All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=669176

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-7qVpgkM


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

android-nexus9

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8994717134484740432

=== Auto-CCing suspected CL author enne@chromium.org ===

Hi enne@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Move SaturatedArithmetic from Blink to base
Author  : enne
Commit description:
  
This is a more efficient implementation than what gfx is doing in most
cases.  This CL moves it to base, updates coding style, and switches
gfx to using it.

TBR=thakis@chromium.org
BUG= 659381 

Review-Url: https://codereview.chromium.org/2499783002
Cr-Commit-Position: refs/heads/master@{#433474}
Commit  : 42cb2e5d1ee0e5a9e1073307b93df913d2df913f
Date    : Mon Nov 21 04:36:21 2016


===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N   Good?
chromium@433467  0.224787  0.0218394  25  good
chromium@433471  0.227628  0.041602   25  good
chromium@433473  0.224726  0.028887   25  good
chromium@433474  0.197834  0.0209447  25  bad    <--
chromium@433481  0.194588  0.0252308  25  bad
chromium@433493  0.191847  0.0266994  25  bad

Bisect job ran on: android_nexus9_perf_bisect
Bug ID: 669176

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.events
Test Metric: hit-test-lots-of-layers/hit-test-lots-of-layers
Relative Change: 14.65%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus9_perf_bisect/builds/2286
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994717134484740432


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6378594836676608

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

enne - Can you comment on if your CL caused the regression and what the plan to address it is?

Sure, it seems plausible that it did if an int rect caused it previously.  I am landing https://codereview.chromium.org/2533113003 speculatively to see if that was what changed.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/3f217d164d6f4d0e16cd2b0ca40f7b4f83c5f58a

commit 3f217d164d6f4d0e16cd2b0ca40f7b4f83c5f58a
Author: enne <enne@chromium.org>
Date: Wed Nov 30 08:52:37 2016

Add back missing ALWAYS_INLINE to saturated math

BUG= 669176 

Review-Url: https://codereview.chromium.org/2533113003
Cr-Commit-Position: refs/heads/master@{#435180}

[modify] https://crrev.com/3f217d164d6f4d0e16cd2b0ca40f7b4f83c5f58a/base/numerics/saturated_arithmetic.h

Two data points make it look like this has recovered now.  I was also informed that Clank builds turn off inlining for binary size, so it makes some sense that forcing inlining would help this case specifically.