Closed morrolinux closed 3 years ago
What is your version of tensortt, cudnn and CUDA?
TensorRT-6.0.1.5 cudnn 7 CUDA 10.0
But I've got the feeling that the wrong version of tensorrt is picked up. How can I check that? I see no cmake message about it
Hi @morrolinux
check that:
cmake ..
. If it does not print it, delete the folder and redo it. Thanks @mive93 , that was it! I had unaligned versions of cuDNN and TensorRT
Sorry to resurrect closed issues, but I'm getting the exact same error, but don't have unaligned versions of cuDNN and TensorRT
Also Ubuntu 18.04. LD_LIBRARY_PATH includes the directories found by cmake.
Hi. @nightduck Have you solved this problem? I met the same issue...
dpkg -l | grep cudnn
ii libcudnn8 8.6.0.166-1+cuda11.4 arm64 cuDNN runtime libraries
ii libcudnn8-dev 8.6.0.166-1+cuda11.4 arm64 cuDNN development libraries and headers
ii libcudnn8-samples 8.6.0.166-1+cuda11.4 arm64 cuDNN samples
dpkg -l | grep nvinfer
ii libnvinfer-bin 8.5.2-1+cuda11.4 arm64 TensorRT binaries
ii libnvinfer-dev 8.5.2-1+cuda11.4 arm64 TensorRT development libraries and headers
ii libnvinfer-plugin-dev 8.5.2-1+cuda11.4 arm64 TensorRT plugin libraries
ii libnvinfer-plugin8 8.5.2-1+cuda11.4 arm64 TensorRT plugin libraries
ii libnvinfer-samples 8.5.2-1+cuda11.4 all TensorRT samples
ii libnvinfer8 8.5.2-1+cuda11.4 arm64 TensorRT runtime libraries
ii python3-libnvinfer 8.5.2-1+cuda11.4 arm64 Python 3 bindings for TensorRT
ii python3-libnvinfer-dev 8.5.2-1+cuda11.4 arm64 Python 3 development package for TensorRT
nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Sun_Oct_23_22:16:07_PDT_2022
Cuda compilation tools, release 11.4, V11.4.315
Build cuda_11.4.r11.4/compiler.31964100_0
but Ubuntu 20.04
@JaneLin-0409 I was running in a docker container. And after software updates in the background on the host machine, docker was no longer able to use CUDA drivers. It was solved with a reboot of the host machine.
It's also important to make sure the version of libcudnn on your docker machine is compatible with the Nvidia drivers on your host machine. CUDA 11.4 requires driver 450 or higher.
Hi, I'm running Ubuntu 18.04 and installed the dependencies as stated in the readme. The
cmake ..
configuration went fine, butmake
fails with: Any idea about what could be wrong? PS: I've got multiple versions of CUDA, CuDNN, TensorRT etc and I'm (supposedly) pointing to the right paths with$LD_LIBRARY_PATH
.