Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964841AbcJQMaY (ORCPT ); Mon, 17 Oct 2016 08:30:24 -0400 Received: from bear.ext.ti.com ([198.47.19.11]:36658 "EHLO bear.ext.ti.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932911AbcJQMaP (ORCPT ); Mon, 17 Oct 2016 08:30:15 -0400 Subject: Re: [PATCH 2/2] ARM: dts: da850: add a node for the LCD controller To: Laurent Pinchart References: <1475672732-17111-1-git-send-email-bgolaszewski@baylibre.com> <65fde145-0820-794d-d345-621f66cdacc0@ti.com> <7a2ffcd0-fe1d-c887-53b7-7cb5e1e61222@ti.com> <4975084.EGQPv58AK6@avalon> CC: Sekhar Nori , Bartosz Golaszewski , Kevin Hilman , Michael Turquette , Rob Herring , Frank Rowand , Mark Rutland , Peter Ujfalusi , Russell King , Karl Beldan , LKML , arm-soc , linux-drm , linux-devicetree , Jyri Sarha , David Airlie , Maxime Ripard , Karl Beldan From: Tomi Valkeinen Message-ID: Date: Mon, 17 Oct 2016 15:29:23 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.3.0 MIME-Version: 1.0 In-Reply-To: <4975084.EGQPv58AK6@avalon> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="l6bSklRD2LiOxRSLtCQjIH3TSIMxR933q" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 4740 Lines: 119 --l6bSklRD2LiOxRSLtCQjIH3TSIMxR933q Content-Type: multipart/mixed; boundary="qJh569mBsJEcvuhkT5OprdgUlUGd5FrL9"; protected-headers="v1" From: Tomi Valkeinen To: Laurent Pinchart Cc: Sekhar Nori , Bartosz Golaszewski , Kevin Hilman , Michael Turquette , Rob Herring , Frank Rowand , Mark Rutland , Peter Ujfalusi , Russell King , Karl Beldan , LKML , arm-soc , linux-drm , linux-devicetree , Jyri Sarha , David Airlie , Maxime Ripard , Karl Beldan Message-ID: Subject: Re: [PATCH 2/2] ARM: dts: da850: add a node for the LCD controller References: <1475672732-17111-1-git-send-email-bgolaszewski@baylibre.com> <65fde145-0820-794d-d345-621f66cdacc0@ti.com> <7a2ffcd0-fe1d-c887-53b7-7cb5e1e61222@ti.com> <4975084.EGQPv58AK6@avalon> In-Reply-To: <4975084.EGQPv58AK6@avalon> --qJh569mBsJEcvuhkT5OprdgUlUGd5FrL9 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 17/10/16 14:40, Laurent Pinchart wrote: > Hello, >=20 > On Monday 17 Oct 2016 10:33:58 Tomi Valkeinen wrote: >> On 17/10/16 10:12, Sekhar Nori wrote: >>> On Monday 17 October 2016 11:26 AM, Tomi Valkeinen wrote: >>>> On 15/10/16 20:42, Sekhar Nori wrote: >>>>>> diff --git a/arch/arm/boot/dts/da850.dtsi >>>>>> b/arch/arm/boot/dts/da850.dtsi >>>>>> index f79e1b9..32908ae 100644 >>>>>> --- a/arch/arm/boot/dts/da850.dtsi >>>>>> +++ b/arch/arm/boot/dts/da850.dtsi >>>>>> @@ -399,6 +420,14 @@ >>>>>> <&edma0 0 1>; >>>>>> dma-names =3D "tx", "rx"; >>>>>> }; >>>>>> + >>>>>> + display: display@213000 { >>>>>> + compatible =3D "ti,am33xx-tilcdc", "ti,da850-tilcdc"; >>>>> >>>>> This should instead be: >>>>> >>>>> compatible =3D "ti,da850-tilcdc", "ti,am33xx-tilcdc"; >>>>> >>>>> as the closest match should appear first in the list. >>>> >>>> Actually I don't think that's correct. The LCDC on da850 is not >>>> compatible with the LCDC on AM335x. I think it should be just >>>> "ti,da850-tilcdc". >>> >>> So if "ti,am33xx-tilcdc" is used, the display wont work at all? If th= ats >>> the case, I wonder how the patch passed testing. Bartosz? >> >> AM3 has "version 2" of LCDC, whereas DA850 is v1. They are quite >> similar, but different. >> >> The driver gets the version number from LCDC's register, and acts base= d >> on that, so afaik the compatible string doesn't really affect the >> functionality (as long as it matches). >> >> But even if it works with the current driver, I don't think >> "ti,am33xx-tilcdc" and "ti,da850-tilcdc" are compatible in the HW leve= l. >=20 > If the hardware provides IP revision information, how about just "ti,lc= dc" ? Maybe, and I agree that's the "correct" way, but looking at the history, it's not just once or twice when we've suddenly found out some difference or bug or such in an IP revision, or the integration to a SoC, that can't be found based on the IP revision. That's why I feel it's usually safer to have the SoC revision there in the compatible string. That said, we have only a few different old SoCs with LCDC (compared to, say, OMAP DSS) so in this case perhaps just "ti,lcdc" would be fine. Tomi --qJh569mBsJEcvuhkT5OprdgUlUGd5FrL9-- --l6bSklRD2LiOxRSLtCQjIH3TSIMxR933q Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJYBMQjAAoJEPo9qoy8lh71X2AP+gPWYfHK5yATUCODohFwpTeZ 9b2bm4rcvetJkd4AMK+kz5dxOtpGbvbICxlWRYiWiydaISUEQs4uZusR9bXiFdvy U8tbsyxP2/D9VdLQiEJi8Yn+X74H56VgPji1Ar6Sm1daNvqNW9ESKev0lINSUs0H tg34+txYm1YFr0fbBPoWQGZyROEafmRIDBDhSpIbxCa9uh6nwvHTXaftBZngLhH4 PEUzThaaORK0aXU9cUa5XKpkD0wTq9kE5IRwDkxPh+vF3TY4EpodnsPLIRMjjK7L 9lnwJZma/N1FQ2X7q+/y+QnnOdbm3cl2jxDAzZTpV7qaKOKArq9462yX7wncfX6Z vV3J6j3FJxwliHYsB3vY55INeiIKBIVYOjMmt0rv12S9WvHinIL6JfTEn29fXKpg scPMN4jM+9EW+IJ//jJcIx77hopmutBdsgOqRlYcN/Pc0f3z6+9jNveCp2kuUml4 WCNk3bW8mDAaanFU+K9Ev9DH19tXE0eRiv/PjKCUWKPRDB1694Yt6fE5T6lGvEBx X4EBpoZX8RaiN9M7FYmbvG5uy37pnB2tc8tEYN2lUQQaL7lVk4BV1fM0xm5VI2TP 2ttvREugavWkezF/7x4Ga2ri+cNdvTrSOJyLmXrFTzR0Y1EiQUUOCWUwT6Voa+lq Mu+bL3bP7QH++BOTmj8T =YCm/ -----END PGP SIGNATURE----- --l6bSklRD2LiOxRSLtCQjIH3TSIMxR933q--