Closed Bug 1406896 Opened 7 years ago Closed 7 years ago

3.52 - 6.68% tart (linux64, windows10-64, windows7-32) regression on push 40ef1df3618b (Sat Oct 7 2017)

Categories

(Firefox :: Theme, defect)

defect
Not set
normal

Tracking

()

RESOLVED FIXED
Firefox 58
Tracking Status
firefox-esr52 --- unaffected
firefox56 --- unaffected
firefox57 --- unaffected
firefox58 --- fixed

People

(Reporter: igoldan, Assigned: johannh)

References

Details

(Keywords: perf, regression, talos-regression)

Talos has detected a Firefox performance regression from push: https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=c18b5243c8705c4b73319d346e8628ab5dcb9228&tochange=40ef1df3618bc5ec673464bfac9187f8b62ccae4 As author of one of the patches included in that push, we need your help to address this regression. Regressions: 7% tart summary windows10-64 opt e10s 4.10 -> 4.37 6% tart summary windows10-64 pgo e10s 3.51 -> 3.70 5% tart summary linux64 pgo e10s 4.42 -> 4.63 4% tart summary windows7-32 pgo e10s 3.91 -> 4.06 4% tart summary linux64 opt e10s 4.96 -> 5.14 4% tart summary windows7-32 opt e10s 4.85 -> 5.02 You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=9873 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
:johannh I see you are the owner for bug 1391539. Can we resolve this regression or should we accept it?
Flags: needinfo?(jhofmann)
I'm leaning towards this is something we have to accept, but I'd like to narrow the cause for this regression down a little further before committing to it. I'll do a Talos push for bug 1406691 which prevents tabs from re-scaling and potentially another few pushes to figure out where this is coming from. Thanks!
Let's see if there are any Talos follow-ups to bug 1406691...
Depends on: 1406691
Flags: needinfo?(jhofmann)
Good news! Push from https://bugzilla.mozilla.org/show_bug.cgi?id=1406691#c44 fixed this regression: == Change summary for alert #9994 (as of October 12 2017 12:30 UTC) == Improvements: 5% tart summary linux64 pgo e10s 4.62 -> 4.41 5% tart summary linux64 opt e10s 5.23 -> 4.99 4% tart summary windows7-32 opt e10s 5.04 -> 4.84 3% tart summary windows10-64 opt e10s4.38 -> 4.27 2% tart summary windows7-32 pgo e10s 4.02 -> 3.95 For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=9994
Unfortunately, on Windows 10 the regression was partially fixed, as you can see here: https://treeherder.mozilla.org/perf.html#/graphs?series=autoland,1500250,1,1&series=autoland,1501631,1,1 :johannh Do you know why this happens specifically on Win10 platform?
Flags: needinfo?(jhofmann)
No, I have no idea. I don't think I can narrow this down significantly further right now, and I'm not sure to what end (this behavior isn't really optional and the regression is very slight). I'm proposing to close this bug since we resolved the main impact. Does anyone disagree?
Flags: needinfo?(jhofmann)
Looks like some other bugs *almost* resolved the Windows 10 regressions. This made the performance baseline resemble its original state, before the regression landed. I will mark this as RESOLVED.
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
Assignee: nobody → jhofmann
Target Milestone: --- → Firefox 58
You need to log in before you can comment on or make changes to this bug.