From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lars-Peter Clausen Subject: Re: [PATCH v3 0/2] i2c/of: Populate multiplexed i2c busses from the device tree. Date: Fri, 13 Apr 2012 13:09:20 +0200 Message-ID: <4F880960.8010803@metafoo.de> References: <1334265263-13549-1-git-send-email-ddaney.cavm@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1334265263-13549-1-git-send-email-ddaney.cavm@gmail.com> Sender: linux-kernel-owner@vger.kernel.org To: David Daney Cc: Grant Likely , devicetree-discuss@lists.ozlabs.org, "Jean Delvare (PC drivers core)" , "Ben Dooks (embedded platforms)" , Wolfram Sang , Peter Korsgaard , Guenter Roeck , linux-i2c@vger.kernel.org, Rob Herring , linux-kernel@vger.kernel.org, David Daney List-Id: linux-i2c@vger.kernel.org On 04/12/2012 11:14 PM, David Daney wrote: > From: David Daney > > v3: Integrate changes from Lars-Peter Clausen to make better use of > the of_*() infrastructure. Get rid of ugly #ifdefs. > > v2: Update bindings to use "reg" insutead of "cell-index" > > v1: Unchanged from the original RFC where I said: > > We need to populate our I2C devices from the device tree, but some > of our systems have multiplexers which currently break this process. > > The basic idea is to have the generic i2c-mux framework propagate > the device_node for the child bus into the corresponding > i2c_adapter, and then call of_i2c_register_devices(). This means > that the device tree bindings for *all* i2c muxes must have some > common properties. I try to document these in mux.txt. > > This is now tested against 3.4-rc2 and is still working well. > I've been using these patches with a pca9548 and a pca9546 for a while now. Both: Tested-by: Lars-Peter Clausen