All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=872165

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=3162188268c0db95f8e62e04aa03c8de99f5f0398f2a54e378c32d8a7c17de10


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

android-nexus5x-perf
linux-perf

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16d6c5a4640000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16d6c5a4640000

Roll AFDO from 70.0.3510.3_rc-r1 to 70.0.3512.0_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/ee593cc28b00bf0e3b5a73b32da4c3429edb325c
8.82 → 15.42 (+6.598)

Assigning to sheriff gbiv@chromium.org because "Roll AFDO from 70.0.3510.3_rc-r1 to 70.0.3512.0_rc-r1" is a roll.

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

Thanks! Looks like the roll in r580797 fixed this.

The recent flakiness of this is moderately concerning, though. I made a note in  issue 872165 , which tracks denoising efforts as a whole. If this flakiness remains, we may be able to do a targeted fix, depending on what's causing it, exactly.

(I guess WontFix is *technically* correct, but ... :) )

:)

thank you for taking a look!