=== Auto-CCing suspected CL author wangxianzhu@chromium.org ===
Hi wangxianzhu@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 : Xianzhu Wang
Commit : 7d51399398db180fe1a7a83a95820e56a9fac755
Date : Thu Jun 29 18:28:49 2017
Subject: Devirtualize LayoutBoxModelObject logical orientation border/margin/padding accessors
Bisect Details
Configuration: winx64_high_dpi_perf_bisect
Benchmark : blink_perf.layout
Metric : floats_2_100_nested/floats_2_100_nested
Change : 19.64% | 157.191833333 -> 188.060333333
Revision Result N
chromium@483416 157.192 +- 1.18147 6 good
chromium@483425 154.259 +- 2.6891 6 good
chromium@483430 153.39 +- 2.13364 6 good
chromium@483431 155.047 +- 1.18452 6 good
chromium@483432 190.28 +- 6.49793 6 bad <--
chromium@483434 193.369 +- 36.5085 6 bad
chromium@483451 187.875 +- 2.98011 6 bad
chromium@483486 186.015 +- 2.82797 6 bad
chromium@483555 188.06 +- 0.770882 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.layout
More information on addressing performance regressions:
http://g.co/ChromePerformanceRegressions
Debug information about this bisect:
https://chromeperf.appspot.com/buildbucket_job_status/8975010211115927504
For feedback, file a bug with component Speed>Bisection
These graphs have all gone back to baseline.
This regression was primarily on Windows. I wonder if the patch just needs to do something special for the windows compiler.
Comment 1 by rmcilroy@chromium.org
, Jul 4 2017