libsanitizer/sanitizer_common/sanitizer_symbolizer_internal.h

   1 //===-- sanitizer_symbolizer_internal.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 // Header for internal classes and functions to be used by implementations of
  10 // symbolizers.
  11 //
  12 //===----------------------------------------------------------------------===//
  13 #ifndef SANITIZER_SYMBOLIZER_INTERNAL_H
  14 #define SANITIZER_SYMBOLIZER_INTERNAL_H
  15
  16 #include "sanitizer_symbolizer.h"
  17 #include "sanitizer_file.h"
  18 #include "sanitizer_vector.h"
  19
  20 namespace __sanitizer {
  21
  22 // Parsing helpers, 'str' is searched for delimiter(s) and a string or uptr
  23 // is extracted. When extracting a string, a newly allocated (using
  24 // InternalAlloc) and null-terminataed buffer is returned. They return a pointer
  25 // to the next characted after the found delimiter.
  26 const char *ExtractToken(const char *str, const char *delims, char **result);
  27 const char *ExtractInt(const char *str, const char *delims, int *result);
  28 const char *ExtractUptr(const char *str, const char *delims, uptr *result);
  29 const char *ExtractTokenUpToDelimiter(const char *str, const char *delimiter,
  30                                       char **result);
  31
  32 const char *DemangleSwiftAndCXX(const char *name);
  33
  34 // SymbolizerTool is an interface that is implemented by individual "tools"
  35 // that can perform symbolication (external llvm-symbolizer, libbacktrace,
  36 // Windows DbgHelp symbolizer, etc.).
  37 class SymbolizerTool {
  38  public:
  39   // The main |Symbolizer| class implements a "fallback chain" of symbolizer
  40   // tools. In a request to symbolize an address, if one tool returns false,
  41   // the next tool in the chain will be tried.
  42   SymbolizerTool *next;
  43
  44   SymbolizerTool() : next(nullptr) { }
  45
  46   // Can't declare pure virtual functions in sanitizer runtimes:
  47   // __cxa_pure_virtual might be unavailable.
  48
  49   // The |stack| parameter is inout. It is pre-filled with the address,
  50   // module base and module offset values and is to be used to construct
  51   // other stack frames.
  52   virtual bool SymbolizePC(uptr addr, SymbolizedStack *stack) {
  53     UNIMPLEMENTED();
  54   }
  55
  56   // The |info| parameter is inout. It is pre-filled with the module base
  57   // and module offset values.
  58   virtual bool SymbolizeData(uptr addr, DataInfo *info) {
  59     UNIMPLEMENTED();
  60   }
  61
  62   virtual bool SymbolizeFrame(uptr addr, FrameInfo *info) {
  63     return false;
  64   }
  65
  66   virtual void Flush() {}
  67
  68   // Return nullptr to fallback to the default platform-specific demangler.
  69   virtual const char *Demangle(const char *name) {
  70     return nullptr;
  71   }
  72 };
  73
  74 // SymbolizerProcess encapsulates communication between the tool and
  75 // external symbolizer program, running in a different subprocess.
  76 // SymbolizerProcess may not be used from two threads simultaneously.
  77 class SymbolizerProcess {
  78  public:
  79   explicit SymbolizerProcess(const char *path, bool use_posix_spawn = false);
  80   const char *SendCommand(const char *command);
  81
  82  protected:
  83   /// The maximum number of arguments required to invoke a tool process.
  84   static const unsigned kArgVMax = 6;
  85
  86   // Customizable by subclasses.
  87   virtual bool StartSymbolizerSubprocess();
  88   virtual bool ReadFromSymbolizer(char *buffer, uptr max_length);
  89
  90  private:
  91   virtual bool ReachedEndOfOutput(const char *buffer, uptr length) const {
  92     UNIMPLEMENTED();
  93   }
  94
  95   /// Fill in an argv array to invoke the child process.
  96   virtual void GetArgV(const char *path_to_binary,
  97                        const char *(&argv)[kArgVMax]) const {
  98     UNIMPLEMENTED();
  99   }
 100
 101   bool Restart();
 102   const char *SendCommandImpl(const char *command);
 103   bool WriteToSymbolizer(const char *buffer, uptr length);
 104
 105   const char *path_;
 106   fd_t input_fd_;
 107   fd_t output_fd_;
 108
 109   static const uptr kBufferSize = 16 * 1024;
 110   char buffer_[kBufferSize];
 111
 112   static const uptr kMaxTimesRestarted = 5;
 113   static const int kSymbolizerStartupTimeMillis = 10;
 114   uptr times_restarted_;
 115   bool failed_to_start_;
 116   bool reported_invalid_path_;
 117   bool use_posix_spawn_;
 118 };
 119
 120 class LLVMSymbolizerProcess;
 121
 122 // This tool invokes llvm-symbolizer in a subprocess. It should be as portable
 123 // as the llvm-symbolizer tool is.
 124 class LLVMSymbolizer : public SymbolizerTool {
 125  public:
 126   explicit LLVMSymbolizer(const char *path, LowLevelAllocator *allocator);
 127
 128   bool SymbolizePC(uptr addr, SymbolizedStack *stack) override;
 129   bool SymbolizeData(uptr addr, DataInfo *info) override;
 130   bool SymbolizeFrame(uptr addr, FrameInfo *info) override;
 131
 132  private:
 133   const char *FormatAndSendCommand(const char *command_prefix,
 134                                    const char *module_name, uptr module_offset,
 135                                    ModuleArch arch);
 136
 137   LLVMSymbolizerProcess *symbolizer_process_;
 138   static const uptr kBufferSize = 16 * 1024;
 139   char buffer_[kBufferSize];
 140 };
 141
 142 // Parses one or more two-line strings in the following format:
 143 //   <function_name>
 144 //   <file_name>:<line_number>[:<column_number>]
 145 // Used by LLVMSymbolizer, Addr2LinePool and InternalSymbolizer, since all of
 146 // them use the same output format.  Returns true if any useful debug
 147 // information was found.
 148 void ParseSymbolizePCOutput(const char *str, SymbolizedStack *res);
 149
 150 // Parses a two-line string in the following format:
 151 //   <symbol_name>
 152 //   <start_address> <size>
 153 // Used by LLVMSymbolizer and InternalSymbolizer.
 154 void ParseSymbolizeDataOutput(const char *str, DataInfo *info);
 155
 156 }  // namespace __sanitizer
 157
 158 #endif  // SANITIZER_SYMBOLIZER_INTERNAL_H