Bug 444166

Summary:	Fall Apart Effect triggered too often on Wayland
Product:	[Plasma] kwin	Reporter:	Dashon <dashonwwIII>
Component:	effects-various	Assignee:	KWin default assignee <kwin-bugs-null>
Status:	RESOLVED FIXED
Severity:	normal	CC:	dashonwwIII, nate
Priority:	LO
Version First Reported In:	5.23.1
Target Milestone:	---
Platform:	Other
OS:	Linux
See Also:	https://bugs.kde.org/show_bug.cgi?id=498848
Latest Commit:		Version Fixed/Implemented In:
Sentry Crash Report:

Description Dashon 2021-10-21 04:06:31 UTC

SUMMARY
It's hard to describe without a video. I did make a screen recording that is about 9.3MB which is too large to upload to bugs.kde.org. Do let me know of a platform to upload to if it helps. Anyways the effect seems to be triggered by every little thing. Opening and closing and yakuake on Xorg used to slide up and down like the panel applets do, but on wayland it uses the fall apart effect. Logging in and out seems to trigger the fall apart effect on wayland as well. Snapping a window to a screen edge triggers the fall apart effect as well. It just seems to be triggered a lot more on wayland than it is on Xorg.

STEPS TO REPRODUCE
1. Enable "Fall Apart" in Desktop Effects
2. Switch to a wayland session.
3. I know this isn't usually a step to follow, but try using the desktop normally for a while with the effect enabled.

OBSERVED RESULT
If you follow those three steps, you should notice the effect being triggered a lot more than it would on Xorg.

EXPECTED RESULT
It should behave the same way it does on Xorg which seems to only trigger it on the closing of normal windows and maybe a few others as far as I can tell.

Operating System: EndeavourOS
KDE Plasma Version: 5.23.1
KDE Frameworks Version: 5.87.0
Qt Version: 5.15.2
Kernel Version: 5.14.12-zen1-1-zen (64-bit)
Graphics Platform: Wayland
Processors: 4 × Intel® Core™ i7-7500U CPU @ 2.70GHz
Memory: 15.4 GiB of RAM
Graphics Processor: Mesa Intel® HD Graphics 620

Comment 1 Vlad Zahorodnii 2021-10-21 07:36:04 UTC

X11 and Wayland are different in general. In particular, how some windows are identified and window types, etc.

Comment 2 Nate Graham 2021-10-21 15:00:38 UTC

FWIW the effect looks fine to me on Wayland.

Comment 3 Dashon 2021-10-21 15:07:14 UTC

(In reply to Nate Graham from comment #2)
> FWIW the effect looks fine to me on Wayland.

It looks and works just fine, but my concern was that it was triggered far more often in wayland than on xorg. I tried to list a few scenarios in an x11 session in which the effect is not triggered. For example, when you drag the window to a screen edge to snap it. Just to name a a prominent one. Vlad has explained that the differences are caused because of the differences between window types and how they are identified in wayland and x11. I suspect that there are bunch of windows as well as window states that now fall into the category that triggers the "Fall Apart" effect.

Comment 4 Dashon 2021-10-21 15:09:19 UTC

(In reply to Nate Graham from comment #2)
> FWIW the effect looks fine to me on Wayland.

Does it not trigger much more often for you. Such as the drag to screen edge to snap, opening and closing yakuake, logging in and out, etc?

Comment 5 Nate Graham 2021-10-21 15:12:43 UTC

Yes, I do see the effect when I unlock the screen. I don't see it when I tile a window though.

Comment 6 Dashon 2021-10-21 15:16:00 UTC

(In reply to Nate Graham from comment #5)
> Yes, I do see the effect when I unlock the screen. I don't see it when I
> tile a window though.

With the window tiling, it only happens when I do it with the mouse. I have to drag it to a screen edge and tile it. It doesn't happen if I snap it to a screen edge using the keyboard. Does it not occur in either scenario for you?

Comment 7 Nate Graham 2021-10-21 15:21:50 UTC

For me, in neither case.

Comment 8 Dashon 2021-10-21 15:23:31 UTC

(In reply to Nate Graham from comment #7)
> For me, in neither case.

Alright, I have a kde neon unstable vm. I'll update it and try and reproduce in that.

Comment 9 Dashon 2021-10-21 15:40:52 UTC

(In reply to Nate Graham from comment #7)
> For me, in neither case.

I managed to reproduce it in a kde neon unstable virtual machine. I also recorded a video if that helps. Just let me know of a suitable hosting platform if that helps.

Comment 10 Dashon 2024-02-02 23:51:46 UTC

Looks like this has been fixed in plasma 6. None of the conditions that triggered this back when I reported it are still a problem.