All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=846275

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=28b8704040859476e76166914e8d7369e9a146314bb3af704e2875ebf99bc129


Bot(s) for this bug's original alert(s):

chromium-rel-mac11-pro

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/118921c2240000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/118921c2240000

Reland "Use Skia's new simplified read/write pixels code paths" by bsalomon@chromium.org
https://chromium.googlesource.com/chromium/src/+/d74706673fcd86bc4c6fd6ff880f104379602ca9

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

I've tried reproducing this on two Macs so far. One gave a consistent 11K runs/second with or without the change. The other gave wildly inconsistent results (2K to 13K runs/sec) with and without the change. I'll try some more Macs.

I was never able to reproduce this and it seems this test is no longer being run on this bot.