From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263429AbTEMVVg (ORCPT ); Tue, 13 May 2003 17:21:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263437AbTEMVVe (ORCPT ); Tue, 13 May 2003 17:21:34 -0400 Received: from air-2.osdl.org ([65.172.181.6]:46211 "EHLO mail.osdl.org") by vger.kernel.org with ESMTP id S263429AbTEMVUZ (ORCPT ); Tue, 13 May 2003 17:20:25 -0400 Date: Tue, 13 May 2003 14:33:25 -0700 (PDT) From: Patrick Mochel X-X-Sender: mochel@cherise To: Matt Domsch cc: Greg KH , , , Subject: Re: [RFC][PATCH] Dynamic PCI Device IDs In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 13 May 2003, Patrick Mochel wrote: > > On Tue, 6 May 2003, Matt Domsch wrote: > > > > You can't just call driver_attach(), as the bus semaphore needs to be > > > locked before doing so.  In short, you almost need to duplicate > > > bus_add_driver(), but not quite :) > > > > Right, and it seems to work. I made driver_attach non-static, declared > > it extern in pci.h, and call it in pci-driver.c while holding the bus > > semaphore and references to the driver and the bus. This also let me > > delete my probe_each_pci_dev() function and let the driver core > > handle it. > > > > Pat, can you ack the changes to bus.c and device.h please? > > ACK. I'll add them to my tree. I take that back, since it's already checked into your BK tree. Go forth and merge. -pat