[lld-macho] Support loading of zippered dylibs

[int3]

Summary

ld64 can emit dylibs that support more than one platform (typically macOS and
macCatalyst). This diff allows LLD to read in those dylibs. Note that this is a
super bare-bones implementation -- in particular, I haven't added support for
LLD to emit those multi-platform dylibs, nor have I added a variety of
validation checks that ld64 does. Until we have a use-case for emitting zippered
dylibs, I think this is good enough.

Fixes PR49597.

[oontvoo]

lld/MachO/InputFiles.cpp

163

perhaps needs pulling and updating to removeSimulator(...) == ...?

(Sorry for the merge conflicts!)

[int3]

rebase

lld/MachO/InputFiles.cpp

163

no worries, I already had it in mind while reviewing your diff :)

[oontvoo]

lld/MachO/InputFiles.h

229

It's a bit unfortunate that this function and the findCommand right above it are very similar ...
could the other one be refactored to return zero, one, or more somehow?

[int3]

lld/MachO/InputFiles.h

229

fair point... done :)

lld/MachO/InputFiles.h

public:
 
   std::unique_ptr<llvm::lto::InputFile> obj;
 };
 
 extern llvm::SetVector<InputFile *> inputFiles;
 
 llvm::Optional<MemoryBufferRef> readFile(StringRef path);
 
-// anyHdr should be a pointer to either mach_header or mach_header_64
-template <class CommandType = llvm::MachO::load_command, class... Types>
-const CommandType *findCommand(const void *anyHdr, Types... types) {
+namespace detail {
+
+template <class CommandType, class... Types>
+std::vector<const CommandType *>
+findCommands(const void *anyHdr, size_t maxCommands, Types... types) {
+  std::vector<const CommandType *> cmds;
   std::initializer_list<uint32_t> typesList{types...};
   const auto *hdr = reinterpret_cast<const llvm::MachO::mach_header *>(anyHdr);
   const uint8_t *p =
       reinterpret_cast<const uint8_t *>(hdr) + target->headerSize;
   for (uint32_t i = 0, n = hdr->ncmds; i < n; ++i) {
     auto *cmd = reinterpret_cast<const CommandType *>(p);
-    if (llvm::is_contained(typesList, cmd->cmd))
-      return cmd;
+    if (llvm::is_contained(typesList, cmd->cmd)) {
+      cmds.push_back(cmd);
+      if (cmds.size() == maxCommands)
+        return cmds;
+    }
     p += cmd->cmdsize;
   }
-  return nullptr;
+  return cmds;
+}
+
+} // namespace detail
+
+// anyHdr should be a pointer to either mach_header or mach_header_64
+template <class CommandType = llvm::MachO::load_command, class... Types>
+const CommandType *findCommand(const void *anyHdr, Types... types) {
+  std::vector<const CommandType *> cmds =
+      detail::findCommands<CommandType>(anyHdr, 1, types...);
+  return cmds.size() ? cmds[0] : nullptr;
+}
+
+template <class CommandType = llvm::MachO::load_command, class... Types>
+std::vector<const CommandType *> findCommands(const void *anyHdr,
+                                              Types... types) {
+  return detail::findCommands<CommandType>(anyHdr, 0, types...);
 }
 
 } // namespace macho
 
 std::string toString(const macho::InputFile *file);
 } // namespace lld
 
 #endif

lld/MachO/InputFiles.cpp

 
 static VersionTuple decodeVersion(uint32_t version) {
   unsigned major = version >> 16;
   unsigned minor = (version >> 8) & 0xffu;
   unsigned subMinor = version & 0xffu;
   return VersionTuple(major, minor, subMinor);
 }
 
-static Optional<PlatformInfo> getPlatformInfo(const InputFile *input) {
+static std::vector<PlatformInfo> getPlatformInfos(const InputFile *input) {
   if (!isa<ObjFile>(input) && !isa<DylibFile>(input))
-    return None;
+    return {};
 
   const char *hdr = input->mb.getBufferStart();
 
-  PlatformInfo platformInfo;
-  if (const auto *cmd =
-          findCommand<build_version_command>(hdr, LC_BUILD_VERSION)) {
-    platformInfo.target.Platform = static_cast<PlatformKind>(cmd->platform);
-    platformInfo.minimum = decodeVersion(cmd->minos);
-    return platformInfo;
+  std::vector<PlatformInfo> platformInfos;
+  for (auto *cmd : findCommands<build_version_command>(hdr, LC_BUILD_VERSION)) {
+    PlatformInfo info;
+    info.target.Platform = static_cast<PlatformKind>(cmd->platform);
+    info.minimum = decodeVersion(cmd->minos);
+    platformInfos.emplace_back(std::move(info));
   }
-  if (const auto *cmd = findCommand<version_min_command>(
+  for (auto *cmd : findCommands<version_min_command>(
            hdr, LC_VERSION_MIN_MACOSX, LC_VERSION_MIN_IPHONEOS,
            LC_VERSION_MIN_TVOS, LC_VERSION_MIN_WATCHOS)) {
+    PlatformInfo info;
     switch (cmd->cmd) {
     case LC_VERSION_MIN_MACOSX:
-      platformInfo.target.Platform = PlatformKind::macOS;
+      info.target.Platform = PlatformKind::macOS;
       break;
     case LC_VERSION_MIN_IPHONEOS:
-      platformInfo.target.Platform = PlatformKind::iOS;
+      info.target.Platform = PlatformKind::iOS;
       break;
     case LC_VERSION_MIN_TVOS:
-      platformInfo.target.Platform = PlatformKind::tvOS;
+      info.target.Platform = PlatformKind::tvOS;
       break;
     case LC_VERSION_MIN_WATCHOS:
-      platformInfo.target.Platform = PlatformKind::watchOS;
+      info.target.Platform = PlatformKind::watchOS;
       break;
     }
-    platformInfo.minimum = decodeVersion(cmd->version);
-    return platformInfo;
+    info.minimum = decodeVersion(cmd->version);
+    platformInfos.emplace_back(std::move(info));
   }
 
-  return None;
+  return platformInfos;
 }
 
 static PlatformKind removeSimulator(PlatformKind platform) {
   // Mapping of platform to simulator and vice-versa.
   static const std::map<PlatformKind, PlatformKind> platformMap = {
       {PlatformKind::iOSSimulator, PlatformKind::iOS},
       {PlatformKind::tvOSSimulator, PlatformKind::tvOS},
       {PlatformKind::watchOSSimulator, PlatformKind::watchOS}};
 
   auto iter = platformMap.find(platform);
   if (iter == platformMap.end())
     return platform;
   return iter->second;
 }
 
 static bool checkCompatibility(const InputFile *input) {
-  Optional<PlatformInfo> platformInfo = getPlatformInfo(input);
-  if (!platformInfo)
+  std::vector<PlatformInfo> platformInfos = getPlatformInfos(input);
+  if (platformInfos.empty())
     return true;
 
-  if (removeSimulator(config->platform()) !=
-      removeSimulator(platformInfo->target.Platform)) {
-    error(toString(input) + " has platform " +
-          getPlatformName(platformInfo->target.Platform) +
+  auto it = find_if(platformInfos, [&](const PlatformInfo &info) {
+    return removeSimulator(info.target.Platform) ==
+           removeSimulator(config->platform());
+  });
+  if (it == platformInfos.end()) {
+    std::string platformNames;
+    raw_string_ostream os(platformNames);
+    interleave(
+        platformInfos, os,
+        [&](const PlatformInfo &info) {
+          os << getPlatformName(info.target.Platform);
+        },
+        "/");
+    error(toString(input) + " has platform " + platformNames +
           Twine(", which is different from target platform ") +
           getPlatformName(config->platform()));
     return false;
   }
-  if (platformInfo->minimum <= config->platformInfo.minimum)
+
+  if (it->minimum <= config->platformInfo.minimum)
     return true;
-  error(toString(input) + " has version " +
-        platformInfo->minimum.getAsString() +
+
+  error(toString(input) + " has version " + it->minimum.getAsString() +
         ", which is newer than target minimum of " +
         config->platformInfo.minimum.getAsString());
   return false;
 }
 
 // Open a given file path and return it as a memory-mapped file.
 Optional<MemoryBufferRef> macho::readFile(StringRef path) {
   // Open a file.

lld/test/MachO/Inputs/MacOSX.sdk/usr/lib/libSystem.tbd

 --- !tapi-tbd
 tbd-version:      4
-targets:          [ x86_64-macos, arm64-macos ]
+targets:          [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
 uuids:
   - target:       x86_64-macos
     value:        00000000-0000-0000-0000-000000000000
+  - target:       x86_64-maccatalyst
+    value:        00000000-0000-0000-0000-000000000000
   - target:       arm64-macos
     value:        00000000-0000-0000-0000-000000000001
 install-name:     '/usr/lib/libSystem.dylib'
 current-version:  0001.001.1
 reexported-libraries:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     libraries:    [ '/usr/lib/system/libdyld.dylib',
                     '/usr/lib/system/libsystem_c.dylib',
                     '/usr/lib/system/libsystem_m.dylib' ]
 --- !tapi-tbd
 tbd-version:      4
-targets:          [ x86_64-macos, arm64-macos ]
+targets:          [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
 uuids:
   - target:       x86_64-macos
     value:        00000000-0000-0000-0000-000000000002
+  - target:       x86_64-maccatalyst
+    value:        00000000-0000-0000-0000-000000000000
   - target:       arm64-macos
     value:        00000000-0000-0000-0000-000000000003
 install-name:     '/usr/lib/system/libdyld.dylib'
 current-version:  0001.001.1
 parent-umbrella:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     umbrella:     System
 exports:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     symbols:      [ dyld_stub_binder, __tlv_bootstrap ]
 --- !tapi-tbd
 tbd-version:      4
-targets:          [ x86_64-macos, arm64-macos ]
+targets:          [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
 uuids:
   - target:       x86_64-macos
     value:        00000000-0000-0000-0000-000000000003
+  - target:       x86_64-maccatalyst
+    value:        00000000-0000-0000-0000-000000000000
   - target:       arm64-macos
     value:        00000000-0000-0000-0000-000000000004
 install-name:     '/usr/lib/system/libsystem_c.dylib'
 current-version:  0001.001.1
 parent-umbrella:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     umbrella:     System
 exports:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     symbols:      [ ]
 --- !tapi-tbd
 tbd-version:      4
-targets:          [ x86_64-macos, arm64-macos ]
+targets:          [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
 uuids:
   - target:       x86_64-macos
     value:        00000000-0000-0000-0000-000000000004
+  - target:       x86_64-maccatalyst
+    value:        00000000-0000-0000-0000-000000000000
   - target:       arm64-macos
     value:        00000000-0000-0000-0000-000000000005
 install-name:     '/usr/lib/system/libsystem_m.dylib'
 current-version:  0001.001.1
 parent-umbrella:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     umbrella:     System
 exports:
-  - targets:      [ x86_64-macos, arm64-macos ]
+  - targets:      [ x86_64-macos, x86_64-maccatalyst, arm64-macos ]
     symbols:      [ ___nan ]
 ...

lld/test/MachO/zippered.yaml

+# REQUIRES: x86
+# RUN: rm -rf %t; mkdir %t
+# RUN: yaml2obj %s > %t/test.dylib
+# RUN: echo "" | llvm-mc -filetype=obj -triple=x86_64-apple-macos10.15 -o %t/test_macos.o
+# RUN: echo "" | llvm-mc -filetype=obj -triple=x86_64-apple-ios13.15.0-macabi -o %t/test_maccatalyst.o
+# RUN: echo "" | llvm-mc -filetype=obj -triple=x86_64-apple-ios13.15.0 -o %t/test_ios.o
+
+# RUN: %lld -lSystem -dylib %t/test.dylib %t/test_macos.o -o /dev/null
+# RUN: %lld -lSystem -dylib -platform_version mac-catalyst 13.15.0 14.0 %t/test.dylib %t/test_maccatalyst.o -o /dev/null
+
+# RUN: not %lld -lSystem -dylib -platform_version ios 13.15.0 14.0 %t/test.dylib %t/test_ios.o -o /dev/null 2>&1 | FileCheck %s
+# CHECK: test.dylib has platform macOS/macCatalyst, which is different from target platform iOS
+
+--- !mach-o
+FileHeader:
+  magic:           0xFEEDFACF
+  cputype:         0x1000007
+  cpusubtype:      0x3
+  filetype:        0x6
+  ncmds:           4
+  sizeofcmds:      600
+  flags:           0x100085
+  reserved:        0x0
+LoadCommands:
+  - cmd:             LC_ID_DYLIB
+    cmdsize:         32
+    dylib:
+      name:            24
+      timestamp:       1
+      current_version: 0
+      compatibility_version: 0
+    PayloadString:   test
+    ZeroPadBytes:    4
+  - cmd:             LC_DYLD_INFO_ONLY
+    cmdsize:         48
+    rebase_off:      0
+    rebase_size:     0
+    bind_off:        0
+    bind_size:       0
+    weak_bind_off:   0
+    weak_bind_size:  0
+    lazy_bind_off:   0
+    lazy_bind_size:  0
+    export_off:      0
+    export_size:     0
+  - cmd:             LC_BUILD_VERSION
+    cmdsize:         32
+    platform:        1
+    minos:           659200
+    sdk:             720896
+    ntools:          1
+    Tools:
+      - tool:            3
+        version:         39913472
+  - cmd:             LC_BUILD_VERSION
+    cmdsize:         32
+    platform:        6
+    minos:           855808
+    sdk:             917504
+    ntools:          1
+    Tools:
+      - tool:            3
+        version:         39913472
+...