Created attachment 144813 [details] xorg 0 log After I switched from modesetting to the intel driver (due to bug#111105), I see a lot of video stuttering. That is, sometimes the video jumps back and forth. Only when played fullscreen, it seems. mpv disables compositing in that case, I think. And not only the video. This firefox window on another screeen shows the same symptom while I am typing -- letters dismiss and appear again. Well it still can be bug#111105 after all -- showing an old buffer on the screen under some circumstances. kernel-default-5.2.1-2.1.gbf5c01b.x86_64 mpv-0.29.1-28.9.x86_64 plasma5-desktop-5.16.2-1.2.x86_64 xf86-video-intel-2.99.917+git8674.25c9a2fcc-3.4.x86_64
Looks like dri3/present is active? Such a stutter is that it is showing a backbuffer, normally that front/back is confused between client&server. If it is using Present (as it appears), please note that 1.20.5 contains a regression 87491ced98705a14be2bb5a3ce7e41cb5ffc11d0. Fwiw, you can use Option "DRI" "2" to quickly test that supposition.
Having run through and corrected all the false assertions after the client does a double SwapBuffers, how bad is it now? That the client is misbehaving quite frequently does not bode well for a stutter-free experience. We could try removing the delay after a skip?
(In reply to Chris Wilson from comment #2) > Having run through and corrected all the false assertions after the client > does a double SwapBuffers, how bad is it now? That the client is misbehaving > quite frequently does not bode well for a stutter-free experience. We could > try removing the delay after a skip? So I don't see any asserts being hit, but the video still stutters. Also, when playing the video, the whole content of this firefox window (on another screen) dismisses. And sometimes it appears back. Or I have to switch tabs to redraw it. This is with DRI 2.
Also, when plasma is started, the splashscreen (the bulb on openSUSE) does not disappear on secondary monitor. I have to turn composition off and on to get rid of it. Switching desktops does not help either. This does not happen with modesetting.
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-intel/issues/165.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.