All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=812294

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=a00238eec4deafe45a3f4d2a3303aed7b8e96030dea71097cdf42b6d3b1744e8


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

android-nexus5
android-nexus5X
android-nexus6
android-nexus7v2
android-webview-nexus5X

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14a228d3840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14c6da17840000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14c6da17840000

Ensure that an OOPIF renderer knows its size before layout. by lfg@chromium.org
https://chromium.googlesource.com/chromium/src/+/e5d27a36230aea4b3664f8b5345b09bb0690a3ff

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

 Issue 812378  has been merged into this issue.

 Issue 814164  has been merged into this issue.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/471160f9f79e96e11de88bf9acddbb07995be31a

commit 471160f9f79e96e11de88bf9acddbb07995be31a
Author: Lucas Gadani <lfg@chromium.org>
Date: Thu Mar 01 03:13:21 2018

Only replicate the frame's size to the browser process.

Instead of replicating the entire frame rect, which also contains
the frame's position, this patch changes it to replicate only the
frame's size. This should reduce the number of IPCs sent from the
renderer to the browser, especially while scrolling.

The frame's size is needed by the browser process when spawning
an out-of-process iframe.

Bug:  812294 

Change-Id: I05aaf42880824abedac9c1cc655cc5c53390925b
Reviewed-on: https://chromium-review.googlesource.com/938651
Commit-Queue: Lucas Gadani <lfg@chromium.org>
Reviewed-by: Charlie Reis <creis@chromium.org>
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#540018}
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/browser/frame_host/render_frame_host_impl.cc
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/browser/frame_host/render_frame_host_impl.h
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/browser/frame_host/render_frame_host_manager.cc
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/browser/frame_host/render_frame_proxy_host.cc
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/browser/frame_host/render_frame_proxy_host.h
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/common/frame.mojom
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/renderer/render_frame_impl.cc
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/renderer/render_frame_impl.h
[modify] https://crrev.com/471160f9f79e96e11de88bf9acddbb07995be31a/content/test/test_render_frame.cc

 Issue 814174  has been merged into this issue.

 Issue 814051  has been merged into this issue.

The graphs seem to have recovered after my patch landed yesterday.

 Issue 812316  has been merged into this issue.