Weird stuttering with a second monitor

Elizabeth · 2026-09-10 21:32:29

seth wrote:

You could try to confirm this w/ https://wiki.archlinux.org/title/Sway#Tearing

Ill do this tmrw cause again its a bit late here, but i wanted to ask what do you mean by this

seth wrote:

and oc. try to bisect the kernel [...]

Do you mean like I could try reading the code to see whats wrong?

seth · 2026-09-10 23:07:21

https://wiki.archlinux.org/title/Bisect … s_with_Git

But see https://bbs.archlinux.org/viewtopic.php … 9#p2309389
loqs can hand you a kernel, you check and report its performance and then you get a new one

Elizabeth · 2026-09-11 14:35:31

seth wrote:

You could try to confirm this w/ https://wiki.archlinux.org/title/Sway#Tearing

After doing this I didnt notice any tearing, the stuttering remained the exact same.

seth · 2026-09-11 14:43:34

Which output did you try?
Did you try "output * allow_tearing yes" ?

Elizabeth · 2026-09-11 15:05:32

I tried both "output * allow_tearing yes" and "output DP-3 allow_tearing yes" (DP-3 being the external monitor ive been reporting stuttering on)
Also im working on bisecting rn and ive found at least one version where the tearing in xorg appears despite the stuttering in wayland being gone

Elizabeth · 2026-09-11 15:45:14

Posting this a separate reply because it feels significant enough
Ive done the bisecting and these are the results:

7.0.0-1-mainline
- Wayland (Stutter) (niri/weston) - no/no
- Xorg (Tearing) (i3) - yes

7.1.0-1-mainline
- Wayland (Stutter) (niri/weston) - no/no
- Xorg (Tearing) (i3) - absent but intermittently shows up???

7.1.0-rc5-1-mainline
- Wayland (Stutter) (niri/weston) - no/no
- Xorg (Tearing) (i3) - no

7.1.0-rc7-1-mainline
- Wayland (Stutter) (niri/weston) - no/no
- Xorg (Tearing) (i3) - yes, and it wouldnt go away no matter how long I looked

7.2.0-1-mainline
- Wayland (Stutter) (niri/weston) - yes/yes
- Xorg (Tearing) (i3) - at first yes then it dissapears?

So it seems that 7.2 introduced the stutter. As for Xorg, I just dont know. There doesn't seem to be any pattern and it might just be totally unrelated and/or a consequence of me doing a bad installation of it. I really just dont know.
Kernels are taken from https://pkgbuild.com/~gromit/linux-bisection-kernels/

Edit: just for clarity's sake the order of the results here is not the order of testing, I just found it to be prettier to present the kernels sequentially rather than in testing order

Last edited by Elizabeth (2026-09-11 15:49:04)

seth · 2026-09-12 19:17:57

Does https://pkgbuild.com//~gromit/linux-bis … kg.tar.zst already have the issue?
Ultimately however, unless https://gitlab.freedesktop.org/drm/amdgpu can identify the likely cause, it needs to be properly bisected.

Elizabeth · 2026-09-12 20:49:45

seth wrote:

Does https://pkgbuild.com//~gromit/linux-bis … kg.tar.zst already have the issue?
Ultimately however, unless https://gitlab.freedesktop.org/drm/amdgpu can identify the likely cause, it needs to be properly bisected.

Im confused, isnt the kernel you linked newer than 7.2.0?
Also I tried looking into how to report this issue on the gitlab site but frankly I have no idea how this works. Theres talk of mailing lists and stuff that I just do not get.

seth · 2026-09-13 06:36:24

No, that's a release candidate - look at the timestamps

linux-mainline-7.2-1-x86_64.pkg.tar.zst            17-Aug-2026 08:09           155118076
linux-mainline-7.2rc1-1-x86_64.pkg.tar.zst         29-Jun-2026 08:18           154953675

https://gitlab.freedesktop.org/drm/amd/-/work_items/
But you'll have to register a gitlab account there to file anything (should™ work better than the archlinux gitlab, though)

Elizabeth · 2026-09-13 10:28:41

seth wrote:

No, that's a release candidate - look at the timestamps
linux-mainline-7.2-1-x86_64.pkg.tar.zst            17-Aug-2026 08:09           155118076
linux-mainline-7.2rc1-1-x86_64.pkg.tar.zst         29-Jun-2026 08:18           154953675
https://gitlab.freedesktop.org/drm/amd/-/work_items/
But you'll have to register a gitlab account there to file anything (should™ work better than the archlinux gitlab, though)

Ah okay, ill try to go trough the release candidates today then, also thanks for linking the repo so directly.

EDIT:
7.2.0-rc1-1-mainline
- Wayland (Stutter) (niri/weston) - yes/yes

So it seems that even the first rc had introduced it.

Last edited by Elizabeth (2026-09-13 10:34:37)

Elizabeth · Today 07:05:42

Posting this reply separately due to its significance,
I have opened an issue at https://gitlab.freedesktop.org/drm/amd/ … items/5839, and the discussion will be moved there.

I'm not sure whether to mark this as solved yet or to wait for a proper patch to come out.

Arch Linux

#26 2026-09-10 21:32:29

Re: Weird stuttering with a second monitor

#27 2026-09-10 23:07:21

Re: Weird stuttering with a second monitor

#28 2026-09-11 14:35:31

Re: Weird stuttering with a second monitor

#29 2026-09-11 14:43:34

Re: Weird stuttering with a second monitor

#30 2026-09-11 15:05:32

Re: Weird stuttering with a second monitor

#31 2026-09-11 15:45:14

Re: Weird stuttering with a second monitor

#32 2026-09-12 19:17:57

Re: Weird stuttering with a second monitor

#33 2026-09-12 20:49:45

Re: Weird stuttering with a second monitor

#34 2026-09-13 06:36:24

Re: Weird stuttering with a second monitor

#35 2026-09-13 10:28:41

Re: Weird stuttering with a second monitor

#36 Today 07:05:42

Re: Weird stuttering with a second monitor

Board footer