From owner-freebsd-hardware@FreeBSD.ORG Mon Mar 28 06:56:54 2005 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7ED2C16A4CE for ; Mon, 28 Mar 2005 06:56:54 +0000 (GMT) Received: from mail13.syd.optusnet.com.au (mail13.syd.optusnet.com.au [211.29.132.194]) by mx1.FreeBSD.org (Postfix) with ESMTP id ACB5C43D4C for ; Mon, 28 Mar 2005 06:56:53 +0000 (GMT) (envelope-from PeterJeremy@optushome.com.au) Received: from cirb503493.alcatel.com.au (c211-30-75-229.belrs2.nsw.optusnet.com.au [211.30.75.229]) j2S6upGU025765 (version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NO) for ; Mon, 28 Mar 2005 16:56:52 +1000 Received: from cirb503493.alcatel.com.au (localhost.alcatel.com.au [127.0.0.1])j2S6up7l068409 for ; Mon, 28 Mar 2005 16:56:51 +1000 (EST) (envelope-from pjeremy@cirb503493.alcatel.com.au) Received: (from pjeremy@localhost)j2S6uodj068408 for freebsd-hardware@freebsd.org; Mon, 28 Mar 2005 16:56:50 +1000 (EST) (envelope-from pjeremy) Date: Mon, 28 Mar 2005 16:56:50 +1000 From: Peter Jeremy To: freebsd-hardware@freebsd.org Message-ID: <20050328065650.GZ57256@cirb503493.alcatel.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2i Subject: Wierd hardware instability X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 28 Mar 2005 06:56:54 -0000 I've been experimenting with using a 13MHz oven crystal oscillator to replace the 14.318MHz master reference in an old Asus P5A-B motherboard. (The board uses an ICS8148-53 clock generator). If I set TIMER_FREQ=1083342 and "overclock" by 10% then the CPU is running very close to nominal. I'm running 5.3p5 and using "make buildworld" as a stress test and have found some fairly wierd behaviour: If I let the system boot normally from power-on then it is unstable - the buildworld will crash with internal compiler errors and I eventually wind up with a panic. If I manually reset the system then it becomes rock solid - it has been doing "make -j 4 buildworld" in a loop for about 4 days without a problem but as soon as I do a power-on restart, it becomes unstable again. The system was reliable before I started, so it's presumably something I've done but I can't see why a power-on reset should have different behaviour to pressing the reset button. Does anyone have any suggestions as to the cause. If anyone wants more detail on what I've done, feel free to ask - I will probably post details at some stage. -- Peter Jeremy