On 06/21/2017 06:19 AM, Chris Cooper wrote:
On Tuesday, June 20, 2017 at 1:28:50 PM UTC-4, Ehsan Akhgari wrote:
My understanding is that the slowdown cannot be reproduced on local
developer machines, but can be reproduced on loaner machines from
infra.
Huh. That's interesting and even more puzzling...
I don't think anybody has tried profiling on infra to see
where time differences are.
That seems like the obvious next step to investigate to me. We should
*really* only talk about stripping builds as the last resort IMO, since
we have way too many developers using OSX every day...
One possible avenue we're considering is shipping the unstripped cross-compiled
builds to the nightly audience (or some subset) and seeing if the performance
issue manifests in the wild. We're confident in our ability to repatriate these
users via nightly updates.
It seems like that we have an answer now in the bug!
https://bugzilla.mozilla.org/show_bug.cgi?id=1338651#c129
FWIW these questions aren't really all that easy to answer on telemetry,
especially when you have ways of reproducing the issue locally and you
don't necessarily know which probe if any is expected to show a change
on telemetry when this hits real users. In practice, deciding to ship
the unstripped build to Nightly and seeing if a change would be observed
in the wild wouldn't have been all that different from deciding to take
the regression in practice. :-)
Cheers,
Ehsan
_______________________________________________
dev-platform mailing list
dev-platform@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-platform
