WebKit should always check the return value of pthread_key_create().
Created attachment 295894 [details] Patch v1
The other place that the return value of pthread_key_create() is currently checked is here: Source/WTF/wtf/ThreadSpecific.h The idiom used in this code has caught bugs recently (potential leaks of JSC::MachineThreads::MachineThreads objects).
Comment on attachment 295894 [details] Patch v1 Attachment 295894 [details] did not pass mac-debug-ews (mac): Output: http://webkit-queues.webkit.org/results/2603847 New failing tests: transitions/default-timing-function.html
Created attachment 295909 [details] Archive of layout-test-results from ews117 for mac-yosemite The attached test failures were seen while running run-webkit-tests on the mac-debug-ews. Bot: ews117 Port: mac-yosemite Platform: Mac OS X 10.10.5
(In reply to comment #3) > Comment on attachment 295894 [details] > Patch v1 > > Attachment 295894 [details] did not pass mac-debug-ews (mac): > Output: http://webkit-queues.webkit.org/results/2603847 > > New failing tests: > transitions/default-timing-function.html I don't see the crashes or failing tests when I run locally trunk r209565. Going to land the patch manually.
Committed r209590: <http://trac.webkit.org/changeset/209590>