Open Bug 1927052 Opened 12 days ago Updated 6 days ago

2.07 - 1.93% speedometer EmberJS-Debug-TodoMVC/CompletingAllItems/Sync / speedometer EmberJS-Debug-TodoMVC/CompletingAllItems/Sync + 3 more (Windows) regression on Sat October 12 2024

Categories

(Core :: JavaScript: GC, defect, P3)

defect

Tracking

()

Tracking Status
firefox-esr128 --- unaffected
firefox131 --- unaffected
firefox132 --- unaffected
firefox133 --- wontfix
firefox134 --- affected

People

(Reporter: intermittent-bug-filer, Unassigned)

References

(Blocks 2 open bugs, Regression)

Details

(Keywords: perf, perf-alert, regression)

Perfherder has detected a browsertime performance regression from push 55d6abd13184906edeb6361febc83b665b995d4e. 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
2% speedometer EmberJS-Debug-TodoMVC/CompletingAllItems windows11-64-shippable-qr fission webrender 97.28 -> 99.29 Before/After
2% speedometer EmberJS-Debug-TodoMVC/CompletingAllItems/Sync windows11-64-shippable-qr fission webrender 95.28 -> 97.26 Before/After
2% speedometer EmberJS-Debug-TodoMVC/CompletingAllItems/Sync windows11-64-nightlyasrelease-qr fission webrender 91.46 -> 93.31 Before/After
2% speedometer EmberJS-Debug-TodoMVC/CompletingAllItems windows11-64-nightlyasrelease-qr fission webrender 93.31 -> 95.18 Before/After
2% speedometer EmberJS-Debug-TodoMVC/CompletingAllItems/Sync windows11-64-nightlyasrelease-qr fission webrender 92.37 -> 94.15 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. 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.

You can run all of these tests on try with ./mach try perf --alert 2593

The following documentation link provides more information about this command.

For more information on performance sheriffing please see our FAQ.

If you have any questions, please do not hesitate to reach out to afinder@mozilla.com.

Flags: needinfo?(mgaudet)

Set release status flags based on info from the regressing bug 1892764

This is interesting; I think it would be an interesting experiment to look at this in a little more detail, but it's going to be hard I suspect to get a good heuristic that works just here.

If anyone does want a closer look, the dom_alloc_site pref does exist still in tree, and so we could investigate this.

For now, I am however going to mark this as S3/P3 as I don't think it's worth backing the code out on the basis of this.

Severity: -- → S3
Flags: needinfo?(mgaudet)
Priority: -- → P3

If this was regressing sp3, then we would back out, but if the regression is sp2.1 only, then less clear.

Speedometer is possibly rather bad (and unrealistic) testcase for bug 1892764. It creates a realm and does something in it and discards it then before GC has run. And then does the same for another subtest.

Set release status flags based on info from the regressing bug 1892764

:mgaudet do you plan on investigating this for Fx133?
Wondering if we should track this or wontfix for Fx133

Flags: needinfo?(mgaudet)

Done.

Flags: needinfo?(mgaudet)
You need to log in before you can comment on or make changes to this bug.