From owner-freebsd-stable Fri Aug 9 13:39:38 2002 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1F12837B400 for ; Fri, 9 Aug 2002 13:39:36 -0700 (PDT) Received: from isris.pair.com (isris.pair.com [209.68.2.39]) by mx1.FreeBSD.org (Postfix) with SMTP id 5809343E70 for ; Fri, 9 Aug 2002 13:39:35 -0700 (PDT) (envelope-from rooneg@isris.pair.com) Received: (qmail 96587 invoked by uid 3130); 9 Aug 2002 20:39:34 -0000 Date: Fri, 9 Aug 2002 16:39:34 -0400 From: Garrett Rooney To: Morten Rodal Cc: David Schultz , kpieckiel@smartrafficenter.org, Bosko Milekic , Mario Pranjic , freebsd-stable@FreeBSD.ORG Subject: Re: SMP kernel: FreeBSD vs. Linux 2.4.x Message-ID: <20020809203934.GA94313@electricjellyfish.net> References: <20020809091008.A87124@unixdaemons.com> <20020809164411.GC78503@pacer.dmz.smartrafficenter.org> <20020809171743.GB290@HAL9000.homeunix.com> <20020809203355.GE6050@slurp.rodal.no> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20020809203355.GE6050@slurp.rodal.no> User-Agent: Mutt/1.3.25i Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Fri, Aug 09, 2002 at 10:33:55PM +0200, Morten Rodal wrote: > On Fri, Aug 09, 2002 at 10:17:43AM -0700, David Schultz wrote: > > Unix was originally designed for uniprocessor systems. > > Consequently, some assumptions were made that are reasonable and > > result in lower locking overhead for uniprocessors, but that > > aren't valid for multiprocessors. > > > > http://www.lemis.com/~grog/SMPng/USENIX/ > > > > > Second, what are KSEs? > > > > cf. Scheduler Activations: > > > > http://citeseer.nj.nec.com/anderson92scheduler.html > > > > How does this compare to the approach NetBSD is taking? The docs above > helped, but I am still a bit puzzled. > > http://daily.daemonnews.org/view_story.php3?story_id=2969 NetBSD is following the approach set forth in the anderson paper, while FreeBSD is trying some new ideas that should provide for better performance, but are less well tested (i.e. nobody has implemented them before). -garrett -- garrett rooney Remember, any design flaw you're rooneg@electricjellyfish.net sufficiently snide about becomes http://electricjellyfish.net/ a feature. -- Dan Sugalski To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message