All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=809437

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=4f897a1985e8cecab76516fb314c428409d6947ea3efc14833c6eb29683946c1


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

chromium-rel-win7-dual

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1483dd91840000

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/1483dd91840000

Over half of these recovered--re-triaging those to "Ignore". Kicking off a bisect on the remaining ones.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/128b67f4440000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/17955694440000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16fcd2fc440000

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/128b67f4440000

📍 Found significant differences after each of 5 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/17955694440000

Add SSLVersionMin policy. by svaldez@chromium.org
https://chromium.googlesource.com/chromium/src/+/e35c49d42c64683ded8c145e7b23ab73eec5aaad

[RootLayerScrolls] Flip RuntimeEnabledFeatures flag to experimental by szager@chromium.org
https://chromium.googlesource.com/chromium/src/+/fe8c4db4003022f5cb12506a15cc78ea6633bbdf

Replace NavigationURLLoader unittest [1/n] by arthursonzogni@chromium.org
https://chromium.googlesource.com/chromium/src/+/e37f76e58eab6dfb5417c7a361ea026dd48dd890

Remove DidNotProduceFrame view message by dtapuska@chromium.org
https://chromium.googlesource.com/chromium/src/+/c8f1338f54e530e7865ec0d2f015725c48605de1

[gin] Implement v8::Platform::GetPageAllocator. by bbudge@chromium.org
https://chromium.googlesource.com/chromium/src/+/1cc19640473a99e99ef5d4223a7470765a3616ef

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

This looks like it's Mac-specific. The one alert for Android (Nexus5x) shows a crazy 6x improvement which I doubt is real.

My CL changes how large blocks of memory are allocated from the host OS. It is unlikely to cause significant performance regressions.

The second CL above looks more suspicious, since it changes how layout tests run. Perhaps tests need to be re-baselined?

The CL link:

https://chromium.googlesource.com/chromium/src/+/fe8c4db4003022f5cb12506a15cc78ea6633bbdf

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/16fcd2fc440000

Limited regression for a single platform for a micro benchmark without corresponding regression elsewhere. Closing.