From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 15 May 2002 10:03:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 15 May 2002 10:03:33 -0400 Received: from zeke.inet.com ([199.171.211.198]:64702 "EHLO zeke.inet.com") by vger.kernel.org with ESMTP id ; Wed, 15 May 2002 10:03:32 -0400 From: "Jordan Breeding" To: "Linux Kernel" Cc: "Jordan Breeding" , "Jordan Breeding" Subject: Problem with nmi watchdog Date: Wed, 15 May 2002 09:01:58 -0500 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hello, I have a Tyan Thunder K7 (S2462UNG) which does not like using the nmi watchdog automatically. If I boot up nmi_watchdog=1 then the machine show no interrupt in the nmi columns of /proc/interrupts. While checking the boot log I see the message about CPU#0 being stuck. This is made more interesting by the fact that if I boot using nmi_watchdog=2 then everything works fine. Why will the automatic detection using nmi_watchdog=1 not work (my current kernel is 2.5.15-dj1)? Thanks. Jordan Breeding