Bug 1217748

Created attachment 871112 [details]
possibly overflowed dmesg taken while corruptions happening

Created attachment 871113 [details]
Screenshot of the gnome-terminal.

This is the only time I could take a screenshot with glitches happening. Several others show the screen is fine, There is absolutely no transparency/hacks etc applied. I am using unmodified gnome with only additional keyboard layout.
I have to add that the glitches are "animated" syncing with cursor flash.

Created attachment 871114 [details]
photo of terminal (via phone)

As the screenshots look fine when taken digitally, I captured photo of terminal. Terminal developers say Gnome Terminal can't display these even if it wanted to.

Created attachment 871115 [details]
Different looking issue with kde system monitor

This is kde system monitor. Corruption looks different. Kate etc are all fine.

As it's a nouveau problem on TW, the best would be to report to the upstream,
e.g. gitlab.freedesktop.org Issues
  https://gitlab.freedesktop.org/drm/nouveau/-/issues/

(In reply to Takashi Iwai from comment #5)
> As it's a nouveau problem on TW, the best would be to report to the upstream,
> e.g. gitlab.freedesktop.org Issues
>   https://gitlab.freedesktop.org/drm/nouveau/-/issues/

As I am mainly a KDE user I almost never launch Gnome-Terminal. On my home laptop having Intel 5500 GPU which already have some freeze issues and debugging enabled I launched gnome-terminal by mistake and to my surprise it displayed the exact same glitch. As it won't appear in screen digital recording, I took a video of it. I am making some space for distrobox right now so I can test another similar distributions gnome-terminal.

Did you open the gitlab issue entry for nouveau already and discussed with developers there?  If yes, please give the URL, so that we can note it here, too as a reference.

In anyway, if the same problem is seen on i915, it's likely a different problem.
You can try to report either to KDE or GNOME at first, although this can be rather a rendering problem in a lower layer.

(In reply to Takashi Iwai from comment #7)
> Did you open the gitlab issue entry for nouveau already and discussed with
> developers there?  If yes, please give the URL, so that we can note it here,
> too as a reference.
> 
> In anyway, if the same problem is seen on i915, it's likely a different
> problem.
> You can try to report either to KDE or GNOME at first, although this can be
> rather a rendering problem in a lower layer.

I managed to create the exact same issue on a fresh user account and Fedora 39 gnome-terminal via distrobox does have the same issue too.

It is possible to compress the fresh user account home directory to a file as it doesn't have any personal info in case it is needed.

gitlab/nouveau bug report is at https://gitlab.freedesktop.org/drm/nouveau/-/issues/295

I am pretty much confused now since I reproduced the same issue with a fresh Tumbleweed install with a fresh home on Intel 5200U HD5500 GPU. This time it happens under KDE current (TW 20240108) with Microsoft Edge and Google Chrome.
I am OK to test anything.
It is exactly the same with HD5500, e.g. when you screenshot or screen record, issue isn't there in the recording but you can see it. I took a video of it.

I've been seeing SDDM look like attachment #871115 [details] on multiple installations, but was ignoring it until stumbling here. If the following seems related, I will have to return, after determining where among 15.5, 15.6, Slowroll, TW and/or other Mageia or Fedora, and what CPUs/GPUs, other than the last seen:

TW20240301 & 6.6.18 on a Cedar Mill Pentium 641 CPU and PCI Tesla 10de:06e4 GPU with dmesg flooding:
(kernel cmdline ...ipv6.disable=1 net.ifnames=0 noresume consoleblank=0 mitigations=off drm.debug=0x1e log_bug_len=2M)
# dmesg
...
[  437.498754] nouveau 0000:0b:00.0: DRM: base-0 cleanup: 0000000058e43383
[  437.498762] nouveau 0000:0b:00.0: DRM: curs-0 cleanup: 0000000000000000
[  437.498770] nouveau 0000:0b:00.0: [drm:drm_atomic_state_default_clear] Clearing atomic state 0000000051b08724
[  437.498793] nouveau 0000:0b:00.0: [drm:__drm_atomic_state_free] Freeing atomic state 0000000051b08724
[  437.498842] nouveau 0000:0b:00.0: [drm:drm_atomic_state_init] Allocated atomic state 000000009deb8fe9
[  437.498854] nouveau 0000:0b:00.0: [drm:drm_atomic_get_plane_state] Added [PLANE:57:curs-1] 0000000053232996 state to 000000009deb8fe9
[  437.498868] nouveau 0000:0b:00.0: [drm:drm_atomic_set_fb_for_plane] Set [NOFB] for [PLANE:57:curs-1] state 0000000053232996
[  437.498877] nouveau 0000:0b:00.0: [drm:drm_atomic_check_only] checking 000000009deb8fe9
[  437.498891] nouveau 0000:0b:00.0: DRM: curs-1 atomic_check
[  437.498899] nouveau 0000:0b:00.0: [drm:drm_atomic_commit] committing 000000009deb8fe9
[  437.498911] nouveau 0000:0b:00.0: DRM: curs-1 prepare: 0000000000000000
[  437.498920] nouveau 0000:0b:00.0: DRM: commit 0 0
[  437.498927] nouveau 0000:0b:00.0: DRM: curs-1: clr 00 (set 00)
[  437.498934] nouveau 0000:0b:00.0: DRM: curs-1: set 00 (clr 00)
[  437.498941] nouveau 0000:0b:00.0: DRM: curs-1 cleanup: 0000000000000000
[  437.498948] nouveau 0000:0b:00.0: [drm:drm_atomic_state_default_clear] Clearing atomic state 000000009deb8fe9
[  437.498967] nouveau 0000:0b:00.0: [drm:__drm_atomic_state_free] Freeing atomic state 000000009deb8fe9
[  437.526894] nouveau 0000:0b:00.0: DRM: [Xorg.bin[807]/00000003:fenceNonStallIntr] [ALLOW]
[  437.526922] nouveau 0000:0b:00.0: DRM: [Xorg.bin[807]/00000003:fenceNonStallIntr] [BLOCK]
[  437.528031] nouveau 0000:0b:00.0: DRM: [Xorg.bin[807]/00000003:fenceNonStallIntr] [ALLOW]
[  437.528534] nouveau 0000:0b:00.0: DRM: [Xorg.bin[807]/00000003:fenceNonStallIntr] [BLOCK]
[  437.552990] nouveau 0000:0b:00.0: DRM: [Xorg.bin[807]/00000006:fenceNonStallIntr] [ALLOW]
[  437.553022] nouveau 0000:0b:00.0: DRM: [Xorg.bin[807]/00000006:fenceNonStallIntr] [BLOCK]
...