New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 798372 link

Starred by 1 user

Issue metadata

Status: Assigned
Owner:
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

14% regression in blink_perf.bindings at 526215:526232

Project Member Reported by alexclarke@chromium.org, Jan 2 2018

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=798372

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=fe662f8e87fbd2af0330f05eff95fde647a31f1d28e7072d7afb3523d3c5e1db


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

android-webview-nexus6
Cc: ishell@chromium.org
Owner: ishell@chromium.org
Status: Assigned (was: Untriaged)

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

Hi ishell@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 : Igor Sheludko
  Commit : efe12782a4aa85f29d698489584c6a78c3e0bc60
  Date   : Thu Dec 21 10:42:55 2017
  Subject: [ic] Restructure LoadGlobalIC generator code.

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : blink_perf.bindings
  Metric       : undefined-first-child/undefined-first-child
  Change       : 9.03% | 261.4927287 -> 237.875563532

Revision                           Result                   N
chromium@526214                    261.493 +- 3.10705       6      good
chromium@526219                    261.592 +- 0.888229      6      good
chromium@526220                    261.193 +- 1.42348       6      good
chromium@526220,v8@efe12782a4      241.647 +- 20.9705       6      bad       <--
chromium@526220,v8@a93b735a1a      233.369 +- 23.3203       6      bad
chromium@526220,v8@5d10735e18      234.152 +- 21.2943       6      bad
chromium@526220,v8@14cd4b5455      237.948 +- 1.256         6      bad
chromium@526220,v8@88a9495c51      237.234 +- 2.22969       6      bad
chromium@526220,v8@a130438e13      237.426 +- 2.04748       6      bad
chromium@526221                    238.178 +- 0.853213      6      bad
chromium@526223                    238.334 +- 1.17336       6      bad
chromium@526232                    237.876 +- 1.87997       6      bad

Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

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

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8958512927558395888


For feedback, file a bug with component Speed>Bisection
Cc: jbroman@chromium.org
ishell: any progress on this performance regression? Adding blink_perf.bindings owner jbroman to comment on severity.

Comment 5 by ishell@chromium.org, Jan 25 2018

No progress so far. Planning to look at it next week.
#4: not much to say from me; this is a microbenchmark, but this regression probably indicates that V8->Blink calls have gotten more expensive which may affect perf in general
Project Member

Comment 7 by bugdroid1@chromium.org, Apr 4 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/v8/v8.git/+/7aa7c5dd245e9ee7301b4966410f4592e81e47af

commit 7aa7c5dd245e9ee7301b4966410f4592e81e47af
Author: Igor Sheludko <ishell@chromium.org>
Date: Wed Apr 04 15:57:18 2018

[ic] Fix performance regression caused by LoadGlobalIC refactoring.

Bug: chromium:798372
Change-Id: I76d7317ef66dd5005250a10961d6732c39d4d108
Reviewed-on: https://chromium-review.googlesource.com/995445
Reviewed-by: Mythri Alle <mythria@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#52359}
[modify] https://crrev.com/7aa7c5dd245e9ee7301b4966410f4592e81e47af/src/ic/accessor-assembler.h
[modify] https://crrev.com/7aa7c5dd245e9ee7301b4966410f4592e81e47af/src/interpreter/interpreter-generator.cc

Project Member

Comment 9 by 42576172...@developer.gserviceaccount.com, Apr 10 2018

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/12c5879cc40000
Project Member

Comment 11 by 42576172...@developer.gserviceaccount.com, Apr 11 2018

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/11809e5cc40000
Project Member

Comment 13 by 42576172...@developer.gserviceaccount.com, Apr 12 2018

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/1286e4a2c40000
These was failing with swarming timeouts, not enough devices, retrying since we've increased capacity.
Project Member

Comment 16 by 42576172...@developer.gserviceaccount.com, Apr 18 2018

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/12b557cac40000
Project Member

Comment 17 by 42576172...@developer.gserviceaccount.com, Apr 18 2018

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/12b557cac40000

Sign in to add a comment