Since https://bugs.webkit.org/show_bug.cgi?id=50080, cookiesEnabled() returns true even when the cookie jar is null.
Created attachment 78266 [details] Patch I think it is ok regarding the thread safety because this check is in the thread that has the possibility to change the cookie jar, so both action cannot happen simultaneously.
The commit-queue encountered the following flaky tests while processing attachment 78266 [details]: fast/workers/storage/use-same-database-in-page-and-workers.html bug 50995 (author: dumi@chromium.org) The commit-queue is continuing to process your patch.
Comment on attachment 78266 [details] Patch Clearing flags on attachment: 78266 Committed r75337: <http://trac.webkit.org/changeset/75337>
All reviewed patches have been landed. Closing bug.
It made fast/dom/navigator-detached-no-crash.html crash.
Created attachment 78361 [details] GDB backtrace in debug mode
Rolled out. Reopening...
Created attachment 78421 [details] Patch
(In reply to comment #5) > It made fast/dom/navigator-detached-no-crash.html crash. It is really a strange case. The function can be called on an invalid object, I must test for the validity of the manager before calling anything on it. I am glad the autotest have catched that.
Comment on attachment 78421 [details] Patch Should work this time around :-)
Comment on attachment 78421 [details] Patch Clearing flags on attachment: 78421 Committed r75411: <http://trac.webkit.org/changeset/75411>