1382663 - 1.63 - 3.85% cart (windows7-32) regression on push 5b29bf0b577c (Mon Jul 17 2017)

Status: RESOLVED INVALID

(Core :: Graphics: Layers, defect)

Description

[Ionuț Goldan [:igoldan]]

Talos has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?changeset=8f7663ac2068ae33e99bb2ce637b98a9f9c283a2

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

Regressions:

  4%  cart summary windows7-32 pgo e10s     8.53 -> 8.86
  2%  cart summary windows7-32 opt e10s     10.92 -> 11.10


You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=8038

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]]

:dvander Is there something we can do to resolve this Win7 performance regression?

Comment 2

[Milan Sreckovic [:milan] (needinfo for best results)]

That pushlog link points to bug 1366361?

Comment 3

[Ionuț Goldan [:igoldan]]

No. That highlighted alert was detected automatically, but the real one if from the previous changeset (5b29bf0b577c).

Comment 4

[Ionuț Goldan [:igoldan]]

The link for it: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=2fee42d106b3bf5ea490154137ce5a65a4d60065&tochange=5b29bf0b577c67d9486aa95a09f36a39c0a67754

Comment 5

[Mike Conley (:mconley) (:⚙️)]

FWIW, I'm very skeptical about the cart test on Nightly - especially since I don't think it's measuring anything very useful on Nightly these days (though it is when Photon is disabled). See bug 1381037 for details.

Comment 6

[David Anderson [:dvander] - inactive, e-mail if emergency]

Advanced Layers isn't riding the 56 train, so if we're going to disable CART on that release then it sounds like this issue will resolve itself?

Comment 7

[David Anderson [:dvander] - inactive, e-mail if emergency]

I'm going to see if anything obvious comes up in a profile, just in case.

Comment 8

[Ionuț Goldan [:igoldan]]

David, have you reached any conclusion? If not, I'm thinking of marking this bug as INVALID.
Mike, are you ok with that?

Comment 9

[Milan Sreckovic [:milan] (needinfo for best results)]

In person, David and I talked about making this invalid, as in, taking the regression, but I will let him formally put that in the bug.

Comment 10

[Mike Conley (:mconley) (:⚙️)]

(In reply to Ionuț Goldan [:igoldan], Performance Sheriffing from comment #8)
> David, have you reached any conclusion? If not, I'm thinking of marking this
> bug as INVALID.
> Mike, are you ok with that?

Yep, this sounds right to me.

Comment 11

[David Anderson [:dvander] - inactive, e-mail if emergency]

Yeah, I'm okay with that. I haven't had time to investigate yet as I need to get a Windows 7 environment going.