You are not logged in.
I receive these messages in the journal
journalctl -b | grep nvidia
Apr 12 08:15:09 I-NET kernel: nvidia: module license 'NVIDIA' taints kernel.
Apr 12 08:15:10 I-NET kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 238
Apr 12 08:15:10 I-NET kernel: nvidia 0000:01:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem
Apr 12 08:15:10 I-NET (udev-worker)[371]: nvidia: Process '/usr/bin/bash -c '/usr/bin/mknod -Z -m 666 /dev/nvidiactl c $(grep nvidia-frontend /proc/devices | cut -d \ -f 1) 255'' failed with exit code 1.
Apr 12 08:15:10 I-NET kernel: nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
Apr 12 08:15:10 I-NET (udev-worker)[371]: nvidia: Process '/usr/bin/bash -c 'for i in $(cat /proc/driver/nvidia/gpus/*/information | grep Minor | cut -d \ -f 4); do /usr/bin/mknod -Z -m 666 /dev/nvidia${i} c $(grep nvidia-frontend /proc/devices | cut -d \ -f 1) ${i}; done'' failed with exit code 1.
Apr 12 08:15:10 I-NET kernel: nvidia-uvm: Loaded the UVM driver, major device number 235.
Apr 12 08:15:10 I-NET systemd-modules-load[326]: Inserted module 'nvidia_uvm'
Apr 12 08:15:10 I-NET kernel: nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 530.41.03 Thu Mar 16 19:23:04 UTC 2023
Apr 12 08:15:10 I-NET kernel: [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
Apr 12 08:15:10 I-NET kernel: [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
What is it?
What do I have to do?
Offline
Nothing that's some script in the nvidia driver that fails when some device nodes aren't present but it doesn't appear to have a negative effect/from context might be what the logic uses to determine it needs to load nvidia_uvm.
Offline
I have removed the "60-nvidia.rules" and the error has disappeared.
Everything is running...
Offline
Which will break power management for NVIDIA PRIME/Optimus cases and will break Wayland for the devices that it was able to create and that matter in that context. And apparently potentially Vulkan: https://bugs.archlinux.org/task/72306
Last edited by V1del (2023-04-12 09:14:59)
Offline
I do not use Nvidia Prime and Wayland.
I do not know what else to do!?
Offline
Well use it that way, but be weary if you notice programs not working that you willing and intentionally messed with packaged files for artifical reasons.
You could also just drop the two rules that had an error but these aren't inherently problematic and you should just not bother with those that much. If it's just about silencing the logs but not logically breaking what the udev rule is trying to do you can add an || exit 0 to the end of the bash commands. Maybe actually check what the commands try to find by checking the ouptut you get from
grep nvidia-frontend /proc/devices | cut -d \ -f 1
for example.
Offline
I will continue to observe the whole thing.
At the moment I have no idea.
Thank you!
Offline