From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753258AbaILI6p (ORCPT ); Fri, 12 Sep 2014 04:58:45 -0400 Received: from metis.ext.pengutronix.de ([92.198.50.35]:40867 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753084AbaILI6o (ORCPT ); Fri, 12 Sep 2014 04:58:44 -0400 Date: Fri, 12 Sep 2014 10:58:36 +0200 From: Uwe =?iso-8859-1?Q?Kleine-K=F6nig?= To: Wolfram Sang Cc: linux-i2c@vger.kernel.org, Jean Delvare , Magnus Damm , linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-sh@vger.kernel.org Subject: Re: [RFC 0/4] i2c: slave support framework for Linux devices Message-ID: <20140912085836.GS3755@pengutronix.de> References: <1410274470-12712-1-git-send-email-wsa@the-dreams.de> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <1410274470-12712-1-git-send-email-wsa@the-dreams.de> User-Agent: Mutt/1.5.21 (2010-09-15) X-SA-Exim-Connect-IP: 2001:67c:670:100:1d::c0 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-kernel@vger.kernel.org Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello Wolfram, On Tue, Sep 09, 2014 at 04:54:26PM +0200, Wolfram Sang wrote: > Finally, here is my take on the often desired feature that Linux can not only > be an I2C master, but also an I2C slave. Compared to my draft sent out last > week, this RFC has been tested on hardware (Renesas Lager board) and works \o/ > > One big part still missing is documentation, so brave ones need to "use the > source". However, this approach turned out to be even less intrusive than > expected, so that is hopefully a good sign. Some thoughts from reading through your patch set: - If I understand correctly you cannot register an i2c slave without also registering a master, right? I don't think this is troubling in practice, is it? For abstraction I would prefer to make these different concepts though. - The IMHO most (even only?) useful slave is "i2cslvdev", i.e. a userspace device. With that you wouldn't need device tree stuff, the driver would just offer the device if supported. Userspace then could care about the rest (slave address, functionality, ...). Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König | Industrial Linux Solutions | http://www.pengutronix.de/ |