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 21 namespace lld { 22 namespace wasm { 23 24 class InputSegment; 25 26 // SymbolTable is a bucket of all known symbols, including defined, 27 // undefined, or lazy symbols (the last one is symbols in archive 28 // files whose archive members are not yet loaded). 29 // 30 // We put all symbols of all files to a SymbolTable, and the 31 // SymbolTable selects the "best" symbols if there are name 32 // conflicts. For example, obviously, a defined symbol is better than 33 // an undefined symbol. Or, if there's a conflict between a lazy and a 34 // undefined, it'll read an archive member to read a real definition 35 // to replace the lazy symbol. The logic is implemented in the 36 // add*() functions, which are called by input files as they are parsed. 37 // There is one add* function per symbol type. 38 class SymbolTable { 39 public: symbols()40 ArrayRef<Symbol *> symbols() const { return symVector; } 41 42 void wrap(Symbol *sym, Symbol *real, Symbol *wrap); 43 44 void addFile(InputFile *file); 45 46 void compileBitcodeFiles(); 47 48 Symbol *find(StringRef name); 49 50 void replace(StringRef name, Symbol* sym); 51 52 void trace(StringRef name); 53 54 Symbol *addDefinedFunction(StringRef name, uint32_t flags, InputFile *file, 55 InputFunction *function); 56 Symbol *addDefinedData(StringRef name, uint32_t flags, InputFile *file, 57 InputChunk *segment, uint64_t address, uint64_t size); 58 Symbol *addDefinedGlobal(StringRef name, uint32_t flags, InputFile *file, 59 InputGlobal *g); 60 Symbol *addDefinedTag(StringRef name, uint32_t flags, InputFile *file, 61 InputTag *t); 62 Symbol *addDefinedTable(StringRef name, uint32_t flags, InputFile *file, 63 InputTable *t); 64 65 Symbol *addUndefinedFunction(StringRef name, 66 std::optional<StringRef> importName, 67 std::optional<StringRef> importModule, 68 uint32_t flags, InputFile *file, 69 const WasmSignature *signature, 70 bool isCalledDirectly); 71 Symbol *addUndefinedData(StringRef name, uint32_t flags, InputFile *file); 72 Symbol *addUndefinedGlobal(StringRef name, 73 std::optional<StringRef> importName, 74 std::optional<StringRef> importModule, 75 uint32_t flags, InputFile *file, 76 const WasmGlobalType *type); 77 Symbol *addUndefinedTable(StringRef name, std::optional<StringRef> importName, 78 std::optional<StringRef> importModule, 79 uint32_t flags, InputFile *file, 80 const WasmTableType *type); 81 Symbol *addUndefinedTag(StringRef name, std::optional<StringRef> importName, 82 std::optional<StringRef> importModule, uint32_t flags, 83 InputFile *file, const WasmSignature *sig); 84 85 TableSymbol *resolveIndirectFunctionTable(bool required); 86 87 void addLazy(ArchiveFile *f, const llvm::object::Archive::Symbol *sym); 88 89 bool addComdat(StringRef name); 90 91 DefinedData *addSyntheticDataSymbol(StringRef name, uint32_t flags); 92 DefinedGlobal *addSyntheticGlobal(StringRef name, uint32_t flags, 93 InputGlobal *global); 94 DefinedFunction *addSyntheticFunction(StringRef name, uint32_t flags, 95 InputFunction *function); 96 DefinedData *addOptionalDataSymbol(StringRef name, uint64_t value = 0); 97 DefinedGlobal *addOptionalGlobalSymbol(StringRef name, InputGlobal *global); 98 DefinedTable *addSyntheticTable(StringRef name, uint32_t flags, 99 InputTable *global); 100 101 void handleSymbolVariants(); 102 void handleWeakUndefines(); 103 DefinedFunction *createUndefinedStub(const WasmSignature &sig); 104 105 std::vector<ObjFile *> objectFiles; 106 std::vector<StubFile *> stubFiles; 107 std::vector<SharedFile *> sharedFiles; 108 std::vector<BitcodeFile *> bitcodeFiles; 109 std::vector<InputFunction *> syntheticFunctions; 110 std::vector<InputGlobal *> syntheticGlobals; 111 std::vector<InputTable *> syntheticTables; 112 113 private: 114 std::pair<Symbol *, bool> insert(StringRef name, const InputFile *file); 115 std::pair<Symbol *, bool> insertName(StringRef name); 116 117 bool getFunctionVariant(Symbol* sym, const WasmSignature *sig, 118 const InputFile *file, Symbol **out); 119 InputFunction *replaceWithUnreachable(Symbol *sym, const WasmSignature &sig, 120 StringRef debugName); 121 void replaceWithUndefined(Symbol *sym); 122 123 TableSymbol *createDefinedIndirectFunctionTable(StringRef name); 124 TableSymbol *createUndefinedIndirectFunctionTable(StringRef name); 125 126 // Maps symbol names to index into the symVector. -1 means that symbols 127 // is to not yet in the vector but it should have tracing enabled if it is 128 // ever added. 129 llvm::DenseMap<llvm::CachedHashStringRef, int> symMap; 130 std::vector<Symbol *> symVector; 131 132 // For certain symbols types, e.g. function symbols, we allow for multiple 133 // variants of the same symbol with different signatures. 134 llvm::DenseMap<llvm::CachedHashStringRef, std::vector<Symbol *>> symVariants; 135 llvm::DenseMap<WasmSignature, DefinedFunction *> stubFunctions; 136 137 // Comdat groups define "link once" sections. If two comdat groups have the 138 // same name, only one of them is linked, and the other is ignored. This set 139 // is used to uniquify them. 140 llvm::DenseSet<llvm::CachedHashStringRef> comdatGroups; 141 142 // For LTO. 143 std::unique_ptr<BitcodeCompiler> lto; 144 }; 145 146 extern SymbolTable *symtab; 147 148 } // namespace wasm 149 } // namespace lld 150 151 #endif 152