Jetson-nx use cuda10.2 in docker

[conggua641]

1.I download cuda10.2 (cudnn8.0) form jetpack
[图片]But wen I install the cuda
have a mistake
“”“”"
root@be28b43454e2:/home/NX_environment/sdkm_downloads# dpkg -i libcudnn8_8.0.0.180-1+cuda10.2_arm64.deb
(Reading database … 27245 files and directories currently installed.)
Preparing to unpack libcudnn8_8.0.0.180-1+cuda10.2_arm64.deb …
Unpacking libcudnn8 (8.0.0.180-1+cuda10.2) …
dpkg: error processing archive libcudnn8_8.0.0.180-1+cuda10.2_arm64.deb (–install):
unable to make backup link of ‘./usr/lib/aarch64-linux-gnu/libcudnn.so.8.0.0’ before installing new version: Invalid cross-device link
dpkg-deb: error: paste subprocess was killed by signal (Broken pipe)
Processing triggers for libc-bin (2.27-3ubuntu1.4) …
Errors were encountered while processing:
libcudnn8_8.0.0.180-1+cuda10.2_arm64.deb
root@be28b43454e2:/home/NX_environment/sdkm_downloads#
“”“”“”“”“”“”
2. Through these DEB packages, what commands should I use to install cuda10 2 and cudnn8 0
3.
At the same time, I also want to install tensorrt. What files do I need? Where can I download them

The most important thing is that I use jetson- NX

[AastaLLL]

Hi,

May I know which JetPack version you use?

The simplest way is to install the packages with the apt command directly.
Please try if below can meet your requirement.

$ sudo apt update
$ sudo apt install nvidia-tensorrt

Thanks.

[conggua641]

I want to install CUDA and cudnn tensorrt in the docker container of jetson -NX

[

不知道伙伴号 无法登录

conggua641@163.com

](https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=不知道伙伴号+无法登录&uid=conggua641%40163.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fsmc5cb86a293cb78cade6a8fefe3c47067.jpg&items=["conggua641%40163.com"])

签名由 网易邮箱大师 定制

On 4/26/2022 11:07，AastaLLL via NVIDIA Developer Forumsnvidia@discoursemail.com wrote：

[AastaLLL]

Hi,

Have you tried the below container? It already has the required libraries installed.

NVIDIA L4T TensorRT | NVIDIA NGC

For JetPack 4.6.1, you can use nvcr.io/nvidia/l4t-tensorrt:r8.2.1-runtime container:

$ sudo docker run -it --rm --net=host --runtime nvidia -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix nvcr.io/nvidia/l4t-tensorrt:r8.2.1-runtime

Thanks.

[conggua641]

Thank you very much for your reply. During your reply, we found a container with jetpack label to solve our problem

[

不知道伙伴号 无法登录

conggua641@163.com

](https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=不知道伙伴号+无法登录&uid=conggua641%40163.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fsmc5cb86a293cb78cade6a8fefe3c47067.jpg&items=["conggua641%40163.com"])

签名由 网易邮箱大师 定制

On 5/5/2022 14:02，AastaLLL via NVIDIA Developer Forumsnvidia@discoursemail.com wrote：