From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756854AbXEQNfR (ORCPT ); Thu, 17 May 2007 09:35:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754312AbXEQNfG (ORCPT ); Thu, 17 May 2007 09:35:06 -0400 Received: from pentafluge.infradead.org ([213.146.154.40]:52961 "EHLO pentafluge.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753995AbXEQNfF (ORCPT ); Thu, 17 May 2007 09:35:05 -0400 Date: Thu, 17 May 2007 06:32:36 -0700 From: Greg KH To: Ben Collins Cc: Roland Dreier , Linus Torvalds , linux-kernel@vger.kernel.org, linux-usb-devel@lists.sourceforge.net, Ganesh Varadarajan Subject: Re: [linux-usb-devel] [PATCH] Remove duplicate ID in ipaq driver Message-ID: <20070517133236.GA16009@kroah.com> References: <1179348840.6510.16.camel@cunning> <1179351408.6510.24.camel@cunning> <20070517124333.GB20055@kroah.com> <1179406940.6510.90.camel@cunning> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1179406940.6510.90.camel@cunning> User-Agent: Mutt/1.5.15 (2007-04-06) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, May 17, 2007 at 09:02:20AM -0400, Ben Collins wrote: > On Thu, 2007-05-17 at 05:43 -0700, Greg KH wrote: > > On Wed, May 16, 2007 at 05:36:48PM -0400, Ben Collins wrote: > > > On Wed, 2007-05-16 at 13:59 -0700, Roland Dreier wrote: > > > > > /* The first entry is a placeholder for the insmod-specified device */ > > > > > - { USB_DEVICE(0x049F, 0x0003) }, > > > > > > > > Is it obvious why this patch is correct? Especially given the > > > > comment just before the line you delete, and the code > > > > > > > > if (vendor) { > > > > ipaq_id_table[0].idVendor = vendor; > > > > ipaq_id_table[0].idProduct = product; > > > > } > > > > > > > > in ipaq_init()? > > > > > > My mistake, quick on the patching going through this dupe list. > > > > > > Might I add that this is terrible use of the device table, though. > > > Clutters userspace, and adds processing to module-init-tools programs. > > > > It's a hold-over from the times when we didn't have the sysfs "add a new > > id" interface for usb-serial drivers, which only recently was created. > > > > So we just have to live with it, and the infinitesimal speed hit it > > creates :) > > Any objection to adding it to planned-for-removal and spitting out a > printk when someone uses the "feature"? No, it's a module parameter and it's much easier to use for some systems than the sysfs file way. It's just not worth breaking userspace APIs for no good reason. thanks, greg k-h