[PDB] Dump extra info about the publics stream

This includes the hash table, the address map, and the thunk table and
section offset table. The last two are only used for incremental
linking, which LLD doesn't support, so they are less interesting. The
hash table is particularly important to get right, since this is the one
of the streams that debuggers use to translate addresses to symbols.

llvm-svn: 308764

llvm/include/llvm/DebugInfo/PDB/Native/PublicsStream.h

@@ -36,6 +36,8 @@ public:
   Expected<const codeview::CVSymbolArray &> getSymbolArray() const;
   iterator_range<codeview::CVSymbolArray::Iterator>
   getSymbols(bool *HadError) const;
+  FixedStreamArray<PSHashRecord> getHashRecords() const { return HashRecords; }
+  FixedStreamArray<PSHashRecord> getHashBitmap() const { return HashBitmap; }
   FixedStreamArray<support::ulittle32_t> getHashBuckets() const {
     return HashBuckets;
   }
@@ -56,8 +58,8 @@ private:
 
   std::unique_ptr<msf::MappedBlockStream> Stream;
   uint32_t NumBuckets = 0;
-  ArrayRef<uint8_t> Bitmap;
   FixedStreamArray<PSHashRecord> HashRecords;
+  ArrayRef<uint8_t> HashBitmap;
   FixedStreamArray<support::ulittle32_t> HashBuckets;
   FixedStreamArray<support::ulittle32_t> AddressMap;
   FixedStreamArray<support::ulittle32_t> ThunkMap;

llvm/lib/DebugInfo/PDB/Native/GSI.cpp

@@ -29,6 +29,7 @@ static Error checkHashHdrVersion(const GSIHashHeader *HashHdr) {
 }
 
 Error readGSIHashBuckets(FixedStreamArray<support::ulittle32_t> &HashBuckets,
+                         ArrayRef<uint8_t> &HashBitmap,
                          const GSIHashHeader *HashHdr,
                          BinaryStreamReader &Reader) {
   if (auto EC = checkHashHdrVersion(HashHdr))
@@ -36,15 +37,14 @@ Error readGSIHashBuckets(FixedStreamArray<support::ulittle32_t> &HashBuckets,
 
   // Before the actual hash buckets, there is a bitmap of length determined by
   // IPHR_HASH.
-  ArrayRef<uint8_t> Bitmap;
   size_t BitmapSizeInBits = alignTo(IPHR_HASH + 1, 32);
   uint32_t NumBitmapEntries = BitmapSizeInBits / 8;
-  if (auto EC = Reader.readBytes(Bitmap, NumBitmapEntries))
+  if (auto EC = Reader.readBytes(HashBitmap, NumBitmapEntries))
     return joinErrors(std::move(EC),
                       make_error<RawError>(raw_error_code::corrupt_file,
                                            "Could not read a bitmap."));
   uint32_t NumBuckets = 0;
-  for (uint8_t B : Bitmap)
+  for (uint8_t B : HashBitmap)
     NumBuckets += countPopulation(B);
 
   // Hash buckets follow.

llvm/lib/DebugInfo/PDB/Native/GSI.h

@@ -41,7 +41,7 @@ namespace pdb {
 static const unsigned IPHR_HASH = 4096;
 
 /// Header of the hash tables found in the globals and publics sections.
-/// Based on GSIHashHeader in
+/// Based on GSIHashHdr in
 /// https://github.com/Microsoft/microsoft-pdb/blob/master/PDB/dbi/gsi.h
 struct GSIHashHeader {
   enum : unsigned {

llvm/lib/DebugInfo/PDB/Native/PublicsStream.cpp

@@ -75,7 +75,7 @@ Error PublicsStream::reload() {
   if (auto EC = readGSIHashRecords(HashRecords, HashHdr, Reader))
     return EC;
 
-  if (auto EC = readGSIHashBuckets(HashBuckets, HashHdr, Reader))
+  if (auto EC = readGSIHashBuckets(HashBuckets, HashBitmap, HashHdr, Reader))
     return EC;
   NumBuckets = HashBuckets.size();
 

llvm/tools/llvm-pdbutil/DumpOutputStyle.cpp

@@ -882,6 +882,52 @@ Error DumpOutputStyle::dumpPublics() {
     P.formatLine("Error while processing public symbol records.  {0}",
                  toString(std::move(EC)));
 
+  // Return early if we aren't dumping public hash table and address map info.
+  if (!opts::dump::DumpPublicExtras)
+    return Error::success();
+
+  P.formatLine("Hash Records");
+  {
+    AutoIndent Indent2(P);
+    for (const PSHashRecord &HR : Publics.getHashRecords())
+      P.formatLine("off = {0}, refcnt = {1}", uint32_t(HR.Off),
+                   uint32_t(HR.CRef));
+  }
+
+  // FIXME: Dump the bitmap.
+
+  P.formatLine("Hash Buckets");
+  {
+    AutoIndent Indent2(P);
+    for (uint32_t Hash : Publics.getHashBuckets())
+      P.formatLine("{0:x8}", Hash);
+  }
+
+  P.formatLine("Address Map");
+  {
+    // These are offsets into the publics stream sorted by secidx:secrel.
+    AutoIndent Indent2(P);
+    for (uint32_t Addr : Publics.getAddressMap())
+      P.formatLine("off = {0}", Addr);
+  }
+
+  // The thunk map is optional debug info used for ILT thunks.
+  if (!Publics.getThunkMap().empty()) {
+    P.formatLine("Thunk Map");
+    AutoIndent Indent2(P);
+    for (uint32_t Addr : Publics.getThunkMap())
+      P.formatLine("{0:x8}", Addr);
+  }
+
+  // The section offsets table appears to be empty when incremental linking
+  // isn't in use.
+  if (!Publics.getSectionOffsets().empty()) {
+    P.formatLine("Section Offsets");
+    AutoIndent Indent2(P);
+    for (const SectionOffset &SO : Publics.getSectionOffsets())
+      P.formatLine("{0:x4}:{1:x8}", uint16_t(SO.Isect), uint32_t(SO.Off));
+  }
+
   return Error::success();
 }
 

llvm/tools/llvm-pdbutil/llvm-pdbutil.cpp

@@ -452,6 +452,9 @@ cl::opt<bool> DumpTypeDependents(
 // SYMBOL OPTIONS
 cl::opt<bool> DumpPublics("publics", cl::desc("dump Publics stream data"),
                           cl::cat(SymbolOptions), cl::sub(DumpSubcommand));
+cl::opt<bool> DumpPublicExtras("public-extras",
+                               cl::desc("dump Publics hashes and address maps"),
+                               cl::cat(SymbolOptions), cl::sub(DumpSubcommand));
 cl::opt<bool> DumpSymbols("symbols", cl::desc("dump module symbols"),
                           cl::cat(SymbolOptions), cl::sub(DumpSubcommand));
 

llvm/tools/llvm-pdbutil/llvm-pdbutil.h

@@ -144,6 +144,7 @@ extern llvm::cl::list<uint32_t> DumpIdIndex;
 extern llvm::cl::opt<bool> DumpSymbols;
 extern llvm::cl::opt<bool> DumpSymRecordBytes;
 extern llvm::cl::opt<bool> DumpPublics;
+extern llvm::cl::opt<bool> DumpPublicExtras;
 extern llvm::cl::opt<bool> DumpSectionContribs;
 extern llvm::cl::opt<bool> DumpSectionMap;
 extern llvm::cl::opt<bool> DumpModules;