Install cuda 11.7 on Jetson Orin NX

Hi all,

My hardware is the Jetson Orin NX (16GB), and I have installed the Petpack 5.1.2 using the SDK Manager. I want to update CUDA from version 11.4 to 11.7.

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/sbsa/cuda-ubuntu2004.pin
sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/11.7.0/local_installers/cuda-repo-ubuntu2004-11-7-local_11.7.0-515.43.04-1_arm64.deb
sudo dpkg -i cuda-repo-ubuntu2004-11-7-local_11.7.0-515.43.04-1_arm64.deb
sudo cp /var/cuda-repo-ubuntu2004-11-7-local/cuda-*-keyring.gpg /usr/share/keyrings/
sudo apt-get update
sudo apt-get -y install cuda

The last command gives these errors:

Preparing to unpack …/72-nvidia-prime_0.8.16~0.20.04.2_all.deb …
Unpacking nvidia-prime (0.8.16~0.20.04.2) …
Errors were encountered while processing:
/tmp/apt-dpkg-install-AI2eBd/13-libnvidia-compute-515_515.43.04-0ubuntu1_arm64.deb
/tmp/apt-dpkg-install-AI2eBd/17-libnvidia-gl-515_515.43.04-0ubuntu1_arm64.deb
/tmp/apt-dpkg-install-AI2eBd/24-nvidia-utils-515_515.43.04-0ubuntu1_arm64.deb

I tried to command “sudo apt --fix-broken install” :

Preparing to unpack …/nvidia-utils-515_515.43.04-0ubuntu1_arm64.deb …
Unpacking nvidia-utils-515 (515.43.04-0ubuntu1) …
dpkg: error processing archive /var/cuda-repo-ubuntu2004-11-7-local/./nvidia-utils-515_515.43.04-0ubuntu1_arm64.deb (–unpack):
trying to overwrite ‘/usr/bin/nvidia-xconfig’, which is also in package nvidia-l4t-x11 35.4.1-20230801124926
dpkg-deb: error: paste subprocess was killed by signal (Broken pipe)
Errors were encountered while processing:
/var/cuda-repo-ubuntu2004-11-7-local/./libnvidia-compute-515_515.43.04-0ubuntu1_arm64.deb
/var/cuda-repo-ubuntu2004-11-7-local/./libnvidia-gl-515_515.43.04-0ubuntu1_arm64.deb
/var/cuda-repo-ubuntu2004-11-7-local/./nvidia-utils-515_515.43.04-0ubuntu1_arm64.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)

How can I address these issues and install CUDA 11.7?
Thank you

Hi,
Manually upgrading CUDA to another version is not supported. Please use the version installed through SDKManager. This is the setup tested in our SQA coverage.

1 Like

Thanks for your answer

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.