Support Universal dSYM files in llvm-objdump

[mtrent]

Summary

Commonly programmers use llvm-objdump to disassemble Mach-O target
binaries with Mach-O dSYMS. While llvm-objdump allows programmers to
disassemble Universal binaries, it previously did not recognize
Universal dSYM files. This change updates llvm-objdump to support
passing in Universal files via the -dsym option. Now, when
disassembling a Mach-O file either as a stand alone file or as an entry
in a Universal binariy, llvm-objdump will search through a Universal
dSYM for a Mach-O matching the architecture flag of the file being
disassembled.

[mtrent]

Increasing error checking to catch symbolic files that are neither
Mach-O nor Universal object files.

[pete]

LGTM.

tools/llvm-objdump/MachODump.cpp

7294 ↗

(On Diff #202200)

The formatting looks off here compared to the rest of the code.

[mtrent]

Remove tabs inserted by emacs (grr)

[mtrent]

tools/llvm-objdump/MachODump.cpp

7294 ↗

(On Diff #202200)

stupid emacs.

llvm/trunk/test/tools/llvm-objdump/X86/macho-disassemble-g-dsym.test

 // RUN: dsymutil -f -oso-prepend-path=%p/../../dsymutil/ %p/../../dsymutil/Inputs/basic.macho.x86_64 -o %t1.dSYM
 // RUN: llvm-objdump -d -g -dsym=%t1.dSYM %p/../../dsymutil/Inputs/basic.macho.x86_64 >%t0
 // RUN: llvm-objdump -dg -dsym=%t1.dSYM %p/../../dsymutil/Inputs/basic.macho.x86_64 >%t1
 // RUN: cmp %t0 %t1
 // RUN: FileCheck --input-file %t0 %s
 
 CHECK: Disassembly of section __TEXT,__text:
+
+// RUN: dsymutil %p/../../dsymutil/Inputs/fat-test.dylib -o fat-test.dylib.dSYM
+// RUN: llvm-objdump -m -d -g -dsym fat-test.dylib.dSYM/Contents/Resources/DWARF/fat-test.dylib %p/../../dsymutil/Inputs/fat-test.dylib | FileCheck -check-prefix MACHO_DSYM %s
+// RUN: dsymutil %p/../../dsymutil/Inputs/basic.macho.x86_64 -o basic.macho.x86_64.dSYM
+// RUN: llvm-objdump -m -d -g -dsym basic.macho.x86_64.dSYM/Contents/Resources/DWARF/basic.macho.x86_64 %p/../../dsymutil/Inputs/basic.macho.x86_64 | FileCheck -check-prefix MACHO_DSYM %s
+
+MACHO_DSYM: (__TEXT,__text) section
+
+// RUN: llvm-objdump -m -d -g -dsym %p/../Inputs/libbogus11.a %p/../../dsymutil/Inputs/basic.macho.x86_64 2>&1 | FileCheck -check-prefix BAD_INPUT %s
+
+BAD_INPUT: is not a Mach-O or Universal file type.

llvm/trunk/tools/llvm-objdump/MachODump.cpp

static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF,
   array_pod_sort(Dices.begin(), Dices.end());
 
 #ifndef NDEBUG
   raw_ostream &DebugOut = DebugFlag ? dbgs() : nulls();
 #else
   raw_ostream &DebugOut = nulls();
 #endif
 
+  // Try to find debug info and set up the DIContext for it.
   std::unique_ptr<DIContext> diContext;
-  ObjectFile *DbgObj = MachOOF;
+  std::unique_ptr<Binary> DSYMBinary;
   std::unique_ptr<MemoryBuffer> DSYMBuf;
-  // Try to find debug info and set up the DIContext for it.
   if (UseDbg) {
+    ObjectFile *DbgObj = MachOOF;
+
     // A separate DSym file path was specified, parse it as a macho file,
     // get the sections and supply it to the section name parsing machinery.
     if (!DSYMFile.empty()) {
       ErrorOr<std::unique_ptr<MemoryBuffer>> BufOrErr =
           MemoryBuffer::getFileOrSTDIN(DSYMFile);
       if (std::error_code EC = BufOrErr.getError()) {
         report_error(errorCodeToError(EC), DSYMFile);
         return;
       }
 
-      std::unique_ptr<MachOObjectFile> DbgObjCheck = unwrapOrError(
-          ObjectFile::createMachOObjectFile(BufOrErr.get()->getMemBufferRef()),
-          DSYMFile.getValue());
-      DbgObj = DbgObjCheck.release();
       // We need to keep the file alive, because we're replacing DbgObj with it.
       DSYMBuf = std::move(BufOrErr.get());
+
+      Expected<std::unique_ptr<Binary>> BinaryOrErr =
+      createBinary(DSYMBuf.get()->getMemBufferRef());
+      if (!BinaryOrErr) {
+        report_error(BinaryOrErr.takeError(), DSYMFile);
+        return;
+      }
+
+      // We need to keep the Binary elive with the buffer
+      DSYMBinary = std::move(BinaryOrErr.get());
+    
+      if (ObjectFile *O = dyn_cast<ObjectFile>(DSYMBinary.get())) {
+        // this is a Mach-O object file, use it
+        if (MachOObjectFile *MachDSYM = dyn_cast<MachOObjectFile>(&*O)) {
+          DbgObj = MachDSYM;
+        }
+        else {
+          WithColor::error(errs(), "llvm-objdump")
+            << DSYMFile << " is not a Mach-O file type.\n";
+          return;
+        }
+      }
+      else if (auto UB = dyn_cast<MachOUniversalBinary>(DSYMBinary.get())){
+        // this is a Universal Binary, find a Mach-O for this architecture
+        uint32_t CPUType, CPUSubType;
+        const char *ArchFlag;
+        if (MachOOF->is64Bit()) {
+          const MachO::mach_header_64 H_64 = MachOOF->getHeader64();
+          CPUType = H_64.cputype;
+          CPUSubType = H_64.cpusubtype;
+        } else {
+          const MachO::mach_header H = MachOOF->getHeader();
+          CPUType = H.cputype;
+          CPUSubType = H.cpusubtype;
+        }
+        Triple T = MachOObjectFile::getArchTriple(CPUType, CPUSubType, nullptr,
+                                                  &ArchFlag);
+        Expected<std::unique_ptr<MachOObjectFile>> MachDSYM =
+            UB->getObjectForArch(ArchFlag);
+        if (!MachDSYM) {
+          report_error(MachDSYM.takeError(), DSYMFile);
+          return;
+        }
+    
+        // We need to keep the Binary elive with the buffer
+        DbgObj = &*MachDSYM.get();
+        DSYMBinary = std::move(*MachDSYM);
+      }
+      else {
+        WithColor::error(errs(), "llvm-objdump")
+          << DSYMFile << " is not a Mach-O or Universal file type.\n";
+        return;
+      }
     }
 
     // Setup the DIContext
     diContext = DWARFContext::create(*DbgObj);
   }
 
   if (FilterSections.empty())
     outs() << "(" << DisSegName << "," << DisSectName << ") section\n";