All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=917267

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=467db0a1bef53072bf40a83dd68d49c905321eddd5f7d3af109fca4c9be076da


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

android-nexus5x-perf

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

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/129d6aee140000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/129d6aee140000

tracing: Emit metadata at the end of tracing instead of beginning by eseckler@chromium.org
https://chromium.googlesource.com/chromium/src/+/e49e90a64dbbd25089c1faf972312a9ed4c08b34
cpu_time_percentage: 1.425 → 1.696 (+0.2713)

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

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

WAI.

Sorry, not a duplicate. It's possible that above patch causes a change in metrics because emitted Trace events may differ. There's still some work out to be done to address remaining perfetto problems, see e.g.  issue 907069 .