From mboxrd@z Thu Jan 1 00:00:00 1970 From: Simon Horman Subject: Re: [PATCH v4 07/14] ARM: dts: koelsch: use demuxer for I2C4 Date: Wed, 16 Nov 2016 14:47:37 +0100 Message-ID: <20161116134736.GA20704@verge.net.au> References: <20161106202032.5227-1-wsa+renesas@sang-engineering.com> <20161106202032.5227-8-wsa+renesas@sang-engineering.com> <20161115174450.GA21715@verge.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from kirsty.vergenet.net ([202.4.237.240]:34959 "EHLO kirsty.vergenet.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752203AbcKPNrp (ORCPT ); Wed, 16 Nov 2016 08:47:45 -0500 Content-Disposition: inline In-Reply-To: Sender: linux-i2c-owner@vger.kernel.org List-Id: linux-i2c@vger.kernel.org To: Geert Uytterhoeven Cc: Wolfram Sang , Linux-Renesas , Linux I2C , Ulrich Hecht On Wed, Nov 16, 2016 at 11:35:22AM +0100, Geert Uytterhoeven wrote: > Hi Simon, > > On Tue, Nov 15, 2016 at 6:44 PM, Simon Horman wrote: > >> i2c4 shares pins with vin0. Hence enabling the former breaks the latter: > >> > >> sh-pfc e6060000.pfc: pin GP_4_13 already requested by > >> e6520000.i2c; cannot claim for e6ef0000.video > >> sh-pfc e6060000.pfc: pin-141 (e6ef0000.video) status -22 > >> sh-pfc e6060000.pfc: could not request pin 141 (GP_4_13) from > >> group vin0_data24 on device sh-pfc > >> rcar-vin e6ef0000.video: Error applying setting, reverse things back > >> rcar-vin: probe of e6ef0000.video failed with error -22 > >> > >> There may be similar issues on other boards. Haven't checked yet. > > > > Thanks, I will drop this patch for now. > > > > I checked the boot logs of other boards with similar patches and didn't see > > anything there. > > Do you have CONFIG_VIDEO_ADV7604=y? No, mainly because its not in shmobile_defconfig. I tried again with that option enabled and still didn't see anything of interest in kernel log. > I expect the issue to be present on gose, too. So far I didn't see it > there, yet. > But the vin DTS of gose is different from koelsch, causing vin0 not to be > initialized? It seems so.