Bug 227367

Summary: REGRESSION (r278746): [Mac wk1] media/media-fullscreen-return-to-inline.html is a flaky timeout
Product: WebKit Reporter: Eric Hutchison <ehutchison>
Component: MediaAssignee: Chris Dumez <cdumez>
Status: REOPENED    
Severity: Normal CC: cdumez, ehutchison, peng.liu6, tsavell, webkit-bot-watchers-bugzilla, webkit-bug-importer, wenson_hsieh
Priority: P2 Keywords: InRadar
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
See Also: https://bugs.webkit.org/show_bug.cgi?id=226899
https://bugs.webkit.org/show_bug.cgi?id=226701
https://bugs.webkit.org/show_bug.cgi?id=226734
https://bugs.webkit.org/show_bug.cgi?id=226672
https://bugs.webkit.org/show_bug.cgi?id=157912
Attachments:
Description Flags
Expectations none

Eric Hutchison
Reported 2021-06-24 11:19:58 PDT
media/media-fullscreen-return-to-inline.html flaky timeout on Mac wk1 queues. It appears timeout began around r278753. History: https://results.webkit.org/?suite=layout-tests&suite=layout-tests&test=media%2Fmedia-fullscreen-return-to-inline.html&test=media%2Fmedia-fullscreen-return-to-inline.html
Attachments
Expectations (1.58 KB, patch)
2021-06-25 16:06 PDT, Eric Hutchison 		no flags
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Radar WebKit Bug Importer
Comment 1 2021-06-24 11:20:32 PDT
<rdar://problem/79733822>
Eric Hutchison
Comment 2 2021-06-25 13:43:07 PDT
Was able to bisect issue to available builds 278743 to 278751.
Truitt Savell
Comment 3 2021-06-25 13:45:06 PDT
Looking at the revisions in this range, it seems like either r278746 or r278747 could have caused this. Chris or Wenson could you see if one of your changes could be responsible for this regression?
Wenson Hsieh
Comment 4 2021-06-25 13:47:19 PDT
(In reply to Truitt Savell from comment #3) > Looking at the revisions in this range, it seems like either r278746 or > r278747 could have caused this. > > Chris or Wenson could you see if one of your changes could be responsible > for this regression? It's definitely not r278747, since the codepaths that were changed there aren't compiled on macOS earlier than Monterey.
Alexey Proskuryakov
Comment 5 2021-06-25 15:30:52 PDT
Re-titled based on above comments.
Eric Hutchison
Comment 6 2021-06-25 16:06:32 PDT
Created attachment 432305 [details] Expectations
EWS
Comment 7 2021-06-25 17:24:05 PDT
Committed r279302 (239178@main): <https://commits.webkit.org/239178@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 432305 [details].
Ryan Haddad
Comment 8 2021-06-25 17:46:52 PDT
Reopening since the patch only updated test expectations.
Chris Dumez
Comment 9 2021-06-28 08:50:33 PDT
--- /Volumes/Data/worker/catalina-release-tests-wk1/build/layout-test-results/media/media-fullscreen-return-to-inline-expected.txt +++ /Volumes/Data/worker/catalina-release-tests-wk1/build/layout-test-results/media/media-fullscreen-return-to-inline-actual.txt @@ -1,3 +1,4 @@ +FAIL: Timed out waiting for notifyDone to be called Test that a <video> pauses after exiting fullscreen when the internal property "allowsInlineMediaPlaybackAfterFullscreen" is false. @@ -11,15 +12,6 @@ EVENT(webkitpresentationmodechanged) EXPECTED (internals.isChangingPresentationMode(video) == 'false') OK EXPECTED (video.webkitDisplayingFullscreen == 'false') OK -EXPECTED (video.paused == 'true') OK +EXPECTED (video.paused == 'true'), OBSERVED 'false' FAIL RUN(video.play()) -EVENT(webkitpresentationmodechanged) -EXPECTED (internals.isChangingPresentationMode(video) == 'false') OK -EXPECTED (video.webkitDisplayingFullscreen == 'true') OK -RUN(video.webkitExitFullscreen()) -EVENT(webkitpresentationmodechanged) -EXPECTED (internals.isChangingPresentationMode(video) == 'false') OK -EXPECTED (video.webkitDisplayingFullscreen == 'false') OK -EXPECTED (video.paused == 'false') OK -END OF TEST
Chris Dumez
Comment 10 2021-06-28 09:51:22 PDT
Sadly, I am having trouble reproducing. I tried both release and debug builds with WK1. I tried running tests sequentially or in parallel.
Note You need to log in before you can comment on or make changes to this bug.