0.26 - 40.13% build times / compiler_metrics num_static_constructors / installer size (android-4-0-armv7-api16, osx-shippable, windows2012-64-shippable) regression on push 82f9e4bad64592efb70ded65f86e566d3ca2c664 (Tue May 12 2020)
Categories
(Firefox Build System :: Toolchains, defect)
Tracking
(firefox-esr68 unaffected, firefox76 unaffected, firefox77 unaffected, firefox78 fixed)
Tracking | Status | |
---|---|---|
firefox-esr68 | --- | unaffected |
firefox76 | --- | unaffected |
firefox77 | --- | unaffected |
firefox78 | --- | fixed |
People
(Reporter: marauder, Unassigned)
References
(Regression)
Details
(Keywords: perf-alert, regression)
Perfherder has detected a build_metrics performance regression from push 82f9e4bad64592efb70ded65f86e566d3ca2c664. As author of one of the patches included in that push, we need your help to address this regression.
Regressions:
40% build times android-4-0-armv7-api16 debug gcp taskcluster-projects/970387039909/machineTypes/custom-32-73728 979.93 -> 1,373.15
2% compiler_metrics num_static_constructors windows2012-64-shippable opt instrumented 129.00 -> 132.00
0.47% installer size osx-shippable opt instrumented 115,756,000.08 -> 116,295,132.00
0.26% installer size osx-shippable opt nightly 82,972,682.92 -> 83,186,034.08
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 | ||
Comment 1•4 years ago
|
||
The backout of Bug 1616692 caused a few regressions, is this expected ?
Thanks!
Updated•4 years ago
|
There's this frustrating pattern where perfherder generates regression alerts for backouts without generating improvement alerts for the initial landings: https://treeherder.mozilla.org/perf.html#/graphs?highlightAlerts=1&series=autoland,1921012,1,2&timerange=1209600&zoom=1589296790470,1589328651419,82670870.67530864,83533361.41358025 Could you please investigate how we can improve this?
In the meantime, we're going to re-land bug 1616692 anyway, so I wouldn't worry about these unless they persist after re-landing.
Comment 3•4 years ago
|
||
Not really a regression since this was from the backout. And fixed since it's re-landed
Updated•4 years ago
|
Description
•