From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S272392AbTGYXt2 (ORCPT ); Fri, 25 Jul 2003 19:49:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S272393AbTGYXtZ (ORCPT ); Fri, 25 Jul 2003 19:49:25 -0400 Received: from pc2-cwma1-4-cust86.swan.cable.ntl.com ([213.105.254.86]:48512 "EHLO lxorguk.ukuu.org.uk") by vger.kernel.org with ESMTP id S272392AbTGYXtU (ORCPT ); Fri, 25 Jul 2003 19:49:20 -0400 Subject: Re: [PATCH] Remove module reference counting. From: Alan Cox To: Stephen Hemminger Cc: Rusty Russell , davem@redhat.com, arjanv@redhat.com, Linus Torvalds , greg@kroah.com, Linux Kernel Mailing List In-Reply-To: <20030725122651.4aedc768.shemminger@osdl.org> References: <20030725173900.D7DE12C2A9@lists.samba.org> <20030725122651.4aedc768.shemminger@osdl.org> Content-Type: text/plain Content-Transfer-Encoding: 7bit Organization: Message-Id: <1059175489.1206.11.camel@dhcp22.swansea.linux.org.uk> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 (1.2.2-5) Date: 26 Jul 2003 00:24:49 +0100 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Gwe, 2003-07-25 at 20:26, Stephen Hemminger wrote: > > If module removal is to be a rare and unusual event, it > > doesn't seem so sensible to go to great lengths in the code to handle > > just that case. In fact, it's easier to leave the module memory in > > place, and not have the concept of parts of the kernel text (and some > > types of kernel data) vanishing. Uggh. There is a difference between taking the approach that some stuff is hard to handle and gets into trouble for using MOD_INC/DEC so is unsafe, and doing the locking from the caller, or arranging that you know the device is quiescent in the unload path and not allowing unloading to work properly. I've got drivers that use MOD_INC/DEC and are technically unsafe, they by default now don't unload and its an incentive to fix them. I'd hate to have my box cluttering up and have to keep rebooting to test drivers because of inept implementations however.