[ObjectYAML][DX] Add dxcontainer2yaml support

[beanz]

Summary

This change finishes fleshing out the ObjectYAML tools to support
converting DXContainer files into yaml representations.

Depends on D124944

[lhames]

LGTM.

llvm/tools/obj2yaml/dxcontainer2yaml.cpp

20

Returning a raw pointer? Is this a yaml-ism?

22

Looks like this is accessible via dxcontainer2yaml without a runtime check on the magic? For parsing API surface like this I would usually go with an llvm::Error rather than an assert, since I don't trust API clients to verify the magic.

In this case you're unlikely to have any clients outside obj2yaml though, so this is mostly pedantry on my part. Go with whichever approach you prefer.

llvm/include/llvm/Object/DXContainer.h

public:
   // The PartIterator is a wrapper around the iterator for the PartOffsets
   // member of the DXContainer. It contains a refernce to the container, and the
   // current iterator value, as well as storage for a parsed part header.
   class PartIterator {
     const DXContainer &Container;
     SmallVectorImpl<uint32_t>::const_iterator OffsetIt;
     struct PartData {
       dxbc::PartHeader Part;
+      uint32_t Offset;
       StringRef Data;
     } IteratorState;
 
     friend class DXContainer;
 
     PartIterator(const DXContainer &C,
                  SmallVectorImpl<uint32_t>::const_iterator It)
         : Container(C), OffsetIt(It) {

llvm/lib/Object/DXContainer.cpp

 void DXContainer::PartIterator::updateIteratorImpl(const uint32_t Offset) {
   StringRef Buffer = Container.Data.getBuffer();
   const char *Current = Buffer.data() + Offset;
   // Offsets are validated during parsing, so all offsets in the container are
   // valid and contain enough readable data to read a header.
   cantFail(readStruct(Buffer, Current, IteratorState.Part));
   IteratorState.Data =
       StringRef(Current + sizeof(dxbc::PartHeader), IteratorState.Part.Size);
+  IteratorState.Offset = Offset;
 }

llvm/test/tools/obj2yaml/DXContainer/ExplicitSizeAndOffsets.yaml

+# RUN: yaml2obj %s | obj2yaml | FileCheck %s 
+
+--- !dxcontainer
+Header:
+  Hash:            [ 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 
+                     0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ]
+  Version:
+    Major:           1
+    Minor:           0
+  FileSize:        172
+  PartCount:       7
+  PartOffsets:     [ 60, 76, 92, 108, 124, 140, 156 ]
+Parts:
+  - Name:            SFI0
+    Size:            8
+  - Name:            ISG1
+    Size:            8
+  - Name:            OSG1
+    Size:            8
+  - Name:            PSV0
+    Size:            8
+  - Name:            STAT
+    Size:            8
+  - Name:            HASH
+    Size:            8
+  - Name:            CXIL
+    Size:            8
+...
+
+# CHECK: --- !dxcontainer
+# CHECK-NEXT: Header:
+# CHECK-NEXT:   Hash:            [ 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 
+# CHECK-NEXT:                      0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ]
+# CHECK-NEXT:   Version:
+# CHECK-NEXT:     Major:           1
+# CHECK-NEXT:     Minor:           0
+# CHECK-NEXT:   FileSize:        172
+# CHECK-NEXT:   PartCount:       7
+# CHECK-NEXT:   PartOffsets:     [ 60, 76, 92, 108, 124, 140, 156 ]
+# CHECK-NEXT: Parts:
+# CHECK-NEXT:   - Name:            SFI0
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            ISG1
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            OSG1
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            PSV0
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            STAT
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            HASH
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            CXIL
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT: ...

llvm/test/tools/obj2yaml/DXContainer/OmitSizeAndOffsets.yaml

+# RUN: yaml2obj %s | obj2yaml | FileCheck %s 
+
+--- !dxcontainer
+Header:
+  Hash:            [ 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 
+                     0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ]
+  Version:
+    Major:           1
+    Minor:           0
+  PartCount:       7
+Parts:
+  - Name:            SFI0
+    Size:            8
+  - Name:            ISG1
+    Size:            8
+  - Name:            OSG1
+    Size:            8
+  - Name:            PSV0
+    Size:            8
+  - Name:            STAT
+    Size:            8
+  - Name:            HASH
+    Size:            8
+  - Name:            CXIL
+    Size:            8
+...
+
+# CHECK: --- !dxcontainer
+# CHECK-NEXT: Header:
+# CHECK-NEXT:   Hash:            [ 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 
+# CHECK-NEXT:                      0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ]
+# CHECK-NEXT:   Version:
+# CHECK-NEXT:     Major:           1
+# CHECK-NEXT:     Minor:           0
+# CHECK-NEXT:   FileSize:        172
+# CHECK-NEXT:   PartCount:       7
+# CHECK-NEXT:   PartOffsets:     [ 60, 76, 92, 108, 124, 140, 156 ]
+# CHECK-NEXT: Parts:
+# CHECK-NEXT:   - Name:            SFI0
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            ISG1
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            OSG1
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            PSV0
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            STAT
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            HASH
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT:   - Name:            CXIL
+# CHECK-NEXT:     Size:            8
+# CHECK-NEXT: ...

llvm/tools/obj2yaml/CMakeLists.txt

 set(LLVM_LINK_COMPONENTS
   BinaryFormat
   DebugInfoCodeView
   DebugInfoDWARF
   Object
   ObjectYAML
   Support
   )
 
 add_llvm_utility(obj2yaml
   archive2yaml.cpp
   obj2yaml.cpp
   coff2yaml.cpp
   dwarf2yaml.cpp
+  dxcontainer2yaml.cpp
   elf2yaml.cpp
   macho2yaml.cpp
   minidump2yaml.cpp
   xcoff2yaml.cpp
   wasm2yaml.cpp
   )

llvm/tools/obj2yaml/dxcontainer2yaml.cpp

+//===------ dxcontainer2yaml.cpp - obj2yaml conversion tool -----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "obj2yaml.h"
+#include "llvm/Object/DXContainer.h"
+#include "llvm/ObjectYAML/DXContainerYAML.h"
+#include "llvm/Support/Error.h"
+
+#include <algorithm>
+
+using namespace llvm;
+using namespace llvm::object;
+
+static Expected<DXContainerYAML::Object *>
+dumpDXContainer(MemoryBufferRef Source) {
+  assert(file_magic::dxcontainer_object == identify_magic(Source.getBuffer()));
+
+  Expected<DXContainer> ExDXC = DXContainer::create(Source);
+  if (!ExDXC)
+    return ExDXC.takeError();
+  DXContainer Container = *ExDXC;
+
+  std::unique_ptr<DXContainerYAML::Object> Obj =
+      std::make_unique<DXContainerYAML::Object>();
+
+  for (uint8_t Byte : Container.getHeader().FileHash.Digest)
+    Obj->Header.Hash.push_back(Byte);
+  Obj->Header.Version.Major = Container.getHeader().Version.Major;
+  Obj->Header.Version.Minor = Container.getHeader().Version.Minor;
+  Obj->Header.FileSize = Container.getHeader().FileSize;
+  Obj->Header.PartCount = Container.getHeader().PartCount;
+
+  Obj->Header.PartOffsets = std::vector<uint32_t>();
+  for (const auto P : Container) {
+    Obj->Header.PartOffsets->push_back(P.Offset);
+    Obj->Parts.push_back(
+        DXContainerYAML::Part{P.Part.getName().str(), P.Part.Size});
+  }
+
+  return Obj.release();
+}
+
+llvm::Error dxcontainer2yaml(llvm::raw_ostream &Out,
+                             llvm::MemoryBufferRef Source) {
+  Expected<DXContainerYAML::Object *> YAMLOrErr = dumpDXContainer(Source);
+  if (!YAMLOrErr)
+    return YAMLOrErr.takeError();
+
+  std::unique_ptr<DXContainerYAML::Object> YAML(YAMLOrErr.get());
+  yaml::Output Yout(Out);
+  Yout << *YAML;
+
+  return Error::success();
+}

llvm/tools/obj2yaml/obj2yaml.h

llvm::Error macho2yaml(llvm::raw_ostream &Out, const llvm::object::Binary &Obj,
                        unsigned RawSegments);
 llvm::Error minidump2yaml(llvm::raw_ostream &Out,
                           const llvm::object::MinidumpFile &Obj);
 llvm::Error xcoff2yaml(llvm::raw_ostream &Out,
                        const llvm::object::XCOFFObjectFile &Obj);
 std::error_code wasm2yaml(llvm::raw_ostream &Out,
                           const llvm::object::WasmObjectFile &Obj);
 llvm::Error archive2yaml(llvm::raw_ostream &Out, llvm::MemoryBufferRef Source);
+llvm::Error dxcontainer2yaml(llvm::raw_ostream &Out,
+                             llvm::MemoryBufferRef Source);
 
 // Forward decls for dwarf2yaml
 namespace llvm {
 class DWARFContext;
 namespace DWARFYAML {
 struct Data;
 }
 }

llvm/tools/obj2yaml/obj2yaml.cpp

 static Error dumpInput(StringRef File) {
   ErrorOr<std::unique_ptr<MemoryBuffer>> FileOrErr =
       MemoryBuffer::getFileOrSTDIN(File, /*IsText=*/false,
                                    /*RequiresNullTerminator=*/false);
   if (std::error_code EC = FileOrErr.getError())
     return errorCodeToError(EC);
   std::unique_ptr<MemoryBuffer> &Buffer = FileOrErr.get();
   MemoryBufferRef MemBuf = Buffer->getMemBufferRef();
-  if (file_magic::archive == identify_magic(MemBuf.getBuffer()))
+  switch (identify_magic(MemBuf.getBuffer())) {
+  case file_magic::archive:
     return archive2yaml(outs(), MemBuf);
+  case file_magic::dxcontainer_object:
+    return dxcontainer2yaml(outs(), MemBuf);
+  default:
+    break;
+  }
 
   Expected<std::unique_ptr<Binary>> BinOrErr =
       createBinary(MemBuf, /*Context=*/nullptr);
   if (!BinOrErr)
     return BinOrErr.takeError();
 
   Binary &Binary = *BinOrErr->get();
   // Universal MachO is not a subclass of ObjectFile, so it needs to be handled