All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=785564

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=86e3c1d665a8d4e0e5d7dc7dbb827990cccf54e120359839dabaee77a688369e


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

mac

There was a sharp jump at r516838, but that change shouldn't have caused this regression. Of the changes around that point, none really pop-out. Since there's no way to bisect, I'm not sure what else I can do as sheriff.

Assigning to test owner...

It looks like it switched to a new buildslave 

https://build.chromium.org/p/chromium.perf/builders/Mac%20Builder/builds/137856 vm159-m1
https://build.chromium.org/p/chromium.perf/builders/Mac%20Builder/builds/137857 vm169-m1

I don't see anything obvious in the configs though

- https://build.chromium.org/p/chromium.perf/buildslaves/vm159-m1
- https://build.chromium.org/p/chromium.perf/buildslaves/vm169-m1

Could be due to slightly different xcode versions.

This is probably fine to ignore unless it starts jumping back and forth.

weeird. There was a 0.9% sizes regression on Linux a couple hundred commits earlier: https://bugs.chromium.org/p/chromium/issues/detail?id=785567&q=owner%3Aajwong%40chromium.org&sort=-modified&colspec=ID%20Pri%20M%20Iteration%20ReleaseBlock%20Cr%20Status%20Owner%20Summary%20OS%20Modified

but that's expected and WAI. I wonder if this is just a weird coincidence? Agree that it should be fine to ignore.

Yeah, seemed like just a coincidence. The other one was due to a Linux-specific compiler flag change.