[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87bm4rtq1c.fsf@concordia.ellerman.id.au>
Date: Tue, 08 Jan 2019 21:14:07 +1100
From: Michael Ellerman <mpe@...erman.id.au>
To: Rob Herring <robh+dt@...nel.org>,
Corentin Labbe <clabbe@...libre.com>
Cc: Benjamin Herrenschmidt <benh@...nel.crashing.org>,
Mark Rutland <mark.rutland@....com>,
Paul Mackerras <paulus@...ba.org>, devicetree@...r.kernel.org,
"linux-kernel\@vger.kernel.org" <linux-kernel@...r.kernel.org>,
linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>
Subject: Re: [PATCH] powerpc: build dtb even without COMPILE_TEST
Rob Herring <robh+dt@...nel.org> writes:
> On Fri, Jan 4, 2019 at 3:58 AM Corentin Labbe <clabbe@...libre.com> wrote:
>>
>> I wanted to test the virtex440-ml507 qemu machine and found that the dtb
>> for it was not builded.
>
> Just do:
>
> make virtex440-ml507.dtb
I actually thought you had to do that, so I've never done anything different.
>> All powerpc DTB are only built when CONFIG_OF_ALL_DTBS is set which depend on
>> COMPILE_TEST.
>> But building DTB is not related to a "compile build test".
>
> But it is. We normally only build the dtbs for enabled platforms just
> like we only build platform/driver code for enabled platforms. It's
> hidden behind COMPILE_TEST so it only gets enabled for
> allmodconfig/allyesconfig builds.
>
>> So this patch made building of DTB independent of COMPILE_TEST (by
>> depending only on the PPC arch)
>> A better selection of which DTB to build could be done in the future
>> like that do the ARM arch.
>
> No reason you can't start that now for the platform you care about.
> You don't have to convert everyone.
Yeah that seems like the best option. I'm happy to take a patch just for
that target.
cheers
Powered by blists - more mailing lists