Something in a webkit roll two weeks ago seemed to cause a performance regression in chromium. So far, it looks like it's caused by http://trac.webkit.org/changeset/58765 , although I have no idea why (and neither does johnnyg ). We'll try reverting the change to see if that speeds up the tests or not.
Created attachment 56420 [details] Patch
Attachment 56420 [details] did not build on chromium: Build output: http://webkit-commit-queue.appspot.com/results/2312298
Comment on attachment 56420 [details] Patch ok.
Committed r59782: <http://trac.webkit.org/changeset/59782>
http://trac.webkit.org/changeset/59782 might have broken Chromium Linux Release