if two repaint requests get issued on the same frame with async zoom the second will be completely dropped
Categories
(Core :: Panning and Zooming, defect)
Tracking
()
Tracking | Status | |
---|---|---|
firefox90 | --- | fixed |
People
(Reporter: tnikkel, Assigned: tnikkel)
References
Details
Attachments
(2 files, 1 obsolete file)
This is needed to be able to write a decent test for bug 1710050.
Assignee | ||
Comment 1•3 years ago
|
||
During double tap animations I was seeing two repaint requests per frame. The first one had scroll update type eNone, the second has type eUserAction and there was a layout scroll position update. Both had the same !=1 async zoom. When the callback helper gets the first it updates the presshell resolution. Then when it gets the second the presshell resolution no longer matches and it returns ignoring the repaint request entirely. When it's only clear that it should ignore the resolution part.
Updated•3 years ago
|
Assignee | ||
Comment 2•3 years ago
|
||
With the patch for this bug we need to modify an existing test.
I haven't investigated this, but I think the scrollY value is always between 0 and 2 and I guess the patch makes it move around a bit more non-monotonically. Not sure it's worth spending time on, but I can if you think otherwise.
Updated•3 years ago
|
Updated•3 years ago
|
Assignee | ||
Comment 3•3 years ago
|
||
We need to further modify this test. This time the resolution.
We calculate the new presshell resolution as (async zoom) * (last known by apz presshell resolution).
Before the patch of this bug (last known by apz presshell resolution) had to be (almost) equal to the actual presshell resolution. After the patch of this bug (last known by apz presshell resolution) can differ, allowing more floating point error to come in.
Updated•3 years ago
|
Comment 5•3 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/187c9f7c4653
https://hg.mozilla.org/mozilla-central/rev/d3b02e2c82d9
Description
•