Closed Bug 1559332 Opened 3 months ago Closed 3 months ago

3.71% tp5o_scroll (windows10-64-shippable) regression on push b9e525d8b1aa6659c6cc95956dafae65c28daa62 (Mon June 3 2019)

Categories

(Core :: Graphics, defect)

x86_64
Windows 10
defect
Not set

Tracking

()

RESOLVED WONTFIX

People

(Reporter: igoldan, Unassigned)

References

(Blocks 1 open bug, Regression)

Details

(Keywords: perf, regression, talos-regression)

Talos has detected a Firefox performance regression from push:

https://hg.mozilla.org/releases/mozilla-beta/pushloghtml?fromchange=fa85a070d83364097722aeb109b110eba62b51f8&tochange=b9e525d8b1aa6659c6cc95956dafae65c28daa62

We need your help to address this regression.

Regressions:

4% tp5o_scroll windows10-64-shippable opt e10s stylo 1.22 -> 1.27

You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=21342

On the page above you can see an alert for each affected platform as well as a link to a graph showing the history of scores for this test. There is also a link to a treeherder page showing the Talos jobs in a pushlog format.

To learn more about the regressing test(s), please see: https://wiki.mozilla.org/Performance_sheriffing/Talos/Tests

For information on reproducing and debugging the regression, either on try or locally, see: https://wiki.mozilla.org/Performance_sheriffing/Talos/Running

*** Please let us know your plans within 3 business days, or the offending patch(es) will be backed out! ***

Our wiki page outlines the common responses and expectations: https://wiki.mozilla.org/Performance_sheriffing/Talos/RegressionBugsHandling

Somehow, I'm not able to bisect via Try the entire patch. So I need your help on this.
Out of the 4 suspects patches which I've managed to narrow down, which one is the likely cause?

Flags: needinfo?(yuki)
Flags: needinfo?(nika)
Flags: needinfo?(haftandilian)
Flags: needinfo?(aryx.bugmail)

Scrolling relates to page navigation, which seems related to Core :: Document Navigation.
Would this further narrow the list to just bug 1555688 & bug 1554217?

Should be from bug 1555956. Bug 1555688 only modifies a test file, bug 1554217 had landed before and got backed out but there is no talos regression for that push. Also bug 1555956 is about disabling a Windows feature (direct3d11) and the regression is on Windows.

Flags: needinfo?(aryx.bugmail)
Regressed by: 1555956

(In reply to Sebastian Hengst [:aryx] (needinfo on intermittent or backout) from comment #3)

Should be from bug 1555956.

:kats can you take a look over this issue & confirm it?

Flags: needinfo?(kats)

(In reply to Sebastian Hengst [:aryx] (needinfo on intermittent or backout) from comment #3)

Should be from bug 1555956. Bug 1555688 only modifies a test file, bug 1554217 had landed before and got backed out but there is no talos regression for that push. Also bug 1555956 is about disabling a Windows feature (direct3d11) and the regression is on Windows.

And bug 1555168 is Mac-specific.

Flags: needinfo?(haftandilian)

Yeah it's quite plausible that the regression is from disabling double-buffering. But double buffering was only recently enabled in bug 1547775, and the feature quality is not good enough to let it ride the trains, so this perf regression is a wontfix.

Status: NEW → RESOLVED
Closed: 3 months ago
Flags: needinfo?(yuki)
Flags: needinfo?(nika)
Flags: needinfo?(kats)
Resolution: --- → WONTFIX
No longer regressed by: 1554217, 1555168, 1555584, 1555688
Component: General → Graphics
Product: Testing → Core
You need to log in before you can comment on or make changes to this bug.