Hi, our custom board want to install cuda、cudnn & tensorRT and we don’t have SDKmanager to do it so we want to install them by deb.
Our version is R32.7.3 and there is a website provided deb file but there is too much file that we don’t know which file to download
https://repo.download.nvidia.cn/jetson/
could you please tell me the right file need to download?
Thanks!
Hi,
Do you have an x86 machine available?
If yes, please download the packages (which are packed into a big deb) with SDKmanager and copy them to the Xavier.
Thanks.
But I only need CUAD CUDNN & tensorRT, because our storge is limited, I can see only can download all jetpack tool in SDKmanager
csp@csp:/NVME/Jetson/pdpb_test/jetpack$ ls
cuda-repo-cross-aarch64-ubuntu1804-10-2-local_10.2.460-1_all.deb nvidia-container-csv-tensorrt_8.2_arm64.deb
cuda-repo-l4t-10-2-local_10.2.460-1_arm64.deb nvidia-container-csv-visionworks_1.6.0.501_arm64.deb
cuda-repo-ubuntu1804-10-2-local_10.2.460-450.115-1_amd64.deb nvidia-container-runtime_3.7.0-1_all.deb
graphsurgeon-tf_8.2.1-1+cuda10.2_arm64.deb nvidia-container-toolkit_1.7.0-1_arm64.deb
libcudnn8_8.2.1.32-1+cuda10.2_arm64.deb nvidia-docker2_2.8.0-1_all.deb
libcudnn8-dev_8.2.1.32-1+cuda10.2_arm64.deb nvidia-l4t-jetson-multimedia-api_32.7.3-20221122092958_arm64.deb
libcudnn8-samples_8.2.1.32-1+cuda10.2_arm64.deb NVIDIA_Nsight_Graphics_L4T_Public_2021.2.21249.deb
libnvidia-container0_0.10.0+jetpack_arm64.deb NVIDIA_VisionWorks_References.zip
libnvidia-container1_1.7.0-1_arm64.deb OpenCV-4.1.1-2-gd5a58aa75-aarch64-dev.deb
libnvidia-container-tools_1.7.0-1_arm64.deb OpenCV-4.1.1-2-gd5a58aa75-aarch64-libs.deb
libnvinfer8_8.2.1-1+cuda10.2_arm64.deb OpenCV-4.1.1-2-gd5a58aa75-aarch64-licenses.deb
libnvinfer-bin_8.2.1-1+cuda10.2_arm64.deb OpenCV-4.1.1-2-gd5a58aa75-aarch64-python.deb
libnvinfer-dev_8.2.1-1+cuda10.2_arm64.deb OpenCV-4.1.1-2-gd5a58aa75-aarch64-samples.deb
libnvinfer-doc_8.2.1-1+cuda10.2_all.deb python3-libnvinfer_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-plugin8_8.2.1-1+cuda10.2_arm64.deb python3-libnvinfer-dev_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-plugin-dev_8.2.1-1+cuda10.2_arm64.deb sdkml3_jetpack_463.json
libnvinfer-samples_8.2.1-1+cuda10.2_all.deb tensorrt_8.2.1.9-1+cuda10.2_arm64.deb
libnvonnxparsers8_8.2.1-1+cuda10.2_arm64.deb uff-converter-tf_8.2.1-1+cuda10.2_arm64.deb
libnvonnxparsers-dev_8.2.1-1+cuda10.2_arm64.deb vpi-cross-aarch64-l4t-1.2.3-cuda10-x86_64-linux.deb
libnvparsers8_8.2.1-1+cuda10.2_arm64.deb vpi-demos-1.2.3-aarch64-l4t.deb
libnvparsers-dev_8.2.1-1+cuda10.2_arm64.deb vpi-demos-1.2.3-cuda10-x86_64-linux-ubuntu1804.deb
libvisionworks-repo_1.6.0.501_amd64.deb vpi-dev-1.2.3-aarch64-l4t.deb
libvisionworks-repo_1.6.0.501_arm64.deb vpi-dev-1.2.3-cuda10-x86_64-linux.deb
libvisionworks-sfm-repo_0.90.4.501_amd64.deb vpi-lib-1.2.3-aarch64-l4t.deb
libvisionworks-sfm-repo_0.90.4.501_arm64.deb vpi-lib-1.2.3-cuda10-x86_64-linux.deb
libvisionworks-tracking-repo_0.88.2.501_amd64.deb vpi-python2.7-1.2.3-aarch64-l4t.deb
libvisionworks-tracking-repo_0.88.2.501_arm64.deb vpi-python2.7-1.2.3-cuda10-x86_64-linux.deb
NsightSystems-linux-public-2021.5.4.19-e642d4b.deb vpi-python3.6-1.2.3-aarch64-l4t.deb
nsight-systems-linux-tegra-public-2021.5.4.19-e642d4b_2021.5.4.19-1_arm64.deb vpi-python3.6-1.2.3-cuda10-x86_64-linux.deb
nvidia-container-csv-cuda_10.2.460-1_arm64.deb vpi-samples-1.2.3-aarch64-l4t.deb
nvidia-container-csv-cudnn_8.2.1.32-1+cuda10.2_arm64.deb vpi-samples-1.2.3-cuda10-x86_64-linux.deb
There are too much pack and I don’t know what is CUDA CUDNN & tensorRT need
Hi,
cuda-repo-l4t-10-2-local_10.2.460-1_arm64.deb
libcudnn8_8.2.1.32-1+cuda10.2_arm64.deb
libcudnn8-dev_8.2.1.32-1+cuda10.2_arm64.deb
libcudnn8-samples_8.2.1.32-1+cuda10.2_arm64.deb (optional)
libnvinfer8_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-dev_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-plugin8_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-plugin-dev_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-bin_8.2.1-1+cuda10.2_arm64.deb
libnvinfer-doc_8.2.1-1+cuda10.2_all.deb
tensorrt_8.2.1.9-1+cuda10.2_arm64.deb
libnvonnxparsers8_8.2.1-1+cuda10.2_arm64.deb
libnvonnxparsers-dev_8.2.1-1+cuda10.2_arm64.deb
libnvparsers8_8.2.1-1+cuda10.2_arm64.deb
libnvparsers-dev_8.2.1-1+cuda10.2_arm64.deb
uff-converter-tf_8.2.1-1+cuda10.2_arm64.deb
python3-libnvinfer_8.2.1-1+cuda10.2_arm64.deb
python3-libnvinfer-dev_8.2.1-1+cuda10.2_arm64.deb
Thanks.
Thanks. that help a lot!
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.