[Windows] Use llvm-symbolizer before using dbghelp

Summary:
llvm-symbolizer understands both PDBs and DWARF, so it's a better bet if
it's available. It prints out the function parameter types and column
numbers, so I needed to churn the expected test output a bit.

This makes most of the llvm-symbolizer subprocessing code
target-independent. Pipes on all platforms use fd_t, and we can use the
portable ReadFromFile / WriteToFile wrappers in symbolizer_sanitizer.cc.
Only the pipe creation and process spawning is Windows-specific.

Please check that the libcdep layering is still correct. I don't know
how to reproduce the build configuration that relies on that.

Reviewers: samsonov

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D11791

llvm-svn: 244616

1 files changed, 49 insertions, 8 deletions

diff --git a/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_internal.h b/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_internal.h
index 350e2a785dd..ac9a5e4a766 100644
--- a/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_internal.h
+++ b/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_internal.h
@@ -74,21 +74,27 @@ class SymbolizerProcess {
   explicit SymbolizerProcess(const char *path, bool use_forkpty = false);
   const char *SendCommand(const char *command);
 
- private:
-  bool Restart();
-  const char *SendCommandImpl(const char *command);
-  bool ReadFromSymbolizer(char *buffer, uptr max_length);
-  bool WriteToSymbolizer(const char *buffer, uptr length);
-  bool StartSymbolizerSubprocess();
-
+ protected:
   virtual bool ReachedEndOfOutput(const char *buffer, uptr length) const {
     UNIMPLEMENTED();
   }
 
-  virtual void ExecuteWithDefaultArgs(const char *path_to_binary) const {
+  /// The maximum number of arguments required to invoke a tool process.
+  enum { kArgVMax = 6 };
+
+  /// Fill in an argv array to invoke the child process.
+  virtual void GetArgV(const char *path_to_binary,
+                       const char *(&argv)[kArgVMax]) const {
     UNIMPLEMENTED();
   }
 
+ private:
+  bool Restart();
+  const char *SendCommandImpl(const char *command);
+  bool ReadFromSymbolizer(char *buffer, uptr max_length);
+  bool WriteToSymbolizer(const char *buffer, uptr length);
+  bool StartSymbolizerSubprocess();
+
   const char *path_;
   fd_t input_fd_;
   fd_t output_fd_;
@@ -104,6 +110,41 @@ class SymbolizerProcess {
   bool use_forkpty_;
 };
 
+class LLVMSymbolizerProcess;
+
+// This tool invokes llvm-symbolizer in a subprocess. It should be as portable
+// as the llvm-symbolizer tool is.
+class LLVMSymbolizer : public SymbolizerTool {
+ public:
+  explicit LLVMSymbolizer(const char *path, LowLevelAllocator *allocator);
+
+  bool SymbolizePC(uptr addr, SymbolizedStack *stack) override;
+
+  bool SymbolizeData(uptr addr, DataInfo *info) override;
+
+ private:
+  const char *SendCommand(bool is_data, const char *module_name,
+                          uptr module_offset);
+
+  LLVMSymbolizerProcess *symbolizer_process_;
+  static const uptr kBufferSize = 16 * 1024;
+  char buffer_[kBufferSize];
+};
+
+// Parses one or more two-line strings in the following format:
+//   <function_name>
+//   <file_name>:<line_number>[:<column_number>]
+// Used by LLVMSymbolizer, Addr2LinePool and InternalSymbolizer, since all of
+// them use the same output format.  Returns true if any useful debug
+// information was found.
+void ParseSymbolizePCOutput(const char *str, SymbolizedStack *res);
+
+// Parses a two-line string in the following format:
+//   <symbol_name>
+//   <start_address> <size>
+// Used by LLVMSymbolizer and InternalSymbolizer.
+void ParseSymbolizeDataOutput(const char *str, DataInfo *info);
+
 }  // namespace __sanitizer
 
 #endif  // SANITIZER_SYMBOLIZER_INTERNAL_H