From owner-freebsd-questions Sat Feb 17 14:29:51 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id OAA02659 for questions-outgoing; Sat, 17 Feb 1996 14:29:51 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id OAA02654 for ; Sat, 17 Feb 1996 14:29:45 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA09667; Sat, 17 Feb 1996 15:26:44 -0700 From: Terry Lambert Message-Id: <199602172226.PAA09667@phaeton.artisoft.com> Subject: Re: SMP support? To: graichen@omega.physik.fu-berlin.de (Thomas Graichen) Date: Sat, 17 Feb 1996 15:26:44 -0700 (MST) Cc: questions@freebsd.org In-Reply-To: <4g5h3o$g2k@prospero.physik.fu-berlin.de> from "Thomas Graichen" at Feb 17, 96 09:25:12 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-questions@freebsd.org Precedence: bulk > : I will send you patches; you will not be able to trigger the > : buzz-flag to throw the second processor into the scheduler, > : however. > : The second processor LDT/GDT is incorrectly setup in my patch set; > : I would be happy to have you fix it for me. > : I'm not going to write right now (basically a BSD version of > : "SoftIce(tm)"), or to sit down and line-by-line analyze what > > maybe a look into a recent linux kernel (they have SMP working) or an email to > alan cox (he has written much of of the linux SMP stuff) may help ? I'll get to it eventually. Since I have a running system with older kernel and VM code, I'd just as soon address the easy-to-address (for me anyway) issues of kernel reeentrancy and FS multithreading. The SMP in the Linux kernel is low-grain at present. I want kernel threads and no significant degradation from 2 to 4 to 8 processors (my desired target is a 32 processor PPC machine from a company in Germany... think they'll send me one for testing? 8-)). There's a P6 machine with "only" 4 processors that I think would be nice as well... when I get excited, I count based on number of processors instead of per-processor power. 8-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.