That is different issue than this topic (monitor freeze with “flip event timeout” in logs), but it is probably worth it to create new topic for it, I will confirm it there, since I got the same with 2 different games.
I’ve been running nvidia-open 570.133.07-1 on Linux 6.13.7.arch1-1 for a week now and no more issues after monitor wake up. Thanks!
Oh wait, I though I posted in the general 570 discussions, also I think I have found a fix. I will confirm in a few minutes (nearing the 40 minute mark now).
Yes, you should follow the instructions on this post, and it seems to be totally unrelated (AMD has the issue too).
fixed for me in 570.133.07 hopefully also soon in Ubuntu PPA
Still happening on nvidia-open 570.133.07-1, RTX 5080, running EndeavourOS with kernel 6.12.21-1-lts. Using KDE w/ Wayland on a single monitor (Samsung Odyssey Neo G9 49").
The screen freezes but audio/etc. continues to work. It happens anywhere from 1-3 times a day, seemingly randomly. I haven’t correlated it with any specific action or program. The only fix is to switch to a virtual console and reboot the machine from there. Trying to switch back to the desktop without a reboot just shows a black screen with a cursor.
When the freeze occurs I get the pageflip timeout error over and over:
Apr 02 16:12:33 eos-desktop kwin_wayland[1220]: kwin_wayland_drm: Pageflip timed out! This is a bug in the nvidia-drm kernel driver
Apr 02 16:12:33 eos-desktop kwin_wayland[1220]: kwin_wayland_drm: Please report this at https://forums.developer.nvidia.com/c/gpu-graphics/linux
Apr 02 16:12:33 eos-desktop kwin_wayland[1220]: kwin_wayland_drm: With the output of 'sudo dmesg' and 'journalctl --user-unit plasma-kwin_wayland --boot 0'
After switching to a virtual console that error stops and it logs one final error:
Apr 02 16:13:19 eos-desktop kwin_wayland[1220]: kwin_wayland_drm: Pageflip arrived after all, 46755ms after the commit
Apr 02 16:13:19 eos-desktop kwin_wayland[1220]: kwin_wayland_drm: atomic commit failed: Permission denied
journalctl_out.txt (23.1 KB)
dmesg_out.txt (117.1 KB)
Same here 570.133.07, RTX 5080 FE, Arch using KDE w/ Wayland on a single HDMI monitor. Initially is seemed random to me but I think I have narrow it down a little bit.
- Triggering fullscreen in video or gaming app causes crash. Hard reboot necessary and pageflip errors logged.
- Sleep/Resume MUST have been used previously for the issue to occur. I can resume from sleep and play a game in windowed mode for a hour without issues, but once I try full screen “crash”. Otherwise I can do the same task without issue in fullscreen UNTIL I sleep/wake up.
journalctl_out.txt (4.9 KB)
Have the same issue when VRR activates. Disable VRR as a temporary fix
Same issue here.
With 570.133.07 got Flip event timeout and freeze of Wayland almost instantly after login.
Reproduces only with RTX 5080. Not Reproduced with RTX 4060.
With 570.124.04 - All works until game decides to go into fullscreen mode. Then I have same pageflip timed out.
Reproduces only with RTX 5080. Not Reproduced with RTX 4060.
In both cases - single monitor, DP, HDR on. Used KDE 6.3.4
Issue is still present on 570.144.
dmesg.log (82.8 KB)
journalctl.log (12.5 KB)
On Plasma 6.3.4 (Wayland, Ubuntu 25.04) Kubuntu, multi-monitor (3x DP) setup with NVIDIA RTX 3080 Ti on driver 570.133.07. After idle, only 1 screen unfreezes, others freeze until I TTY switch. Sometimes it takes multiple tries. Journal shows repeated [nvidia-drm] Flip event timeout on head X.
– this happens without the monitors going to sleep.
– Update I resolved the issue. In Power Management System Settings I set Dim automatically to “Never”.
I still have the issue (screen freezing randomly and needing a log out or something like that). For dual screen, using the other screen to change the display mode (like removing one display and immediately reverting it) help solving the problem without logging out.
Specs: Fedora 42, nvidia 570.144 (I had this bug for at least two mounts), NVIDIA GeForce RTX 3060.
The bug only occurs when using two screens, and it always starts on my monitor shenzhen KTC M27T20S.
Having the same issue.
RTX5080, dual monitor, fedora 42 with NVIDIA closed drivers on Wayland.
Same issue here. I tested on both Fedora 42 and Ubuntu 25.04 (6.14.0-15-generic)
It seems pressing Ctrl-Alt-F1 (TTY restart) does the job for now.
[49389.536590] [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
[49392.671667] [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Flip event timeout on head 0
I’m also having this issue and error. Wasn’t doing anything special. YouTube video playing in the background on Librewolf. Everything suddenly froze. Mouse and keyboard weren’t responding. Cursor disappeared. Sound continued playing in the background. Alternating between TTY eventually resolved the issue, or maybe it was a coincidence, I don’t know.
Arch / Linux 6.14.6-arch1-1
nvidia-open-dkms 570.144
4070 Ti Super / Single monitor / Display Port
Screensaver and suspend are disabled
Any updates? It’s still happening for me same as it ever did. I even tried reinstalling the OS from scratch in a desperate attempt to fix this but it’s still happening on a clean Arch w/ KDE. Latest kernel (6.14.6.arch1-1), latest nvidia-dkms (570.144-3), Plasma on Wayland (6.3.5-1), same logs as everyone else above.
The easiest way for me to reproduce the bug is by messing with the night light feature of KDE: if you change the Day (or Night) light temperature it will preview the color while you’re moving the slider, and I assume that is what causes the timeout. Just a second or two of jerking the temperature slider around causes the display to freeze. Obviously that’s not the only scenario when it happens, but it’s 100% reproduceable for me.
The only solution I found is downgrading to 570.86.16-2, but it’s been so long that it also requires downgrading the kernel to 6.13 because 6.14 doesn’t work with nvidia-dkms that old.
Please fix this, I’m begging you.
This has started happening to me as well, sometimes it will not happen for a whole day, other times it will happen multiple times in the span of an hour. Switching to another tty and then back to Gnome fixes it.
Fedora 42 on Gnome Wayland
RTX 4090 on nvidia 570.144
Intel i9 13900K
Same issue here.
TTY restart fix it also most of the time.
Fedora 42 Gnome Wayland with Kernel 6.14.6
Nvidia closed driver 570.153.02
RTX 4060 Ti 16GB
Intel i5 12400F
Same thing here. Though it started to be a problem for me since a few versions ago
Tested out the beta driver 575.51.02 and it seems to have fixed this issue for me! Can’t reproduce it with my usual method.
Hallelujah!
Sadly for me it’s no better. Running KDE Plasma 6.3.4, Linux 6.12.27 on Debian Testing with Xorg and the beta driver still crashes my display. I can remotely ssh to the machine, but the keyboard is inactive and I cannot switch tty and back again.