38.17 - 52.26% raptor-tp6m-facebook-geckoview-cold / raptor-tp6m-facebook-geckoview-cold fcp / raptor-tp6m-instagram-geckoview-cold fcp regression on push 3bf1f379d77e758092f84df765e8c8726b0f3b98 (Wed May 8 2019)
Categories
(Testing :: General, defect)
Tracking
(Not tracked)
People
(Reporter: Bebe, Unassigned)
References
(Regression)
Details
(Keywords: perf, regression)
Raptor has detected a Firefox performance regression from push:
As author of one of the patches included in that push, we need your help to address this regression.
Regressions:
52% raptor-tp6m-facebook-geckoview-cold fcp android-hw-p2-8-0-android-aarch64 opt 853.61 -> 1,299.75
51% raptor-tp6m-instagram-geckoview-cold fcp android-hw-p2-8-0-android-aarch64 opt 312.94 -> 473.75
43% raptor-tp6m-facebook-geckoview-cold android-hw-p2-8-0-android-aarch64 opt 1,102.69 -> 1,576.37
38% raptor-tp6m-instagram-geckoview-cold android-hw-p2-8-0-android-aarch64 opt 544.23 -> 751.95
You can find links to graphs and comparison views for each of the above tests at: https://treeherder.mozilla.org/perf.html#/alerts?id=20864
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 Raptor jobs in a pushlog format.
To learn more about the regressing test(s) or reproducing them, please see: https://wiki.mozilla.org/Performance_sheriffing/Raptor
*** 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
Reporter | ||
Updated•5 years ago
|
Comment 1•5 years ago
|
||
Hi Florin,
I'm having hard time seeing how my patch was at fault for this regression. The patch adds a change to a devtools panel functionality that needs to be explicitly triggered, and the devtools/panel open.
Comment 2•5 years ago
|
||
(In reply to Yura Zenevich [:yzen] from comment #1)
Hi Florin,
I'm having hard time seeing how my patch was at fault for this regression. The patch adds a change to a devtools panel functionality that needs to be explicitly triggered, and the devtools/panel open.
I agree with Yura. The autoland pattern hints a pretty likely backout. I suggest we investigate closer on what happened around 3bf1f379d77e and 8fea66166287, by retriggering/backfilling more.
Comment 3•5 years ago
|
||
E.g. The improvements landed via 8fea66166287, which backed out bug 1547813. I think that's a likely suspect.
Comment 4•5 years ago
|
||
Between May 7 and 8 we added new Android phone to our device pool, which temporarily messed up our perf tests.
Thus, these aren't real perf regressions.
Updated•5 years ago
|
Comment hidden (Intermittent Failures Robot) |
Updated•2 years ago
|
Description
•