From owner-freebsd-amd64@FreeBSD.ORG Sat Apr 3 12:49:17 2004 Return-Path: Delivered-To: freebsd-amd64@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C54D216A4CE for ; Sat, 3 Apr 2004 12:49:17 -0800 (PST) Received: from mail.jrv.org (rrcs-sw-24-73-246-106.biz.rr.com [24.73.246.106]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7146D43D2D for ; Sat, 3 Apr 2004 12:49:11 -0800 (PST) (envelope-from james@jrv.org) Received: from jrv.org (zippy.jrv.org [192.168.3.156]) (authenticated bits=0) by dogfood.jrv.org (8.12.11/8.12.10) with ESMTP id i337QaAc067033 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sat, 3 Apr 2004 01:26:36 -0600 (CST) (envelope-from james@jrv.org) Message-ID: <406E672C.4050100@jrv.org> Date: Sat, 03 Apr 2004 01:26:36 -0600 From: "James R. Van Artsalen" User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en-us, en MIME-Version: 1.0 To: "Cassidy B. Larson" References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: "freebsd-amd64@freebsd.org" Subject: Re: SMP issues on K8S Pro S2882G3NR X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 03 Apr 2004 20:49:17 -0000 Cassidy B. Larson wrote: >One more question I had. > >What's the typical temperature of two 246 opterons? We have it in a 1U box >with fans and ventilation galore, but when I push the box for more then 20 >seconds with 0 CPU free, it freezes. A quick power cycle into the BIOS >details that I'm running at over 95 degrees centigrade. Letting it cool >down will let me run my CPU test another 2-3 times then freezes again. > > The rated case temperature is 70 C. There is a hardware-enforced shutdown mechanism at "125 C" but I don't know where this is measured. I don't know if the ROM Setup number is accurate, or what it is measuring. I have no confidence in this number. There are two pins to an on-die thermal diode they could monitor, but I don't know how to relate this value to the rated thermal maximums. >I looked through the Tyan manual and found nothing of the halt temperature, >nor AMD's site. > > Try http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/23932.pdf , the AMD Opteron Processor Data Sheet, the Power Management sections and the Miscellaneous Pin Descriptions for THERMDA, THERMDC and THERMTRIP_L. >If needs be I may send this back and get a 2U case with better cooling. >Argh. > > Make sure CPU cooler exhaust air is not recirculating back into the CPU cooler air intake. Most ATX thermal schemes are completely incompetent, as is the BTX proposal. The focus should be on getting hot exhaust air out of the chassis before it heats any component, especially the CPU. If you have to get another case evaluate thermals by seeing how they evacuate hot exhaust from the CPU (it should not recirculate or bake any other component).