Window Occlusion Notifications on Mac mark the WebKitTestRunner's WebView as not visible as soon as created. On Mac this enables JavaScript timer throttling and hence causes the following layouts tests to fail. fast/dom/timer-increase-min-interval-and-reset-part-2.html fast/dom/timer-increase-then-decrease-min-interval-repeating.html fast/dom/timer-increase-then-decrease-min-interval.html fast/dom/timer-throttling-hidden-page.html This issue was initially fixed by just setting the page visibility to "visible" after creating the WebView in https://bugs.webkit.org/show_bug.cgi?id=111025. But, it is not very robust, since the view still thinks it is invisible and this might cause other tests to fail. Hence, using the private API being added by https://bugs.webkit.org/show_bug.cgi?id=111107, to disable window occlusion detection entirely for the WebKitTestRunner's web view is a better solution.
<rdar://problem/13305561>
Created attachment 190862 [details] Patch
Comment on attachment 190862 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=190862&action=review > Tools/ChangeLog:11 > + Window occlusion notifications on Mac cause the WebKitTestRunner's > + Web View to be treated as invisible and this automatically enables > + JavaScript timer throttling, causing some layout tests exercising > + JavaScript timers to fail. Hence, disable window occlusion detection. This doesn't seem to match the patch (yes, I know it's a follow-up but should not be copied from the other bug).
Created attachment 190869 [details] Patch
(In reply to comment #3) > (From update of attachment 190862 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=190862&action=review > > > Tools/ChangeLog:11 > > + Window occlusion notifications on Mac cause the WebKitTestRunner's > > + Web View to be treated as invisible and this automatically enables > > + JavaScript timer throttling, causing some layout tests exercising > > + JavaScript timers to fail. Hence, disable window occlusion detection. > > This doesn't seem to match the patch (yes, I know it's a follow-up but should not be copied from the other bug). You're right, I wrote the change log as if I was fixing the tests, while they have already been fixed. Corrected the ChangeLog.
Comment on attachment 190869 [details] Patch Clearing flags on attachment: 190869 Committed r144415: <http://trac.webkit.org/changeset/144415>
All reviewed patches have been landed. Closing bug.