[Webkit-unassigned] [Bug 235114] REGRESSION (iOS 15): Massive variation in microbenchmark results when tab returns to foreground

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Feb 3 09:30:10 PST 2022


https://bugs.webkit.org/show_bug.cgi?id=235114

--- Comment #12 from Simon Taylor <simontaylor1 at ntlworld.com> ---
I've also added a "manual" benchmark runner page that just takes one set of 200 samples and shows them all each time you click a button:
https://tango-bravo.net/webkit-bug-235114/manual.html

You can observe the CPU frequency ramping up as the early samples have lower counts than the later ones, and the later ones are still quite consistent. For whatever reason this bug isn't as obvious on that page though, even on iOS 15.2 devices, though a do still see a couple of clear operating points, it's not the 3x difference that is visible in the rAF-based test. So the rAF test page is probably the more useful test case but thought I might as well share this one too.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20220203/cf05a9cc/attachment.htm>


More information about the webkit-unassigned mailing list