Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 4 Oct 2002 13:58:49 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 4 Oct 2002 13:58:48 -0400 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:63750 "EHLO www.linux.org.uk") by vger.kernel.org with ESMTP id ; Fri, 4 Oct 2002 13:58:46 -0400 Message-ID: <3D9DD801.4090603@pobox.com> Date: Fri, 04 Oct 2002 14:03:45 -0400 From: Jeff Garzik User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Linus Torvalds CC: Greg KH , linux-kernel@vger.kernel.org Subject: Re: [BK PATCH] pcibios_* removals for 2.5.40 References: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1634 Lines: 40 Linus Torvalds wrote: > On Thu, 3 Oct 2002, Greg KH wrote: > >>Here's some changesets that remove the pcibios_find_class(), >>pci_find_device(), and pcibios_present() functions. These functions >>have been marked as obsolete since the 2.2 kernel, so it's about time >>that we removed them. > > > They are still in use by a lot of drivers.. I hate to break even more > drivers at this point in 2.5.x, and so quite frankly I'd rather just do > this in early 2.7.x instead. Unless somebody really steps up to the plate > and also fixes the drivers ("it's a ton of fun, and imagine all the > adoration you'll get from teenage girls/boys/ninja turtles for doing it") Removing pcibios_present() makes a lot of sense, I have considered it deprecated for quite a while now. Further (more a note to Greg), often the pci[bios]_present() call can be removed completely: when it is followed by a pci_find_xxx or pci_register_driver() probe, which will obviously not find anything if PCI bus is absent, the pci_present() call can be removed completely. I would love to deprecate pcibios_{read,write}_foo but I don't think we can remove them yet. Likewise for pcibios_find_xxx... I haven't look at all the examples, but I would be worried about breaking ancient (but working) code. That said, if you wanna do the cleanup, I wouldn't object to removing pcibios_find_xxx... Jeff - 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/