Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 29 Oct 2001 06:17:02 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 29 Oct 2001 06:16:52 -0500 Received: from workplace.tp1.ruhr-uni-bochum.de ([134.147.240.2]:28169 "EHLO localhost.localdomain") by vger.kernel.org with ESMTP id ; Mon, 29 Oct 2001 06:16:35 -0500 Date: Mon, 29 Oct 2001 12:17:02 +0100 (CET) From: Kai Germaschewski X-X-Sender: To: Greg KH cc: Keith Owens , Subject: Re: 2.4.13 errors and warnings In-Reply-To: <20011028100317.C8059@kroah.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 28 Oct 2001, Greg KH wrote: > These, and lots of the other pci_id table warnings are due to the tables > being used for MODULE_DEVICE_TABLE() information. When the code is not > compiled as modules, those tables are not needed. > > Hm, I guess I should look into some kind of macro to keep this from > happening... A couple of months ago I thought about this and could think of two possible solutions: o add a __moddevtable which expands to __devinitdata __attribute((unused)). Drawback: Needs changing of all drivers which produce the warning. o add a variable which references the table within MODULE_DEVICE_TABLE. I implemented this, minor drawback is that it costs 4 bytes per table. However, IIRC Keith didn't like it at this time. The best option, of course, is to move drivers to the new-style pci or whatever interface, such that the table actually gets used. But the middle of a stable series is not necessarily the best time to do so. --Kai - 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/