Message283077
> 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. |
|
Date |
User |
Action |
Args |
2016-12-13 06:36:04 | methane | set | recipients:
+ methane, 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 06:36:04 | methane | set | messageid: <1481610964.32.0.631172527736.issue26110@psf.upfronthosting.co.za> |
2016-12-13 06:36:04 | methane | link | issue26110 messages |
2016-12-13 06:36:04 | methane | create | |
|