From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753990AbbIRRRQ (ORCPT ); Fri, 18 Sep 2015 13:17:16 -0400 Received: from lists.s-osg.org ([54.187.51.154]:38144 "EHLO lists.s-osg.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751596AbbIRRRO (ORCPT ); Fri, 18 Sep 2015 13:17:14 -0400 Subject: Re: [PATCH] gpio: mention in DT binding doc that -gpio is also supported To: Alexandre Courbot References: <1442453638-14246-1-git-send-email-javier@osg.samsung.com> From: Javier Martinez Canillas X-Enigmail-Draft-Status: N1110 Cc: Linux Kernel Mailing List , "linux-gpio@vger.kernel.org" , Linus Walleij , Alexandre Courbot , "devicetree@vger.kernel.org" Message-ID: <55FC4714.1010804@osg.samsung.com> Date: Fri, 18 Sep 2015 19:17:08 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.0.1 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello Alexandre, On 09/18/2015 05:44 PM, Alexandre Courbot wrote: > On Thu, Sep 17, 2015 at 10:33 AM, Javier Martinez Canillas > wrote: >> The GPIO DT binding doc mentions that GPIO are mapped by defining >> a -gpios property in the consumer device's node but a -gpio >> sufix is also supported after commit: >> >> dd34c37aa3e8 ("gpio: of: Allow -gpio suffix for property names") >> >> Update the DT binding documentation to match the implementation. >> >> Signed-off-by: Javier Martinez Canillas >> >> --- >> Hello, >> >> The GPIO documentation was updated to mention that the -gpio sufix >> is also supported in patch https://lkml.org/lkml/2015/9/1/117 that >> already landed in Torvalds tree. >> >> I now noticed that the DT binding also only mentions -gpios so I'm >> posting this patch that adds -gpio to the DT binding documentation. > > I think I remember that -gpio is considered obsolete and its use > should thus not be encouraged, which is the reason why the > documentation does not mention it. We could mention it and add a note > saying that it should not be used for new bindings, but all in all > isn't it better to keep the documentation clear of such use cases that > will not be accepted for new patches anyway? > I agree that if that's the case then it should be documented. Currently by reading the docs there is no way to tell if -gpio was only added to support old DT bindings and should not be used or if is that the docs were not updated when -gpio parsing was added to gpiolib. I can re-spin the patch making it clear that even when the -gpio suffix is supported, it's only there for compatibility reasons and should not be used for newer bindings. And also Documentation/gpio/board.txt has to be updated now since now it mentions -gpio but does not say that should not be used. Best regards, -- Javier Martinez Canillas Open Source Group Samsung Research America