Fedora 38 (KDE Plasma)
RTX 3070 Ti
Drivers installed via RPMFusion
As per the subject. Since installing Fedora 38 (KDE) it has become increasingly difficult to boot to a desktop environment as the majority of the time I will just get a black screen instead of the login prompt. If you see the output from /var/log/messages below you can see that a number of checks fail during boot resulting in an attempt to fallback to nouveau.
I suspect this must be some sort of race condition because multiple restart attempts will eventually boot successfully and everything works perfectly. Not sure if it is coincidence but I have discovered that if I enable the grub menu to always show, wait 3 seconds and then select the first boot option I can often manage to get to the desktop environment in maybe 1 in 5 times.
The following is shown when the following command is run:
sudo cat /var/log/messages | grep nvidia
Aug 13 11:59:49 fedora systemd[1]: Starting nvidia-powerd.service - nvidia-powerd service…
Aug 13 11:59:49 fedora /usr/bin/nvidia-powerd[1041]: nvidia-powerd version:1.0(build 1)
Aug 13 11:59:49 fedora kernel: nvidia: loading out-of-tree module taints kernel.
Aug 13 11:59:49 fedora kernel: nvidia: module license ‘NVIDIA’ taints kernel.
Aug 13 11:59:49 fedora kernel: nvidia: module verification failed: signature and/or required key missing - tai
nting kernel
Aug 13 11:59:49 fedora kernel: nvidia: module license taints kernel.
Aug 13 11:59:49 fedora kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 511
Aug 13 11:59:49 fedora kernel: nvidia 0000:01:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=non
e:owns=none
Aug 13 11:59:50 fedora (udev-worker)[740]: nvidia: Process ‘/usr/bin/bash -c ‘/usr/bin/mknod -Z -m 666 /dev/nv
idiactl c $(grep nvidia-frontend /proc/devices | cut -d \ -f 1) 255’’ failed with exit code 1.
Aug 13 11:59:50 fedora (udev-worker)[780]: nvidia: Process ‘/usr/bin/bash -c ‘for i in $(cat /proc/driver/nvid
ia/gpus/*/information | grep Minor | cut -d \ -f 4); do /usr/bin/mknod -Z -m 666 /dev/nvidia${i} c $(grep nvi
dia-frontend /proc/devices | cut -d \ -f 1) ${i}; done’’ failed with exit code 1.
Aug 13 11:59:50 fedora kernel: nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietar
y module nvidia, inheriting taint.
Aug 13 11:59:50 fedora kernel: nvidia-uvm: Loaded the UVM driver, major device number 509.
Aug 13 11:59:50 fedora kernel: nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 5
35.86.05 Fri Jul 14 20:20:58 UTC 2023
Aug 13 11:59:50 fedora kernel: [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
Aug 13 11:59:51 fedora /usr/bin/nvidia-powerd[1041]: No matching GPU found
Aug 13 11:59:51 fedora /usr/bin/nvidia-powerd[1041]: Failed to initialize RM Client
Aug 13 11:59:51 fedora systemd[1]: nvidia-powerd.service: Main process exited, code=exited, status=1/FAILURE
Aug 13 11:59:51 fedora systemd[1]: nvidia-powerd.service: Failed with result ‘exit-code’.
Aug 13 11:59:51 fedora systemd[1]: Failed to start nvidia-powerd.service - nvidia-powerd service.
Aug 13 11:59:51 fedora systemd[1]: nvidia-powerd.service: Consumed 1.114s CPU time.
Aug 13 11:59:51 fedora audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system
_r:init_t:s0 msg=‘unit=nvidia-powerd comm=“systemd” exe=“/usr/lib/systemd/systemd” hostname=? addr=? terminal=
? res=failed’
Aug 13 11:59:51 fedora kernel: [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
Aug 13 11:59:51 fedora systemd[1]: nvidia-fallback.service - Fallback to nouveau as nvidia did not load was sk
ipped because of an unmet condition check (ConditionPathExists=!/sys/module/nvidia).
As you can see, a number of Process calls fail with exit code 1 during boot. I have never had any issues with Fedora 37 (KDE).
Bug report attached
nvidia-bug-report.log.gz (491.9 KB)