Fedora 41 KDE Plasma 6.3.1
Kernel 6.12.15
GPU 3090
Primary Monitor Dell S2716DG
Secondary ASUS VG248
Intermittent issue with secondary monitor freezing. Moving windows to the primary monitor will still be usable. Most recent issue occured when playing plateup on secondary monitor and discord on primary. Turning off the monitor does not resolve the issue. I have had the same issue when playing path of exile 2 and a youtube video on secondary monitor or discord on secondary monitor but didn’t report it. Goes back to at least Kernel 6.12.11 on Fedora’s release schedule.
Attempting to change resolution causes main monitor to freeze. However, I was able to get around it by pressing alt+f2 and blind typing kwin --replace to get control back. Second monitor remained frozen.
Feb 25 00:38:19 fedora kwin_wayland[4092]: kwin_wayland_drm: Please report this at https://forums.developer.nvidia.com/c/gpu-graphics/linux
Feb 25 00:38:19 fedora kwin_wayland[4092]: kwin_wayland_drm: With the output of 'sudo dmesg' and 'journalctl --user-unit plasma-kwin_wayland --boot 0'
The problem with the issue is that it can either never happen or happen very frequently. Back when I was playing a lot of Path of Exile I could guarantee it would happen at least once, sometimes multiple times within a 3 hour session.
I haven’t been utilizing the latest drivers so unfortunately I can’t say.
currently not experiencing the bug anymore after changing my display manager configuration a bit but before that i’d been experiencing the issue right around 13 hours of uptime and not one of my sessions would go past 15 hours of uptime without a screen freezing up
I am able to make my issue repeatable by changing the input on my primary monitor to my other computer, then changing it back to this computer again after 10 minutes. Occurs on the latest drivers (570.124.04)
Repeatable on a different distro but same DE and similar monitor. The difference is that the ASUS is primary, but still the monitor that is freezing. It seems to have something to do with VRR and async flips, from some digging.
Manjaro KDE Plasma 6.3.4
Kernel 6.12.21
GPU 2080 Ti
Primary Monitor ASUS VG278
Secondary LG FHD
Nvidia Driver Version: 570.133.07
This can be reproduced instantly by toggling Night Light on.
I’ve got the same issue with various monitors freezing and the same logs.
It is not restricted to one monitor. It can be one monitor one day, and another the next.
All 3 of my monitors are from different brands and era. 2 of my monitors are connected through DisplayPort (both of them had FreeSync enabled). The other, connected through HDMI also froze, but does not support FreeSync or G-Sync.
It happens with many applications. I normally have at least one 3d application running at any given time through WINE. I also am consistently running at least a couple electron apps, and a firefox-based web browser.
I generally see it when my system has been running for at least an hour. Timing is highly variable. I haven’t managed to strongly correlate it with any application. Switching terminals with Control + Alt F2 or F5 or etc, then back to the main vty with F1 does seem to fix it – sometimes. Occasionally, but not always, after I switch VTYs like that, I will need to disconnect and reconnect my mouse and keyboard to gain input back. Switching VTYs doesn’t work again when input is lost, the system seems to ignore the devices entirely (though I haven’t tried the magic sysrq). about 5-10% of the time, the entire system locks up and needs to be rebooted at least from the physical terminal – SSH still works to the system typically. Usually I can continue using the system after the intervention, outside of the latter case.
Operating System: Nobara Linux 41
KDE Plasma Version: 6.3.4
KDE Frameworks Version: 6.13.0
Qt Version: 6.8.2
Kernel Version: 6.14.2-201.nobara.fc41.x86_64 (64-bit)
Graphics Platform: Wayland
Processors: 48 × AMD Ryzen Threadripper 3960X 24-Core Processor
Memory: 62.7 GiB of RAM
Graphics Processor: NVIDIA TITAN RTX
DRIVER version : 570.133.07
I have disabled FreeSync on my monitors and will check back if it resolves the issue in a few days.
Inside the journalctl log, at the very bottom, I have included log lines after the terminal switch intervention.
After setting all of my monitors to 60hz no HDR, VRR disabled in the monitor hardware; I haven’t had a crash like this in 15 days. I have had plasmashell crash but I have never had to restart my system due to a hard lock. I also haven’t had to re-attach my peripherals as often. Having non-unified refresh rates also had problems; within the same day. 144/144/60 hz did not work. I wonder if there is some frame counter or something that messes up
Having the same issue on a 3070 TI
Hoping to give more info to get an update out to fix the issue either on the driver side or plasma side. screen_freezing_logs.tar.gz (148.1 KB)
Additional info:
Nobara 41
Ryzen 7 5800X3D
32 gigs of RAM
MEG B550 UNIFY-X motherboard
Hi,
I see similar error logs reported by another user on different thread and he no longer sees this issue with beta drivers (currently 575.51.02).
Could someone please verify it once and update.