llvm/lib/ObjCopy/MachO/MachOObject.h

   1 //===- MachOObject.h - Mach-O object file model -----------------*- C++ -*-===//
   2 //
   3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   4 // See https://llvm.org/LICENSE.txt for license information.
   5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   6 //
   7 //===----------------------------------------------------------------------===//
   8
   9 #ifndef LLVM_LIB_OBJCOPY_MACHO_MACHOOBJECT_H
  10 #define LLVM_LIB_OBJCOPY_MACHO_MACHOOBJECT_H
  11
  12 #include "llvm/ADT/StringRef.h"
  13 #include "llvm/BinaryFormat/MachO.h"
  14 #include "llvm/MC/StringTableBuilder.h"
  15 #include "llvm/ObjectYAML/DWARFYAML.h"
  16 #include "llvm/Support/StringSaver.h"
  17 #include "llvm/Support/YAMLTraits.h"
  18 #include <cstdint>
  19 #include <string>
  20 #include <vector>
  21
  22 namespace llvm {
  23 namespace objcopy {
  24 namespace macho {
  25
  26 struct MachHeader {
  27   uint32_t Magic;
  28   uint32_t CPUType;
  29   uint32_t CPUSubType;
  30   uint32_t FileType;
  31   uint32_t NCmds;
  32   uint32_t SizeOfCmds;
  33   uint32_t Flags;
  34   uint32_t Reserved = 0;
  35 };
  36
  37 struct RelocationInfo;
  38 struct Section {
  39   uint32_t Index;
  40   std::string Segname;
  41   std::string Sectname;
  42   // CanonicalName is a string formatted as “<Segname>,<Sectname>".
  43   std::string CanonicalName;
  44   uint64_t Addr = 0;
  45   uint64_t Size = 0;
  46   // Offset in the input file.
  47   std::optional<uint32_t> OriginalOffset;
  48   uint32_t Offset = 0;
  49   uint32_t Align = 0;
  50   uint32_t RelOff = 0;
  51   uint32_t NReloc = 0;
  52   uint32_t Flags = 0;
  53   uint32_t Reserved1 = 0;
  54   uint32_t Reserved2 = 0;
  55   uint32_t Reserved3 = 0;
  56   StringRef Content;
  57   std::vector<RelocationInfo> Relocations;
  58
  59   Section(StringRef SegName, StringRef SectName);
  60
  61   Section(StringRef SegName, StringRef SectName, StringRef Content);
  62
  63   MachO::SectionType getType() const {
  64     return static_cast<MachO::SectionType>(Flags & MachO::SECTION_TYPE);
  65   }
  66
  67   bool isVirtualSection() const {
  68     return (getType() == MachO::S_ZEROFILL ||
  69             getType() == MachO::S_GB_ZEROFILL ||
  70             getType() == MachO::S_THREAD_LOCAL_ZEROFILL);
  71   }
  72
  73   bool hasValidOffset() const {
  74     return !(isVirtualSection() || (OriginalOffset && *OriginalOffset == 0));
  75   }
  76 };
  77
  78 struct LoadCommand {
  79   // The type MachO::macho_load_command is defined in llvm/BinaryFormat/MachO.h
  80   // and it is a union of all the structs corresponding to various load
  81   // commands.
  82   MachO::macho_load_command MachOLoadCommand;
  83
  84   // The raw content of the payload of the load command (located right after the
  85   // corresponding struct). In some cases it is either empty or can be
  86   // copied-over without digging into its structure.
  87   std::vector<uint8_t> Payload;
  88
  89   // Some load commands can contain (inside the payload) an array of sections,
  90   // though the contents of the sections are stored separately. The struct
  91   // Section describes only sections' metadata and where to find the
  92   // corresponding content inside the binary.
  93   std::vector<std::unique_ptr<Section>> Sections;
  94
  95   // Returns the segment name if the load command is a segment command.
  96   std::optional<StringRef> getSegmentName() const;
  97
  98   // Returns the segment vm address if the load command is a segment command.
  99   std::optional<uint64_t> getSegmentVMAddr() const;
 100 };
 101
 102 // A symbol information. Fields which starts with "n_" are same as them in the
 103 // nlist.
 104 struct SymbolEntry {
 105   std::string Name;
 106   bool Referenced = false;
 107   uint32_t Index;
 108   uint8_t n_type;
 109   uint8_t n_sect;
 110   uint16_t n_desc;
 111   uint64_t n_value;
 112
 113   bool isExternalSymbol() const { return n_type & MachO::N_EXT; }
 114
 115   bool isLocalSymbol() const { return !isExternalSymbol(); }
 116
 117   bool isUndefinedSymbol() const {
 118     return (n_type & MachO::N_TYPE) == MachO::N_UNDF;
 119   }
 120
 121   bool isSwiftSymbol() const {
 122     return StringRef(Name).starts_with("_$s") ||
 123            StringRef(Name).starts_with("_$S");
 124   }
 125
 126   std::optional<uint32_t> section() const {
 127     return n_sect == MachO::NO_SECT ? std::nullopt
 128                                     : std::optional<uint32_t>(n_sect);
 129   }
 130 };
 131
 132 /// The location of the symbol table inside the binary is described by LC_SYMTAB
 133 /// load command.
 134 struct SymbolTable {
 135   std::vector<std::unique_ptr<SymbolEntry>> Symbols;
 136
 137   using iterator = pointee_iterator<
 138       std::vector<std::unique_ptr<SymbolEntry>>::const_iterator>;
 139
 140   iterator begin() const { return iterator(Symbols.begin()); }
 141   iterator end() const { return iterator(Symbols.end()); }
 142
 143   const SymbolEntry *getSymbolByIndex(uint32_t Index) const;
 144   SymbolEntry *getSymbolByIndex(uint32_t Index);
 145   void removeSymbols(
 146       function_ref<bool(const std::unique_ptr<SymbolEntry> &)> ToRemove);
 147 };
 148
 149 struct IndirectSymbolEntry {
 150   // The original value in an indirect symbol table. Higher bits encode extra
 151   // information (INDIRECT_SYMBOL_LOCAL and INDIRECT_SYMBOL_ABS).
 152   uint32_t OriginalIndex;
 153   /// The Symbol referenced by this entry. It's std::nullopt if the index is
 154   /// INDIRECT_SYMBOL_LOCAL or INDIRECT_SYMBOL_ABS.
 155   std::optional<SymbolEntry *> Symbol;
 156
 157   IndirectSymbolEntry(uint32_t OriginalIndex,
 158                       std::optional<SymbolEntry *> Symbol)
 159       : OriginalIndex(OriginalIndex), Symbol(Symbol) {}
 160 };
 161
 162 struct IndirectSymbolTable {
 163   std::vector<IndirectSymbolEntry> Symbols;
 164 };
 165
 166 /// The location of the string table inside the binary is described by LC_SYMTAB
 167 /// load command.
 168 struct StringTable {
 169   std::vector<std::string> Strings;
 170 };
 171
 172 struct RelocationInfo {
 173   // The referenced symbol entry. Set if !Scattered && Extern.
 174   std::optional<const SymbolEntry *> Symbol;
 175   // The referenced section. Set if !Scattered && !Extern.
 176   std::optional<const Section *> Sec;
 177   // True if Info is a scattered_relocation_info.
 178   bool Scattered;
 179   // True if the type is an ADDEND. r_symbolnum holds the addend instead of a
 180   // symbol index.
 181   bool IsAddend;
 182   // True if the r_symbolnum points to a section number (i.e. r_extern=0).
 183   bool Extern;
 184   MachO::any_relocation_info Info;
 185
 186   unsigned getPlainRelocationSymbolNum(bool IsLittleEndian) {
 187     if (IsLittleEndian)
 188       return Info.r_word1 & 0xffffff;
 189     return Info.r_word1 >> 8;
 190   }
 191
 192   void setPlainRelocationSymbolNum(unsigned SymbolNum, bool IsLittleEndian) {
 193     assert(SymbolNum < (1 << 24) && "SymbolNum out of range");
 194     if (IsLittleEndian)
 195       Info.r_word1 = (Info.r_word1 & ~0x00ffffff) | SymbolNum;
 196     else
 197       Info.r_word1 = (Info.r_word1 & ~0xffffff00) | (SymbolNum << 8);
 198   }
 199 };
 200
 201 /// The location of the rebase info inside the binary is described by
 202 /// LC_DYLD_INFO load command. Dyld rebases an image whenever dyld loads it at
 203 /// an address different from its preferred address.  The rebase information is
 204 /// a stream of byte sized opcodes whose symbolic names start with
 205 /// REBASE_OPCODE_. Conceptually the rebase information is a table of tuples:
 206 ///   <seg-index, seg-offset, type>
 207 /// The opcodes are a compressed way to encode the table by only
 208 /// encoding when a column changes.  In addition simple patterns
 209 /// like "every n'th offset for m times" can be encoded in a few
 210 /// bytes.
 211 struct RebaseInfo {
 212   // At the moment we do not parse this info (and it is simply copied over),
 213   // but the proper support will be added later.
 214   ArrayRef<uint8_t> Opcodes;
 215 };
 216
 217 /// The location of the bind info inside the binary is described by
 218 /// LC_DYLD_INFO load command. Dyld binds an image during the loading process,
 219 /// if the image requires any pointers to be initialized to symbols in other
 220 /// images. The bind information is a stream of byte sized opcodes whose
 221 /// symbolic names start with BIND_OPCODE_. Conceptually the bind information is
 222 /// a table of tuples: <seg-index, seg-offset, type, symbol-library-ordinal,
 223 /// symbol-name, addend> The opcodes are a compressed way to encode the table by
 224 /// only encoding when a column changes.  In addition simple patterns like for
 225 /// runs of pointers initialized to the same value can be encoded in a few
 226 /// bytes.
 227 struct BindInfo {
 228   // At the moment we do not parse this info (and it is simply copied over),
 229   // but the proper support will be added later.
 230   ArrayRef<uint8_t> Opcodes;
 231 };
 232
 233 /// The location of the weak bind info inside the binary is described by
 234 /// LC_DYLD_INFO load command. Some C++ programs require dyld to unique symbols
 235 /// so that all images in the process use the same copy of some code/data. This
 236 /// step is done after binding. The content of the weak_bind info is an opcode
 237 /// stream like the bind_info.  But it is sorted alphabetically by symbol name.
 238 /// This enable dyld to walk all images with weak binding information in order
 239 /// and look for collisions.  If there are no collisions, dyld does no updating.
 240 /// That means that some fixups are also encoded in the bind_info.  For
 241 /// instance, all calls to "operator new" are first bound to libstdc++.dylib
 242 /// using the information in bind_info.  Then if some image overrides operator
 243 /// new that is detected when the weak_bind information is processed and the
 244 /// call to operator new is then rebound.
 245 struct WeakBindInfo {
 246   // At the moment we do not parse this info (and it is simply copied over),
 247   // but the proper support will be added later.
 248   ArrayRef<uint8_t> Opcodes;
 249 };
 250
 251 /// The location of the lazy bind info inside the binary is described by
 252 /// LC_DYLD_INFO load command. Some uses of external symbols do not need to be
 253 /// bound immediately. Instead they can be lazily bound on first use.  The
 254 /// lazy_bind contains a stream of BIND opcodes to bind all lazy symbols. Normal
 255 /// use is that dyld ignores the lazy_bind section when loading an image.
 256 /// Instead the static linker arranged for the lazy pointer to initially point
 257 /// to a helper function which pushes the offset into the lazy_bind area for the
 258 /// symbol needing to be bound, then jumps to dyld which simply adds the offset
 259 /// to lazy_bind_off to get the information on what to bind.
 260 struct LazyBindInfo {
 261   ArrayRef<uint8_t> Opcodes;
 262 };
 263
 264 /// The location of the export info inside the binary is described by
 265 /// LC_DYLD_INFO load command. The symbols exported by a dylib are encoded in a
 266 /// trie.  This is a compact representation that factors out common prefixes. It
 267 /// also reduces LINKEDIT pages in RAM because it encodes all information (name,
 268 /// address, flags) in one small, contiguous range. The export area is a stream
 269 /// of nodes.  The first node sequentially is the start node for the trie. Nodes
 270 /// for a symbol start with a uleb128 that is the length of the exported symbol
 271 /// information for the string so far. If there is no exported symbol, the node
 272 /// starts with a zero byte. If there is exported info, it follows the length.
 273 /// First is a uleb128 containing flags. Normally, it is followed by
 274 /// a uleb128 encoded offset which is location of the content named
 275 /// by the symbol from the mach_header for the image.  If the flags
 276 /// is EXPORT_SYMBOL_FLAGS_REEXPORT, then following the flags is
 277 /// a uleb128 encoded library ordinal, then a zero terminated
 278 /// UTF8 string.  If the string is zero length, then the symbol
 279 /// is re-export from the specified dylib with the same name.
 280 /// If the flags is EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER, then following
 281 /// the flags is two uleb128s: the stub offset and the resolver offset.
 282 /// The stub is used by non-lazy pointers.  The resolver is used
 283 /// by lazy pointers and must be called to get the actual address to use.
 284 /// After the optional exported symbol information is a byte of
 285 /// how many edges (0-255) that this node has leaving it,
 286 /// followed by each edge.
 287 /// Each edge is a zero terminated UTF8 of the addition chars
 288 /// in the symbol, followed by a uleb128 offset for the node that
 289 /// edge points to.
 290 struct ExportInfo {
 291   ArrayRef<uint8_t> Trie;
 292 };
 293
 294 struct LinkData {
 295   ArrayRef<uint8_t> Data;
 296 };
 297
 298 struct Object {
 299   MachHeader Header;
 300   std::vector<LoadCommand> LoadCommands;
 301
 302   SymbolTable SymTable;
 303   StringTable StrTable;
 304
 305   RebaseInfo Rebases;
 306   BindInfo Binds;
 307   WeakBindInfo WeakBinds;
 308   LazyBindInfo LazyBinds;
 309   ExportInfo Exports;
 310   IndirectSymbolTable IndirectSymTable;
 311   LinkData DataInCode;
 312   LinkData LinkerOptimizationHint;
 313   LinkData FunctionStarts;
 314   LinkData ExportsTrie;
 315   LinkData ChainedFixups;
 316   LinkData DylibCodeSignDRs;
 317
 318   std::optional<uint32_t> SwiftVersion;
 319
 320   /// The index of LC_CODE_SIGNATURE load command if present.
 321   std::optional<size_t> CodeSignatureCommandIndex;
 322   /// The index of LC_DYLIB_CODE_SIGN_DRS load command if present.
 323   std::optional<size_t> DylibCodeSignDRsIndex;
 324   /// The index of LC_SYMTAB load command if present.
 325   std::optional<size_t> SymTabCommandIndex;
 326   /// The index of LC_DYLD_INFO or LC_DYLD_INFO_ONLY load command if present.
 327   std::optional<size_t> DyLdInfoCommandIndex;
 328   /// The index LC_DYSYMTAB load command if present.
 329   std::optional<size_t> DySymTabCommandIndex;
 330   /// The index LC_DATA_IN_CODE load command if present.
 331   std::optional<size_t> DataInCodeCommandIndex;
 332   /// The index of LC_LINKER_OPTIMIZATIN_HINT load command if present.
 333   std::optional<size_t> LinkerOptimizationHintCommandIndex;
 334   /// The index LC_FUNCTION_STARTS load command if present.
 335   std::optional<size_t> FunctionStartsCommandIndex;
 336   /// The index LC_DYLD_CHAINED_FIXUPS load command if present.
 337   std::optional<size_t> ChainedFixupsCommandIndex;
 338   /// The index LC_DYLD_EXPORTS_TRIE load command if present.
 339   std::optional<size_t> ExportsTrieCommandIndex;
 340   /// The index of the LC_SEGMENT or LC_SEGMENT_64 load command
 341   /// corresponding to the __TEXT segment.
 342   std::optional<size_t> TextSegmentCommandIndex;
 343
 344   BumpPtrAllocator Alloc;
 345   StringSaver NewSectionsContents;
 346
 347   Object() : NewSectionsContents(Alloc) {}
 348
 349   Error
 350   removeSections(function_ref<bool(const std::unique_ptr<Section> &)> ToRemove);
 351
 352   Error removeLoadCommands(function_ref<bool(const LoadCommand &)> ToRemove);
 353
 354   void updateLoadCommandIndexes();
 355
 356   /// Creates a new segment load command in the object and returns a reference
 357   /// to the newly created load command. The caller should verify that SegName
 358   /// is not too long (SegName.size() should be less than or equal to 16).
 359   LoadCommand &addSegment(StringRef SegName, uint64_t SegVMSize);
 360
 361   bool is64Bit() const {
 362     return Header.Magic == MachO::MH_MAGIC_64 ||
 363            Header.Magic == MachO::MH_CIGAM_64;
 364   }
 365
 366   uint64_t nextAvailableSegmentAddress() const;
 367 };
 368
 369 } // end namespace macho
 370 } // end namespace objcopy
 371 } // end namespace llvm
 372
 373 #endif // LLVM_LIB_OBJCOPY_MACHO_MACHOOBJECT_H