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

Issue 814757 link

Starred by 1 user
Status: Fixed
Owner:
chrishtr@chromium.org
Closed: Mar 2018
Cc:
schenney@chromium.org
█ alexclarke@chromium.org
skobes@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

10.8%-18% regression in system_health.common_mobile at 538115:538191

Project Member Reported by alexclarke@chromium.org, Feb 22 2018 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Feb 22 2018 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=814757

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=d1e306e8f77706e356420ccc079a90a3800c7f2b6d8185e79eb54fe0836d1f5a


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

android-nexus5
android-nexus6
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Feb 25 2018 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/14db7a9f840000
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Feb 28 2018

Cc: schenney@chromium.org skobes@chromium.org
Owner: schenney@chromium.org
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/11e0b3b8440000

Revert "[RLS] Skip compositing input step if a scroll can be applied directly." by schenney@chromium.org
https://chromium.googlesource.com/chromium/src/+/e0c819ef2ef21b3580dbf5458cc3f966c2dbc9e1

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Comment 6 by schenney@chromium.org, Feb 28 2018

Owner: chrishtr@chromium.org
The patch that was reverted itself caused regressions more severe than this across more metrics. I'll leave it up to chrishtr@ to decide which metrics we care more about or see if there a way to get the benefits without the costs.

The the reversion changelist for links to the other bugs.
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Mar 4 2018

Owner: schenney@chromium.org
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1038b7a4440000

Revert "[RLS] Skip compositing input step if a scroll can be applied directly." by schenney@chromium.org
https://chromium.googlesource.com/chromium/src/+/e0c819ef2ef21b3580dbf5458cc3f966c2dbc9e1

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Comment 9 by schenney@chromium.org, Mar 5 2018

Owner: chrishtr@chromium.org

Comment 10 by chrishtr@chromium.org, Mar 5 2018

Status: Fixed (was: Assigned)
I re-landed the patch on Friday - performance back to the moar optimized
state.
►

Sign in to add a comment