libkern/mach-o/loader.h

   1 /*
   2  * Copyright (c) 2000 Apple Computer, Inc. All rights reserved.
   3  *
   4  * @APPLE_LICENSE_OSREFERENCE_HEADER_START@
   5  *
   6  * This file contains Original Code and/or Modifications of Original Code
   7  * as defined in and that are subject to the Apple Public Source License
   8  * Version 2.0 (the 'License'). You may not use this file except in
   9  * compliance with the License.  The rights granted to you under the
  10  * License may not be used to create, or enable the creation or
  11  * redistribution of, unlawful or unlicensed copies of an Apple operating
  12  * system, or to circumvent, violate, or enable the circumvention or
  13  * violation of, any terms of an Apple operating system software license
  14  * agreement.
  15  *
  16  * Please obtain a copy of the License at
  17  * http://www.opensource.apple.com/apsl/ and read it before using this
  18  * file.
  19  *
  20  * The Original Code and all software distributed under the License are
  21  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  22  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  23  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  24  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  25  * Please see the License for the specific language governing rights and
  26  * limitations under the License.
  27  *
  28  * @APPLE_LICENSE_OSREFERENCE_HEADER_END@
  29  */
  30 #ifndef _MACHO_LOADER_H_
  31 #define _MACHO_LOADER_H_
  32
  33 /*
  34  * This file describes the format of mach object files.
  35  */
  36
  37 /*
  38  * <mach/machine.h> is needed here for the cpu_type_t and cpu_subtype_t types
  39  * and contains the constants for the possible values of these types.
  40  */
  41 #include <mach/machine.h>
  42
  43 /*
  44  * <mach/vm_prot.h> is needed here for the vm_prot_t type and contains the
  45  * constants that are or'ed together for the possible values of this type.
  46  */
  47 #include <mach/vm_prot.h>
  48
  49 /*
  50  * <machine/thread_status.h> is expected to define the flavors of the thread
  51  * states and the structures of those flavors for each machine.
  52  */
  53 #include <mach/machine/thread_status.h>
  54
  55 /*
  56  * The mach header appears at the very beginning of the object file; it
  57  * is the same for both 32-bit and 64-bit architectures.
  58  */
  59 struct mach_header {
  60         uint32_t        magic;          /* mach magic number identifier */
  61         cpu_type_t      cputype;        /* cpu specifier */
  62         cpu_subtype_t   cpusubtype;     /* machine specifier */
  63         uint32_t        filetype;       /* type of file */
  64         uint32_t        ncmds;          /* number of load commands */
  65         uint32_t        sizeofcmds;     /* the size of all the load commands */
  66         uint32_t        flags;          /* flags */
  67 };
  68
  69 /* Constant for the magic field of the mach_header (32-bit architectures) */
  70 #define MH_MAGIC        0xfeedface      /* the mach magic number */
  71 #define MH_CIGAM        0xcefaedfe
  72
  73 /* Constant for the magic field of the mach_header_64 (64-bit architectures) */
  74 #define MH_MAGIC_64     0xfeedfacf      /* the 64-bit mach magic number */
  75 #define MH_CIGAM_64     0xcffaedfe
  76
  77 /* Constants for the cmd field of new load commands, the type */
  78 #define LC_SEGMENT_64   0x19    /* 64-bit segment of this file to be mapped */
  79 #define LC_ROUTINES_64  0x1a    /* 64-bit image routines */
  80
  81
  82 /*
  83  * The layout of the file depends on the filetype.  For all but the MH_OBJECT
  84  * file type the segments are padded out and aligned on a segment alignment
  85  * boundary for efficient demand pageing.  The MH_EXECUTE, MH_FVMLIB, MH_DYLIB,
  86  * MH_DYLINKER and MH_BUNDLE file types also have the headers included as part
  87  * of their first segment.
  88  *
  89  * The file type MH_OBJECT is a compact format intended as output of the
  90  * assembler and input (and possibly output) of the link editor (the .o
  91  * format).  All sections are in one unnamed segment with no segment padding.
  92  * This format is used as an executable format when the file is so small the
  93  * segment padding greatly increases it's size.
  94  *
  95  * The file type MH_PRELOAD is an executable format intended for things that
  96  * not executed under the kernel (proms, stand alones, kernels, etc).  The
  97  * format can be executed under the kernel but may demand paged it and not
  98  * preload it before execution.
  99  *
 100  * A core file is in MH_CORE format and can be any in an arbritray legal
 101  * Mach-O file.
 102  *
 103  * Constants for the filetype field of the mach_header
 104  */
 105 #define MH_OBJECT       0x1             /* relocatable object file */
 106 #define MH_EXECUTE      0x2             /* demand paged executable file */
 107 #define MH_FVMLIB       0x3             /* fixed VM shared library file */
 108 #define MH_CORE         0x4             /* core file */
 109 #define MH_PRELOAD      0x5             /* preloaded executable file */
 110 #define MH_DYLIB        0x6             /* dynamicly bound shared library file*/
 111 #define MH_DYLINKER     0x7             /* dynamic link editor */
 112 #define MH_BUNDLE       0x8             /* dynamicly bound bundle file */
 113
 114 /* Constants for the flags field of the mach_header */
 115 #define MH_NOUNDEFS     0x1             /* the object file has no undefined
 116                                            references, can be executed */
 117 #define MH_INCRLINK     0x2             /* the object file is the output of an
 118                                            incremental link against a base file
 119                                            and can't be link edited again */
 120 #define MH_DYLDLINK     0x4             /* the object file is input for the
 121                                            dynamic linker and can't be staticly
 122                                            link edited again */
 123 #define MH_BINDATLOAD   0x8             /* the object file's undefined
 124                                            references are bound by the dynamic
 125                                            linker when loaded. */
 126 #define MH_PREBOUND     0x10            /* the file has it's dynamic undefined
 127                                            references prebound. */
 128
 129 /*
 130  * The load commands directly follow the mach_header.  The total size of all
 131  * of the commands is given by the sizeofcmds field in the mach_header.  All
 132  * load commands must have as their first two fields cmd and cmdsize.  The cmd
 133  * field is filled in with a constant for that command type.  Each command type
 134  * has a structure specifically for it.  The cmdsize field is the size in bytes
 135  * of the particular load command structure plus anything that follows it that
 136  * is a part of the load command (i.e. section structures, strings, etc.).  To
 137  * advance to the next load command the cmdsize can be added to the offset or
 138  * pointer of the current load command.  The cmdsize for 32-bit architectures
 139  * MUST be a multiple of 4 bytes and for 64-bit architectures MUST be a multiple
 140  * of 8 bytes (these are forever the maximum alignment of any load commands).
 141  * sizeof(long) (this is forever the maximum alignment of any load commands).
 142  * The padded bytes must be zero.  All tables in the object file must also
 143  * follow these rules so the file can be memory mapped.  Otherwise the pointers
 144  * to these tables will not work well or at all on some machines.  With all
 145  * padding zeroed like objects will compare byte for byte.
 146  */
 147 struct load_command {
 148         unsigned long cmd;              /* type of load command */
 149         unsigned long cmdsize;          /* total size of command in bytes */
 150 };
 151
 152 /* Constants for the cmd field of all load commands, the type */
 153 #define LC_SEGMENT      0x1     /* segment of this file to be mapped */
 154 #define LC_SYMTAB       0x2     /* link-edit stab symbol table info */
 155 #define LC_SYMSEG       0x3     /* link-edit gdb symbol table info (obsolete) */
 156 #define LC_THREAD       0x4     /* thread */
 157 #define LC_UNIXTHREAD   0x5     /* unix thread (includes a stack) */
 158 #define LC_LOADFVMLIB   0x6     /* load a specified fixed VM shared library */
 159 #define LC_IDFVMLIB     0x7     /* fixed VM shared library identification */
 160 #define LC_IDENT        0x8     /* object identification info (obsolete) */
 161 #define LC_FVMFILE      0x9     /* fixed VM file inclusion (internal use) */
 162 #define LC_PREPAGE      0xa     /* prepage command (internal use) */
 163 #define LC_DYSYMTAB     0xb     /* dynamic link-edit symbol table info */
 164 #define LC_LOAD_DYLIB   0xc     /* load a dynamicly linked shared library */
 165 #define LC_ID_DYLIB     0xd     /* dynamicly linked shared lib identification */
 166 #define LC_LOAD_DYLINKER 0xe    /* load a dynamic linker */
 167 #define LC_ID_DYLINKER  0xf     /* dynamic linker identification */
 168 #define LC_PREBOUND_DYLIB 0x10  /* modules prebound for a dynamicly */
 169                                 /*  linked shared library */
 170
 171 /*
 172  * A variable length string in a load command is represented by an lc_str
 173  * union.  The strings are stored just after the load command structure and
 174  * the offset is from the start of the load command structure.  The size
 175  * of the string is reflected in the cmdsize field of the load command.
 176  * Once again any padded bytes to bring the cmdsize field to a multiple
 177  * of sizeof(long) must be zero.
 178  */
 179 union lc_str {
 180         unsigned long   offset; /* offset to the string */
 181         char            *ptr;   /* pointer to the string */
 182 };
 183
 184 /*
 185  * The segment load command indicates that a part of this file is to be
 186  * mapped into the task's address space.  The size of this segment in memory,
 187  * vmsize, maybe equal to or larger than the amount to map from this file,
 188  * filesize.  The file is mapped starting at fileoff to the beginning of
 189  * the segment in memory, vmaddr.  The rest of the memory of the segment,
 190  * if any, is allocated zero fill on demand.  The segment's maximum virtual
 191  * memory protection and initial virtual memory protection are specified
 192  * by the maxprot and initprot fields.  If the segment has sections then the
 193  * section structures directly follow the segment command and their size is
 194  * reflected in cmdsize.
 195  */
 196 struct segment_command {        /* for 32-bit architectures */
 197         unsigned long   cmd;            /* LC_SEGMENT */
 198         unsigned long   cmdsize;        /* includes sizeof section structs */
 199         char            segname[16];    /* segment name */
 200         unsigned long   vmaddr;         /* memory address of this segment */
 201         unsigned long   vmsize;         /* memory size of this segment */
 202         unsigned long   fileoff;        /* file offset of this segment */
 203         unsigned long   filesize;       /* amount to map from the file */
 204         vm_prot_t       maxprot;        /* maximum VM protection */
 205         vm_prot_t       initprot;       /* initial VM protection */
 206         unsigned long   nsects;         /* number of sections in segment */
 207         unsigned long   flags;          /* flags */
 208 };
 209
 210 /*
 211  * The 64-bit segment load command indicates that a part of this file is to be
 212  * mapped into a 64-bit task's address space.  If the 64-bit segment has
 213  * sections then section_64 structures directly follow the 64-bit segment
 214  * command and their size is reflected in cmdsize.
 215  */
 216 struct segment_command_64 {     /* for 64-bit architectures */
 217         uint32_t        cmd;            /* LC_SEGMENT_64 */
 218         uint32_t        cmdsize;        /* includes sizeof section_64 structs */
 219         char            segname[16];    /* segment name */
 220         uint64_t        vmaddr;         /* memory address of this segment */
 221         uint64_t        vmsize;         /* memory size of this segment */
 222         uint32_t        fileoff;        /* file offset of this segment */
 223         uint32_t        filesize;       /* amount to map from the file */
 224         vm_prot_t       maxprot;        /* maximum VM protection */
 225         vm_prot_t       initprot;       /* initial VM protection */
 226         uint32_t        nsects;         /* number of sections in segment */
 227         uint32_t        flags;          /* flags */
 228 };
 229
 230
 231 /* Constants for the flags field of the segment_command */
 232 #define SG_HIGHVM       0x1     /* the file contents for this segment is for
 233                                    the high part of the VM space, the low part
 234                                    is zero filled (for stacks in core files) */
 235 #define SG_FVMLIB       0x2     /* this segment is the VM that is allocated by
 236                                    a fixed VM library, for overlap checking in
 237                                    the link editor */
 238 #define SG_NORELOC      0x4     /* this segment has nothing that was relocated
 239                                    in it and nothing relocated to it, that is
 240                                    it maybe safely replaced without relocation*/
 241
 242 /*
 243  * A segment is made up of zero or more sections.  Non-MH_OBJECT files have
 244  * all of their segments with the proper sections in each, and padded to the
 245  * specified segment alignment when produced by the link editor.  The first
 246  * segment of a MH_EXECUTE and MH_FVMLIB format file contains the mach_header
 247  * and load commands of the object file before it's first section.  The zero
 248  * fill sections are always last in their segment (in all formats).  This
 249  * allows the zeroed segment padding to be mapped into memory where zero fill
 250  * sections might be. The gigabyte zero fill sections, those with the section
 251  * type S_GB_ZEROFILL, can only be in a segment with sections of this type.
 252  * These segments are then placed after all other segments.
 253  *
 254  * The MH_OBJECT format has all of it's sections in one segment for
 255  * compactness.  There is no padding to a specified segment boundary and the
 256  * mach_header and load commands are not part of the segment.
 257  *
 258  * Sections with the same section name, sectname, going into the same segment,
 259  * segname, are combined by the link editor.  The resulting section is aligned
 260  * to the maximum alignment of the combined sections and is the new section's
 261  * alignment.  The combined sections are aligned to their original alignment in
 262  * the combined section.  Any padded bytes to get the specified alignment are
 263  * zeroed.
 264  *
 265  * The format of the relocation entries referenced by the reloff and nreloc
 266  * fields of the section structure for mach object files is described in the
 267  * header file <reloc.h>.
 268  */
 269 struct section {                /* for 32-bit architectures */
 270         char            sectname[16];   /* name of this section */
 271         char            segname[16];    /* segment this section goes in */
 272         unsigned long   addr;           /* memory address of this section */
 273         unsigned long   size;           /* size in bytes of this section */
 274         unsigned long   offset;         /* file offset of this section */
 275         unsigned long   align;          /* section alignment (power of 2) */
 276         unsigned long   reloff;         /* file offset of relocation entries */
 277         unsigned long   nreloc;         /* number of relocation entries */
 278         unsigned long   flags;          /* flags (section type and attributes)*/
 279         unsigned long   reserved1;      /* reserved */
 280         unsigned long   reserved2;      /* reserved */
 281 };
 282
 283 struct section_64 { /* for 64-bit architectures */
 284         char            sectname[16];   /* name of this section */
 285         char            segname[16];    /* segment this section goes in */
 286         uint64_t        addr;           /* memory address of this section */
 287         uint64_t        size;           /* size in bytes of this section */
 288         uint32_t        offset;         /* file offset of this section */
 289         uint32_t        align;          /* section alignment (power of 2) */
 290         uint32_t        reloff;         /* file offset of relocation entries */
 291         uint32_t        nreloc;         /* number of relocation entries */
 292         uint32_t        flags;          /* flags (section type and attributes)*/
 293         uint32_t        reserved1;      /* reserved (for offset or index) */
 294         uint32_t        reserved2;      /* reserved (for count or sizeof) */
 295         uint32_t        reserved3;      /* reserved */
 296 };
 297
 298
 299 /*
 300  * The flags field of a section structure is separated into two parts a section
 301  * type and section attributes.  The section types are mutually exclusive (it
 302  * can only have one type) but the section attributes are not (it may have more
 303  * than one attribute).
 304  */
 305 #define SECTION_TYPE             0x000000ff     /* 256 section types */
 306 #define SECTION_ATTRIBUTES       0xffffff00     /*  24 section attributes */
 307
 308 /* Constants for the type of a section */
 309 #define S_REGULAR               0x0     /* regular section */
 310 #define S_ZEROFILL              0x1     /* zero fill on demand section */
 311 #define S_CSTRING_LITERALS      0x2     /* section with only literal C strings*/
 312 #define S_4BYTE_LITERALS        0x3     /* section with only 4 byte literals */
 313 #define S_8BYTE_LITERALS        0x4     /* section with only 8 byte literals */
 314 #define S_LITERAL_POINTERS      0x5     /* section with only pointers to */
 315                                         /*  literals */
 316 /*
 317  * For the two types of symbol pointers sections and the symbol stubs section
 318  * they have indirect symbol table entries.  For each of the entries in the
 319  * section the indirect symbol table entries, in corresponding order in the
 320  * indirect symbol table, start at the index stored in the reserved1 field
 321  * of the section structure.  Since the indirect symbol table entries
 322  * correspond to the entries in the section the number of indirect symbol table
 323  * entries is inferred from the size of the section divided by the size of the
 324  * entries in the section.  For symbol pointers sections the size of the entries
 325  * in the section is 4 bytes and for symbol stubs sections the byte size of the
 326  * stubs is stored in the reserved2 field of the section structure.
 327  */
 328 #define S_NON_LAZY_SYMBOL_POINTERS      0x6     /* section with only non-lazy
 329                                                    symbol pointers */
 330 #define S_LAZY_SYMBOL_POINTERS          0x7     /* section with only lazy symbol
 331                                                    pointers */
 332 #define S_SYMBOL_STUBS                  0x8     /* section with only symbol
 333                                                    stubs, byte size of stub in
 334                                                    the reserved2 field */
 335 #define S_MOD_INIT_FUNC_POINTERS        0x9     /* section with only function
 336                                                    pointers for initialization*/
 337 /*
 338  * Constants for the section attributes part of the flags field of a section
 339  * structure.
 340  */
 341 #define SECTION_ATTRIBUTES_USR   0xff000000     /* User setable attributes */
 342 #define S_ATTR_PURE_INSTRUCTIONS 0x80000000     /* section contains only true
 343                                                    machine instructions */
 344 #define SECTION_ATTRIBUTES_SYS   0x00ffff00     /* system setable attributes */
 345 #define S_ATTR_SOME_INSTRUCTIONS 0x00000400     /* section contains some
 346                                                    machine instructions */
 347 #define S_ATTR_EXT_RELOC         0x00000200     /* section has external
 348                                                    relocation entries */
 349 #define S_ATTR_LOC_RELOC         0x00000100     /* section has local
 350                                                    relocation entries */
 351
 352
 353 /*
 354  * The names of segments and sections in them are mostly meaningless to the
 355  * link-editor.  But there are few things to support traditional UNIX
 356  * executables that require the link-editor and assembler to use some names
 357  * agreed upon by convention.
 358  *
 359  * The initial protection of the "__TEXT" segment has write protection turned
 360  * off (not writeable).
 361  *
 362  * The link-editor will allocate common symbols at the end of the "__common"
 363  * section in the "__DATA" segment.  It will create the section and segment
 364  * if needed.
 365  */
 366
 367 /* The currently known segment names and the section names in those segments */
 368
 369 #define SEG_PAGEZERO    "__PAGEZERO"    /* the pagezero segment which has no */
 370                                         /* protections and catches NULL */
 371                                         /* references for MH_EXECUTE files */
 372
 373
 374 #define SEG_TEXT        "__TEXT"        /* the tradition UNIX text segment */
 375 #define SECT_TEXT       "__text"        /* the real text part of the text */
 376                                         /* section no headers, and no padding */
 377 #define SECT_FVMLIB_INIT0 "__fvmlib_init0"      /* the fvmlib initialization */
 378                                                 /*  section */
 379 #define SECT_FVMLIB_INIT1 "__fvmlib_init1"      /* the section following the */
 380                                                 /*  fvmlib initialization */
 381                                                 /*  section */
 382
 383 #define SEG_DATA        "__DATA"        /* the tradition UNIX data segment */
 384 #define SECT_DATA       "__data"        /* the real initialized data section */
 385                                         /* no padding, no bss overlap */
 386 #define SECT_BSS        "__bss"         /* the real uninitialized data section*/
 387                                         /* no padding */
 388 #define SECT_COMMON     "__common"      /* the section common symbols are */
 389                                         /* allocated in by the link editor */
 390
 391 #define SEG_OBJC        "__OBJC"        /* objective-C runtime segment */
 392 #define SECT_OBJC_SYMBOLS "__symbol_table"      /* symbol table */
 393 #define SECT_OBJC_MODULES "__module_info"       /* module information */
 394 #define SECT_OBJC_STRINGS "__selector_strs"     /* string table */
 395 #define SECT_OBJC_REFS "__selector_refs"        /* string table */
 396
 397 #define SEG_ICON         "__ICON"       /* the NeXT icon segment */
 398 #define SECT_ICON_HEADER "__header"     /* the icon headers */
 399 #define SECT_ICON_TIFF   "__tiff"       /* the icons in tiff format */
 400
 401 #define SEG_LINKEDIT    "__LINKEDIT"    /* the segment containing all structs */
 402                                         /* created and maintained by the link */
 403                                         /* editor.  Created with -seglinkedit */
 404                                         /* option to ld(1) for MH_EXECUTE and */
 405                                         /* FVMLIB file types only */
 406
 407 #define SEG_UNIXSTACK   "__UNIXSTACK"   /* the unix stack segment */
 408
 409 /*
 410  * Fixed virtual memory shared libraries are identified by two things.  The
 411  * target pathname (the name of the library as found for execution), and the
 412  * minor version number.  The address of where the headers are loaded is in
 413  * header_addr.
 414  */
 415 struct fvmlib {
 416         union lc_str    name;           /* library's target pathname */
 417         unsigned long   minor_version;  /* library's minor version number */
 418         unsigned long   header_addr;    /* library's header address */
 419 };
 420
 421 /*
 422  * A fixed virtual shared library (filetype == MH_FVMLIB in the mach header)
 423  * contains a fvmlib_command (cmd == LC_IDFVMLIB) to identify the library.
 424  * An object that uses a fixed virtual shared library also contains a
 425  * fvmlib_command (cmd == LC_LOADFVMLIB) for each library it uses.
 426  */
 427 struct fvmlib_command {
 428         unsigned long   cmd;            /* LC_IDFVMLIB or LC_LOADFVMLIB */
 429         unsigned long   cmdsize;        /* includes pathname string */
 430         struct fvmlib   fvmlib;         /* the library identification */
 431 };
 432
 433 /*
 434  * Dynamicly linked shared libraries are identified by two things.  The
 435  * pathname (the name of the library as found for execution), and the
 436  * compatibility version number.  The pathname must match and the compatibility
 437  * number in the user of the library must be greater than or equal to the
 438  * library being used.  The time stamp is used to record the time a library was
 439  * built and copied into user so it can be use to determined if the library used
 440  * at runtime is exactly the same as used to built the program.
 441  */
 442 struct dylib {
 443     union lc_str  name;                 /* library's path name */
 444     unsigned long timestamp;            /* library's build time stamp */
 445     unsigned long current_version;      /* library's current version number */
 446     unsigned long compatibility_version;/* library's compatibility vers number*/
 447 };
 448
 449 /*
 450  * A dynamicly linked shared library (filetype == MH_DYLIB in the mach header)
 451  * contains a dylib_command (cmd == LC_ID_DYLIB) to identify the library.
 452  * An object that uses a dynamicly linked shared library also contains a
 453  * dylib_command (cmd == LC_LOAD_DYLIB) for each library it uses.
 454  */
 455 struct dylib_command {
 456         unsigned long   cmd;            /* LC_ID_DYLIB or LC_LOAD_DYLIB */
 457         unsigned long   cmdsize;        /* includes pathname string */
 458         struct dylib    dylib;          /* the library identification */
 459 };
 460
 461 /*
 462  * A program (filetype == MH_EXECUTE) or bundle (filetype == MH_BUNDLE) that is
 463  * prebound to it's dynamic libraries has one of these for each library that
 464  * the static linker used in prebinding.  It contains a bit vector for the
 465  * modules in the library.  The bits indicate which modules are bound (1) and
 466  * which are not (0) from the library.  The bit for module 0 is the low bit
 467  * of the first byte.  So the bit for the Nth module is:
 468  * (linked_modules[N/8] >> N%8) & 1
 469  */
 470 struct prebound_dylib_command {
 471         unsigned long   cmd;            /* LC_PREBOUND_DYLIB */
 472         unsigned long   cmdsize;        /* includes strings */
 473         union lc_str    name;           /* library's path name */
 474         unsigned long   nmodules;       /* number of modules in library */
 475         union lc_str    linked_modules; /* bit vector of linked modules */
 476 };
 477
 478 /*
 479  * A program that uses a dynamic linker contains a dylinker_command to identify
 480  * the name of the dynamic linker (LC_LOAD_DYLINKER).  And a dynamic linker
 481  * contains a dylinker_command to identify the dynamic linker (LC_ID_DYLINKER).
 482  * A file can have at most one of these.
 483  */
 484 struct dylinker_command {
 485         unsigned long   cmd;            /* LC_ID_DYLINKER or LC_LOAD_DYLINKER */
 486         unsigned long   cmdsize;        /* includes pathname string */
 487         union lc_str    name;           /* dynamic linker's path name */
 488 };
 489
 490 /*
 491  * Thread commands contain machine-specific data structures suitable for
 492  * use in the thread state primitives.  The machine specific data structures
 493  * follow the struct thread_command as follows.
 494  * Each flavor of machine specific data structure is preceded by an unsigned
 495  * long constant for the flavor of that data structure, an unsigned long
 496  * that is the count of longs of the size of the state data structure and then
 497  * the state data structure follows.  This triple may be repeated for many
 498  * flavors.  The constants for the flavors, counts and state data structure
 499  * definitions are expected to be in the header file <machine/thread_status.h>.
 500  * These machine specific data structures sizes must be multiples of
 501  * sizeof(long).  The cmdsize reflects the total size of the thread_command
 502  * and all of the sizes of the constants for the flavors, counts and state
 503  * data structures.
 504  *
 505  * For executable objects that are unix processes there will be one
 506  * thread_command (cmd == LC_UNIXTHREAD) created for it by the link-editor.
 507  * This is the same as a LC_THREAD, except that a stack is automatically
 508  * created (based on the shell's limit for the stack size).  Command arguments
 509  * and environment variables are copied onto that stack.
 510  */
 511 struct thread_command {
 512         unsigned long   cmd;            /* LC_THREAD or  LC_UNIXTHREAD */
 513         unsigned long   cmdsize;        /* total size of this command */
 514         /* unsigned long flavor            flavor of thread state */
 515         /* unsigned long count             count of longs in thread state */
 516         /* struct XXX_thread_state state   thread state for this flavor */
 517         /* ... */
 518 };
 519
 520 /*
 521  * The symtab_command contains the offsets and sizes of the link-edit 4.3BSD
 522  * "stab" style symbol table information as described in the header files
 523  * <nlist.h> and <stab.h>.
 524  */
 525 struct symtab_command {
 526         unsigned long   cmd;            /* LC_SYMTAB */
 527         unsigned long   cmdsize;        /* sizeof(struct symtab_command) */
 528         unsigned long   symoff;         /* symbol table offset */
 529         unsigned long   nsyms;          /* number of symbol table entries */
 530         unsigned long   stroff;         /* string table offset */
 531         unsigned long   strsize;        /* string table size in bytes */
 532 };
 533
 534 /*
 535  * This is the second set of the symbolic information which is used to support
 536  * the data structures for the dynamicly link editor.
 537  *
 538  * The original set of symbolic information in the symtab_command which contains
 539  * the symbol and string tables must also be present when this load command is
 540  * present.  When this load command is present the symbol table is organized
 541  * into three groups of symbols:
 542  *      local symbols (static and debugging symbols) - grouped by module
 543  *      defined external symbols - grouped by module (sorted by name if not lib)
 544  *      undefined external symbols (sorted by name)
 545  * In this load command there are offsets and counts to each of the three groups
 546  * of symbols.
 547  *
 548  * This load command contains a the offsets and sizes of the following new
 549  * symbolic information tables:
 550  *      table of contents
 551  *      module table
 552  *      reference symbol table
 553  *      indirect symbol table
 554  * The first three tables above (the table of contents, module table and
 555  * reference symbol table) are only present if the file is a dynamicly linked
 556  * shared library.  For executable and object modules, which are files
 557  * containing only one module, the information that would be in these three
 558  * tables is determined as follows:
 559  *      table of contents - the defined external symbols are sorted by name
 560  *      module table - the file contains only one module so everything in the
 561  *                     file is part of the module.
 562  *      reference symbol table - is the defined and undefined external symbols
 563  *
 564  * For dynamicly linked shared library files this load command also contains
 565  * offsets and sizes to the pool of relocation entries for all sections
 566  * separated into two groups:
 567  *      external relocation entries
 568  *      local relocation entries
 569  * For executable and object modules the relocation entries continue to hang
 570  * off the section structures.
 571  */
 572 struct dysymtab_command {
 573     unsigned long cmd;          /* LC_DYSYMTAB */
 574     unsigned long cmdsize;      /* sizeof(struct dysymtab_command) */
 575
 576     /*
 577      * The symbols indicated by symoff and nsyms of the LC_SYMTAB load command
 578      * are grouped into the following three groups:
 579      *    local symbols (further grouped by the module they are from)
 580      *    defined external symbols (further grouped by the module they are from)
 581      *    undefined symbols
 582      *
 583      * The local symbols are used only for debugging.  The dynamic binding
 584      * process may have to use them to indicate to the debugger the local
 585      * symbols for a module that is being bound.
 586      *
 587      * The last two groups are used by the dynamic binding process to do the
 588      * binding (indirectly through the module table and the reference symbol
 589      * table when this is a dynamicly linked shared library file).
 590      */
 591     unsigned long ilocalsym;    /* index to local symbols */
 592     unsigned long nlocalsym;    /* number of local symbols */
 593
 594     unsigned long iextdefsym;   /* index to externally defined symbols */
 595     unsigned long nextdefsym;   /* number of externally defined symbols */
 596
 597     unsigned long iundefsym;    /* index to undefined symbols */
 598     unsigned long nundefsym;    /* number of undefined symbols */
 599
 600     /*
 601      * For the for the dynamic binding process to find which module a symbol
 602      * is defined in the table of contents is used (analogous to the ranlib
 603      * structure in an archive) which maps defined external symbols to modules
 604      * they are defined in.  This exists only in a dynamicly linked shared
 605      * library file.  For executable and object modules the defined external
 606      * symbols are sorted by name and is use as the table of contents.
 607      */
 608     unsigned long tocoff;       /* file offset to table of contents */
 609     unsigned long ntoc;         /* number of entries in table of contents */
 610
 611     /*
 612      * To support dynamic binding of "modules" (whole object files) the symbol
 613      * table must reflect the modules that the file was created from.  This is
 614      * done by having a module table that has indexes and counts into the merged
 615      * tables for each module.  The module structure that these two entries
 616      * refer to is described below.  This exists only in a dynamicly linked
 617      * shared library file.  For executable and object modules the file only
 618      * contains one module so everything in the file belongs to the module.
 619      */
 620     unsigned long modtaboff;    /* file offset to module table */
 621     unsigned long nmodtab;      /* number of module table entries */
 622
 623     /*
 624      * To support dynamic module binding the module structure for each module
 625      * indicates the external references (defined and undefined) each module
 626      * makes.  For each module there is an offset and a count into the
 627      * reference symbol table for the symbols that the module references.
 628      * This exists only in a dynamicly linked shared library file.  For
 629      * executable and object modules the defined external symbols and the
 630      * undefined external symbols indicates the external references.
 631      */
 632     unsigned long extrefsymoff;  /* offset to referenced symbol table */
 633     unsigned long nextrefsyms;   /* number of referenced symbol table entries */
 634
 635     /*
 636      * The sections that contain "symbol pointers" and "routine stubs" have
 637      * indexes and (implied counts based on the size of the section and fixed
 638      * size of the entry) into the "indirect symbol" table for each pointer
 639      * and stub.  For every section of these two types the index into the
 640      * indirect symbol table is stored in the section header in the field
 641      * reserved1.  An indirect symbol table entry is simply a 32bit index into
 642      * the symbol table to the symbol that the pointer or stub is referring to.
 643      * The indirect symbol table is ordered to match the entries in the section.
 644      */
 645     unsigned long indirectsymoff; /* file offset to the indirect symbol table */
 646     unsigned long nindirectsyms;  /* number of indirect symbol table entries */
 647
 648     /*
 649      * To support relocating an individual module in a library file quickly the
 650      * external relocation entries for each module in the library need to be
 651      * accessed efficiently.  Since the relocation entries can't be accessed
 652      * through the section headers for a library file they are separated into
 653      * groups of local and external entries further grouped by module.  In this
 654      * case the presents of this load command who's extreloff, nextrel,
 655      * locreloff and nlocrel fields are non-zero indicates that the relocation
 656      * entries of non-merged sections are not referenced through the section
 657      * structures (and the reloff and nreloc fields in the section headers are
 658      * set to zero).
 659      *
 660      * Since the relocation entries are not accessed through the section headers
 661      * this requires the r_address field to be something other than a section
 662      * offset to identify the item to be relocated.  In this case r_address is
 663      * set to the offset from the vmaddr of the first LC_SEGMENT command.
 664      *
 665      * The relocation entries are grouped by module and the module table
 666      * entries have indexes and counts into them for the group of external
 667      * relocation entries for that the module.
 668      *
 669      * For sections that are merged across modules there must not be any
 670      * remaining external relocation entries for them (for merged sections
 671      * remaining relocation entries must be local).
 672      */
 673     unsigned long extreloff;    /* offset to external relocation entries */
 674     unsigned long nextrel;      /* number of external relocation entries */
 675
 676     /*
 677      * All the local relocation entries are grouped together (they are not
 678      * grouped by their module since they are only used if the object is moved
 679      * from it staticly link edited address).
 680      */
 681     unsigned long locreloff;    /* offset to local relocation entries */
 682     unsigned long nlocrel;      /* number of local relocation entries */
 683
 684 };
 685
 686 /*
 687  * An indirect symbol table entry is simply a 32bit index into the symbol table
 688  * to the symbol that the pointer or stub is refering to.  Unless it is for a
 689  * non-lazy symbol pointer section for a defined symbol which strip(1) as
 690  * removed.  In which case it has the value INDIRECT_SYMBOL_LOCAL.  If the
 691  * symbol was also absolute INDIRECT_SYMBOL_ABS is or'ed with that.
 692  */
 693 #define INDIRECT_SYMBOL_LOCAL   0x80000000
 694 #define INDIRECT_SYMBOL_ABS     0x40000000
 695
 696
 697 /* a table of contents entry */
 698 struct dylib_table_of_contents {
 699     unsigned long symbol_index; /* the defined external symbol
 700                                    (index into the symbol table) */
 701     unsigned long module_index; /* index into the module table this symbol
 702                                    is defined in */
 703 };
 704
 705 /* a module table entry */
 706 struct dylib_module {
 707     unsigned long module_name;  /* the module name (index into string table) */
 708
 709     unsigned long iextdefsym;   /* index into externally defined symbols */
 710     unsigned long nextdefsym;   /* number of externally defined symbols */
 711     unsigned long irefsym;              /* index into reference symbol table */
 712     unsigned long nrefsym;      /* number of reference symbol table entries */
 713     unsigned long ilocalsym;    /* index into symbols for local symbols */
 714     unsigned long nlocalsym;    /* number of local symbols */
 715
 716     unsigned long iextrel;      /* index into external relocation entries */
 717     unsigned long nextrel;      /* number of external relocation entries */
 718
 719     unsigned long iinit;        /* index into the init section */
 720     unsigned long ninit;        /* number of init section entries */
 721
 722     unsigned long               /* for this module address of the start of */
 723         objc_module_info_addr;  /*  the (__OBJC,__module_info) section */
 724     unsigned long               /* for this module size of */
 725         objc_module_info_size;  /*  the (__OBJC,__module_info) section */
 726 };
 727
 728 /* a 64-bit module table entry */
 729 struct dylib_module_64 {
 730         uint32_t module_name;   /* the module name (index into string table) */
 731
 732         uint32_t iextdefsym;    /* index into externally defined symbols */
 733         uint32_t nextdefsym;    /* number of externally defined symbols */
 734         uint32_t irefsym;       /* index into reference symbol table */
 735         uint32_t nrefsym;       /* number of reference symbol table entries */
 736         uint32_t ilocalsym;     /* index into symbols for local symbols */
 737         uint32_t nlocalsym;     /* number of local symbols */
 738
 739         uint32_t iextrel;       /* index into external relocation entries */
 740         uint32_t nextrel;       /* number of external relocation entries */
 741
 742         uint32_t iinit_iterm;   /* low 16 bits are the index into the init
 743                                    section, high 16 bits are the index into
 744                                    the term section */
 745         uint32_t ninit_nterm;   /* low 16 bits are the number of init section
 746                                    entries, high 16 bits are the number of
 747                                    term section entries */
 748
 749         uint32_t                /* for this module size of the */
 750                 objc_module_info_size;  /* (__OBJC,__module_info) section */
 751         uint64_t                /* for this module address of the start of */
 752                 objc_module_info_addr;  /* the (__OBJC,__module_info) section */
 753 };
 754
 755
 756 /*
 757  * The entries in the reference symbol table are used when loading the module
 758  * (both by the static and dynamic link editors) and if the module is unloaded
 759  * or replaced.  Therefore all external symbols (defined and undefined) are
 760  * listed in the module's reference table.  The flags describe the type of
 761  * reference that is being made.  The constants for the flags are defined in
 762  * <mach-o/nlist.h> as they are also used for symbol table entries.
 763  */
 764 struct dylib_reference {
 765     unsigned long isym:24,      /* index into the symbol table */
 766                   flags:8;      /* flags to indicate the type of reference */
 767 };
 768
 769 /*
 770  * The symseg_command contains the offset and size of the GNU style
 771  * symbol table information as described in the header file <symseg.h>.
 772  * The symbol roots of the symbol segments must also be aligned properly
 773  * in the file.  So the requirement of keeping the offsets aligned to a
 774  * multiple of a sizeof(long) translates to the length field of the symbol
 775  * roots also being a multiple of a long.  Also the padding must again be
 776  * zeroed. (THIS IS OBSOLETE and no longer supported).
 777  */
 778 struct symseg_command {
 779         unsigned long   cmd;            /* LC_SYMSEG */
 780         unsigned long   cmdsize;        /* sizeof(struct symseg_command) */
 781         unsigned long   offset;         /* symbol segment offset */
 782         unsigned long   size;           /* symbol segment size in bytes */
 783 };
 784
 785 /*
 786  * The ident_command contains a free format string table following the
 787  * ident_command structure.  The strings are null terminated and the size of
 788  * the command is padded out with zero bytes to a multiple of sizeof(long).
 789  * (THIS IS OBSOLETE and no longer supported).
 790  */
 791 struct ident_command {
 792         unsigned long cmd;              /* LC_IDENT */
 793         unsigned long cmdsize;          /* strings that follow this command */
 794 };
 795
 796 /*
 797  * The fvmfile_command contains a reference to a file to be loaded at the
 798  * specified virtual address.  (Presently, this command is reserved for NeXT
 799  * internal use.  The kernel ignores this command when loading a program into
 800  * memory).
 801  */
 802 struct fvmfile_command {
 803         unsigned long cmd;              /* LC_FVMFILE */
 804         unsigned long cmdsize;          /* includes pathname string */
 805         union lc_str    name;           /* files pathname */
 806         unsigned long   header_addr;    /* files virtual address */
 807 };
 808
 809 #endif /*_MACHO_LOADER_H_*/