From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755017AbZHTWK7 (ORCPT ); Thu, 20 Aug 2009 18:10:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754867AbZHTWK7 (ORCPT ); Thu, 20 Aug 2009 18:10:59 -0400 Received: from smtp1.linux-foundation.org ([140.211.169.13]:45436 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754863AbZHTWK6 (ORCPT ); Thu, 20 Aug 2009 18:10:58 -0400 Date: Thu, 20 Aug 2009 15:10:24 -0700 From: Andrew Morton To: "tina.yang" Cc: jkosina@suse.cz, linux-kernel@vger.kernel.org Subject: Re: [PATCH] sysrq: add debug message to reboot event Message-Id: <20090820151024.a60567c7.akpm@linux-foundation.org> In-Reply-To: <4A8CB358.4090001@oracle.com> References: <4A89AC97.6030109@oracle.com> <20090819135827.168621d8.akpm@linux-foundation.org> <4A8CB358.4090001@oracle.com> X-Mailer: Sylpheed version 2.2.4 (GTK+ 2.8.20; i486-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 19 Aug 2009 19:22:16 -0700 "tina.yang" wrote: > Add debug message to detect keyboard vs non-keyboard triggered sysrq-b events. > This is to assist postmortem debugging on complicated computing setup with > large number of applications involved where reboot event had occurred, but > unclear of its origin. I'm still struggling to understand the motivation for the change. There are a large number of ways in which a machine can be rebooted, all the way down to a triple-fault. So it seems fairly arbitrary to add additional information to discriminate between just two of those ways. I assume that somewhere in your setup you have a script which does `echo b /proc/sysrq-trigger' and it took ages to work out that this was happening and you felt that having this code in place would have helped you debug that problem, yes? If so, I wonder what is the likelihood that someone else will have the same problem and will find this change useful. Perhaps we should do this for all sysrq events rather than just sysrq-b? > --- linux-2.6.18.i686/drivers/char/sysrq.c.orig 2009-08-13 10:55:57.526459000 -0700 > +++ linux-2.6.18.i686/drivers/char/sysrq.c 2009-08-13 10:58:10.798739000 -0700 2.6.18 is truly ancient and this patch doesn't apply at all to current development kernels.