All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=624619

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg4uLHpwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg4tSmtgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg4pKBoAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgkpTQugoM


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

chromium-rel-win7-dual
chromium-rel-win7-gpu-intel
chromium-rel-win7-x64-dual
chromium-rel-win8-dual

Started bisect: https://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1281

Not sure what happened to that job. Started a new bisect:
https://chromeperf.appspot.com/buildbucket_job_status/9008378477604573728

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

OK that bisect FAILED but it didn't update this bug.

There is some possibility that the instrumentation here caused a regression:
https://codereview.chromium.org/2107773002

Since the test exercises postMessage. If that's true we should expect the perf to recover when this change cycles through perf bots: https://codereview.chromium.org/2133743002/

We got our perf back after https://codereview.chromium.org/2133743002/.