This is an automatically generated bug from the commit-queue. svg/animations/smil-leak-elements.svg has been flaky on the commit-queue. svg/animations/smil-leak-elements.svg was authored by fmalita@chromium.org and timothy_horton@apple.com. http://trac.webkit.org/browser/trunk/LayoutTests/svg/animations/smil-leak-elements.svg The commit-queue just saw svg/animations/smil-leak-elements.svg flake (text diff) while processing attachment 197034 [details] on bug 85934. Bot: webkit-cq-02 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.8.3 The bots will update this with information from each new failure. If you believe this bug to be fixed or invalid, feel free to close. The bots will re-open if the flake re-occurs. If you would like to track this test fix with another bug, please close this bug as a duplicate. The bots will follow the duplicate chain when making future comments.
Created attachment 197070 [details] Archive of layout-test-results from webkit-cq-02
The commit-queue just saw svg/animations/smil-leak-elements.svg flake (text diff) while processing attachment 198387 [details] on bug 108980. Bot: webkit-cq-02 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.8.3
Created attachment 198394 [details] Archive of layout-test-results from webkit-cq-02
The commit-queue just saw svg/animations/smil-leak-elements.svg flake (text diff) while processing attachment 198850 [details] on bug 109588. Bot: webkit-cq-03 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.8.3
Created attachment 198887 [details] Archive of layout-test-results from webkit-cq-03
The commit-queue just saw svg/animations/smil-leak-elements.svg flake (text diff) while processing attachment 199517 [details] on bug 114218. Bot: webkit-cq-03 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.8.3
Created attachment 199605 [details] Archive of layout-test-results from webkit-cq-03
The commit-queue just saw svg/animations/smil-leak-elements.svg flake (text diff) while processing attachment 200427 [details] on bug 114742. Bot: webkit-cq-02 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.8.3
Created attachment 200678 [details] Archive of layout-test-results from webkit-cq-02
The commit-queue just saw svg/animations/smil-leak-elements.svg flake (text diff) while processing attachment 202536 [details] on bug 116605. Bot: webkit-cq-01 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.8.3
Created attachment 202540 [details] Archive of layout-test-results from webkit-cq-01
*** Bug 114281 has been marked as a duplicate of this bug. ***
Created attachment 211471 [details] speculative fix
Comment on attachment 211471 [details] speculative fix Seems reasonable to me.
Committed <http://trac.webkit.org/r155648>.
This didn't help.
Increased the number of nodes in the tests 10x in <http://trac.webkit.org/r155728> to see what happens (if the number of leaked nodes grows accordingly, we're more likely to have a real bug).
Created attachment 211808 [details] another speculative fix
Comment on attachment 211808 [details] another speculative fix r=me
Committed <http://trac.webkit.org/r155882>.
And that didn't help either. Changed timeouts from 0 to 100 ms as suggested by Geoff on IRC: <http://trac.webkit.org/r155890>.
Still failing, no change at all (seeing between 4 and 14 leaked nodes).
Marked as flaky in <http://trac.webkit.org/r155897>. This seems more like a real WebKit bug now, the tests are doing everything possible to perform a complete garbage collection.
> Marked as flaky in <http://trac.webkit.org/r155897>. I really don't like the "flaky" as a solution to a failing test. It's net negative information. If a bot watcher or engineer ever sets aside productive work to diagnose a red bot because of these broken tests, the tests will have been a net detriment to the project. It would be better not to run these tests, or to run them to test for crashes, and completely ignore other output. Another better option would be to give the tests a tolerance for some percentage of surviving Nodes. Then, they could still test for a traditional storage leak, which would exceed the tolerance.
> or to run them to test for crashes, and completely ignore other output. That's pretty much exactly what marking as flaky does - the test won't make bots red unless it starts to crash. Plus flakiness console may one day tell us that the underlying bug disappeared, so we can remove the entries from TestExpectations.
Committed r181401: <http://trac.webkit.org/changeset/181401>
Change http://trac.webkit.org/changeset/181401 marks these test to be flaky on all ports.
The commit-queue just saw svg/animations/smil-leak-dynamically-added-element-instances.svg flake (text diff) while processing attachment 316629 [details] on bug 173419. Bot: webkit-cq-03 Port: <class 'webkitpy.common.config.ports.MacPort'> Platform: Mac OS X 10.11.6
Created attachment 316631 [details] Archive of layout-test-results from webkit-cq-03
Created attachment 460693 [details] Patch
*** Bug 175867 has been marked as a duplicate of this bug. ***
*** Bug 175886 has been marked as a duplicate of this bug. ***
*** Bug 174180 has been marked as a duplicate of this bug. ***
Created attachment 460697 [details] Patch
Created attachment 460703 [details] Patch
Committed 252260@main (56cbe674213b): <https://commits.webkit.org/252260@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 460703 [details].
<rdar://problem/96648276>