1692962 - 3.75 - 3.77% installer size (osx-aarch64-shippable) regression on push 31d00b6fe69f72bd8ec516af02ed3ffbbb3768f0 (Sat February 13 2021)

Status: RESOLVED WONTFIX

(Firefox Build System :: Toolchains, defect)

Description

[Alexandru Ionescu (needinfo me) [:alexandrui]]

Perfherder has detected a build_metrics performance regression from push 31d00b6fe69f72bd8ec516af02ed3ffbbb3768f0. 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)
4%	installer size		osx-aarch64-shippable	aarch64 nightly	79,931,096.50 -> 82,942,606.08
4%	installer size		osx-aarch64-shippable	aarch64 nightly	79,943,994.12 -> 82,941,573.08

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.

Comment 1

[Mike Hommey [:glandium]]

The change in bug 1692418 added a signature to binaries that didn't exist before. That signature is actually dropped further down the releng stack, and the build is resigned with a proper key. That signature was already happening before, so the actual installer size from nightlies didn't change.

The nightly before this landing: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/MX79GC8pREWXtPtTY-t6Lw/runs/0/artifacts/public/build/target.dmg
The nightly after this landing: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/HBV21ww8TSGRqfPPVzVAUQ/runs/0/artifacts/public/build/target.dmg

145,610,339 -> 145,714,876

That's less than 0.1% difference.