From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 1 Jun 2001 10:01:03 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 1 Jun 2001 10:00:53 -0400 Received: from [203.34.97.3] ([203.34.97.3]:15364 "HELO mail.ocs.com.au") by vger.kernel.org with SMTP id ; Fri, 1 Jun 2001 10:00:45 -0400 X-Mailer: exmh version 2.1.1 10/15/1999 From: Keith Owens To: Matt Chapman cc: Dag Brattli , linux-kernel@vger.kernel.org, linux-irda@pasta.cs.uit.no Subject: Re: [PATCH] for Linux IRDA initialisation bug 2.4.5 In-Reply-To: Your message of "Fri, 01 Jun 2001 23:32:46 +1000." <20010601233245.A10478@cse.unsw.edu.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 01 Jun 2001 23:59:13 +1000 Message-ID: <6679.991403953@ocs3.ocs-net> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 1 Jun 2001 23:32:46 +1000, Matt Chapman wrote: >I've found that if you compile IRDA into the kernel, irda_proto_init >gets called twice - once at do_initcalls time, and once explicitly >in do_basic_setup - eventually resulting in a hang (as >register_netdevice_notifier gets called twice with the same struct, >and it's list becomes circular). The suggested patch has one non-obvious side effect which somebody in irda needs to verify is OK. Previously irda_proto_init() and irda_device_init() were called after every other driver had initialized. Now irda_proto_init() is called based on the object order in the top level Makefile, so irda is initialized before i2c, telephony, acpi and mddev. Is this a valid initialization order? If not, move DRIVERS-$(CONFIG_IRDA) += drivers/net/irda/irda.o to the end of the drivers list and document why it needs to be there.