Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Build] Failed to compile against TensoRT 10.5 #22387

Closed
mc-nv opened this issue Oct 10, 2024 · 4 comments
Closed

[Build] Failed to compile against TensoRT 10.5 #22387

mc-nv opened this issue Oct 10, 2024 · 4 comments
Labels
build build issues; typically submitted using template ep:TensorRT issues related to TensorRT execution provider

Comments

@mc-nv
Copy link
Contributor

mc-nv commented Oct 10, 2024

Describe the issue

Recent build of ONNX Runtime 1.19.2 against TensorRT 10.5.0.18 was unsuccessful, due to an API deprecation.

BUILDTOOLS_VERSION 17.12.35323.107
CMAKE_VERSION 3.30.0
CUDA_VERSION 12.5.1
CUDNN_VERSION 9.5.0.50
PYTHON_VERSION 3.10.11
TENSORRT_VERSION 10.5.0.18
VCPGK_VERSION 2024.03.19

Here is a documentation reference:
https://docs.nvidia.com/deeplearning/tensorrt/release-notes/index.html
https://docs.nvidia.com/deeplearning/tensorrt/api/c_api/classnvinfer1_1_1_i_builder.html#a5758c218e1d9f36bea5b9f7bc63b9ca9

Urgency

Could you please prioritize it for next release.

Target platform

Windows, haven't test against Linux

Build script

${env:ONNXRUNTIME_TAG}="v1.19.2"
${env:CUDAARCHS}= "60;61;70;75;80;86;90"
git clone --recursive --single-branch -b ${env:ONNXRUNTIME_TAG} https://github.com/microsoft/onnxruntime.git onnxruntime
onnxruntime\tools\ci_build\build.py --cmake_generator "Visual Studio 17 2022" --config Release --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=${env:CUDAARCHS}" --skip_submodule_sync --parallel --build_shared_lib --compile_no_warning_as_error --skip_tests --update --build --build_dir /workspace/build --use_cuda --cuda_home "$env:CUDA_PATH" --cudnn_home "$env:CUDA_PATH" --use_tensorrt --tensorrt_home "/tensorrt"

Error / output


Error ouptut:

C:\workspace\onnxruntime\onnxruntime\core\providers\tensorrt\tensorrt_execution_provider.cc(3547,50): error C4996: 'nvinfer1::IBuilder::platformHasFastInt8': was declared deprecated [C:\workspace\build\Release\onnxruntime_p
roviders_tensorrt.vcxproj]
C:\workspace\onnxruntime\onnxruntime\core\providers\tensorrt\tensorrt_execution_provider.cc(2910,23): error C4996: 'nvinfer1::IBuilder::platformHasFastFp16': was declared deprecated [C:\workspace\build\Release\onnxruntime_p
roviders_tensorrt.vcxproj]
C:\workspace\onnxruntime\onnxruntime\core\providers\tensorrt\tensorrt_execution_provider.cc(2917,23): error C4996: 'nvinfer1::IBuilder::platformHasFastInt8': was declared deprecated [C:\workspace\build\Release\onnxruntime_p
roviders_tensorrt.vcxproj]
C:\workspace\onnxruntime\onnxruntime\core\providers\tensorrt\tensorrt_execution_provider.cc(3134,42): error C4996: 'nvinfer1::IBuilder::platformHasFastInt8': was declared deprecated [C:\workspace\build\Release\onnxruntime_p
roviders_tensorrt.vcxproj]

Visual Studio Version

17.12.35323.107

GCC / Compiler Version

No response

@mc-nv mc-nv added the build build issues; typically submitted using template label Oct 10, 2024
@snnn snnn added the ep:TensorRT issues related to TensorRT execution provider label Oct 10, 2024
@edgchen1
Copy link
Contributor

#22374 was just merged, can you try again with main?

@snnn
Copy link
Member

snnn commented Oct 22, 2024

@mc-nv , could you please help confirm if the problem is solved? We are preparing a new release. We hope the issue can be addressed in the release.

@mc-nv
Copy link
Contributor Author

mc-nv commented Oct 22, 2024

In our build scenario change #22374 has been cherry-picked into rel-1.19.2 and it help us to compile project.

@snnn
Copy link
Member

snnn commented Oct 22, 2024

Thanks!

@snnn snnn closed this as completed Oct 22, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
build build issues; typically submitted using template ep:TensorRT issues related to TensorRT execution provider
Projects
None yet
Development

No branches or pull requests

3 participants