All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=895537

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=dd7e12596bb6d4c3809c490c7044cceab8a96c6a2c95e4bcac95d323cf28b6eb


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

linux-perf

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14780a6ce40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14780a6ce40000

Roll AFDO from 71.0.3578.2_rc-r1 to 72.0.3579.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/c5dcad34966ab45f11d8826212660265a3445881
clockless_video_playback_vp9: 32.89 → 32.43 (-0.4633)

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

It appears you were the Sheriff on the specific roll, could please review and reassign as appropriate.

AFDO rolls are known to cause noise sometimes; looks like this recovered quite quickly, and I can't see any recent AFDO-induced noise, so there's not much to really do here.

Normally the spike back up coincides with another AFDO roll, though that isn't the case here, which is... mildly interesting. In fact, the only two commits I can see on the 'recovered' edge are a Clank roll and version bump... Neither of which should really cause perf differences for Linux. Hm.

In any case, like said, this appears to be a one-off thing that we've already recovered from fully, so I'm declaring victory and moving on :)