DragonFly kernel List (threaded) for 2006-12
DragonFly BSD
DragonFly kernel List (threaded) for 2006-12
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

Re: jemalloc?


From: "Simon 'corecode' Schubert" <corecode@xxxxxxxxxxxx>
Date: Thu, 14 Dec 2006 12:13:09 +0100

Vlad Galu wrote:
  It looks like the extra spage usage was fixed by Jason Evans around
March. The Google link provided by Freddie Cash pretty much covers
every info you needed. I can perform further benchmarks if needed. But
I think I'll not do it, given the reluctance on the list. I'm
certainly not the best judge for this level of technical detail.

I can't spot a definite reluctance. Some people are conservative, some people say "why not, if it is good". I think it shows promising numbers, but we will probably not experience the reduced contention because our userland threading is still 1:N.


In any case, we slowly need a formalized benchmark suite to spot regressions and prove that new code does what it advertises. An allocator benchmark is needed for sure as well.

I think that jemalloc will perform better than phkmalloc in almost all cases. However I get the feeling that it is a little bit complicated, but maybe I just don't have the right knowledge about allocators. If nothing breaks I'm for including jemalloc, maybe not as default (I wouldn't see why not), but as runtime switchable allocator (/etc/malloc.type -> jemalloc)

cheers
 simon

--
Serve - BSD     +++  RENT this banner advert  +++    ASCII Ribbon   /"\
Work - Mac      +++  space for low €€€ NOW!1  +++      Campaign     \ /
Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \
Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \

Attachment: signature.asc
Description: OpenPGP digital signature



[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]