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

Issue 804323 link

Starred by 1 user
Status: Fixed
Owner:
yhirano@chromium.org
Closed: Apr 2018
Cc:
yhirano@chromium.org
█ pmeenan@chromium.org
dtu@chromium.org
tdres...@chromium.org
sullivan@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

99.2% regression in loading.desktop at 530148:530262

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

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

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=804323

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=ef6ea9381a01321e8de919cd8f17354a9fc85e51229665d938fe1c4aab845d33


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

chromium-rel-win7-dual
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Jan 22 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/149d0da8840000
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Jan 23 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14c5e5a4840000

Comment 6 by sullivan@chromium.org, Jan 29 2018

Cc: sullivan@chromium.org tdres...@chromium.org
+tdresser: another loading regression too noisy to bisect. Is there anything we can do here, or should we close the bug?

Comment 7 by tdres...@chromium.org, Jan 30 2018 

I've bumped one of the alerts back a few revisions, as I think the regression occurred earlier than the alert indicates.

We'll see if a bisect there succeeds.
Project Member

Comment 9 by 42576172...@developer.gserviceaccount.com, Jan 30 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/16b133ea840000

Comment 11 by tdres...@chromium.org, Jan 30 2018 

This looks like a difference to me...
https://pinpoint-dot-chromeperf.appspot.com/job/16b133ea840000

Comment 12 by sullivan@chromium.org, Jan 30 2018

Cc: dtu@chromium.org
+dtu can you take a look at #11?
Project Member

Comment 13 by 42576172...@developer.gserviceaccount.com, Jan 30 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/148d2146840000

Comment 14 by tdres...@chromium.org, Jan 30 2018 

And 13 looks like the opposite of 11.

Marking WontFix (though I am curious about #11).

Comment 15 by dtu@chromium.org, Jan 31 2018 

I'm getting really high p-values for this distribution. Mann-Whitney U is not great at detecting this kind of distribution change, where both the low values get lower and the high values get higher. I've made a note in https://github.com/catapult-project/catapult/issues/3898

Comment 16 by dtu@chromium.org, Jan 31 2018 

I've deployed an experimental fix that implements the Kolmogorov-Smirnov test:
https://chromium-review.googlesource.com/c/catapult/+/896396

And will try re-running both jobs now.
Project Member

Comment 19 by 42576172...@developer.gserviceaccount.com, Feb 1 2018

Cc: isherman@chromium.org yhirano@chromium.org
Owner: yhirano@chromium.org
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16aeff26840000

Update RendererSideResourceScheduler testing config (Enabled_3_1024)
By yhirano@chromium.org · Wed Jan 17 06:00:00 2018
chromium @ e6395d1c63e9f382fbd13de8bc94796712fd88b2

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

Comment 20 by isherman@chromium.org, Feb 1 2018

Cc: -isherman@chromium.org
Project Member

Comment 21 by 42576172...@developer.gserviceaccount.com, Feb 1 2018

Cc: isherman@chromium.org
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/15abb5f6840000

Update RendererSideResourceScheduler testing config (Enabled_3_1024)
By yhirano@chromium.org · Wed Jan 17 06:00:00 2018
chromium @ e6395d1c63e9f382fbd13de8bc94796712fd88b2

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

Comment 22 by dtu@chromium.org, Feb 1 2018

Cc: -isherman@chromium.org
Project Member

Comment 27 by 42576172...@developer.gserviceaccount.com, Feb 2 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/13cfcec6840000
Project Member

Comment 28 by 42576172...@developer.gserviceaccount.com, Feb 2 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/12a63976840000
Project Member

Comment 29 by 42576172...@developer.gserviceaccount.com, Feb 2 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14780f46840000
Project Member

Comment 30 by 42576172...@developer.gserviceaccount.com, Feb 2 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14d6604e840000

Comment 31 by sullivan@chromium.org, Feb 2 2018 

yhirano: can you take a look at this bug? I know it's pretty noisy, but the bisects in #19 and #21 did see a regression at your CL.

Comment 32 by yhirano@chromium.org, Feb 5 2018 

I reverted the change but I would like to keep this issue open for a while for further investigation.

Comment 33 by sullivan@chromium.org, Feb 5 2018

Labels: -Performance-Sheriff
Thanks! I'm taking the Performance-Sheriff label off the bug so we stop pinging.

Comment 34 by yhirano@chromium.org, Feb 7 2018 

--
Project Member

Comment 37 by 42576172...@developer.gserviceaccount.com, Feb 7 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14ac6349840000
Project Member

Comment 38 by 42576172...@developer.gserviceaccount.com, Feb 7 2018 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/11e43221840000

Comment 39 by dtu@chromium.org, Feb 9 2018 

Sorry, I think we're seeing no repros here because of go/catabug/4226

Comment 40 by yhirano@chromium.org, Apr 2 2018

Status: Fixed (was: Assigned)
I reverted the change.
►

Sign in to add a comment