1867071 - 0.16% installer size (OSX) regression on Mon November 20 2023

Status: RESOLVED WONTFIX

(Core :: Graphics: WebGPU, defect)

Description

[Alex Finder]

Perfherder has detected a build_metrics performance regression from push 9763af9939468e874ac0d5f1c4ad2016e12ceafc. 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)
0.16%	installer size	osx-aarch64-shippable	aarch64 nightly	84,216,501.08 -> 84,353,684.42

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.

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 1865585

Comment 2

[Erich Gubler [:ErichDonGubler]]

Alternatively requesting NI from :jimb.

Comment 3

[Donal Meehan [:dmeehan]]

:ErichDonGubler is this only nightly impact or will it impact when Fx122 goes to beta/release? Looks like Nical hasn't looked at it?
Fx122 soft code freeze starts next week

Comment 4

[Erich Gubler [:ErichDonGubler]]

:dmeehan: AFAIK, this will also impact release. I've pinged :nical and :jimb in our team channel to elicit a response here.

Comment 5

[Nicolas Silva [:nical]]

This is happening in an upstream dependency. I'll look at whether there are low hanging fruits to bring the binary size of the upstream dep but we can't afford to back the update out.

Comment 6

[Donal Meehan [:dmeehan]]

:nical any updates since Comment 5?
The Fx122 nightly soft code freeze starts on 2023-12-14, with Fx122 going to beta on 2023-12-18.

Comment 7

[Nicolas Silva [:nical]]

Sorry I should have reported back. There wasn't a localized regression in the code size, the extra bloat is smeared all over wgpu due to a large refactoring so we couldn't find low hanging fruits to pick and will have to live with the regression for now.

Comment 8

[Donal Meehan [:dmeehan]]

(In reply to Nicolas Silva [:nical] from comment #7)

Sorry I should have reported back. There wasn't a localized regression in the code size, the extra bloat is smeared all over wgpu due to a large refactoring so we couldn't find low hanging fruits to pick and will have to live with the regression for now.

I'll set 122 to wontfix, but guessing you would like to resolve this as wontfix also?