[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87tw4vf9wg.fsf@eliezer.anholt.net>
Date: Mon, 08 May 2017 12:03:59 -0700
From: Eric Anholt <eric@...olt.net>
To: Florian Fainelli <f.fainelli@...il.com>,
dri-devel@...ts.freedesktop.org,
bcm-kernel-feedback-list@...adcom.com, Ray Jui <rjui@...adcom.com>,
Scott Branden <sbranden@...adcom.com>,
Jon Mason <jonmason@...adcom.com>
Cc: linux-kernel@...r.kernel.org, mircea.carausu@...adcom.com
Subject: Re: [PATCH] drm/vc4: Enable selection in Kconfig on any BCM platform.
Florian Fainelli <f.fainelli@...il.com> writes:
> On 05/08/2017 11:18 AM, Eric Anholt wrote:
>> With the Cygnus port, we needed to add at least "|| ARCH_BCM_CYGNUS"
>> to let the module get built on a cygnus-only kernel. However, I
>> anticipate having a port for Kona soon, so just present the module on
>> all of BCM.
>
> This seems reasonable, but by replacing ARCH_BCM2835 which is common to
> ARM/Linux and ARM64/Linux, you are no longer allowing an ARM64 systems
> to benefit from this driver unless COMPILE_TEST is also selected, right?
>
> This could be:
>
> depends on COMPILE_TEST
> depends on ARCH_BCM # 32-bit ARM
> depends on ARCH_BCM2835 || ARCH_BCM_IPROC # 64-bit ARM
>
> or maybe down to just:
>
> depends on COMPILE_TEST
>
> and let DRM, COMMON_CLK and SND drive the bulk of the dependencies?
That would disable it on ARCH_BCM2835 && !COMPILE_TEST, right?
COMPILE_TEST is typically disabled.
I think I should use ARCH_BCM2835 || ARCH_BCM || COMPILE_TEST. Does
that sound good?
Download attachment "signature.asc" of type "application/pgp-signature" (833 bytes)
Powered by blists - more mailing lists