1//===- SymbolTable.h --------------------------------------------*- C++ -*-===//
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#ifndef LLD_WASM_SYMBOL_TABLE_H
10#define LLD_WASM_SYMBOL_TABLE_H
11
12#include "InputFiles.h"
13#include "LTO.h"
14#include "Symbols.h"
15#include "lld/Common/LLVM.h"
16#include "llvm/ADT/CachedHashString.h"
17#include "llvm/ADT/DenseSet.h"
18#include "llvm/BinaryFormat/WasmTraits.h"
19#include <optional>
20
21namespace lld::wasm {
22
23class InputSegment;
24
25// SymbolTable is a bucket of all known symbols, including defined,
26// undefined, or lazy symbols (the last one is symbols in archive
27// files whose archive members are not yet loaded).
28//
29// We put all symbols of all files to a SymbolTable, and the
30// SymbolTable selects the "best" symbols if there are name
31// conflicts. For example, obviously, a defined symbol is better than
32// an undefined symbol. Or, if there's a conflict between a lazy and a
33// undefined, it'll read an archive member to read a real definition
34// to replace the lazy symbol. The logic is implemented in the
35// add*() functions, which are called by input files as they are parsed.
36// There is one add* function per symbol type.
37class SymbolTable {
38public:
39 ArrayRef<Symbol *> symbols() const { return symVector; }
40
41 void wrap(Symbol *sym, Symbol *real, Symbol *wrap);
42
43 void addFile(InputFile *file, StringRef symName = {});
44
45 void compileBitcodeFiles();
46
47 Symbol *find(StringRef name);
48
49 void replace(StringRef name, Symbol* sym);
50
51 void trace(StringRef name);
52
53 Symbol *addSharedFunction(StringRef name, uint32_t flags, InputFile *file,
54 const WasmSignature *sig);
55 Symbol *addSharedData(StringRef name, uint32_t flags, InputFile *file);
56 Symbol *addDefinedFunction(StringRef name, uint32_t flags, InputFile *file,
57 InputFunction *function);
58 Symbol *addDefinedData(StringRef name, uint32_t flags, InputFile *file,
59 InputChunk *segment, uint64_t address, uint64_t size);
60 Symbol *addDefinedGlobal(StringRef name, uint32_t flags, InputFile *file,
61 InputGlobal *g);
62 Symbol *addDefinedTag(StringRef name, uint32_t flags, InputFile *file,
63 InputTag *t);
64 Symbol *addDefinedTable(StringRef name, uint32_t flags, InputFile *file,
65 InputTable *t);
66
67 Symbol *addUndefinedFunction(StringRef name,
68 std::optional<StringRef> importName,
69 std::optional<StringRef> importModule,
70 uint32_t flags, InputFile *file,
71 const WasmSignature *signature,
72 bool isCalledDirectly);
73 Symbol *addUndefinedData(StringRef name, uint32_t flags, InputFile *file);
74 Symbol *addUndefinedGlobal(StringRef name,
75 std::optional<StringRef> importName,
76 std::optional<StringRef> importModule,
77 uint32_t flags, InputFile *file,
78 const WasmGlobalType *type);
79 Symbol *addUndefinedTable(StringRef name, std::optional<StringRef> importName,
80 std::optional<StringRef> importModule,
81 uint32_t flags, InputFile *file,
82 const WasmTableType *type);
83 Symbol *addUndefinedTag(StringRef name, std::optional<StringRef> importName,
84 std::optional<StringRef> importModule, uint32_t flags,
85 InputFile *file, const WasmSignature *sig);
86
87 TableSymbol *resolveIndirectFunctionTable(bool required);
88
89 void addLazy(StringRef name, InputFile *f);
90
91 bool addComdat(StringRef name);
92
93 DefinedData *addSyntheticDataSymbol(StringRef name, uint32_t flags);
94 DefinedGlobal *addSyntheticGlobal(StringRef name, uint32_t flags,
95 InputGlobal *global);
96 DefinedFunction *addSyntheticFunction(StringRef name, uint32_t flags,
97 InputFunction *function);
98 DefinedData *addOptionalDataSymbol(StringRef name, uint64_t value = 0);
99 DefinedGlobal *addOptionalGlobalSymbol(StringRef name, InputGlobal *global);
100 DefinedTable *addSyntheticTable(StringRef name, uint32_t flags,
101 InputTable *global);
102
103 void handleSymbolVariants();
104 void handleWeakUndefines();
105 DefinedFunction *createUndefinedStub(const WasmSignature &sig);
106
107private:
108 std::pair<Symbol *, bool> insert(StringRef name, const InputFile *file);
109 std::pair<Symbol *, bool> insertName(StringRef name);
110
111 bool getFunctionVariant(Symbol* sym, const WasmSignature *sig,
112 const InputFile *file, Symbol **out);
113 InputFunction *replaceWithUnreachable(Symbol *sym, const WasmSignature &sig,
114 StringRef debugName);
115 void replaceWithUndefined(Symbol *sym);
116
117 TableSymbol *createDefinedIndirectFunctionTable(StringRef name);
118 TableSymbol *createUndefinedIndirectFunctionTable(StringRef name);
119
120 // Maps symbol names to index into the symVector. -1 means that symbols
121 // is to not yet in the vector but it should have tracing enabled if it is
122 // ever added.
123 llvm::DenseMap<llvm::CachedHashStringRef, int> symMap;
124 std::vector<Symbol *> symVector;
125
126 // For certain symbols types, e.g. function symbols, we allow for multiple
127 // variants of the same symbol with different signatures.
128 llvm::DenseMap<llvm::CachedHashStringRef, std::vector<Symbol *>> symVariants;
129 llvm::DenseMap<WasmSignature, DefinedFunction *> stubFunctions;
130
131 // Comdat groups define "link once" sections. If two comdat groups have the
132 // same name, only one of them is linked, and the other is ignored. This set
133 // is used to uniquify them.
134 llvm::DenseSet<llvm::CachedHashStringRef> comdatGroups;
135
136 // For LTO.
137 std::unique_ptr<BitcodeCompiler> lto;
138};
139
140extern SymbolTable *symtab;
141
142} // namespace lld::wasm
143
144#endif
145