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

Issue 776745 link

Starred by 1 user
Status: WontFix
Owner:
kerrnel@chromium.org
Closed: Oct 2017
Cc:
kerrnel@chromium.org
█ pmeenan@chromium.org
rsesek@chromium.org
█ benhenry@chromium.org
█ holte@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

3.9%-4.3% regression in rasterize_and_record_micro.top_25 at 508438:508569

Project Member Reported by pmeenan@chromium.org, Oct 20 2017 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Oct 20 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=776745

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=460b3288fa8df7a4e824cf97223d1870b2cb438d5b299328b955143ac2c0a1fd


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

chromium-rel-mac11-air
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Oct 20 2017 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/14a584b7780000

Comment 4 by simonhatch@chromium.org, Oct 20 2017 

Broke some jobs with a dev version, restarting.
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Oct 20 2017 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/11d3075b780000
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Oct 23 2017

Cc: rsesek@chromium.org kerrnel@chromium.org holte@chromium.org
Owner: kerrnel@chromium.org
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14b5af80f80000

Enable Finch Trial Test for MacV2Sandbox.
By kerrnel@chromium.org · Thu Oct 12 22:48:50 2017
chromium @ fcd7b94faa40dfc3e4e4be26f4e854dd14ef650c

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

Comment 9 by kerrnel@chromium.org, Oct 24 2017

Cc: benhenry@chromium.org
pmeenan@, in this case, is it correct to say that the regression only affects 10.11?

I pulled the graphs for 10.12 and don't see any slowdown:
https://chromeperf.appspot.com/report?sid=24138a6df2174455f0985492339a2ee2eb8b59729e892b7e37b0b0146ccb24a0

Understanding if it only affects 10.11 will be really helpful in understanding where the problem is, and how many users it actually affects.

Comment 10 by pmeenan@chromium.org, Oct 24 2017 

Possibly though it's hard to tell with this test and the bots because the change is pretty subtle and the benchmark is a bit noisy and the 10.11 impact looks like it is mostly on the Air and not the other devices.

If you squint really hard 10.12 and the other 10.11 devices look like they may be at a slightly higher level but it's really subtle.

https://chromeperf.appspot.com/report?sid=1d428d34bdc514a9d2519a8ea9a88d9bb24624ed0da882fa88f9afe0ad5abc9b

Comment 11 by kerrnel@chromium.org, Oct 26 2017 

Looking at the graphs, it looks like there is actually some change in chromium-rel-mac12, right?

https://chromeperf.appspot.com/report?sid=1d428d34bdc514a9d2519a8ea9a88d9bb24624ed0da882fa88f9afe0ad5abc9b

My question though is if the change from 4.45ms to 4.55ms is of any actual significance or real impact on user experience?

Comment 12 by kerrnel@chromium.org, Oct 27 2017

Status: WontFix (was: Assigned)
I talked to Ben. Since this is a .1ms regression across the board, and please correct me if that's wrong, I think this is a WontFix.
►

Sign in to add a comment