All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=923371

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=85fc3ab31978cc14f9cf2466f06e72f5e3a949ecfe79d5ffe0bc05a9857e732e


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

Android Nexus5 Perf

blink_perf.events - Benchmark documentation link:
  https://bit.ly/blink-perf-benchmarks

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/11e6f9f6540000

📍 Found significant differences after each of 6 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/11e6f9f6540000

Let InspectorLayerTreeAgent use cc::Layer instead of GraphicsLayer by wangxianzhu@chromium.org
https://chromium.googlesource.com/chromium/src/+/e0e0e3b5f844cbe32fea55f63300b8f727f50282
EventsDispatchingInV1ShadowTrees: 1.827e+04 → 1.794e+04 (-332.9)

Port VideoResourceUpdater to shared images by piman@chromium.org
https://chromium.googlesource.com/chromium/src/+/986ad9d9f120c1a8eaf46a94d7e0c31fb022f5ef
EventsDispatchingInV1ShadowTrees: 1.789e+04 → 1.872e+04 (+832.4)

Roll AFDO from 73.0.3671.0_rc-r1 to 73.0.3672.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/2c7da71a5d429c476dbe4351fa3b8fdec82ecefa
EventsDispatchingInV1ShadowTrees: 1.872e+04 → 1.659e+04 (-2124)

Partial cleanup use of base::Value in printing backend. by vkuzkokov@chromium.org
https://chromium.googlesource.com/chromium/src/+/490ab0c8364fdf4b25829ff56b3df401b988bec8
EventsDispatchingInV1ShadowTrees: 1.668e+04 → 1.616e+04 (-515.3)

Fixed IPC memory leak of prerender alias urls messages. by a-v-y@yandex-team.ru
https://chromium.googlesource.com/chromium/src/+/6b058fecd94d99f75aec20fa7eb05cbc3bcc1c41
EventsDispatchingInV1ShadowTrees: 1.617e+04 → 1.628e+04 (+113.7)

drawText is deprecated, use drawSimpleText or drawTextBlob by reed@google.com
https://chromium.googlesource.com/chromium/src/+/f1639a837a7242ab03ff36399277bd47ccaf1d4a
EventsDispatchingInV1ShadowTrees: 1.64e+04 → 1.676e+04 (+356.5)

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

Benchmark documentation link:
  https://bit.ly/blink-perf-benchmarks

Based on the pinpoint result, he regression is caused by the AFDO roll.

This recovered in the AFDO roll at r624467. I don't see any other obvious AFDO-related noise in the history of this chart, so I'm just duping this against the general AFDO noise bug.

Thanks!