1958241 - 3.61 - 0.36% Base Content Heap Unclassified / Base Content JS (Windows) regression on Wed April 2 2025

Status: RESOLVED WONTFIX

(Core :: Privacy: Anti-Tracking, defect, P3)

Description

[Treeherder Bug Filer]

Perfherder has detected a awsy performance regression from push 1dcb66eab4412818d5d0bbf36e8501d1b75b850d. 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)
4%	Base Content Heap Unclassified	windows11-64-24h2-shippable	fission	1,949,907.33 -> 2,020,216.67
0.36%	Base Content JS	windows11-64-24h2-shippable	fission	1,540,629.33 -> 1,546,178.67

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 44574

The following documentation link provides more information about this command.

Comment 1

[Tooru Fujisawa [:arai]]

The push contains many patches: https://hg-edge.mozilla.org/integration/autoland/pushloghtml?changeset=1dcb66eab4412818d5d0bbf36e8501d1b75b850d
And as far as I can see, bug 1956976 patch can also affect awsy score.

Do we have comparison data between each revision inside the push?

Comment 2

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

Set release status flags based on info from the regressing bug 1948288

Comment 3

[Tooru Fujisawa [:arai]]

I'll trigger some try runs to see where the regression happens in the stack,
but in any case, if there's a fix necessary, this will need to be coordinated with the version bump.

:twisniewski, can I have your opinion for how to address this?

Comment 4

[Thomas Wisniewski [:twisniewski]]

I don't believe there is anything we can really do here. If it's indeed this stack causing increased RAM usage, it's almost certainly caused simply by the overhead of us doing more during the webcompat addon's startup (registering more content scripts to do webcompat fixes on sites). I don't see us wanting to stop doing the things the webcompat addon is doing, so if this much RAM usage increase is truly considered a problem, we will likely need the folks maintaining the webextensions framework to optimize the RAM usage of this scenario in general (since we're going to be adding more webcompat interventions for the foreseeable future).

Comment 5

[Tooru Fujisawa [:arai]]

so far, the regression comes from other patch in the patch stack than bug 1948288:

before vs Bug 195697 to Bug 1902403: https://perf.compare/compare-results?baseRev=65ac3a586161069bd14f6ce4fca87d3f2f286c03&baseRepo=try&newRev=4a2acd6edac07986b695a439960cd41c76997ac0&newRepo=try&framework=4

Bug 1902403 vs Bug 1948288: https://perf.compare/compare-results?baseRev=4a2acd6edac07986b695a439960cd41c76997ac0&baseRepo=try&newRev=6af74049924fe0cbf7accf36b4fcce0dacde7102&newRepo=try&framework=4

Bug 1957506 and bug 1934189 are test-only, and shouldn't affect.
Bug 1844503, bug 1894682, bug 1786404, and bug 1886591 are removal of the rules, so this won't result in increase, unless there's something unusual happening.
Bug 1902403 is addition of the rule, but this would most likely compensate with the above removal.
Bug 1956976 adds an internal functionality, and this may be causing something, but so far I don't find anything that can directly contribute to "heap unclassified", except for something in messagemanager, or maybe something around script source handling in JS engine.

Anyway, if necessary, I'll trigger some more try runs to bisect.

Comment 10

[Thomas Wisniewski [:twisniewski]]

With luck this will be fixed by the patch which just landed in bug 1960374.

Comment 12

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

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

:arai, since you are the author of the regressor, bug 1948288, 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 13

[Thomas Wisniewski [:twisniewski]]

If this was not solved by bug 1960374, and it isn't caused by this bug (as implied by comment 5), then there is nothing else we can really do here, as the rest of the patches in the stack were for webcompat fixes. So we might as well just close this bug as WONTFIX.