Closed Bug 1316910 Opened 8 years ago Closed 8 years ago

3% damp (windows7-32) regression on push b3ad07be866f (Thu Nov 10 2016)

Categories

(Firefox :: Untriaged, defect, P1)

52 Branch
defect

Tracking

()

RESOLVED FIXED
Iteration:
53.1 - Nov 28
Tracking Status
firefox52 --- unaffected

People

(Reporter: jmaher, Unassigned)

References

Details

(Keywords: perf, regression, talos-regression, Whiteboard: [devtools-html])

Talos has detected a Firefox performance regression from push b3ad07be866f. 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 257.08 -> 264.8 You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=4078 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
I did the opt retriggers, but pgo had only one data point, thanks for a few retriggers!
added a few more test types here, specifically win7/8 opt.
Maybe unrelated, but since the end of last week we seem to have a spike in timeout intermittents on debug platforms. See for instance the failures in linux debug for the following job https://treeherder.mozilla.org/#/jobs?repo=try&revision=76f159e59c4ed6fbaa1cdc3264eaf2471d42a642
Flags: qe-verify-
Priority: -- → P2
Whiteboard: [devtools-html]
Joel, trying to narrow down if this is affecting e10s. I see win8 64 e10s in the alert summary, but when I try to look at details on the test, it looks like there are 0 g2 runs on the autolanded commit (https://treeherder.mozilla.org/perf.html#/comparesubtest?originalProject=autoland&originalRevision=827b1805b72469f2612649b41df015359777fe0a&newProject=autoland&newRevision=b3ad07be866f0d891fdfca9b8e6077431b5b52a7&originalSignature=f1b8974c1b07db58b52be38dbc65a8d6f44e8d53&newSignature=f1b8974c1b07db58b52be38dbc65a8d6f44e8d53&framework=1). Then from treeherder I don't see any g2 tests and don't know how to add them (https://treeherder.mozilla.org/#/jobs?repo=autoland&revision=b3ad07be866f0d891fdfca9b8e6077431b5b52a7&selectedJob=6462063). Does it not matter if there are runs on the commit itself to the system detecting regressions?
Flags: needinfo?(jmaher)
:bgrins, I assumed the win8 one was related, but you did point out a flaw in my work! I have went into the buildbot api to get the g2 jobs scheduled for win8 (opt + e10s), so in another hour or two there will be all proper data.
Flags: needinfo?(jmaher)
Priority: P2 → P1
As a note, this regression patch has been merged into Aurora, so please make sure to uplift any fixes to Aurora, thank you. For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=4129
Bug 1310416 has been backed out and reopened, so we should be expecting this regression to disappear now
confirmed on graph server. We can keep this open if there is a need to track any performance work prior to relanding- otherwise closing this helps keep lists of regressions reduced.
Let's close it
Flags: needinfo?(ttromey)
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → FIXED
Iteration: --- → 53.1 - Nov 28
You need to log in before you can comment on or make changes to this bug.