1//===- DWARFGdbIndex.cpp --------------------------------------------------===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8
9#include "llvm/DebugInfo/DWARF/DWARFGdbIndex.h"
10#include "llvm/ADT/SmallVector.h"
11#include "llvm/ADT/StringRef.h"
12#include "llvm/Support/Format.h"
13#include "llvm/Support/FormatVariadic.h"
14#include "llvm/Support/raw_ostream.h"
15#include <algorithm>
16#include <cassert>
17#include <cinttypes>
18#include <cstdint>
19#include <utility>
20
21using namespace llvm;
22
23// .gdb_index section format reference:
24// https://sourceware.org/gdb/onlinedocs/gdb/Index-Section-Format.html
25
26void DWARFGdbIndex::dumpCUList(raw_ostream &OS) const {
27  OS << format("\n  CU list offset = 0x%x, has %" PRId64 " entries:",
28               CuListOffset, (uint64_t)CuList.size())
29     << '\n';
30  uint32_t I = 0;
31  for (const CompUnitEntry &CU : CuList)
32    OS << format("    %d: Offset = 0x%llx, Length = 0x%llx\n", I++, CU.Offset,
33                 CU.Length);
34}
35
36void DWARFGdbIndex::dumpTUList(raw_ostream &OS) const {
37  OS << formatv("\n  Types CU list offset = {0:x}, has {1} entries:\n",
38                TuListOffset, TuList.size());
39  uint32_t I = 0;
40  for (const TypeUnitEntry &TU : TuList)
41    OS << formatv("    {0}: offset = {1:x8}, type_offset = {2:x8}, "
42                  "type_signature = {3:x16}\n",
43                  I++, TU.Offset, TU.TypeOffset, TU.TypeSignature);
44}
45
46void DWARFGdbIndex::dumpAddressArea(raw_ostream &OS) const {
47  OS << format("\n  Address area offset = 0x%x, has %" PRId64 " entries:",
48               AddressAreaOffset, (uint64_t)AddressArea.size())
49     << '\n';
50  for (const AddressEntry &Addr : AddressArea)
51    OS << format(
52        "    Low/High address = [0x%llx, 0x%llx) (Size: 0x%llx), CU id = %d\n",
53        Addr.LowAddress, Addr.HighAddress, Addr.HighAddress - Addr.LowAddress,
54        Addr.CuIndex);
55}
56
57void DWARFGdbIndex::dumpSymbolTable(raw_ostream &OS) const {
58  OS << format("\n  Symbol table offset = 0x%x, size = %" PRId64
59               ", filled slots:",
60               SymbolTableOffset, (uint64_t)SymbolTable.size())
61     << '\n';
62  uint32_t I = -1;
63  for (const SymTableEntry &E : SymbolTable) {
64    ++I;
65    if (!E.NameOffset && !E.VecOffset)
66      continue;
67
68    OS << format("    %d: Name offset = 0x%x, CU vector offset = 0x%x\n", I,
69                 E.NameOffset, E.VecOffset);
70
71    StringRef Name = ConstantPoolStrings.substr(
72        ConstantPoolOffset - StringPoolOffset + E.NameOffset);
73
74    auto CuVector = llvm::find_if(
75        ConstantPoolVectors,
76        [&](const std::pair<uint32_t, SmallVector<uint32_t, 0>> &V) {
77          return V.first == E.VecOffset;
78        });
79    assert(CuVector != ConstantPoolVectors.end() && "Invalid symbol table");
80    uint32_t CuVectorId = CuVector - ConstantPoolVectors.begin();
81    OS << format("      String name: %s, CU vector index: %d\n", Name.data(),
82                 CuVectorId);
83  }
84}
85
86void DWARFGdbIndex::dumpConstantPool(raw_ostream &OS) const {
87  OS << format("\n  Constant pool offset = 0x%x, has %" PRId64 " CU vectors:",
88               ConstantPoolOffset, (uint64_t)ConstantPoolVectors.size());
89  uint32_t I = 0;
90  for (const auto &V : ConstantPoolVectors) {
91    OS << format("\n    %d(0x%x): ", I++, V.first);
92    for (uint32_t Val : V.second)
93      OS << format("0x%x ", Val);
94  }
95  OS << '\n';
96}
97
98void DWARFGdbIndex::dump(raw_ostream &OS) {
99  if (HasError) {
100    OS << "\n<error parsing>\n";
101    return;
102  }
103
104  if (HasContent) {
105    OS << "  Version = " << Version << '\n';
106    dumpCUList(OS);
107    dumpTUList(OS);
108    dumpAddressArea(OS);
109    dumpSymbolTable(OS);
110    dumpConstantPool(OS);
111  }
112}
113
114bool DWARFGdbIndex::parseImpl(DataExtractor Data) {
115  uint64_t Offset = 0;
116
117  // Only version 7 is supported at this moment.
118  Version = Data.getU32(&Offset);
119  if (Version != 7)
120    return false;
121
122  CuListOffset = Data.getU32(&Offset);
123  TuListOffset = Data.getU32(&Offset);
124  AddressAreaOffset = Data.getU32(&Offset);
125  SymbolTableOffset = Data.getU32(&Offset);
126  ConstantPoolOffset = Data.getU32(&Offset);
127
128  if (Offset != CuListOffset)
129    return false;
130
131  uint32_t CuListSize = (TuListOffset - CuListOffset) / 16;
132  CuList.reserve(CuListSize);
133  for (uint32_t i = 0; i < CuListSize; ++i) {
134    uint64_t CuOffset = Data.getU64(&Offset);
135    uint64_t CuLength = Data.getU64(&Offset);
136    CuList.push_back({CuOffset, CuLength});
137  }
138
139  // CU Types are no longer needed as DWARF skeleton type units never made it
140  // into the standard.
141  uint32_t TuListSize = (AddressAreaOffset - TuListOffset) / 24;
142  TuList.resize(TuListSize);
143  for (uint32_t I = 0; I < TuListSize; ++I) {
144    uint64_t CuOffset = Data.getU64(&Offset);
145    uint64_t TypeOffset = Data.getU64(&Offset);
146    uint64_t Signature = Data.getU64(&Offset);
147    TuList[I] = {CuOffset, TypeOffset, Signature};
148  }
149
150  uint32_t AddressAreaSize = (SymbolTableOffset - AddressAreaOffset) / 20;
151  AddressArea.reserve(AddressAreaSize);
152  for (uint32_t i = 0; i < AddressAreaSize; ++i) {
153    uint64_t LowAddress = Data.getU64(&Offset);
154    uint64_t HighAddress = Data.getU64(&Offset);
155    uint32_t CuIndex = Data.getU32(&Offset);
156    AddressArea.push_back({LowAddress, HighAddress, CuIndex});
157  }
158
159  // The symbol table. This is an open addressed hash table. The size of the
160  // hash table is always a power of 2.
161  // Each slot in the hash table consists of a pair of offset_type values. The
162  // first value is the offset of the symbol's name in the constant pool. The
163  // second value is the offset of the CU vector in the constant pool.
164  // If both values are 0, then this slot in the hash table is empty. This is ok
165  // because while 0 is a valid constant pool index, it cannot be a valid index
166  // for both a string and a CU vector.
167  uint32_t SymTableSize = (ConstantPoolOffset - SymbolTableOffset) / 8;
168  SymbolTable.reserve(SymTableSize);
169  uint32_t CuVectorsTotal = 0;
170  for (uint32_t i = 0; i < SymTableSize; ++i) {
171    uint32_t NameOffset = Data.getU32(&Offset);
172    uint32_t CuVecOffset = Data.getU32(&Offset);
173    SymbolTable.push_back({NameOffset, CuVecOffset});
174    if (NameOffset || CuVecOffset)
175      ++CuVectorsTotal;
176  }
177
178  // The constant pool. CU vectors are stored first, followed by strings.
179  // The first value is the number of CU indices in the vector. Each subsequent
180  // value is the index and symbol attributes of a CU in the CU list.
181  for (uint32_t i = 0; i < CuVectorsTotal; ++i) {
182    ConstantPoolVectors.emplace_back(0, SmallVector<uint32_t, 0>());
183    auto &Vec = ConstantPoolVectors.back();
184    Vec.first = Offset - ConstantPoolOffset;
185
186    uint32_t Num = Data.getU32(&Offset);
187    for (uint32_t j = 0; j < Num; ++j)
188      Vec.second.push_back(Data.getU32(&Offset));
189  }
190
191  ConstantPoolStrings = Data.getData().drop_front(Offset);
192  StringPoolOffset = Offset;
193  return true;
194}
195
196void DWARFGdbIndex::parse(DataExtractor Data) {
197  HasContent = !Data.getData().empty();
198  HasError = HasContent && !parseImpl(Data);
199}
200