1991421 - 25.45 - 2.1% speedometer3 Editor-TipTap/Long/Sync / speedometer3 NewsSite-Nuxt/total + 58 more (Android, Linux, OSX, Windows) regression on Tue September 23 2025

Status: RESOLVED DUPLICATE of bug 1987997

(Core :: Layout: Text and Fonts, defect, P3)

Description

[Treeherder Bug Filer]

Perfherder has detected a browsertime performance regression from push ff7bf296a64e46a6e33cc3cd0ebe683d8ae8556f. As author of one of the patches included in that push, we need your help to address this regression.

Please acknowledge, and begin investigating this alert within 3 business days, or the patch(es) may be backed out in accordance with our regression policy. Our guide to handling regression bugs has information about how you can proceed with this investigation.

If you have any questions or need any help with the investigation, please reach out to [e-mail]. Alternatively, you can find help on Slack by joining #perf-help, and on Matrix you can find help by joining #perftest.

Regressions:

Ratio	Test	Platform	Options	Absolute values (old vs new)	Performance Profiles
25%	speedometer3 Editor-TipTap/Long/Sync	macosx1500-aarch64-shippable	fission webrender	24.95 -> 31.30	Before/After
25%	speedometer3 Editor-TipTap/Long/total	macosx1500-aarch64-shippable	fission webrender	25.58 -> 31.97	Before/After
23%	speedometer3 Editor-TipTap/Long/Sync	windows11-64-24h2-shippable	fission webrender	38.31 -> 47.06
23%	speedometer3 Editor-TipTap/total	macosx1500-aarch64-shippable	fission webrender	47.28 -> 57.94	Before/After
22%	speedometer3 Editor-TipTap/Long/total	windows11-64-24h2-shippable	fission webrender	39.87 -> 48.71
22%	speedometer3 Editor-TipTap/Long/Sync	linux1804-64-shippable-qr	fission webrender	82.92 -> 101.30	Before/After
22%	speedometer3 Editor-TipTap/Long/total	linux1804-64-shippable-qr	fission webrender	86.50 -> 105.09	Before/After
21%	speedometer3 Editor-TipTap/Highlight/Sync	macosx1500-aarch64-shippable	fission webrender	20.35 -> 24.62	Before/After
21%	speedometer3 Editor-TipTap/Highlight/Sync	windows11-64-24h2-shippable	fission webrender	30.65 -> 37.00
20%	speedometer3 Editor-TipTap/total	windows11-64-24h2-shippable	fission webrender	73.92 -> 88.89
...	...	...	...	...	...
2%	speedometer jQuery-TodoMVC/CompletingAllItems/Async	linux1804-64-shippable-qr	fission webrender	3.04 -> 3.12	Before/After
2%	speedometer VanillaJS-TodoMVC/Adding100Items/Async	linux1804-64-shippable-qr	fission webrender	2.90 -> 2.97	Before/After
2%	speedometer3 NewsSite-Nuxt/NavigateToWorld/Sync	macosx1500-aarch64-shippable	fission webrender	4.28 -> 4.37	Before/After
2%	speedometer3 NewsSite-Nuxt/NavigateToUS/Async	macosx1500-aarch64-shippable	fission webrender	9.43 -> 9.64
2%	speedometer3 NewsSite-Nuxt/total	macosx1500-aarch64-shippable	fission webrender	39.91 -> 40.75	Before/After

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)	Performance Profiles
2%	speedometer3 perfstats-MinorGC	macosx1500-aarch64-shippable	fission webrender	267.93 -> 261.82	Before/After

Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests.

If you need the profiling jobs you can trigger them yourself from treeherder job view or ask [e-mail] to do that for you.

You can run all of these tests on try with ./mach try perf --alert 46914

The following documentation link provides more information about this command.

Comment 1

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

Set release status flags based on info from the regressing bug 1988596

Comment 2

[Markus Stange [:mstange]]

I think this is a nightly-only regression, and it was expected in bug 1988596. We expect this to be fixed by bug 1987997 (once it lands).

Comment 3

[Ting-Yu Lin [:TYLin] (PST, UTC-8)]

Here is a performance comparison of current Nightly vs Nightly + bug 1987997:

• MacOS
• Windows

Bug 1987997 will bring the performance of these tests on par with Nightly build with text-autospace feature disabled.

Comment 4

[Emilio Cobos Álvarez [:emilio]]

Can we land bug 1987997 right away? Having speedometer regressions lingering around is slightly unfortunate.

Comment 5

[Celeste Liu [:Coelacanthus]]

(In reply to Emilio Cobos Álvarez [:emilio] from comment #4)

Can we land bug 1987997 right away? Having speedometer regressions lingering around is slightly unfortunate.

I think it's not a good idea. I think Jonathan Kew [:jfkthame] introduced this time period to see whether we can optimize the code and make perf regression small enough to land text-autospace: normal directly, and some people have been optimizing recently and have gained perf improvement indeed. For example, in https://bugzilla.mozilla.org/show_bug.cgi?id=1991943, Jonathan Kew [:jfkthame] improved perf for around 10~13% for Editor-TipTap testcase, which is the main issue reported in this perf regression. I think we should be patient to see if we can get more perf improvement before we reach the day to land text-autospace: no-autospace.

Comment 6

[Ting-Yu Lin [:TYLin] (PST, UTC-8)]

(In reply to Emilio Cobos Álvarez [:emilio] from comment #4)

Can we land bug 1987997 right away? Having speedometer regressions lingering around is slightly unfortunate.

We have scheduled to land bug 1987997 in 10 days. Jonathan, what do you think? Do you have any other ideas to improve the performance?

Comment 7

[Jonathan Kew [:jfkthame]]

We've landed some improvements in bug 1991943 that partially mitigated the regressions here. Perfherder doesn't seem to have reported on the changes yet, but graphs like https://treeherder.mozilla.org/perfherder/graphs?timerange=1209600&series=autoland,5352629,1,13 clearly show the gain.

Regarding bug 1987997, which will fully address the reported regression, I'd prefer to stick with our existing plan to land after another week or so. This is not just about observing (and potentially working on) the performance issues, but also to give a window of opportunity for any feedback from Nightly users on the normal value of the property, given the open questions regarding what should be spec'd/shipped as the initial value.

Comment 8

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

It has been over 7 days with no activity on this performance regression.

:TYLin, since you are the author of the regressor, bug 1988596, which triggered this performance alert, could you please provide a progress update?

If this regression is something that fixes a bug, changes the baseline of the regression metrics, or otherwise will not be fixed, please consider closing it as WONTFIX. See this documentation for more information on how to handle regressions.

For additional information/help, please needinfo the performance sheriff who filed this alert (they can be found in comment #0), or reach out in #perftest, or #perfsheriffs on Element.

For more information, please visit BugBot documentation.

Comment 9

[Ting-Yu Lin [:TYLin] (PST, UTC-8)]

text-autospace is not enabled on Firefox 145, so changing the tracking status to disabled. I'm planning to land bug 1987997 once I have the discussion with the team on Oct 14.

Comment 10

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

Set release status flags based on info from the regressing bug 1988596

Comment 11

[Ting-Yu Lin [:TYLin] (PST, UTC-8)]

This regression is fixed by bug 1987997. Resolving as a duplicate.