Closed Bug 1908328 Opened 1 year ago Closed 1 year ago

2.24 - 15.53% speedometer3 TodoMVC-Angular-Complex-DOM/CompletingAllItems/Async / speedometer3 TodoMVC-Angular-Complex-DOM/CompletingAllItems and more on Tue Jul 09

Categories

(Toolkit :: Form Autofill, defect)

Product:
Toolkit
Component:
Form Autofill
Type:
defect

Tracking

()

Status:
RESOLVED WONTFIX
Tracking Flags:
Tracking Status
firefox-esr115 --- unaffected
firefox-esr128 --- unaffected
firefox128 --- unaffected
firefox129 --- unaffected
firefox130 --- affected

People

(Reporter: fbilt, Unassigned)

References

(Regression)

Details

(Keywords: perf, perf-alert, regression)

Perfherder has detected a browsertime performance change from push 21237438a5301c7a5ba65a854974614422126b31.

Regressions:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
16% speedometer3 TodoMVC-JavaScript-ES5/CompletingAllItems/Async android-hw-a51-11-0-aarch64-shippable-qr webrender 11.60 -> 13.41 Before/After
10% speedometer3 TodoMVC-Angular-Complex-DOM/CompletingAllItems/Async android-hw-a51-11-0-aarch64-shippable-qr webrender 23.90 -> 26.26 Before/After
10% speedometer3 TodoMVC-Angular-Complex-DOM/CompletingAllItems/Async android-hw-a51-11-0-aarch64-shippable-qr webrender 23.82 -> 26.13 Before/After
9% speedometer3 TodoMVC-Svelte-Complex-DOM/Adding100Items/Sync macosx1400-64-shippable-qr fission webrender 1.44 -> 1.57 Before/After
7% speedometer3 TodoMVC-Svelte-Complex-DOM/Adding100Items/Sync macosx1015-64-shippable-qr fission webrender 3.15 -> 3.38 Before/After
7% speedometer3 TodoMVC-Svelte-Complex-DOM/Adding100Items/Sync windows11-64-shippable-qr fission webrender 2.02 -> 2.16 Before/After
7% speedometer3 TodoMVC-Svelte-Complex-DOM/Adding100Items/Sync windows11-64-nightlyasrelease-qr fission webrender 1.97 -> 2.10 Before/After
4% speedometer3 TodoMVC-Angular-Complex-DOM/CompletingAllItems/total android-hw-a51-11-0-aarch64-shippable-qr webrender 93.50 -> 97.36 Before/After
2% speedometer3 TodoMVC-Preact-Complex-DOM/Adding100Items/Async windows11-64-nightlyasrelease-qr fission webrender 6.43 -> 6.58 Before/After

Improvements:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
87% speedometer EmberJS-Debug-TodoMVC/Adding100Items/Async linux1804-64-shippable-qr fission webrender 32.55 -> 4.38 Before/After
86% speedometer EmberJS-Debug-TodoMVC/Adding100Items/Async linux1804-64-nightlyasrelease-qr fission webrender 33.88 -> 4.76 Before/After
75% speedometer EmberJS-Debug-TodoMVC/Adding100Items/Async macosx1015-64-nightlyasrelease-qr fission webrender 17.54 -> 4.42 Before/After
73% speedometer Flight-TodoMVC/Adding100Items/Async windows11-64-shippable-qr fission webrender 8.54 -> 2.28 Before/After
65% speedometer EmberJS-TodoMVC/Adding100Items/Async windows11-64-nightlyasrelease-qr fission webrender 6.37 -> 2.23 Before/After
... ... ... ... ... ...
2% speedometer AngularJS-TodoMVC/Adding100Items/Async macosx1015-64-shippable-qr fission webrender 7.38 -> 7.22 Before/After

As author of one of the patches included in that push, we need your help to address this regression.
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 these tests on try with ./mach try perf --alert 1141

For more information on performance sheriffing please see our FAQ.

Flags: needinfo?(dlee)

Set release status flags based on info from the regressing bug 1898745

status-firefox128: --- → unaffected
status-firefox129: --- → unaffected
status-firefox-esr115: --- → unaffected
status-firefox-esr128: --- → unaffected

The patches in Bug 1898745 move the field identification logic from the content process to the parent process. This change means that some optimizations previously applied in the content process are no longer in effect, and we now need to pass the information to the parent to determine what to do next. However, moving the logic from the content to the parent process also has the benefit of reducing the workload on the content process.

Hi Florin, Based on the alert summary, it looks like that the patch introduced both regressions and improvements. I'm not very familiar with Browsertime performance changes, but it seems that overall, the patch improves performance. Is that correct? If so, could we consider closing this issue? thank you!

Flags: needinfo?(dlee) → needinfo?(fbilt)

Hello
If the summary contains a single regression, then we consider it overall as regression. Indeed, this summary also contains many improvements, but I am not in a position to say if it is worth it.

Flags: needinfo?(fbilt) → needinfo?(dlee)

(In reply to Florin Bilt from comment #3)

Hello
If the summary contains a single regression, then we consider it overall as regression. Indeed, this summary also contains many improvements, but I am not in a position to say if it is worth it.

In that case, after investigation (See Comment 2), i think we can close this bug because this patch actually improves performance a bit.

Status: NEW → RESOLVED
Closed: 1 year ago
Flags: needinfo?(dlee)
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.