Date: Fri, 11 Jan 2002 09:44:12 +1030 From: Greg Lehey <grog@FreeBSD.org> To: Terry Lambert <tlambert2@mindspring.com> Cc: Garance A Drosihn <drosih@rpi.edu>, smp@FreeBSD.ORG Subject: Re: Linux moves to per CPU run queues Message-ID: <20020111094412.B42285@wantadilla.lemis.com> In-Reply-To: <3C3DE767.8A1EBD8B@mindspring.com> References: <3C3CF60B.CE5F3E6C@mindspring.com> <p05101007b862dc9d2809@[128.113.24.47]> <3C3DE767.8A1EBD8B@mindspring.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday, 10 January 2002 at 11:11:35 -0800, Terry Lambert wrote: > Garance A Drosihn wrote: >> At 6:01 PM -0800 1/9/02, Terry Lambert wrote: >>> As I have been suggesting for FreeBSD for some time, Linux has >>> just moved to per CPU run queues to reduce scheduler contention >>> and to improve affinity. >>> >>> http://www.uwsg.iu.edu/hypermail/linux/kernel/0201.0/0810.html >> >> Well, it's still in the testing stage, it's not like the switch >> has "gone production" yet. >> >> It does look very interesting, and it will be good to keep track >> of how well it performs as more people beat on it. There clearly >> seemed to be a lot of interest for this change in the above >> mailing list. There seem to be several "doubters" in the list, >> so I'm sure they'll come up with several different ways to >> benchmark the results. >> >> Always good to have a little competition in the world of ideas... > > Or we could look at the benchmarks published about a similar > approach in Dynix, back in 1991, on a 32 CPU system, and just > decide that it's the right way to do things... Sure, because the rest of Dynix worked the same way, and the hardware was the same. It's got to be the right way! Greg -- See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020111094412.B42285>