How to install onnruntime successfully?
We build onnxruntime with experimental TensorRT support.
Compilation always fails when we use ./build.sh --config Release --update --build --parallel --use_tensorrt --cuda_home /usr/local/cuda --cudnn_home /usr/lib/aarch64-linux-gnu --tensorrt_home /usr/lib/aarch64-linux-gnu.
We're not sure where the problem is.Hi,
You can find the onnxruntime for Jetson below:
https://elinux.org/Jetson_Zoo#ONNX_Runtime
Thanks.
[ 70%] Linking CXX static library libonnxruntime_providers.a
[ 70%] Built target onnxruntime_providers
Makefile:181: recipe for target ‘all’ failed
make: *** [all] Error 2
Traceback (most recent call last):
File “/home/nvidia/onnxruntime/tools/ci_build/build.py”, line 2114, in
sys.exit(main())
File “/home/nvidia/onnxruntime/tools/ci_build/build.py”, line 2040, in main
build_targets(args, cmake_path, build_dir, configs, num_parallel_jobs, args.target)
File “/home/nvidia/onnxruntime/tools/ci_build/build.py”, line 1062, in build_targets
run_subprocess(cmd_args, env=env)
File “/home/nvidia/onnxruntime/tools/ci_build/build.py”, line 578, in run_subprocess
return run(*args, cwd=cwd, capture_stdout=capture_stdout, shell=shell, env=my_env)
File “/home/nvidia/onnxruntime/tools/python/util/run.py”, line 44, in run
env=env, shell=shell)
File “/home/nvidia/miniforge-pypy3/lib-python/3/subprocess.py”, line 512, in run
output=stdout, stderr=stderr)
subprocess.CalledProcessError: Command ‘[’/usr/bin/cmake’, ‘–build’, ‘/home/nvidia/onnxruntime/build/Linux/Release’, ‘–config’, ‘Release’, ‘–’, ‘-j4’]’ returned non-zero exit status 2.
Hi,
Since TensorRT is enabled, please also select the CUDA to see if it works.
Thanks.
