1 //===- StringTableBuilder.cpp - String table building utility -------------===//
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
7 //===----------------------------------------------------------------------===//
9 #include "llvm/MC/StringTableBuilder.h"
10 #include "llvm/ADT/CachedHashString.h"
11 #include "llvm/ADT/SmallString.h"
12 #include "llvm/ADT/StringRef.h"
13 #include "llvm/BinaryFormat/COFF.h"
14 #include "llvm/Support/Endian.h"
15 #include "llvm/Support/MathExtras.h"
16 #include "llvm/Support/raw_ostream.h"
26 StringTableBuilder::~StringTableBuilder() = default;
28 void StringTableBuilder::initSize() {
29 // Account for leading bytes in table so that offsets returned from add are
38 // Start the table with a NUL byte.
43 // Make room to write the table size later.
49 StringTableBuilder::StringTableBuilder(Kind K, unsigned Alignment)
50 : K(K), Alignment(Alignment) {
54 void StringTableBuilder::write(raw_ostream &OS) const {
55 assert(isFinalized());
57 Data.resize(getSize());
58 write((uint8_t *)Data.data());
62 using StringPair = std::pair<CachedHashStringRef, size_t>;
64 void StringTableBuilder::write(uint8_t *Buf) const {
65 assert(isFinalized());
66 for (const StringPair &P : StringIndexMap) {
67 StringRef Data = P.first.val();
69 memcpy(Buf + P.second, Data.data(), Data.size());
71 // The COFF formats store the size of the string table in the first 4 bytes.
72 // For Windows, the format is little-endian; for AIX, it is big-endian.
74 support::endian::write32le(Buf, Size);
76 support::endian::write32be(Buf, Size);
79 // Returns the character at Pos from end of a string.
80 static int charTailAt(StringPair *P, size_t Pos) {
81 StringRef S = P->first.val();
84 return (unsigned char)S[S.size() - Pos - 1];
87 // Three-way radix quicksort. This is much faster than std::sort with strcmp
88 // because it does not compare characters that we already know the same.
89 static void multikeySort(MutableArrayRef<StringPair *> Vec, int Pos) {
94 // Partition items so that items in [0, I) are greater than the pivot,
95 // [I, J) are the same as the pivot, and [J, Vec.size()) are less than
97 int Pivot = charTailAt(Vec[0], Pos);
99 size_t J = Vec.size();
100 for (size_t K = 1; K < J;) {
101 int C = charTailAt(Vec[K], Pos);
103 std::swap(Vec[I++], Vec[K++]);
105 std::swap(Vec[--J], Vec[K]);
110 multikeySort(Vec.slice(0, I), Pos);
111 multikeySort(Vec.slice(J), Pos);
113 // multikeySort(Vec.slice(I, J - I), Pos + 1), but with
114 // tail call optimization.
116 Vec = Vec.slice(I, J - I);
122 void StringTableBuilder::finalize() {
124 finalizeStringTable(/*Optimize=*/true);
127 void StringTableBuilder::finalizeInOrder() {
128 finalizeStringTable(/*Optimize=*/false);
131 void StringTableBuilder::finalizeStringTable(bool Optimize) {
135 std::vector<StringPair *> Strings;
136 Strings.reserve(StringIndexMap.size());
137 for (StringPair &P : StringIndexMap)
138 Strings.push_back(&P);
140 multikeySort(Strings, 0);
144 for (StringPair *P : Strings) {
145 StringRef S = P->first.val();
146 if (Previous.endswith(S)) {
147 size_t Pos = Size - S.size() - (K != RAW);
148 if (!(Pos & (Alignment - 1))) {
154 Size = alignTo(Size, Alignment);
165 Size = alignTo(Size, 4); // Pad to multiple of 4.
167 // The first byte in an ELF string table must be null, according to the ELF
168 // specification. In 'initSize()' we reserved the first byte to hold null for
169 // this purpose and here we actually add the string to allow 'getOffset()' to
170 // be called on an empty string.
172 StringIndexMap[CachedHashStringRef("")] = 0;
175 void StringTableBuilder::clear() {
177 StringIndexMap.clear();
180 size_t StringTableBuilder::getOffset(CachedHashStringRef S) const {
181 assert(isFinalized());
182 auto I = StringIndexMap.find(S);
183 assert(I != StringIndexMap.end() && "String is not in table!");
187 size_t StringTableBuilder::add(CachedHashStringRef S) {
189 assert(S.size() > COFF::NameSize && "Short string in COFF string table!");
191 assert(!isFinalized());
192 auto P = StringIndexMap.insert(std::make_pair(S, 0));
194 size_t Start = alignTo(Size, Alignment);
195 P.first->second = Start;
196 Size = Start + S.size() + (K != RAW);
198 return P.first->second;