10b57cec5SDimitry Andric //===- SymbolTable.h --------------------------------------------*- C++ -*-===// 20b57cec5SDimitry Andric // 30b57cec5SDimitry Andric // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. 40b57cec5SDimitry Andric // See https://llvm.org/LICENSE.txt for license information. 50b57cec5SDimitry Andric // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception 60b57cec5SDimitry Andric // 70b57cec5SDimitry Andric //===----------------------------------------------------------------------===// 80b57cec5SDimitry Andric 90b57cec5SDimitry Andric #ifndef LLD_ELF_SYMBOL_TABLE_H 100b57cec5SDimitry Andric #define LLD_ELF_SYMBOL_TABLE_H 110b57cec5SDimitry Andric 120b57cec5SDimitry Andric #include "InputFiles.h" 130b57cec5SDimitry Andric #include "Symbols.h" 140b57cec5SDimitry Andric #include "lld/Common/Strings.h" 150b57cec5SDimitry Andric #include "llvm/ADT/CachedHashString.h" 160b57cec5SDimitry Andric #include "llvm/ADT/DenseMap.h" 170b57cec5SDimitry Andric #include "llvm/ADT/STLExtras.h" 180b57cec5SDimitry Andric 190b57cec5SDimitry Andric namespace lld { 200b57cec5SDimitry Andric namespace elf { 210b57cec5SDimitry Andric 220b57cec5SDimitry Andric // SymbolTable is a bucket of all known symbols, including defined, 230b57cec5SDimitry Andric // undefined, or lazy symbols (the last one is symbols in archive 240b57cec5SDimitry Andric // files whose archive members are not yet loaded). 250b57cec5SDimitry Andric // 260b57cec5SDimitry Andric // We put all symbols of all files to a SymbolTable, and the 270b57cec5SDimitry Andric // SymbolTable selects the "best" symbols if there are name 280b57cec5SDimitry Andric // conflicts. For example, obviously, a defined symbol is better than 290b57cec5SDimitry Andric // an undefined symbol. Or, if there's a conflict between a lazy and a 300b57cec5SDimitry Andric // undefined, it'll read an archive member to read a real definition 310b57cec5SDimitry Andric // to replace the lazy symbol. The logic is implemented in the 320b57cec5SDimitry Andric // add*() functions, which are called by input files as they are parsed. There 330b57cec5SDimitry Andric // is one add* function per symbol type. 340b57cec5SDimitry Andric class SymbolTable { 35480093f4SDimitry Andric struct FilterOutPlaceholder { 36480093f4SDimitry Andric bool operator()(Symbol *S) const { return !S->isPlaceholder(); } 37480093f4SDimitry Andric }; 38480093f4SDimitry Andric using iterator = llvm::filter_iterator<std::vector<Symbol *>::const_iterator, 39480093f4SDimitry Andric FilterOutPlaceholder>; 400b57cec5SDimitry Andric 41480093f4SDimitry Andric public: 42480093f4SDimitry Andric llvm::iterator_range<iterator> symbols() const { 43480093f4SDimitry Andric return llvm::make_filter_range(symVector, FilterOutPlaceholder()); 440b57cec5SDimitry Andric } 450b57cec5SDimitry Andric 46480093f4SDimitry Andric void wrap(Symbol *sym, Symbol *real, Symbol *wrap); 47480093f4SDimitry Andric 480b57cec5SDimitry Andric Symbol *insert(StringRef name); 490b57cec5SDimitry Andric 5085868e8aSDimitry Andric Symbol *addSymbol(const Symbol &newSym); 510b57cec5SDimitry Andric 520b57cec5SDimitry Andric void scanVersionScript(); 530b57cec5SDimitry Andric 540b57cec5SDimitry Andric Symbol *find(StringRef name); 550b57cec5SDimitry Andric 560b57cec5SDimitry Andric void handleDynamicList(); 570b57cec5SDimitry Andric 580b57cec5SDimitry Andric // Set of .so files to not link the same shared object file more than once. 590b57cec5SDimitry Andric llvm::DenseMap<StringRef, SharedFile *> soNames; 600b57cec5SDimitry Andric 610b57cec5SDimitry Andric // Comdat groups define "link once" sections. If two comdat groups have the 620b57cec5SDimitry Andric // same name, only one of them is linked, and the other is ignored. This map 630b57cec5SDimitry Andric // is used to uniquify them. 640b57cec5SDimitry Andric llvm::DenseMap<llvm::CachedHashStringRef, const InputFile *> comdatGroups; 650b57cec5SDimitry Andric 660b57cec5SDimitry Andric private: 670b57cec5SDimitry Andric std::vector<Symbol *> findByVersion(SymbolVersion ver); 68*6e75b2fbSDimitry Andric std::vector<Symbol *> findAllByVersion(SymbolVersion ver, 69*6e75b2fbSDimitry Andric bool includeNonDefault); 700b57cec5SDimitry Andric 710b57cec5SDimitry Andric llvm::StringMap<std::vector<Symbol *>> &getDemangledSyms(); 72*6e75b2fbSDimitry Andric bool assignExactVersion(SymbolVersion ver, uint16_t versionId, 73*6e75b2fbSDimitry Andric StringRef versionName, bool includeNonDefault); 74*6e75b2fbSDimitry Andric void assignWildcardVersion(SymbolVersion ver, uint16_t versionId, 75*6e75b2fbSDimitry Andric bool includeNonDefault); 760b57cec5SDimitry Andric 770b57cec5SDimitry Andric // The order the global symbols are in is not defined. We can use an arbitrary 780b57cec5SDimitry Andric // order, but it has to be reproducible. That is true even when cross linking. 790b57cec5SDimitry Andric // The default hashing of StringRef produces different results on 32 and 64 800b57cec5SDimitry Andric // bit systems so we use a map to a vector. That is arbitrary, deterministic 810b57cec5SDimitry Andric // but a bit inefficient. 820b57cec5SDimitry Andric // FIXME: Experiment with passing in a custom hashing or sorting the symbols 830b57cec5SDimitry Andric // once symbol resolution is finished. 840b57cec5SDimitry Andric llvm::DenseMap<llvm::CachedHashStringRef, int> symMap; 850b57cec5SDimitry Andric std::vector<Symbol *> symVector; 860b57cec5SDimitry Andric 870b57cec5SDimitry Andric // A map from demangled symbol names to their symbol objects. 880b57cec5SDimitry Andric // This mapping is 1:N because two symbols with different versions 890b57cec5SDimitry Andric // can have the same name. We use this map to handle "extern C++ {}" 900b57cec5SDimitry Andric // directive in version scripts. 910b57cec5SDimitry Andric llvm::Optional<llvm::StringMap<std::vector<Symbol *>>> demangledSyms; 920b57cec5SDimitry Andric }; 930b57cec5SDimitry Andric 940b57cec5SDimitry Andric extern SymbolTable *symtab; 950b57cec5SDimitry Andric 960b57cec5SDimitry Andric } // namespace elf 970b57cec5SDimitry Andric } // namespace lld 980b57cec5SDimitry Andric 990b57cec5SDimitry Andric #endif 100