[BOLT][DWARF] Add support for .debug_names

DWARF5 spec supports the .debug_names acceleration table. This is the formalized
version of combination of gdb-index/pubnames/types. Added implementation of it
to BOLT. It supports both monolothic and split dwarf, with and without Type
Units. It does not include parent indices. This will be in followup PR. Unlike
LLVM output this will put all the CUs and TUs into one Module.

Author: ayermolo

bolt/include/bolt/Core/BinaryContext.h

@@ -997,6 +997,10 @@ class BinaryContext {
     return getUniqueSectionByName(".gdb_index");
   }
 
+  ErrorOr<BinarySection &> getDebugNamesSection() const {
+    return getUniqueSectionByName(".debug_names");
+  }
+
   /// @}
 
   /// Register \p TargetFunction as a fragment of \p Function if checks pass:

bolt/include/bolt/Core/DIEBuilder.h

@@ -16,6 +16,7 @@
 #define BOLT_CORE_DIE_BUILDER_H
 
 #include "bolt/Core/BinaryContext.h"
+#include "bolt/Core/DebugNames.h"
 #include "llvm/CodeGen/DIE.h"
 #include "llvm/DebugInfo/DWARF/DWARFAbbreviationDeclaration.h"
 #include "llvm/DebugInfo/DWARF/DWARFDie.h"
@@ -124,9 +125,10 @@ class DIEBuilder {
   std::vector<std::unique_ptr<DIEAbbrev>> Abbreviations;
   BinaryContext &BC;
   DWARFContext *DwarfContext{nullptr};
-  bool IsDWO{false};
+  DWARFUnit *SkeletonCU{nullptr};
   uint64_t UnitSize{0};
   llvm::DenseSet<uint64_t> AllProcessed;
+  DWARF5AcceleratorTable &DebugNamesTable;
 
   /// Returns current state of the DIEBuilder
   State &getState() { return *BuilderState.get(); }
@@ -207,7 +209,7 @@ class DIEBuilder {
   void updateReferences();
 
   /// Update the Offset and Size of DIE.
-  uint32_t computeDIEOffset(const DWARFUnit &CU, DIE &Die, uint32_t &CurOffset);
+  uint32_t computeDIEOffset(DWARFUnit &CU, DIE &Die, uint32_t &CurOffset);
 
   void registerUnit(DWARFUnit &DU, bool NeedSort);
 
@@ -264,8 +266,13 @@ class DIEBuilder {
   /// current Section.
   DIE *constructDIEFast(DWARFDie &DDie, DWARFUnit &U, uint32_t UnitId);
 
+  /// Returns true if this DIEBUilder is for DWO Unit.
+  bool isDWO() const { return SkeletonCU != nullptr; }
+
 public:
-  DIEBuilder(BinaryContext &BC, DWARFContext *DwarfContext, bool IsDWO = false);
+  DIEBuilder(BinaryContext &BC, DWARFContext *DwarfContext,
+             DWARF5AcceleratorTable &DebugNamesTable,
+             DWARFUnit *SkeletonCU = nullptr);
 
   /// Returns enum to what we are currently processing.
   ProcessingType getCurrentProcessingState() { return getState().Type; }

bolt/include/bolt/Core/DebugData.h

@@ -439,6 +439,12 @@ class DebugStrOffsetsWriter {
   /// Update Str offset in .debug_str in .debug_str_offsets.
   void updateAddressMap(uint32_t Index, uint32_t Address);
 
+  /// Get offset for given index in original .debug_str_offsets section.
+  uint64_t getOffset(uint32_t Index) const {
+    assert(StrOffsets.size() > Index && "Index is out of bounds.");
+    return StrOffsets[Index];
+  }
+
   /// Writes out current sections entry into .debug_str_offsets.
   void finalizeSection(DWARFUnit &Unit, DIEBuilder &DIEBldr);
 
@@ -450,10 +456,16 @@ class DebugStrOffsetsWriter {
     return std::move(StrOffsetsBuffer);
   }
 
-private:
   /// Initializes Buffer and Stream.
   void initialize(DWARFUnit &Unit);
 
+  /// Clear data.
+  void clear() {
+    IndexToAddressMap.clear();
+    StrOffsets.clear();
+  }
+
+private:
   std::unique_ptr<DebugStrOffsetsBufferVector> StrOffsetsBuffer;
   std::unique_ptr<raw_svector_ostream> StrOffsetsStream;
   std::map<uint32_t, uint32_t> IndexToAddressMap;
@@ -478,11 +490,12 @@ class DebugStrWriter {
   /// Returns False if no strings were added to .debug_str.
   bool isInitialized() const { return !StrBuffer->empty(); }
 
+  /// Initializes Buffer and Stream.
+  void initialize();
+
 private:
   /// Mutex used for parallel processing of debug info.
   std::mutex WriterMutex;
-  /// Initializes Buffer and Stream.
-  void initialize();
   /// Creates internal data structures.
   void create();
   std::unique_ptr<DebugStrBufferVector> StrBuffer;
@@ -798,6 +811,7 @@ class DwarfLineTable {
   // Returns DWARF Version for this line table.
   uint16_t getDwarfVersion() const { return DwarfVersion; }
 };
+
 } // namespace bolt
 } // namespace llvm
 

bolt/include/bolt/Core/DebugNames.h

@@ -0,0 +1,193 @@
+//===- bolt/Core/DebugNames.cpp - Debugging information handling ---*- 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
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains declaration of classes required for generation of
+// .debug_names section.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef BOLT_CORE_DEBUG_NAMES_H
+#define BOLT_CORE_DEBUG_NAMES_H
+
+#include "DebugData.h"
+#include "llvm/CodeGen/AccelTable.h"
+
+namespace llvm {
+namespace bolt {
+class BOLTDWARF5AccelTableData : DWARF5AccelTableData {
+public:
+  BOLTDWARF5AccelTableData(const uint64_t DieOffset,
+                           const std::optional<uint64_t> DefiningParentOffset,
+                           const unsigned DieTag, const unsigned UnitID,
+                           const bool IsTU,
+                           const std::optional<unsigned> SecondUnitID)
+      : DWARF5AccelTableData(DieOffset, DefiningParentOffset, DieTag, UnitID,
+                             IsTU),
+        SecondUnitID(SecondUnitID) {}
+
+  uint64_t getDieOffset() const { return DWARF5AccelTableData::getDieOffset(); }
+  unsigned getDieTag() const { return DWARF5AccelTableData::getDieTag(); }
+  unsigned getUnitID() const { return DWARF5AccelTableData::getUnitID(); }
+  bool isTU() const { return DWARF5AccelTableData::isTU(); }
+  std::optional<unsigned> getSecondUnitID() const { return SecondUnitID; }
+
+private:
+  std::optional<unsigned> SecondUnitID;
+};
+
+class DWARF5AcceleratorTable {
+public:
+  DWARF5AcceleratorTable(const bool CreateDebugNames, BinaryContext &BC,
+                         DebugStrWriter &MainBinaryStrWriter);
+  /// Add DWARF5 Accelerator table entry.
+  /// Input is DWARFUnit being processed, DIE that belongs to it, and potential
+  /// SkeletonCU if the Unit comes from a DWO section.
+  void addAccelTableEntry(DWARFUnit &Unit, const DIE &Die,
+                          const std::optional<uint64_t> &DWOID);
+  /// Set current unit being processed.
+  void setCurrentUnit(DWARFUnit &Unit, const uint64_t UnitStartOffset);
+  /// Emit Accelerator table.
+  void emitAccelTable();
+  /// Returns true if the table was crated.
+  bool isCreated() const { return NeedToCreate; }
+  /// Returns buffer containing the accelerator table.
+  std::unique_ptr<DebugBufferVector> releaseBuffer() {
+    return std::move(FullTableBuffer);
+  }
+
+private:
+  union AbbrevDescriptor {
+    struct {
+      uint32_t CompUnit : 1;
+      uint32_t TypeUnit : 1;
+      uint32_t DieOffset : 1;
+      uint32_t Parent : 1;
+      uint32_t TypeHash : 2;
+      uint32_t Tag : 26;
+    } Bits;
+    uint32_t Value = 0;
+  };
+  class TagIndex {
+  public:
+    uint32_t DieTag;
+    uint32_t Index;
+  };
+  struct cmpByTagIndex {
+    bool operator()(const TagIndex &LHS, const TagIndex &RHS) const {
+      return LHS.Index < RHS.Index;
+    }
+  };
+  bool NeedToCreate = false;
+  BumpPtrAllocator Allocator;
+  DebugStrWriter &MainBinaryStrWriter;
+  DebugStrOffsetsWriter StrOffsetsWriter;
+  StringRef StrSection;
+  uint64_t CurrentUnitOffset = 0;
+  const DWARFUnit *CurrentUnit = nullptr;
+  std::unordered_map<uint32_t, uint32_t> AbbrevTagToIndexMap;
+
+  /// Represents a group of entries with identical name (and hence, hash value).
+  struct HashData {
+    uint64_t StrOffset;
+    uint32_t HashValue;
+    uint32_t EntryOffset;
+    std::vector<BOLTDWARF5AccelTableData *> Values;
+  };
+  using HashList = std::vector<HashData *>;
+  using BucketList = std::vector<HashList>;
+  /// Contains all the offsets of CUs.
+  SmallVector<uint32_t, 1> CUList;
+  /// Contains all the offsets of local TUs.
+  SmallVector<uint32_t, 1> LocalTUList;
+  /// Contains all the type hashes for split dwarf TUs.
+  SmallVector<uint64_t, 1> ForeignTUList;
+  using StringEntries =
+      MapVector<std::string, HashData, llvm::StringMap<unsigned>>;
+  StringEntries Entries;
+  std::map<TagIndex, SmallVector<DWARF5AccelTableData::AttributeEncoding, 2>,
+           cmpByTagIndex>
+      Abbreviations;
+  uint32_t BucketCount = 0;
+  uint32_t UniqueHashCount = 0;
+  uint32_t AbbrevTableSize = 0;
+  uint32_t CUIndexEncodingSize = 4;
+  uint32_t TUIndexEncodingSize = 4;
+  uint32_t AugmentationStringSize = 0;
+  dwarf::Form CUIndexForm = dwarf::DW_FORM_data4;
+  dwarf::Form TUIndexForm = dwarf::DW_FORM_data4;
+
+  BucketList Buckets;
+
+  std::unique_ptr<DebugBufferVector> FullTableBuffer;
+  std::unique_ptr<raw_svector_ostream> FullTableStream;
+  std::unique_ptr<DebugBufferVector> StrBuffer;
+  std::unique_ptr<raw_svector_ostream> StrStream;
+  std::unique_ptr<DebugBufferVector> EntriesBuffer;
+  std::unique_ptr<raw_svector_ostream> Entriestream;
+  std::unique_ptr<DebugBufferVector> AugStringBuffer;
+  std::unique_ptr<raw_svector_ostream> AugStringtream;
+  llvm::DenseMap<llvm::hash_code, uint64_t> StrCacheToOffsetMap;
+  // Contains DWO ID to CUList Index.
+  llvm::DenseMap<uint64_t, uint32_t> CUOffsetsToPatch;
+  /// Adds Unit to either CUList, LocalTUList or ForeignTUList.
+  /// Input Unit being processed, and DWO ID if Unit is being processed comes
+  /// from a DWO section.
+  void addUnit(DWARFUnit &Unit, const std::optional<uint64_t> &DWOID);
+  /// Returns number of buckets in .debug_name table.
+  ArrayRef<HashList> getBuckets() const { return Buckets; }
+  /// Constructs and returns a unique AbbrevTag that captures what a DIE
+  /// accesses.
+  TagIndex getAbbrevIndex(
+      const unsigned DieTag,
+      const std::optional<DWARF5AccelTable::UnitIndexAndEncoding> &EntryRet,
+      const std::optional<DWARF5AccelTable::UnitIndexAndEncoding>
+          &SecondEntryRe);
+  /// Get encoding for a given attribute.
+  std::optional<DWARF5AccelTable::UnitIndexAndEncoding>
+  getIndexForEntry(const BOLTDWARF5AccelTableData &Value) const;
+  /// Get encoding for a given attribute for second index.
+  /// Returns nullopt if there is no second index.
+  std::optional<DWARF5AccelTable::UnitIndexAndEncoding>
+  getSecondIndexForEntry(const BOLTDWARF5AccelTableData &Value) const;
+  /// Uniquify Entries.
+  void finalize();
+  /// Computes bucket count.
+  void computeBucketCount();
+  /// Populate Abbreviations Map.
+  void populateAbbrevsMap();
+  /// Write Entries.
+  void writeEntries();
+  /// Write an Entry;
+  void writeEntry(const BOLTDWARF5AccelTableData &Entry);
+  /// Write augmentation_string for BOLT.
+  void writeAugmentationString();
+  /// Emit out Header for DWARF5 Accelerator table.
+  void emitHeader() const;
+  /// Emit out CU list.
+  void emitCUList() const;
+  /// Emit out TU List. Combination of LocalTUList and ForeignTUList.
+  void emitTUList() const;
+  /// Emit buckets.
+  void emitBuckets() const;
+  /// Emit hashes for hash table.
+  void emitHashes() const;
+  /// Emit string offsets for hash table.
+  void emitStringOffsets() const;
+  /// Emit Entry Offsets for hash table.
+  void emitOffsets() const;
+  /// Emit abbreviation table.
+  void emitAbbrevs();
+  /// Emit entries.
+  void emitData();
+  /// Emit augmentation string.
+  void emitAugmentationString() const;
+};
+} // namespace bolt
+} // namespace llvm
+#endif

bolt/include/bolt/Rewrite/DWARFRewriter.h

@@ -11,6 +11,7 @@
 
 #include "bolt/Core/DIEBuilder.h"
 #include "bolt/Core/DebugData.h"
+#include "bolt/Core/DebugNames.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/CodeGen/DIE.h"
 #include "llvm/DWP/DWP.h"
@@ -140,8 +141,10 @@ class DWARFRewriter {
                             const std::list<DWARFUnit *> &CUs);
 
   /// Finalize debug sections in the main binary.
-  void finalizeDebugSections(DIEBuilder &DIEBlder, DIEStreamer &Streamer,
-                             raw_svector_ostream &ObjOS, CUOffsetMap &CUMap);
+  void finalizeDebugSections(DIEBuilder &DIEBlder,
+                             DWARF5AcceleratorTable &DebugNamesTable,
+                             DIEStreamer &Streamer, raw_svector_ostream &ObjOS,
+                             CUOffsetMap &CUMap);
 
   /// Patches the binary for DWARF address ranges (e.g. in functions and lexical
   /// blocks) to be updated.

bolt/lib/Core/CMakeLists.txt

@@ -20,6 +20,7 @@ add_llvm_library(LLVMBOLTCore
   BinaryFunctionProfile.cpp
   BinarySection.cpp
   DebugData.cpp
+  DebugNames.cpp
   DIEBuilder.cpp
   DynoStats.cpp
   Exceptions.cpp