All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=718028

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg8uyYpgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg8tjssAkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg0on6gwkM


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

chromium-rel-win10
chromium-rel-win7-gpu-intel
linux-release

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8980605284043362368

This could be related to  Issue 717128 , very close in commit position.

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

Hi chrishtr@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : chrishtr
  Commit : 91dea332b757072ca557cd386fdbf5383c0aa2b7
  Date   : Fri Apr 28 17:53:35 2017
  Subject: Introduce PaintPropertyTreeBuilderFragmentContext and use it throughout.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : blink_perf.paint
  Metric       : complex-content-slow-scroll/complex-content-slow-scroll
  Change       : 7.35% | 213.7945 -> 229.515083333

Revision             Result                   N
chromium@467971      213.795 +- 4.2052        6      good
chromium@468017      215.831 +- 0.882441      6      good
chromium@468040      212.075 +- 4.01208       6      good
chromium@468046      211.525 +- 1.22544       6      good
chromium@468049      214.294 +- 1.74929       6      good
chromium@468050      213.913 +- 1.99906       6      good
chromium@468051      234.372 +- 4.48255       6      bad       <--
chromium@468052      234.871 +- 1.80299       6      bad
chromium@468063      229.515 +- 0.956549      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.paint

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8980605284043362368

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6661506327904256


| 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 Speed>Bisection.  Thank you!

I think a proposed fix to  issue 718045  might also resolve this one.

 Issue 718045  has been merged into this issue.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/0f10830d5d83bf380b6a1c9978473b2d4e1720b4

commit 0f10830d5d83bf380b6a1c9978473b2d4e1720b4
Author: Chris Harrelson <chrishtr@chromium.org>
Date: Fri Jun 09 20:45:48 2017

Reserve one entry in the fragments list in PaintPropertyTreeBuilderContext.

There is always at least one fragment.

Bug:  718028 
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
Change-Id: I2cbae47b62e537cb44dc655b2f40a23c58b32f90
Reviewed-on: https://chromium-review.googlesource.com/529586
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Commit-Queue: Chris harrelson <chrishtr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#478404}
[modify] https://crrev.com/0f10830d5d83bf380b6a1c9978473b2d4e1720b4/third_party/WebKit/Source/core/paint/PaintPropertyTreeBuilder.h