1545015 - 2.9 - 3.25% raptor-tp6m-espn-geckoview / raptor-tp6m-espn-geckoview fcp (android-hw-p2-8-0-android-aarch64) regression on push 8eaa269b362fd429c04779eaff8e2e7cd0075ad7 (Tue Apr 9 2019)

Status: VERIFIED FIXED

(Core :: JavaScript Engine, defect)

Description

[Ionuț Goldan [:igoldan]]

Raptor has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?changeset=8eaa269b362fd429c04779eaff8e2e7cd0075ad7

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

Regressions:

3% raptor-tp6m-espn-geckoview fcp android-hw-p2-8-0-android-aarch64 opt 665.92 -> 687.58
3% raptor-tp6m-espn-geckoview android-hw-p2-8-0-android-aarch64 opt 782.61 -> 805.34

You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=20392

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 Raptor jobs in a pushlog format.

To learn more about the regressing test(s) or reproducing them, please see: https://wiki.mozilla.org/Performance_sheriffing/Raptor

*** 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/Performance_sheriffing/Talos/RegressionBugsHandling

Comment 1

[Jeff Walden [:Waldo]]

FYI: I'm traveling visiting family Wednesday to Wednesday now. And while I had planned to work remotely for much of this stretch (probably taking a half day here or there), at the airport in the security line I found myself inadvertently laptop-less. I've attempted to see if someone at home could ship it to me, but this is looking less and less likely by the day given I haven't heard back on the text I sent.

So for the moment, I'm betting it's likely I won't be able to look into this till next Thursday. :-(

Comment 2

[Jeff Walden [:Waldo]]

Back now, looking into the bug 1545015/bug 1545034/bug 1546313 triad that are almost certainly all an algorithmic concern with the column-computation approach. Just so anyone watching knows...

Comment 3

[Jeff Walden [:Waldo]]

The regressing patch ended up being backed out, and a fresh approach (bug 1551916) has not received any reports of regressions similar to this.