1747197 - 46.42 - 2.05% welcome fcp / welcome PerceptualSpeedIndex + 5 more (Linux, OSX) regression on Tue December 14 2021

Status: RESOLVED WONTFIX

(Firefox :: Messaging System, defect)

Description

[Alexandru Ionescu (needinfo me) [:alexandrui]]

Perfherder has detected a browsertime performance regression from push 0331c56022c9747db6b9abeb3ab6920189cb1044. 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)
46%	welcome fcp	macosx1014-64-shippable-qr	cold webrender	123.56 -> 180.92
46%	welcome fcp	macosx1014-64-shippable-qr	cold webrender	117.48 -> 171.04
4%	welcome ContentfulSpeedIndex	linux1804-64-shippable-qr	cold webrender	1,469.17 -> 1,524.67
4%	welcome SpeedIndex	macosx1014-64-shippable-qr	cold webrender	1,453.17 -> 1,505.50
3%	welcome ContentfulSpeedIndex	macosx1015-64-shippable-qr	cold webrender	1,539.62 -> 1,590.83
3%	welcome SpeedIndex	linux1804-64-shippable-qr	cold webrender	1,243.92 -> 1,283.67
2%	welcome PerceptualSpeedIndex	macosx1014-64-shippable-qr	cold webrender	1,585.38 -> 1,617.88

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)
7%	amazon PerceptualSpeedIndex	macosx1014-64-shippable-qr	cold fission webrender	893.46 -> 835.08
3%	welcome ContentfulSpeedIndex	macosx1014-64-shippable-qr	warm webrender	1,383.89 -> 1,343.00

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 1

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

Set release status flags based on info from the regressing bug 1705799

Comment 2

[Punam Dahiya [:pdahiya]]

Patch switched jpeg background image used on about:welcome to avif. NI @jbauman for feedback on regression reported due to avif switch and if thats expected. Thanks

Comment 3

[Jon Bauman [:jbauman:]]

I'm not very familiar with these tests. If I'm understanding correctly this is indicating an ~50 millisecond regression. Given the relative sizes of the JPEG (~200KB) and the AVIF which replaced it (~100KB) and the comparative decoding performance of the two formats, I'd say this is expected and acceptable.

Comment 4

[Alexandru Ionescu (needinfo me) [:alexandrui]]

Given that 100ms change is only perceptible by the human eyes, it's ok to close as wontfix.