1481837 - 2.65% tresize (osx-10-10) regression on push b5ba45e6ac73574c9e6bee886929f684e39de2c8 (Tue Aug 7 2018)

Status: RESOLVED FIXED

(Firefox :: Address Bar, defect, P1)

Description

[Ionuț Goldan [:igoldan]]

Talos has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/autoland/pushloghtml?changeset=b5ba45e6ac73574c9e6bee886929f684e39de2c8

As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

  3%  tresize osx-10-10 opt e10s stylo     27.43 -> 28.15

Improvements:

  2%  tresize linux64-qr opt e10s stylo     13.18 -> 12.88


You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=14805

On the page above you can see an alert for each affected platform as well as a link to a graph showing the history of scores for this test. There is also a link to a treeherder page showing the Talos jobs in a pushlog format.

To learn more about the regressing test(s), please see: https://wiki.mozilla.org/Buildbot/Talos/Tests

For information on reproducing and debugging the regression, either on try or locally, see: https://wiki.mozilla.org/Buildbot/Talos/Running

*** Please let us know your plans within 3 business days, or the offending patch(es) will be backed out! ***

Our wiki page outlines the common responses and expectations: https://wiki.mozilla.org/Buildbot/Talos/RegressionBugsHandling

Comment 1

[Ionuț Goldan [:igoldan]]

Needinfo?ing :adw, as :mak is on PTO.

Comment 2

[Drew Willcoxon :adw]

The patch for bug 1480349 did add a resize listener, so this is probably a legit regression.  As I understand it, this bug is new in 63 since the fix for the bug that caused it, bug 1419391, landed in 63.

Maybe we should back this out for now, even if that means bug 1480349 may end up in 63.  Marco is away for a while.  I can take a look at it, with the goal of re-landing it in time for 63 without regressing talos.  I'm not sure when I'll be able to get to it, maybe later this week or next.

Comment 3

[Ionuț Goldan [:igoldan]]

:adw to provide you with some assistance, here are the Gecko profiles for tresize, on OS X:

before: https://perf-html.io/from-url/https%3A%2F%2Fqueue.taskcluster.net%2Fv1%2Ftask%2FYUCdTxISRDGdI38gVBke3g%2Fruns%2F0%2Fartifacts%2Fpublic%2Ftest_info%2Fprofile_tresize.zip

after bug 1480349 landed: https://perf-html.io/from-url/https%3A%2F%2Fqueue.taskcluster.net%2Fv1%2Ftask%2FKBaPUHrrQ8ithpcyuuYqFQ%2Fruns%2F0%2Fartifacts%2Fpublic%2Ftest_info%2Fprofile_tresize.zip

Comment 4

[Marco Bonardo [:mak] (Away 10-21 August)]

Are you sure this was on the relanding and not on the FIRST landing? that bug has been pushed, backed out and pushed again with a slightly different behavior.

Comment 5

[Marco Bonardo [:mak] (Away 10-21 August)]

ok, the graph looks clear. going back to PTO and thanks for looking into this :)

Comment 6

[Drew Willcoxon :adw]

Let's close this bug -- the patch was backed out, so the regression was fixed -- and work on a new patch in the reopened bug 1480349.