14.08% tp5o responsiveness (linux64-shippable) regression on push 4fc932c5974e567ed1568bd3cba75345fb9c36c3 (Wed December 2 2020)
Categories
(Core :: JavaScript Engine: JIT, defect)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox85 | --- | wontfix |
People
(Reporter: alexandrui, Unassigned)
References
(Regression)
Details
(4 keywords, Whiteboard: [perf:alert:0])
Perfherder has detected a talos performance regression from push 4fc932c5974e567ed1568bd3cba75345fb9c36c3. As author of one of the patches included in that push, we need your help to address this regression.
Regressions:
| Ratio | Suite | Test | Platform | Options | Absolute values (old vs new) |
|---|---|---|---|---|---|
| 14% | tp5o | responsiveness | linux64-shippable | e10s stylo | 0.87 -> 0.99 |
Improvements:
| Ratio | Suite | Test | Platform | Options | Absolute values (old vs new) |
|---|---|---|---|---|---|
| 5% | kraken | windows10-64-shippable | e10s stylo | 1,410.10 -> 1,341.84 |
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.
For more information on performance sheriffing please see our FAQ.
| Reporter | ||
Updated•5 years ago
|
Comment 1•5 years ago
|
||
== Change summary for alert #27956 (as of Thu, 03 Dec 2020 22:57:56 GMT) ==
Improvements:
| Ratio | Suite | Test | Platform | Options | Absolute values (old vs new) |
|---|---|---|---|---|---|
| 4% | raptor-tp6-office-firefox-cold | loadtime | windows10-64-shippable-qr | webrender | 2,261.46 -> 2,172.33 |
For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=27956
Comment 2•5 years ago
|
||
== Change summary for alert #27958 (as of Thu, 03 Dec 2020 23:32:50 GMT) ==
Regressions:
| Ratio | Suite | Test | Platform | Options | Absolute values (old vs new) |
|---|---|---|---|---|---|
| 10% | twitch | FirstVisualChange | linux64-shippable-qr | nocondprof warm webrender | 695.00 -> 766.67 |
| 8% | twitch | SpeedIndex | linux64-shippable-qr | nocondprof warm webrender | 754.50 -> 818.00 |
| 8% | twitch | PerceptualSpeedIndex | linux64-shippable-qr | nocondprof warm webrender | 806.50 -> 870.83 |
| 3% | twitch | linux64-shippable-qr | nocondprof warm webrender | 194.66 -> 200.18 |
Improvements:
| Ratio | Suite | Test | Platform | Options | Absolute values (old vs new) |
|---|---|---|---|---|---|
| 55% | google-sheets | LastVisualChange | linux64-shippable-qr | nocondprof warm webrender | 5,268.33 -> 2,350.00 |
| 4% | speedometer | android-hw-g5-7-0-arm7-api-16-shippable | geckoview nocondprof webrender | 10.10 -> 10.48 |
For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=27958
Comment 3•5 years ago
|
||
I'm going to bisect the patches I pushed, on Try, for the tp5o-responsiveness regression. Hopefully that will narrow it down a bit and tell us more.
Comment 4•5 years ago
|
||
I didn't see any regressions on Try so I looked at the autoland data. Turns out the alert is wrong, it's based on one data point that was an outlier. Retriggers and backfills point to push e4c8bc7feaa6 instead.
Can you move and change this? Thanks.
| Reporter | ||
Updated•5 years ago
|
| Reporter | ||
Comment 5•5 years ago
|
||
Makes sense. Thanks Jan.
| Reporter | ||
Updated•5 years ago
|
Updated•5 years ago
|
Updated•5 years ago
|
Description
•