1901213 - 10.98% speedometer3 Perf-Dashboard/SelectingRange/Sync (OSX) regression on Tue June 4 2024

Status: RESOLVED INVALID

(Core :: Layout, defect)

Description

[Florin Bilt]

Perfherder has detected a browsertime performance regression from push 653d00ce1d0f45e9ff5b812cb783df0216e6ffb5. As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

Ratio	Test	Platform	Options	Absolute values (old vs new)	Performance Profiles
11%	speedometer3 Perf-Dashboard/SelectingRange/Sync	macosx1400-64-shippable-qr	fission webrender	9.92 -> 11.01	Before/After

Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests. Please follow our guide to handling regression bugs and let us know your plans within 3 business days, or the patch(es) may be backed out in accordance with our regression policy.

If you need the profiling jobs you can trigger them yourself from treeherder job view or ask a sheriff to do that for you.

You can run these tests on try with ./mach try perf --alert 635

For more information on performance sheriffing please see our FAQ.

Comment 1

[Emilio Cobos Álvarez [:emilio]]

Yeah that regression comes from backing out bug 1850834. When it relanded, it went back to the new (better) value.

So I think this is invalid unless I'm missing something?

Comment 2

[Daniel Holbert [:dholbert]]

Yeah, it looks like the "after" regression in the alert here is https://hg.mozilla.org/integration/autoland/rev/653d00ce1d0f45e9ff5b812cb783df0216e6ffb5 which is where we backed out bug 1850834.

So what happened is: we landed bug 1850834, and briefly got an improvement on this subtest, and then had to back it out (losing the improvement, i.e. regressing), and that spawned this perf alert & resulted in this bug. And then later, we re-landed and got the improvement back (see bug 1850834 comment 39).

Comment 3

[Florin Bilt]

Thanks for the update, I have set the alert as won't fix.