From owner-freebsd-hackers Tue Feb 16 06:03:49 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id GAA20360 for freebsd-hackers-outgoing; Tue, 16 Feb 1999 06:03:49 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from spooky.rwwa.com (rwwa.com [198.115.177.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA20354 for ; Tue, 16 Feb 1999 06:03:46 -0800 (PST) (envelope-from witr@rwwa.com) Received: from spooky.rwwa.com (localhost.rwwa.com [127.0.0.1]) by spooky.rwwa.com (8.8.7/8.8.7) with ESMTP id JAA02654 for ; Tue, 16 Feb 1999 09:08:28 -0500 (EST) (envelope-from witr@rwwa.com) Message-Id: <199902161408.JAA02654@spooky.rwwa.com> X-Mailer: exmh version 2.0zeta 7/24/97 To: hackers@FreeBSD.ORG Subject: Re: Processor affinity? In-reply-to: Your message of "Mon, 15 Feb 1999 17:13:40 EST." <199902152213.RAA01910@y.dyson.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 16 Feb 1999 09:08:28 -0500 From: Robert Withrow Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG dyson@iquest.net said: :- The main problem that I was trying to solve was the silly bouncing. But it can only be called "silly" if the scheduling latency introduced by bouncing to another CPU is greater than that introduced by waiting to schedule the process on the "current" CPU, right? Have you actually measured that? It seems reasonable to me that a scheduler might cause processes to periodically bounce from CPU to CPU, and still be optimizing "global" latency...depending on what the processes are doing. Or are you using some other measure of scheduling performance than latency? --------------------------------------------------------------------- Robert Withrow, R.W. Withrow Associates, Swampscott MA, witr@rwwa.COM To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message