From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262039AbVBURCu (ORCPT ); Mon, 21 Feb 2005 12:02:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262041AbVBURCu (ORCPT ); Mon, 21 Feb 2005 12:02:50 -0500 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:15317 "EHLO parcelfarce.linux.theplanet.co.uk") by vger.kernel.org with ESMTP id S262039AbVBURCs (ORCPT ); Mon, 21 Feb 2005 12:02:48 -0500 Message-ID: <421A142A.1060302@pobox.com> Date: Mon, 21 Feb 2005 12:02:34 -0500 From: Jeff Garzik User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040922 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Jon Smirl CC: lkml , fbdev , Greg KH Subject: Re: Problem: how to sequence reset of PCI hardware References: <9e47339105022023242e2fd9ce@mail.gmail.com> <42199DD9.10807@pobox.com> <9e47339105022108527e3c679d@mail.gmail.com> In-Reply-To: <9e47339105022108527e3c679d@mail.gmail.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Jon Smirl wrote: > On Mon, 21 Feb 2005 03:37:45 -0500, Jeff Garzik wrote: > >>You either need to execute the video BIOS to initialize the hardware >>registers, or initialize the hardware registers themselves. > > > That is what the user mode reset program does. > > The problem is, how do I get it to run before calling the device's > probe function? Most of the framebuffer drivers assume that the > hardware has already been reset in their probe code. You do precisely what you just said: run it before the device's probe function. That typically means either initramfs addition or using 'install command...' in /etc/modprobe.conf. Jeff