Intermittent black screen on fullscreen games after suspend or turning off/on monitor

Hi! I’ve been facing an issue on Fedora 41 whereby launching a game (both Wine and native ones) in fullscreen mode after I have in some way turned off the monitor beforehand (either suspending the PC or simply turning the display off and back on) will make the screen entirely black sometimes permanently as long as the game is on-screen, other times in weird bouts where it is black for a while, then I can see the content again and then black, like this back and forth.
I don’t know what’s happened to the system to provoke this change as it had been working correctly before, though I can’t say at which point exactly the behaviour started appearing.
Right now my system is in this state:

$ rpm-ostree status 
State: idle
Deployments:
● fedora:fedora/41/x86_64/kinoite
                  Version: 41.20241222.0 (2024-12-22T00:52:10Z)
               BaseCommit: b822534f80bb6fd0b80465ec2acb82aec788c6e545af4d8e8f0465a42b4a6217
             GPGSignature: Valid signature by 466CF2D8B60BC3057AA9453ED0622462E99D6AD1
      RemovedBasePackages: noopenh264 0.1.0~openh264_2.4.1-2.fc41
          LayeredPackages: distrobox drm_info gstreamer-plugins-espeak gstreamer1-plugin-openh264 gwenview intel-media-driver libvirt-daemon libvirt-daemon-config-network libvirt-daemon-driver-interface libvirt-daemon-driver-network
                           libvirt-daemon-driver-nwfilter libvirt-daemon-driver-qemu libvirt-daemon-driver-secret libvirt-daemon-driver-storage-core lm_sensors mozilla-openh264 mpv-libs PackageKit-gstreamer-plugin pipx
                           plasma-workspace-x11 qemu-kvm rpmfusion-nonfree-release steam-devices vnstat vorta waydroid

With the following configuration as seen in Plasma’s system settings:

Operating System: Fedora Linux 41
KDE Plasma Version: 6.2.4
KDE Frameworks Version: 6.9.0
Qt Version: 6.8.1
Kernel Version: 6.12.5-200.fc41.x86_64 (64-bit)
Graphics Platform: Wayland
Processors: 16 × AMD Ryzen 7 5700G with Radeon Graphics
Memory: 14,9 GiB of RAM
Graphics Processor: Mesa Intel® Arc
Manufacturer: ASUS

Hi, we will need a journalctl log to help you

Ayy, we meet again :smiley:
Anyways, yes, what operations should I do before extracting its output to capture the interesting things in the log?

Hi there!

Not sure, just start journalctl -f --priority err and then do whatever you do, then exit and press ctrl+c

You can also filter by application but we dont know what that is yet.

Alright, here’s what I get from that:

Dec 24 20:35:11 fedora kernel: snd_hda_intel 0000:04:00.0: Unknown capability 0
Dec 24 20:35:11 fedora (udev-worker)[1182]: event13: Failed to call EVIOCSKEYCODE with scan code 0x7c, and key code 190: Invalid argument
Dec 24 20:35:11 fedora kernel: ee1004 12-0053: probe with driver ee1004 failed with error -5
Dec 24 20:35:12 fedora systemd-tmpfiles[1451]: "/home" already exists and is not a directory.
Dec 24 20:35:12 fedora systemd-tmpfiles[1451]: "/srv" already exists and is not a directory.
Dec 24 20:35:12 fedora systemd-tmpfiles[1451]: "/root" already exists and is not a directory.
Dec 24 20:35:22 fedora systemd[1]: Failed to start vnstat.service - vnStat network traffic monitor.
Dec 24 20:35:29 fedora org_kde_powerdevil[2411]: busno=8, sleep-multiplier =  2,00. Testing for supported feature 0x10 returned Error_Info[DDCRC_RETRIES in ddc_write_read_with_retry, causes: DDCRC_NULL_RESPONSE(10)]
1 Like

Does the issue happen if you boot from a previous version of the kernel, say 6.12.4? I know there have been a lot of regressions and issues with 6.12.5 that’s causing screen blinking, blanking and other not fun stuff.

Ah that’s already a good indicator, let me try

Actually I was reading up on how to do that and I’m kind of lost, so do I have to do as instructed here? os/docs/faq.md at master · openshift/os · GitHub
If so, how do I get packages of the version I want? I read elsewhere that’s Koji providing them, right?

I figured out how to do that now, I rolled the kernel back to version 6.12.4 with this command:

rpm-ostree override replace https://koji.fedoraproject.org/koji/buildinfo?buildID=2599252

with this as the final report:

Upgraded:
  xorg-x11-server-Xorg 21.1.14-2.fc41 -> 21.1.15-1.fc41
  xorg-x11-server-common 21.1.14-2.fc41 -> 21.1.15-1.fc41
Downgraded:
  kernel 6.12.5-200.fc41 -> 6.12.4-200.fc41
  kernel-core 6.12.5-200.fc41 -> 6.12.4-200.fc41
  kernel-modules 6.12.5-200.fc41 -> 6.12.4-200.fc41
  kernel-modules-core 6.12.5-200.fc41 -> 6.12.4-200.fc41
  kernel-modules-extra 6.12.5-200.fc41 -> 6.12.4-200.fc41
  kernel-tools 6.12.5-200.fc41 -> 6.12.4-200.fc41
  kernel-tools-libs 6.12.5-200.fc41 -> 6.12.4-200.fc41
  python3-perf 6.12.5-200.fc41 -> 6.12.4-200.fc41

So I rebooted and tested this again, unfortunately the result is the same, I keep getting black screens even on the older kernel, this is the output of journalctl

$ journalctl -f --priority err
Dec 26 16:05:24 fedora kernel: mei i915.mei-gscfi.768-e2c2afa2-3817-4d19-9d95-06b16b588a5d: cannot connect
Dec 26 16:05:24 fedora (udev-worker)[1201]: event13: Failed to call EVIOCSKEYCODE with scan code 0x7c, and key code 190: Invalid argument
Dec 26 16:05:24 fedora kernel: snd_hda_intel 0000:04:00.0: Unknown capability 0
Dec 26 16:05:24 fedora systemd-tmpfiles[1462]: "/home" already exists and is not a directory.
Dec 26 16:05:24 fedora systemd-tmpfiles[1462]: "/srv" already exists and is not a directory.
Dec 26 16:05:24 fedora systemd-tmpfiles[1462]: "/root" already exists and is not a directory.
Dec 26 16:05:36 fedora systemd[1]: Failed to start vnstat.service - vnStat network traffic monitor.
Dec 26 16:08:06 fedora kwin_wayland[2136]: kwin_scene_opengl: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
Dec 26 16:08:09 fedora kwin_wayland[2136]: kwin_scene_opengl: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
1 Like

I share the same problem - since about Christmas, fullscreen apps black out intermittently. The problem seems to be related to wayland - X11 sessions are not affected.

I tried to rollback my latest updates to see which one introduced the problem, but failed - it seems the packages are not available anymore.

1 Like

Which packages are you looking for?

I am not completely sure what my problem is or if I am attempting is even possible .
I tried to go back a couple of dnf transactions like this:
sudo dnf history rollback 14

The idea is to try to pin the problem to an exact point in time to be able to compare the versions where it works and where it broke.

However, I am getting the plethora of messages listed below and am not sure how to interpret them .

Updating and loading repositories:
Repositories loaded.
Failed to resolve the transaction:
Transaction merge error: 'Action 'Install' 'kernel-modules-extra-6.11.11-300.fc41.x86_64' cannot be merged because it is already installed in version 'kernel-modules-extra-6.12.4-200.fc41.x86_64' -> keeping the action from older transaction with 'kernel-modules-extra-6.11.11-300.fc41.x86_64'.'
Transaction merge error: 'Action 'Install' 'kernel-modules-core-6.11.11-300.fc41.x86_64' cannot be merged because it is already installed in version 'kernel-modules-core-6.12.4-200.fc41.x86_64' -> keeping the action from older transaction with 'kernel-modules-core-6.11.11-300.fc41.x86_64'.'
Transaction merge error: 'Action 'Install' 'kernel-modules-6.11.11-300.fc41.x86_64' cannot be merged because it is already installed in version 'kernel-modules-6.12.4-200.fc41.x86_64' -> keeping the action from older transaction with 'kernel-modules-6.11.11-300.fc41.x86_64'.'
Transaction merge error: 'Action 'Install' 'kernel-devel-6.11.11-300.fc41.x86_64' cannot be merged because it is already installed in version 'kernel-devel-6.12.4-200.fc41.x86_64' -> keeping the action from older transaction with 'kernel-devel-6.11.11-300.fc41.x86_64'.'
Transaction merge error: 'Action 'Install' 'kernel-core-6.11.11-300.fc41.x86_64' cannot be merged because it is already installed in version 'kernel-core-6.12.4-200.fc41.x86_64' -> keeping the action from older transaction with 'kernel-core-6.11.11-300.fc41.x86_64'.'
Cannot perform Install action, no match for: kernel-6.11.11-300.fc41.x86_64.
Cannot perform Install action, no match for: kernel-core-6.11.11-300.fc41.x86_64.
Cannot perform Install action, no match for: kernel-devel-6.11.11-300.fc41.x86_64.
Cannot perform Install action, no match for: kernel-modules-core-6.11.11-300.fc41.x86_64.
Cannot perform Install action, no match for: kernel-modules-extra-6.11.11-300.fc41.x86_64.
Cannot perform Install action, no match for: kernel-modules-6.11.11-300.fc41.x86_64.
Cannot perform Install action, no match for: libxcrypt-devel-4.4.36-11.fc41.x86_64.
Cannot perform Install action, no match for: kernel-tools-6.12.5-200.fc41.x86_64.
Cannot perform Install action, no match for: python3-perf-6.12.5-200.fc41.x86_64.
Cannot perform Install action, no match for: xorg-x11-server-common-21.1.14-2.fc41.x86_64.
Cannot perform Install action, no match for: flatpak-1.15.12-1.fc41.x86_64.
Cannot perform Install action, no match for: kernel-tools-libs-6.12.5-200.fc41.x86_64.
Cannot perform Install action, no match for: kmod-nvidia-6.11.11-300.fc41.x86_64-3:565.77-1.fc41.x86_64.
Cannot perform Install action, no match for: libnfsidmap-1:2.8.1-2.rc2.fc41.x86_64.
Cannot perform Install action, no match for: libxcrypt-4.4.36-11.fc41.x86_64.
Cannot perform Install action, no match for: nfs-utils-1:2.8.1-2.rc2.fc41.x86_64.
Cannot perform Install action, no match for: flatpak-selinux-1.15.12-1.fc41.noarch.
Cannot perform Install action, no match for: tigervnc-license-1.14.1-2.fc41.noarch.
Cannot perform Install action, no match for: tigervnc-server-minimal-1.14.1-2.fc41.x86_64.
Cannot perform Install action, no match for: xorg-x11-server-Xephyr-21.1.14-2.fc41.x86_64.
Cannot perform Install action, no match for: kernel-devel-matched-6.12.5-200.fc41.x86_64.
Cannot perform Install action, no match for: ibus-typing-booster-2.26.12-1.fc41.noarch.
Cannot perform Install action, no match for: flatpak-session-helper-1.15.12-1.fc41.x86_64.
Cannot perform Install action, no match for: flatpak-libs-1.15.12-1.fc41.x86_64.
Problem: problem with installed package
  - installed package systemd-256.10-1.fc41.x86_64 requires libcrypt.so.2()(64bit), but none of the providers can be installed
  - installed package systemd-256.10-1.fc41.x86_64 requires libcrypt.so.2(XCRYPT_2.0)(64bit), but none of the providers can be installed
  - installed package systemd-256.10-1.fc41.x86_64 requires libcrypt.so.2(XCRYPT_4.4)(64bit), but none of the providers can be installed
  - package systemd-256.7-1.fc41.x86_64 from fedora requires libcrypt.so.2()(64bit), but none of the providers can be installed
  - package systemd-256.7-1.fc41.x86_64 from fedora requires libcrypt.so.2(XCRYPT_2.0)(64bit), but none of the providers can be installed
  - package systemd-256.7-1.fc41.x86_64 from fedora requires libcrypt.so.2(XCRYPT_4.4)(64bit), but none of the providers can be installed
  - package systemd-256.10-1.fc41.x86_64 from updates requires libcrypt.so.2()(64bit), but none of the providers can be installed
  - package systemd-256.10-1.fc41.x86_64 from updates requires libcrypt.so.2(XCRYPT_2.0)(64bit), but none of the providers can be installed
  - package systemd-256.10-1.fc41.x86_64 from updates requires libcrypt.so.2(XCRYPT_4.4)(64bit), but none of the providers can be installed
  - conflicting requests
You can try to add to command line:
  --skip-unavailable to skip unavailable packages

Can you check if this is the same?

Sorry, I’m not sure how to help there since that’s standard Fedora, it’s very different from how you would work with rpm-ostree

Can you try setting the KDE power profile to performance to see if it resolves the issue.

Thanks

I’ve already tried that after digging through the links connected to the page you linked, it didn’t make a difference unfortunately.
Should I try reverting the kernel to the latest available version and retry to rule this one out completely?

6,12.4 should be unaffected AFAIK.

I don’t know, I’ve run uname -r and got that version as output, behaviour is still the same, so unless there is something wrong with my configuration, that kernel is affected too.
I’ve also done the test of Performance mode on the latest kernel 6.12.6-200.fc41.x86_64 and the problem happens still.

I don’t know if it is relevant at all, but I noticed some similar flickering on Windows 11 on this same system that happens after coming back from suspension.
It seemed to happen when I updated to the latest driver there, in that case it is temporary and it does that without loading up any specific application in fullscreen, it just does it all across.
I wonder if there is an issue with both Intel drivers introduced recently.

By trying I found out that logging out and back in (still in the Wayland session) makes the problem go away until the screen is turned off again, so it effectively resets the status of whatever is breaking, quite a bit better than having to reboot fully.
Anyway, has anyone figured out a way to avoid that step too, maybe?