RESOLVED FIXED 179085
updateMaskedAncestorShouldIsolateBlending() should check the Nullability of the computedStyle() of the element's ancestors
https://bugs.webkit.org/show_bug.cgi?id=179085
Summary updateMaskedAncestorShouldIsolateBlending() should check the Nullability of t...
Said Abou-Hallawa
Reported 2017-10-31 16:20:55 PDT
This should not happen unless recalcStyle() is called in the middle of changing the DOM tree and the render tree has not been updated yet. In this case, a renderer will be associated with an element which is not connected to the DOM tree.
Attachments
Patch (2.20 KB, patch)
2017-10-31 16:23 PDT, Said Abou-Hallawa
no flags
Archive of layout-test-results from ews107 for mac-elcapitan-wk2 (1.14 MB, application/zip)
2017-10-31 17:37 PDT, Build Bot
no flags
Patch (2.22 KB, patch)
2017-10-31 18:08 PDT, Said Abou-Hallawa
no flags
Said Abou-Hallawa
Comment 1 2017-10-31 16:22:02 PDT
Said Abou-Hallawa
Comment 2 2017-10-31 16:23:49 PDT
Ryosuke Niwa
Comment 3 2017-10-31 17:30:18 PDT
Comment on attachment 325518 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=325518&action=review > Source/WebCore/ChangeLog:11 > + while the DOM tree is being changed an event fires and this event causes "DOM tree is being changed an event fires"? You mean like "a script runs in response to a DOM mutation"?
Build Bot
Comment 4 2017-10-31 17:37:03 PDT
Comment on attachment 325518 [details] Patch Attachment 325518 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.webkit.org/results/5057128 New failing tests: imported/w3c/web-platform-tests/service-workers/cache-storage/serviceworker/cache-match.https.html
Build Bot
Comment 5 2017-10-31 17:37:05 PDT
Created attachment 325528 [details] Archive of layout-test-results from ews107 for mac-elcapitan-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews107 Port: mac-elcapitan-wk2 Platform: Mac OS X 10.11.6
Said Abou-Hallawa
Comment 6 2017-10-31 18:08:30 PDT
Said Abou-Hallawa
Comment 7 2017-10-31 18:19:27 PDT
Comment on attachment 325518 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=325518&action=review >> Source/WebCore/ChangeLog:11 >> + while the DOM tree is being changed an event fires and this event causes > > "DOM tree is being changed an event fires"? > You mean like "a script runs in response to a DOM mutation"? Not necessarily. For example while waiting for a response from the network process, WebKit may decide to process another message from the UI process. Processing the UI message will cause an event to fire. In this case styleReclac()/layout() may happen. I changed the ChangeLog to make the wording more generic.
WebKit Commit Bot
Comment 8 2017-10-31 18:39:12 PDT
Comment on attachment 325531 [details] Patch Clearing flags on attachment: 325531 Committed r224269: <https://trac.webkit.org/changeset/224269>
WebKit Commit Bot
Comment 9 2017-10-31 18:39:14 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.