From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 30 Dec 2002 17:05:46 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 30 Dec 2002 17:05:45 -0500 Received: from 12-231-249-244.client.attbi.com ([12.231.249.244]:47887 "HELO kroah.com") by vger.kernel.org with SMTP id ; Mon, 30 Dec 2002 17:05:44 -0500 Date: Mon, 30 Dec 2002 14:09:12 -0800 From: Greg KH To: Jaroslav Kysela Cc: Adam Belay , LKML Subject: Re: [PATCH] proposed PnP layer changes/fixes and cleanups Message-ID: <20021230220911.GD32324@kroah.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Dec 29, 2002 at 12:20:15PM +0100, Jaroslav Kysela wrote: > Hi, > > I've revised the PnP code and found numerous of simple bugs. Also, > my opinion is that the resctriction of resources to only ones hardcoded by > hardware vendors is not a good idea. Hardware usually supports more > configurations (especially ISA PnP devices). > I've tried to implement a configuration template, so driver can > pass it's own resource map in the probe phase. Also, I've added a new > member to the driver structures - flags. The flag > PNP_DRIVER_DO_NOT_ACTIVATE instructs the pnp code that the device > shouldn't be activated before the probe() callback is entered. It's > necessary to implement this behaviour to allow passing of a new > configuration template (thus calling pnp_activate_dev() from the probe() > callback). > Also, the resources can be locked now (I've changed the ro flag to > lock_resources). These changes look good, I'll send them on to Linus. thanks, greg k-h