All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=847450

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=472ec8b3d903857b02cf7c301a05b5ddffa67479bcfab83fd02aafcb0d115a59


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

chromium-rel-win7-gpu-nvidia

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14fec09a240000

📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14fec09a240000

Do not consider pending queries as idle work in passthrough decoder by sunnyps@chromium.org
https://chromium.googlesource.com/chromium/src/+/ef1b3ab1cc2498425c46c021e9344c514392026c

Remove idle work polling interval override by sunnyps@chromium.org
https://chromium.googlesource.com/chromium/src/+/db2744fe6d080718ac44c3810dee794e4d3400d7

<b>Fix end_time < start_time in traces.</b> by jdarpinian@chromium.org
https://chromium.googlesource.com/chromium/src/+/95c992e0a0ee9246d56ad41a86aa149262b3c530

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

Is this expected from your change?

This is a "regression" in tasks_per_frame_total_all, so not really a regression at all.

I've reached out to speed infra owners off bug about the incorrect monitoring here.