Andrew Scull | 0372a57 | 2018-11-16 15:47:06 +0000 | [diff] [blame] | 1 | //===- SymbolRemappingReader.h - Read symbol remapping file -----*- C++ -*-===// |
| 2 | // |
Andrew Walbran | 16937d0 | 2019-10-22 13:54:20 +0100 | [diff] [blame^] | 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 |
Andrew Scull | 0372a57 | 2018-11-16 15:47:06 +0000 | [diff] [blame] | 6 | // |
| 7 | //===----------------------------------------------------------------------===// |
| 8 | // |
| 9 | // This file contains definitions needed for reading and applying symbol |
| 10 | // remapping files. |
| 11 | // |
| 12 | // Support is provided only for the Itanium C++ name mangling scheme for now. |
| 13 | // |
| 14 | // NOTE: If you are making changes to this file format, please remember |
| 15 | // to document them in the Clang documentation at |
| 16 | // tools/clang/docs/UsersManual.rst. |
| 17 | // |
| 18 | // File format |
| 19 | // ----------- |
| 20 | // |
| 21 | // The symbol remappings are written as an ASCII text file. Blank lines and |
| 22 | // lines starting with a # are ignored. All other lines specify a kind of |
| 23 | // mangled name fragment, along with two fragments of that kind that should |
| 24 | // be treated as equivalent, separated by spaces. |
| 25 | // |
| 26 | // See http://itanium-cxx-abi.github.io/cxx-abi/abi.html#mangling for a |
| 27 | // description of the Itanium name mangling scheme. |
| 28 | // |
| 29 | // The accepted fragment kinds are: |
| 30 | // |
| 31 | // * name A <name>, such as 6foobar or St3__1 |
| 32 | // * type A <type>, such as Ss or N4llvm9StringRefE |
| 33 | // * encoding An <encoding> (a complete mangling without the leading _Z) |
| 34 | // |
| 35 | // For example: |
| 36 | // |
| 37 | // # Ignore int / long differences to treat symbols from 32-bit and 64-bit |
| 38 | // # builds with differing size_t / ptrdiff_t / intptr_t as equivalent. |
| 39 | // type i l |
| 40 | // type j m |
| 41 | // |
| 42 | // # Ignore differences between libc++ and libstdc++, and between libstdc++'s |
| 43 | // # C++98 and C++11 ABIs. |
| 44 | // name 3std St3__1 |
| 45 | // name 3std St7__cxx11 |
| 46 | // |
| 47 | // # Remap a function overload to a specialization of a template (including |
| 48 | // # any local symbols declared within it). |
| 49 | // encoding N2NS1fEi N2NS1fIiEEvT_ |
| 50 | // |
| 51 | // # Substitutions must be remapped separately from namespace 'std' for now. |
| 52 | // name Sa NSt3__19allocatorE |
| 53 | // name Sb NSt3__112basic_stringE |
| 54 | // type Ss NSt3__112basic_stringIcSt11char_traitsIcESaE |
| 55 | // # ... |
| 56 | // |
| 57 | //===----------------------------------------------------------------------===// |
| 58 | |
| 59 | #ifndef LLVM_SUPPORT_SYMBOLREMAPPINGREADER_H |
| 60 | #define LLVM_SUPPORT_SYMBOLREMAPPINGREADER_H |
| 61 | |
| 62 | #include "llvm/ADT/StringRef.h" |
| 63 | #include "llvm/Support/Error.h" |
| 64 | #include "llvm/Support/ItaniumManglingCanonicalizer.h" |
| 65 | #include "llvm/Support/MemoryBuffer.h" |
| 66 | |
| 67 | namespace llvm { |
| 68 | |
| 69 | class SymbolRemappingParseError : public ErrorInfo<SymbolRemappingParseError> { |
| 70 | public: |
| 71 | SymbolRemappingParseError(StringRef File, int64_t Line, Twine Message) |
| 72 | : File(File), Line(Line), Message(Message.str()) {} |
| 73 | |
| 74 | void log(llvm::raw_ostream &OS) const override { |
| 75 | OS << File << ':' << Line << ": " << Message; |
| 76 | } |
| 77 | std::error_code convertToErrorCode() const override { |
| 78 | return llvm::inconvertibleErrorCode(); |
| 79 | } |
| 80 | |
| 81 | StringRef getFileName() const { return File; } |
| 82 | int64_t getLineNum() const { return Line; } |
| 83 | StringRef getMessage() const { return Message; } |
| 84 | |
| 85 | static char ID; |
| 86 | |
| 87 | private: |
| 88 | std::string File; |
| 89 | int64_t Line; |
| 90 | std::string Message; |
| 91 | }; |
| 92 | |
| 93 | /// Reader for symbol remapping files. |
| 94 | /// |
| 95 | /// Remaps the symbol names in profile data to match those in the program |
| 96 | /// according to a set of rules specified in a given file. |
| 97 | class SymbolRemappingReader { |
| 98 | public: |
| 99 | /// Read remappings from the given buffer, which must live as long as |
| 100 | /// the remapper. |
| 101 | Error read(MemoryBuffer &B); |
| 102 | |
| 103 | /// A Key represents an equivalence class of symbol names. |
| 104 | using Key = uintptr_t; |
| 105 | |
| 106 | /// Construct a key for the given symbol, or return an existing one if an |
| 107 | /// equivalent name has already been inserted. The symbol name must live |
| 108 | /// as long as the remapper. |
| 109 | /// |
| 110 | /// The result will be Key() if the name cannot be remapped (typically |
| 111 | /// because it is not a valid mangled name). |
| 112 | Key insert(StringRef FunctionName) { |
| 113 | return Canonicalizer.canonicalize(FunctionName); |
| 114 | } |
| 115 | |
| 116 | /// Map the given symbol name into the key for the corresponding equivalence |
| 117 | /// class. |
| 118 | /// |
| 119 | /// The result will typically be Key() if no equivalent symbol has been |
| 120 | /// inserted, but this is not guaranteed: a Key different from all keys ever |
| 121 | /// returned by \c insert may be returned instead. |
| 122 | Key lookup(StringRef FunctionName) { |
| 123 | return Canonicalizer.lookup(FunctionName); |
| 124 | } |
| 125 | |
| 126 | private: |
| 127 | ItaniumManglingCanonicalizer Canonicalizer; |
| 128 | }; |
| 129 | |
| 130 | } // end namespace llvm |
| 131 | |
| 132 | #endif // LLVM_SUPPORT_SYMBOLREMAPPINGREADER_H |