http://trac.webkit.org/changeset/57081 broke the build: We think it triggered two tests to fail on Tiger because of race conditions (Requested by abarth on #webkit). This is an automatic bug report generated by the sheriff-bot. If this bug report was created because of a flaky test, please file a bug for the flaky test (if we don't already have one on file) and dup this bug against that bug so that we can track how often these flaky tests case pain. "Only you can prevent forest fires." -- Smokey the Bear
Created attachment 52584 [details] ROLLOUT of r57081
Comment on attachment 52584 [details] ROLLOUT of r57081 Clearing flags on attachment: 52584 Committed r57100: <http://trac.webkit.org/changeset/57100>
All reviewed patches have been landed. Closing bug.
This also affected Qt. Ossy reproduced the failure locally and confirmed that the rollout caused the failure to go away. I've not yet been able to reproduce the failure on Leopard. I don't have access to a Tiger machine.
Our understanding of this failure is that this regression made it possible for load events to fire on the outermost document, before the inner <object> elements had a chance to schedule their loads. <object> load scheduling is crazy. it happens in the renderers via updateWidget which is scheduled from attach().
Looks like this rollout fixed both the Qt bot and Tiger.