All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=828892

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=9b5544562c6284708ef8fa36382dc2b2251331223e64f19c6927bef60cfd4fd3


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

mac

tapted, as sizes owner can you help out? There was a build breakage so there is a big range of CLs: http://test-results.appspot.com/revision_range?start=547688&end=547782&n=1000

-> rbpotter, r547708 is my main suspect.

If this is mac-specific, r547730 stands out as a potential sizes _improvement_... Otherwise, there's very little mac-specific stuff in that bisect.

But there's actually no change in GoogleCodeFramework. That leaves resource bundle stuff mainly, or other non-code data.

Maybe the src-internal roll - https://chromium-review.googlesource.com/992882

Or some generated javascript in https://chromium-review.googlesource.com/c/chromium/src/+/985911

There's a delta in resources.pak on linux/win, but it's only ~10k vs this, 200k on Mac.

The src-internal roll updated chromeos code, so seems unlikely it's related to mac size changes.

Regarding r547708, I don't see how it can cause a 200kb increase, since it does not add (or generate) that much output. Un-assigning for now.

Looking through the large range of commits, did not spot anything obvious. Perhaps r547746 or r547776 (moves some code in components/ but not sure if this is included in CrOS only)? CC'ing more people.

Elly: do you think this is something we can address at this point, or should we close?

Downgrading P2s that haven't been modified in more than 6 months, which have no component or owner.