Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932509AbdIFOAG (ORCPT ); Wed, 6 Sep 2017 10:00:06 -0400 Received: from mail-wm0-f65.google.com ([74.125.82.65]:37453 "EHLO mail-wm0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932378AbdIFOAC (ORCPT ); Wed, 6 Sep 2017 10:00:02 -0400 X-Google-Smtp-Source: ADKCNb6SQvyxKPkV07QUucW/Z8Ka/Cntxr6yT9zpExR55YCSsIDpDnhOGTcMI/rEJtQ2YV9X0SugwQ== Date: Wed, 6 Sep 2017 15:59:59 +0200 From: Rocco Folino To: Alban Cc: Ralf Baechle , robh+dt@kernel.org, mark.rutland@arm.com, linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, devicetree@vger.kernel.org, Antony Pavlov , John Crispin Subject: Re: [PATCH] MIPS: ath79: support devicetree selection Message-ID: <20170906135959.GA25439@void> Mail-Followup-To: Alban , Ralf Baechle , robh+dt@kernel.org, mark.rutland@arm.com, linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, devicetree@vger.kernel.org, Antony Pavlov , John Crispin References: <20170906111435.GA1856@linux-mips.org> <20170906142005.67586253@avionic-0141> <20170906123200.GA21761@void> <20170906153543.412774d5@avionic-0141> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170906153543.412774d5@avionic-0141> User-Agent: Mutt/1.8.3 (2017-05-23) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1172 Lines: 31 On Wed, Sep 06, 2017 at 03:35:43PM +0200, Alban wrote: > On Wed, 6 Sep 2017 14:32:00 +0200 > Rocco Folino wrote: > > > On Wed, Sep 06, 2017 at 02:20:05PM +0200, Alban wrote: > > > On Wed, 6 Sep 2017 13:14:35 +0200 > > > Ralf Baechle wrote: > > > > > > > On Fri, Aug 18, 2017 at 05:32:42PM +0200, Rocco Folino wrote: > > > > > > > > > Allow to choose devicetrees from Kconfig. > > > > > > > > > > Signed-off-by: Rocco Folino > > > > > > I don't really see the point of this patch. Building the dtb doesn't > > > take any significant time, so why add this extra complexity? > > > > Because you need to select the SoC type in order to enable some > > drivers, for example on the AR9331 to use the serial you need the > > CONFIG_SERIAL_AR933X which depends on the CONFIG_SOC_AR933X. > > Seeing as this driver is the only one that make use of CONFIG_SOC_AR933X > I would prefer removing this dependency. It would also open the way to have > the driver built in COMPILE_TEST. A few more fixes might be needed but that > would be better than such a workaround. Yes, I agree! Thanks, Rocco