New issue
Advanced search Search tips

Issue 812719 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Mar 2018
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocked on:
issue 813881

Blocking:
issue 694855



Sign in to add a comment

13.7%-34.4% regression in media.desktop at 536407:536577

Project Member Reported by wolenetz@chromium.org, Feb 15 2018

Issue description

These 3 alerts look like they were valid. I suspect the underlying commit has been reverted already (https://chromium-review.googlesource.com/c/chromium/src/+/920481), corroborated by the most recent datapoint for each of these looks like it's returned to pre-alert state.

Notably, there are 12 other *improvements* in similar metrics in the same range. Group: https://chromeperf.appspot.com/group_report?sid=51158e203526b738d45a4cadcfeaebf4a1bc5b5c43db74200f74f761c04b3492 (expand the "15" to see the 12 improvements beyond these 3 alerts).

Starting a bisect just in case there's something else causing this particular set of 3 alerts...
 
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Feb 15 2018

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=812719

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=f0cb743c389af2a044874b948ddee25f3945a56a5786e2e8afa5480e41568145


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

chromium-rel-mac11-air
chromium-rel-win7-gpu-ati
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Feb 15 2018

Cc: danakj@chromium.org ericwilligers@chromium.org sky@chromium.org zhuoyu.q...@samsung.com
Owner: zhuoyu.q...@samsung.com
Status: Assigned (was: Untriaged)
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/12ad2d93840000

Adds base::MatchesUniquePtr to ease lookup by pointer by sky@chromium.org
https://chromium.googlesource.com/chromium/src/+/f9025d4abc34a32a4c7449005cdae14d87205c19

Put CSSRayValue and CSSContentDistributionValue into blink::cssvalue. by zhuoyu.qian@samsung.com
https://chromium.googlesource.com/chromium/src/+/e1f5e9674ded3bb7455242ddcac01dcad95dfb48

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
@#3 that pinpoint job had many problems. Ignore #3. I'll start a job with same range.
Owner: ----
Status: Unconfirmed (was: Assigned)
#6 is a retry, but on a different bot/metric that regressed in similar range:  ChromiumPerf/chromium-rel-win7-gpu-ati/media.desktop / cpu_time_percentage_avg /
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Feb 16 2018

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14e2a163840000
Hmm. I think pinpoint infra might have led to bad #3 result. I'll retry precisely that bisect now.
Blockedon: 813881
Project Member

Comment 12 by 42576172...@developer.gserviceaccount.com, Feb 20 2018

Cc: mlamouri@chromium.org liber...@chromium.org
Owner: dalecur...@chromium.org
Status: Assigned (was: Unconfirmed)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14b389c7840000

Remove kHaveFutureData restriction from play state notifications. by dalecurtis@chromium.org
https://chromium.googlesource.com/chromium/src/+/b0ec93f6f1c9dd4754c813f198336a7a4c2f22a4

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -danakj@chromium.org
Blocking: 694855
Status: Fixed (was: Assigned)
Fixed since was reverted. We can reevaluate if we attempt to remove this lie again.

Sign in to add a comment