CodeView: extract the OMF Directory Header

[compnerd]

Summary

The DebugDirectory contains a pointer to the CodeView info structure which is a
derivative of the OMF debug directory. The structure has evolved a bit over
time, and PDB 2.0 used a slightly different definition from PDB 7.0. Both of
these are specific to CodeView and not COFF. Reflect this by moving the
structure definitions into the DebugInfo/CodeView headers. Define a generic
DebugInfo union type that can be used to pass around a reference to the
DebugInfo irrespective of the versioning. NFC.

[majnemer]

lib/DebugInfo/CodeView/CVDebugRecord.cpp

14–21 ↗

(On Diff #67121)

This is pretty irregular, we usually have the constant have a normal type. I would expect this to be an enum.

[compnerd]

lib/DebugInfo/CodeView/CVDebugRecord.cpp

14–21 ↗

(On Diff #67121)

I didnt consider making that an enum. I like that better. Ill change this.

[compnerd]

Use an enum as suggested by @majnemer

[majnemer]

include/llvm/DebugInfo/CodeView/CVDebugRecord.h

18

I'd give the enum an underlying type of uint32_t.

[majnemer]

LGTM

include/llvm/DebugInfo/CodeView/CVDebugRecord.h

+//===- CVDebugRecord.h ------------------------------------------*- C++ -*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_CODEVIEW_CVDEBUGRECORD_H
+#define LLVM_DEBUGINFO_CODEVIEW_CVDEBUGRECORD_H
+
+#include "llvm/Support/Endian.h"
+
+namespace llvm {
+namespace OMF {
+struct Signature {
+  enum ID {
+    PDB70 = 0x53445352, // RSDS
+    PDB20 = 0x3031424e, // NB10
+    CV50 = 0x3131424e,  // NB11
+    CV41 = 0x3930424e,  // NB09
+  };
+
+  support::ulittle32_t CVSignature;
+  support::ulittle32_t Offset;
+};
+}
+
+namespace codeview {
+struct PDB70DebugInfo {
+  support::ulittle32_t CVSignature;
+  uint8_t Signature[16];
+  support::ulittle32_t Age;
+  // char PDBFileName[];
+};
+
+struct PDB20DebugInfo {
+  support::ulittle32_t CVSignature;
+  support::ulittle32_t Offset;
+  support::ulittle32_t Signature;
+  support::ulittle32_t Age;
+  // char PDBFileName[];
+};
+
+union DebugInfo {
+  struct OMF::Signature Signature;
+  struct PDB20DebugInfo PDB20;
+  struct PDB70DebugInfo PDB70;
+};
+}
+}
+
+#endif
+

include/llvm/Object/COFF.h

 #define LLVM_OBJECT_COFF_H
 
 #include "llvm/ADT/PointerUnion.h"
+#include "llvm/DebugInfo/CodeView/CVDebugRecord.h"
 #include "llvm/Object/ObjectFile.h"
 #include "llvm/Support/COFF.h"
 #include "llvm/Support/Endian.h"

   support::ulittle32_t PointerToRawData;
 };
 
-/// Information that is resent in debug_directory::AddressOfRawData if Type is
-/// IMAGE_DEBUG_TYPE_CODEVIEW.
-struct debug_pdb_info {
-  support::ulittle32_t Signature;
-  uint8_t Guid[16];
-  support::ulittle32_t Age;
-  // PDBFileName: The null-terminated PDB file name follows.
-};
-
 template <typename IntTy>
 struct import_lookup_table_entry {
   IntTy Data;

 
   /// Get PDB information out of a codeview debug directory entry.
   std::error_code getDebugPDBInfo(const debug_directory *DebugDir,
-                                  const debug_pdb_info *&Info,
+                                  const codeview::DebugInfo *&Info,
                                   StringRef &PDBFileName) const;
 
   /// Get PDB information from an executable. If the information is not present,
   /// Info will be set to nullptr and PDBFileName will be empty. An error is
   /// returned only on corrupt object files. Convenience accessor that can be
   /// used if the debug directory is not already handy.
-  std::error_code getDebugPDBInfo(const debug_pdb_info *&Info,
+  std::error_code getDebugPDBInfo(const codeview::DebugInfo *&Info,
                                   StringRef &PDBFileName) const;
 
   bool isRelocatableObject() const override;

lib/DebugInfo/Symbolize/Symbolize.cpp

   // If this is a COFF object containing PDB info, use a PDBContext to
   // symbolize. Otherwise, use DWARF.
   if (auto CoffObject = dyn_cast<COFFObjectFile>(Objects.first)) {
-    const debug_pdb_info *PDBInfo;
+    const codeview::DebugInfo *DebugInfo;
     StringRef PDBFileName;
-    auto EC = CoffObject->getDebugPDBInfo(PDBInfo, PDBFileName);
-    if (!EC && PDBInfo != nullptr) {
+    auto EC = CoffObject->getDebugPDBInfo(DebugInfo, PDBFileName);
+    if (!EC && DebugInfo != nullptr) {
       using namespace pdb;
       std::unique_ptr<IPDBSession> Session;
       if (auto Err = loadDataForEXE(PDB_ReaderType::DIA,

lib/Object/COFFObjectFile.cpp

   return std::error_code();
 }
 
-std::error_code COFFObjectFile::getDebugPDBInfo(const debug_directory *DebugDir,
-                                                const debug_pdb_info *&PDBInfo,
-                                                StringRef &PDBFileName) const {
+std::error_code
+COFFObjectFile::getDebugPDBInfo(const debug_directory *DebugDir,
+                                const codeview::DebugInfo *&PDBInfo,
+                                StringRef &PDBFileName) const {
   ArrayRef<uint8_t> InfoBytes;
   if (std::error_code EC = getRvaAndSizeAsBytes(
           DebugDir->AddressOfRawData, DebugDir->SizeOfData, InfoBytes))
     return EC;
-  if (InfoBytes.size() < sizeof(debug_pdb_info) + 1)
+  if (InfoBytes.size() < sizeof(*PDBInfo) + 1)
     return object_error::parse_failed;
-  PDBInfo = reinterpret_cast<const debug_pdb_info *>(InfoBytes.data());
-  InfoBytes = InfoBytes.drop_front(sizeof(debug_pdb_info));
+  PDBInfo = reinterpret_cast<const codeview::DebugInfo *>(InfoBytes.data());
+  InfoBytes = InfoBytes.drop_front(sizeof(*PDBInfo));
   PDBFileName = StringRef(reinterpret_cast<const char *>(InfoBytes.data()),
                           InfoBytes.size());
   // Truncate the name at the first null byte. Ignore any padding.

   return std::error_code();
 }
 
-std::error_code COFFObjectFile::getDebugPDBInfo(const debug_pdb_info *&PDBInfo,
-                                                StringRef &PDBFileName) const {
+std::error_code
+COFFObjectFile::getDebugPDBInfo(const codeview::DebugInfo *&PDBInfo,
+                                StringRef &PDBFileName) const {
   for (const debug_directory &D : debug_directories())
     if (D.Type == COFF::IMAGE_DEBUG_TYPE_CODEVIEW)
       return getDebugPDBInfo(&D, PDBInfo, PDBFileName);

tools/llvm-readobj/COFFDumper.cpp

     W.printHex("AddressOfRawData", D.AddressOfRawData);
     W.printHex("PointerToRawData", D.PointerToRawData);
     if (D.Type == COFF::IMAGE_DEBUG_TYPE_CODEVIEW) {
-      const debug_pdb_info *PDBInfo;
+      const codeview::DebugInfo *DebugInfo;
       StringRef PDBFileName;
-      error(Obj->getDebugPDBInfo(&D, PDBInfo, PDBFileName));
+      error(Obj->getDebugPDBInfo(&D, DebugInfo, PDBFileName));
       DictScope PDBScope(W, "PDBInfo");
-      W.printHex("PDBSignature", PDBInfo->Signature);
-      W.printBinary("PDBGUID", makeArrayRef(PDBInfo->Guid));
-      W.printNumber("PDBAge", PDBInfo->Age);
-      W.printString("PDBFileName", PDBFileName);
+      W.printHex("PDBSignature", DebugInfo->Signature.CVSignature);
+      if (DebugInfo->Signature.CVSignature == OMF::Signature::PDB70) {
+        W.printBinary("PDBGUID", makeArrayRef(DebugInfo->PDB70.Signature));
+        W.printNumber("PDBAge", DebugInfo->PDB70.Age);
+        W.printString("PDBFileName", PDBFileName);
+      }
     } else {
       // FIXME: Type values of 12 and 13 are commonly observed but are not in
       // the documented type enum.  Figure out what they mean.