[llvm-objcopy][MachO] Add llvm-strip support for newer load commands

Previously llvm-strip would fail because of unknown commands.

Fixes https://bugs.llvm.org/show_bug.cgi?id=50044

Differential Revision: https://reviews.llvm.org/D113734

llvm/test/tools/llvm-objcopy/MachO/Inputs/strip-chained-fixups.yaml

@@ -0,0 +1,238 @@
+--- !mach-o
+FileHeader:
+  magic:           0xFEEDFACF
+  cputype:         0x100000C
+  cpusubtype:      0x0
+  filetype:        0x2
+  ncmds:           16
+  sizeofcmds:      744
+  flags:           0x200085
+  reserved:        0x0
+LoadCommands:
+  - cmd:             LC_SEGMENT_64
+    cmdsize:         72
+    segname:         __PAGEZERO
+    vmaddr:          0
+    vmsize:          4294967296
+    fileoff:         0
+    filesize:        0
+    maxprot:         0
+    initprot:        0
+    nsects:          0
+    flags:           0
+  - cmd:             LC_SEGMENT_64
+    cmdsize:         232
+    segname:         __TEXT
+    vmaddr:          4294967296
+    vmsize:          16384
+    fileoff:         0
+    filesize:        16384
+    maxprot:         5
+    initprot:        5
+    nsects:          2
+    flags:           0
+    Sections:
+      - sectname:        __text
+        segname:         __TEXT
+        addr:            0x100003F98
+        size:            24
+        offset:          0x3F98
+        align:           2
+        reloff:          0x0
+        nreloc:          0
+        flags:           0x80000400
+        reserved1:       0x0
+        reserved2:       0x0
+        reserved3:       0x0
+        content:         C0035FD6FF4300D100008052FF0F00B9FF430091C0035FD6
+      - sectname:        __unwind_info
+        segname:         __TEXT
+        addr:            0x100003FB0
+        size:            80
+        offset:          0x3FB0
+        align:           2
+        reloff:          0x0
+        nreloc:          0
+        flags:           0x0
+        reserved1:       0x0
+        reserved2:       0x0
+        reserved3:       0x0
+        content:         010000001C000000000000001C000000000000001C00000002000000983F00003400000034000000B13F00000000000034000000030000000C0002001400020000000001040000000010000200000002
+  - cmd:             LC_SEGMENT_64
+    cmdsize:         72
+    segname:         __LINKEDIT
+    vmaddr:          4294983680
+    vmsize:          16384
+    fileoff:         16384
+    filesize:        753
+    maxprot:         1
+    initprot:        1
+    nsects:          0
+    flags:           0
+  - cmd:             LC_DYLD_CHAINED_FIXUPS
+    cmdsize:         16
+    dataoff:         16384
+    datasize:        56
+  - cmd:             LC_DYLD_EXPORTS_TRIE
+    cmdsize:         16
+    dataoff:         16440
+    datasize:        56
+  - cmd:             LC_SYMTAB
+    cmdsize:         24
+    symoff:          16504
+    nsyms:           15
+    stroff:          16744
+    strsize:         120
+  - cmd:             LC_DYSYMTAB
+    cmdsize:         80
+    ilocalsym:       0
+    nlocalsym:       12
+    iextdefsym:      12
+    nextdefsym:      3
+    iundefsym:       15
+    nundefsym:       0
+    tocoff:          0
+    ntoc:            0
+    modtaboff:       0
+    nmodtab:         0
+    extrefsymoff:    0
+    nextrefsyms:     0
+    indirectsymoff:  0
+    nindirectsyms:   0
+    extreloff:       0
+    nextrel:         0
+    locreloff:       0
+    nlocrel:         0
+  - cmd:             LC_LOAD_DYLINKER
+    cmdsize:         32
+    name:            12
+    Content:         '/usr/lib/dyld'
+    ZeroPadBytes:    7
+  - cmd:             LC_UUID
+    cmdsize:         24
+    uuid:            F445529E-643C-3A38-8F59-AB64566BCAFF
+  - cmd:             LC_BUILD_VERSION
+    cmdsize:         32
+    platform:        1
+    minos:           786432
+    sdk:             786432
+    ntools:          1
+    Tools:
+      - tool:            3
+        version:         46596096
+  - cmd:             LC_SOURCE_VERSION
+    cmdsize:         16
+    version:         0
+  - cmd:             LC_MAIN
+    cmdsize:         24
+    entryoff:        16284
+    stacksize:       0
+  - cmd:             LC_LOAD_DYLIB
+    cmdsize:         56
+    dylib:
+      name:            24
+      timestamp:       2
+      current_version: 85917696
+      compatibility_version: 65536
+    Content:         '/usr/lib/libSystem.B.dylib'
+    ZeroPadBytes:    6
+  - cmd:             LC_FUNCTION_STARTS
+    cmdsize:         16
+    dataoff:         16496
+    datasize:        8
+  - cmd:             LC_DATA_IN_CODE
+    cmdsize:         16
+    dataoff:         16504
+    datasize:        0
+  - cmd:             LC_CODE_SIGNATURE
+    cmdsize:         16
+    dataoff:         16864
+    datasize:        273
+LinkEditData:
+  NameList:
+    - n_strx:          33
+      n_type:          0x64
+      n_sect:          0
+      n_desc:          0
+      n_value:         0
+    - n_strx:          39
+      n_type:          0x64
+      n_sect:          0
+      n_desc:          0
+      n_value:         0
+    - n_strx:          46
+      n_type:          0x66
+      n_sect:          0
+      n_desc:          1
+      n_value:         1636754403
+    - n_strx:          1
+      n_type:          0x2E
+      n_sect:          1
+      n_desc:          0
+      n_value:         4294983576
+    - n_strx:          109
+      n_type:          0x24
+      n_sect:          1
+      n_desc:          0
+      n_value:         4294983576
+    - n_strx:          1
+      n_type:          0x24
+      n_sect:          0
+      n_desc:          0
+      n_value:         4
+    - n_strx:          1
+      n_type:          0x4E
+      n_sect:          1
+      n_desc:          0
+      n_value:         4
+    - n_strx:          1
+      n_type:          0x2E
+      n_sect:          1
+      n_desc:          0
+      n_value:         4294983580
+    - n_strx:          114
+      n_type:          0x24
+      n_sect:          1
+      n_desc:          0
+      n_value:         4294983580
+    - n_strx:          1
+      n_type:          0x24
+      n_sect:          0
+      n_desc:          0
+      n_value:         20
+    - n_strx:          1
+      n_type:          0x4E
+      n_sect:          1
+      n_desc:          0
+      n_value:         20
+    - n_strx:          1
+      n_type:          0x64
+      n_sect:          1
+      n_desc:          0
+      n_value:         0
+    - n_strx:          2
+      n_type:          0xF
+      n_sect:          1
+      n_desc:          16
+      n_value:         4294967296
+    - n_strx:          22
+      n_type:          0xF
+      n_sect:          1
+      n_desc:          0
+      n_value:         4294983576
+    - n_strx:          27
+      n_type:          0xF
+      n_sect:          1
+      n_desc:          0
+      n_value:         4294983580
+  StringTable:
+    - ' '
+    - __mh_execute_header
+    - _foo
+    - _main
+    - '/tmp/'
+    - main.c
+    - '/var/folders/gj/wf3swl0x215b2sq1qy84kzkm0000gn/T/main-e32fe7.o'
+    - _foo
+    - _main
+...

llvm/test/tools/llvm-objcopy/MachO/strip-all.test

@@ -2,6 +2,7 @@
 
 # RUN: yaml2obj %p/Inputs/strip-all.yaml -o %t.exec
 # RUN: yaml2obj %p/Inputs/strip-all-with-dwarf.yaml -o %t.dwarf
+# RUN: yaml2obj %p/Inputs/strip-chained-fixups.yaml -o %t.fixups
 
 ## Check that the symbol list satisfies the order: local / defined external /
 ## undefined external, otherwise llvm-objcopy will fail.
@@ -15,6 +16,10 @@
 # RUN: llvm-readobj --sections --relocations --symbols %t.dwarf.stripped \
 # RUN:   | FileCheck --check-prefixes=COMMON,DWARF %s
 
+# RUN: llvm-objcopy --strip-all %t.fixups %t.fixups.stripped
+# RUN: llvm-readobj --sections --relocations --symbols %t.fixups.stripped \
+# RUN:   | FileCheck --check-prefix=COMMON %s
+
 ## The output of "llvm-strip" should be identical with that of
 ## "llvm-strip --strip-all" and "llvm-objcopy --strip-all".
 # RUN: llvm-strip %t.exec -o %t2
@@ -32,6 +37,11 @@
 # RUN: llvm-lipo %t.dwarf.universal.stripped -thin x86_64 -output %t6
 # RUN: cmp %t6 %t.dwarf.stripped
 
+# RUN: llvm-strip %t.exec -o %t7
+# RUN: llvm-strip --strip-all %t.exec -o %t8
+# cmp %t7 %t.fixups.stripped
+# cmp %t8 %t.fixups.stripped
+
 ## Make sure that debug sections are removed.
 # DWARF:       Sections [
 # DWARF-NOT:     Name: __debug_str

llvm/tools/llvm-objcopy/MachO/MachOLayoutBuilder.cpp

@@ -249,8 +249,12 @@ Error MachOLayoutBuilder::layoutTail(uint64_t Offset) {
   uint64_t StartOfExportTrie =
       StartOfLazyBindingInfo + O.LazyBinds.Opcodes.size();
   uint64_t StartOfFunctionStarts = StartOfExportTrie + O.Exports.Trie.size();
-  uint64_t StartOfDataInCode =
+  uint64_t StartOfDyldExportsTrie =
       StartOfFunctionStarts + O.FunctionStarts.Data.size();
+  uint64_t StartOfChainedFixups =
+      StartOfDyldExportsTrie + O.ExportsTrie.Data.size();
+  uint64_t StartOfDataInCode =
+      StartOfChainedFixups + O.ChainedFixups.Data.size();
   uint64_t StartOfLinkerOptimizationHint =
       StartOfDataInCode + O.DataInCode.Data.size();
   uint64_t StartOfSymbols =
@@ -353,6 +357,14 @@ Error MachOLayoutBuilder::layoutTail(uint64_t Offset) {
       MLC.linkedit_data_command_data.dataoff = StartOfFunctionStarts;
       MLC.linkedit_data_command_data.datasize = O.FunctionStarts.Data.size();
       break;
+    case MachO::LC_DYLD_CHAINED_FIXUPS:
+      MLC.linkedit_data_command_data.dataoff = StartOfChainedFixups;
+      MLC.linkedit_data_command_data.datasize = O.ChainedFixups.Data.size();
+      break;
+    case MachO::LC_DYLD_EXPORTS_TRIE:
+      MLC.linkedit_data_command_data.dataoff = StartOfDyldExportsTrie;
+      MLC.linkedit_data_command_data.datasize = O.ExportsTrie.Data.size();
+      break;
     case MachO::LC_DYLD_INFO:
     case MachO::LC_DYLD_INFO_ONLY:
       MLC.dyld_info_command_data.rebase_off =

llvm/tools/llvm-objcopy/MachO/MachOReader.cpp

@@ -174,6 +174,12 @@ Error MachOReader::readLoadCommands(Object &O) const {
     case MachO::LC_FUNCTION_STARTS:
       O.FunctionStartsCommandIndex = O.LoadCommands.size();
       break;
+    case MachO::LC_DYLD_EXPORTS_TRIE:
+      O.ExportsTrieCommandIndex = O.LoadCommands.size();
+      break;
+    case MachO::LC_DYLD_CHAINED_FIXUPS:
+      O.ChainedFixupsCommandIndex = O.LoadCommands.size();
+      break;
     }
 #define HANDLE_LOAD_COMMAND(LCName, LCValue, LCStruct)                         \
   case MachO::LCName:                                                          \
@@ -301,6 +307,14 @@ void MachOReader::readFunctionStartsData(Object &O) const {
   return readLinkData(O, O.FunctionStartsCommandIndex, O.FunctionStarts);
 }
 
+void MachOReader::readExportsTrie(Object &O) const {
+  return readLinkData(O, O.ExportsTrieCommandIndex, O.ExportsTrie);
+}
+
+void MachOReader::readChainedFixups(Object &O) const {
+  return readLinkData(O, O.ChainedFixupsCommandIndex, O.ChainedFixups);
+}
+
 void MachOReader::readIndirectSymbolTable(Object &O) const {
   MachO::dysymtab_command DySymTab = MachOObj.getDysymtabLoadCommand();
   constexpr uint32_t AbsOrLocalMask =
@@ -352,6 +366,8 @@ Expected<std::unique_ptr<Object>> MachOReader::create() const {
   readDataInCodeData(*Obj);
   readLinkerOptimizationHint(*Obj);
   readFunctionStartsData(*Obj);
+  readExportsTrie(*Obj);
+  readChainedFixups(*Obj);
   readIndirectSymbolTable(*Obj);
   readSwiftVersion(*Obj);
   return std::move(Obj);

llvm/tools/llvm-objcopy/MachO/MachOReader.h

@@ -41,6 +41,8 @@ class MachOReader : public Reader {
   void readDataInCodeData(Object &O) const;
   void readLinkerOptimizationHint(Object &O) const;
   void readFunctionStartsData(Object &O) const;
+  void readExportsTrie(Object &O) const;
+  void readChainedFixups(Object &O) const;
   void readIndirectSymbolTable(Object &O) const;
   void readSwiftVersion(Object &O) const;
 

llvm/tools/llvm-objcopy/MachO/MachOWriter.cpp

@@ -133,6 +133,26 @@ size_t MachOWriter::totalSize() const {
                      LinkEditDataCommand.datasize);
   }
 
+  if (O.ChainedFixupsCommandIndex) {
+    const MachO::linkedit_data_command &LinkEditDataCommand =
+        O.LoadCommands[*O.ChainedFixupsCommandIndex]
+            .MachOLoadCommand.linkedit_data_command_data;
+
+    if (LinkEditDataCommand.dataoff)
+      Ends.push_back(LinkEditDataCommand.dataoff +
+                     LinkEditDataCommand.datasize);
+  }
+
+  if (O.ExportsTrieCommandIndex) {
+    const MachO::linkedit_data_command &LinkEditDataCommand =
+        O.LoadCommands[*O.ExportsTrieCommandIndex]
+            .MachOLoadCommand.linkedit_data_command_data;
+
+    if (LinkEditDataCommand.dataoff)
+      Ends.push_back(LinkEditDataCommand.dataoff +
+                     LinkEditDataCommand.datasize);
+  }
+
   // Otherwise, use the last section / reloction.
   for (const LoadCommand &LC : O.LoadCommands)
     for (const std::unique_ptr<Section> &S : LC.Sections) {
@@ -585,6 +605,14 @@ void MachOWriter::writeFunctionStartsData() {
   return writeLinkData(O.FunctionStartsCommandIndex, O.FunctionStarts);
 }
 
+void MachOWriter::writeChainedFixupsData() {
+  return writeLinkData(O.ChainedFixupsCommandIndex, O.ChainedFixups);
+}
+
+void MachOWriter::writeExportsTrieData() {
+  return writeLinkData(O.ExportsTrieCommandIndex, O.ExportsTrie);
+}
+
 void MachOWriter::writeTail() {
   typedef void (MachOWriter::*WriteHandlerType)(void);
   typedef std::pair<uint64_t, WriteHandlerType> WriteOperation;
@@ -670,6 +698,26 @@ void MachOWriter::writeTail() {
                          &MachOWriter::writeFunctionStartsData);
   }
 
+  if (O.ChainedFixupsCommandIndex) {
+    const MachO::linkedit_data_command &LinkEditDataCommand =
+        O.LoadCommands[*O.ChainedFixupsCommandIndex]
+            .MachOLoadCommand.linkedit_data_command_data;
+
+    if (LinkEditDataCommand.dataoff)
+      Queue.emplace_back(LinkEditDataCommand.dataoff,
+                         &MachOWriter::writeChainedFixupsData);
+  }
+
+  if (O.ExportsTrieCommandIndex) {
+    const MachO::linkedit_data_command &LinkEditDataCommand =
+        O.LoadCommands[*O.ExportsTrieCommandIndex]
+            .MachOLoadCommand.linkedit_data_command_data;
+
+    if (LinkEditDataCommand.dataoff)
+      Queue.emplace_back(LinkEditDataCommand.dataoff,
+                         &MachOWriter::writeExportsTrieData);
+  }
+
   llvm::sort(Queue, [](const WriteOperation &LHS, const WriteOperation &RHS) {
     return LHS.first < RHS.first;
   });

llvm/tools/llvm-objcopy/MachO/MachOWriter.h

@@ -50,6 +50,8 @@ class MachOWriter {
   void writeDataInCodeData();
   void writeLinkerOptimizationHint();
   void writeFunctionStartsData();
+  void writeChainedFixupsData();
+  void writeExportsTrieData();
   void writeTail();
 
 public:

llvm/tools/llvm-objcopy/MachO/Object.cpp

@@ -66,6 +66,12 @@ void Object::updateLoadCommandIndexes() {
     case MachO::LC_FUNCTION_STARTS:
       FunctionStartsCommandIndex = Index;
       break;
+    case MachO::LC_DYLD_CHAINED_FIXUPS:
+      ChainedFixupsCommandIndex = Index;
+      break;
+    case MachO::LC_DYLD_EXPORTS_TRIE:
+      ExportsTrieCommandIndex = Index;
+      break;
     }
   }
 }

llvm/tools/llvm-objcopy/MachO/Object.h

@@ -315,6 +315,8 @@ struct Object {
   LinkData DataInCode;
   LinkData LinkerOptimizationHint;
   LinkData FunctionStarts;
+  LinkData ExportsTrie;
+  LinkData ChainedFixups;
 
   Optional<uint32_t> SwiftVersion;
 
@@ -332,6 +334,10 @@ struct Object {
   Optional<size_t> LinkerOptimizationHintCommandIndex;
   /// The index LC_FUNCTION_STARTS load comamnd if present.
   Optional<size_t> FunctionStartsCommandIndex;
+  /// The index LC_DYLD_CHAINED_FIXUPS load comamnd if present.
+  Optional<size_t> ChainedFixupsCommandIndex;
+  /// The index LC_DYLD_EXPORTS_TRIE load comamnd if present.
+  Optional<size_t> ExportsTrieCommandIndex;
   /// The index of the LC_SEGMENT or LC_SEGMENT_64 load command
   /// corresponding to the __TEXT segment.
   Optional<size_t> TextSegmentCommandIndex;