[Wayland] Use timeout for frame callbacks for basic compositor
Categories
(Core :: Widget: Gtk, defect, P3)
Tracking
()
Tracking | Status | |
---|---|---|
firefox83 | --- | fixed |
People
(Reporter: stransky, Assigned: stransky)
References
(Blocks 1 open bug)
Details
Attachments
(1 file)
When basic compositor commits, a frame callback is generated and next commit is performed in the frame callback. But when the frame callback is not triggered by compositor (buggy compositor or missing event or so), we fail to draw the cached buffer.
Assignee | ||
Comment 1•4 years ago
|
||
Pushed by nerli@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/ca0b10a6ebb8 [Wayland] Use timeout for frame callbacks for basic compositor, r=jhorak
Comment 3•4 years ago
|
||
I'd like to highlight that such a timeout works around bugs in the compositor at the cost of higher energy consumption. I.e. the compositor should not send any frame callbacks if the window is hidden - potentially for minutes or hours.
So I hope we don't need to introduce something like this when using WR when we switch to frame callbacks there.
Assignee | ||
Comment 4•4 years ago
|
||
(In reply to Robert Mader [:rmader] from comment #3)
I'd like to highlight that such a timeout works around bugs in the compositor at the cost of higher energy consumption. I.e. the compositor should not send any frame callbacks if the window is hidden - potentially for minutes or hours.
The patch here does not trigger the frame callback/drawing. It just remove the pending one. So if the drawing happens (after some delay), it will happen immediately. Also it's used for basic compositor only.
We can put it under preference if you like but it should solve major usability issues (Bug 1653850) right now .
Comment 5•4 years ago
|
||
The patch here does not trigger the frame callback/drawing. It just remove the pending one. So if the drawing happens (after some delay), it will happen immediately.
Oh, I misunderstood that. Never mind, then :)
Comment 6•4 years ago
|
||
bugherder |
Assignee | ||
Comment 7•4 years ago
|
||
We need to revert this one as it causes https://bugzilla.redhat.com/show_bug.cgi?id=1888920
Description
•