There are some counters which make sense for workers as well (e.g. event listener count) and since workers run in their own threads it should be safe to increment/decrement counters from there.
Created attachment 129910 [details] Patch
Dromaeo results for Chromium run with the patch applied: http://dromaeo.com/?id=164681 and without it: http://dromaeo.com/?id=164673
Comment on attachment 129910 [details] Patch Attachment 129910 [details] did not pass qt-ews (qt): Output: http://queues.webkit.org/results/11802009
Comment on attachment 129910 [details] Patch Attachment 129910 [details] did not pass win-ews (win): Output: http://queues.webkit.org/results/11797038
(In reply to comment #2) > Dromaeo results for Chromium run > with the patch applied: http://dromaeo.com/?id=164681 > and without it: http://dromaeo.com/?id=164673 So you are saying it got faster? Sounds like you are breaking ews.
Comment on attachment 129910 [details] Patch r- for breaking build.
(In reply to comment #5) > (In reply to comment #2) > > Dromaeo results for Chromium run > > with the patch applied: http://dromaeo.com/?id=164681 > > and without it: http://dromaeo.com/?id=164673 > > So you are saying it got faster? > Of cause not, it just one of the sampling errors. In two other runs the results were about the same: http://dromaeo.com/?id=164676 http://dromaeo.com/?id=164678 > Sounds like you are breaking ews. Will fix.
Created attachment 130107 [details] Patch
Created attachment 130385 [details] Patch
Committed r109922: <http://trac.webkit.org/changeset/109922>