reference, declarationdefinition
definition → references, declarations, derived classes, virtual overrides
reference to multiple definitions → definitions
unreferenced
    1
    2
    3
    4
    5
    6
    7
    8
    9
   10
   11
   12
   13
   14
   15
   16
   17
   18
   19
   20
   21
   22
   23
   24
   25
   26
   27
   28
   29
   30
   31
   32
   33
   34
   35
   36
   37
   38
   39
   40
   41
   42
   43
   44
   45
   46
   47
   48
   49
   50
   51
   52
   53
   54
   55
   56
   57
   58
   59
   60
   61
   62
   63
   64
   65
   66
   67
   68
   69
   70
   71
   72
   73
   74
   75
   76
   77
   78
   79
   80
   81
   82
   83
   84
   85
   86
   87
   88
   89
   90
   91
   92
   93
   94
   95
   96
   97
   98
   99
  100
  101
  102
  103
  104
  105
  106
  107
  108
  109
  110
  111
  112
  113
  114
  115
  116
  117
  118
  119
  120
  121
  122
  123
  124
  125
  126
  127
  128
  129
  130
  131
  132
  133
  134
  135
  136
  137
  138
  139
  140
  141
  142
  143
  144
  145
  146
  147
  148
  149
  150
  151
  152
  153
  154
  155
  156
  157
  158
  159
  160
  161
  162
  163
  164
  165
  166
  167
  168
  169
  170
  171
  172
  173
  174
  175
  176
  177
  178
  179
  180
  181
//===- GlobalsStream.cpp - PDB Index of Symbols by Name ---------*- 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
//
//===----------------------------------------------------------------------===//
//
// The on-disk structores used in this file are based on the reference
// implementation which is available at
// https://github.com/Microsoft/microsoft-pdb/blob/master/PDB/dbi/gsi.h
//
// When you are reading the reference source code, you'd find the
// information below useful.
//
//  - ppdb1->m_fMinimalDbgInfo seems to be always true.
//  - SMALLBUCKETS macro is defined.
//
//===----------------------------------------------------------------------===//

#include "llvm/DebugInfo/PDB/Native/GlobalsStream.h"

#include "llvm/DebugInfo/CodeView/RecordName.h"
#include "llvm/DebugInfo/PDB/Native/Hash.h"
#include "llvm/DebugInfo/PDB/Native/RawError.h"
#include "llvm/DebugInfo/PDB/Native/SymbolStream.h"
#include "llvm/Support/BinaryStreamReader.h"
#include "llvm/Support/Error.h"
#include <algorithm>

using namespace llvm;
using namespace llvm::msf;
using namespace llvm::pdb;

GlobalsStream::GlobalsStream(std::unique_ptr<MappedBlockStream> Stream)
    : Stream(std::move(Stream)) {}

GlobalsStream::~GlobalsStream() = default;

Error GlobalsStream::reload() {
  BinaryStreamReader Reader(*Stream);
  if (auto E = GlobalsTable.read(Reader))
    return E;
  return Error::success();
}

std::vector<std::pair<uint32_t, codeview::CVSymbol>>
GlobalsStream::findRecordsByName(StringRef Name,
                                 const SymbolStream &Symbols) const {
  std::vector<std::pair<uint32_t, codeview::CVSymbol>> Result;

  // Hash the name to figure out which bucket this goes into.
  size_t ExpandedBucketIndex = hashStringV1(Name) % IPHR_HASH;
  int32_t CompressedBucketIndex = GlobalsTable.BucketMap[ExpandedBucketIndex];
  if (CompressedBucketIndex == -1)
    return Result;

  uint32_t LastBucketIndex = GlobalsTable.HashBuckets.size() - 1;
  uint32_t StartRecordIndex =
      GlobalsTable.HashBuckets[CompressedBucketIndex] / 12;
  uint32_t EndRecordIndex = 0;
  if (LLVM_LIKELY(uint32_t(CompressedBucketIndex) < LastBucketIndex)) {
    EndRecordIndex = GlobalsTable.HashBuckets[CompressedBucketIndex + 1];
  } else {
    // If this is the last bucket, it consists of all hash records until the end
    // of the HashRecords array.
    EndRecordIndex = GlobalsTable.HashRecords.size() * 12;
  }

  EndRecordIndex /= 12;

  assert(EndRecordIndex <= GlobalsTable.HashRecords.size());
  while (StartRecordIndex < EndRecordIndex) {
    PSHashRecord PSH = GlobalsTable.HashRecords[StartRecordIndex];
    uint32_t Off = PSH.Off - 1;
    codeview::CVSymbol Record = Symbols.readRecord(Off);
    if (codeview::getSymbolName(Record) == Name)
      Result.push_back(std::make_pair(Off, std::move(Record)));
    ++StartRecordIndex;
  }
  return Result;
}

static Error checkHashHdrVersion(const GSIHashHeader *HashHdr) {
  if (HashHdr->VerHdr != GSIHashHeader::HdrVersion)
    return make_error<RawError>(
        raw_error_code::feature_unsupported,
        "Encountered unsupported globals stream version.");

  return Error::success();
}

static Error readGSIHashHeader(const GSIHashHeader *&HashHdr,
                               BinaryStreamReader &Reader) {
  if (Reader.readObject(HashHdr))
    return make_error<RawError>(raw_error_code::corrupt_file,
                                "Stream does not contain a GSIHashHeader.");

  if (HashHdr->VerSignature != GSIHashHeader::HdrSignature)
    return make_error<RawError>(
        raw_error_code::feature_unsupported,
        "GSIHashHeader signature (0xffffffff) not found.");

  return Error::success();
}

static Error readGSIHashRecords(FixedStreamArray<PSHashRecord> &HashRecords,
                                const GSIHashHeader *HashHdr,
                                BinaryStreamReader &Reader) {
  if (auto EC = checkHashHdrVersion(HashHdr))
    return EC;

  // HashHdr->HrSize specifies the number of bytes of PSHashRecords we have.
  // Verify that we can read them all.
  if (HashHdr->HrSize % sizeof(PSHashRecord))
    return make_error<RawError>(raw_error_code::corrupt_file,
                                "Invalid HR array size.");
  uint32_t NumHashRecords = HashHdr->HrSize / sizeof(PSHashRecord);
  if (auto EC = Reader.readArray(HashRecords, NumHashRecords))
    return joinErrors(std::move(EC),
                      make_error<RawError>(raw_error_code::corrupt_file,
                                           "Error reading hash records."));

  return Error::success();
}

static Error
readGSIHashBuckets(FixedStreamArray<support::ulittle32_t> &HashBuckets,
                   FixedStreamArray<support::ulittle32_t> &HashBitmap,
                   const GSIHashHeader *HashHdr,
                   MutableArrayRef<int32_t> BucketMap,
                   BinaryStreamReader &Reader) {
  if (auto EC = checkHashHdrVersion(HashHdr))
    return EC;

  // Before the actual hash buckets, there is a bitmap of length determined by
  // IPHR_HASH.
  size_t BitmapSizeInBits = alignTo(IPHR_HASH + 1, 32);
  uint32_t NumBitmapEntries = BitmapSizeInBits / 32;
  if (auto EC = Reader.readArray(HashBitmap, NumBitmapEntries))
    return joinErrors(std::move(EC),
                      make_error<RawError>(raw_error_code::corrupt_file,
                                           "Could not read a bitmap."));
  uint32_t NumBuckets1 = 0;
  uint32_t CompressedBucketIdx = 0;
  for (uint32_t I = 0; I <= IPHR_HASH; ++I) {
    uint8_t WordIdx = I / 32;
    uint8_t BitIdx = I % 32;
    bool IsSet = HashBitmap[WordIdx] & (1U << BitIdx);
    if (IsSet) {
      ++NumBuckets1;
      BucketMap[I] = CompressedBucketIdx++;
    } else {
      BucketMap[I] = -1;
    }
  }

  uint32_t NumBuckets = 0;
  for (uint32_t B : HashBitmap)
    NumBuckets += countPopulation(B);

  // Hash buckets follow.
  if (auto EC = Reader.readArray(HashBuckets, NumBuckets))
    return joinErrors(std::move(EC),
                      make_error<RawError>(raw_error_code::corrupt_file,
                                           "Hash buckets corrupted."));

  return Error::success();
}

Error GSIHashTable::read(BinaryStreamReader &Reader) {
  if (auto EC = readGSIHashHeader(HashHdr, Reader))
    return EC;
  if (auto EC = readGSIHashRecords(HashRecords, HashHdr, Reader))
    return EC;
  if (HashHdr->HrSize > 0)
    if (auto EC = readGSIHashBuckets(HashBuckets, HashBitmap, HashHdr,
                                     BucketMap, Reader))
      return EC;
  return Error::success();
}