Speed up SizeOfType

test=develop

paddle/fluid/framework/data_type.cc

@@ -14,6 +14,7 @@
 
 #include "paddle/fluid/framework/data_type.h"
 #include <stdint.h>
+#include <map>
 #include <string>
 #include <unordered_map>
 
@@ -23,10 +24,10 @@ namespace paddle {
 namespace framework {
 
 struct DataTypeMap {
-  std::unordered_map<std::type_index, proto::VarType::Type> cpp_to_proto_;
+  std::map<const char*, proto::VarType::Type> cpp_to_proto_;
   std::unordered_map<int, std::type_index> proto_to_cpp_;
   std::unordered_map<int, std::string> proto_to_str_;
-  std::unordered_map<std::type_index, size_t> cpp_to_size_;
+  std::map<const char* /*name pointer*/, size_t> cpp_to_size_;
 };
 
 static DataTypeMap* InitDataTypeMap();
@@ -43,9 +44,9 @@ static inline void RegisterType(DataTypeMap* map,
                                 proto::VarType::Type proto_type,
                                 const std::string& name) {
   map->proto_to_cpp_.emplace(static_cast<int>(proto_type), typeid(T));
-  map->cpp_to_proto_.emplace(typeid(T), proto_type);
+  map->cpp_to_proto_.emplace(typeid(T).name(), proto_type);
   map->proto_to_str_.emplace(static_cast<int>(proto_type), name);
-  map->cpp_to_size_.emplace(typeid(T), sizeof(T));
+  map->cpp_to_size_.emplace(typeid(T).name(), sizeof(T));
 }
 
 static DataTypeMap* InitDataTypeMap() {
@@ -71,7 +72,7 @@ static DataTypeMap* InitDataTypeMap() {
 }
 
 proto::VarType::Type ToDataType(std::type_index type) {
-  auto it = gDataTypeMap().cpp_to_proto_.find(type);
+  auto it = gDataTypeMap().cpp_to_proto_.find(type.name());
   if (it != gDataTypeMap().cpp_to_proto_.end()) {
     return it->second;
   }
@@ -97,8 +98,8 @@ std::string DataTypeToString(const proto::VarType::Type type) {
 }
 
 size_t SizeOfType(std::type_index type) {
-  auto it = gDataTypeMap().cpp_to_size_.find(type);
-  if (it != gDataTypeMap().cpp_to_size_.end()) {
+  auto it = gDataTypeMap().cpp_to_size_.find(type.name());
+  if (LIKELY(it != gDataTypeMap().cpp_to_size_.end())) {
     return it->second;
   }
   PADDLE_THROW("Not support %s as tensor type", type.name());

paddle/fluid/framework/parallel_executor.cc

@@ -31,9 +31,9 @@ limitations under the License. */
 #include "paddle/fluid/platform/profiler.h"
 
 #ifdef WITH_GPERFTOOLS
-#include "google/gperftools.h"
+#include "gperftools/profiler.h"
 #endif
-DEFINE_string(PEProfileFName, "",
+DEFINE_string(pe_profile_fname, "",
               "Profiler filename for PE, which generated by gperftools."
               "Only valid when compiled `WITH_PRIFILER=ON`. Empty if disable.");
 
@@ -45,14 +45,14 @@ class ParallelExecutorPrivate {
  public:
   explicit ParallelExecutorPrivate(const std::vector<platform::Place> &places)
       : places_(places) {
-    if (!FLAGS_PEProfileFName.empty()) {
+    if (!FLAGS_pe_profile_fname.empty()) {
       std::call_once(gProfileOnce, [] {
 #ifdef WITH_GPERFTOOLS
-        ProfilerStart(FLAGS_PEProfileFName.c_str());
+        ProfilerStart(FLAGS_pe_profile_fname.c_str());
         gProfileStarted = true;
 #else
         LOG(WARNING) << "Paddle is not compiled with gperftools. "
-                        "FLAGS_PEProfileFName will be ignored";
+                        "FLAGS_pe_profile_fname will be ignored";
 #endif
       });
     }

paddle/fluid/platform/enforce.h

@@ -62,45 +62,54 @@ inline std::string demangle(std::string name) { return name; }
 #endif
 
 struct EnforceNotMet : public std::exception {
-  std::exception_ptr exp_;
   std::string err_str_;
-  EnforceNotMet(std::exception_ptr e, const char* f, int l) : exp_(e) {
-    static constexpr int TRACE_STACK_LIMIT = 100;
+  EnforceNotMet(std::exception_ptr e, const char* f, int l) {
     try {
-      std::rethrow_exception(exp_);
-    } catch (const std::exception& exp) {
-      std::ostringstream sout;
+      std::rethrow_exception(e);
+    } catch (std::exception& e) {
+      Init(e.what(), f, l);
+    }
+  }
 
-      sout << string::Sprintf("%s at [%s:%d]", exp.what(), f, l) << std::endl;
-      sout << "PaddlePaddle Call Stacks: " << std::endl;
+  template <typename... ARGS>
+  EnforceNotMet(const char* f, int l, ARGS... args) {
+    Init(string::Sprintf(args...), f, l);
+  }
+
+  const char* what() const noexcept override { return err_str_.c_str(); }
+
+ private:
+  template <typename StrType>
+  inline void Init(StrType what, const char* f, int l) {
+    static constexpr int TRACE_STACK_LIMIT = 100;
+    std::ostringstream sout;
+
+    sout << string::Sprintf("%s at [%s:%d]", what, f, l) << std::endl;
+    sout << "PaddlePaddle Call Stacks: " << std::endl;
 #if !defined(_WIN32)
-      void* call_stack[TRACE_STACK_LIMIT];
-      auto size = backtrace(call_stack, TRACE_STACK_LIMIT);
-      auto symbols = backtrace_symbols(call_stack, size);
-
-      Dl_info info;
-      for (int i = 0; i < size; ++i) {
-        if (dladdr(call_stack[i], &info) && info.dli_sname) {
-          auto demangled = demangle(info.dli_sname);
-          auto addr_offset = static_cast<char*>(call_stack[i]) -
-                             static_cast<char*>(info.dli_saddr);
-          sout << string::Sprintf("%-3d %*0p %s + %zd\n", i,
-                                  2 + sizeof(void*) * 2, call_stack[i],
-                                  demangled, addr_offset);
-        } else {
-          sout << string::Sprintf("%-3d %*0p\n", i, 2 + sizeof(void*) * 2,
-                                  call_stack[i]);
-        }
+    void* call_stack[TRACE_STACK_LIMIT];
+    auto size = backtrace(call_stack, TRACE_STACK_LIMIT);
+    auto symbols = backtrace_symbols(call_stack, size);
+    Dl_info info;
+    for (int i = 0; i < size; ++i) {
+      if (dladdr(call_stack[i], &info) && info.dli_sname) {
+        auto demangled = demangle(info.dli_sname);
+        auto addr_offset = static_cast<char*>(call_stack[i]) -
+                           static_cast<char*>(info.dli_saddr);
+        sout << string::Sprintf("%-3d %*0p %s + %zd\n", i,
+                                2 + sizeof(void*) * 2, call_stack[i], demangled,
+                                addr_offset);
+      } else {
+        sout << string::Sprintf("%-3d %*0p\n", i, 2 + sizeof(void*) * 2,
+                                call_stack[i]);
       }
-      free(symbols);
+    }
+    free(symbols);
 #else
-      sout << "Windows not support stack backtrace yet.";
+    sout << "Windows not support stack backtrace yet.";
 #endif
-      err_str_ = sout.str();
-    }
+    err_str_ = sout.str();
   }
-
-  const char* what() const noexcept { return err_str_.c_str(); }
 };
 
 struct EOFException : public std::exception {
@@ -242,13 +251,8 @@ inline void throw_on_error(T e) {
   throw_on_error(e, "");
 }
 
-#define PADDLE_THROW(...)                                              \
-  do {                                                                 \
-    throw ::paddle::platform::EnforceNotMet(                           \
-        std::make_exception_ptr(                                       \
-            std::runtime_error(paddle::string::Sprintf(__VA_ARGS__))), \
-        __FILE__, __LINE__);                                           \
-  } while (false)
+#define PADDLE_THROW(...) \
+  throw ::paddle::platform::EnforceNotMet(__FILE__, __LINE__, __VA_ARGS__)
 
 #ifndef REPLACE_ENFORCE_GLOG
 #define PADDLE_ENFORCE(...)                                             \

python/paddle/fluid/__init__.py

@@ -127,7 +127,8 @@ def __bootstrap__():
         'use_ngraph', 'initial_cpu_memory_in_mb', 'init_allocated_mem',
         'free_idle_memory', 'paddle_num_threads', "dist_threadpool_size",
         'eager_delete_tensor_gb', 'allocator_strategy',
-        'reader_queue_speed_test_mode', 'print_sub_graph_dir'
+        'reader_queue_speed_test_mode', 'print_sub_graph_dir',
+        'pe_profile_fname'
     ]
     if 'Darwin' not in sysstr:
         read_env_flags.append('use_pinned_memory')