All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=813994

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=74dbd54d342fd689a549b88fe98f19912b8c551a789a9b8b5710a884f500ae1f


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

chromium-rel-mac11-air
chromium-rel-mac11-pro
chromium-rel-mac12
chromium-rel-win10
chromium-rel-win7-gpu-ati
chromium-rel-win7-gpu-intel
chromium-rel-win7-gpu-nvidia
chromium-rel-win8-dual
linux-release
win-high-dpi

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/11e37c8f840000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/11e37c8f840000

ResourceLoadScheduler: Count the number of throttlable requests by yhirano@chromium.org
https://chromium.googlesource.com/chromium/src/+/392731364d805ab0ee42e785382f23e0707f9f9a

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

According to ksakamoto@ cpuTimeToFirstMeaningfulPaint_avg itself is not a optimization target, so I ignore them.

There are some FMP / FCP regressions, but all of them were improved by my previous change[1] before this regression, so they are not real regressions. 

1: https://chromium.googlesource.com/chromium/src/+/1f25bbc692f06964b5e10c7312f9f81377cdea32