summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--llvm/include/llvm/DebugInfo/PDB/Raw/ModInfo.h8
-rw-r--r--llvm/include/llvm/DebugInfo/PDB/Raw/PDBDbiStream.h12
-rw-r--r--llvm/lib/DebugInfo/PDB/Raw/PDBDbiStream.cpp135
-rw-r--r--llvm/test/DebugInfo/PDB/pdbdump-headers.test3
-rw-r--r--llvm/tools/llvm-pdbdump/llvm-pdbdump.cpp35
5 files changed, 166 insertions, 27 deletions
diff --git a/llvm/include/llvm/DebugInfo/PDB/Raw/ModInfo.h b/llvm/include/llvm/DebugInfo/PDB/Raw/ModInfo.h
index 8a7efd0af38..f26e29d4792 100644
--- a/llvm/include/llvm/DebugInfo/PDB/Raw/ModInfo.h
+++ b/llvm/include/llvm/DebugInfo/PDB/Raw/ModInfo.h
@@ -13,6 +13,7 @@
#include "llvm/ADT/StringRef.h"
#include <stdint.h>
+#include <vector>
namespace llvm {
class PDBFile;
@@ -42,6 +43,13 @@ private:
const FileLayout *Layout;
};
+struct ModuleInfoEx {
+ ModuleInfoEx(ModInfo Module) : Info(Module) {}
+
+ ModInfo Info;
+ std::vector<StringRef> SourceFiles;
+};
+
class ModInfoIterator {
public:
ModInfoIterator(const uint8_t *Stream);
diff --git a/llvm/include/llvm/DebugInfo/PDB/Raw/PDBDbiStream.h b/llvm/include/llvm/DebugInfo/PDB/Raw/PDBDbiStream.h
index 179797f2be8..eac7e4bbb75 100644
--- a/llvm/include/llvm/DebugInfo/PDB/Raw/PDBDbiStream.h
+++ b/llvm/include/llvm/DebugInfo/PDB/Raw/PDBDbiStream.h
@@ -43,13 +43,23 @@ public:
PDB_Machine getMachineType() const;
- llvm::iterator_range<ModInfoIterator> modules() const;
+ ArrayRef<ModuleInfoEx> modules() const;
private:
+ std::error_code readSubstream(std::vector<uint8_t> &Bytes, uint32_t Size);
+ std::error_code initializeFileInfo();
+
PDBFile &Pdb;
PDBStream Stream;
+ std::vector<ModuleInfoEx> ModuleInfos;
+
std::vector<uint8_t> ModInfoSubstream;
+ std::vector<uint8_t> SecContrSubstream;
+ std::vector<uint8_t> SecMapSubstream;
+ std::vector<uint8_t> FileInfoSubstream;
+ std::vector<uint8_t> TypeServerMapSubstream;
+ std::vector<uint8_t> ECSubstream;
std::unique_ptr<HeaderInfo> Header;
};
}
diff --git a/llvm/lib/DebugInfo/PDB/Raw/PDBDbiStream.cpp b/llvm/lib/DebugInfo/PDB/Raw/PDBDbiStream.cpp
index db84192bc36..c6f16d30b31 100644
--- a/llvm/lib/DebugInfo/PDB/Raw/PDBDbiStream.cpp
+++ b/llvm/lib/DebugInfo/PDB/Raw/PDBDbiStream.cpp
@@ -48,16 +48,16 @@ struct PDBDbiStream::HeaderInfo {
little32_t VersionSignature;
ulittle32_t VersionHeader;
ulittle32_t Age; // Should match PDBInfoStream.
- ulittle16_t GSSyms;
- ulittle16_t BuildNumber; // See DbiBuildNo structure.
- ulittle16_t PSSyms;
+ ulittle16_t GSSyms; // Number of global symbols
+ ulittle16_t BuildNumber; // See DbiBuildNo structure.
+ ulittle16_t PSSyms; // Number of public symbols
ulittle16_t PdbDllVersion; // version of mspdbNNN.dll
ulittle16_t SymRecords; // Number of symbols
ulittle16_t PdbDllRbld; // rbld number of mspdbNNN.dll
little32_t ModiSubstreamSize; // Size of module info stream
little32_t SecContrSubstreamSize; // Size of sec. contribution stream
- little32_t SectionMapSize;
- little32_t FileInfoSize;
+ little32_t SectionMapSize; // Size of sec. map substream
+ little32_t FileInfoSize; // Size of file info substream
little32_t TypeServerSize; // Size of type server map
ulittle32_t MFCTypeServerIndex; // Index of MFC Type Server
little32_t OptionalDbgHdrSize; // Size of DbgHeader info
@@ -101,12 +101,42 @@ std::error_code PDBDbiStream::reload() {
Header->OptionalDbgHdrSize + Header->ECSubstreamSize)
return std::make_error_code(std::errc::illegal_byte_sequence);
+ // Only certain substreams are guaranteed to be aligned. Validate
+ // them here.
if (Header->ModiSubstreamSize % sizeof(uint32_t) != 0)
return std::make_error_code(std::errc::illegal_byte_sequence);
+ if (Header->SecContrSubstreamSize % sizeof(uint32_t) != 0)
+ return std::make_error_code(std::errc::illegal_byte_sequence);
+ if (Header->SectionMapSize % sizeof(uint32_t) != 0)
+ return std::make_error_code(std::errc::illegal_byte_sequence);
+ if (Header->FileInfoSize % sizeof(uint32_t) != 0)
+ return std::make_error_code(std::errc::illegal_byte_sequence);
+ if (Header->TypeServerSize % sizeof(uint32_t) != 0)
+ return std::make_error_code(std::errc::illegal_byte_sequence);
+
+ std::error_code EC;
+ if (EC = readSubstream(ModInfoSubstream, Header->ModiSubstreamSize))
+ return EC;
+
+ // Since each ModInfo in the stream is a variable length, we have to iterate
+ // them to know how many there actually are.
+ auto Range = llvm::make_range(ModInfoIterator(&ModInfoSubstream.front()),
+ ModInfoIterator(&ModInfoSubstream.back() + 1));
+ for (auto Info : Range)
+ ModuleInfos.push_back(ModuleInfoEx(Info));
- ModInfoSubstream.resize(Header->ModiSubstreamSize);
- if (auto EC =
- Stream.readBytes(&ModInfoSubstream[0], Header->ModiSubstreamSize))
+ if (EC = readSubstream(SecContrSubstream, Header->SecContrSubstreamSize))
+ return EC;
+ if (EC = readSubstream(SecMapSubstream, Header->SectionMapSize))
+ return EC;
+ if (EC = readSubstream(FileInfoSubstream, Header->FileInfoSize))
+ return EC;
+ if (EC = readSubstream(TypeServerMapSubstream, Header->TypeServerSize))
+ return EC;
+ if (EC = readSubstream(ECSubstream, Header->ECSubstreamSize))
+ return EC;
+
+ if (EC = initializeFileInfo())
return EC;
return std::error_code();
@@ -150,7 +180,90 @@ PDB_Machine PDBDbiStream::getMachineType() const {
return static_cast<PDB_Machine>(Machine);
}
-llvm::iterator_range<ModInfoIterator> PDBDbiStream::modules() const {
- return llvm::make_range(ModInfoIterator(&ModInfoSubstream.front()),
- ModInfoIterator(&ModInfoSubstream.back() + 1));
+ArrayRef<ModuleInfoEx> PDBDbiStream::modules() const { return ModuleInfos; }
+
+std::error_code PDBDbiStream::readSubstream(std::vector<uint8_t> &Bytes, uint32_t Size) {
+ Bytes.clear();
+ if (Size == 0)
+ return std::error_code();
+
+ Bytes.resize(Size);
+ return Stream.readBytes(&Bytes[0], Size);
+}
+
+std::error_code PDBDbiStream::initializeFileInfo() {
+ struct FileInfoSubstreamHeader {
+ ulittle16_t NumModules; // Total # of modules, should match number of
+ // records in the ModuleInfo substream.
+ ulittle16_t NumSourceFiles; // Total # of source files. This value is not
+ // accurate because PDB actually supports more
+ // than 64k source files, so we ignore it and
+ // compute the value from other stream fields.
+ };
+
+ // The layout of the FileInfoSubstream is like this:
+ // struct {
+ // ulittle16_t NumModules;
+ // ulittle16_t NumSourceFiles;
+ // ulittle16_t ModIndices[NumModules];
+ // ulittle16_t ModFileCounts[NumModules];
+ // ulittle32_t FileNameOffsets[NumSourceFiles];
+ // char Names[][NumSourceFiles];
+ // };
+ // with the caveat that `NumSourceFiles` cannot be trusted, so
+ // it is computed by summing `ModFileCounts`.
+ //
+ const uint8_t *Buf = &FileInfoSubstream[0];
+ auto FI = reinterpret_cast<const FileInfoSubstreamHeader *>(Buf);
+ Buf += sizeof(FileInfoSubstreamHeader);
+ // The number of modules in the stream should be the same as reported by
+ // the FileInfoSubstreamHeader.
+ if (FI->NumModules != ModuleInfos.size())
+ return std::make_error_code(std::errc::illegal_byte_sequence);
+
+ // First is an array of `NumModules` module indices. This is not used for the
+ // same reason that `NumSourceFiles` is not used. It's an array of uint16's,
+ // but it's possible there are more than 64k source files, which would imply
+ // more than 64k modules (e.g. object files) as well. So we ignore this
+ // field.
+ llvm::ArrayRef<ulittle16_t> ModIndexArray(
+ reinterpret_cast<const ulittle16_t *>(Buf), ModuleInfos.size());
+
+ llvm::ArrayRef<ulittle16_t> ModFileCountArray(ModIndexArray.end(),
+ ModuleInfos.size());
+
+ // Compute the real number of source files.
+ uint32_t NumSourceFiles = 0;
+ for (auto Count : ModFileCountArray)
+ NumSourceFiles += Count;
+
+ // This is the array that in the reference implementation corresponds to
+ // `ModInfo::FileLayout::FileNameOffs`, which is commented there as being a
+ // pointer. Due to the mentioned problems of pointers causing difficulty
+ // when reading from the file on 64-bit systems, we continue to ignore that
+ // field in `ModInfo`, and instead build a vector of StringRefs and stores
+ // them in `ModuleInfoEx`. The value written to and read from the file is
+ // not used anyway, it is only there as a way to store the offsets for the
+ // purposes of later accessing the names at runtime.
+ llvm::ArrayRef<little32_t> FileNameOffsets(
+ reinterpret_cast<const little32_t *>(ModFileCountArray.end()),
+ NumSourceFiles);
+
+ const char *Names = reinterpret_cast<const char *>(FileNameOffsets.end());
+
+ // We go through each ModuleInfo, determine the number N of source files for
+ // that module, and then get the next N offsets from the Offsets array, using
+ // them to get the corresponding N names from the Names buffer and associating
+ // each one with the corresponding module.
+ uint32_t NextFileIndex = 0;
+ for (size_t I = 0; I < ModuleInfos.size(); ++I) {
+ uint32_t NumFiles = ModFileCountArray[I];
+ ModuleInfos[I].SourceFiles.resize(NumFiles);
+ for (size_t J = 0; J < NumFiles; ++J, ++NextFileIndex) {
+ uint32_t FileIndex = FileNameOffsets[NextFileIndex];
+ ModuleInfos[I].SourceFiles[J] = StringRef(Names + FileIndex);
+ }
+ }
+
+ return std::error_code();
}
diff --git a/llvm/test/DebugInfo/PDB/pdbdump-headers.test b/llvm/test/DebugInfo/PDB/pdbdump-headers.test
index 1932378fe06..384fcee332a 100644
--- a/llvm/test/DebugInfo/PDB/pdbdump-headers.test
+++ b/llvm/test/DebugInfo/PDB/pdbdump-headers.test
@@ -41,6 +41,8 @@
; CHECK-NEXT: Symbol Byte Size: 208
; CHECK-NEXT: Type Server Index: 0
; CHECK-NEXT: Has EC Info: 0
+; CHECK-NEXT: 1 Contributing Source Files:
+; CHECK-NEXT: d:\src\llvm\test\debuginfo\pdb\inputs\empty.cpp
; CHECK-NEXT: * Linker *
; CHECK-NEXT: Debug Stream Index: 14
; CHECK-NEXT: Object File:
@@ -52,3 +54,4 @@
; CHECK-NEXT: Symbol Byte Size: 516
; CHECK-NEXT: Type Server Index: 0
; CHECK-NEXT: Has EC Info: 0
+; CHECK-NEXT: 0 Contributing Source Files:
diff --git a/llvm/tools/llvm-pdbdump/llvm-pdbdump.cpp b/llvm/tools/llvm-pdbdump/llvm-pdbdump.cpp
index e051c85c5dc..4d4987611b9 100644
--- a/llvm/tools/llvm-pdbdump/llvm-pdbdump.cpp
+++ b/llvm/tools/llvm-pdbdump/llvm-pdbdump.cpp
@@ -281,25 +281,30 @@ static void dumpStructure(RawSession &RS) {
<< Minor << "." << DbiStream.getPdbDllVersion() << '\n';
outs() << "Modules: \n";
- for (auto Modi : DbiStream.modules()) {
- outs() << Modi.getModuleName() << '\n';
- outs().indent(4) << "Debug Stream Index: " << Modi.getModuleStreamIndex()
- << '\n';
- outs().indent(4) << "Object File: " << Modi.getObjFileName() << '\n';
- outs().indent(4) << "Num Files: " << Modi.getNumberOfFiles() << '\n';
+ for (auto &Modi : DbiStream.modules()) {
+ outs() << Modi.Info.getModuleName() << '\n';
+ outs().indent(4) << "Debug Stream Index: "
+ << Modi.Info.getModuleStreamIndex() << '\n';
+ outs().indent(4) << "Object File: " << Modi.Info.getObjFileName() << '\n';
+ outs().indent(4) << "Num Files: " << Modi.Info.getNumberOfFiles() << '\n';
outs().indent(4) << "Source File Name Idx: "
- << Modi.getSourceFileNameIndex() << '\n';
- outs().indent(4) << "Pdb File Name Idx: " << Modi.getPdbFilePathNameIndex()
- << '\n';
- outs().indent(4) << "Line Info Byte Size: " << Modi.getLineInfoByteSize()
- << '\n';
+ << Modi.Info.getSourceFileNameIndex() << '\n';
+ outs().indent(4) << "Pdb File Name Idx: "
+ << Modi.Info.getPdbFilePathNameIndex() << '\n';
+ outs().indent(4) << "Line Info Byte Size: "
+ << Modi.Info.getLineInfoByteSize() << '\n';
outs().indent(4) << "C13 Line Info Byte Size: "
- << Modi.getC13LineInfoByteSize() << '\n';
+ << Modi.Info.getC13LineInfoByteSize() << '\n';
outs().indent(4) << "Symbol Byte Size: "
- << Modi.getSymbolDebugInfoByteSize() << '\n';
- outs().indent(4) << "Type Server Index: " << Modi.getTypeServerIndex()
+ << Modi.Info.getSymbolDebugInfoByteSize() << '\n';
+ outs().indent(4) << "Type Server Index: " << Modi.Info.getTypeServerIndex()
<< '\n';
- outs().indent(4) << "Has EC Info: " << Modi.hasECInfo() << '\n';
+ outs().indent(4) << "Has EC Info: " << Modi.Info.hasECInfo() << '\n';
+ outs().indent(4) << Modi.SourceFiles.size()
+ << " Contributing Source Files: \n";
+ for (auto File : Modi.SourceFiles) {
+ outs().indent(8) << File << '\n';
+ }
}
}
OpenPOWER on IntegriCloud