Page MenuHomePhabricator

eo: move all call assignment
ClosedPublic

Authored by bu5hm4n on Mar 25 2020, 2:19 AM.

Details

Summary

this improves the cache performance a lot. Caches are only invalidated
once, and not multiple times.
Depends on D11591

Diff Detail

Repository
rEFL core/efl
Lint
Automatic diff as part of commit; lint not applicable.
Unit
Automatic diff as part of commit; unit tests not applicable.
bu5hm4n created this revision.Mar 25 2020, 2:19 AM
bu5hm4n requested review of this revision.Mar 25 2020, 2:19 AM
bu5hm4n added a child revision: D11593: eo: use eina_prefetch.

This looks fine to me, but I wonder how are you measuring the cache hits and misses?

raster accepted this revision.Mar 26 2020, 5:38 AM
This revision is now accepted and ready to land.Mar 26 2020, 5:38 AM

@stefan_schmidt, these numbers are messured with perf stat -e cache-misses elementary_perf.

Closed by commit rEFL8e20d28dfe63: eo: move all call assignment (authored by Marcel Hollerbach <mail@marcel-hollerbach.de>). · Explain WhyMar 26 2020, 5:59 AM
This revision was automatically updated to reflect the committed changes.