Podman user and nvidia-container-toolkit

This problem (to me it’s a problem) seems specific to openSUSE. From what I’ve read other hosts don’t have this happen.

Ultimately I need to run my podman container rootless and it needs to have a normal user in the container using gpu acceleration. But the normal user cannot use the gpu, I think, because of this… nobody:nogroup.

user@91585e6a257a:/$ ls -l /dev/nvidia*
crw-rw----+ 1 nobody nogroup 195, 0 Jan 26 20:12 /dev/nvidia0
crw-rw----+ 1 nobody nogroup 195, 255 Jan 26 20:12 /dev/nvidiactl
crw-rw----+ 1 nobody nogroup 195, 254 Jan 26 20:12 /dev/nvidia-modeset
crw-rw-rw-+ 1 nobody nogroup 237, 0 Jan 26 20:12 /dev/nvidia-uvm
crw-rw-rw-+ 1 nobody nogroup 237, 1 Jan 26 20:12 /dev/nvidia-uvm-tools

Thank you for any help!