From owner-freebsd-hackers  Tue Feb 16  9:57: 6 1999
Received: from ns.oeno.com (ns.oeno.com [194.100.99.145])
          by hub.freebsd.org (8.8.8/8.8.8) with SMTP id JAA15954
          for <hackers@FreeBSD.ORG>; Tue, 16 Feb 1999 09:57:04 -0800 (PST)
          (envelope-from will@ns.oeno.com)
Received: (qmail 23404 invoked by uid 1001); 16 Feb 1999 17:57:02 -0000
Date: 16 Feb 1999 17:57:01 -0000
Message-ID: <19990216175701.23401.qmail@ns.oeno.com>
From: Ville-Pertti Keinonen <will@iki.fi>
To: tlambert@primenet.com
CC: dyson@iquest.net, hackers@FreeBSD.ORG
In-reply-to: <199902160045.RAA22056@usr02.primenet.com> (message from Terry
	Lambert on Tue, 16 Feb 1999 00:45:20 +0000 (GMT))
Subject: Re: Processor affinity?
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG


> A very trivial affinity soloution is to have per CPU scheduler
> queues, and to keep a "quantum count" per CPU as well.

Any kind of processor load calculation was what I was referring to by
"moderately complex".  Not that I don't think that it's worth it.

> Processes becoming "ready to run" are queued on the processor with
> the highest quantum count (e.g., highest process turnover rate,
> indicating relatively higher I/O binding).

Ignoring long-running, low-priority processes?  They shouldn't affect
the turnover rate because they can typically be pre-empted when an I/O
bound process wakes up.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message