Intermittent browser/base/content/test/performance/browser_tabswitch.js | unexpected changed rect: ({x1:417, x2:472, y1:0, y2:32, w:56, h:33}), window width: 1280 -
Categories
(Firefox :: General, defect, P5)
Tracking
()
People
(Reporter: intermittent-bug-filer, Unassigned)
Details
(Keywords: intermittent-failure)
| Comment hidden (Intermittent Failures Robot) |
Comment 2•7 years ago
|
||
Comment 3•7 years ago
|
||
Fail reappeared on autoland.
Log link: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=221816991&repo=autoland&lineNumber=5692
| Comment hidden (Intermittent Failures Robot) |
Comment 5•7 years ago
|
||
:florian, we have 15 open bugs against this testcase:
https://bugzilla.mozilla.org/buglist.cgi?quicksearch=browser_tabswitch.js&list_id=14525157
could we make this a single bug, or does it make sense to unique bugs for the different values in the rect.
Comment 6•7 years ago
|
||
(In reply to Narcis Beleuzu [:NarcisB] from comment #3)
Fail reappeared on autoland.
Log link: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=221816991&repo=autoland&lineNumber=5692
This is showing the second tab that's still animating. I don't expect that within a tab switch test.
(In reply to Joel Maher ( :jmaher ) (UTC-4) from comment #5)
:florian, we have 15 open bugs against this testcase:
https://bugzilla.mozilla.org/buglist.cgi?quicksearch=browser_tabswitch.js&list_id=14525157could we make this a single bug, or does it make sense to unique bugs for the different values in the rect.
Bug 1455054 is a different issue.
Bug 1465574 is similar to bug 1502255.
All the other bugs with rects of height 33px or 10x10px or 14x14px rects look like the second tab is still animating when we attempt to begin the tabswitch test.
I think at https://searchfox.org/mozilla-central/rev/c035ee7d3a5cd6913e7143e1bce549ffb4a566ff/browser/base/content/test/performance/browser_tabswitch.js#37 we should either do something to skip the animation, or await something to ensure the animation is finished before proceeding.
Felipe, do you have a recommendation? Is this a regression from your recent tab animation work, or was this before your work landed? This bug was filed 2 months, but the failures in all the other similar bugs seem to have spiked this week.
Comment 7•7 years ago
|
||
Not a regression from my work because it got backed out.
Since this is a reflow test, I don't think we should skip the animation. We can just wait for the transitionend event, like other tests do. e.g.: https://searchfox.org/mozilla-central/rev/465dbfe030dfec7756b9b523029e90d48dd5ecce/browser/base/content/test/performance/browser_tabclose.js#39-40
Once the groundwork for the new tab animations land again, I'll be replacing these listeners for transitionend to listeners for a new event, TabAnimationEnd
| Comment hidden (Intermittent Failures Robot) |
| Comment hidden (Intermittent Failures Robot) |
| Comment hidden (Intermittent Failures Robot) |
Updated•6 years ago
|
Description
•