Since this member is hit frequently in StyleResolver, this looks worth killed by a bit flag, which is much efficient.
Created attachment 163493 [details] Patch
Comment on attachment 163493 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=163493&action=review > Source/WebCore/ChangeLog:10 > + Instead of tracking the number of certain node, this chagne compute the number chagne->change > Source/WebCore/html/HTMLStyleElement.cpp:128 > + // Note: We cannot rely on the 'scoped' element still being present when this method is invoked. I think he meant 'scoped' _attribute_?
Comment on attachment 163493 [details] Patch Attachment 163493 [details] did not pass win-ews (win): Output: http://queues.webkit.org/results/13832193
Comment on attachment 163493 [details] Patch Attachment 163493 [details] did not pass mac-ews (mac): Output: http://queues.webkit.org/results/13827484
Dimitri, thank for taking a look at such a late at night! I'll address the points and land the change. (In reply to comment #2) > (From update of attachment 163493 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=163493&action=review > > > Source/WebCore/ChangeLog:10 > > + Instead of tracking the number of certain node, this chagne compute the number > > chagne->change Will fix. hmm. > > > Source/WebCore/html/HTMLStyleElement.cpp:128 > > + // Note: We cannot rely on the 'scoped' element still being present when this method is invoked. > > I think he meant 'scoped' _attribute_? Well, I just moved this around. But yes, it looks. Will fix.
Created attachment 163646 [details] Patch for landing
Comment on attachment 163646 [details] Patch for landing Clearing flags on attachment: 163646 Committed r128331: <http://trac.webkit.org/changeset/128331>
All reviewed patches have been landed. Closing bug.