This is a meta bug of tracking down each revision. I haven't investigated whether these regression affect other platforms.
http://build.chromium.org/f/chromium/perf/xp-release-dual-core/intl1/report.html?history=1200&rev=103677
There are some other small regressions, but those are the big potatoes.
http://code.google.com/p/chromium/issues/detail?id=98904
Greetings, It seems the "Linux Perf (1)" bot also reports similar perf regression (*1). (*1) http://build.chromium.org/f/chromium/perf/linux-release/intl1/report.html?history=1200&rev=103677 To read WebKit changes (*2) rolled by r102911 <http://crrev.com/102911> (which caused the regression on Sep 27), the most possible change is r96066 (*3). (*2) http://trac.webkit.org/log/trunk/?action=stop_on_copy&mode=stop_on_copy&rev=96091&stop_rev=96065&limit=100&verbose=on (*3) http://trac.webkit.org/changeset/96066/trunk I wish it helps. Regards, Hironori Bono
Marking test failures as WontFix. Bug is still accessible and recording in TestExpectations.