1763768 - 4.14% ebay-kleinanzeigen ContentfulSpeedIndex (Android) regression on Tue March 29 2022

Status: RESOLVED WONTFIX

(Testing :: Performance, defect)

Description

[Alex Finder]

Perfherder has detected a browsertime performance regression from push ddf7a48e3bf19b0b4e8c5d34df1e7d31bbee65c5. 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)
4%	ebay-kleinanzeigen ContentfulSpeedIndex	android-hw-g5-7-0-arm7-shippable-qr	warm webrender	2,249.17 -> 2,342.33

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 offending patch(es) will be backed out in accordance with our regression policy.

For more information on performance sheriffing please see our FAQ.

Comment 2

[BugBot [:suhaib / :marco/ :calixte]]

:emilio, since you are the author of the regressor, bug 1760674, could you take a look?
For more information, please visit auto_nag documentation.

Comment 3

[Emilio Cobos Álvarez [:emilio]]

I'm slightly surprised that patch caused a regression (since it should do generally less work for visited links).

But it's a warm page load test so there might be links to that same page which are considered visited and now are processed a bit faster, causing an extra paint before load.

Unfortunately there's, I think, not too much we can do other than taking it, we need to do the extra work in order to make visited not observable to the page via timing. Given it's only a regression from a single test on a single platform, and it's relatively minor, I think taking this regression is acceptable, but let me know if you disagree.