1556412 - 3.2 - 4.09% raptor-stylebench-firefox (linux64-shippable, linux64-shippable-qr) regression on push 899d0c1363e0f250406ca7f0a02ad798726abb94 (Thu May 30 2019)

Status: RESOLVED DUPLICATE of bug 1555867

(Core :: CSS Parsing and Computation, defect)

Description

[Florin Strugariu [:Bebe]]

Raptor has detected a Firefox performance regression from push:

https://hg.mozilla.org/integration/autoland/pushloghtml?changeset=2428766459c6339fcafdc437dfb45beba59e6de2

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

Regressions:

4% raptor-stylebench-firefox linux64-shippable opt 43.10 -> 41.28
4% raptor-stylebench-firefox linux64-shippable opt 43.12 -> 41.36
3% raptor-stylebench-firefox linux64-shippable-qr opt 42.15 -> 40.80

You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=21247

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

[Emilio Cobos Álvarez (:emilio)]

This is sort-of expected given the change, but I'll take a look at whether we can optimize some of it away after bug 1555867 lands.

Comment 2

[Bobby Holley (:bholley)]

(In reply to Emilio Cobos Álvarez (:emilio) from comment #1)

This is sort-of expected given the change, but I'll take a look at whether we can optimize some of it away after bug 1555867 lands.

The premise in bug 1493420 was that only microbenchmarks would be affected, whereas I don't think we think of StyleBench as a microbenchmark.

Would be good to understand:

• What the Windows graphs look like.
• Whether bug 1555867 fixes this.

Comment 3

[Emilio Cobos Álvarez (:emilio)]

Yes, in particular, I don't think that my push to talos in bug 1493420 regressed stylebench. But that was a long while ago.

Comment 4

[Florin Strugariu [:Bebe]]

== Change summary for alert #21282 (as of Tue, 04 Jun 2019 09:21:50 GMT) ==

Improvements:

3% raptor-stylebench-firefox linux64-shippable-qr opt 40.78 -> 41.83
2% raptor-stylebench-firefox linux64-shippable opt 41.52 -> 42.54

For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=21282

Comment 5

[Emilio Cobos Álvarez (:emilio)]

Great, thanks... I can do another round of profiling of stylebench, but that's mostly the same magnitude as the regression.