From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail-ew0-f207.google.com ([209.85.219.207]:33467 "EHLO mail-ew0-f207.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753108AbZKHTfP (ORCPT ); Sun, 8 Nov 2009 14:35:15 -0500 Received: by ewy3 with SMTP id 3so2544885ewy.37 for ; Sun, 08 Nov 2009 11:35:19 -0800 (PST) From: Ivo van Doorn To: Bartlomiej Zolnierkiewicz Subject: Re: [PATCH 8/9] rt2800: add eFuse EEPROM support code to rt2800lib Date: Sun, 8 Nov 2009 20:35:17 +0100 Cc: linux-wireless@vger.kernel.org, Gertjan van Wingerde References: <20091108133854.23584.86842.sendpatchset@localhost.localdomain> <200911081947.53754.bzolnier@gmail.com> <200911082021.32455.bzolnier@gmail.com> In-Reply-To: <200911082021.32455.bzolnier@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-15" Message-Id: <200911082035.18197.IvDoorn@gmail.com> Sender: linux-wireless-owner@vger.kernel.org List-ID: > > > > I think that they shouldn't have been added in the first place and I'll be > > > > happy to add patch removing them to rt2800 tree (since code savings seem to > > > > be really marginal and not worth the maintenance cost). > > > > > > Removing the ifdefs entirely would be fine. > > > > Like I said before -- this would mean driver's behavior change. Even though > > WISOC code is currently dead (RALINK_RT288X and RALINK_RT305X are never set) > > I prefer to not "overload" patches with logically different changes. > > > > If you feel strongly about it please fix it in rt2x00 code and rt2800 tree > > will deal with it, or alternatively please send me an incremental patch. > > BTW the patch's impact is _320_ bytes increase of rt2800lib (on x86-64 so > it is probably much less on the affected embedded architectures): > > text data bss dec hex filename > before: > 16916 0 0 16916 4214 drivers/net/wireless/rt2x00/rt2800lib.o > after: > 17281 0 0 17281 4381 drivers/net/wireless/rt2x00/rt2800lib.o > > for the _completely_ dead code (because embedded WISOC support is never > enabled) that probably will be changed over anyway later during development. As mentioned: Wisoc is not dead code, it is there for the platform devices with integrated rt2800 chipsets. If you look at KConfig you can see the RALINK_RT288X RALINK_RT305X are dependencies from configuration options... > IOW something like this is completely not worth to be worried about > and I would personally just skip it during review to not waste people's > time needlessly.. Well dependencies from/on other drivers is something important... Ivo