Open Bug 1678543 Opened 4 years ago Updated 5 months ago

4.84 - 8.77% startup_about_home_paint_realworld_webextensions (windows10-64-shippable-qr) regression on push 6ca8cea4044a74c89e0137660c243bd542714abc (Mon November 16 2020)

Categories

(Firefox :: Session Restore, defect)

Firefox 85
defect

Tracking

()

ASSIGNED
Tracking Status
firefox-esr78 --- unaffected
firefox83 --- unaffected
firefox84 --- unaffected
firefox85 --- wontfix
firefox86 --- wontfix
firefox87 --- wontfix
firefox88 --- wontfix
firefox89 --- wontfix

People

(Reporter: alexandrui, Assigned: nika)

References

(Regression)

Details

(4 keywords)

Perfherder has detected a talos performance regression from push 6ca8cea4044a74c89e0137660c243bd542714abc. 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)
9% startup_about_home_paint_realworld_webextensions startup_about_home_paint_realworld_webextensions windows10-64-shippable-qr e10s stylo webrender 572.75 -> 623.00
5% startup_about_home_paint_realworld_webextensions startup_about_home_paint_realworld_webextensions windows10-64-shippable-qr e10s stylo webrender 594.50 -> 623.25

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.

Flags: needinfo?(nika)
Component: Performance → Session Restore
Product: Testing → Firefox

Set release status flags based on info from the regressing bug 1675334

I'm guessing I caused us to do a process switch later or similar. Will investigate a bit more to figure out what exactly this test is doing.

This probably won't be done until the week after next, due to next week largely being taken off.

Assignee: nobody → nika
Flags: needinfo?(nika)

Probably actually regressed by bug 1671983 from the same push

Regressed by: 1671983
No longer regressed by: 1675334
Has Regression Range: --- → yes

I've triggered some retries with profiles between these two revisions:

before: https://treeherder.mozilla.org/jobs?repo=autoland&revision=346bbd581eb8d0da384d248bef8a3f4733f4e391&selectedTaskRun=UhGrREZ7RTOb8NA-nQHn4g.0
after: https://treeherder.mozilla.org/jobs?repo=autoland&revision=6ca8cea4044a74c89e0137660c243bd542714abc&selectedTaskRun=byeaJ_3RTdyE6MkCQzcfQg.0

Unfortunately, it seems that running the profiler caused the output parser to break, meaning that you need to open the raw log files in order to get the specific results from these test runs.

I looked at the profiles a bit, and didn't see anything obvious. I can try to give it a more in-depth look after the break.

Flags: needinfo?(nika)
Status: NEW → ASSIGNED
Flags: needinfo?(nika)
You need to log in before you can comment on or make changes to this bug.