New issue
Advanced search Search tips

Issue 788340 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner:
Closed: Nov 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

25.9% regression in thread_times.key_silk_cases at 518283:518481

Project Member Reported by hjd@google.com, Nov 24 2017

Issue description

See the link to graphs below.
 
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Nov 24 2017

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=788340

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=aaf129732cb98aa85429c4a760dcfe946771ebb477ffb8a13f469e663daf205d


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

android-nexus5
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Nov 24 2017

Cc: f...@opera.com
Owner: f...@opera.com
Status: Assigned (was: Untriaged)

=== Auto-CCing suspected CL author fs@opera.com ===

Hi fs@opera.com, 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 : Fredrik Söderquist
  Commit : e9bd34c60d38b0b5bcee52a6cd475b9124974aab
  Date   : Tue Nov 21 21:09:18 2017
  Subject: Use approximate stroke bounding boxes for SVG shapes

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : thread_times.key_silk_cases
  Metric       : thread_raster_cpu_time_per_frame/http___wiltzius.github.io_shape-shifter_
  Change       : 27.85% | 0.983602973681 -> 1.25753245596

Revision             Result                      N
chromium@518282      0.983603 +- 0.0245776       6      good
chromium@518382      0.974923 +- 0.0197803       6      good
chromium@518389      0.973669 +- 0.0102292       6      good
chromium@518391      0.976606 +- 0.00954395      6      good
chromium@518392      1.246 +- 0.0265017          6      bad       <--
chromium@518395      1.24189 +- 0.0230783        6      bad
chromium@518407      1.24986 +- 0.0403475        6      bad
chromium@518432      1.24745 +- 0.0254844        6      bad
chromium@518481      1.25753 +- 0.0361044        6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...wiltzius.github.io.shape.shifter. thread_times.key_silk_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8962047777858978848


For feedback, file a bug with component Speed>Bisection
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, Nov 24 2017

 Issue 788339  has been merged into this issue.
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Nov 25 2017

 Issue 788397  has been merged into this issue.

Comment 6 by f...@opera.com, Nov 27 2017

 Issue 788334  has been merged into this issue.

Comment 7 by f...@opera.com, Nov 27 2017

The blink_perf.svg:SvgHitTesting instance is known and expected (and will be swallowed.)
thread_times.key_silk_cases looks like it may be raster related, and can probably be improved upon.
Project Member

Comment 8 by bugdroid1@chromium.org, Nov 28 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/8f54d372df2a11c8ed90da000b7a2e7e574b6543

commit 8f54d372df2a11c8ed90da000b7a2e7e574b6543
Author: Fredrik Söderquist <fs@opera.com>
Date: Tue Nov 28 19:11:15 2017

Make stroke bounding boxes for <line> tighter

Since <line> elements are just a two point line segment, it does not
have any line joins and thus needn't apply any correction for the miter.

Bug: 435097,  788340 
Change-Id: Ie7e9d89e0f7c741cf7cf40b3af82521869bb5b54
Reviewed-on: https://chromium-review.googlesource.com/790294
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Fredrik Söderquist <fs@opera.com>
Cr-Commit-Position: refs/heads/master@{#519779}
[modify] https://crrev.com/8f54d372df2a11c8ed90da000b7a2e7e574b6543/third_party/WebKit/LayoutTests/platform/linux/paint/invalidation/svg/window-expected.txt
[delete] https://crrev.com/4ce19321a1f9fe98d5d692477981ac14a3ae13cf/third_party/WebKit/LayoutTests/platform/linux/virtual/spv175/paint/invalidation/svg/window-expected.txt
[modify] https://crrev.com/8f54d372df2a11c8ed90da000b7a2e7e574b6543/third_party/WebKit/LayoutTests/platform/mac/paint/invalidation/svg/window-expected.txt
[modify] https://crrev.com/8f54d372df2a11c8ed90da000b7a2e7e574b6543/third_party/WebKit/LayoutTests/platform/win/paint/invalidation/svg/window-expected.txt
[modify] https://crrev.com/8f54d372df2a11c8ed90da000b7a2e7e574b6543/third_party/WebKit/Source/core/layout/svg/LayoutSVGShape.cpp
[modify] https://crrev.com/8f54d372df2a11c8ed90da000b7a2e7e574b6543/third_party/WebKit/Source/core/layout/svg/LayoutSVGShape.h

Comment 9 by f...@opera.com, Nov 30 2017

Status: WontFix (was: Assigned)
CL in c#8 helped a little bit, but I don't think we will achieve much better.

Sign in to add a comment