nvidia gpu device acess in containers

[viprob-ai]

Hi I am facing an issue in setting up the gpu inside container.

OS: ubuntu 22.04 podman versions : 3.4.4 nvidia-ctk version : 1.17

Followed all the instruction as per the below link but still no success : https://github.com/NVIDIA/nvidia-container-runtime/issues/85#issuecomment-1772741424

• nvidia-ctk cdi generate --output=/etc/cdi/nvidia.yaml

• podman run --rm --device nvidia.com/gpu=all --security-opt=label=disable ubuntu nvidia-smi -L

Error: stat nvidia.com/gpu=all: no such file or directory

[ericthomasca]

Did you try running sudo nvidia-ctk cdi generate --output=/etc/cdi/nvidia.yaml?

[viprob-ai]

Yes after trying many things with no success finally I upgraded the Podman version to 5.0 and everything works.