Nvidia issues after updating driver to version 555

UnluckyNinja · 2024-07-05 07:06:27

I'm using a GTX 1070 TI and I updated nvidia-dkms and nvidia-utils to version 555.58.02-1, but after reboot I'm having many issues. Applications like kitty, mpv and supertuxkart do not open. Betterbird (a fork of Thunderbird) opens but crashes every few minutes. Firefox crashes when I open it.

When I try to open kitty or mpv via xfce4-terminal I get this output:

zsh: segmentation fault (core dumped)

mpv outputs more stuff:

[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu] Force-disabling compute shaders as an FBO format was not available! See your FBO format configuration!
[vo/gpu] High bit depth FBOs unsupported. Enabling dumb mode.
[vo/gpu] Most extended features will be disabled.
[vo/gpu/opengl] mapping buffer: OpenGL error OUT_OF_MEMORY.
AO: [pipewire] 48000Hz stereo 2ch floatp
VO: [gpu] 480x702 yuv420p
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu/opengl] Error: framebuffer completeness check failed (error=36061).
[vo/gpu] Error: texture could not be created.
[vo/gpu] Force-disabling compute shaders as an FBO format was not available! See your FBO format configuration!
zsh: segmentation fault (core dumped) mpv Videos/video.mp4

I'm using the Chromium browser right now and it works completely fine, I can even watch 4k videos on YouTube. Telegram-desktop and Spotify also seem to be running completely fine.

lspci output:

01:00.0 VGA compatible controller: NVIDIA Corporation GP104 [GeForce GTX 1070 Ti] (rev a1)
Subsystem: eVga.com. Corp. Device 5671
Kernel driver in use: nvidia
Kernel modules: nouveau, nvidia_drm, nvidia

lsmod | grep nvidia output:

nvidia_uvm 6672384 0
nvidia_drm 122880 9
nvidia_modeset 1617920 23 nvidia_drm
video 77824 1 nvidia_modeset
nvidia 60997632 279 nvidia_uvm,nvidia_modeset

dmesg | grep -i nvidia output:

[ 1.598524] nvidia: loading out-of-tree module taints kernel.
[ 1.598531] nvidia: module license 'NVIDIA' taints kernel.
[ 1.598535] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[ 1.598536] nvidia: module license taints kernel.
[ 2.070389] nvidia-nvlink: Nvlink Core is being initialized, major device number 237
[ 2.074609] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=io+mem
[ 2.192323] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 555.58.02 Tue Jun 25 01:39:15 UTC 2024
[ 2.224995] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 555.58.02 Tue Jun 25 01:10:21 UTC 2024
[ 2.228115] nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
[ 2.228893] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 2.228896] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 1
[ 2.383395] nvidia-uvm: Loaded the UVM driver, major device number 235.
[ 13.910813] input: HDA NVidia HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input13
[ 13.910863] input: HDA NVidia HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input14
[ 13.910906] input: HDA NVidia HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input15
[ 13.910952] input: HDA NVidia HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input16
[ 18.788242] caller _nv000733rm+0x11e/0x255 [nvidia] mapping multiple BARs
[ 1452.452882] vo[13900]: segfault at 4 ip 000074ab9a8778c7 sp 000074abb56c2b40 error 4 in libnvidia-eglcore.so.555.58.02[74ab9a3ea000+c0c000] likely on CPU 1 (core 1, socket 0)
[ 1470.380350] vo[14034]: segfault at 4 ip 00007ad12e8778c7 sp 00007ad15d491b40 error 4 in libnvidia-eglcore.so.555.58.02[7ad12e3ea000+c0c000] likely on CPU 5 (core 1, socket 0)

This libnvidia-eglcore error wasn't there when I started writing this post.

nvidia-smi command shows that GPU memory is being utilized by some applications.

I don't have any kernel commandline parameter for nvidia and I also do not have a nvidia.conf file in /etc/modprobe.d/ so it is all defaults.

Anyone knows if there is a fix for this other than downgrading the nvidia drivers?

Last edited by UnluckyNinja (2024-07-05 07:15:07)

UnluckyNinja · 2024-07-05 07:30:01

UPDATE:

I don't know why but I was able to open the video that was outputting errors with mpv using the very same command that was not working before o.O mpv is working now.
I'm still having issues with other applications (kitty, firefox, steam, supertuxkart and possibly others too).
GIMP seems to be working and electron applications like Vesktop and Ferdium too.

Gilis95 · 2024-07-05 09:14:53

Same here, on the same GPU, here's the stacktrace:
```
#0 0x0000000000000000 in ?? ()
#1 0x00007f138289a71f in ?? () from /usr/lib/libnvidia-glcore.so.555.58.02
#2 0x00007f138289abc1 in ?? () from /usr/lib/libnvidia-glcore.so.555.58.02
#3 0x00007f139e979aa4 in ?? () from /usr/lib/libGLX_nvidia.so.0
#4 0x00007f139e948330 in ?? () from /usr/lib/libGLX_nvidia.so.0
#5 0x00007f138da09bf2 in QGLXContext::swapBuffers(QPlatformSurface*) () from /opt/zoom/Qt/plugins/xcbglintegrations/libqxcb-glx-integration.so
#6 0x00007f139ebb796c in QOpenGLContext::swapBuffers(QSurface*) () from /opt/zoom/Qt/lib/libQt5Gui.so.5
#7 0x00007f13a0c056f0 in QSGRenderThread::syncAndRender(QImage*) () from /opt/zoom/Qt/lib/libQt5Quick.so.5
#8 0x00007f13a0c08bcb in QSGRenderThread::run() () from /opt/zoom/Qt/lib/libQt5Quick.so.5
#9 0x00007f139c2b470d in QThreadPrivate::start(void*) () from /opt/zoom/Qt/lib/libQt5Core.so.5
#10 0x00007f13a12a6ded in start_thread (arg=<optimized out>) at pthread_create.c:447
#11 0x00007f13a132a0dc in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:78

```

LksNuns · 2024-07-22 11:42:34

I am facing the same problem, with the same GPU too... my issues are with firefox / zoom so far.

Does anybody knows if I can downgrade this without much impact on the system?

Last edited by LksNuns (2024-07-22 13:02:40)

seth · 2024-07-22 14:47:23

You cannot downgrade the nvidia module w/o also downgrading the kernel.
Instead use https://aur.archlinux.org/packages/nvidia-535xx-dkms

clements · 2024-07-29 16:23:05

I'm using a GTX 1080 and was encountering the same kind of issue with a SDL/wgpu-native project.
Using the nvidia-535xx-dkms driver seem to have solved the issue.

Maybe all Pascal based Nvidia GPU are impacted.

Arch Linux

#1 2024-07-05 07:06:27

Nvidia issues after updating driver to version 555

#2 2024-07-05 07:30:01

Re: Nvidia issues after updating driver to version 555

#3 2024-07-05 09:14:53

Re: Nvidia issues after updating driver to version 555

#4 2024-07-22 11:42:34

Re: Nvidia issues after updating driver to version 555

#5 2024-07-22 14:47:23

Re: Nvidia issues after updating driver to version 555

#6 2024-07-29 16:23:05

Re: Nvidia issues after updating driver to version 555

Board footer