Currently, this test: Is skipped on Mac Passes ~100% on Linux Fails ~100% on Win The chromium baseline is the mac baseline + a console error message. Win is failing because it matches the upstream expectation, aka, it does not print out an error. Linux passes because it does have the error. Matching upstream expectation and updating test_expectations for new failures seems correct, rather than adding a linux baseline with the error message.
FWIW, I'm trying to switch to using the webkit.org test plugin. I've already made the switch for chromium DRT mac, so it would be good to know if this test passes with chromium DRT mac.
Created attachment 71364 [details] Patch
Comment on attachment 71364 [details] Patch Not sure what happened with ChangeLog there. Fixing.
Created attachment 71368 [details] Patch
I don't have a Mac, but I could try unskipping in test_expectations and see what happens? I don't see a good reason for this to be skipped anyway.
Comment on attachment 71368 [details] Patch Deleting the wrong baseline seems fine. It would be good to try to unskip this test on Mac, but that can be a separate change.
The commit-queue encountered the following flaky tests while processing attachment 71368 [details]: java/lc3/JSObject/ToObject-001.html Please file bugs against the tests. The author(s) of the test(s) are ap@webkit.org and ap@webkit.org. The commit-queue is continuing to process your patch.
Comment on attachment 71368 [details] Patch Clearing flags on attachment: 71368 Committed r70271: <http://trac.webkit.org/changeset/70271>
All reviewed patches have been landed. Closing bug.