From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751580AbXDLTKh (ORCPT ); Thu, 12 Apr 2007 15:10:37 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751600AbXDLTKh (ORCPT ); Thu, 12 Apr 2007 15:10:37 -0400 Received: from smtp.osdl.org ([65.172.181.24]:55698 "EHLO smtp.osdl.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751564AbXDLTKg (ORCPT ); Thu, 12 Apr 2007 15:10:36 -0400 Date: Thu, 12 Apr 2007 12:10:28 -0700 From: Andrew Morton To: Andi Kleen Cc: linux-kernel@vger.kernel.org Subject: Re: [PATCH] Add keyboard blink driver Message-Id: <20070412121028.d7e5c4b6.akpm@linux-foundation.org> In-Reply-To: <200704121727.27892.ak@novell.com> References: <200704121727.27892.ak@novell.com> X-Mailer: Sylpheed version 2.2.7 (GTK+ 2.8.6; i686-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 X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 12 Apr 2007 17:27:27 +0200 Andi Kleen wrote: > Add a blink driver for debugging > > Simple driver that blinks the keyboard LEDs when loaded. Useful > for checking that the kernel is still alive or for crashdumping. > The kdump kernel currently typically doesn't give any feedback > on the screen because it often cannot reinitialize the graphics > state. This patch allows it to at least show blinking lights > like panic. OK.. > --- /dev/null > +++ linux/drivers/misc/blink.c > @@ -0,0 +1,26 @@ > +#include > +#include > +#include > +#include > + > +static void do_blink(unsigned long data); > + > +static DEFINE_TIMER(blink_timer, do_blink, 0 ,0); > + > +static void do_blink(unsigned long data) > +{ > + static long count; > + panic_blink(count++); > + blink_timer.expires = jiffies + msecs_to_jiffies(1); > + add_timer(&blink_timer); > +} > + > +static int blink_init(void) > +{ > + printk(KERN_INFO "Enabling keyboard blinking\n"); > + do_blink(0); > + return 0; > +} > + > +module_init(blink_init); panic_blink is NULL in lots of setups. `modprobe blink' will be deadly. I wonder if this facility would be more effective if it were to use schedule_delayed_work(). That way, we're using more of the kernel and it will more reliably detect kernel deadness.