1395058 - 1.98 - 2.46% damp (linux64, windows7-32) regression on push 8a5a636429ea (Mon Aug 28 2017)

Status: RESOLVED WONTFIX

(Core :: DOM: Navigation, defect)

Description

[Ionuț Goldan [:igoldan]]

Talos has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=0a864923a1b47b2959a4ea949f14a8e7cb396cd1&tochange=8a5a636429ea000a738efd1d9189c2a5b696de8d

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

Regressions:

  2%  damp summary linux64 opt e10s     254.97 -> 261.24
  2%  damp summary windows7-32 opt e10s 265.30 -> 270.54


You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=9058

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]]

Samael, I see you are the owner of bug 1375833. Could you take a look over this perf regression? Is this something we can resolve or should we accept it?

Comment 2

[Samael Wang [:freesamael] (away for now)]

I'm a bit surprised about the number 2.46%, but we're making necessary behavioral changes on reloading in the bug. I reviewed the patch again. The patch is as simple as I can do to achieve the goal, so I believe there's not much room for improvement.

To give you more context, in bug 1326251, we made gecko drop all subframe history entries on reloading. Turns out that was too aggressive and should consider a behavioral regression, so we made another fix in bug 1375833. In other words, we were causing data loss in bug 1326251, and that data loss may have made PageLoader test slightly faster before, but it's not a correct behavior.

Olli, do you have different opinions or comments on this?

Comment 3

[Olli Pettay [:smaug][[e-mail]]]

Yeah, very surprising. Did we see any damp changes when bug 1326251 landed?
I don't think there is much we can do here.

Comment 4

[Andrew Overholt [:overholt]]

So ... WONTFIX?

Comment 5

[Samael Wang [:freesamael] (away for now)]

If no one raises other concerns, yes.