fast/images/animated-gif-iframe-webkit-transform.html is flaky - mostly on debug bots, but also on release sometimes. -PASS isFirstImagePaused() is false +FAIL isFirstImagePaused() should be false. Was true. https://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&revision=179444&showExpectations=true&tests=fast%2Fimages%2Fanimated-gif-iframe-webkit-transform.html
This is a fairly new test. Chris, do you know what could be wrong here? I couldn't reproduce locally, even with many parallel processes.
I wrote this test. I'll take a look tomorrow. Like other gif animations test relying on internals.hasPausedImageAnimations(), it doesn't work (reliably) on WK1. We already have a section in wk1 TestsExpectations to skip those. However, on WK2 it should work so I'll have to investigate. My initial feeling is that we'll like need to use shouldBecomeEqual() for isFirstImagePaused(), like we already do for isSecondImagePaused().
Created attachment 246222 [details] Patch
Comment on attachment 246222 [details] Patch rs=me
Comment on attachment 246222 [details] Patch Clearing flags on attachment: 246222 Committed r179787: <http://trac.webkit.org/changeset/179787>
All reviewed patches have been landed. Closing bug.
Thanks, this fixed it.