1 //===- SymbolRemappingReader.h - Read symbol remapping file -----*- C++ -*-===//
2 //
3 //                      The LLVM Compiler Infrastructure
4 //
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
7 //
8 //===----------------------------------------------------------------------===//
9 //
10 // This file contains definitions needed for reading and applying symbol
11 // remapping files.
12 //
13 // Support is provided only for the Itanium C++ name mangling scheme for now.
14 //
15 // NOTE: If you are making changes to this file format, please remember
16 //       to document them in the Clang documentation at
17 //       tools/clang/docs/UsersManual.rst.
18 //
19 // File format
20 // -----------
21 //
22 // The symbol remappings are written as an ASCII text file. Blank lines and
23 // lines starting with a # are ignored. All other lines specify a kind of
24 // mangled name fragment, along with two fragments of that kind that should
25 // be treated as equivalent, separated by spaces.
26 //
27 // See http://itanium-cxx-abi.github.io/cxx-abi/abi.html#mangling for a
28 // description of the Itanium name mangling scheme.
29 //
30 // The accepted fragment kinds are:
31 //
32 //  * name  A <name>, such as 6foobar or St3__1
33 //  * type  A <type>, such as Ss or N4llvm9StringRefE
34 //  * encoding  An <encoding> (a complete mangling without the leading _Z)
35 //
36 // For example:
37 //
38 // # Ignore int / long differences to treat symbols from 32-bit and 64-bit
39 // # builds with differing size_t / ptrdiff_t / intptr_t as equivalent.
40 // type i l
41 // type j m
42 //
43 // # Ignore differences between libc++ and libstdc++, and between libstdc++'s
44 // # C++98 and C++11 ABIs.
45 // name 3std St3__1
46 // name 3std St7__cxx11
47 //
48 // # Remap a function overload to a specialization of a template (including
49 // # any local symbols declared within it).
50 // encoding N2NS1fEi N2NS1fIiEEvT_
51 //
52 // # Substitutions must be remapped separately from namespace 'std' for now.
53 // name Sa NSt3__19allocatorE
54 // name Sb NSt3__112basic_stringE
55 // type Ss NSt3__112basic_stringIcSt11char_traitsIcESaE
56 // # ...
57 //
58 //===----------------------------------------------------------------------===//
59 
60 #ifndef LLVM_SUPPORT_SYMBOLREMAPPINGREADER_H
61 #define LLVM_SUPPORT_SYMBOLREMAPPINGREADER_H
62 
63 #include "llvm/ADT/StringRef.h"
64 #include "llvm/Support/Error.h"
65 #include "llvm/Support/ItaniumManglingCanonicalizer.h"
66 #include "llvm/Support/MemoryBuffer.h"
67 
68 namespace llvm {
69 
70 class SymbolRemappingParseError : public ErrorInfo<SymbolRemappingParseError> {
71 public:
SymbolRemappingParseError(StringRef File,int64_t Line,Twine Message)72   SymbolRemappingParseError(StringRef File, int64_t Line, Twine Message)
73       : File(File), Line(Line), Message(Message.str()) {}
74 
log(llvm::raw_ostream & OS)75   void log(llvm::raw_ostream &OS) const override {
76     OS << File << ':' << Line << ": " << Message;
77   }
convertToErrorCode()78   std::error_code convertToErrorCode() const override {
79     return llvm::inconvertibleErrorCode();
80   }
81 
getFileName()82   StringRef getFileName() const { return File; }
getLineNum()83   int64_t getLineNum() const { return Line; }
getMessage()84   StringRef getMessage() const { return Message; }
85 
86   static char ID;
87 
88 private:
89   std::string File;
90   int64_t Line;
91   std::string Message;
92 };
93 
94 /// Reader for symbol remapping files.
95 ///
96 /// Remaps the symbol names in profile data to match those in the program
97 /// according to a set of rules specified in a given file.
98 class SymbolRemappingReader {
99 public:
100   /// Read remappings from the given buffer, which must live as long as
101   /// the remapper.
102   Error read(MemoryBuffer &B);
103 
104   /// A Key represents an equivalence class of symbol names.
105   using Key = uintptr_t;
106 
107   /// Construct a key for the given symbol, or return an existing one if an
108   /// equivalent name has already been inserted. The symbol name must live
109   /// as long as the remapper.
110   ///
111   /// The result will be Key() if the name cannot be remapped (typically
112   /// because it is not a valid mangled name).
insert(StringRef FunctionName)113   Key insert(StringRef FunctionName) {
114     return Canonicalizer.canonicalize(FunctionName);
115   }
116 
117   /// Map the given symbol name into the key for the corresponding equivalence
118   /// class.
119   ///
120   /// The result will typically be Key() if no equivalent symbol has been
121   /// inserted, but this is not guaranteed: a Key different from all keys ever
122   /// returned by \c insert may be returned instead.
lookup(StringRef FunctionName)123   Key lookup(StringRef FunctionName) {
124     return Canonicalizer.lookup(FunctionName);
125   }
126 
127 private:
128   ItaniumManglingCanonicalizer Canonicalizer;
129 };
130 
131 } // end namespace llvm
132 
133 #endif // LLVM_SUPPORT_SYMBOLREMAPPINGREADER_H
134