Closed Bug 1770437 Opened 3 years ago Closed 3 years ago

14.82 - 14.6% perf_reftest_singletons bloom-basic-2.html / perf_reftest_singletons bloom-basic.html (OSX) regression on Sun May 8 2022

Categories

(Core :: DOM: Copy & Paste and Drag & Drop, defect)

Unspecified
macOS
defect

Tracking

()

RESOLVED INVALID
Tracking Status
firefox-esr91 --- unaffected
firefox100 --- unaffected
firefox101 --- wontfix
firefox102 --- wontfix

People

(Reporter: afinder, Unassigned)

References

(Regression)

Details

(4 keywords)

Perfherder has detected a talos performance regression from push 0c6cb53ce2ec9e5b1357566286b9498f5c02c0f2. 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)
15% perf_reftest_singletons bloom-basic-2.html macosx1015-64-shippable-qr e10s fission stylo webrender 37.19 -> 42.70
15% perf_reftest_singletons bloom-basic.html macosx1015-64-shippable-qr e10s fission stylo webrender 37.00 -> 42.40

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 offending patch(es) will 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.

Flags: needinfo?(emilio)

I don't understand how these can come from that push. There's no changes to the style system in that push.

Flags: needinfo?(emilio)

Set release status flags based on info from the regressing bug 1767938

I'm not an expert at reading these graphs, but it looks like this might have just been an outlier/anomaly?

The samples following this one seem to all be in the normal range.

It also looks like this was retriggered for the same changeset, and all of those retriggers seem to be in the normal range as well.

(I'm looking at the graphs linked from the alert summary, with the little "graph" icons underneath the stars on the left side of the alert-summary-page. By default, that shows 90 days worth of data. If you view the last 30 days, you can see the individual measurements more clearly, and it's even more-clear that this particular measurement seems to have been an odd one-off outlier.)

This test suite is notoriously noisy and as noted by dholbert, the one point from the alert looks like an obvious outlier.

Status: NEW → RESOLVED
Closed: 3 years ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.