1867258 - 4.86% google-slides loadtime (Linux) regression on Mon November 27 2023

Status: RESOLVED FIXED

(Core :: Layout, defect)

Description

[Acasandrei Beatrice (needinfo me)]

Perfherder has detected a browsertime performance regression from push 95679f6ce2544b21ec9cfd0d19961058d60155d6. 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)	Performance Profiles
5%	google-slides loadtime	linux1804-64-shippable-qr	bytecode-cached fission warm webrender	922.81 -> 967.63	Before/After

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)	Performance Profiles
51%	cnn-nav.world FirstVisualChange	macosx1015-64-shippable-qr	cold fission webrender	316.00 -> 154.38
51%	cnn-nav.world FirstVisualChange	windows10-64-shippable-qr	cold fission webrender	352.25 -> 173.73
47%	facebook-nav.marketplace FirstVisualChange	macosx1015-64-shippable-qr	cold fission webrender	279.61 -> 147.23
47%	facebook-nav.marketplace FirstVisualChange	windows10-64-shippable-qr	cold fission webrender	249.48 -> 131.38
41%	cnn-nav.world SpeedIndex	linux1804-64-shippable-qr	cold fission webrender	639.22 -> 379.37
...	...	...	...	...	...
5%	google-slides PerceptualSpeedIndex	windows10-64-shippable-qr	bytecode-cached fission warm webrender	464.87 -> 442.29

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.

If you need the profiling jobs you can trigger them yourself from treeherder job view or ask a sheriff to do that for you.

For more information on performance sheriffing please see our FAQ.

Comment 1

[Emilio Cobos Álvarez [:emilio]]

Seems the graph has recovered after bug 1867042 landed, can you confirm?

Comment 2

[Acasandrei Beatrice (needinfo me)]

(In reply to Emilio Cobos Álvarez (:emilio) from comment #1)

Seems the graph has recovered after bug 1867042 landed, can you confirm?

It looks that way, yes. I've backfilled the missing datapoints from the graph, we should get an alert soon, if not, I will trigger one manually. Thanks.

Comment 3

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

:fredw, since you are the author of the regressor, bug 1807253, could you take a look?

For more information, please visit BugBot documentation.

Comment 4

[Frédéric Wang Nélar]

BugBot is asking me again, so here are my not-helpful analysis from yesterday. It looks like we have many perf improvements (not sure it's related to my patch though) including for google-slides tests. The one regression is specifically about google-slides loadtime, which seems really related to bug 1867042 (doing the new work too early). I had checked the graph and as mentioned above it seems it has recovered after the fix for this bug landed. I understand from comment 2 there is nothing more to do from my side for now, just pending confirmation that things have recovered.

Comment 5

[Emilio Cobos Álvarez [:emilio]]

Yeah the graph is pretty clear.