1738009 - 74.9% google-slides dcf (Windows) regression on Mon October 25 2021

Status: RESOLVED WONTFIX

(Core :: Performance: General, defect)

Description

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

Perfherder has detected a browsertime performance regression from push 5c29a315aa4aa1a19d5e2d65e5ba57df0c8501a2. 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)
75%	google-slides dcf	windows10-64-shippable-qr	fission warm webrender	726.42 -> 1,270.50

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)
6%	google-slides fcp	windows10-64-shippable-qr	fission warm webrender	229.04 -> 215.21
6%	google-slides fnbpaint	windows10-64-shippable-qr	fission warm webrender	233.77 -> 220.17

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 1736057

Comment 2

[Denis Palmeiro [:denispal]]

Unfortunately, this is a situation where no network delay actually hurts us in testing. We are now utilizing this delay time to full parse scripts off thread so that does not need to happen on the main thread. So there are actually some expected regressions here. Surprised to see that there are actually a significant number of improvements as well, however. Since telemetry is showing good results, I think we might need to live with these CI regressions for the time being. Some of the followup off thread delazification work from stencil should help improve these tests though.