Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 5 Dec 2001 09:19:03 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 5 Dec 2001 09:18:44 -0500 Received: from mail207.mail.bellsouth.net ([205.152.58.147]:24549 "EHLO imf07bis.bellsouth.net") by vger.kernel.org with ESMTP id ; Wed, 5 Dec 2001 09:18:41 -0500 Message-ID: <3C0E2CBA.57412C69@mandrakesoft.com> Date: Wed, 05 Dec 2001 09:18:34 -0500 From: Jeff Garzik Organization: MandrakeSoft X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.16 i686) X-Accept-Language: en MIME-Version: 1.0 To: Eric Lammerts CC: Alan Cox , Andrew Morton , Josh McKinney , linux-kernel@vger.kernel.org Subject: Re: Fwd: binutils in debian unstable is broken. In-Reply-To: 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 Eric Lammerts wrote: > change the drivers like this: > > #ifdef DEVEXIT_LINKED > remove: firestream_remove_one, > #endif > > to this: > > remove: DEVEXIT_FUNC(firestream_remove_one), The overall situation is not great, but I think I prefer this, or something like this, to the ifdef. I'm tempted to say we should add a per-driver function that calls BUG() instead of a generic function which calls panic(), though. That will be a bit more informative to users. For the case where the real xxx_remove_one is dropped, it is a kernel bug if that code is -ever- called... Jeff -- Jeff Garzik | Only so many songs can be sung Building 1024 | with two lips, two lungs, and one tongue. MandrakeSoft | - nomeansno - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/