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

Issue 794557 link

Starred by 1 user
Status: WontFix
Owner:
wangxianzhu@chromium.org
Closed: Dec 2017
Cc:
wangxianzhu@chromium.org
█ kraynov@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

14.8% regression in smoothness.key_mobile_sites_smooth at 523016:523020

Project Member Reported by kraynov@chromium.org, Dec 13 2017 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Dec 13 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=794557

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=c3ec1b5cffa288867e6c96f078e39cb2ba9dd1ad846e86b9799099ffacf98bac


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

android-nexus5X
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Dec 13 2017

Cc: wangxianzhu@chromium.org
Owner: wangxianzhu@chromium.org
Status: Assigned (was: Untriaged)

=== 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 : 94b89f4fe30c693692988e3360acf3c8952c532e
  Date   : Sun Dec 10 06:24:43 2017
  Subject: [SPv175/PE] SetLayerState for AncestorClippingMaskLayer and ChildClippingMaskLayer

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : smoothness.key_mobile_sites_smooth
  Metric       : mean_input_event_latency/http___worldjournal.com_
  Change       : 18.69% | 17.5796666667 -> 20.8656666667

Revision             Result                   N
chromium@523015      17.5797 +- 0.843708      6      good
chromium@523018      18.0108 +- 1.20132       6      good
chromium@523019      17.5947 +- 1.30343       6      good
chromium@523020      20.8657 +- 1.13822       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...worldjournal.com. smoothness.key_mobile_sites_smooth

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8960315716067935840


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

Comment 4 by 42576172...@developer.gserviceaccount.com, Dec 13 2017 

 Issue 794545  has been merged into this issue.

Comment 6 by wangxianzhu@chromium.org, Dec 13 2017

Status: WontFix (was: Assigned)
The CL fixed a bug caused by https://chromium.googlesource.com/chromium/src/+/88986f832c463a210324d7d9cf9b5213f7b69f63 which also caused the false progression. The regression actually returns the performance to the level before the bug.
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Dec 13 2017 


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

Suspected Commit
  Author : Xianzhu Wang
  Commit : 94b89f4fe30c693692988e3360acf3c8952c532e
  Date   : Sun Dec 10 06:24:43 2017
  Subject: [SPv175/PE] SetLayerState for AncestorClippingMaskLayer and ChildClippingMaskLayer

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : smoothness.key_mobile_sites_smooth
  Metric       : mean_input_event_latency/http___worldjournal.com_
  Change       : 26.23% | 23.1151666667 -> 29.1783333333

Revision             Result                  N
chromium@523016      23.1152 +- 2.97525      6      good
chromium@523018      22.2005 +- 1.8253       6      good
chromium@523019      22.4133 +- 3.35866      6      good
chromium@523020      29.1783 +- 2.92584      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...worldjournal.com. smoothness.key_mobile_sites_smooth

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8960305881920538240


For feedback, file a bug with component Speed>Bisection
►

Sign in to add a comment