1711970 - 6.97 - 6.96% allrecipes loadtime / allrecipes loadtime (Android) regression on Tue May 11 2021

Status: RESOLVED WONTFIX

(Core :: JavaScript: GC, defect, P5)

Description

[Florin Strugariu [:Bebe]]

Perfherder has detected a browsertime performance regression from push f7b553bd9c0b77b0130d377f9a9c696c8c611a01. As author of one of the patches included in that push, we need your help to address this regression.

Regressions:

Ratio	Suite	Test	Platform	Options	Absolute values (old vs new)
7%	allrecipes	loadtime	android-hw-g5-7-0-arm7-api-16-shippable	warm	3,377.42 -> 3,612.71
7%	allrecipes	loadtime	android-hw-g5-7-0-arm7-api-16-shippable	warm	3,374.88 -> 3,609.79

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 1694538

Comment 2

[Steve Fink [:sfink] [:s:]]

Looks fine to me. The allrecipes regression is balanced with an improvement for espn, and I would accept a fair amount of regression in order to keep the code simplification. (There is a followup correctness fix, but I doubt it'll impact perf.)

In general, I would expect this to improve perf, though perhaps not noticeably. The most likely reason it could regress things is that it's a little more conservative, so could hold onto some memory for longer in edge cases. That said, it doesn't seem to make much of a perf difference either way.

Comment 3

[Steve Fink [:sfink] [:s:]]

We accepted this regression.