[mach-o] Support linker synthesized mach_header symbols.

On darwin in final linked images, the __TEXT segment covers that start of the
file.  That means in memory a process can see the mach_header (and load commands)
for every loaded image in a process.  There are APIs that take and return the
mach_header addresses as a way to specify a particular loaded image.

For completeness, any code can get the address of the mach_header of the image
it is in by using &__dso_handle.  In addition there are mach-o type specific
symbols like __mh_execute_header.

The linker needs to supply a definition for any of these symbols if used.  But
the address the symbol it resolves to is not in any section.  Instead it is the
address of the start of the __TEXT segment.

I needed to make a small change to SimpleFileNode to not override
resetNextIndex() because the Driver creates a SimpleFileNode to hold the
internal/implicit files that the context/writer can create. For some reason
SimpleFileNode overrode resetNextIndex() to do nothing instead of reseting
the index (which mach-o needs if the internal file is an archive).

llvm-svn: 221822

10 files changed, 257 insertions, 4 deletions

diff --git a/lld/include/lld/Core/DefinedAtom.h b/lld/include/lld/Core/DefinedAtom.h
index 38e58382eed..cc3b9e674f0 100644
--- a/lld/include/lld/Core/DefinedAtom.h
+++ b/lld/include/lld/Core/DefinedAtom.h
@@ -143,6 +143,7 @@ public:
     typeTLVInitialData,     // initial data for a TLV [Darwin]
     typeTLVInitialZeroFill, // TLV initial zero fill data [Darwin]
     typeTLVInitializerPtr,  // pointer to thread local initializer [Darwin]
+    typeMachHeader,         // atom representing mach_header [Darwin]
     typeThreadZeroFill,     // Uninitialized thread local data(TBSS) [ELF]
     typeThreadData,         // Initialized thread local data(TDATA) [ELF]
     typeRONote,             // Identifies readonly note sections [ELF]
diff --git a/lld/include/lld/Core/InputGraph.h b/lld/include/lld/Core/InputGraph.h
index ef408f07b6f..f72a2120e21 100644
--- a/lld/include/lld/Core/InputGraph.h
+++ b/lld/include/lld/Core/InputGraph.h
@@ -296,9 +296,6 @@ public:
       return make_error_code(InputGraphError::no_more_files);
     return *_files[_nextFileIndex++];
   }
-
-  // Do nothing here.
-  void resetNextIndex() override {}
 };
 } // namespace lld
 
diff --git a/lld/lib/Core/DefinedAtom.cpp b/lld/lib/Core/DefinedAtom.cpp
index fa2fa45f695..298a768b111 100644
--- a/lld/lib/Core/DefinedAtom.cpp
+++ b/lld/lib/Core/DefinedAtom.cpp
@@ -28,6 +28,7 @@ DefinedAtom::ContentPermissions DefinedAtom::permissions(ContentType type) {
   case typeBranchShim:
   case typeStub:
   case typeStubHelper:
+  case typeMachHeader:
     return permR_X;
 
   case typeConstant:
diff --git a/lld/lib/ReaderWriter/MachO/ExecutableAtoms.hpp b/lld/lib/ReaderWriter/MachO/ExecutableAtoms.hpp
index f0788fa3da1..693a428d0df 100644
--- a/lld/lib/ReaderWriter/MachO/ExecutableAtoms.hpp
+++ b/lld/lib/ReaderWriter/MachO/ExecutableAtoms.hpp
@@ -10,8 +10,11 @@
 #ifndef LLD_READER_WRITER_MACHO_EXECUTABLE_ATOMS_H
 #define LLD_READER_WRITER_MACHO_EXECUTABLE_ATOMS_H
 
+#include "Atoms.h"
+
 #include "llvm/Support/MachO.h"
 
+#include "lld/Core/ArchiveLibraryFile.h"
 #include "lld/Core/DefinedAtom.h"
 #include "lld/Core/File.h"
 #include "lld/Core/LinkingContext.h"
@@ -58,8 +61,75 @@ private:
 };
 
 
+//
+// MachHeaderAliasFile lazily instantiates the magic symbols that mark the start
+// of the mach_header for final linked images.
+//
+class MachHeaderAliasFile : public ArchiveLibraryFile {
+public:
+  MachHeaderAliasFile(const MachOLinkingContext &context)
+      : ArchiveLibraryFile("mach_header symbols") {
+      switch (context.outputMachOType()) {
+      case llvm::MachO::MH_EXECUTE:
+        _machHeaderSymbolName = "__mh_execute_header";
+        break;
+      case llvm::MachO::MH_DYLIB:
+        _machHeaderSymbolName = "__mh_dylib_header";
+        break;
+      case llvm::MachO::MH_BUNDLE:
+        _machHeaderSymbolName = "__mh_bundle_header";
+        break;
+      case llvm::MachO::MH_DYLINKER:
+        _machHeaderSymbolName = "__mh_dylinker_header";
+        break;
+      case llvm::MachO::MH_PRELOAD:
+        _machHeaderSymbolName = "__mh_preload_header";
+        break;
+      default:
+        llvm_unreachable("no mach_header symbol for file type");
+      }
+  }
+
+  std::error_code
+  parseAllMembers(std::vector<std::unique_ptr<File>> &result) const override {
+    return std::error_code();
+  }
+
+  const File *find(StringRef sym, bool dataSymbolOnly) const override {
+    if (sym.equals("___dso_handle") || sym.equals(_machHeaderSymbolName)) {
+      _definedAtoms._atoms.push_back(new (_alloc) MachODefinedAtom(
+          *this, sym, DefinedAtom::scopeLinkageUnit,
+          DefinedAtom::typeMachHeader, DefinedAtom::mergeNo, false, false,
+          ArrayRef<uint8_t>()));
+      return this;
+    }
+    return nullptr;
+  }
+
+  const atom_collection<DefinedAtom> &defined() const override {
+    return _definedAtoms;
+  }
+  const atom_collection<UndefinedAtom> &undefined() const override {
+    return _undefinedAtoms;
+  }
 
+  const atom_collection<SharedLibraryAtom> &sharedLibrary() const override {
+    return _sharedLibraryAtoms;
+  }
 
+  const atom_collection<AbsoluteAtom> &absolute() const override {
+    return _absoluteAtoms;
+  }
+
+
+private:
+  mutable atom_collection_vector<DefinedAtom> _definedAtoms;
+  atom_collection_vector<UndefinedAtom> _undefinedAtoms;
+  atom_collection_vector<SharedLibraryAtom> _sharedLibraryAtoms;
+  atom_collection_vector<AbsoluteAtom> _absoluteAtoms;
+  StringRef _machHeaderSymbolName;
+  mutable llvm::BumpPtrAllocator _alloc;
+};
 
 } // namespace mach_o
 } // namespace lld
diff --git a/lld/lib/ReaderWriter/MachO/MachONormalizedFileFromAtoms.cpp b/lld/lib/ReaderWriter/MachO/MachONormalizedFileFromAtoms.cpp
index 8333d099ccc..780c9c02fc8 100644
--- a/lld/lib/ReaderWriter/MachO/MachONormalizedFileFromAtoms.cpp
+++ b/lld/lib/ReaderWriter/MachO/MachONormalizedFileFromAtoms.cpp
@@ -167,6 +167,7 @@ private:
   DylibPathToInfo               _dylibInfo;
   const DefinedAtom            *_entryAtom;
   AtomToIndex                   _atomToSymbolIndex;
+  std::vector<const Atom *>     _machHeaderAliasAtoms;
 };
 
 
@@ -331,7 +332,10 @@ void Util::appendAtom(SectionInfo *sect, const DefinedAtom *atom) {
 
 void Util::assignAtomsToSections(const lld::File &atomFile) {
   for (const DefinedAtom *atom : atomFile.defined()) {
-    appendAtom(sectionForAtom(atom), atom);
+    if (atom->contentType() == DefinedAtom::typeMachHeader)
+      _machHeaderAliasAtoms.push_back(atom);
+    else
+      appendAtom(sectionForAtom(atom), atom);
   }
 }
 
@@ -653,6 +657,14 @@ void Util::buildAtomToAddressMap() {
               << " name=" << info.atom->name() << "\n");
     }
   }
+  for (const Atom *atom : _machHeaderAliasAtoms) {
+    _atomToAddress[atom] = _context.baseAddress();
+    DEBUG_WITH_TYPE("WriterMachO-address", llvm::dbgs()
+              << "   address="
+              << llvm::format("0x%016X", _atomToAddress[atom])
+              << " atom=" << atom
+              << " name=" << atom->name() << "\n");
+  }
 }
 
 uint16_t Util::descBits(const DefinedAtom* atom) {
diff --git a/lld/lib/ReaderWriter/MachO/WriterMachO.cpp b/lld/lib/ReaderWriter/MachO/WriterMachO.cpp
index 17102df6349..a0e9d434c9a 100644
--- a/lld/lib/ReaderWriter/MachO/WriterMachO.cpp
+++ b/lld/lib/ReaderWriter/MachO/WriterMachO.cpp
@@ -52,6 +52,9 @@ public:
     // If this can link with dylibs, need helper function (dyld_stub_binder).
     if (_context.needsStubsPass())
       r.emplace_back(new StubHelperFile(_context));
+    // Final linked images can access a symbol for their mach_header.
+    if (_context.outputMachOType() != llvm::MachO::MH_OBJECT)
+      r.emplace_back(new MachHeaderAliasFile(_context));
 
     return true;
   }
diff --git a/lld/lib/ReaderWriter/YAML/ReaderWriterYAML.cpp b/lld/lib/ReaderWriter/YAML/ReaderWriterYAML.cpp
index 72537b94cd3..6993ee4a099 100644
--- a/lld/lib/ReaderWriter/YAML/ReaderWriterYAML.cpp
+++ b/lld/lib/ReaderWriter/YAML/ReaderWriterYAML.cpp
@@ -447,6 +447,7 @@ template <> struct ScalarEnumerationTraits<lld::DefinedAtom::ContentType> {
     io.enumCase(value, "tlv-zero-fill",   DefinedAtom::typeTLVInitialZeroFill);
     io.enumCase(value, "tlv-initializer-ptr",
                                           DefinedAtom::typeTLVInitializerPtr);
+    io.enumCase(value, "mach_header",     DefinedAtom::typeMachHeader);
     io.enumCase(value, "thread-data",     DefinedAtom::typeThreadData);
     io.enumCase(value, "thread-zero-fill",DefinedAtom::typeThreadZeroFill);
     io.enumCase(value, "ro-note",         DefinedAtom::typeRONote);
diff --git a/lld/test/mach-o/dso_handle.yaml b/lld/test/mach-o/dso_handle.yaml
new file mode 100644
index 00000000000..d39d2c23080
--- /dev/null
+++ b/lld/test/mach-o/dso_handle.yaml
@@ -0,0 +1,62 @@
+# RUN: lld -flavor darwin -arch x86_64  %s  %p/Inputs/libSystem.yaml -o %t1
+# RUN: llvm-nm -m -n %t1 | FileCheck %s
+#
+# RUN: lld -flavor darwin -arch x86_64  %s %p/Inputs/libSystem.yaml -dead_strip -o %t2
+# RUN: llvm-nm -m -n %t2 | FileCheck %s
+#
+# RUN: lld -flavor darwin -arch x86_64  %s %p/Inputs/libSystem.yaml -dylib -o %t3
+# RUN: llvm-nm -m -n %t3 | FileCheck %s
+#
+# RUN: lld -flavor darwin -arch x86_64  %s %p/Inputs/libSystem.yaml -bundle -o %t4
+# RUN: llvm-nm -m -n %t4 | FileCheck %s
+#
+# Test that ___dso_handle symbol is available for executables, bundles, and dylibs
+#
+
+--- !mach-o
+arch:            x86_64
+file-type:       MH_OBJECT
+flags:           [ MH_SUBSECTIONS_VIA_SYMBOLS ]
+sections:
+  - segment:         __TEXT
+    section:         __text
+    type:            S_REGULAR
+    attributes:      [ S_ATTR_PURE_INSTRUCTIONS, S_ATTR_SOME_INSTRUCTIONS ]
+    address:         0x0000000000000000
+    content:         [ 0x55, 0x48, 0x89, 0xE5, 0x31, 0xC0, 0x5D, 0xC3 ]
+  - segment:         __DATA
+    section:         __data
+    type:            S_REGULAR
+    attributes:      [  ]
+    alignment:       3
+    address:         0x0000000000000008
+    content:         [ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ]
+    relocations:
+      - offset:          0x00000000
+        type:            X86_64_RELOC_UNSIGNED
+        length:          3
+        pc-rel:          false
+        extern:          true
+        symbol:          2
+global-symbols:
+  - name:            _d
+    type:            N_SECT
+    scope:           [ N_EXT ]
+    sect:            2
+    value:           0x0000000000000008
+  - name:            _main
+    type:            N_SECT
+    scope:           [ N_EXT ]
+    sect:            1
+    value:           0x0000000000000000
+undefined-symbols:
+  - name:            ___dso_handle
+    type:            N_UNDF
+    scope:           [ N_EXT ]
+    value:           0x0000000000000000
+
+
+...
+
+# CHECK_NOT:	___dso_handle
+# CHECK:      _main
diff --git a/lld/test/mach-o/mh_bundle_header.yaml b/lld/test/mach-o/mh_bundle_header.yaml
new file mode 100644
index 00000000000..558df2ca2e9
--- /dev/null
+++ b/lld/test/mach-o/mh_bundle_header.yaml
@@ -0,0 +1,53 @@
+# RUN: lld -flavor darwin -arch x86_64  %s -bundle -o %t %p/Inputs/libSystem.yaml
+# RUN: llvm-nm -m -n %t | FileCheck %s
+#
+# Test that __mh_bundle_header symbol is available for bundles
+#
+
+--- !mach-o
+arch:            x86_64
+file-type:       MH_OBJECT
+flags:           [ MH_SUBSECTIONS_VIA_SYMBOLS ]
+sections:
+  - segment:         __TEXT
+    section:         __text
+    type:            S_REGULAR
+    attributes:      [ S_ATTR_PURE_INSTRUCTIONS, S_ATTR_SOME_INSTRUCTIONS ]
+    address:         0x0000000000000000
+    content:         [ 0x55, 0x48, 0x89, 0xE5, 0x31, 0xC0, 0x5D, 0xC3 ]
+  - segment:         __DATA
+    section:         __data
+    type:            S_REGULAR
+    attributes:      [  ]
+    alignment:       3
+    address:         0x0000000000000008
+    content:         [ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ]
+    relocations:
+      - offset:          0x00000000
+        type:            X86_64_RELOC_UNSIGNED
+        length:          3
+        pc-rel:          false
+        extern:          true
+        symbol:          2
+global-symbols:
+  - name:            _d
+    type:            N_SECT
+    scope:           [ N_EXT ]
+    sect:            2
+    value:           0x0000000000000008
+  - name:            _foo
+    type:            N_SECT
+    scope:           [ N_EXT ]
+    sect:            1
+    value:           0x0000000000000000
+undefined-symbols:
+  - name:            __mh_bundle_header
+    type:            N_UNDF
+    scope:           [ N_EXT ]
+    value:           0x0000000000000000
+
+
+...
+
+# CHECK_NOT:	__mh_bundle_header
+# CHECK:      _foo
diff --git a/lld/test/mach-o/mh_dylib_header.yaml b/lld/test/mach-o/mh_dylib_header.yaml
new file mode 100644
index 00000000000..07429b30c94
--- /dev/null
+++ b/lld/test/mach-o/mh_dylib_header.yaml
@@ -0,0 +1,53 @@
+# RUN: lld -flavor darwin -arch x86_64  %s -dylib -o %t %p/Inputs/libSystem.yaml
+# RUN: llvm-nm -m -n %t | FileCheck %s
+#
+# Test that __mh_dylib_header symbol is available for dylibs
+#
+
+--- !mach-o
+arch:            x86_64
+file-type:       MH_OBJECT
+flags:           [ MH_SUBSECTIONS_VIA_SYMBOLS ]
+sections:
+  - segment:         __TEXT
+    section:         __text
+    type:            S_REGULAR
+    attributes:      [ S_ATTR_PURE_INSTRUCTIONS, S_ATTR_SOME_INSTRUCTIONS ]
+    address:         0x0000000000000000
+    content:         [ 0x55, 0x48, 0x89, 0xE5, 0x31, 0xC0, 0x5D, 0xC3 ]
+  - segment:         __DATA
+    section:         __data
+    type:            S_REGULAR
+    attributes:      [  ]
+    alignment:       3
+    address:         0x0000000000000008
+    content:         [ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 ]
+    relocations:
+      - offset:          0x00000000
+        type:            X86_64_RELOC_UNSIGNED
+        length:          3
+        pc-rel:          false
+        extern:          true
+        symbol:          2
+global-symbols:
+  - name:            _d
+    type:            N_SECT
+    scope:           [ N_EXT ]
+    sect:            2
+    value:           0x0000000000000008
+  - name:            _foo
+    type:            N_SECT
+    scope:           [ N_EXT ]
+    sect:            1
+    value:           0x0000000000000000
+undefined-symbols:
+  - name:            __mh_dylib_header
+    type:            N_UNDF
+    scope:           [ N_EXT ]
+    value:           0x0000000000000000
+
+
+...
+
+# CHECK_NOT:	__mh_dylib_header
+# CHECK:      _foo