0.68 - 0.08% JS / Resident Memory (Windows) regression on Wed December 11 2024
Categories
(Remote Protocol :: Agent, defect)
Tracking
(firefox-esr128 unaffected, firefox133 unaffected, firefox134 unaffected, firefox135 affected)
| Tracking | Status | |
|---|---|---|
| firefox-esr128 | --- | unaffected |
| firefox133 | --- | unaffected |
| firefox134 | --- | unaffected |
| firefox135 | --- | affected |
People
(Reporter: intermittent-bug-filer, Unassigned)
References
(Regression)
Details
(Keywords: perf, perf-alert, regression)
Perfherder has detected a awsy performance regression from push ee733c6049eb17133d0e7885c758272f5f842bc7. As author of one of the patches included in that push, we need your help to address this regression.
Improvements:
| Ratio | Test | Platform | Options | Absolute values (old vs new) |
|---|---|---|---|---|
| 1% | JS | windows11-64-2009-shippable-qr | fission tp6 | 172,865,013.83 -> 171,681,768.04 |
| 0.08% | Resident Memory | windows11-64-2009-shippable-qr | fission tp6 | 886,863,344.17 -> 886,180,728.97 |
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 43125
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.
|
||
Comment 1•2 months ago
|
||
Just to clarify for the comment 0 details, it seems that the alert detection system classified this incorrectly. The alert mentioned in comment 0 is actually a regression, not an improvement.
|
||
Comment 2•2 months ago
|
||
If it's a regression that it should be for bug 1922077 and not bug 1935324 which only disabled some tests.
The changes as landed are a major architectural rewrite of the actions processing and dispatching logic using more IPC communication now. As such a slight increase of memory is expected as well, and as it can be seen it's very low.
So should those two changes have triggered an alert? We would vote for accepting it. Greg, what do you think?
|
||
Comment 3•2 months ago
|
||
Set release status flags based on info from the regressing bug 1922077
|
||
Comment 4•2 months ago
|
||
I agree, this can be closed as a wontfix given that they're both under the 2% threshold - they would not have alerted if it wasn't for the noise that triggered the initial improvement alert. I'll close it based on comment #2.
Description
•