Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264337AbUDORJq (ORCPT ); Thu, 15 Apr 2004 13:09:46 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264354AbUDORJp (ORCPT ); Thu, 15 Apr 2004 13:09:45 -0400 Received: from smtp2.rz.tu-harburg.de ([134.28.205.13]:61922 "EHLO smtp2.rz.tu-harburg.de") by vger.kernel.org with ESMTP id S264337AbUDORJm (ORCPT ); Thu, 15 Apr 2004 13:09:42 -0400 From: Christian =?iso-8859-1?q?Kr=F6ner?= To: linux-kernel@vger.kernel.org Subject: Re: IO-APIC on nforce2 [PATCH] Date: Thu, 15 Apr 2004 19:09:40 +0200 User-Agent: KMail/1.6.1 Cc: Len Brown , ross@datscreative.com.au, "Maciej W. Rozycki" References: <200404131117.31306.ross@datscreative.com.au> <1081988224.15062.75.camel@dhcppc4> <200404151148.53187.ross@datscreative.com.au> In-Reply-To: <200404151148.53187.ross@datscreative.com.au> MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <200404151909.40636.christian.kroener@tu-harburg.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1421 Lines: 40 On Thursday 15 April 2004 10:17, Len Brown wrote: > I'm okay putting the bootparam and the workaround into the kernel, > for it is generic and we may find other platforms need it. Thats more than what I could have expected, thanks. Ross, I tested my kernel with nmi_watchdog=1 and nmi_watchdog=2 getting only 2 to work. output: nmi_watchdog=1 activating NMI Watchdog ... done. testing NMI watchdog ... CPU#0: NMI appears to be stuck! output: nmi_watchdog=2 testing NMI watchdog ... OK. This is on 2.6.5-mm5-1. Concerning the timer, well I tested it against my radio-controlled clock, setting it with ntpdate first and letting the system run (with ntpd off) and my system is kinda faster than my radio-clock. After about one hour my system was off by +14s compared to the radio-clock. I don't know if that is pretty shitty or simply normal for these bad pc-clocks... I'm now compiling 2.6.6-rc1 with the nmi_watchdog=1 workaround. One question about the C1idle-patch, does this add a new feature or is it just a workaround for locked up nforce2-systems (since I never experienced lockups on my system, I wouldn't need it then)? thanks for now, christian. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/