F40 KDE Nvidia Driver Problems

Ask Fedora
, , , , , , , ,
1

I’m Completely new to Linux. Nvidia seems to be unstable. I made the installation in X11 following this github. https://github.com/roworu/nvidia-fedora-secureboot

When I first log into launch X11 - It takes longer than usual to get in. After about a minute Plasma Crashes, or sometimes even a black screen. After force rebooting and launching back into X11 the next time and consecutive times after - System mostly seems to work fine.

I’ve only tested games in X11. Something like Helldivers works. SteamVR not so much. Headset not working. Xorg window not responding at times.

Switching to Wayland appears to not stutter anymore. Normal activity works on consecutive launches, However, Same problem returns to X11 when I switch back.

System:
  Kernel: 6.10.4-200.fc40.x86_64 arch: x86_64 bits: 64 compiler: gcc
    v: 2.41-37.fc40
  Desktop: KDE Plasma v: 6.1.4 Distro: Fedora Linux 40 (KDE Plasma)
Machine:
  Type: Desktop Mobo: Micro-Star model: MAG B650 TOMAHAWK WIFI (MS-7D75)
    v: 1.0 serial: <superuser required> UEFI: American Megatrends LLC. v: 1.J0
    date: 07/24/2024
CPU:
  Info: 12-core model: AMD Ryzen 9 7900X3D bits: 64 type: MT MCP MCM
    arch: Zen 4 rev: 2 cache: L1: 768 KiB L2: 12 MiB L3: 128 MiB
  Speed (MHz): avg: 1108 high: 4798 min/max: 545/5660 cores: 1: 545 2: 545
    3: 545 4: 545 5: 545 6: 545 7: 545 8: 3407 9: 4798 10: 4271 11: 545 12: 545
    13: 545 14: 545 15: 545 16: 545 17: 545 18: 545 19: 545 20: 545 21: 545
    22: 545 23: 3236 24: 545 bogomips: 211199
  Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm
Graphics:
  Device-1: NVIDIA AD103 [GeForce RTX 4070 Ti SUPER] vendor: ZOTAC
    driver: nvidia v: 555.58.02 arch: Lovelace bus-ID: 01:00.0
  Device-2: AMD Raphael vendor: Micro-Star MSI driver: amdgpu v: kernel
    arch: RDNA-2 bus-ID: 10:00.0 temp: 44.0 C
  Display: x11 server: X.Org v: 1.20.14 with: Xwayland v: 24.1.2 driver: X:
    loaded: modesetting,nvidia dri: radeonsi gpu: nvidia,nvidia-nvswitch
    resolution: 2560x1440~144Hz
  API: EGL v: 1.5 drivers: kms_swrast,nvidia,radeonsi,swrast platforms:
    active: gbm,x11,surfaceless,device inactive: wayland,device-2
  API: OpenGL v: 4.6.0 compat-v: 4.5 vendor: nvidia mesa v: 555.58.02
    glx-v: 1.4 direct-render: yes renderer: NVIDIA GeForce RTX 4070 Ti
    SUPER/PCIe/SSE2
  API: Vulkan v: 1.3.283 drivers: N/A surfaces: xcb,xlib devices: 3
Drives:
  Local Storage: total: 3.64 TiB used: 81.04 GiB (2.2%)
  ID-1: /dev/nvme0n1 vendor: Crucial model: CT4000P3PSSD8 size: 3.64 TiB
    temp: 35.9 C
Info:
  Memory: total: 32 GiB note: est. available: 30.42 GiB used: 6.24 GiB (20.5%)
  Processes: 560 Uptime: 54m Init: systemd target: graphical (5)
  Packages: N/A Compilers: gcc: 14.2.1 Shell: Bash v: 5.2.26 inxi: 3.3.34

If there is more relevant info I can provide. Please let me know and how.

2

A new install of Fedora 40 KDE does not have the X11 desktop enabled.
It needs to be installed by the user
dnf install plasma-workspace-x11

I see from the inxi output that you seem to have x11 running, so please be more specific as to what is the actual problem if that package is already installed.

3

Wayland would freeze up randomly and frequently. First thing I ran was

sudo dnf update
sudo dnf install plasma-workspace-x11 kwin-x11

Rebooted and selected x11. Running it currently. The Stuttering in Wayland seems to be gone after installing the Nvidia Driver though.

Could there be a problem between the Nvidia Driver and Xserver? I just ran this and got these EE and WW.

grep -E "(EE|WW)" ~/.local/share/xorg/Xorg.0.log
[   950.265] Current Operating System: Linux fedora 6.10.4-200.fc40.x86_64 #1 SMP PREEMPT_DYNAMIC Sun Aug 11 15:32:50 UTC 2024 x86_64
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   951.410] (WW) Warning, couldn't open module nouveau
[   951.410] (EE) Failed to load module "nouveau" (module does not exist, 0)
[   951.410] (WW) Warning, couldn't open module nv
[   951.410] (EE) Failed to load module "nv" (module does not exist, 0)
[   951.410] (WW) Warning, couldn't open module ati
[   951.410] (EE) Failed to load module "ati" (module does not exist, 0)
[   951.411] (WW) Warning, couldn't open module fbdev
[   951.411] (EE) Failed to load module "fbdev" (module does not exist, 0)
[   951.411] (WW) Warning, couldn't open module vesa
[   951.411] (EE) Failed to load module "vesa" (module does not exist, 0)
[   952.273] (WW) Falling back to old probe method for modesetting
[   952.273] (EE) open /dev/dri/card0: No such file or directory
[   952.273] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support
[   952.273] (WW) NVIDIA(0): Invalid SLI option: 'Auto'; using single GPU rendering.
[   952.281] (WW) NVIDIA(0): Failed to initialize Base Mosaic!  Reason: Only one GPU
[   952.282] (WW) NVIDIA(0):     detected.  Only one GPU will be used for this X screen.
[   952.682] (WW) NVIDIA: Failed to bind sideband socket to
[   952.682] (WW) NVIDIA:     '/var/run/nvidia-xdriver-ad7e231b' Permission denied
[   953.411] (II) Initializing extension MIT-SCREEN-SAVER

I also don’t seem to have /.var/log/Xorg.0.log
Don’t know if that’s normal.

4

normal. It is the one you showed above under .local.

5

I’m puzzled on what to do for the next steps of troubleshooting. Tested the same games on Steam in Wayland. Maybe not quite as good as on X11, but functional.

Switching back to X11 doesn’t seem to always make it crash initially like I thought. I really don’t know how or what to check for the cause of the crashing in X11.

6

Does the long load time occur every time you boot? Or is it only the first boot after and update?

If you do lsmod | grep nvidia and it returns several lines then the modules are loaded and functioning.

I don’t understand if you are getting crashes or not. there seems some ambiguity there in what you most recently said.

If it was crashing before you installed the drivers and not now, that is progress and may only require watching for crashes in the future.

7 
nvidia_drm            135168  39
nvidia_modeset       1617920  119 nvidia_drm
nvidia_uvm           6815744  0
nvidia              61018112  1202 nvidia_uvm,nvidia_modeset
video                  81920  2 amdgpu,nvidia_modeset

Sorry if I’m not clear. Before installing the Nvidia Driver. Wayland would stutter- X11 wouldn’t. Never had any crashes right after login.

After installing the Nvidia Driver. Neither Wayland would stutter. X11 might crash. After logging in, I’m greeted with a black screen and command line. Maybe 15-25 seconds, before the splash screen. A minute later and the Plasma Shell Crashes -Blackscreen.

I haven’t been able to replicate it again.

8

If you are not able to repeat the issue then it would appear to be solved.

9

Last night the issue appeared again after one of my restarts messing with SteamVR. Don’t know if there’s a correlation. It’s happened before without even touching Steam. No swapping between X11 and wayland on restart. Just X11.

10

Another instance. I have a better idea of the timeframe. Somewhere around Aug 21 15:50:00 to Aug 21 15:57:00 for login and reboot.

After login - had a blackscreen with some text 20 to 30 seconds give or take - splash screen - Had Firefox, System Monitor, maybe Problem Reporting open in background - Sudden Blackscreen - Couldn’t open Terminals - Rebooted pressing power button on and off.

I’m trying to learn more about JournalCTL and how to use it. What Warnings, Errors, etc should I be looking for? Nvidia, Plasma, X. org.? Really want to figure out why this happens on some restarts.