From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thierry Reding Subject: Re: [PATCH v3] of: replace Asahi Kasei Corp venter prefix Date: Fri, 14 Nov 2014 10:30:17 +0100 Message-ID: <20141114093016.GB3810@ulmo> References: <20140922130026.GK3290@leverpostej> <20140922132743.GM3290@leverpostej> <87y4rfhmgl.wl%kuninori.morimoto.gx@renesas.com> <1415845309.4269.6.camel@perches.com> <20141113025105.GA16652@verge.net.au> <87vbmjhj00.wl%kuninori.morimoto.gx@renesas.com> <20141113110740.GE3815@sirena.org.uk> <87k32yd0bd.wl%kuninori.morimoto.gx@renesas.com> <5465BFA2.9050706@nvidia.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="PmA2V3Z32TCmWXqI" Return-path: Content-Disposition: inline In-Reply-To: <5465BFA2.9050706@nvidia.com> Sender: linux-sh-owner@vger.kernel.org To: Alexandre Courbot Cc: Kuninori Morimoto , Mark Brown , Mark Rutland , Simon Horman , Joe Perches , Geert Uytterhoeven , Kuninori Morimoto , Olof Johansson , Stephen Warren , Rob Herring , "linux-kernel@vger.kernel.org" , "devicetree@vger.kernel.org" , ALSA Development Mailing List , Linux-sh list , Magnus , "grant.likely@linaro.org" , linux-arm-kernel@lists.infradead.org, linux-tegra@vger.kernel.org List-Id: devicetree@vger.kernel.org --PmA2V3Z32TCmWXqI Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Nov 14, 2014 at 05:38:58PM +0900, Alexandre Courbot wrote: > On 11/14/2014 10:43 AM, Kuninori Morimoto wrote: [...] > >diff --git a/arch/arm/boot/dts/tegra20-seaboard.dts b/arch/arm/boot/dts/= tegra20-seaboard.dts > >index a1d4bf9..7f5cf80 100644 > >--- a/arch/arm/boot/dts/tegra20-seaboard.dts > >+++ b/arch/arm/boot/dts/tegra20-seaboard.dts > >@@ -405,7 +405,7 @@ > > clock-frequency =3D <400000>; > > > > magnetometer@c { > >- compatible =3D "ak,ak8975"; > >+ compatible =3D "asahi-kasei,ak8975"; >=20 > Mmm. So does this mean this device was never probed because the driver did > not recognize its compatible property? I2C (like SPI) has a fallback in case the OF table yields no match. The of_modalias_node() that it uses extracts the module name from the device tree node's compatible string by skipping the vendor prefix. The resulting string is then set as the I2C client's name and causes the I2C standard ID matching to succeed (see of_i2c_register_devices() for how this is done in detail). > I cannot find "ak,ak8975" anywhere else in the kernel. >=20 > If so, >=20 > Acked-by: Alexandre Courbot Technically this is breaking backwards-compatibility from a DT perspective, but I think it's safe to do because the Linux kernel has a way of dealing with this and other operating systems should be able to deal with this in a similar way (or even easier by letting the driver match on the ak prefix as well for this particular chip). So: Acked-by: Thierry Reding --PmA2V3Z32TCmWXqI Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJUZcuoAAoJEN0jrNd/PrOhijEP/0fWBqWZTvD71aj6UyH4xq3U 9EmL389EXehLKgnx7cjPW98QN0MPEDHotakxbVPD3Vn2IhU/SI2/c1w7BAQH8OdM FZjm1ujoHRk5ePM27/23OWPZd5L/9p54K7ko5qpZvxDj4tmxcMiOWL694RgSRW8q HkuFd+WcVOdD67uQRcY5xRNRB/egnLzFI5N2jaQUZE8adL9OTWo6objaOigTXAFl mMlqIpIwK9oGhuJygi05ESp3LcJYMA8i3kdKJv45YH8ctKonRBOmOQqjgxjgMeF6 drzGcz/oiIgd4vV41N9OsKWyTY5gamX/6t604ur+UyKlf/7vgiW72EGMflwcMyKd reps8t+ZrRQfDBgqy9/+cpQsSnhnLLhFjl/b3jkveNZoEFwf7MWmNdF4VdMdD/mV BKPNkHHQDI5xbAZJpdZfD9o5zWupkxFt0PrvU8fg78UPXg/pG8Pa91m6S8w77iD6 NyJmWdc8x3R0ADT4DtLelemJalcPaJzcWABYlgDQMBSCms6o6KpFVfIc0VqQUtup 1TzB+QkXNhJdWd6cVbK/XF9wyuiwzCfmXR0+gAK1hOxHfj8IKt+4MtiRDSKmdgSH S3Cz7OcHCqElXp4QQpuKtNkp1CQffH9rzqF4/aH3zXxayCyz+8hI7a3Twet9UauE Z967FL/gqnn5QOLMuUfF =BXjP -----END PGP SIGNATURE----- --PmA2V3Z32TCmWXqI--