All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=893864

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=62fef9231e80ab34cb98f333c3cfff43c4560af13d7db852e8b08ad3d76ceda6


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

Android Nexus5 Perf
Android Nexus5X WebView Perf
Android Nexus6 WebView Perf

system_health.memory_mobile - Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

rendering.mobile - Benchmark documentation link:
  https://bit.ly/rendering-benchmarks

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/128ce112e40000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/128ce112e40000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/106eb1cae40000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16da8c8ee40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/106eb1cae40000

NavigationPredictor: Skip storage of duplicate anchor elements by tbansal@chromium.org
https://chromium.googlesource.com/chromium/src/+/70974c1a00b4ef70f2df877bddcc842ff30c91a2
memory:webview:all_processes:reported_by_chrome:cc:effective_size: 7.259e+07 → 1.1e+08 (+3.741e+07)

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

Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16da8c8ee40000

NavigationPredictor: Skip storage of duplicate anchor elements by tbansal@chromium.org
https://chromium.googlesource.com/chromium/src/+/70974c1a00b4ef70f2df877bddcc842ff30c91a2
memory:webview:all_processes:reported_by_chrome:cc:effective_size: 7.24e+07 → 1.096e+08 (+3.72e+07)

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

Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

The CL that triggered this alert is behind a finch experiment. As of now, the experiment is not enabled on Stable channel. Currently, the code is running in the dry run mode: i.e., there is some computational overhead involved, but the feature does not take any actual action (besides recording metrics).

Before we start ramping up on Stable, we plan to move it out of the dry run mode. At that point, hopefully, the actions taken by the experiment would more than compensate the computational overhead.

Since the experiment is behind finch, and we are going to compare the usual performance and heartbeat metrics against the Control group (e.g., time to interactive, Event.Latency.*), I am inclined to close this as WontFix since there is nothing really actionable here.