target mach_o_be_vec;
487extern const bfd_target mach_o_le_vec;
488extern const bfd_target mach_o_fat_vec;
489
490#endif /* _BFD_MACH_O_H_ */
491