From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263737AbTHBNmv (ORCPT ); Sat, 2 Aug 2003 09:42:51 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263738AbTHBNmv (ORCPT ); Sat, 2 Aug 2003 09:42:51 -0400 Received: from sunpizz1.rvs.uni-bielefeld.de ([129.70.123.31]:42208 "EHLO mail.rvs.uni-bielefeld.de") by vger.kernel.org with ESMTP id S263737AbTHBNmu (ORCPT ); Sat, 2 Aug 2003 09:42:50 -0400 Subject: Re: Linux 2.4.22-pre10-ac1 From: Marcel Holtmann To: Alan Cox Cc: Manuel Estrada Sainz , "Barry K. Nathan" , Linux Kernel Mailing List In-Reply-To: <200308021229.h72CT5128965@devserv.devel.redhat.com> References: <200308021229.h72CT5128965@devserv.devel.redhat.com> Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Ximian Evolution 1.0.5 Date: 02 Aug 2003 15:42:27 +0200 Message-Id: <1059831754.22190.102.camel@pegasus> Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi Alan, > > not quite true. If hotplug is not enabled it tells the driver that the > > firmware can't be loaded. It is the same if hotplug_path is zero, or you > > The ifdef should be there, or firmware should depend on hotplug, and > probably the firmware users should also depend on hotplug I definitively prefer the #ifdef, because the firmware loader should automaticly selected and compiled if a driver needs it. But let a driver depend on hotplug can not be the solution, because some drivers maybe also work if the firmware loading fails. Regards Marcel