From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264289AbTEPACm (ORCPT ); Thu, 15 May 2003 20:02:42 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264292AbTEPACl (ORCPT ); Thu, 15 May 2003 20:02:41 -0400 Received: from pizda.ninka.net ([216.101.162.242]:36264 "EHLO pizda.ninka.net") by vger.kernel.org with ESMTP id S264289AbTEPACk (ORCPT ); Thu, 15 May 2003 20:02:40 -0400 Date: Thu, 15 May 2003 17:14:57 -0700 (PDT) Message-Id: <20030515.171457.02279102.davem@redhat.com> To: chas@locutus.cmf.nrl.navy.mil Cc: linux-kernel@vger.kernel.org Subject: Re: [PATCH][ATM] add reference counting to atm_dev From: "David S. Miller" In-Reply-To: <200305152020.h4FKKfGi014696@locutus.cmf.nrl.navy.mil> References: <20030515.131021.104054490.davem@redhat.com> <200305152020.h4FKKfGi014696@locutus.cmf.nrl.navy.mil> X-FalunGong: Information control. X-Mailer: Mew version 2.1 on Emacs 21.1 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: chas williams Date: Thu, 15 May 2003 16:20:41 -0400 i suspect atm doesnt use rtnl because its register/unregister function dont use the underlying netdevice layer (when perhaps it should). There is no correlation between whether you use netdevice or not and whether network configuration changes should be synchronized using the RTNL semaphore :-) thanks for the pointer. No problem. here is the the atm_dev locking patch with the __module_get change: Ok, I'll apply this. But long term we really need to clean out the cobwebs here, use RTNL, do solid module refcounting etc.