From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751002AbdE3Gi5 (ORCPT ); Tue, 30 May 2017 02:38:57 -0400 Received: from mail.free-electrons.com ([62.4.15.54]:41576 "EHLO mail.free-electrons.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750733AbdE3Giz (ORCPT ); Tue, 30 May 2017 02:38:55 -0400 Date: Tue, 30 May 2017 08:38:43 +0200 From: Maxime Ripard To: Karsten Merker Cc: Jagan Teki , Jagan Teki , Chen-Yu Tsai , Sean Wang , Icenowy Zheng , Rob Herring , Mark Rutland , Catalin Marinas , Will Deacon , Michael Trimarchi , "linux-arm-kernel@lists.infradead.org" , devicetree@vger.kernel.org, "linux-kernel@vger.kernel.org" , linux-sunxi , Jagan Teki Subject: Re: [linux-sunxi] [PATCH 1/3] ARM: dts: sun7i-a20: Rename bananapi as bananapi m1 Message-ID: <20170530063843.yujrog2ctjkatnlk@flea.lan> References: <1496086228-2806-1-git-send-email-jagan@openedev.com> <20170529214551.GA1990@excalibur.cnev.de> <20170530062244.GA1962@excalibur.cnev.de> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="wixjneie55tyuu34" Content-Disposition: inline In-Reply-To: <20170530062244.GA1962@excalibur.cnev.de> User-Agent: NeoMutt/20170428 (1.8.2) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --wixjneie55tyuu34 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, May 30, 2017 at 08:22:44AM +0200, Karsten Merker wrote: > On Tue, May 30, 2017 at 10:00:49AM +0530, Jagan Teki wrote: > > On Tue, May 30, 2017 at 3:15 AM, Karsten Merker wro= te: > > > On Mon, May 29, 2017 at 07:30:26PM +0000, Jagan Teki wrote: > > >> From: Jagan Teki > > >> > > >> from BPI(BIPAI KEJI LIMITED) products the Bananapi board > > >> is named as 'Bananapi M1' and this is the starting > > >> bananapi board from M1 series. > > >> > > >> So rename dts and suffix 'M1' on model for the same, > > >> so-that next sequence on bananapi starts like M1 Plus, M2 and so..on > > >> > > >> Signed-off-by: Jagan Teki > > >> --- > > >> Note: Bananapi BPI product site > > >> http://www.banana-pi.org/product.html > > >> > > >> arch/arm/boot/dts/Makefile | 2 +- > > >> arch/arm/boot/dts/sun7i-a20-bananapi-m1.dts | 286 +++++++++++++++++= +++++++++++ > > >> arch/arm/boot/dts/sun7i-a20-bananapi.dts | 286 -----------------= ----------- > > >> 3 files changed, 287 insertions(+), 287 deletions(-) > > >> create mode 100644 arch/arm/boot/dts/sun7i-a20-bananapi-m1.dts > > >> delete mode 100644 arch/arm/boot/dts/sun7i-a20-bananapi.dts > > >> > > >> diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile > > >> index 45c6e65..1b086f0 100644 > > >> --- a/arch/arm/boot/dts/Makefile > > >> +++ b/arch/arm/boot/dts/Makefile > > >> @@ -851,7 +851,7 @@ dtb-$(CONFIG_MACH_SUN6I) +=3D \ > > >> sun6i-a31s-sinovoip-bpi-m2.dtb \ > > >> sun6i-a31s-yones-toptech-bs1078-v2.dtb > > >> dtb-$(CONFIG_MACH_SUN7I) +=3D \ > > >> - sun7i-a20-bananapi.dtb \ > > >> + sun7i-a20-bananapi-m1.dtb \ > > >> sun7i-a20-bananapi-m1-plus.dtb \ > > >> sun7i-a20-bananapro.dtb \ > > >> sun7i-a20-cubieboard2.dtb \ > > >> diff --git a/arch/arm/boot/dts/sun7i-a20-bananapi-m1.dts b/arch/arm/= boot/dts/sun7i-a20-bananapi-m1.dts > > >> new file mode 100644 > > >> index 0000000..8b97b89 > > >> --- /dev/null > > >> +++ b/arch/arm/boot/dts/sun7i-a20-bananapi-m1.dts > > >> @@ -0,0 +1,286 @@ > > >> +/* > > >> + * Copyright 2014 Hans de Goede > > >> + * > > >> + * Hans de Goede > > > [...] > > >> +/dts-v1/; > > >> +#include "sun7i-a20.dtsi" > > >> +#include "sunxi-common-regulators.dtsi" > > >> + > > >> +#include > > >> +#include > > >> + > > >> +/ { > > >> + model =3D "LeMaker Banana Pi M1"; > > >> + compatible =3D "lemaker,bananapi", "allwinner,sun7i-a20"; > > > [...] > > >> diff --git a/arch/arm/boot/dts/sun7i-a20-bananapi.dts b/arch/arm/boo= t/dts/sun7i-a20-bananapi.dts > > >> deleted file mode 100644 > > >> index ed2f35a..0000000 > > >> --- a/arch/arm/boot/dts/sun7i-a20-bananapi.dts > > >> +++ /dev/null > > > > > > NACK! > > > > > > Please neither rename the dts nor change the model string. Such a > > > change would make newer kernels unusable on many existing > > > installations without manual fixups by the end user. Linux > > > distributions use databases with model-specific setup information > > > (such as the dtb file name, the platform-specific bootscript to > > > use, usable kernel flavours (lpae or non-lpae), etc.) on kernel > > > installations and kernel upgrades, and those use the model string > > > as their key for finding the relevant information. If you change > > > either the dts file name or the model string inside the dts, > > > you'll effectively break the proper installation of newer kernel > > > versions on existing end user systems. > >=20 > > I understand your concerns about distribution change, but with new > > change in 'bananapi' brand owned by BIPAI KEJI(BPI) the model must > > need to update and this is not technically as Bananapi board it is > > Bananapi M1 [1] > >=20 > > These are generic changes based on the hardware vendor info. > >=20 > > [1] http://www.banana-pi.org/m1.html >=20 > Hello Jagan, >=20 > I have to disagree here. Whatever BIPAI KEJI(BPI) chooses to name > their products today or in the future doesn't change history.=20 > The "original" Banana Pi was sold under the LeMaker brand and it > was named "Banana Pi" and not "Banana Pi M1". The fact that > LeMaker has stopped selling their Banana Pi board and BIPAI > KEJI(BPI) (a different company than LeMaker!) now sells a board > that is compatible to the original "LeMaker Banana Pi" under the > name "Banana Pi M1" (and not "LeMaker Banana Pi M1" as you claim > in your modified dts) doesn't matter at all for an existing dts > and is in no way a valid reason to make an incompatible change > that breaks existing systems. >=20 > If you wanted to add a new (technically identical) dts with a > different model string under a new dts file name and keep the > existing one unchanged, I would find that unneccessary but at > least acceptable insofar as it wouldn't break existing end user > systems, but breaking existing systems in the wild is clearly > unacceptable. I assume that the kernel dts maintainers would > probably still object to such an addition as it is unneccessary > from a purely technical point of view if the boards are 100% > compatible, but that would of course be their decision to make. I fully agree, we shouldn't have to modify either the DT itself, the compatibles, or anything really, to keep up with the crazy naming scheme some random vendor invents. Maxime --=20 Maxime Ripard, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com --wixjneie55tyuu34 Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIcBAEBAgAGBQJZLRNzAAoJEBx+YmzsjxAgqyYP/11Hy30FMz4l5Uk/S46QOA6a CshkOKE0BK9DaXDlvwvShHifV6IQBSkWstCn2q0DebXFul1Qh1qqyA3Ncuc8e2FO LJwNd7PBPcfvJRKywO4AEXNJbaKgflz9yO3UVShCMzKrYMZzVL/4Q0aBup4O5v0i 3d6Jph3c0do6Ua0jEz9ERxyB3z0lxI2RelSi8VCr91slo768vU2u+cjEI+FxFPlc cz/FcUPAKIaA4YVEb/k+mTnz78ImaWDGhzETRqlOi/QACeCmt6oqjqGTUHyOt1jU BVIHE+dEukVykU8i7E827zM18Tl70gDSjCggTZimdiokLR3F/muuKYB0P7nDYonW mABeuWMGmK+BoTjiVGhGLPhCGXhSCC5DEBfFV2yZ6MU7fjSdDLCj9x1aXgfWdqba Q6svz0EmeIRM8Q+LrisTJhUXtgeX8IyGzPsNHHT71EqsaLTfOBjY2hGBk0tCbMog awqYTISU9A8RSwBqxVIDsmkxcvRIXcF3Gy+67iojZpObCJKQt6g2p91XzbEDToOW PaaQ7PZhqDmMdUxdOviTf6OIfJX2SfV2ydpgcZQrxMgO4W9+uEIVjghPAaddlub+ xiQ1cGGQZ7zwOySSDHsA9DWlTR5AkItk9c6RnZQBTw+EMR4TFzRIjMzv+Lmc61du BmJ+w7sd9iQvG+Wenbq3 =Q6bF -----END PGP SIGNATURE----- --wixjneie55tyuu34--