test

microsoft · Dec 13, 2023 · ab1d616 · ab1d616
1 parent 156d02c
commit ab1d616
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 21 deletions.
diff --git a/onnxruntime/core/providers/tensorrt/tensorrt_execution_provider.cc b/onnxruntime/core/providers/tensorrt/tensorrt_execution_provider.cc
@@ -2489,11 +2489,7 @@ common::Status TensorrtExecutionProvider::Compile(const std::vector<FusedNodeAnd
             &networks_[context->node_name], input_info_[context->node_name], output_info_[context->node_name],
             input_shape_ranges_[context->node_name], &tensorrt_mu_, &max_workspace_size_,
             trt_node_name_with_precision, runtime_.get(), profiles_[context->node_name], &max_ctx_mem_size_,
-            dynamic_range_map, !tactic_sources_.empty(), tactics,
-            // Below: class private members
-            fp16_enable_, int8_enable_, int8_calibration_cache_available_, dla_enable_,
-            dla_core_, engine_cache_enable_, cache_path_
-      };
+            dynamic_range_map, !tactic_sources_.empty(), tactics};
       *state = p.release();
       return 0;
     };
@@ -2547,7 +2543,7 @@ common::Status TensorrtExecutionProvider::Compile(const std::vector<FusedNodeAnd
       std::string compute_capability = GetComputeCapacity(prop);
 
       // Prepare cache name
-      const std::string cache_path = GetCachePath(trt_state->engine_cache_path, trt_state->trt_node_name_with_precision);
+      const std::string cache_path = GetCachePath(cache_path_, trt_state->trt_node_name_with_precision);
       const std::string engine_cache_path = cache_path + "_sm" + compute_capability + ".engine";
       const std::string encrypted_engine_cache_path = engine_cache_path + ".encrypted";
       const std::string profile_cache_path = cache_path + "_sm" + compute_capability + ".profile";
@@ -2557,7 +2553,7 @@ common::Status TensorrtExecutionProvider::Compile(const std::vector<FusedNodeAnd
       }
 
       // Load serialized engine
-      if (trt_state->engine_cache_enable && trt_engine == nullptr) {
+      if (engine_cache_enable_ && trt_engine == nullptr) {
         std::ifstream engine_file(engine_cache_path, std::ios::binary | std::ios::in);
         std::ifstream profile_file(profile_cache_path, std::ios::binary | std::ios::in);
         if (engine_file && !engine_decryption_enable_ && profile_file) {
@@ -2641,28 +2637,28 @@ common::Status TensorrtExecutionProvider::Compile(const std::vector<FusedNodeAnd
         }
 
         // Set INT8 Per Tensor Dynamic range
-        if (trt_state->int8_enable && trt_builder->platformHasFastInt8() && trt_state->int8_calibration_cache_available) {
+        if (int8_enable_ && trt_builder->platformHasFastInt8() && int8_calibration_cache_available_) {
           trt_config->setInt8Calibrator(nullptr);
           if (!SetDynamicRange(*trt_state->network->get(), trt_state->dynamic_range_map)) {
             return ORT_MAKE_STATUS(ONNXRUNTIME, EP_FAIL, "TensorRT EP failed to set INT8 dynamic range.");
           }
         }
 
         // Set precision
-        if (trt_state->fp16_enable && trt_state->int8_enable) {
+        if (fp16_enable_ && int8_enable_) {
           trt_config->setFlags(1U << static_cast<uint32_t>(nvinfer1::BuilderFlag::kFP16) | 1U << static_cast<uint32_t>(nvinfer1::BuilderFlag::kINT8));
-        } else if (trt_state->fp16_enable) {
+        } else if (fp16_enable_) {
           trt_config->setFlag(nvinfer1::BuilderFlag::kFP16);
-        } else if (trt_state->int8_enable) {
+        } else if (int8_enable_) {
           trt_config->setFlag(nvinfer1::BuilderFlag::kINT8);
         }
 
         // Set DLA (DLA can only run with FP16 or INT8)
-        if ((trt_state->fp16_enable || trt_state->int8_enable) && trt_state->dla_enable) {
-          LOGS_DEFAULT(VERBOSE) << "[TensorRT EP] use DLA core " << trt_state->dla_core;
+        if ((fp16_enable_ || int8_enable_) && dla_enable_) {
+          LOGS_DEFAULT(VERBOSE) << "[TensorRT EP] use DLA core " << dla_core_;
           trt_config->setFlag(nvinfer1::BuilderFlag::kGPU_FALLBACK);
           trt_config->setDefaultDeviceType(nvinfer1::DeviceType::kDLA);
-          trt_config->setDLACore(trt_state->dla_core);
+          trt_config->setDLACore(dla_core_);
         }
 
         // enable sparse weights
@@ -2737,7 +2733,7 @@ common::Status TensorrtExecutionProvider::Compile(const std::vector<FusedNodeAnd
           return ORT_MAKE_STATUS(ONNXRUNTIME, EP_FAIL, "TensorRT EP Failed to Build Engine.");
         }
         trt_engine = trt_state->engine->get();
-        if (trt_state->engine_cache_enable) {
+        if (engine_cache_enable_) {
           // Serialize engine profile
           SerializeProfileV2(profile_cache_path, shape_ranges);
           LOGS_DEFAULT(VERBOSE) << "[TensorRT EP] Serialized " + profile_cache_path;

diff --git a/onnxruntime/core/providers/tensorrt/tensorrt_execution_provider.h b/onnxruntime/core/providers/tensorrt/tensorrt_execution_provider.h
@@ -123,12 +123,7 @@ struct TensorrtFuncState {
   bool filter_tactic_sources = false;
   nvinfer1::TacticSources tactic_sources;
   // Below: class private members
-  bool fp16_enable = false;
-  bool int8_enable = false;
-  bool int8_calibration_cache_available = false;
-  bool dla_enable = false;
-  int dla_core = 0;
-  bool engine_cache_enable = false;
+
   std::string engine_cache_path;
 };