Author inada.naoki
Recipients Ryan May, abarry, alecsandru.patrascu, benjamin.peterson, brett.cannon, dino.viehland, eric.fahlgren, florin.papa, gvanrossum, inada.naoki, josh.r, jstasiak, maciej.szulik, ncoghlan, santagada, serhiy.storchaka, twouters, vstinner, yselivanov
Date 2016-12-13.06:36:04
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1481610964.32.0.631172527736.issue26110@psf.upfronthosting.co.za>
In-reply-to
Content
> Technically the patch LGTM. But we should find the cause of the regression in some benchmarks.

The benchmark is on Sandy Bridge (Core i5 2400) and I didn't use PGO build.
perf_event reported branch-miss rate increase at cpickle's save function.

I'll rerun benchmark with PGO build. I hope PGO is friendly with CPU branch
prediction, like L1/L2 cache.

Anyway, recent amd64 CPUs have more large branch history.

> And would be nice to extend the optimization to C functions.
> In any case this optimization is worth mentioning in What's New.

I'll do them.
History
Date User Action Args
2016-12-13 06:36:04inada.naokisetrecipients: + inada.naoki, gvanrossum, twouters, brett.cannon, ncoghlan, vstinner, benjamin.peterson, dino.viehland, santagada, serhiy.storchaka, maciej.szulik, yselivanov, jstasiak, josh.r, abarry, alecsandru.patrascu, florin.papa, eric.fahlgren, Ryan May
2016-12-13 06:36:04inada.naokisetmessageid: <1481610964.32.0.631172527736.issue26110@psf.upfronthosting.co.za>
2016-12-13 06:36:04inada.naokilinkissue26110 messages
2016-12-13 06:36:04inada.naokicreate