From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753891AbaBMJ57 (ORCPT ); Thu, 13 Feb 2014 04:57:59 -0500 Received: from cpsmtpb-ews07.kpnxchange.com ([213.75.39.10]:56344 "EHLO cpsmtpb-ews07.kpnxchange.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753382AbaBMJ54 (ORCPT ); Thu, 13 Feb 2014 04:57:56 -0500 Message-ID: <1392285474.30853.28.camel@x220> Subject: Re: [PATCH] Blackfin: bf537: rename "CONFIG_ADT7310" and friends From: Paul Bolle To: Lars-Peter Clausen Cc: Guenter Roeck , Steven Miao , adi-buildroot-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org Date: Thu, 13 Feb 2014 10:57:54 +0100 In-Reply-To: <51507126.2040002@metafoo.de> References: <1364211831.1390.285.camel@x61.thuisdomein> <20130325152258.GA17467@roeck-us.net> <51507126.2040002@metafoo.de> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.10.3 (3.10.3-1.fc20) Mime-Version: 1.0 Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 13 Feb 2014 09:57:55.0052 (UTC) FILETIME=[10E882C0:01CF28A2] X-RcptDomain: vger.kernel.org Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org [Replaced previous maintainer and list with current maintainer and list.] On Mon, 2013-03-25 at 16:45 +0100, Lars-Peter Clausen wrote: > On 03/25/2013 04:22 PM, Guenter Roeck wrote: > > On Mon, Mar 25, 2013 at 12:43:51PM +0100, Paul Bolle wrote: > >> The Kconfig symbol ADT7310 got removed in commit > >> 2b0c856ad9571013db8fc369194b7108dff3c18e ("staging:iio: Consolidate > >> adt7310 and adt7410 driver"). That commit did not touch the references > >> to CONFIG_ADT7310 and CONFIG_ADT7310_MODULE in the BF537-STAMP code. > >> Convert these now, to their ADT7410 equivalents. > >> > >> Signed-off-by: Paul Bolle > >> --- > >> 0) Untested. This took a bit of guesswork. I guessed that multiple AD > >> drivers can be enabled at the same time in this code. And I also guessed > >> that the "adt7310" modalias is still correct. > >> > >> 1) If this passes testing it could go in stable (3.8.y). > >> > >> arch/blackfin/mach-bf537/boards/stamp.c | 4 ++-- > >> 1 file changed, 2 insertions(+), 2 deletions(-) > >> > >> diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c > >> index 95114ed..23bb55d 100644 > >> --- a/arch/blackfin/mach-bf537/boards/stamp.c > >> +++ b/arch/blackfin/mach-bf537/boards/stamp.c > >> @@ -700,7 +700,7 @@ static struct bfin5xx_spi_chip ad7816_spi_chip_info = { > >> }; > >> #endif > >> > >> -#if defined(CONFIG_ADT7310) || defined(CONFIG_ADT7310_MODULE) > >> +#if defined(CONFIG_ADT7410) || defined(CONFIG_ADT7410_MODULE) > > > > It might be easier and more future-proof to just drop the conditional. > > If the driver isn't there, it won't be loaded anyway. It would also make > > back-porting much easier. > > > > Guenter > > > The situation is unfortunately really messy. The issue is that these are > add-on board and the same SPI chip select pin is used for multiple devices. > If we add the spi_board_info for all devices unconditionally the SPI core > will scream at us because we tried to register multiple devices with the > same chip select pin and none of the devices gets actually registered. This issue is still present in v3.14-rc2. Guenter's suggestion is the easiest way out. Should I submit a trivial patch that just removes the dead code depending on never defined CONFIG_ADT7310 and CONFIG_ADT7310_MODULE? Paul Bolle