Bug: 570.124.04 - freeze on monitor wakeup - "Flip event timeout"

Same absolutely crippling issue with:

  • NVIDIA GeForce RTX 3070 Ti/PCIe/SSE2
    • NVIDIA Driver Version: 570.124.04
    • NVML Version: 12.570.124.04

I am using two monitors and I have used the KDE Plasma monitor brightness dimming options. Both monitors connected only to the Nvidia controller. Nothing unique from the other observations above so I’ll skip posting the diagnostic capture.

I can not leave my machine to sleep; only save to shutdown and restart. CTRL ALT 4 doesn’t always open new terminal login and is not “real time” responsible. Occasionally, I can’t reach the machine via SSH to do a safe shutdown.

Michael

570.133.07 is now out. Let’s see how this goes.

Hi all,

Please help test 570.133.07 when possible. Please let me know if you see the reported failure on this release.

Thank you

I installed 570.133.07 today. Put my laptop to sleep 3 times in a row. On the third sleep, the display didn’t come back to life. Only way to get some life back was to SSH into my machine and restart gdm3. Rebooted the computer, tried again, and this time the display didn’t come back after the second time going to sleep. I have the output of “dmesg” and “journalctl” after the display crash, but don’t have time to cull any identifying info from it before posting online. Happy to send directly to you if that’s helpful.

In my case the combo that triggered no video after monitor wakeup has been 570.124.04 and kernel 6.13.6. The same driver didn’t have any issues with kernel 6.13.5:

[2025-03-05T09:09:16+0100] [ALPM] upgraded nvidia (565.77-11 -> 570.124.04-2) for linux 6.13.5.arch1-1
[2025-03-14T10:28:15+0100] [ALPM] upgraded nvidia (570.124.04-2 -> 570.124.04-3) for linux 6.13.6.arch1-1
[2025-03-19T08:41:03+0100] [ALPM] upgraded nvidia (570.124.04-3 -> 570.124.04-4) for linux 6.13.7.arch1-1
λ journalctl --since "200 days ago" | grep -E "nvidia-modeset.*ERROR" | head
(standard input):1130747:Mar 14 12:08:52 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000c57d:0:0:1221
(standard input):1130748:Mar 14 12:09:24 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000c57e:6:0:1230
(standard input):1130749:Mar 14 12:09:32 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000c57e:7:0:1230
(standard input):1131279:Mar 14 13:41:07 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600
(standard input):1131280:Mar 14 13:41:12 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600
(standard input):1131281:Mar 14 13:41:17 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600
(standard input):1131282:Mar 14 13:41:22 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600
(standard input):1131283:Mar 14 13:41:27 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600
(standard input):1131326:Mar 14 13:41:32 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600
(standard input):1131337:Mar 14 13:41:37 outstanding kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c57d:0 2:0:3608:3600

Will report back about 570.133.07 once it’s available on Arch.

I’ve been experiencing this issue, and I can now report as of driver version 570.133.07 that the issue seems to be resolved! Kernel 6.13.7, KDE Plasma (Wayland)

it also does not happen to me any more with nvidia drivers 570.133.07 and kernel 6.13.7-gentoo-dist

As per above, didn’t resolve it for me with 570.133.07 but I’m on kernel 6.12.17 (Debian 12 testing). I’ll have to wait until they release a kernel update to the 6.13 branch and see if that improves things.

Still happening on 570.133.07.
6.13.7-3-cachyos

Mar 20 22:27:10 p-cachyos kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000900] Flip event timeout on head 0
Mar 20 22:27:10 p-cachyos kwin_wayland[1225]: kwin_wayland_drm: Pageflip timed out! This is a bug in the nvidia-drm kernel driver
Mar 20 22:27:10 p-cachyos kwin_wayland[1225]: kwin_wayland_drm: Please report this at https://forums.developer.nvidia.com/c/gpu-graphics/linux
Mar 20 22:27:10 p-cachyos kwin_wayland[1225]: kwin_wayland_drm: With the output of 'sudo dmesg' and 'journalctl --user-unit plasma-kwin_wayland --boot 0'

I confirm that 570.133.07 fixes my problems on fedora 6.13.6-200.fc41
I couldn’t use my ultrawide monitor, but now I can.

still having the issue with kernel 6.13.7-3-cachyos and 570.133.07

plasmakwin.txt (17 Bytes)
dmesg.txt (102.7 KB)
nvidia-bug-report.log.old.gz (791.7 KB)

Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kwin_wayland[1791]: kwin_wayland_drm: Pageflip timed out! This is a bug in the nvidia-drm kernel driver
Mar 22 11:33:44 deskt0p kwin_wayland[1791]: kwin_wayland_drm: Please report this at https://forums.developer.nvidia.com/c/gpu-graphics/linux
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
Mar 22 11:33:44 deskt0p kernel: [drm:nv_drm_gem_alloc_nvkms_memory_ioctl [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000800] Failed to allocate NVKMS>
lines 923-961/1000 97%

this is an issue in 570.86.16 as well for me

I haven’t had to do this for a long time but I have to mask the latest two drivers in gentoo.

=x11-drivers/nvidia-drivers-570.124.04
=x11-drivers/nvidia-drivers-570.133.07

570.124.04 would freeze but I could get it back. 570.133.07 freezes and crashes out to console. For the latter I added “nvidia-drm.modeset=1 nvidia-drm.fbdev=0” to test, worked for a few days but crashed out today. Didn’t even have to touch it. Turned on the comp, walked away, came back to it frozen, left it to see if it would recover by itself, crashed out to console.

[ 674.134761] [drm:nv_drm_atomic_commit [nvidia_drm]] ERROR [nvidia-drm] [GPU ID 0x00002d00] Flip event timeout on head 0
[ 677.206474] [drm:nv_drm_atomic_commit [nvidia_drm]] ERROR [nvidia-drm] [GPU ID 0x00002d00] Flip event timeout on head 0

If you need more granular traces, I might be able to help. This is on a 4090; driver version seems to have been 570.124.04.

What I can offer for now: I have an output from nvidia-debugdump -D executed with 100 ms pauses in between, which ends up being once every 164 ms. The sequence I have right now is from after turning on the monitor and it failing to come up, but it shows the point where the system freezes when I turn off the monitor that failed to come up. I don’t know if that point in time is interesting to you, although of course it would be ideal if it didn’t cause a full system crash :)

Since I seem to be able to reproduce this at least reasonably often without trying, I could work to get you a similar dump of the monitor being turned on and failing to come up?

I repacked the .zips into a .tar.zst since this made the 617 MB of .zips fit into 11 MB: https://sliedes.kapsi.fi/nv/system-hang-seq.tar.zst

The dumps are numbered from 0 onwards. Before each invocation of nvidia-debugdump -D, I printed the number on the other display; i.e. something like for ((i=0;; i++)); do echo $i; nvidia-debugdump -D; mv dump.zip $i.zip; done. The last number displayed on my monitor before the hang (that happened maybe a couple of seconds after I turned off the monitor that failed to come up) was 83.
Here’s a bug report log, but note that this is after a reboot; I’ll try to do better next time.
nvidia-bug-report.log.gz (701.2 KB)

Mar 23 11:47:03 poyta kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002100] Flip event timeout on head 0
Mar 23 11:47:07 poyta kernel: [drm:nv_drm_atomic_commit [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002100] Failed to apply atomic modeset.  Error code: -22

I was one of the people reporting this issue on RTX 4070 with single monitor via DP on 570.124.04. Now with 570.133.07 (on latest Arch) it seems to be fixed for me - I’ve tried suspending and waking up system, just lock and wait for monitor to go stand by and wake it up; several times - everything works so far.

The issue is fixed for me as well on 570.133.07.

However, there is a new problem.

During boot, sometimes only one of the monitor is activated. I think this occurs about 20-40% of the times.
Workaround is either to reboot and try again, or to unplug and replug the other monitor.

Not fixed for me on 570.133.07, though mine never occurred on monitor wakeup. It occurs completely randomly.

This bug is still happening on driver 570.133.07. I have tested it with the new the drivers in both KDE and Gnome (both on Wayland) and the second screen freezes randomly.

After putting the PC to sleep, waking up and then using spectacle to crop a part of the screen (meta + shift + print screen) causes Pageflip timed out event. Please see the attached log from journalctl
Distro: Fedora 41
KDE 6.3.3
Kernel: 6.13.8
Nvidia driver: 570.133.07
Monitors: 2x LG UltraGear 1440p 144Hz (27GL850-B)
GPU: Asus TUF 5070 TI
nvidia-error.txt (122.6 KB)

I’m getting this issue from time to time, usually just after signing out and then signing back in. Happens in both GNOME and KDE. I’m running NVIDIA driver version 570.133.07 and Kernel version 6.13.8-200.fc41.x86_64

nvidia-bug-report.log.gz (1.1 MB)

I’m getting this also at random times, 570.124.04, open kernel drivers, kernel 6.11.0-21-generic #21~24.04.1-Ubuntu, KDE Neon 6.3.3, RTX 3060, Wayland. Single monitor, LG 32GK650F-B.

The workaround for this, after display freezes, is to go to the virtual console (e.g. Ctrl+Alt+F3) and back to desktop session (e.g. Alt+F2). After this, desktop continues to work without any problem… until next time.

For me it usually happens while randomly using the browser (Chromium based, X11 client).