Date: Wed, 18 Sep 2002 04:53:06 -0400 (EDT) From: Jeff Roberson <jroberson@chesapeake.net> To: Jeff Roberson <jeff@FreeBSD.org> Cc: cvs-committers@FreeBSD.org, <cvs-all@FreeBSD.org> Subject: Re: cvs commit: src/sys/kern kern_malloc.c src/sys/vm uma.h uma_core.c uma_dbg.c uma_int.h vm_page.h Message-ID: <20020918042829.K43954-100000@mail.chesapeake.net> In-Reply-To: <200209180826.g8I8QUXO077424@freefall.freebsd.org>
index | next in thread | previous in thread | raw e-mail
On Wed, 18 Sep 2002, Jeff Roberson wrote: > jeff 2002/09/18 01:26:30 PDT > > vm_page.h > Log: > - Split UMA_ZFLAG_OFFPAGE into UMA_ZFLAG_OFFPAGE and UMA_ZFLAG_HASH. > - Remove all instances of the mallochash. > - Stash the slab pointer in the vm page's object pointer when allocating from > the kmem_obj. > - Use the overloaded object pointer to find slabs for malloced memory. > Removing mallochash got rid of one more contended global resource. The only remaining global contention points are the malloc_type stats, but that's not nearly as serious. Also, removing the malloc hash saved a big chunk of memory. This yielded slightly faster kernel times for buildworlds (~2%). More importantly, it paves the way for more effecient large object representation within UMA. Oh, and, I forgot. I tested this on x86 and alpha. Jeff To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the messagehelp
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020918042829.K43954-100000>