New issue
Advanced search Search tips

Issue 699741 link

Starred by 1 user

Issue metadata

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



Sign in to add a comment

22.5% regression in smoothness.gpu_rasterization.tough_filters_cases at 455245:455368

Project Member Reported by benhenry@google.com, Mar 8 2017

Issue description

Might dupe into a regression to memory. Running as this as metrics are different.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=699741

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDglJKA5AoM


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

android-webview-nexus5X
Cc: reed@google.com
Owner: reed@google.com

=== Auto-CCing suspected CL author reed@google.com ===

Hi reed@google.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 : Mike Reed
  Commit : ca49067a3378b5c0364817c4198c063de05a331a
  Date   : Tue Mar 07 20:51:34 2017
  Subject: optimize intersect, use getType to utilize fast-case in preTranslate

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : smoothness.gpu_rasterization.tough_filters_cases
  Metric       : frame_times/http___web.archive.org_web_20150502135732_http___ie.microsoft.com_testdrive_Performance_Pirates_Default.html
  Change       : 23.27% | 27.0278255335 -> 33.3161177878

Revision                             Result                  N
chromium@455244                      27.0278 +- 3.46331      6      good
chromium@455260                      26.9076 +- 3.13123      6      good
chromium@455268                      26.8637 +- 2.95186      6      good
chromium@455268,skia@fdf3bbe821      26.3265 +- 1.97057      6      good
chromium@455268,skia@ca49067a33      31.1112 +- 4.10333      6      bad       <--
chromium@455268,skia@99a339069c      31.1953 +- 3.05465      6      bad
chromium@455269                      30.9994 +- 3.99634      6      bad
chromium@455270                      31.2279 +- 3.79901      6      bad
chromium@455272                      31.6709 +- 3.68706      6      bad
chromium@455275                      31.54 +- 4.37049        6      bad
chromium@455306                      31.7007 +- 4.10137      6      bad
chromium@455368                      33.3161 +- 3.89626      6      bad

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 --story-filter=http...web.archive.org.web.20150502135732.http...ie.microsoft.com.testdrive.Performance.Pirates.Default.html smoothness.gpu_rasterization.tough_filters_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8985651670314702688

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5007658911793152


| 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!
Status: Assigned (was: Untriaged)
This is a rather significant regression. Are there better ways to solve the problem without affecting smoothness?
Project Member

Comment 5 by sheriffbot@chromium.org, Jul 25 2017

Labels: Hotlist-Google
Status: WontFix (was: Assigned)
These metrics eventually came back down.

Sign in to add a comment