From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758275Ab2CBRWU (ORCPT ); Fri, 2 Mar 2012 12:22:20 -0500 Received: from hqemgate03.nvidia.com ([216.228.121.140]:13204 "EHLO hqemgate03.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756885Ab2CBRWT convert rfc822-to-8bit (ORCPT ); Fri, 2 Mar 2012 12:22:19 -0500 X-PGP-Universal: processed; by hqnvupgp07.nvidia.com on Fri, 02 Mar 2012 09:22:15 -0800 From: Stephen Warren To: Linus Walleij , "linux-kernel@vger.kernel.org" , "linux-arm-kernel@lists.infradead.org" CC: Grant Likely , Barry Song <21cnbao@gmail.com>, Shawn Guo , Thomas Abraham , Dong Aisheng , Rajendra Nayak , Haojian Zhuang , Linus Walleij Date: Fri, 2 Mar 2012 09:22:14 -0800 Subject: RE: [PATCH] pinctrl: resurrect verbose pinmux-pins Thread-Topic: [PATCH] pinctrl: resurrect verbose pinmux-pins Thread-Index: Acz4lixCNm7g+kafTBifQmUDymm0IwAAk6Og Message-ID: <74CDBE0F657A3D45AFBB94109FB122FF17BE861978@HQMAIL01.nvidia.com> References: <1330705873-10798-1-git-send-email-linus.walleij@stericsson.com> In-Reply-To: <1330705873-10798-1-git-send-email-linus.walleij@stericsson.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Linus Walleij wrote at riday, March 02, 2012 9:31 AM: > From: Linus Walleij > > This reintroduces the verbosity in the pinmux-pins file, now this > file will list all pins, their names, owning device (and if it's > a hog) and also in applicable cases the function and group making > use of that pin right now. > > Signed-off-by: Linus Walleij It'd be nice if we could defer this fix until after all the API rework, because it'll cause all kinds of conflicts otherwise. The type "struct pinmux_group" you added to struct pin_desc is removed by patches in my series. If this patch goes in before my series, it'd be best if it just stored the char* for the function instead of that type, so the rework by my patche series will be smaller. I don't think it make sense to store a "group" as owning a pin; each pin should be part of one muxing group as defined by the hardware so there isn't a concept of "which group acquired the pin", just "which function is programmed on the pin". Even if multiple "virtual groups" contain a pin (which I think would be a bad idea) then the important information from a HW perspective is the function and not the arbitrary group name. -- nvpublic