Link Search Menu Expand Document

Add a new Execution Provider to ONNX Runtime

Contents

Create the Execution Provider

  1. Create a folder under onnxruntime/core/providers
  2. Create a folder under include/onnxruntime/core/providers, it should has the same name as the first step.
  3. Create a new class, which must inherit from IExecutionProvider. The source code should be put in ‘onnxruntime/core/providers/[your_provider_name]’
  4. Create a new header file under include/onnxruntime/core/providers/[your_provider_name]. The file should provide one function for creating an OrtProviderFactoryInterface. You may use ‘include/onnxruntime/core/providers/cpu/cpu_provider_factory.h’ as a template. You don’t need to provide a function for creating MemoryInfo.
  5. Put a symbols.txt under ‘onnxruntime/core/providers/[your_provider_name]’. The file should contain all the function names that would be exported from you provider. Usually, just a single function for creating provider factory is enough.
  6. Add your provider in onnxruntime_providers.cmake. Build it as a static lib.
  7. Add one line in cmake/onnxruntime.cmake, to the ‘target_link_libraries’ function call. Put your provider there.

Examples:

Use the Execution Provider

  1. Create a factory for that provider, by using the c function you exported in ‘symbols.txt’
  2. Put the provider factory into session options
  3. Create session from that session option

Example:

  OrtEnv* env;
  OrtInitialize(ORT_LOGGING_LEVEL_WARNING, "test", &env)
  OrtSessionOptions* session_option = OrtCreateSessionOptions();
  OrtProviderFactoryInterface** factory;
  OrtCreateCUDAExecutionProviderFactory(0, &factory);
  OrtSessionOptionsAppendExecutionProvider(session_option, factory);
  OrtReleaseObject(factory);
  OrtCreateSession(env, model_path, session_option, &session);

Test the Execution Provider

To ease the testing of your execution provider, you can add a new case for it to the onnx_test_runner command, do this by adding it to onnxruntime/test/onnx/main.cc file, following the pattern for other existing providers.

Once you have this in place, you can run the onnx_test_runner, like this:

$ cd build/PLATFORM/CONFIGURATION
$ ./onnx_test_runner -e YOUR_BACKEND ./testdata/ort_minimal_e2e_test_data/
$ ./onnx_test_runner -e YOUR_BACKEND ./testdata/gemm_activation_fusion/