1323059 - 2.1 - 2.83% damp (linux64, windows7-32) regression on push f09e9013044d352d51c3dea33826884188a59622 (Thu Dec 8 2016)

Status: RESOLVED DUPLICATE of bug 1321749

(DevTools :: Netmonitor, defect)

Description

[Joel Maher ( :jmaher ) (UTC -8)]

Talos has detected a Firefox performance regression from push f09e9013044d352d51c3dea33826884188a59622. As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

  3%  damp summary windows7-32 pgo e10s     229.47 -> 235.96
  2%  damp summary linux64 pgo e10s         251.87 -> 257.81
  2%  damp summary linux64 opt e10s         297.69 -> 303.95


You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=4486

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

[Joel Maher ( :jmaher ) (UTC -8)]

I did a compare view to the previous changeset:
https://treeherder.mozilla.org/perf.html#/compare?originalProject=mozilla-inbound&originalRevision=9a03728016ff&newProject=mozilla-inbound&newRevision=28445d6f646b&framework=1&filter=damp&showOnlyImportant=0

while there is nothing showing >2%, you can see that linux64 (the alert we saw and filed this bug for), is really close to 2%.  In fact the netmonitor tests show the regression nicely:
https://treeherder.mozilla.org/perf.html#/comparesubtest?originalProject=mozilla-inbound&originalRevision=9a03728016ff&newProject=mozilla-inbound&newRevision=28445d6f646b&originalSignature=edaec66500db21d37602c99daa61ac983f21a6ac&newSignature=edaec66500db21d37602c99daa61ac983f21a6ac&framework=1

this is no surprise since the patch in question was about modifying netmonitor.

:jsnajdr, can you take a look at this and see if it makes sense to see this regression based on your patch?  Possibly there is something we can do to reduce the performance regression, or maybe we need to accept it.

Comment 2

[Joel Maher ( :jmaher ) (UTC -8)]

as a note, the damp test regression,improved, and regressed again, following the pattern of the patch as it was backed out and relanded.

Comment 4

[Liz Henry (:lizzard) (relman/hg->git project)]

Marking this fix-optional to get it out of triage; we are tracking it in bug 1321749.