Closed Bug 1766439 Opened 3 years ago Closed 3 years ago

9.99 - 5.95% perf_reftest_singletons parent-basic-singleton.html / perf_reftest_singletons attr-selector-1.html + 1 more (Linux, Windows) regression on Thu April 21 2022

Categories

(Firefox Build System :: Toolchains, defect)

defect

Tracking

(firefox-esr91 unaffected, firefox99 unaffected, firefox100 unaffected, firefox101 wontfix, firefox102 wontfix)

RESOLVED WONTFIX
Tracking Status
firefox-esr91 --- unaffected
firefox99 --- unaffected
firefox100 --- unaffected
firefox101 --- wontfix
firefox102 --- wontfix

People

(Reporter: aglavic, Unassigned)

References

(Regression)

Details

(4 keywords)

Perfherder has detected a talos performance regression from push 5bebcdfc5dd864f94cf6af8e418fd3900995bc28. 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)
10% perf_reftest_singletons parent-basic-singleton.html windows10-64-shippable-qr e10s fission stylo webrender 80.30 -> 88.32
7% perf_reftest_singletons parent-basic-singleton.html linux1804-64-shippable-qr e10s fission stylo webrender 91.48 -> 97.48
6% perf_reftest_singletons attr-selector-1.html linux1804-64-shippable-qr e10s fission stylo webrender 139.56 -> 147.87

Improvements:

Ratio Test Platform Options Absolute values (old vs new)
18% perf_reftest_singletons id-getter-4.html windows10-64-shippable-qr e10s fission stylo webrender 394.74 -> 323.22
18% perf_reftest_singletons id-getter-3.html windows10-64-shippable-qr e10s fission stylo webrender 394.74 -> 323.84
18% perf_reftest_singletons id-getter-6.html windows10-64-shippable-qr e10s fission stylo webrender 394.63 -> 323.75
18% perf_reftest_singletons id-getter-5.html windows10-64-shippable-qr e10s fission stylo webrender 394.94 -> 324.60
18% perf_reftest_singletons id-getter-7.html windows10-64-shippable-qr e10s fission stylo webrender 394.35 -> 324.75
... ... ... ... ...
4% tabswitch linux1804-64-shippable-qr e10s fission stylo webrender 7.23 -> 6.91

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?(mh+mozilla)

Set release status flags based on info from the regressing bug 1758780

Has Regression Range: --- → yes

== Change summary for alert #33979 (as of Fri, 29 Apr 2022 02:39:03 GMT) ==

Regressions:

Ratio Test Platform Options Absolute values (old vs new)
3% microsoft-support LastVisualChange android-hw-g5-7-0-arm7-shippable-qr warm webrender 2,616.38 -> 2,691.10

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=33979

Set release status flags based on info from the regressing bug 1758780

Can't tell about comment 2, but the ones in comment 0 seem to all be a regression in stylo with rustc 1.60. Emilio, do you want to look into this?

Flags: needinfo?(mh+mozilla) → needinfo?(emilio)

I took a look and while I can reproduce they aren't terribly concerning.

Sent a few selector-matching improvements that improve the generated code, but we're in micro-optimization-land at this point.

Flags: needinfo?(emilio)

Doesn't look like the graphs changed much after the various deps landed. Was there more you wanted to try here still or should we close this out as a wontfix?

Flags: needinfo?(emilio)

Yeah, I don't plan to look into it deeper atm.

Status: NEW → RESOLVED
Closed: 3 years ago
Flags: needinfo?(emilio)
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.